flock-botnet/text.txt
text.txtRaw
1PREFACE TO THE FIRST EDITION, 1781
2
3Human reason, in one sphere of its cognition, is called upon to consider
4questions, which it cannot decline, as they are presented by its own
5nature, but which it cannot answer, as they transcend every faculty of
6the mind.
7
8It falls into this difficulty without any fault of its own. It begins
9with principles, which cannot be dispensed with in the field of
10experience, and the truth and sufficiency of which are, at the same
11time, insured by experience. With these principles it rises, in
12obedience to the laws of its own nature, to ever higher and more remote
13conditions. But it quickly discovers that, in this way, its labours must
14remain ever incomplete, because new questions never cease to present
15themselves; and thus it finds itself compelled to have recourse to
16principles which transcend the region of experience, while they are
17regarded by common sense without distrust. It thus falls into confusion
18and contradictions, from which it conjectures the presence of latent
19errors, which, however, it is unable to discover, because the principles
20it employs, transcending the limits of experience, cannot be tested
21by that criterion. The arena of these endless contests is called
22Metaphysic.
23
24Time was, when she was the queen of all the sciences; and, if we take
25the will for the deed, she certainly deserves, so far as regards the
26high importance of her object-matter, this title of honour. Now, it is
27the fashion of the time to heap contempt and scorn upon her; and the
28matron mourns, forlorn and forsaken, like Hecuba:
29
30 Modo maxima rerum,
31 Tot generis, natisque potens...
32 Nunc trahor exul, inops.
33 --Ovid, Metamorphoses. xiii
34
35At first, her government, under the administration of the dogmatists,
36was an absolute despotism. But, as the legislative continued to show
37traces of the ancient barbaric rule, her empire gradually broke up, and
38intestine wars introduced the reign of anarchy; while the sceptics,
39like nomadic tribes, who hate a permanent habitation and settled mode
40of living, attacked from time to time those who had organized themselves
41into civil communities. But their number was, very happily, small; and
42thus they could not entirely put a stop to the exertions of those who
43persisted in raising new edifices, although on no settled or uniform
44plan. In recent times the hope dawned upon us of seeing those disputes
45settled, and the legitimacy of her claims established by a kind of
46physiology of the human understanding--that of the celebrated Locke. But
47it was found that--although it was affirmed that this so-called queen
48could not refer her descent to any higher source than that of common
49experience, a circumstance which necessarily brought suspicion on
50her claims--as this genealogy was incorrect, she persisted in the
51advancement of her claims to sovereignty. Thus metaphysics necessarily
52fell back into the antiquated and rotten constitution of dogmatism, and
53again became obnoxious to the contempt from which efforts had been
54made to save it. At present, as all methods, according to the general
55persuasion, have been tried in vain, there reigns nought but weariness
56and complete indifferentism--the mother of chaos and night in the
57scientific world, but at the same time the source of, or at least the
58prelude to, the re-creation and reinstallation of a science, when it has
59fallen into confusion, obscurity, and disuse from ill directed effort.
60
61For it is in reality vain to profess indifference in regard to such
62inquiries, the object of which cannot be indifferent to humanity.
63Besides, these pretended indifferentists, however much they may try to
64disguise themselves by the assumption of a popular style and by changes
65on the language of the schools, unavoidably fall into metaphysical
66declarations and propositions, which they profess to regard with so much
67contempt. At the same time, this indifference, which has arisen in the
68world of science, and which relates to that kind of knowledge which
69we should wish to see destroyed the last, is a phenomenon that well
70deserves our attention and reflection. It is plainly not the effect of
71the levity, but of the matured judgement* of the age, which refuses to
72be any longer entertained with illusory knowledge, It is, in fact, a
73call to reason, again to undertake the most laborious of all tasks--that
74of self-examination, and to establish a tribunal, which may secure it
75in its well-grounded claims, while it pronounces against all baseless
76assumptions and pretensions, not in an arbitrary manner, but according
77to its own eternal and unchangeable laws. This tribunal is nothing less
78than the critical investigation of pure reason.
79
80[*Footnote: We very often hear complaints of the shallowness of the
81present age, and of the decay of profound science. But I do not think
82that those which rest upon a secure foundation, such as mathematics,
83physical science, etc., in the least deserve this reproach, but that
84they rather maintain their ancient fame, and in the latter case, indeed,
85far surpass it. The same would be the case with the other kinds of
86cognition, if their principles were but firmly established. In the
87absence of this security, indifference, doubt, and finally, severe
88criticism are rather signs of a profound habit of thought. Our age
89is the age of criticism, to which everything must be subjected. The
90sacredness of religion, and the authority of legislation, are by many
91regarded as grounds of exemption from the examination of this tribunal.
92But, if they on they are exempted, they become the subjects of just
93suspicion, and cannot lay claim to sincere respect, which reason accords
94only to that which has stood the test of a free and public examination.]
95
96I do not mean by this a criticism of books and systems, but a critical
97inquiry into the faculty of reason, with reference to the cognitions
98to which it strives to attain without the aid of experience; in other
99words, the solution of the question regarding the possibility or
100impossibility of metaphysics, and the determination of the origin, as
101well as of the extent and limits of this science. All this must be done
102on the basis of principles.
103
104This path--the only one now remaining--has been entered upon by me; and
105I flatter myself that I have, in this way, discovered the cause of--and
106consequently the mode of removing--all the errors which have hitherto
107set reason at variance with itself, in the sphere of non-empirical
108thought. I have not returned an evasive answer to the questions of
109reason, by alleging the inability and limitation of the faculties of the
110mind; I have, on the contrary, examined them completely in the light
111of principles, and, after having discovered the cause of the doubts and
112contradictions into which reason fell, have solved them to its perfect
113satisfaction. It is true, these questions have not been solved as
114dogmatism, in its vain fancies and desires, had expected; for it can
115only be satisfied by the exercise of magical arts, and of these I have
116no knowledge. But neither do these come within the compass of our mental
117powers; and it was the duty of philosophy to destroy the illusions which
118had their origin in misconceptions, whatever darling hopes and valued
119expectations may be ruined by its explanations. My chief aim in this
120work has been thoroughness; and I make bold to say that there is not a
121single metaphysical problem that does not find its solution, or at
122least the key to its solution, here. Pure reason is a perfect unity; and
123therefore, if the principle presented by it prove to be insufficient for
124the solution of even a single one of those questions to which the very
125nature of reason gives birth, we must reject it, as we could not be
126perfectly certain of its sufficiency in the case of the others.
127
128While I say this, I think I see upon the countenance of the reader signs
129of dissatisfaction mingled with contempt, when he hears declarations
130which sound so boastful and extravagant; and yet they are beyond
131comparison more moderate than those advanced by the commonest author of
132the commonest philosophical programme, in which the dogmatist professes
133to demonstrate the simple nature of the soul, or the necessity of a
134primal being. Such a dogmatist promises to extend human knowledge beyond
135the limits of possible experience; while I humbly confess that this
136is completely beyond my power. Instead of any such attempt, I confine
137myself to the examination of reason alone and its pure thought; and I do
138not need to seek far for the sum-total of its cognition, because it
139has its seat in my own mind. Besides, common logic presents me with
140a complete and systematic catalogue of all the simple operations of
141reason; and it is my task to answer the question how far reason can go,
142without the material presented and the aid furnished by experience.
143
144So much for the completeness and thoroughness necessary in the execution
145of the present task. The aims set before us are not arbitrarily
146proposed, but are imposed upon us by the nature of cognition itself.
147
148The above remarks relate to the matter of our critical inquiry. As
149regards the form, there are two indispensable conditions, which any one
150who undertakes so difficult a task as that of a critique of pure reason,
151is bound to fulfil. These conditions are certitude and clearness.
152
153As regards certitude, I have fully convinced myself that, in this sphere
154of thought, opinion is perfectly inadmissible, and that everything which
155bears the least semblance of an hypothesis must be excluded, as of no
156value in such discussions. For it is a necessary condition of every
157cognition that is to be established upon a priori grounds that it shall
158be held to be absolutely necessary; much more is this the case with an
159attempt to determine all pure a priori cognition, and to furnish the
160standard--and consequently an example--of all apodeictic (philosophical)
161certitude. Whether I have succeeded in what I professed to do, it is for
162the reader to determine; it is the author's business merely to adduce
163grounds and reasons, without determining what influence these ought to
164have on the mind of his judges. But, lest anything he may have said may
165become the innocent cause of doubt in their minds, or tend to weaken the
166effect which his arguments might otherwise produce--he may be allowed
167to point out those passages which may occasion mistrust or difficulty,
168although these do not concern the main purpose of the present work. He
169does this solely with the view of removing from the mind of the reader
170any doubts which might affect his judgement of the work as a whole, and
171in regard to its ultimate aim.
172
173I know no investigations more necessary for a full insight into the
174nature of the faculty which we call understanding, and at the same time
175for the determination of the rules and limits of its use, than those
176undertaken in the second chapter of the "Transcendental Analytic," under
177the title of "Deduction of the Pure Conceptions of the Understanding";
178and they have also cost me by far the greatest labour--labour which, I
179hope, will not remain uncompensated. The view there taken, which goes
180somewhat deeply into the subject, has two sides, The one relates to the
181objects of the pure understanding, and is intended to demonstrate and
182to render comprehensible the objective validity of its a priori
183conceptions; and it forms for this reason an essential part of the
184Critique. The other considers the pure understanding itself, its
185possibility and its powers of cognition--that is, from a subjective
186point of view; and, although this exposition is of great importance, it
187does not belong essentially to the main purpose of the work, because the
188grand question is what and how much can reason and understanding, apart
189from experience, cognize, and not, how is the faculty of thought itself
190possible? As the latter is an inquiry into the cause of a given effect,
191and has thus in it some semblance of an hypothesis (although, as I shall
192show on another occasion, this is really not the fact), it would seem
193that, in the present instance, I had allowed myself to enounce a mere
194opinion, and that the reader must therefore be at liberty to hold
195a different opinion. But I beg to remind him that, if my subjective
196deduction does not produce in his mind the conviction of its certitude
197at which I aimed, the objective deduction, with which alone the present
198work is properly concerned, is in every respect satisfactory.
199
200As regards clearness, the reader has a right to demand, in the first
201place, discursive or logical clearness, that is, on the basis of
202conceptions, and, secondly, intuitive or aesthetic clearness, by means
203of intuitions, that is, by examples or other modes of illustration
204in concreto. I have done what I could for the first kind of
205intelligibility. This was essential to my purpose; and it thus became
206the accidental cause of my inability to do complete justice to the
207second requirement. I have been almost always at a loss, during the
208progress of this work, how to settle this question. Examples and
209illustrations always appeared to me necessary, and, in the first sketch
210of the Critique, naturally fell into their proper places. But I very
211soon became aware of the magnitude of my task, and the numerous problems
212with which I should be engaged; and, as I perceived that this critical
213investigation would, even if delivered in the driest scholastic manner,
214be far from being brief, I found it unadvisable to enlarge it still more
215with examples and explanations, which are necessary only from a popular
216point of view. I was induced to take this course from the consideration
217also that the present work is not intended for popular use, that those
218devoted to science do not require such helps, although they are always
219acceptable, and that they would have materially interfered with my
220present purpose. Abbe Terrasson remarks with great justice that, if we
221estimate the size of a work, not from the number of its pages, but from
222the time which we require to make ourselves master of it, it may be said
223of many a book that it would be much shorter, if it were not so short.
224On the other hand, as regards the comprehensibility of a system of
225speculative cognition, connected under a single principle, we may say
226with equal justice: many a book would have been much clearer, if it had
227not been intended to be so very clear. For explanations and examples,
228and other helps to intelligibility, aid us in the comprehension of
229parts, but they distract the attention, dissipate the mental power of
230the reader, and stand in the way of his forming a clear conception of
231the whole; as he cannot attain soon enough to a survey of the system,
232and the colouring and embellishments bestowed upon it prevent his
233observing its articulation or organization--which is the most important
234consideration with him, when he comes to judge of its unity and
235stability.
236
237The reader must naturally have a strong inducement to co-operate
238with the present author, if he has formed the intention of erecting a
239complete and solid edifice of metaphysical science, according to the
240plan now laid before him. Metaphysics, as here represented, is the only
241science which admits of completion--and with little labour, if it
242is united, in a short time; so that nothing will be left to
243future generations except the task of illustrating and applying it
244didactically. For this science is nothing more than the inventory of all
245that is given us by pure reason, systematically arranged. Nothing can
246escape our notice; for what reason produces from itself cannot lie
247concealed, but must be brought to the light by reason itself, so soon
248as we have discovered the common principle of the ideas we seek. The
249perfect unity of this kind of cognitions, which are based upon pure
250conceptions, and uninfluenced by any empirical element, or any peculiar
251intuition leading to determinate experience, renders this completeness
252not only practicable, but also necessary.
253
254 Tecum habita, et noris quam sit tibi curta supellex.
255 --Persius. Satirae iv. 52.
256
257Such a system of pure speculative reason I hope to be able to publish
258under the title of Metaphysic of Nature*. The content of this work
259(which will not be half so long) will be very much richer than that
260of the present Critique, which has to discover the sources of this
261cognition and expose the conditions of its possibility, and at the same
262time to clear and level a fit foundation for the scientific edifice. In
263the present work, I look for the patient hearing and the impartiality
264of a judge; in the other, for the good-will and assistance of a
265co-labourer. For, however complete the list of principles for this
266system may be in the Critique, the correctness of the system requires
267that no deduced conceptions should be absent. These cannot be presented
268a priori, but must be gradually discovered; and, while the synthesis of
269conceptions has been fully exhausted in the Critique, it is necessary
270that, in the proposed work, the same should be the case with their
271analysis. But this will be rather an amusement than a labour.
272
273 [*Footnote: In contradistinction to the Metaphysic of
274 Ethics. This work was never published.]
275
276
277
278PREFACE TO THE SECOND EDITION, 1787
279
280Whether the treatment of that portion of our knowledge which lies within
281the province of pure reason advances with that undeviating certainty
282which characterizes the progress of science, we shall be at no loss to
283determine. If we find those who are engaged in metaphysical pursuits,
284unable to come to an understanding as to the method which they ought
285to follow; if we find them, after the most elaborate preparations,
286invariably brought to a stand before the goal is reached, and compelled
287to retrace their steps and strike into fresh paths, we may then feel
288quite sure that they are far from having attained to the certainty of
289scientific progress and may rather be said to be merely groping about in
290the dark. In these circumstances we shall render an important service to
291reason if we succeed in simply indicating the path along which it must
292travel, in order to arrive at any results--even if it should be found
293necessary to abandon many of those aims which, without reflection, have
294been proposed for its attainment.
295
296That logic has advanced in this sure course, even from the earliest
297times, is apparent from the fact that, since Aristotle, it has been
298unable to advance a step and, thus, to all appearance has reached its
299completion. For, if some of the moderns have thought to enlarge its
300domain by introducing psychological discussions on the mental faculties,
301such as imagination and wit, metaphysical, discussions on the origin
302of knowledge and the different kinds of certitude, according to
303the difference of the objects (idealism, scepticism, and so on), or
304anthropological discussions on prejudices, their causes and remedies:
305this attempt, on the part of these authors, only shows their ignorance
306of the peculiar nature of logical science. We do not enlarge but
307disfigure the sciences when we lose sight of their respective limits and
308allow them to run into one another. Now logic is enclosed within limits
309which admit of perfectly clear definition; it is a science which has for
310its object nothing but the exposition and proof of the formal laws of
311all thought, whether it be a priori or empirical, whatever be its
312origin or its object, and whatever the difficulties--natural or
313accidental--which it encounters in the human mind.
314
315The early success of logic must be attributed exclusively to the
316narrowness of its field, in which abstraction may, or rather must,
317be made of all the objects of cognition with their characteristic
318distinctions, and in which the understanding has only to deal with
319itself and with its own forms. It is, obviously, a much more difficult
320task for reason to strike into the sure path of science, where it has to
321deal not simply with itself, but with objects external to itself. Hence,
322logic is properly only a propaedeutic--forms, as it were, the vestibule
323of the sciences; and while it is necessary to enable us to form a
324correct judgement with regard to the various branches of knowledge,
325still the acquisition of real, substantive knowledge is to be sought
326only in the sciences properly so called, that is, in the objective
327sciences.
328
329Now these sciences, if they can be termed rational at all, must contain
330elements of a priori cognition, and this cognition may stand in a
331twofold relation to its object. Either it may have to determine the
332conception of the object--which must be supplied extraneously, or it
333may have to establish its reality. The former is theoretical, the latter
334practical, rational cognition. In both, the pure or a priori element
335must be treated first, and must be carefully distinguished from that
336which is supplied from other sources. Any other method can only lead to
337irremediable confusion.
338
339Mathematics and physics are the two theoretical sciences which have to
340determine their objects a priori. The former is purely a priori, the
341latter is partially so, but is also dependent on other sources of
342cognition.
343
344In the earliest times of which history affords us any record,
345mathematics had already entered on the sure course of science, among
346that wonderful nation, the Greeks. Still it is not to be supposed that
347it was as easy for this science to strike into, or rather to construct
348for itself, that royal road, as it was for logic, in which reason has
349only to deal with itself. On the contrary, I believe that it must
350have remained long--chiefly among the Egyptians--in the stage of
351blind groping after its true aims and destination, and that it was
352revolutionized by the happy idea of one man, who struck out and
353determined for all time the path which this science must follow,
354and which admits of an indefinite advancement. The history of this
355intellectual revolution--much more important in its results than the
356discovery of the passage round the celebrated Cape of Good Hope--and of
357its author, has not been preserved. But Diogenes Laertius, in naming
358the supposed discoverer of some of the simplest elements of geometrical
359demonstration--elements which, according to the ordinary opinion, do not
360even require to be proved--makes it apparent that the change introduced
361by the first indication of this new path, must have seemed of the utmost
362importance to the mathematicians of that age, and it has thus been
363secured against the chance of oblivion. A new light must have flashed on
364the mind of the first man (Thales, or whatever may have been his name)
365who demonstrated the properties of the isosceles triangle. For he found
366that it was not sufficient to meditate on the figure, as it lay before
367his eyes, or the conception of it, as it existed in his mind, and thus
368endeavour to get at the knowledge of its properties, but that it was
369necessary to produce these properties, as it were, by a positive a
370priori construction; and that, in order to arrive with certainty at
371a priori cognition, he must not attribute to the object any other
372properties than those which necessarily followed from that which he had
373himself, in accordance with his conception, placed in the object.
374
375A much longer period elapsed before physics entered on the highway of
376science. For it is only about a century and a half since the wise Bacon
377gave a new direction to physical studies, or rather--as others were
378already on the right track--imparted fresh vigour to the pursuit of
379this new direction. Here, too, as in the case of mathematics, we find
380evidence of a rapid intellectual revolution. In the remarks which follow
381I shall confine myself to the empirical side of natural science.
382
383When Galilei experimented with balls of a definite weight on the
384inclined plane, when Torricelli caused the air to sustain a weight which
385he had calculated beforehand to be equal to that of a definite column of
386water, or when Stahl, at a later period, converted metals into lime, and
387reconverted lime into metal, by the addition and subtraction of certain
388elements; [Footnote: I do not here follow with exactness the history of
389the experimental method, of which, indeed, the first steps are involved
390in some obscurity.] a light broke upon all natural philosophers. They
391learned that reason only perceives that which it produces after its
392own design; that it must not be content to follow, as it were, in the
393leading-strings of nature, but must proceed in advance with principles
394of judgement according to unvarying laws, and compel nature to reply
395its questions. For accidental observations, made according to no
396preconceived plan, cannot be united under a necessary law. But it is
397this that reason seeks for and requires. It is only the principles of
398reason which can give to concordant phenomena the validity of laws, and
399it is only when experiment is directed by these rational principles that
400it can have any real utility. Reason must approach nature with the view,
401indeed, of receiving information from it, not, however, in the character
402of a pupil, who listens to all that his master chooses to tell him,
403but in that of a judge, who compels the witnesses to reply to those
404questions which he himself thinks fit to propose. To this single idea
405must the revolution be ascribed, by which, after groping in the dark for
406so many centuries, natural science was at length conducted into the path
407of certain progress.
408
409We come now to metaphysics, a purely speculative science, which occupies
410a completely isolated position and is entirely independent of the
411teachings of experience. It deals with mere conceptions--not, like
412mathematics, with conceptions applied to intuition--and in it, reason is
413the pupil of itself alone. It is the oldest of the sciences, and would
414still survive, even if all the rest were swallowed up in the abyss of
415an all-destroying barbarism. But it has not yet had the good fortune to
416attain to the sure scientific method. This will be apparent; if we
417apply the tests which we proposed at the outset. We find that reason
418perpetually comes to a stand, when it attempts to gain a priori the
419perception even of those laws which the most common experience confirms.
420We find it compelled to retrace its steps in innumerable instances, and
421to abandon the path on which it had entered, because this does not
422lead to the desired result. We find, too, that those who are engaged in
423metaphysical pursuits are far from being able to agree among themselves,
424but that, on the contrary, this science appears to furnish an arena
425specially adapted for the display of skill or the exercise of strength
426in mock-contests--a field in which no combatant ever yet succeeded in
427gaining an inch of ground, in which, at least, no victory was ever yet
428crowned with permanent possession.
429
430This leads us to inquire why it is that, in metaphysics, the sure path
431of science has not hitherto been found. Shall we suppose that it is
432impossible to discover it? Why then should nature have visited our
433reason with restless aspirations after it, as if it were one of our
434weightiest concerns? Nay, more, how little cause should we have to place
435confidence in our reason, if it abandons us in a matter about which,
436most of all, we desire to know the truth--and not only so, but even
437allures us to the pursuit of vain phantoms, only to betray us in the
438end? Or, if the path has only hitherto been missed, what indications do
439we possess to guide us in a renewed investigation, and to enable us to
440hope for greater success than has fallen to the lot of our predecessors?
441
442It appears to me that the examples of mathematics and natural
443philosophy, which, as we have seen, were brought into their present
444condition by a sudden revolution, are sufficiently remarkable to fix our
445attention on the essential circumstances of the change which has proved
446so advantageous to them, and to induce us to make the experiment of
447imitating them, so far as the analogy which, as rational sciences, they
448bear to metaphysics may permit. It has hitherto been assumed that our
449cognition must conform to the objects; but all attempts to ascertain
450anything about these objects a priori, by means of conceptions, and thus
451to extend the range of our knowledge, have been rendered abortive by
452this assumption. Let us then make the experiment whether we may not
453be more successful in metaphysics, if we assume that the objects must
454conform to our cognition. This appears, at all events, to accord better
455with the possibility of our gaining the end we have in view, that is to
456say, of arriving at the cognition of objects a priori, of determining
457something with respect to these objects, before they are given to us. We
458here propose to do just what Copernicus did in attempting to explain
459the celestial movements. When he found that he could make no progress by
460assuming that all the heavenly bodies revolved round the spectator,
461he reversed the process, and tried the experiment of assuming that the
462spectator revolved, while the stars remained at rest. We may make
463the same experiment with regard to the intuition of objects. If the
464intuition must conform to the nature of the objects, I do not see how
465we can know anything of them a priori. If, on the other hand, the object
466conforms to the nature of our faculty of intuition, I can then easily
467conceive the possibility of such an a priori knowledge. Now as I cannot
468rest in the mere intuitions, but--if they are to become cognitions--must
469refer them, as representations, to something, as object, and must
470determine the latter by means of the former, here again there are two
471courses open to me. Either, first, I may assume that the conceptions,
472by which I effect this determination, conform to the object--and in this
473case I am reduced to the same perplexity as before; or secondly, I may
474assume that the objects, or, which is the same thing, that experience,
475in which alone as given objects they are cognized, conform to my
476conceptions--and then I am at no loss how to proceed. For experience
477itself is a mode of cognition which requires understanding. Before
478objects, are given to me, that is, a priori, I must presuppose in myself
479laws of the understanding which are expressed in conceptions a
480priori. To these conceptions, then, all the objects of experience must
481necessarily conform. Now there are objects which reason thinks, and
482that necessarily, but which cannot be given in experience, or, at least,
483cannot be given so as reason thinks them. The attempt to think these
484objects will hereafter furnish an excellent test of the new method of
485thought which we have adopted, and which is based on the principle that
486we only cognize in things a priori that which we ourselves place in
487them.*
488
489 [*Footnote: This method, accordingly, which we have borrowed
490 from the natural philosopher, consists in seeking for the
491 elements of pure reason in that which admits of confirmation
492 or refutation by experiment. Now the propositions of pure
493 reason, especially when they transcend the limits of
494 possible experience, do not admit of our making any
495 experiment with their objects, as in natural science. Hence,
496 with regard to those conceptions and principles which we
497 assume a priori, our only course ill be to view them from
498 two different sides. We must regard one and the same
499 conception, on the one hand, in relation to experience as an
500 object of the senses and of the understanding, on the other
501 hand, in relation to reason, isolated and transcending the
502 limits of experience, as an object of mere thought. Now if
503 we find that, when we regard things from this double point
504 of view, the result is in harmony with the principle of pure
505 reason, but that, when we regard them from a single point of
506 view, reason is involved in self-contradiction, then the
507 experiment will establish the correctness of this
508 distinction.]
509
510This attempt succeeds as well as we could desire, and promises to
511metaphysics, in its first part--that is, where it is occupied with
512conceptions a priori, of which the corresponding objects may be given in
513experience--the certain course of science. For by this new method we are
514enabled perfectly to explain the possibility of a priori cognition, and,
515what is more, to demonstrate satisfactorily the laws which lie a
516priori at the foundation of nature, as the sum of the objects of
517experience--neither of which was possible according to the procedure
518hitherto followed. But from this deduction of the faculty of a priori
519cognition in the first part of metaphysics, we derive a surprising
520result, and one which, to all appearance, militates against the great
521end of metaphysics, as treated in the second part. For we come to the
522conclusion that our faculty of cognition is unable to transcend the
523limits of possible experience; and yet this is precisely the most
524essential object of this science. The estimate of our rational cognition
525a priori at which we arrive is that it has only to do with phenomena,
526and that things in themselves, while possessing a real existence,
527lie beyond its sphere. Here we are enabled to put the justice of this
528estimate to the test. For that which of necessity impels us to transcend
529the limits of experience and of all phenomena is the unconditioned,
530which reason absolutely requires in things as they are in themselves,
531in order to complete the series of conditions. Now, if it appears that
532when, on the one hand, we assume that our cognition conforms to its
533objects as things in themselves, the unconditioned cannot be thought
534without contradiction, and that when, on the other hand, we assume that
535our representation of things as they are given to us, does not conform
536to these things as they are in themselves, but that these objects, as
537phenomena, conform to our mode of representation, the contradiction
538disappears: we shall then be convinced of the truth of that which we
539began by assuming for the sake of experiment; we may look upon it as
540established that the unconditioned does not lie in things as we
541know them, or as they are given to us, but in things as they are in
542themselves, beyond the range of our cognition.*
543
544 [*Footnote: This experiment of pure reason has a great
545 similarity to that of the chemists, which they term the
546 experiment of reduction, or, more usually, the synthetic
547 process. The analysis of the metaphysician separates pure
548 cognition a priori into two heterogeneous elements, viz.,
549 the cognition of things as phenomena, and of things in
550 themselves. Dialectic combines these again into harmony with
551 the necessary rational idea of the unconditioned, and finds
552 that this harmony never results except through the above
553 distinction, which is, therefore, concluded to be just.]
554
555But, after we have thus denied the power of speculative reason to make
556any progress in the sphere of the supersensible, it still remains for
557our consideration whether data do not exist in practical cognition
558which may enable us to determine the transcendent conception of the
559unconditioned, to rise beyond the limits of all possible experience
560from a practical point of view, and thus to satisfy the great ends of
561metaphysics. Speculative reason has thus, at least, made room for such
562an extension of our knowledge: and, if it must leave this space vacant,
563still it does not rob us of the liberty to fill it up, if we can,
564by means of practical data--nay, it even challenges us to make the
565attempt.*
566
567 [*Footnote: So the central laws of the movements of the
568 heavenly bodies established the truth of that which
569 Copernicus, first, assumed only as a hypothesis, and, at the
570 same time, brought to light that invisible force (Newtonian
571 attraction) which holds the universe together. The latter
572 would have remained forever undiscovered, if Copernicus had
573 not ventured on the experiment--contrary to the senses but
574 still just--of looking for the observed movements not in the
575 heavenly bodies, but in the spectator. In this Preface I
576 treat the new metaphysical method as a hypothesis with the
577 view of rendering apparent the first attempts at such a
578 change of method, which are always hypothetical. But in the
579 Critique itself it will be demonstrated, not hypothetically,
580 but apodeictically, from the nature of our representations
581 of space and time, and from the elementary conceptions of
582 the understanding.]
583
584
585This attempt to introduce a complete revolution in the procedure
586of metaphysics, after the example of the geometricians and natural
587philosophers, constitutes the aim of the Critique of Pure Speculative
588Reason. It is a treatise on the method to be followed, not a system of
589the science itself. But, at the same time, it marks out and defines both
590the external boundaries and the internal structure of this science.
591For pure speculative reason has this peculiarity, that, in choosing the
592various objects of thought, it is able to define the limits of its own
593faculties, and even to give a complete enumeration of the possible
594modes of proposing problems to itself, and thus to sketch out the entire
595system of metaphysics. For, on the one hand, in cognition a priori,
596nothing must be attributed to the objects but what the thinking subject
597derives from itself; and, on the other hand, reason is, in regard to
598the principles of cognition, a perfectly distinct, independent unity, in
599which, as in an organized body, every member exists for the sake of
600the others, and all for the sake of each, so that no principle can be
601viewed, with safety, in one relationship, unless it is, at the same
602time, viewed in relation to the total use of pure reason. Hence, too,
603metaphysics has this singular advantage--an advantage which falls to the
604lot of no other science which has to do with objects--that, if once it
605is conducted into the sure path of science, by means of this criticism,
606it can then take in the whole sphere of its cognitions, and can thus
607complete its work, and leave it for the use of posterity, as a capital
608which can never receive fresh accessions. For metaphysics has to deal
609only with principles and with the limitations of its own employment as
610determined by these principles. To this perfection it is, therefore,
611bound, as the fundamental science, to attain, and to it the maxim may
612justly be applied:
613
614 Nil actum reputans, si quid superesset agendum.
615
616But, it will be asked, what kind of a treasure is this that we propose
617to bequeath to posterity? What is the real value of this system of
618metaphysics, purified by criticism, and thereby reduced to a permanent
619condition? A cursory view of the present work will lead to the
620supposition that its use is merely negative, that it only serves to
621warn us against venturing, with speculative reason, beyond the limits
622of experience. This is, in fact, its primary use. But this, at once,
623assumes a positive value, when we observe that the principles with which
624speculative reason endeavours to transcend its limits lead inevitably,
625not to the extension, but to the contraction of the use of reason,
626inasmuch as they threaten to extend the limits of sensibility, which
627is their proper sphere, over the entire realm of thought and, thus,
628to supplant the pure (practical) use of reason. So far, then, as this
629criticism is occupied in confining speculative reason within its proper
630bounds, it is only negative; but, inasmuch as it thereby, at the same
631time, removes an obstacle which impedes and even threatens to destroy
632the use of practical reason, it possesses a positive and very important
633value. In order to admit this, we have only to be convinced that there
634is an absolutely necessary use of pure reason--the moral use--in which
635it inevitably transcends the limits of sensibility, without the aid
636of speculation, requiring only to be insured against the effects of a
637speculation which would involve it in contradiction with itself. To deny
638the positive advantage of the service which this criticism renders
639us would be as absurd as to maintain that the system of police is
640productive of no positive benefit, since its main business is to prevent
641the violence which citizen has to apprehend from citizen, that so each
642may pursue his vocation in peace and security. That space and time are
643only forms of sensible intuition, and hence are only conditions of the
644existence of things as phenomena; that, moreover, we have no conceptions
645of the understanding, and, consequently, no elements for the cognition
646of things, except in so far as a corresponding intuition can be given
647to these conceptions; that, accordingly, we can have no cognition of
648an object, as a thing in itself, but only as an object of sensible
649intuition, that is, as phenomenon--all this is proved in the analytical
650part of the Critique; and from this the limitation of all possible
651speculative cognition to the mere objects of experience, follows as a
652necessary result. At the same time, it must be carefully borne in mind
653that, while we surrender the power of cognizing, we still reserve the
654power of thinking objects, as things in themselves.* For, otherwise,
655we should require to affirm the existence of an appearance, without
656something that appears--which would be absurd. Now let us suppose, for a
657moment, that we had not undertaken this criticism and, accordingly,
658had not drawn the necessary distinction between things as objects
659of experience and things as they are in themselves. The principle of
660causality, and, by consequence, the mechanism of nature as determined by
661causality, would then have absolute validity in relation to all things
662as efficient causes. I should then be unable to assert, with regard to
663one and the same being, e.g., the human soul, that its will is free, and
664yet, at the same time, subject to natural necessity, that is, not free,
665without falling into a palpable contradiction, for in both propositions
666I should take the soul in the same signification, as a thing in general,
667as a thing in itself--as, without previous criticism, I could not but
668take it. Suppose now, on the other hand, that we have undertaken this
669criticism, and have learnt that an object may be taken in two senses,
670first, as a phenomenon, secondly, as a thing in itself; and that,
671according to the deduction of the conceptions of the understanding, the
672principle of causality has reference only to things in the first sense.
673We then see how it does not involve any contradiction to assert, on the
674one hand, that the will, in the phenomenal sphere--in visible action--is
675necessarily obedient to the law of nature, and, in so far, not free;
676and, on the other hand, that, as belonging to a thing in itself, it is
677not subject to that law, and, accordingly, is free. Now, it is true that
678I cannot, by means of speculative reason, and still less by empirical
679observation, cognize my soul as a thing in itself and consequently,
680cannot cognize liberty as the property of a being to which I ascribe
681effects in the world of sense. For, to do so, I must cognize this being
682as existing, and yet not in time, which--since I cannot support my
683conception by any intuition--is impossible. At the same time, while
684I cannot cognize, I can quite well think freedom, that is to say, my
685representation of it involves at least no contradiction, if we bear in
686mind the critical distinction of the two modes of representation (the
687sensible and the intellectual) and the consequent limitation of the
688conceptions of the pure understanding and of the principles which flow
689from them. Suppose now that morality necessarily presupposed liberty,
690in the strictest sense, as a property of our will; suppose that reason
691contained certain practical, original principles a priori, which were
692absolutely impossible without this presupposition; and suppose, at the
693same time, that speculative reason had proved that liberty was
694incapable of being thought at all. It would then follow that the
695moral presupposition must give way to the speculative affirmation, the
696opposite of which involves an obvious contradiction, and that liberty
697and, with it, morality must yield to the mechanism of nature; for
698the negation of morality involves no contradiction, except on the
699presupposition of liberty. Now morality does not require the speculative
700cognition of liberty; it is enough that I can think it, that its
701conception involves no contradiction, that it does not interfere with
702the mechanism of nature. But even this requirement we could not satisfy,
703if we had not learnt the twofold sense in which things may be taken; and
704it is only in this way that the doctrine of morality and the doctrine of
705nature are confined within their proper limits. For this result,
706then, we are indebted to a criticism which warns us of our unavoidable
707ignorance with regard to things in themselves, and establishes the
708necessary limitation of our theoretical cognition to mere phenomena.
709
710 [*Footnote: In order to cognize an object, I must be able to
711 prove its possibility, either from its reality as attested
712 by experience, or a priori, by means of reason. But I can
713 think what I please, provided only I do not contradict
714 myself; that is, provided my conception is a possible
715 thought, though I may be unable to answer for the existence
716 of a corresponding object in the sum of possibilities. But
717 something more is required before I can attribute to such a
718 conception objective validity, that is real possibility--the
719 other possibility being merely logical. We are not, however,
720 confined to theoretical sources of cognition for the means
721 of satisfying this additional requirement, but may derive
722 them from practical sources.]
723
724
725The positive value of the critical principles of pure reason in relation
726to the conception of God and of the simple nature of the soul, admits of
727a similar exemplification; but on this point I shall not dwell. I
728cannot even make the assumption--as the practical interests of
729morality require--of God, freedom, and immortality, if I do not deprive
730speculative reason of its pretensions to transcendent insight. For to
731arrive at these, it must make use of principles which, in fact, extend
732only to the objects of possible experience, and which cannot be applied
733to objects beyond this sphere without converting them into phenomena,
734and thus rendering the practical extension of pure reason impossible.
735I must, therefore, abolish knowledge, to make room for belief. The
736dogmatism of metaphysics, that is, the presumption that it is possible
737to advance in metaphysics without previous criticism, is the true source
738of the unbelief (always dogmatic) which militates against morality.
739
740Thus, while it may be no very difficult task to bequeath a legacy
741to posterity, in the shape of a system of metaphysics constructed in
742accordance with the Critique of Pure Reason, still the value of such a
743bequest is not to be depreciated. It will render an important service
744to reason, by substituting the certainty of scientific method for that
745random groping after results without the guidance of principles, which
746has hitherto characterized the pursuit of metaphysical studies. It will
747render an important service to the inquiring mind of youth, by leading
748the student to apply his powers to the cultivation of genuine science,
749instead of wasting them, as at present, on speculations which can never
750lead to any result, or on the idle attempt to invent new ideas and
751opinions. But, above all, it will confer an inestimable benefit on
752morality and religion, by showing that all the objections urged against
753them may be silenced for ever by the Socratic method, that is to say, by
754proving the ignorance of the objector. For, as the world has never been,
755and, no doubt, never will be without a system of metaphysics of one kind
756or another, it is the highest and weightiest concern of philosophy to
757render it powerless for harm, by closing up the sources of error.
758
759This important change in the field of the sciences, this loss of its
760fancied possessions, to which speculative reason must submit, does not
761prove in any way detrimental to the general interests of humanity. The
762advantages which the world has derived from the teachings of pure reason
763are not at all impaired. The loss falls, in its whole extent, on the
764monopoly of the schools, but does not in the slightest degree touch the
765interests of mankind. I appeal to the most obstinate dogmatist, whether
766the proof of the continued existence of the soul after death, derived
767from the simplicity of its substance; of the freedom of the will in
768opposition to the general mechanism of nature, drawn from the subtle but
769impotent distinction of subjective and objective practical necessity;
770or of the existence of God, deduced from the conception of an ens
771realissimum--the contingency of the changeable, and the necessity of
772a prime mover, has ever been able to pass beyond the limits of the
773schools, to penetrate the public mind, or to exercise the slightest
774influence on its convictions. It must be admitted that this has not been
775the case and that, owing to the unfitness of the common understanding
776for such subtle speculations, it can never be expected to take place. On
777the contrary, it is plain that the hope of a future life arises from the
778feeling, which exists in the breast of every man, that the temporal
779is inadequate to meet and satisfy the demands of his nature. In like
780manner, it cannot be doubted that the clear exhibition of duties
781in opposition to all the claims of inclination, gives rise to the
782consciousness of freedom, and that the glorious order, beauty, and
783providential care, everywhere displayed in nature, give rise to the
784belief in a wise and great Author of the Universe. Such is the genesis
785of these general convictions of mankind, so far as they depend on
786rational grounds; and this public property not only remains undisturbed,
787but is even raised to greater importance, by the doctrine that the
788schools have no right to arrogate to themselves a more profound insight
789into a matter of general human concernment than that to which the great
790mass of men, ever held by us in the highest estimation, can without
791difficulty attain, and that the schools should, therefore, confine
792themselves to the elaboration of these universally comprehensible and,
793from a moral point of view, amply satisfactory proofs. The change,
794therefore, affects only the arrogant pretensions of the schools, which
795would gladly retain, in their own exclusive possession, the key to the
796truths which they impart to the public.
797
798Quod mecum nescit, solus vult scire videri.
799
800At the same time it does not deprive the speculative philosopher of
801his just title to be the sole depositor of a science which benefits the
802public without its knowledge--I mean, the Critique of Pure Reason. This
803can never become popular and, indeed, has no occasion to be so; for
804finespun arguments in favour of useful truths make just as little
805impression on the public mind as the equally subtle objections brought
806against these truths. On the other hand, since both inevitably force
807themselves on every man who rises to the height of speculation, it
808becomes the manifest duty of the schools to enter upon a thorough
809investigation of the rights of speculative reason and, thus, to prevent
810the scandal which metaphysical controversies are sure, sooner or later,
811to cause even to the masses. It is only by criticism that metaphysicians
812(and, as such, theologians too) can be saved from these controversies
813and from the consequent perversion of their doctrines. Criticism
814alone can strike a blow at the root of materialism, fatalism, atheism,
815free-thinking, fanaticism, and superstition, which are universally
816injurious--as well as of idealism and scepticism, which are dangerous
817to the schools, but can scarcely pass over to the public. If governments
818think proper to interfere with the affairs of the learned, it would be
819more consistent with a wise regard for the interests of science, as well
820as for those of society, to favour a criticism of this kind, by which
821alone the labours of reason can be established on a firm basis, than to
822support the ridiculous despotism of the schools, which raise a loud cry
823of danger to the public over the destruction of cobwebs, of which the
824public has never taken any notice, and the loss of which, therefore, it
825can never feel.
826
827This critical science is not opposed to the dogmatic procedure of reason
828in pure cognition; for pure cognition must always be dogmatic, that is,
829must rest on strict demonstration from sure principles a priori--but to
830dogmatism, that is, to the presumption that it is possible to make
831any progress with a pure cognition, derived from (philosophical)
832conceptions, according to the principles which reason has long been in
833the habit of employing--without first inquiring in what way and by what
834right reason has come into the possession of these principles. Dogmatism
835is thus the dogmatic procedure of pure reason without previous criticism
836of its own powers, and in opposing this procedure, we must not be
837supposed to lend any countenance to that loquacious shallowness which
838arrogates to itself the name of popularity, nor yet to scepticism, which
839makes short work with the whole science of metaphysics. On the contrary,
840our criticism is the necessary preparation for a thoroughly scientific
841system of metaphysics which must perform its task entirely a priori, to
842the complete satisfaction of speculative reason, and must, therefore,
843be treated, not popularly, but scholastically. In carrying out the
844plan which the Critique prescribes, that is, in the future system
845of metaphysics, we must have recourse to the strict method of the
846celebrated Wolf, the greatest of all dogmatic philosophers. He was the
847first to point out the necessity of establishing fixed principles, of
848clearly defining our conceptions, and of subjecting our demonstrations
849to the most severe scrutiny, instead of rashly jumping at conclusions.
850The example which he set served to awaken that spirit of profound and
851thorough investigation which is not yet extinct in Germany. He would
852have been peculiarly well fitted to give a truly scientific character to
853metaphysical studies, had it occurred to him to prepare the field by a
854criticism of the organum, that is, of pure reason itself. That he failed
855to perceive the necessity of such a procedure must be ascribed to the
856dogmatic mode of thought which characterized his age, and on this point
857the philosophers of his time, as well as of all previous times, have
858nothing to reproach each other with. Those who reject at once the method
859of Wolf, and of the Critique of Pure Reason, can have no other aim
860but to shake off the fetters of science, to change labour into sport,
861certainty into opinion, and philosophy into philodoxy.
862
863In this second edition, I have endeavoured, as far as possible, to
864remove the difficulties and obscurity which, without fault of mine
865perhaps, have given rise to many misconceptions even among acute
866thinkers. In the propositions themselves, and in the demonstrations by
867which they are supported, as well as in the form and the entire plan of
868the work, I have found nothing to alter; which must be attributed
869partly to the long examination to which I had subjected the whole before
870offering it to the public and partly to the nature of the case. For pure
871speculative reason is an organic structure in which there is nothing
872isolated or independent, but every Single part is essential to all the
873rest; and hence, the slightest imperfection, whether defect or positive
874error, could not fail to betray itself in use. I venture, further, to
875hope, that this system will maintain the same unalterable character for
876the future. I am led to entertain this confidence, not by vanity, but by
877the evidence which the equality of the result affords, when we proceed,
878first, from the simplest elements up to the complete whole of pure
879reason and, and then, backwards from the whole to each part. We find
880that the attempt to make the slightest alteration, in any part, leads
881inevitably to contradictions, not merely in this system, but in
882human reason itself. At the same time, there is still much room for
883improvement in the exposition of the doctrines contained in this work.
884In the present edition, I have endeavoured to remove misapprehensions of
885the aesthetical part, especially with regard to the conception of time;
886to clear away the obscurity which has been found in the deduction of
887the conceptions of the understanding; to supply the supposed want of
888sufficient evidence in the demonstration of the principles of the pure
889understanding; and, lastly, to obviate the misunderstanding of the
890paralogisms which immediately precede the rational psychology. Beyond
891this point--the end of the second main division of the "Transcendental
892Dialectic"--I have not extended my alterations,* partly from want
893of time, and partly because I am not aware that any portion of the
894remainder has given rise to misconceptions among intelligent and
895impartial critics, whom I do not here mention with that praise which is
896their due, but who will find that their suggestions have been attended
897to in the work itself.
898
899 [*Footnote: The only addition, properly so called--and that
900 only in the method of proof--which I have made in the
901 present edition, consists of a new refutation of
902 psychological idealism, and a strict demonstration--the only
903 one possible, as I believe--of the objective reality of
904 external intuition. However harmless idealism may be
905 considered--although in reality it is not so--in regard to
906 the essential ends of metaphysics, it must still remain a
907 scandal to philosophy and to the general human reason to be
908 obliged to assume, as an article of mere belief, the
909 existence of things external to ourselves (from which, yet,
910 we derive the whole material of cognition for the internal
911 sense), and not to be able to oppose a satisfactory proof to
912 any one who may call it in question. As there is some
913 obscurity of expression in the demonstration as it stands in
914 the text, I propose to alter the passage in question as
915 follows: "But this permanent cannot be an intuition in me.
916 For all the determining grounds of my existence which can be
917 found in me are representations and, as such, do themselves
918 require a permanent, distinct from them, which may determine
919 my existence in relation to their changes, that is, my
920 existence in time, wherein they change." It may, probably,
921 be urged in opposition to this proof that, after all, I am
922 only conscious immediately of that which is in me, that is,
923 of my representation of external things, and that,
924 consequently, it must always remain uncertain whether
925 anything corresponding to this representation does or does
926 not exist externally to me. But I am conscious, through
927 internal experience, of my existence in time (consequently,
928 also, of the determinability of the former in the latter),
929 and that is more than the simple consciousness of my
930 representation. It is, in fact, the same as the empirical
931 consciousness of my existence, which can only be determined
932 in relation to something, which, while connected with my
933 existence, is external to me. This consciousness of my
934 existence in time is, therefore, identical with the
935 consciousness of a relation to something external to me, and
936 it is, therefore, experience, not fiction, sense, not
937 imagination, which inseparably connects the external with my
938 internal sense. For the external sense is, in itself, the
939 relation of intuition to something real, external to me; and
940 the reality of this something, as opposed to the mere
941 imagination of it, rests solely on its inseparable
942 connection with internal experience as the condition of its
943 possibility. If with the intellectual consciousness of my
944 existence, in the representation: I am, which accompanies
945 all my judgements, and all the operations of my
946 understanding, I could, at the same time, connect a
947 determination of my existence by intellectual intuition,
948 then the consciousness of a relation to something external
949 to me would not be necessary. But the internal intuition in
950 which alone my existence can be determined, though preceded
951 by that purely intellectual consciousness, is itself
952 sensible and attached to the condition of time. Hence this
953 determination of my existence, and consequently my internal
954 experience itself, must depend on something permanent which
955 is not in me, which can be, therefore, only in something
956 external to me, to which I must look upon myself as being
957 related. Thus the reality of the external sense is
958 necessarily connected with that of the internal, in order to
959 the possibility of experience in general; that is, I am just
960 as certainly conscious that there are things external to me
961 related to my sense as I am that I myself exist as
962 determined in time. But in order to ascertain to what given
963 intuitions objects, external me, really correspond, in other
964 words, what intuitions belong to the external sense and not
965 to imagination, I must have recourse, in every particular
966 case, to those rules according to which experience in
967 general (even internal experience) is distinguished from
968 imagination, and which are always based on the proposition
969 that there really is an external experience. We may add the
970 remark that the representation of something permanent in
971 existence, is not the same thing as the permanent
972 representation; for a representation may be very variable
973 and changing--as all our representations, even that of
974 matter, are--and yet refer to something permanent, which
975 must, therefore, be distinct from all my representations and
976 external to me, the existence of which is necessarily
977 included in the determination of my own existence, and with
978 it constitutes one experience--an experience which would not
979 even be possible internally, if it were not also at the same
980 time, in part, external. To the question How? we are no more
981 able to reply, than we are, in general, to think the
982 stationary in time, the coexistence of which with the
983 variable, produces the conception of change.]
984
985
986In attempting to render the exposition of my views as intelligible as
987possible, I have been compelled to leave out or abridge various passages
988which were not essential to the completeness of the work, but which many
989readers might consider useful in other respects, and might be unwilling
990to miss. This trifling loss, which could not be avoided without swelling
991the book beyond due limits, may be supplied, at the pleasure of the
992reader, by a comparison with the first edition, and will, I hope, be
993more than compensated for by the greater clearness of the exposition as
994it now stands.
995
996I have observed, with pleasure and thankfulness, in the pages of
997various reviews and treatises, that the spirit of profound and thorough
998investigation is not extinct in Germany, though it may have been
999overborne and silenced for a time by the fashionable tone of a licence
1000in thinking, which gives itself the airs of genius, and that the
1001difficulties which beset the paths of criticism have not prevented
1002energetic and acute thinkers from making themselves masters of the
1003science of pure reason to which these paths conduct--a science which is
1004not popular, but scholastic in its character, and which alone can hope
1005for a lasting existence or possess an abiding value. To these deserving
1006men, who so happily combine profundity of view with a talent for lucid
1007exposition--a talent which I myself am not conscious of possessing--I
1008leave the task of removing any obscurity which may still adhere to the
1009statement of my doctrines. For, in this case, the danger is not that
1010of being refuted, but of being misunderstood. For my own part, I must
1011henceforward abstain from controversy, although I shall carefully attend
1012to all suggestions, whether from friends or adversaries, which may be
1013of use in the future elaboration of the system of this propaedeutic. As,
1014during these labours, I have advanced pretty far in years this month
1015I reach my sixty-fourth year--it will be necessary for me to economize
1016time, if I am to carry out my plan of elaborating the metaphysics of
1017nature as well as of morals, in confirmation of the correctness of the
1018principles established in this Critique of Pure Reason, both speculative
1019and practical; and I must, therefore, leave the task of clearing up the
1020obscurities of the present work--inevitable, perhaps, at the outset--as
1021well as, the defence of the whole, to those deserving men, who have made
1022my system their own. A philosophical system cannot come forward armed
1023at all points like a mathematical treatise, and hence it may be quite
1024possible to take objection to particular passages, while the organic
1025structure of the system, considered as a unity, has no danger to
1026apprehend. But few possess the ability, and still fewer the inclination,
1027to take a comprehensive view of a new system. By confining the view to
1028particular passages, taking these out of their connection and comparing
1029them with one another, it is easy to pick out apparent contradictions,
1030especially in a work written with any freedom of style. These
1031contradictions place the work in an unfavourable light in the eyes of
1032those who rely on the judgement of others, but are easily reconciled
1033by those who have mastered the idea of the whole. If a theory possesses
1034stability in itself, the action and reaction which seemed at first to
1035threaten its existence serve only, in the course of time, to smooth
1036down any superficial roughness or inequality, and--if men of insight,
1037impartiality, and truly popular gifts, turn their attention to it--to
1038secure to it, in a short time, the requisite elegance also.
1039
1040Konigsberg, April 1787.
1041
1042
1043
1044INTRODUCTION
1045
1046
1047I. Of the difference between Pure and Empirical Knowledge
1048
1049That all our knowledge begins with experience there can be no doubt. For
1050how is it possible that the faculty of cognition should be awakened into
1051exercise otherwise than by means of objects which affect our senses, and
1052partly of themselves produce representations, partly rouse our powers
1053of understanding into activity, to compare to connect, or to separate
1054these, and so to convert the raw material of our sensuous impressions
1055into a knowledge of objects, which is called experience? In respect of
1056time, therefore, no knowledge of ours is antecedent to experience, but
1057begins with it.
1058
1059But, though all our knowledge begins with experience, it by no means
1060follows that all arises out of experience. For, on the contrary, it is
1061quite possible that our empirical knowledge is a compound of that which
1062we receive through impressions, and that which the faculty of cognition
1063supplies from itself (sensuous impressions giving merely the occasion),
1064an addition which we cannot distinguish from the original element given
1065by sense, till long practice has made us attentive to, and skilful
1066in separating it. It is, therefore, a question which requires close
1067investigation, and not to be answered at first sight, whether there
1068exists a knowledge altogether independent of experience, and even of
1069all sensuous impressions? Knowledge of this kind is called a priori,
1070in contradistinction to empirical knowledge, which has its sources a
1071posteriori, that is, in experience.
1072
1073But the expression, "a priori," is not as yet definite enough adequately
1074to indicate the whole meaning of the question above started. For, in
1075speaking of knowledge which has its sources in experience, we are wont
1076to say, that this or that may be known a priori, because we do not
1077derive this knowledge immediately from experience, but from a general
1078rule, which, however, we have itself borrowed from experience. Thus,
1079if a man undermined his house, we say, "he might know a priori that
1080it would have fallen;" that is, he needed not to have waited for the
1081experience that it did actually fall. But still, a priori, he could not
1082know even this much. For, that bodies are heavy, and, consequently, that
1083they fall when their supports are taken away, must have been known to
1084him previously, by means of experience.
1085
1086By the term "knowledge a priori," therefore, we shall in the sequel
1087understand, not such as is independent of this or that kind of
1088experience, but such as is absolutely so of all experience. Opposed
1089to this is empirical knowledge, or that which is possible only a
1090posteriori, that is, through experience. Knowledge a priori is either
1091pure or impure. Pure knowledge a priori is that with which no empirical
1092element is mixed up. For example, the proposition, "Every change has
1093a cause," is a proposition a priori, but impure, because change is a
1094conception which can only be derived from experience.
1095
1096
1097
1098II. The Human Intellect, even in an Unphilosophical State, is in
1099Possession of Certain Cognitions "a priori".
1100
1101The question now is as to a criterion, by which we may securely
1102distinguish a pure from an empirical cognition. Experience no doubt
1103teaches us that this or that object is constituted in such and such a
1104manner, but not that it could not possibly exist otherwise. Now, in
1105the first place, if we have a proposition which contains the idea
1106of necessity in its very conception, it is priori. If, moreover, it is not
1107derived from any other proposition, unless from one equally involving
1108the idea of necessity, it is absolutely priori. Secondly, an empirical
1109judgement never exhibits strict and absolute, but only assumed and
1110comparative universality (by induction); therefore, the most we can say
1111is--so far as we have hitherto observed, there is no exception to this
1112or that rule. If, on the other hand, a judgement carries with it strict
1113and absolute universality, that is, admits of no possible exception, it
1114is not derived from experience, but is valid absolutely a priori.
1115
1116Empirical universality is, therefore, only an arbitrary extension of
1117validity, from that which may be predicated of a proposition valid in
1118most cases, to that which is asserted of a proposition which holds good
1119in all; as, for example, in the affirmation, "All bodies are heavy."
1120When, on the contrary, strict universality characterizes a judgement,
1121it necessarily indicates another peculiar source of knowledge, namely,
1122a faculty of cognition a priori. Necessity and strict universality,
1123therefore, are infallible tests for distinguishing pure from empirical
1124knowledge, and are inseparably connected with each other. But as in the
1125use of these criteria the empirical limitation is sometimes more
1126easily detected than the contingency of the judgement, or the unlimited
1127universality which we attach to a judgement is often a more convincing
1128proof than its necessity, it may be advisable to use the criteria
1129separately, each being by itself infallible.
1130
1131Now, that in the sphere of human cognition we have judgements which are
1132necessary, and in the strictest sense universal, consequently pure a
1133priori, it will be an easy matter to show. If we desire an example from
1134the sciences, we need only take any proposition in mathematics. If we
1135cast our eyes upon the commonest operations of the understanding, the
1136proposition, "Every change must have a cause," will amply serve our
1137purpose. In the latter case, indeed, the conception of a cause so
1138plainly involves the conception of a necessity of connection with an
1139effect, and of a strict universality of the law, that the very notion of
1140a cause would entirely disappear, were we to derive it, like Hume, from
1141a frequent association of what happens with that which precedes; and the
1142habit thence originating of connecting representations--the necessity
1143inherent in the judgement being therefore merely subjective. Besides,
1144without seeking for such examples of principles existing a priori
1145in cognition, we might easily show that such principles are the
1146indispensable basis of the possibility of experience itself, and
1147consequently prove their existence a priori. For whence could our
1148experience itself acquire certainty, if all the rules on which it
1149depends were themselves empirical, and consequently fortuitous? No one,
1150therefore, can admit the validity of the use of such rules as first
1151principles. But, for the present, we may content ourselves with having
1152established the fact, that we do possess and exercise a faculty of pure
1153a priori cognition; and, secondly, with having pointed out the proper
1154tests of such cognition, namely, universality and necessity.
1155
1156Not only in judgements, however, but even in conceptions, is an a
1157priori origin manifest. For example, if we take away by degrees from
1158our conceptions of a body all that can be referred to mere
1159sensuous experience--colour, hardness or softness, weight, even
1160impenetrability--the body will then vanish; but the space which it
1161occupied still remains, and this it is utterly impossible to annihilate
1162in thought. Again, if we take away, in like manner, from our empirical
1163conception of any object, corporeal or incorporeal, all properties which
1164mere experience has taught us to connect with it, still we cannot think
1165away those through which we cogitate it as substance, or adhering to
1166substance, although our conception of substance is more determined than
1167that of an object. Compelled, therefore, by that necessity with which
1168the conception of substance forces itself upon us, we must confess that
1169it has its seat in our faculty of cognition a priori.
1170
1171
1172
1173III. Philosophy stands in need of a Science which shall Determine the
1174Possibility, Principles, and Extent of Human Knowledge "a priori"
1175
1176Of far more importance than all that has been above said, is the
1177consideration that certain of our cognitions rise completely above the
1178sphere of all possible experience, and by means of conceptions, to which
1179there exists in the whole extent of experience no corresponding object,
1180seem to extend the range of our judgements beyond its bounds. And just
1181in this transcendental or supersensible sphere, where experience affords
1182us neither instruction nor guidance, lie the investigations of reason,
1183which, on account of their importance, we consider far preferable to,
1184and as having a far more elevated aim than, all that the understanding
1185can achieve within the sphere of sensuous phenomena. So high a value
1186do we set upon these investigations, that even at the risk of error, we
1187persist in following them out, and permit neither doubt nor disregard
1188nor indifference to restrain us from the pursuit. These unavoidable
1189problems of mere pure reason are God, freedom (of will), and
1190immortality. The science which, with all its preliminaries, has for its
1191especial object the solution of these problems is named metaphysics--a
1192science which is at the very outset dogmatical, that is, it confidently
1193takes upon itself the execution of this task without any previous
1194investigation of the ability or inability of reason for such an
1195undertaking.
1196
1197Now the safe ground of experience being thus abandoned, it seems
1198nevertheless natural that we should hesitate to erect a building with
1199the cognitions we possess, without knowing whence they come, and on the
1200strength of principles, the origin of which is undiscovered. Instead of
1201thus trying to build without a foundation, it is rather to be expected
1202that we should long ago have put the question, how the understanding can
1203arrive at these a priori cognitions, and what is the extent, validity,
1204and worth which they may possess? We say, "This is natural enough,"
1205meaning by the word natural, that which is consistent with a just and
1206reasonable way of thinking; but if we understand by the term, that
1207which usually happens, nothing indeed could be more natural and
1208more comprehensible than that this investigation should be left
1209long unattempted. For one part of our pure knowledge, the science of
1210mathematics, has been long firmly established, and thus leads us to form
1211flattering expectations with regard to others, though these may be of
1212quite a different nature. Besides, when we get beyond the bounds of
1213experience, we are of course safe from opposition in that quarter;
1214and the charm of widening the range of our knowledge is so great that,
1215unless we are brought to a standstill by some evident contradiction, we
1216hurry on undoubtingly in our course. This, however, may be avoided, if
1217we are sufficiently cautious in the construction of our fictions, which
1218are not the less fictions on that account.
1219
1220Mathematical science affords us a brilliant example, how far,
1221independently of all experience, we may carry our a priori knowledge.
1222It is true that the mathematician occupies himself with objects and
1223cognitions only in so far as they can be represented by means of
1224intuition. But this circumstance is easily overlooked, because the said
1225intuition can itself be given a priori, and therefore is hardly to be
1226distinguished from a mere pure conception. Deceived by such a proof of
1227the power of reason, we can perceive no limits to the extension of our
1228knowledge. The light dove cleaving in free flight the thin air, whose
1229resistance it feels, might imagine that her movements would be far
1230more free and rapid in airless space. Just in the same way did Plato,
1231abandoning the world of sense because of the narrow limits it sets to
1232the understanding, venture upon the wings of ideas beyond it, into the
1233void space of pure intellect. He did not reflect that he made no real
1234progress by all his efforts; for he met with no resistance which might
1235serve him for a support, as it were, whereon to rest, and on which he
1236might apply his powers, in order to let the intellect acquire momentum
1237for its progress. It is, indeed, the common fate of human reason in
1238speculation, to finish the imposing edifice of thought as rapidly as
1239possible, and then for the first time to begin to examine whether the
1240foundation is a solid one or no. Arrived at this point, all sorts
1241of excuses are sought after, in order to console us for its want of
1242stability, or rather, indeed, to enable Us to dispense altogether with
1243so late and dangerous an investigation. But what frees us during the
1244process of building from all apprehension or suspicion, and flatters
1245us into the belief of its solidity, is this. A great part, perhaps the
1246greatest part, of the business of our reason consists in the analysation
1247of the conceptions which we already possess of objects. By this means we
1248gain a multitude of cognitions, which although really nothing more than
1249elucidations or explanations of that which (though in a confused manner)
1250was already thought in our conceptions, are, at least in respect of
1251their form, prized as new introspections; whilst, so far as regards
1252their matter or content, we have really made no addition to our
1253conceptions, but only disinvolved them. But as this process does furnish
1254a real priori knowledge, which has a sure progress and useful results,
1255reason, deceived by this, slips in, without being itself aware of it,
1256assertions of a quite different kind; in which, to given conceptions it
1257adds others, a priori indeed, but entirely foreign to them, without our
1258knowing how it arrives at these, and, indeed, without such a question
1259ever suggesting itself. I shall therefore at once proceed to examine the
1260difference between these two modes of knowledge.
1261
1262
1263
1264IV. Of the Difference Between Analytical and Synthetical Judgements.
1265
1266In all judgements wherein the relation of a subject to the predicate is
1267cogitated (I mention affirmative judgements only here; the application
1268to negative will be very easy), this relation is possible in two
1269different ways. Either the predicate B belongs to the subject A, as
1270somewhat which is contained (though covertly) in the conception A; or
1271the predicate B lies completely out of the conception A, although
1272it stands in connection with it. In the first instance, I term the
1273judgement analytical, in the second, synthetical. Analytical judgements
1274(affirmative) are therefore those in which the connection of the
1275predicate with the subject is cogitated through identity; those in which
1276this connection is cogitated without identity, are called synthetical
1277judgements. The former may be called explicative, the latter
1278augmentative judgements; because the former add in the predicate
1279nothing to the conception of the subject, but only analyse it into its
1280constituent conceptions, which were thought already in the subject,
1281although in a confused manner; the latter add to our conceptions of the
1282subject a predicate which was not contained in it, and which no analysis
1283could ever have discovered therein. For example, when I say, "All bodies
1284are extended," this is an analytical judgement. For I need not go beyond
1285the conception of body in order to find extension connected with it, but
1286merely analyse the conception, that is, become conscious of the manifold
1287properties which I think in that conception, in order to discover this
1288predicate in it: it is therefore an analytical judgement. On the other
1289hand, when I say, "All bodies are heavy," the predicate is something
1290totally different from that which I think in the mere conception of
1291a body. By the addition of such a predicate, therefore, it becomes a
1292synthetical judgement.
1293
1294Judgements of experience, as such, are always synthetical. For it would
1295be absurd to think of grounding an analytical judgement on experience,
1296because in forming such a judgement I need not go out of the sphere of
1297my conceptions, and therefore recourse to the testimony of experience
1298is quite unnecessary. That "bodies are extended" is not an empirical
1299judgement, but a proposition which stands firm a priori. For before
1300addressing myself to experience, I already have in my conception all the
1301requisite conditions for the judgement, and I have only to extract
1302the predicate from the conception, according to the principle of
1303contradiction, and thereby at the same time become conscious of the
1304necessity of the judgement, a necessity which I could never learn from
1305experience. On the other hand, though at first I do not at all include
1306the predicate of weight in my conception of body in general, that
1307conception still indicates an object of experience, a part of the
1308totality of experience, to which I can still add other parts; and this I
1309do when I recognize by observation that bodies are heavy. I can
1310cognize beforehand by analysis the conception of body through the
1311characteristics of extension, impenetrability, shape, etc., all which
1312are cogitated in this conception. But now I extend my knowledge, and
1313looking back on experience from which I had derived this conception
1314of body, I find weight at all times connected with the above
1315characteristics, and therefore I synthetically add to my conceptions
1316this as a predicate, and say, "All bodies are heavy." Thus it is
1317experience upon which rests the possibility of the synthesis of
1318the predicate of weight with the conception of body, because both
1319conceptions, although the one is not contained in the other, still
1320belong to one another (only contingently, however), as parts of a whole,
1321namely, of experience, which is itself a synthesis of intuitions.
1322
1323But to synthetical judgements a priori, such aid is entirely wanting. If
1324I go out of and beyond the conception A, in order to recognize another
1325B as connected with it, what foundation have I to rest on, whereby to
1326render the synthesis possible? I have here no longer the advantage of
1327looking out in the sphere of experience for what I want. Let us take,
1328for example, the proposition, "Everything that happens has a cause." In
1329the conception of "something that happens," I indeed think an existence
1330which a certain time antecedes, and from this I can derive analytical
1331judgements. But the conception of a cause lies quite out of the above
1332conception, and indicates something entirely different from "that which
1333happens," and is consequently not contained in that conception. How
1334then am I able to assert concerning the general conception--"that which
1335happens"--something entirely different from that conception, and to
1336recognize the conception of cause although not contained in it, yet as
1337belonging to it, and even necessarily? what is here the unknown = X,
1338upon which the understanding rests when it believes it has found, out of
1339the conception A a foreign predicate B, which it nevertheless considers
1340to be connected with it? It cannot be experience, because the principle
1341adduced annexes the two representations, cause and effect, to the
1342representation existence, not only with universality, which experience
1343cannot give, but also with the expression of necessity, therefore
1344completely a priori and from pure conceptions. Upon such synthetical,
1345that is augmentative propositions, depends the whole aim of our
1346speculative knowledge a priori; for although analytical judgements are
1347indeed highly important and necessary, they are so, only to arrive at
1348that clearness of conceptions which is requisite for a sure and extended
1349synthesis, and this alone is a real acquisition.
1350
1351
1352
1353V. In all Theoretical Sciences of Reason, Synthetical Judgements "a
1354priori" are contained as Principles.
1355
13561. Mathematical judgements are always synthetical. Hitherto this fact,
1357though incontestably true and very important in its consequences, seems
1358to have escaped the analysts of the human mind, nay, to be in
1359complete opposition to all their conjectures. For as it was found that
1360mathematical conclusions all proceed according to the principle of
1361contradiction (which the nature of every apodeictic certainty requires),
1362people became persuaded that the fundamental principles of the science
1363also were recognized and admitted in the same way. But the notion is
1364fallacious; for although a synthetical proposition can certainly be
1365discerned by means of the principle of contradiction, this is possible
1366only when another synthetical proposition precedes, from which the
1367latter is deduced, but never of itself.
1368
1369Before all, be it observed, that proper mathematical propositions are
1370always judgements a priori, and not empirical, because they carry
1371along with them the conception of necessity, which cannot be given by
1372experience. If this be demurred to, it matters not; I will then limit my
1373assertion to pure mathematics, the very conception of which implies that
1374it consists of knowledge altogether non-empirical and a priori.
1375
1376We might, indeed at first suppose that the proposition 7 + 5 = 12 is a
1377merely analytical proposition, following (according to the principle of
1378contradiction) from the conception of a sum of seven and five. But if we
1379regard it more narrowly, we find that our conception of the sum of seven
1380and five contains nothing more than the uniting of both sums into one,
1381whereby it cannot at all be cogitated what this single number is which
1382embraces both. The conception of twelve is by no means obtained by
1383merely cogitating the union of seven and five; and we may analyse our
1384conception of such a possible sum as long as we will, still we shall
1385never discover in it the notion of twelve. We must go beyond these
1386conceptions, and have recourse to an intuition which corresponds to
1387one of the two--our five fingers, for example, or like Segner in his
1388Arithmetic five points, and so by degrees, add the units contained in
1389the five given in the intuition, to the conception of seven. For I first
1390take the number 7, and, for the conception of 5 calling in the aid of
1391the fingers of my hand as objects of intuition, I add the units, which I
1392before took together to make up the number 5, gradually now by means of
1393the material image my hand, to the number 7, and by this process, I
1394at length see the number 12 arise. That 7 should be added to 5, I have
1395certainly cogitated in my conception of a sum = 7 + 5, but not that
1396this sum was equal to 12. Arithmetical propositions are therefore always
1397synthetical, of which we may become more clearly convinced by trying
1398large numbers. For it will thus become quite evident that, turn and
1399twist our conceptions as we may, it is impossible, without having
1400recourse to intuition, to arrive at the sum total or product by means of
1401the mere analysis of our conceptions. Just as little is any principle
1402of pure geometry analytical. "A straight line between two points is the
1403shortest," is a synthetical proposition. For my conception of straight
1404contains no notion of quantity, but is merely qualitative. The
1405conception of the shortest is therefore fore wholly an addition, and by
1406no analysis can it be extracted from our conception of a straight line.
1407Intuition must therefore here lend its aid, by means of which, and thus
1408only, our synthesis is possible.
1409
1410Some few principles preposited by geometricians are, indeed, really
1411analytical, and depend on the principle of contradiction. They serve,
1412however, like identical propositions, as links in the chain of method,
1413not as principles--for example, a = a, the whole is equal to itself,
1414or (a+b) --> a, the whole is greater than its part. And yet even these
1415principles themselves, though they derive their validity from pure
1416conceptions, are only admitted in mathematics because they can be
1417presented in intuition. What causes us here commonly to believe that
1418the predicate of such apodeictic judgements is already contained in our
1419conception, and that the judgement is therefore analytical, is merely
1420the equivocal nature of the expression. We must join in thought a
1421certain predicate to a given conception, and this necessity cleaves
1422already to the conception. But the question is, not what we must join
1423in thought to the given conception, but what we really think therein,
1424though only obscurely, and then it becomes manifest that the predicate
1425pertains to these conceptions, necessarily indeed, yet not as thought
1426in the conception itself, but by virtue of an intuition, which must be
1427added to the conception.
1428
14292. The science of natural philosophy (physics) contains in itself
1430synthetical judgements a priori, as principles. I shall adduce two
1431propositions. For instance, the proposition, "In all changes of the
1432material world, the quantity of matter remains unchanged"; or, that, "In
1433all communication of motion, action and reaction must always be equal."
1434In both of these, not only is the necessity, and therefore their origin
1435a priori clear, but also that they are synthetical propositions. For in
1436the conception of matter, I do not cogitate its permanency, but merely
1437its presence in space, which it fills. I therefore really go out of and
1438beyond the conception of matter, in order to think on to it something
1439a priori, which I did not think in it. The proposition is therefore not
1440analytical, but synthetical, and nevertheless conceived a priori; and so
1441it is with regard to the other propositions of the pure part of natural
1442philosophy.
1443
14443. As to metaphysics, even if we look upon it merely as an attempted
1445science, yet, from the nature of human reason, an indispensable one, we
1446find that it must contain synthetical propositions a priori. It is not
1447merely the duty of metaphysics to dissect, and thereby analytically to
1448illustrate the conceptions which we form a priori of things; but we seek
1449to widen the range of our a priori knowledge. For this purpose, we must
1450avail ourselves of such principles as add something to the original
1451conception--something not identical with, nor contained in it, and by
1452means of synthetical judgements a priori, leave far behind us the limits
1453of experience; for example, in the proposition, "the world must have a
1454beginning," and such like. Thus metaphysics, according to the proper aim
1455of the science, consists merely of synthetical propositions a priori.
1456
1457
1458
1459VI. The Universal Problem of Pure Reason.
1460
1461It is extremely advantageous to be able to bring a number of
1462investigations under the formula of a single problem. For in this
1463manner, we not only facilitate our own labour, inasmuch as we define it
1464clearly to ourselves, but also render it more easy for others to decide
1465whether we have done justice to our undertaking. The proper problem of
1466pure reason, then, is contained in the question: "How are synthetical
1467judgements a priori possible?"
1468
1469That metaphysical science has hitherto remained in so vacillating a
1470state of uncertainty and contradiction, is only to be attributed to the
1471fact that this great problem, and perhaps even the difference between
1472analytical and synthetical judgements, did not sooner suggest itself
1473to philosophers. Upon the solution of this problem, or upon sufficient
1474proof of the impossibility of synthetical knowledge a priori, depends
1475the existence or downfall of the science of metaphysics. Among
1476philosophers, David Hume came the nearest of all to this problem; yet it
1477never acquired in his mind sufficient precision, nor did he regard the
1478question in its universality. On the contrary, he stopped short at the
1479synthetical proposition of the connection of an effect with its cause
1480(principium causalitatis), insisting that such proposition a priori
1481was impossible. According to his conclusions, then, all that we term
1482metaphysical science is a mere delusion, arising from the fancied
1483insight of reason into that which is in truth borrowed from experience,
1484and to which habit has given the appearance of necessity. Against
1485this assertion, destructive to all pure philosophy, he would have been
1486guarded, had he had our problem before his eyes in its universality. For
1487he would then have perceived that, according to his own argument, there
1488likewise could not be any pure mathematical science, which assuredly
1489cannot exist without synthetical propositions a priori--an absurdity
1490from which his good understanding must have saved him.
1491
1492In the solution of the above problem is at the same time comprehended
1493the possibility of the use of pure reason in the foundation and
1494construction of all sciences which contain theoretical knowledge
1495a priori of objects, that is to say, the answer to the following
1496questions:
1497
1498How is pure mathematical science possible?
1499
1500How is pure natural science possible?
1501
1502Respecting these sciences, as they do certainly exist, it may with
1503propriety be asked, how they are possible?--for that they must be
1504possible is shown by the fact of their really existing.* But as to
1505metaphysics, the miserable progress it has hitherto made, and the fact
1506that of no one system yet brought forward, far as regards its true
1507aim, can it be said that this science really exists, leaves any one at
1508liberty to doubt with reason the very possibility of its existence.
1509
1510 [*Footnote: As to the existence of pure natural science, or
1511 physics, perhaps many may still express doubts. But we have
1512 only to look at the different propositions which are
1513 commonly treated of at the commencement of proper
1514 (empirical) physical science--those, for example, relating
1515 to the permanence of the same quantity of matter, the vis
1516 inertiae, the equality of action and reaction, etc.--to be
1517 soon convinced that they form a science of pure physics
1518 (physica pura, or rationalis), which well deserves to be
1519 separately exposed as a special science, in its whole
1520 extent, whether that be great or confined.]
1521
1522
1523Yet, in a certain sense, this kind of knowledge must unquestionably be
1524looked upon as given; in other words, metaphysics must be considered
1525as really existing, if not as a science, nevertheless as a natural
1526disposition of the human mind (metaphysica naturalis). For human
1527reason, without any instigations imputable to the mere vanity of great
1528knowledge, unceasingly progresses, urged on by its own feeling of
1529need, towards such questions as cannot be answered by any empirical
1530application of reason, or principles derived therefrom; and so there
1531has ever really existed in every man some system of metaphysics. It will
1532always exist, so soon as reason awakes to the exercise of its power
1533of speculation. And now the question arises: "How is metaphysics, as a
1534natural disposition, possible?" In other words, how, from the nature
1535of universal human reason, do those questions arise which pure reason
1536proposes to itself, and which it is impelled by its own feeling of need
1537to answer as well as it can?
1538
1539But as in all the attempts hitherto made to answer the questions which
1540reason is prompted by its very nature to propose to itself, for example,
1541whether the world had a beginning, or has existed from eternity, it has
1542always met with unavoidable contradictions, we must not rest satisfied
1543with the mere natural disposition of the mind to metaphysics, that is,
1544with the existence of the faculty of pure reason, whence, indeed, some
1545sort of metaphysical system always arises; but it must be possible to
1546arrive at certainty in regard to the question whether we know or do not
1547know the things of which metaphysics treats. We must be able to arrive
1548at a decision on the subjects of its questions, or on the ability or
1549inability of reason to form any judgement respecting them; and therefore
1550either to extend with confidence the bounds of our pure reason, or to
1551set strictly defined and safe limits to its action. This last question,
1552which arises out of the above universal problem, would properly run
1553thus: "How is metaphysics possible as a science?"
1554
1555Thus, the critique of reason leads at last, naturally and necessarily,
1556to science; and, on the other hand, the dogmatical use of reason without
1557criticism leads to groundless assertions, against which others equally
1558specious can always be set, thus ending unavoidably in scepticism.
1559
1560Besides, this science cannot be of great and formidable prolixity,
1561because it has not to do with objects of reason, the variety of which is
1562inexhaustible, but merely with Reason herself and her problems; problems
1563which arise out of her own bosom, and are not proposed to her by the
1564nature of outward things, but by her own nature. And when once Reason
1565has previously become able completely to understand her own power in
1566regard to objects which she meets with in experience, it will be easy to
1567determine securely the extent and limits of her attempted application to
1568objects beyond the confines of experience.
1569
1570We may and must, therefore, regard the attempts hitherto made to
1571establish metaphysical science dogmatically as non-existent. For what of
1572analysis, that is, mere dissection of conceptions, is contained in one
1573or other, is not the aim of, but only a preparation for metaphysics
1574proper, which has for its object the extension, by means of synthesis,
1575of our a priori knowledge. And for this purpose, mere analysis is
1576of course useless, because it only shows what is contained in these
1577conceptions, but not how we arrive, a priori, at them; and this it is
1578her duty to show, in order to be able afterwards to determine their
1579valid use in regard to all objects of experience, to all knowledge in
1580general. But little self-denial, indeed, is needed to give up these
1581pretensions, seeing the undeniable, and in the dogmatic mode of
1582procedure, inevitable contradictions of Reason with herself, have long
1583since ruined the reputation of every system of metaphysics that has
1584appeared up to this time. It will require more firmness to remain
1585undeterred by difficulty from within, and opposition from without, from
1586endeavouring, by a method quite opposed to all those hitherto followed,
1587to further the growth and fruitfulness of a science indispensable to
1588human reason--a science from which every branch it has borne may be cut
1589away, but whose roots remain indestructible.
1590
1591
1592
1593VII. Idea and Division of a Particular Science, under the Name of a
1594Critique of Pure Reason.
1595
1596From all that has been said, there results the idea of a particular
1597science, which may be called the Critique of Pure Reason. For reason
1598is the faculty which furnishes us with the principles of knowledge a
1599priori. Hence, pure reason is the faculty which contains the principles
1600of cognizing anything absolutely a priori. An organon of pure reason
1601would be a compendium of those principles according to which alone
1602all pure cognitions a priori can be obtained. The completely extended
1603application of such an organon would afford us a system of pure reason.
1604As this, however, is demanding a great deal, and it is yet doubtful
1605whether any extension of our knowledge be here possible, or, if so,
1606in what cases; we can regard a science of the mere criticism of pure
1607reason, its sources and limits, as the propaedeutic to a system of
1608pure reason. Such a science must not be called a doctrine, but only a
1609critique of pure reason; and its use, in regard to speculation, would be
1610only negative, not to enlarge the bounds of, but to purify, our reason,
1611and to shield it against error--which alone is no little gain. I apply
1612the term transcendental to all knowledge which is not so much occupied
1613with objects as with the mode of our cognition of these objects, so
1614far as this mode of cognition is possible a priori. A system of such
1615conceptions would be called transcendental philosophy. But this, again,
1616is still beyond the bounds of our present essay. For as such a science
1617must contain a complete exposition not only of our synthetical a priori,
1618but of our analytical a priori knowledge, it is of too wide a range for
1619our present purpose, because we do not require to carry our analysis
1620any farther than is necessary to understand, in their full extent, the
1621principles of synthesis a priori, with which alone we have to do. This
1622investigation, which we cannot properly call a doctrine, but only a
1623transcendental critique, because it aims not at the enlargement, but
1624at the correction and guidance, of our knowledge, and is to serve as a
1625touchstone of the worth or worthlessness of all knowledge a priori, is
1626the sole object of our present essay. Such a critique is consequently,
1627as far as possible, a preparation for an organon; and if this new
1628organon should be found to fail, at least for a canon of pure reason,
1629according to which the complete system of the philosophy of pure reason,
1630whether it extend or limit the bounds of that reason, might one day
1631be set forth both analytically and synthetically. For that this is
1632possible, nay, that such a system is not of so great extent as to
1633preclude the hope of its ever being completed, is evident. For we have
1634not here to do with the nature of outward objects, which is infinite,
1635but solely with the mind, which judges of the nature of objects, and,
1636again, with the mind only in respect of its cognition a priori. And the
1637object of our investigations, as it is not to be sought without, but,
1638altogether within, ourselves, cannot remain concealed, and in all
1639probability is limited enough to be completely surveyed and fairly
1640estimated, according to its worth or worthlessness. Still less let the
1641reader here expect a critique of books and systems of pure reason; our
1642present object is exclusively a critique of the faculty of pure reason
1643itself. Only when we make this critique our foundation, do we possess
1644a pure touchstone for estimating the philosophical value of ancient
1645and modern writings on this subject; and without this criterion, the
1646incompetent historian or judge decides upon and corrects the groundless
1647assertions of others with his own, which have themselves just as little
1648foundation.
1649
1650Transcendental philosophy is the idea of a science, for which the
1651Critique of Pure Reason must sketch the whole plan architectonically,
1652that is, from principles, with a full guarantee for the validity and
1653stability of all the parts which enter into the building. It is the
1654system of all the principles of pure reason. If this Critique itself
1655does not assume the title of transcendental philosophy, it is only
1656because, to be a complete system, it ought to contain a full analysis of
1657all human knowledge a priori. Our critique must, indeed, lay before us a
1658complete enumeration of all the radical conceptions which constitute the
1659said pure knowledge. But from the complete analysis of these conceptions
1660themselves, as also from a complete investigation of those derived from
1661them, it abstains with reason; partly because it would be deviating from
1662the end in view to occupy itself with this analysis, since this process
1663is not attended with the difficulty and insecurity to be found in the
1664synthesis, to which our critique is entirely devoted, and partly because
1665it would be inconsistent with the unity of our plan to burden this
1666essay with the vindication of the completeness of such an analysis and
1667deduction, with which, after all, we have at present nothing to do. This
1668completeness of the analysis of these radical conceptions, as well as
1669of the deduction from the conceptions a priori which may be given by the
1670analysis, we can, however, easily attain, provided only that we are
1671in possession of all these radical conceptions, which are to serve as
1672principles of the synthesis, and that in respect of this main purpose
1673nothing is wanting.
1674
1675To the Critique of Pure Reason, therefore, belongs all that constitutes
1676transcendental philosophy; and it is the complete idea of transcendental
1677philosophy, but still not the science itself; because it only proceeds
1678so far with the analysis as is necessary to the power of judging
1679completely of our synthetical knowledge a priori.
1680
1681The principal thing we must attend to, in the division of the parts of
1682a science like this, is that no conceptions must enter it which contain
1683aught empirical; in other words, that the knowledge a priori must be
1684completely pure. Hence, although the highest principles and fundamental
1685conceptions of morality are certainly cognitions a priori, yet they do
1686not belong to transcendental philosophy; because, though they certainly
1687do not lay the conceptions of pain, pleasure, desires, inclinations,
1688etc. (which are all of empirical origin), at the foundation of its
1689precepts, yet still into the conception of duty--as an obstacle to
1690be overcome, or as an incitement which should not be made into a
1691motive--these empirical conceptions must necessarily enter, in the
1692construction of a system of pure morality. Transcendental philosophy is
1693consequently a philosophy of the pure and merely speculative reason.
1694For all that is practical, so far as it contains motives, relates to
1695feelings, and these belong to empirical sources of cognition.
1696
1697If we wish to divide this science from the universal point of view of
1698a science in general, it ought to comprehend, first, a Doctrine of the
1699Elements, and, secondly, a Doctrine of the Method of pure reason. Each
1700of these main divisions will have its subdivisions, the separate reasons
1701for which we cannot here particularize. Only so much seems necessary, by
1702way of introduction of premonition, that there are two sources of human
1703knowledge (which probably spring from a common, but to us unknown root),
1704namely, sense and understanding. By the former, objects are given to
1705us; by the latter, thought. So far as the faculty of sense may contain
1706representations a priori, which form the conditions under which objects
1707are given, in so far it belongs to transcendental philosophy. The
1708transcendental doctrine of sense must form the first part of our science
1709of elements, because the conditions under which alone the objects
1710of human knowledge are given must precede those under which they are
1711thought.
1712
1713
1714
1715I. TRANSCENDENTAL DOCTRINE OF ELEMENTS.
1716
1717
1718
1719FIRST PART. TRANSCENDENTAL AESTHETIC.
1720
1721
1722SS I. Introductory.
1723
1724In whatsoever mode, or by whatsoever means, our knowledge may relate
1725to objects, it is at least quite clear that the only manner in which it
1726immediately relates to them is by means of an intuition. To this as the
1727indispensable groundwork, all thought points. But an intuition can take
1728place only in so far as the object is given to us. This, again, is only
1729possible, to man at least, on condition that the object affect the
1730mind in a certain manner. The capacity for receiving representations
1731(receptivity) through the mode in which we are affected by objects,
1732objects, is called sensibility. By means of sensibility, therefore,
1733objects are given to us, and it alone furnishes us with intuitions; by
1734the understanding they are thought, and from it arise conceptions. But
1735an thought must directly, or indirectly, by means of certain signs,
1736relate ultimately to intuitions; consequently, with us, to sensibility,
1737because in no other way can an object be given to us.
1738
1739The effect of an object upon the faculty of representation, so far as
1740we are affected by the said object, is sensation. That sort of intuition
1741which relates to an object by means of sensation is called an empirical
1742intuition. The undetermined object of an empirical intuition is called
1743phenomenon. That which in the phenomenon corresponds to the sensation,
1744I term its matter; but that which effects that the content of the
1745phenomenon can be arranged under certain relations, I call its form. But
1746that in which our sensations are merely arranged, and by which they are
1747susceptible of assuming a certain form, cannot be itself sensation. It
1748is, then, the matter of all phenomena that is given to us a posteriori;
1749the form must lie ready a priori for them in the mind, and consequently
1750can be regarded separately from all sensation.
1751
1752I call all representations pure, in the transcendental meaning of
1753the word, wherein nothing is met with that belongs to sensation. And
1754accordingly we find existing in the mind a priori, the pure form of
1755sensuous intuitions in general, in which all the manifold content of the
1756phenomenal world is arranged and viewed under certain relations. This
1757pure form of sensibility I shall call pure intuition. Thus, if I take
1758away from our representation of a body all that the understanding thinks
1759as belonging to it, as substance, force, divisibility, etc., and also
1760whatever belongs to sensation, as impenetrability, hardness, colour,
1761etc.; yet there is still something left us from this empirical
1762intuition, namely, extension and shape. These belong to pure intuition,
1763which exists a priori in the mind, as a mere form of sensibility, and
1764without any real object of the senses or any sensation.
1765
1766The science of all the principles of sensibility a priori, I call
1767transcendental aesthetic.* There must, then, be such a science
1768forming the first part of the transcendental doctrine of elements, in
1769contradistinction to that part which contains the principles of pure
1770thought, and which is called transcendental logic.
1771
1772 [Footnote: The Germans are the only people who at present
1773 use this word to indicate what others call the critique of
1774 taste. At the foundation of this term lies the disappointed
1775 hope, which the eminent analyst, Baumgarten, conceived, of
1776 subjecting the criticism of the beautiful to principles of
1777 reason, and so of elevating its rules into a science. But
1778 his endeavours were vain. For the said rules or criteria
1779 are, in respect to their chief sources, merely empirical,
1780 consequently never can serve as determinate laws a priori,
1781 by which our judgement in matters of taste is to be
1782 directed. It is rather our judgement which forms the proper
1783 test as to the correctness of the principles. On this
1784 account it is advisable to give up the use of the term as
1785 designating the critique of taste, and to apply it solely to
1786 that doctrine, which is true science--the science of the
1787 laws of sensibility--and thus come nearer to the language
1788 and the sense of the ancients in their well-known division
1789 of the objects of cognition into aiotheta kai noeta, or to
1790 share it with speculative philosophy, and employ it partly
1791 in a transcendental, partly in a psychological
1792 signification.]
1793
1794
1795In the science of transcendental aesthetic accordingly, we shall first
1796isolate sensibility or the sensuous faculty, by separating from it all
1797that is annexed to its perceptions by the conceptions of understanding,
1798so that nothing be left but empirical intuition. In the next place we
1799shall take away from this intuition all that belongs to sensation,
1800so that nothing may remain but pure intuition, and the mere form of
1801phenomena, which is all that the sensibility can afford a priori. From
1802this investigation it will be found that there are two pure forms of
1803sensuous intuition, as principles of knowledge a priori, namely, space
1804and time. To the consideration of these we shall now proceed.
1805
1806
1807SECTION I. Of Space.
1808
1809
1810SS 2. Metaphysical Exposition of this Conception.
1811
1812By means of the external sense (a property of the mind), we represent
1813to ourselves objects as without us, and these all in space. Herein alone
1814are their shape, dimensions, and relations to each other determined
1815or determinable. The internal sense, by means of which the mind
1816contemplates itself or its internal state, gives, indeed, no intuition
1817of the soul as an object; yet there is nevertheless a determinate form,
1818under which alone the contemplation of our internal state is possible,
1819so that all which relates to the inward determinations of the mind is
1820represented in relations of time. Of time we cannot have any external
1821intuition, any more than we can have an internal intuition of space.
1822What then are time and space? Are they real existences? Or, are they
1823merely relations or determinations of things, such, however, as would
1824equally belong to these things in themselves, though they should never
1825become objects of intuition; or, are they such as belong only to the
1826form of intuition, and consequently to the subjective constitution of
1827the mind, without which these predicates of time and space could not be
1828attached to any object? In order to become informed on these points,
1829we shall first give an exposition of the conception of space. By
1830exposition, I mean the clear, though not detailed, representation of
1831that which belongs to a conception; and an exposition is metaphysical
1832when it contains that which represents the conception as given a priori.
1833
18341. Space is not a conception which has been derived from outward
1835experiences. For, in order that certain sensations may relate to
1836something without me (that is, to something which occupies a different
1837part of space from that in which I am); in like manner, in order that
1838I may represent them not merely as without, of, and near to each other,
1839but also in separate places, the representation of space must already
1840exist as a foundation. Consequently, the representation of space cannot
1841be borrowed from the relations of external phenomena through experience;
1842but, on the contrary, this external experience is itself only possible
1843through the said antecedent representation.
1844
18452. Space then is a necessary representation a priori, which serves for
1846the foundation of all external intuitions. We never can imagine or make
1847a representation to ourselves of the non-existence of space, though
1848we may easily enough think that no objects are found in it. It must,
1849therefore, be considered as the condition of the possibility of
1850phenomena, and by no means as a determination dependent on them, and
1851is a representation a priori, which necessarily supplies the basis for
1852external phenomena.
1853
18543. Space is no discursive, or as we say, general conception of the
1855relations of things, but a pure intuition. For, in the first place, we
1856can only represent to ourselves one space, and, when we talk of divers
1857spaces, we mean only parts of one and the same space. Moreover, these
1858parts cannot antecede this one all-embracing space, as the component
1859parts from which the aggregate can be made up, but can be cogitated only
1860as existing in it. Space is essentially one, and multiplicity in it,
1861consequently the general notion of spaces, of this or that space,
1862depends solely upon limitations. Hence it follows that an a priori
1863intuition (which is not empirical) lies at the root of all our
1864conceptions of space. Thus, moreover, the principles of geometry--for
1865example, that "in a triangle, two sides together are greater than the
1866third," are never deduced from general conceptions of line and triangle,
1867but from intuition, and this a priori, with apodeictic certainty.
1868
18694. Space is represented as an infinite given quantity. Now every
1870conception must indeed be considered as a representation which
1871is contained in an infinite multitude of different possible
1872representations, which, therefore, comprises these under itself; but
1873no conception, as such, can be so conceived, as if it contained within
1874itself an infinite multitude of representations. Nevertheless, space
1875is so conceived of, for all parts of space are equally capable of being
1876produced to infinity. Consequently, the original representation of space
1877is an intuition a priori, and not a conception.
1878
1879
1880SS 3. Transcendental Exposition of the Conception of Space.
1881
1882By a transcendental exposition, I mean the explanation of a conception,
1883as a principle, whence can be discerned the possibility of other
1884synthetical a priori cognitions. For this purpose, it is requisite,
1885firstly, that such cognitions do really flow from the given conception;
1886and, secondly, that the said cognitions are only possible under the
1887presupposition of a given mode of explaining this conception.
1888
1889Geometry is a science which determines the properties of space
1890synthetically, and yet a priori. What, then, must be our representation
1891of space, in order that such a cognition of it may be possible? It must
1892be originally intuition, for from a mere conception, no propositions can
1893be deduced which go out beyond the conception, and yet this happens in
1894geometry. (Introd. V.) But this intuition must be found in the mind a
1895priori, that is, before any perception of objects, consequently must be
1896pure, not empirical, intuition. For geometrical principles are always
1897apodeictic, that is, united with the consciousness of their necessity,
1898as: "Space has only three dimensions." But propositions of this kind
1899cannot be empirical judgements, nor conclusions from them. (Introd. II.)
1900Now, how can an external intuition anterior to objects themselves, and
1901in which our conception of objects can be determined a priori, exist
1902in the human mind? Obviously not otherwise than in so far as it has its
1903seat in the subject only, as the formal capacity of the subject's being
1904affected by objects, and thereby of obtaining immediate representation,
1905that is, intuition; consequently, only as the form of the external sense
1906in general.
1907
1908Thus it is only by means of our explanation that the possibility of
1909geometry, as a synthetical science a priori, becomes comprehensible.
1910Every mode of explanation which does not show us this possibility,
1911although in appearance it may be similar to ours, can with the utmost
1912certainty be distinguished from it by these marks.
1913
1914
1915SS 4. Conclusions from the foregoing Conceptions.
1916
1917(a) Space does not represent any property of objects as things in
1918themselves, nor does it represent them in their relations to each other;
1919in other words, space does not represent to us any determination of
1920objects such as attaches to the objects themselves, and would remain,
1921even though all subjective conditions of the intuition were abstracted.
1922For neither absolute nor relative determinations of objects can be
1923intuited prior to the existence of the things to which they belong, and
1924therefore not a priori.
1925
1926(b) Space is nothing else than the form of all phenomena of the external
1927sense, that is, the subjective condition of the sensibility, under which
1928alone external intuition is possible. Now, because the receptivity or
1929capacity of the subject to be affected by objects necessarily antecedes
1930all intuitions of these objects, it is easily understood how the form
1931of all phenomena can be given in the mind previous to all actual
1932perceptions, therefore a priori, and how it, as a pure intuition, in
1933which all objects must be determined, can contain principles of the
1934relations of these objects prior to all experience.
1935
1936It is therefore from the human point of view only that we can speak
1937of space, extended objects, etc. If we depart from the subjective
1938condition, under which alone we can obtain external intuition, or,
1939in other words, by means of which we are affected by objects, the
1940representation of space has no meaning whatsoever. This predicate is
1941only applicable to things in so far as they appear to us, that is, are
1942objects of sensibility. The constant form of this receptivity, which
1943we call sensibility, is a necessary condition of all relations in which
1944objects can be intuited as existing without us, and when abstraction of
1945these objects is made, is a pure intuition, to which we give the name
1946of space. It is clear that we cannot make the special conditions of
1947sensibility into conditions of the possibility of things, but only of
1948the possibility of their existence as far as they are phenomena. And
1949so we may correctly say that space contains all which can appear to us
1950externally, but not all things considered as things in themselves,
1951be they intuited or not, or by whatsoever subject one will. As to the
1952intuitions of other thinking beings, we cannot judge whether they are or
1953are not bound by the same conditions which limit our own intuition,
1954and which for us are universally valid. If we join the limitation of
1955a judgement to the conception of the subject, then the judgement will
1956possess unconditioned validity. For example, the proposition, "All
1957objects are beside each other in space," is valid only under the
1958limitation that these things are taken as objects of our sensuous
1959intuition. But if I join the condition to the conception and say, "All
1960things, as external phenomena, are beside each other in space," then the
1961rule is valid universally, and without any limitation. Our expositions,
1962consequently, teach the reality (i.e., the objective validity) of space
1963in regard of all which can be presented to us externally as object, and
1964at the same time also the ideality of space in regard to objects when
1965they are considered by means of reason as things in themselves, that is,
1966without reference to the constitution of our sensibility. We maintain,
1967therefore, the empirical reality of space in regard to all possible
1968external experience, although we must admit its transcendental ideality;
1969in other words, that it is nothing, so soon as we withdraw the condition
1970upon which the possibility of all experience depends and look upon space
1971as something that belongs to things in themselves.
1972
1973But, with the exception of space, there is no representation, subjective
1974and referring to something external to us, which could be called
1975objective a priori. For there are no other subjective representations
1976from which we can deduce synthetical propositions a priori, as we can
1977from the intuition of space. (See SS 3.) Therefore, to speak accurately,
1978no ideality whatever belongs to these, although they agree in this
1979respect with the representation of space, that they belong merely to the
1980subjective nature of the mode of sensuous perception; such a mode, for
1981example, as that of sight, of hearing, and of feeling, by means of the
1982sensations of colour, sound, and heat, but which, because they are
1983only sensations and not intuitions, do not of themselves give us
1984the cognition of any object, least of all, an a priori cognition. My
1985purpose, in the above remark, is merely this: to guard any one
1986against illustrating the asserted ideality of space by examples quite
1987insufficient, for example, by colour, taste, etc.; for these must be
1988contemplated not as properties of things, but only as changes in the
1989subject, changes which may be different in different men. For, in such
1990a case, that which is originally a mere phenomenon, a rose, for example,
1991is taken by the empirical understanding for a thing in itself, though to
1992every different eye, in respect of its colour, it may appear different.
1993On the contrary, the transcendental conception of phenomena in space is
1994a critical admonition, that, in general, nothing which is intuited in
1995space is a thing in itself, and that space is not a form which belongs
1996as a property to things; but that objects are quite unknown to us in
1997themselves, and what we call outward objects, are nothing else but mere
1998representations of our sensibility, whose form is space, but whose
1999real correlate, the thing in itself, is not known by means of these
2000representations, nor ever can be, but respecting which, in experience,
2001no inquiry is ever made.
2002
2003
2004SECTION II. Of Time.
2005
2006
2007SS 5. Metaphysical Exposition of this Conception.
2008
20091. Time is not an empirical conception. For neither coexistence nor
2010succession would be perceived by us, if the representation of time did
2011not exist as a foundation a priori. Without this presupposition we could
2012not represent to ourselves that things exist together at one and the
2013same time, or at different times, that is, contemporaneously, or in
2014succession.
2015
20162. Time is a necessary representation, lying at the foundation of all
2017our intuitions. With regard to phenomena in general, we cannot think
2018away time from them, and represent them to ourselves as out of and
2019unconnected with time, but we can quite well represent to ourselves time
2020void of phenomena. Time is therefore given a priori. In it alone is all
2021reality of phenomena possible. These may all be annihilated in thought,
2022but time itself, as the universal condition of their possibility, cannot
2023be so annulled.
2024
20253. On this necessity a priori is also founded the possibility of
2026apodeictic principles of the relations of time, or axioms of time in
2027general, such as: "Time has only one dimension," "Different times are
2028not coexistent but successive" (as different spaces are not successive
2029but coexistent). These principles cannot be derived from experience, for
2030it would give neither strict universality, nor apodeictic certainty. We
2031should only be able to say, "so common experience teaches us," but not
2032"it must be so." They are valid as rules, through which, in general,
2033experience is possible; and they instruct us respecting experience, and
2034not by means of it.
2035
20364. Time is not a discursive, or as it is called, general conception, but
2037a pure form of the sensuous intuition. Different times are merely parts
2038of one and the same time. But the representation which can only be
2039given by a single object is an intuition. Besides, the proposition that
2040different times cannot be coexistent could not be derived from a general
2041conception. For this proposition is synthetical, and therefore cannot
2042spring out of conceptions alone. It is therefore contained immediately
2043in the intuition and representation of time.
2044
20455. The infinity of time signifies nothing more than that every
2046determined quantity of time is possible only through limitations of one
2047time lying at the foundation. Consequently, the original representation,
2048time, must be given as unlimited. But as the determinate representation
2049of the parts of time and of every quantity of an object can only be
2050obtained by limitation, the complete representation of time must not
2051be furnished by means of conceptions, for these contain only partial
2052representations. Conceptions, on the contrary, must have immediate
2053intuition for their basis.
2054
2055
2056
2057SS 6 Transcendental Exposition of the Conception of Time.
2058
2059I may here refer to what is said above (SS 5, 3), where, for or sake of
2060brevity, I have placed under the head of metaphysical exposition, that
2061which is properly transcendental. Here I shall add that the conception
2062of change, and with it the conception of motion, as change of place, is
2063possible only through and in the representation of time; that if this
2064representation were not an intuition (internal) a priori, no conception,
2065of whatever kind, could render comprehensible the possibility of change,
2066in other words, of a conjunction of contradictorily opposed predicates
2067in one and the same object, for example, the presence of a thing in a
2068place and the non-presence of the same thing in the same place. It
2069is only in time that it is possible to meet with two contradictorily
2070opposed determinations in one thing, that is, after each other. Thus
2071our conception of time explains the possibility of so much synthetical
2072knowledge a priori, as is exhibited in the general doctrine of motion,
2073which is not a little fruitful.
2074
2075
2076
2077SS 7. Conclusions from the above Conceptions.
2078
2079(a) Time is not something which subsists of itself, or which inheres
2080in things as an objective determination, and therefore remains, when
2081abstraction is made of the subjective conditions of the intuition of
2082things. For in the former case, it would be something real, yet without
2083presenting to any power of perception any real object. In the latter
2084case, as an order or determination inherent in things themselves, it
2085could not be antecedent to things, as their condition, nor discerned or
2086intuited by means of synthetical propositions a priori. But all this is
2087quite possible when we regard time as merely the subjective condition
2088under which all our intuitions take place. For in that case, this form
2089of the inward intuition can be represented prior to the objects, and
2090consequently a priori.
2091
2092(b) Time is nothing else than the form of the internal sense, that is,
2093of the intuitions of self and of our internal state. For time cannot be
2094any determination of outward phenomena. It has to do neither with
2095shape nor position; on the contrary, it determines the relation of
2096representations in our internal state. And precisely because this
2097internal intuition presents to us no shape or form, we endeavour to
2098supply this want by analogies, and represent the course of time by a
2099line progressing to infinity, the content of which constitutes a series
2100which is only of one dimension; and we conclude from the properties of
2101this line as to all the properties of time, with this single exception,
2102that the parts of the line are coexistent, whilst those of time are
2103successive. From this it is clear also that the representation of time
2104is itself an intuition, because all its relations can be expressed in an
2105external intuition.
2106
2107(c) Time is the formal condition a priori of all phenomena whatsoever.
2108Space, as the pure form of external intuition, is limited as a condition
2109a priori to external phenomena alone. On the other hand, because all
2110representations, whether they have or have not external things for their
2111objects, still in themselves, as determinations of the mind, belong to
2112our internal state; and because this internal state is subject to the
2113formal condition of the internal intuition, that is, to time--time is a
2114condition a priori of all phenomena whatsoever--the immediate condition
2115of all internal, and thereby the mediate condition of all external
2116phenomena. If I can say a priori, "All outward phenomena are in space,
2117and determined a priori according to the relations of space," I can
2118also, from the principle of the internal sense, affirm universally, "All
2119phenomena in general, that is, all objects of the senses, are in time
2120and stand necessarily in relations of time."
2121
2122If we abstract our internal intuition of ourselves and all external
2123intuitions, possible only by virtue of this internal intuition and
2124presented to us by our faculty of representation, and consequently take
2125objects as they are in themselves, then time is nothing. It is only
2126of objective validity in regard to phenomena, because these are things
2127which we regard as objects of our senses. It no longer objective we,
2128make abstraction of the sensuousness of our intuition, in other words,
2129of that mode of representation which is peculiar to us, and speak of
2130things in general. Time is therefore merely a subjective condition of
2131our (human) intuition (which is always sensuous, that is, so far as we
2132are affected by objects), and in itself, independently of the mind
2133or subject, is nothing. Nevertheless, in respect of all phenomena,
2134consequently of all things which come within the sphere of our
2135experience, it is necessarily objective. We cannot say, "All things are
2136in time," because in this conception of things in general, we abstract
2137and make no mention of any sort of intuition of things. But this is
2138the proper condition under which time belongs to our representation
2139of objects. If we add the condition to the conception, and say, "All
2140things, as phenomena, that is, objects of sensuous intuition, are
2141in time," then the proposition has its sound objective validity and
2142universality a priori.
2143
2144What we have now set forth teaches, therefore, the empirical reality of
2145time; that is, its objective validity in reference to all objects which
2146can ever be presented to our senses. And as our intuition is always
2147sensuous, no object ever can be presented to us in experience, which
2148does not come under the conditions of time. On the other hand, we deny
2149to time all claim to absolute reality; that is, we deny that it, without
2150having regard to the form of our sensuous intuition, absolutely inheres
2151in things as a condition or property. Such properties as belong to
2152objects as things in themselves never can be presented to us through
2153the medium of the senses. Herein consists, therefore, the transcendental
2154ideality of time, according to which, if we abstract the subjective
2155conditions of sensuous intuition, it is nothing, and cannot be
2156reckoned as subsisting or inhering in objects as things in themselves,
2157independently of its relation to our intuition. This ideality, like that
2158of space, is not to be proved or illustrated by fallacious analogies
2159with sensations, for this reason--that in such arguments or
2160illustrations, we make the presupposition that the phenomenon, in which
2161such and such predicates inhere, has objective reality, while in this
2162case we can only find such an objective reality as is itself empirical,
2163that is, regards the object as a mere phenomenon. In reference to this
2164subject, see the remark in Section I (SS 4)
2165
2166
2167
2168SS 8. Elucidation.
2169
2170Against this theory, which grants empirical reality to time, but denies
2171to it absolute and transcendental reality, I have heard from intelligent
2172men an objection so unanimously urged that I conclude that it must
2173naturally present itself to every reader to whom these considerations
2174are novel. It runs thus: "Changes are real" (this the continual change
2175in our own representations demonstrates, even though the existence of
2176all external phenomena, together with their changes, is denied). Now,
2177changes are only possible in time, and therefore time must be something
2178real. But there is no difficulty in answering this. I grant the whole
2179argument. Time, no doubt, is something real, that is, it is the real
2180form of our internal intuition. It therefore has subjective reality,
2181in reference to our internal experience, that is, I have really
2182the representation of time and of my determinations therein. Time,
2183therefore, is not to be regarded as an object, but as the mode of
2184representation of myself as an object. But if I could intuite myself,
2185or be intuited by another being, without this condition of sensibility,
2186then those very determinations which we now represent to ourselves as
2187changes, would present to us a knowledge in which the representation
2188of time, and consequently of change, would not appear. The empirical
2189reality of time, therefore, remains, as the condition of all our
2190experience. But absolute reality, according to what has been said above,
2191cannot be granted it. Time is nothing but the form of our internal
2192intuition.* If we take away from it the special condition of our
2193sensibility, the conception of time also vanishes; and it inheres not
2194in the objects themselves, but solely in the subject (or mind) which
2195intuites them.
2196
2197 [*Footnote: I can indeed say "my representations follow one
2198 another, or are successive"; but this means only that we are
2199 conscious of them as in a succession, that is, according to
2200 the form of the internal sense. Time, therefore, is not a
2201 thing in itself, nor is it any objective determination
2202 pertaining to, or inherent in things.]
2203
2204
2205But the reason why this objection is so unanimously brought against
2206our doctrine of time, and that too by disputants who cannot start any
2207intelligible arguments against the doctrine of the ideality of space,
2208is this--they have no hope of demonstrating apodeictically the absolute
2209reality of space, because the doctrine of idealism is against them,
2210according to which the reality of external objects is not capable of
2211any strict proof. On the other hand, the reality of the object of
2212our internal sense (that is, myself and my internal state) is clear
2213immediately through consciousness. The former--external objects in
2214space--might be a mere delusion, but the latter--the object of my
2215internal perception--is undeniably real. They do not, however, reflect
2216that both, without question of their reality as representations, belong
2217only to the genus phenomenon, which has always two aspects, the one, the
2218object considered as a thing in itself, without regard to the mode
2219of intuiting it, and the nature of which remains for this very reason
2220problematical, the other, the form of our intuition of the object,
2221which must be sought not in the object as a thing in itself, but in
2222the subject to which it appears--which form of intuition nevertheless
2223belongs really and necessarily to the phenomenal object.
2224
2225Time and space are, therefore, two sources of knowledge, from which, a
2226priori, various synthetical cognitions can be drawn. Of this we find
2227a striking example in the cognitions of space and its relations, which
2228form the foundation of pure mathematics. They are the two pure forms
2229of all intuitions, and thereby make synthetical propositions a priori
2230possible. But these sources of knowledge being merely conditions of our
2231sensibility, do therefore, and as such, strictly determine their own
2232range and purpose, in that they do not and cannot present objects as
2233things in themselves, but are applicable to them solely in so far as
2234they are considered as sensuous phenomena. The sphere of phenomena is
2235the only sphere of their validity, and if we venture out of this, no
2236further objective use can be made of them. For the rest, this formal
2237reality of time and space leaves the validity of our empirical knowledge
2238unshaken; for our certainty in that respect is equally firm, whether
2239these forms necessarily inhere in the things themselves, or only in our
2240intuitions of them. On the other hand, those who maintain the absolute
2241reality of time and space, whether as essentially subsisting, or only
2242inhering, as modifications, in things, must find themselves at utter
2243variance with the principles of experience itself. For, if they decide
2244for the first view, and make space and time into substances, this being
2245the side taken by mathematical natural philosophers, they must admit
2246two self-subsisting nonentities, infinite and eternal, which exist (yet
2247without there being anything real) for the purpose of containing in
2248themselves everything that is real. If they adopt the second view of
2249inherence, which is preferred by some metaphysical natural philosophers,
2250and regard space and time as relations (contiguity in space or
2251succession in time), abstracted from experience, though represented
2252confusedly in this state of separation, they find themselves in that
2253case necessitated to deny the validity of mathematical doctrines a
2254priori in reference to real things (for example, in space)--at all
2255events their apodeictic certainty. For such certainty cannot be found in
2256an a posteriori proposition; and the conceptions a priori of space and
2257time are, according to this opinion, mere creations of the imagination,
2258having their source really in experience, inasmuch as, out of relations
2259abstracted from experience, imagination has made up something which
2260contains, indeed, general statements of these relations, yet of which
2261no application can be made without the restrictions attached thereto by
2262nature. The former of these parties gains this advantage, that they
2263keep the sphere of phenomena free for mathematical science. On the other
2264hand, these very conditions (space and time) embarrass them greatly,
2265when the understanding endeavours to pass the limits of that sphere. The
2266latter has, indeed, this advantage, that the representations of space
2267and time do not come in their way when they wish to judge of objects,
2268not as phenomena, but merely in their relation to the understanding.
2269Devoid, however, of a true and objectively valid a priori intuition,
2270they can neither furnish any basis for the possibility of mathematical
2271cognitions a priori, nor bring the propositions of experience into
2272necessary accordance with those of mathematics. In our theory of
2273the true nature of these two original forms of the sensibility, both
2274difficulties are surmounted.
2275
2276In conclusion, that transcendental aesthetic cannot contain any more
2277than these two elements--space and time, is sufficiently obvious from
2278the fact that all other conceptions appertaining to sensibility,
2279even that of motion, which unites in itself both elements, presuppose
2280something empirical. Motion, for example, presupposes the perception
2281of something movable. But space considered in itself contains nothing
2282movable, consequently motion must be something which is found in space
2283only through experience--in other words, an empirical datum. In like
2284manner, transcendental aesthetic cannot number the conception of change
2285among its data a priori; for time itself does not change, but only
2286something which is in time. To acquire the conception of change,
2287therefore, the perception of some existing object and of the succession
2288of its determinations, in one word, experience, is necessary.
2289
2290
2291
2292SS 9. General Remarks on Transcendental Aesthetic.
2293
2294I. In order to prevent any misunderstanding, it will be requisite,
2295in the first place, to recapitulate, as clearly as possible, what
2296our opinion is with respect to the fundamental nature of our sensuous
2297cognition in general. We have intended, then, to say that all our
2298intuition is nothing but the representation of phenomena; that
2299the things which we intuite, are not in themselves the same as our
2300representations of them in intuition, nor are their relations in
2301themselves so constituted as they appear to us; and that if we take away
2302the subject, or even only the subjective constitution of our senses in
2303general, then not only the nature and relations of objects in space and
2304time, but even space and time themselves disappear; and that these, as
2305phenomena, cannot exist in themselves, but only in us. What may be
2306the nature of objects considered as things in themselves and without
2307reference to the receptivity of our sensibility is quite unknown to us.
2308We know nothing more than our mode of perceiving them, which is peculiar
2309to us, and which, though not of necessity pertaining to every animated
2310being, is so to the whole human race. With this alone we have to do.
2311Space and time are the pure forms thereof; sensation the matter. The
2312former alone can we cognize a priori, that is, antecedent to all actual
2313perception; and for this reason such cognition is called pure intuition.
2314The latter is that in our cognition which is called cognition a
2315posteriori, that is, empirical intuition. The former appertain
2316absolutely and necessarily to our sensibility, of whatsoever kind our
2317sensations may be; the latter may be of very diversified character.
2318Supposing that we should carry our empirical intuition even to the very
2319highest degree of clearness, we should not thereby advance one step
2320nearer to a knowledge of the constitution of objects as things in
2321themselves. For we could only, at best, arrive at a complete cognition
2322of our own mode of intuition, that is of our sensibility, and this
2323always under the conditions originally attaching to the subject, namely,
2324the conditions of space and time; while the question: "What are objects
2325considered as things in themselves?" remains unanswerable even after the
2326most thorough examination of the phenomenal world.
2327
2328To say, then, that all our sensibility is nothing but the confused
2329representation of things containing exclusively that which belongs
2330to them as things in themselves, and this under an accumulation of
2331characteristic marks and partial representations which we cannot
2332distinguish in consciousness, is a falsification of the conception
2333of sensibility and phenomenization, which renders our whole doctrine
2334thereof empty and useless. The difference between a confused and a clear
2335representation is merely logical and has nothing to do with content.
2336No doubt the conception of right, as employed by a sound understanding,
2337contains all that the most subtle investigation could unfold from
2338it, although, in the ordinary practical use of the word, we are not
2339conscious of the manifold representations comprised in the conception.
2340But we cannot for this reason assert that the ordinary conception is a
2341sensuous one, containing a mere phenomenon, for right cannot appear as
2342a phenomenon; but the conception of it lies in the understanding, and
2343represents a property (the moral property) of actions, which belongs to
2344them in themselves. On the other hand, the representation in intuition
2345of a body contains nothing which could belong to an object considered as
2346a thing in itself, but merely the phenomenon or appearance of something,
2347and the mode in which we are affected by that appearance; and this
2348receptivity of our faculty of cognition is called sensibility, and
2349remains toto caelo different from the cognition of an object in itself,
2350even though we should examine the content of the phenomenon to the very
2351bottom.
2352
2353It must be admitted that the Leibnitz-Wolfian philosophy has assigned an
2354entirely erroneous point of view to all investigations into the nature
2355and origin of our cognitions, inasmuch as it regards the distinction
2356between the sensuous and the intellectual as merely logical, whereas
2357it is plainly transcendental, and concerns not merely the clearness
2358or obscurity, but the content and origin of both. For the faculty of
2359sensibility not only does not present us with an indistinct and confused
2360cognition of objects as things in themselves, but, in fact, gives us no
2361knowledge of these at all. On the contrary, so soon as we abstract in
2362thought our own subjective nature, the object represented, with the
2363properties ascribed to it by sensuous intuition, entirely disappears,
2364because it was only this subjective nature that determined the form of
2365the object as a phenomenon.
2366
2367In phenomena, we commonly, indeed, distinguish that which essentially
2368belongs to the intuition of them, and is valid for the sensuous faculty
2369of every human being, from that which belongs to the same intuition
2370accidentally, as valid not for the sensuous faculty in general, but for
2371a particular state or organization of this or that sense. Accordingly,
2372we are accustomed to say that the former is a cognition which represents
2373the object itself, whilst the latter presents only a particular
2374appearance or phenomenon thereof. This distinction, however, is
2375only empirical. If we stop here (as is usual), and do not regard the
2376empirical intuition as itself a mere phenomenon (as we ought to do), in
2377which nothing that can appertain to a thing in itself is to be found,
2378our transcendental distinction is lost, and we believe that we cognize
2379objects as things in themselves, although in the whole range of the
2380sensuous world, investigate the nature of its objects as profoundly
2381as we may, we have to do with nothing but phenomena. Thus, we call the
2382rainbow a mere appearance of phenomenon in a sunny shower, and the
2383rain, the reality or thing in itself; and this is right enough, if we
2384understand the latter conception in a merely physical sense, that is,
2385as that which in universal experience, and under whatever conditions of
2386sensuous perception, is known in intuition to be so and so determined,
2387and not otherwise. But if we consider this empirical datum generally,
2388and inquire, without reference to its accordance with all our senses,
2389whether there can be discovered in it aught which represents an object
2390as a thing in itself (the raindrops of course are not such, for they
2391are, as phenomena, empirical objects), the question of the relation of
2392the representation to the object is transcendental; and not only are the
2393raindrops mere phenomena, but even their circular form, nay, the space
2394itself through which they fall, is nothing in itself, but both are mere
2395modifications or fundamental dispositions of our sensuous intuition,
2396whilst the transcendental object remains for us utterly unknown.
2397
2398The second important concern of our aesthetic is that it does not obtain
2399favour merely as a plausible hypothesis, but possess as undoubted a
2400character of certainty as can be demanded of any theory which is to
2401serve for an organon. In order fully to convince the reader of this
2402certainty, we shall select a case which will serve to make its validity
2403apparent, and also to illustrate what has been said in SS 3.
2404
2405Suppose, then, that space and time are in themselves objective, and
2406conditions of the--possibility of objects as things in themselves. In
2407the first place, it is evident that both present us, with very
2408many apodeictic and synthetic propositions a priori, but especially
2409space--and for this reason we shall prefer it for investigation at
2410present. As the propositions of geometry are cognized synthetically a
2411priori, and with apodeictic certainty, I inquire: Whence do you obtain
2412propositions of this kind, and on what basis does the understanding
2413rest, in order to arrive at such absolutely necessary and universally
2414valid truths?
2415
2416There is no other way than through intuitions or conceptions, as such;
2417and these are given either a priori or a posteriori. The latter, namely,
2418empirical conceptions, together with the empirical intuition on which
2419they are founded, cannot afford any synthetical proposition, except such
2420as is itself also empirical, that is, a proposition of experience. But
2421an empirical proposition cannot possess the qualities of necessity and
2422absolute universality, which, nevertheless, are the characteristics of
2423all geometrical propositions. As to the first and only means to arrive
2424at such cognitions, namely, through mere conceptions or intuitions a
2425priori, it is quite clear that from mere conceptions no synthetical
2426cognitions, but only analytical ones, can be obtained. Take, for
2427example, the proposition: "Two straight lines cannot enclose a space,
2428and with these alone no figure is possible," and try to deduce it
2429from the conception of a straight line and the number two; or take the
2430proposition: "It is possible to construct a figure with three straight
2431lines," and endeavour, in like manner, to deduce it from the mere
2432conception of a straight line and the number three. All your endeavours
2433are in vain, and you find yourself forced to have recourse to intuition,
2434as, in fact, geometry always does. You therefore give yourself an object
2435in intuition. But of what kind is this intuition? Is it a pure a
2436priori, or is it an empirical intuition? If the latter, then neither an
2437universally valid, much less an apodeictic proposition can arise from
2438it, for experience never can give us any such proposition. You must,
2439therefore, give yourself an object a priori in intuition, and upon that
2440ground your synthetical proposition. Now if there did not exist within
2441you a faculty of intuition a priori; if this subjective condition were
2442not in respect to its form also the universal condition a priori under
2443which alone the object of this external intuition is itself possible;
2444if the object (that is, the triangle) were something in itself, without
2445relation to you the subject; how could you affirm that that which
2446lies necessarily in your subjective conditions in order to construct a
2447triangle, must also necessarily belong to the triangle in itself? For
2448to your conceptions of three lines, you could not add anything new (that
2449is, the figure); which, therefore, must necessarily be found in the
2450object, because the object is given before your cognition, and not by
2451means of it. If, therefore, space (and time also) were not a mere form
2452of your intuition, which contains conditions a priori, under which alone
2453things can become external objects for you, and without which subjective
2454conditions the objects are in themselves nothing, you could not
2455construct any synthetical proposition whatsoever regarding external
2456objects. It is therefore not merely possible or probable, but
2457indubitably certain, that space and time, as the necessary conditions
2458of all our external and internal experience, are merely subjective
2459conditions of all our intuitions, in relation to which all objects are
2460therefore mere phenomena, and not things in themselves, presented to us
2461in this particular manner. And for this reason, in respect to the form
2462of phenomena, much may be said a priori, whilst of the thing in itself,
2463which may lie at the foundation of these phenomena, it is impossible to
2464say anything.
2465
2466II. In confirmation of this theory of the ideality of the external as
2467well as internal sense, consequently of all objects of sense, as mere
2468phenomena, we may especially remark that all in our cognition that
2469belongs to intuition contains nothing more than mere relations. (The
2470feelings of pain and pleasure, and the will, which are not cognitions,
2471are excepted.) The relations, to wit, of place in an intuition
2472(extension), change of place (motion), and laws according to which this
2473change is determined (moving forces). That, however, which is present in
2474this or that place, or any operation going on, or result taking place
2475in the things themselves, with the exception of change of place, is not
2476given to us by intuition. Now by means of mere relations, a thing cannot
2477be known in itself; and it may therefore be fairly concluded, that, as
2478through the external sense nothing but mere representations of relations
2479are given us, the said external sense in its representation can contain
2480only the relation of the object to the subject, but not the essential
2481nature of the object as a thing in itself.
2482
2483The same is the case with the internal intuition, not only because,
2484in the internal intuition, the representation of the external senses
2485constitutes the material with which the mind is occupied; but because
2486time, in which we place, and which itself antecedes the consciousness
2487of, these representations in experience, and which, as the formal
2488condition of the mode according to which objects are placed in the mind,
2489lies at the foundation of them, contains relations of the successive,
2490the coexistent, and of that which always must be coexistent with
2491succession, the permanent. Now that which, as representation, can
2492antecede every exercise of thought (of an object), is intuition; and
2493when it contains nothing but relations, it is the form of the intuition,
2494which, as it presents us with no representation, except in so far as
2495something is placed in the mind, can be nothing else than the mode in
2496which the mind is affected by its own activity, to wit--its presenting
2497to itself representations, consequently the mode in which the mind is
2498affected by itself; that is, it can be nothing but an internal sense in
2499respect to its form. Everything that is represented through the medium
2500of sense is so far phenomenal; consequently, we must either refuse
2501altogether to admit an internal sense, or the subject, which is the
2502object of that sense, could only be represented by it as phenomenon, and
2503not as it would judge of itself, if its intuition were pure spontaneous
2504activity, that is, were intellectual. The difficulty here lies wholly in
2505the question: How can the subject have an internal intuition of itself?
2506But this difficulty is common to every theory. The consciousness of
2507self (apperception) is the simple representation of the "ego"; and if by
2508means of that representation alone, all the manifold representations in
2509the subject were spontaneously given, then our internal intuition
2510would be intellectual. This consciousness in man requires an internal
2511perception of the manifold representations which are previously given in
2512the subject; and the manner in which these representations are given in
2513the mind without spontaneity, must, on account of this difference
2514(the want of spontaneity), be called sensibility. If the faculty of
2515self-consciousness is to apprehend what lies in the mind, it must all
2516act that and can in this way alone produce an intuition of self. But the
2517form of this intuition, which lies in the original constitution of the
2518mind, determines, in the representation of time, the manner in which the
2519manifold representations are to combine themselves in the mind;
2520since the subject intuites itself, not as it would represent itself
2521immediately and spontaneously, but according to the manner in which the
2522mind is internally affected, consequently, as it appears, and not as it
2523is.
2524
2525III. When we say that the intuition of external objects, and also the
2526self-intuition of the subject, represent both, objects and subject, in
2527space and time, as they affect our senses, that is, as they appear--this
2528is by no means equivalent to asserting that these objects are mere
2529illusory appearances. For when we speak of things as phenomena, the
2530objects, nay, even the properties which we ascribe to them, are looked
2531upon as really given; only that, in so far as this or that property
2532depends upon the mode of intuition of the subject, in the relation
2533of the given object to the subject, the object as phenomenon is to be
2534distinguished from the object as a thing in itself. Thus I do not say
2535that bodies seem or appear to be external to me, or that my soul seems
2536merely to be given in my self-consciousness, although I maintain that
2537the properties of space and time, in conformity to which I set both, as
2538the condition of their existence, abide in my mode of intuition, and not
2539in the objects in themselves. It would be my own fault, if out of that
2540which I should reckon as phenomenon, I made mere illusory appearance.*
2541But this will not happen, because of our principle of the ideality
2542of all sensuous intuitions. On the contrary, if we ascribe objective
2543reality to these forms of representation, it becomes impossible to avoid
2544changing everything into mere appearance. For if we regard space
2545and time as properties, which must be found in objects as things in
2546themselves, as sine quibus non of the possibility of their existence,
2547and reflect on the absurdities in which we then find ourselves involved,
2548inasmuch as we are compelled to admit the existence of two infinite
2549things, which are nevertheless not substances, nor anything really
2550inhering in substances, nay, to admit that they are the necessary
2551conditions of the existence of all things, and moreover, that they must
2552continue to exist, although all existing things were annihilated--we
2553cannot blame the good Berkeley for degrading bodies to mere illusory
2554appearances. Nay, even our own existence, which would in this case
2555depend upon the self-existent reality of such a mere nonentity as time,
2556would necessarily be changed with it into mere appearance--an absurdity
2557which no one has as yet been guilty of.
2558
2559 [*Footnote: The predicates of the phenomenon can be affixed
2560 to the object itself in relation to our sensuous faculty;
2561 for example, the red colour or the perfume to the rose. But
2562 (illusory) appearance never can be attributed as a predicate
2563 to an object, for this very reason, that it attributes to
2564 this object in itself that which belongs to it only in
2565 relation to our sensuous faculty, or to the subject in
2566 general, e.g., the two handles which were formerly ascribed
2567 to Saturn. That which is never to be found in the object
2568 itself, but always in the relation of the object to the
2569 subject, and which moreover is inseparable from our
2570 representation of the object, we denominate phenomenon. Thus
2571 the predicates of space and time are rightly attributed to
2572 objects of the senses as such, and in this there is no
2573 illusion. On the contrary, if I ascribe redness of the rose
2574 as a thing in itself, or to Saturn his handles, or extension
2575 to all external objects, considered as things in themselves,
2576 without regarding the determinate relation of these objects
2577 to the subject, and without limiting my judgement to that
2578 relation--then, and then only, arises illusion.]
2579
2580
2581IV. In natural theology, where we think of an object--God--which never
2582can be an object of intuition to us, and even to himself can never be
2583an object of sensuous intuition, we carefully avoid attributing to
2584his intuition the conditions of space and time--and intuition all his
2585cognition must be, and not thought, which always includes limitation.
2586But with what right can we do this if we make them forms of objects as
2587things in themselves, and such, moreover, as would continue to exist as
2588a priori conditions of the existence of things, even though the things
2589themselves were annihilated? For as conditions of all existence in
2590general, space and time must be conditions of the existence of the
2591Supreme Being also. But if we do not thus make them objective forms
2592of all things, there is no other way left than to make them subjective
2593forms of our mode of intuition--external and internal; which is called
2594sensuous, because it is not primitive, that is, is not such as gives in
2595itself the existence of the object of the intuition (a mode of intuition
2596which, so far as we can judge, can belong only to the Creator), but is
2597dependent on the existence of the object, is possible, therefore, only
2598on condition that the representative faculty of the subject is affected
2599by the object.
2600
2601It is, moreover, not necessary that we should limit the mode of
2602intuition in space and time to the sensuous faculty of man. It may well
2603be that all finite thinking beings must necessarily in this respect
2604agree with man (though as to this we cannot decide), but sensibility
2605does not on account of this universality cease to be sensibility, for
2606this very reason, that it is a deduced (intuitus derivativus), and not
2607an original (intuitus originarius), consequently not an intellectual
2608intuition, and this intuition, as such, for reasons above mentioned,
2609seems to belong solely to the Supreme Being, but never to a being
2610dependent, quoad its existence, as well as its intuition (which its
2611existence determines and limits relatively to given objects). This
2612latter remark, however, must be taken only as an illustration, and not
2613as any proof of the truth of our aesthetical theory.
2614
2615
2616
2617SS 10. Conclusion of the Transcendental Aesthetic.
2618
2619We have now completely before us one part of the solution of the grand
2620general problem of transcendental philosophy, namely, the question: "How
2621are synthetical propositions a priori possible?" That is to say, we have
2622shown that we are in possession of pure a priori intuitions, namely,
2623space and time, in which we find, when in a judgement a priori we pass
2624out beyond the given conception, something which is not discoverable in
2625that conception, but is certainly found a priori in the intuition which
2626corresponds to the conception, and can be united synthetically with it.
2627But the judgements which these pure intuitions enable us to make, never
2628reach farther than to objects of the senses, and are valid only for
2629objects of possible experience.
2630
2631
2632
2633SECOND PART. TRANSCENDENTAL LOGIC.
2634
2635INTRODUCTION. Idea of a Transcendental Logic.
2636
2637I. Of Logic in General.
2638
2639Our knowledge springs from two main sources in the mind, first of which
2640is the faculty or power of receiving representations (receptivity for
2641impressions); the second is the power of cognizing by means of these
2642representations (spontaneity in the production of conceptions). Through
2643the first an object is given to us; through the second, it is, in
2644relation to the representation (which is a mere determination of the
2645mind), thought. Intuition and conceptions constitute, therefore, the
2646elements of all our knowledge, so that neither conceptions without
2647an intuition in some way corresponding to them, nor intuition without
2648conceptions, can afford us a cognition. Both are either pure or
2649empirical. They are empirical, when sensation (which presupposes the
2650actual presence of the object) is contained in them; and pure, when no
2651sensation is mixed with the representation. Sensations we may call
2652the matter of sensuous cognition. Pure intuition consequently contains
2653merely the form under which something is intuited, and pure conception
2654only the form of the thought of an object. Only pure intuitions and pure
2655conceptions are possible a priori; the empirical only a posteriori.
2656
2657We apply the term sensibility to the receptivity of the mind for
2658impressions, in so far as it is in some way affected; and, on the other
2659hand, we call the faculty of spontaneously producing representations,
2660or the spontaneity of cognition, understanding. Our nature is so
2661constituted that intuition with us never can be other than sensuous,
2662that is, it contains only the mode in which we are affected by objects.
2663On the other hand, the faculty of thinking the object of sensuous
2664intuition is the understanding. Neither of these faculties has a
2665preference over the other. Without the sensuous faculty no object
2666would be given to us, and without the understanding no object would
2667be thought. Thoughts without content are void; intuitions without
2668conceptions, blind. Hence it is as necessary for the mind to make its
2669conceptions sensuous (that is, to join to them the object in intuition),
2670as to make its intuitions intelligible (that is, to bring them under
2671conceptions). Neither of these faculties can exchange its proper
2672function. Understanding cannot intuite, and the sensuous faculty cannot
2673think. In no other way than from the united operation of both, can
2674knowledge arise. But no one ought, on this account, to overlook the
2675difference of the elements contributed by each; we have rather great
2676reason carefully to separate and distinguish them. We therefore
2677distinguish the science of the laws of sensibility, that is, aesthetic,
2678from the science of the laws of the understanding, that is, logic.
2679
2680Now, logic in its turn may be considered as twofold--namely, as logic
2681of the general, or of the particular use of the understanding. The first
2682contains the absolutely necessary laws of thought, without which no use
2683whatsoever of the understanding is possible, and gives laws therefore to
2684the understanding, without regard to the difference of objects on which
2685it may be employed. The logic of the particular use of the understanding
2686contains the laws of correct thinking upon a particular class of
2687objects. The former may be called elemental logic--the latter, the
2688organon of this or that particular science. The latter is for the
2689most part employed in the schools, as a propaedeutic to the sciences,
2690although, indeed, according to the course of human reason, it is the
2691last thing we arrive at, when the science has been already matured, and
2692needs only the finishing touches towards its correction and completion;
2693for our knowledge of the objects of our attempted science must be
2694tolerably extensive and complete before we can indicate the laws by
2695which a science of these objects can be established.
2696
2697General logic is again either pure or applied. In the former, we
2698abstract all the empirical conditions under which the understanding is
2699exercised; for example, the influence of the senses, the play of the
2700fantasy or imagination, the laws of the memory, the force of habit, of
2701inclination, etc., consequently also, the sources of prejudice--in a
2702word, we abstract all causes from which particular cognitions
2703arise, because these causes regard the understanding under certain
2704circumstances of its application, and, to the knowledge of them
2705experience is required. Pure general logic has to do, therefore, merely
2706with pure a priori principles, and is a canon of understanding and
2707reason, but only in respect of the formal part of their use, be the
2708content what it may, empirical or transcendental. General logic is
2709called applied, when it is directed to the laws of the use of the
2710understanding, under the subjective empirical conditions which
2711psychology teaches us. It has therefore empirical principles, although,
2712at the same time, it is in so far general, that it applies to the
2713exercise of the understanding, without regard to the difference of
2714objects. On this account, moreover, it is neither a canon of the
2715understanding in general, nor an organon of a particular science, but
2716merely a cathartic of the human understanding.
2717
2718In general logic, therefore, that part which constitutes pure logic must
2719be carefully distinguished from that which constitutes applied (though
2720still general) logic. The former alone is properly science, although
2721short and dry, as the methodical exposition of an elemental doctrine of
2722the understanding ought to be. In this, therefore, logicians must always
2723bear in mind two rules:
2724
27251. As general logic, it makes abstraction of all content of the
2726cognition of the understanding, and of the difference of objects, and
2727has to do with nothing but the mere form of thought.
2728
27292. As pure logic, it has no empirical principles, and consequently
2730draws nothing (contrary to the common persuasion) from psychology, which
2731therefore has no influence on the canon of the understanding. It is a
2732demonstrated doctrine, and everything in it must be certain completely a
2733priori.
2734
2735What I called applied logic (contrary to the common acceptation of this
2736term, according to which it should contain certain exercises for the
2737scholar, for which pure logic gives the rules), is a representation
2738of the understanding, and of the rules of its necessary employment
2739in concreto, that is to say, under the accidental conditions of the
2740subject, which may either hinder or promote this employment, and which
2741are all given only empirically. Thus applied logic treats of attention,
2742its impediments and consequences, of the origin of error, of the state
2743of doubt, hesitation, conviction, etc., and to it is related pure
2744general logic in the same way that pure morality, which contains only
2745the necessary moral laws of a free will, is related to practical ethics,
2746which considers these laws under all the impediments of feelings,
2747inclinations, and passions to which men are more or less subjected, and
2748which never can furnish us with a true and demonstrated science, because
2749it, as well as applied logic, requires empirical and psychological
2750principles.
2751
2752
2753
2754II. Of Transcendental Logic.
2755
2756General logic, as we have seen, makes abstraction of all content of
2757cognition, that is, of all relation of cognition to its object, and
2758regards only the logical form in the relation of cognitions to each
2759other, that is, the form of thought in general. But as we have both pure
2760and empirical intuitions (as transcendental aesthetic proves), in like
2761manner a distinction might be drawn between pure and empirical thought
2762(of objects). In this case, there would exist a kind of logic, in which
2763we should not make abstraction of all content of cognition; for or logic
2764which should comprise merely the laws of pure thought (of an object),
2765would of course exclude all those cognitions which were of empirical
2766content. This kind of logic would also examine the origin of our
2767cognitions of objects, so far as that origin cannot be ascribed to the
2768objects themselves; while, on the contrary, general logic has nothing
2769to do with the origin of our cognitions, but contemplates our
2770representations, be they given primitively a priori in ourselves, or be
2771they only of empirical origin, solely according to the laws which the
2772understanding observes in employing them in the process of thought, in
2773relation to each other. Consequently, general logic treats of the form
2774of the understanding only, which can be applied to representations, from
2775whatever source they may have arisen.
2776
2777And here I shall make a remark, which the reader must bear well in mind
2778in the course of the following considerations, to wit, that not every
2779cognition a priori, but only those through which we cognize that and how
2780certain representations (intuitions or conceptions) are applied or are
2781possible only a priori; that is to say, the a priori possibility of
2782cognition and the a priori use of it are transcendental. Therefore
2783neither is space, nor any a priori geometrical determination of space,
2784a transcendental Representation, but only the knowledge that such a
2785representation is not of empirical origin, and the possibility of its
2786relating to objects of experience, although itself a priori, can be
2787called transcendental. So also, the application of space to objects
2788in general would be transcendental; but if it be limited to objects of
2789sense it is empirical. Thus, the distinction of the transcendental
2790and empirical belongs only to the critique of cognitions, and does not
2791concern the relation of these to their object.
2792
2793Accordingly, in the expectation that there may perhaps be conceptions
2794which relate a priori to objects, not as pure or sensuous intuitions,
2795but merely as acts of pure thought (which are therefore conceptions,
2796but neither of empirical nor aesthetical origin)--in this expectation,
2797I say, we form to ourselves, by anticipation, the idea of a science
2798of pure understanding and rational cognition, by means of which we may
2799cogitate objects entirely a priori. A science of this kind, which should
2800determine the origin, the extent, and the objective validity of such
2801cognitions, must be called transcendental logic, because it has not,
2802like general logic, to do with the laws of understanding and reason
2803in relation to empirical as well as pure rational cognitions without
2804distinction, but concerns itself with these only in an a priori relation
2805to objects.
2806
2807
2808
2809III. Of the Division of General Logic into Analytic and Dialectic.
2810
2811The old question with which people sought to push logicians into a
2812corner, so that they must either have recourse to pitiful sophisms or
2813confess their ignorance, and consequently the vanity of their whole art,
2814is this: "What is truth?" The definition of the word truth, to wit,
2815"the accordance of the cognition with its object," is presupposed in
2816the question; but we desire to be told, in the answer to it, what is the
2817universal and secure criterion of the truth of every cognition.
2818
2819To know what questions we may reasonably propose is in itself a strong
2820evidence of sagacity and intelligence. For if a question be in itself
2821absurd and unsusceptible of a rational answer, it is attended with the
2822danger--not to mention the shame that falls upon the person who proposes
2823it--of seducing the unguarded listener into making absurd answers, and
2824we are presented with the ridiculous spectacle of one (as the ancients
2825said) "milking the he-goat, and the other holding a sieve."
2826
2827If truth consists in the accordance of a cognition with its object,
2828this object must be, ipso facto, distinguished from all others; for a
2829cognition is false if it does not accord with the object to which it
2830relates, although it contains something which may be affirmed of other
2831objects. Now an universal criterion of truth would be that which is
2832valid for all cognitions, without distinction of their objects. But
2833it is evident that since, in the case of such a criterion, we make
2834abstraction of all the content of a cognition (that is, of all relation
2835to its object), and truth relates precisely to this content, it must
2836be utterly absurd to ask for a mark of the truth of this content of
2837cognition; and that, accordingly, a sufficient, and at the same time
2838universal, test of truth cannot possibly be found. As we have already
2839termed the content of a cognition its matter, we shall say: "Of the
2840truth of our cognitions in respect of their matter, no universal test
2841can be demanded, because such a demand is self-contradictory."
2842
2843On the other hand, with regard to our cognition in respect of its mere
2844form (excluding all content), it is equally manifest that logic, in
2845so far as it exhibits the universal and necessary laws of the
2846understanding, must in these very laws present us with criteria of
2847truth. Whatever contradicts these rules is false, because thereby the
2848understanding is made to contradict its own universal laws of thought;
2849that is, to contradict itself. These criteria, however, apply solely to
2850the form of truth, that is, of thought in general, and in so far they
2851are perfectly accurate, yet not sufficient. For although a cognition
2852may be perfectly accurate as to logical form, that is, not
2853self-contradictory, it is notwithstanding quite possible that it may
2854not stand in agreement with its object. Consequently, the merely logical
2855criterion of truth, namely, the accordance of a cognition with the
2856universal and formal laws of understanding and reason, is nothing more
2857than the conditio sine qua non, or negative condition of all truth.
2858Farther than this logic cannot go, and the error which depends not
2859on the form, but on the content of the cognition, it has no test to
2860discover.
2861
2862General logic, then, resolves the whole formal business of understanding
2863and reason into its elements, and exhibits them as principles of all
2864logical judging of our cognitions. This part of logic may, therefore, be
2865called analytic, and is at least the negative test of truth, because all
2866cognitions must first of an be estimated and tried according to these
2867laws before we proceed to investigate them in respect of their content,
2868in order to discover whether they contain positive truth in regard to
2869their object. Because, however, the mere form of a cognition, accurately
2870as it may accord with logical laws, is insufficient to supply us with
2871material (objective) truth, no one, by means of logic alone, can venture
2872to predicate anything of or decide concerning objects, unless he has
2873obtained, independently of logic, well-grounded information about them,
2874in order afterwards to examine, according to logical laws, into the use
2875and connection, in a cohering whole, of that information, or, what is
2876still better, merely to test it by them. Notwithstanding, there lies so
2877seductive a charm in the possession of a specious art like this--an
2878art which gives to all our cognitions the form of the understanding,
2879although with respect to the content thereof we may be sadly
2880deficient--that general logic, which is merely a canon of judgement, has
2881been employed as an organon for the actual production, or rather for
2882the semblance of production, of objective assertions, and has thus
2883been grossly misapplied. Now general logic, in its assumed character of
2884organon, is called dialectic.
2885
2886Different as are the significations in which the ancients used this
2887term for a science or an art, we may safely infer, from their actual
2888employment of it, that with them it was nothing else than a logic of
2889illusion--a sophistical art for giving ignorance, nay, even intentional
2890sophistries, the colouring of truth, in which the thoroughness of
2891procedure which logic requires was imitated, and their topic employed
2892to cloak the empty pretensions. Now it may be taken as a safe and useful
2893warning, that general logic, considered as an organon, must always be
2894a logic of illusion, that is, be dialectical, for, as it teaches us
2895nothing whatever respecting the content of our cognitions, but merely
2896the formal conditions of their accordance with the understanding, which
2897do not relate to and are quite indifferent in respect of objects, any
2898attempt to employ it as an instrument (organon) in order to extend and
2899enlarge the range of our knowledge must end in mere prating; any one
2900being able to maintain or oppose, with some appearance of truth, any
2901single assertion whatever.
2902
2903Such instruction is quite unbecoming the dignity of philosophy. For
2904these reasons we have chosen to denominate this part of logic dialectic,
2905in the sense of a critique of dialectical illusion, and we wish the term
2906to be so understood in this place.
2907
2908
2909
2910IV. Of the Division of Transcendental Logic into Transcendental
2911 Analytic and Dialectic.
2912
2913In transcendental logic we isolate the understanding (as in
2914transcendental aesthetic the sensibility) and select from our cognition
2915merely that part of thought which has its origin in the understanding
2916alone. The exercise of this pure cognition, however, depends upon this
2917as its condition, that objects to which it may be applied be given to
2918us in intuition, for without intuition the whole of our cognition
2919is without objects, and is therefore quite void. That part of
2920transcendental logic, then, which treats of the elements of pure
2921cognition of the understanding, and of the principles without which no
2922object at all can be thought, is transcendental analytic, and at the
2923same time a logic of truth. For no cognition can contradict it, without
2924losing at the same time all content, that is, losing all reference to an
2925object, and therefore all truth. But because we are very easily seduced
2926into employing these pure cognitions and principles of the understanding
2927by themselves, and that even beyond the boundaries of experience, which
2928yet is the only source whence we can obtain matter (objects) on which
2929those pure conceptions may be employed--understanding runs the risk of
2930making, by means of empty sophisms, a material and objective use of
2931the mere formal principles of the pure understanding, and of passing
2932judgements on objects without distinction--objects which are not given
2933to us, nay, perhaps cannot be given to us in any way. Now, as it ought
2934properly to be only a canon for judging of the empirical use of the
2935understanding, this kind of logic is misused when we seek to employ
2936it as an organon of the universal and unlimited exercise of the
2937understanding, and attempt with the pure understanding alone to judge
2938synthetically, affirm, and determine respecting objects in general. In
2939this case the exercise of the pure understanding becomes dialectical.
2940The second part of our transcendental logic must therefore be a critique
2941of dialectical illusion, and this critique we shall term transcendental
2942dialectic--not meaning it as an art of producing dogmatically
2943such illusion (an art which is unfortunately too current among
2944the practitioners of metaphysical juggling), but as a critique of
2945understanding and reason in regard to their hyperphysical use. This
2946critique will expose the groundless nature of the pretensions of
2947these two faculties, and invalidate their claims to the discovery
2948and enlargement of our cognitions merely by means of transcendental
2949principles, and show that the proper employment of these faculties is to
2950test the judgements made by the pure understanding, and to guard it from
2951sophistical delusion.
2952
2953
2954
2955TRANSCENDENTAL LOGIC. FIRST DIVISION.
2956
2957TRANSCENDENTAL ANALYTIC.
2958
2959SS I.
2960
2961Transcendental analytic is the dissection of the whole of our a priori
2962knowledge into the elements of the pure cognition of the understanding.
2963In order to effect our purpose, it is necessary: (1) That the
2964conceptions be pure and not empirical; (2) That they belong not to
2965intuition and sensibility, but to thought and understanding; (3) That
2966they be elementary conceptions, and as such, quite different from
2967deduced or compound conceptions; (4) That our table of these elementary
2968conceptions be complete, and fill up the whole sphere of the pure
2969understanding. Now this completeness of a science cannot be accepted
2970with confidence on the guarantee of a mere estimate of its existence in
2971an aggregate formed only by means of repeated experiments and attempts.
2972The completeness which we require is possible only by means of an idea
2973of the totality of the a priori cognition of the understanding, and
2974through the thereby determined division of the conceptions which form
2975the said whole; consequently, only by means of their connection in
2976a system. Pure understanding distinguishes itself not merely from
2977everything empirical, but also completely from all sensibility. It is
2978a unity self-subsistent, self-sufficient, and not to be enlarged by any
2979additions from without. Hence the sum of its cognition constitutes
2980a system to be determined by and comprised under an idea; and the
2981completeness and articulation of this system can at the same time
2982serve as a test of the correctness and genuineness of all the parts of
2983cognition that belong to it. The whole of this part of transcendental
2984logic consists of two books, of which the one contains the conceptions,
2985and the other the principles of pure understanding.
2986
2987
2988
2989BOOK I.
2990
2991SS 2. Analytic of Conceptions.
2992
2993By the term Analytic of Conceptions, I do not understand the analysis
2994of these, or the usual process in philosophical investigations of
2995dissecting the conceptions which present themselves, according to
2996their content, and so making them clear; but I mean the hitherto little
2997attempted dissection of the faculty of understanding itself, in order to
2998investigate the possibility of conceptions a priori, by looking for them
2999in the understanding alone, as their birthplace, and analysing the pure
3000use of this faculty. For this is the proper duty of a transcendental
3001philosophy; what remains is the logical treatment of the conceptions in
3002philosophy in general. We shall therefore follow up the pure conceptions
3003even to their germs and beginnings in the human understanding, in which
3004they lie, until they are developed on occasions presented by experience,
3005and, freed by the same understanding from the empirical conditions
3006attaching to them, are set forth in their unalloyed purity.
3007
3008
3009
3010CHAPTER I. Of the Transcendental Clue to the Discovery of all Pure
3011Conceptions of the Understanding.
3012
3013SS 3. Introductory.
3014
3015When we call into play a faculty of cognition, different conceptions
3016manifest themselves according to the different circumstances, and
3017make known this faculty, and assemble themselves into a more or less
3018extensive collection, according to the time or penetration that has been
3019applied to the consideration of them. Where this process, conducted as
3020it is mechanically, so to speak, will end, cannot be determined with
3021certainty. Besides, the conceptions which we discover in this haphazard
3022manner present themselves by no means in order and systematic unity,
3023but are at last coupled together only according to resemblances to
3024each other, and arranged in series, according to the quantity of their
3025content, from the simpler to the more complex--series which are anything
3026but systematic, though not altogether without a certain kind of method
3027in their construction.
3028
3029Transcendental philosophy has the advantage, and moreover the duty, of
3030searching for its conceptions according to a principle; because these
3031conceptions spring pure and unmixed out of the understanding as
3032an absolute unity, and therefore must be connected with each other
3033according to one conception or idea. A connection of this kind, however,
3034furnishes us with a ready prepared rule, by which its proper place
3035may be assigned to every pure conception of the understanding, and the
3036completeness of the system of all be determined a priori--both which
3037would otherwise have been dependent on mere choice or chance.
3038
3039
3040SS 4. SECTION 1. Of defined above Use of understanding in General.
3041
3042The understanding was defined above only negatively, as a non-sensuous
3043faculty of cognition. Now, independently of sensibility, we cannot
3044possibly have any intuition; consequently, the understanding is no
3045faculty of intuition. But besides intuition there is no other mode of
3046cognition, except through conceptions; consequently, the cognition of
3047every, at least of every human, understanding is a cognition through
3048conceptions--not intuitive, but discursive. All intuitions, as sensuous,
3049depend on affections; conceptions, therefore, upon functions. By the
3050word function I understand the unity of the act of arranging diverse
3051representations under one common representation. Conceptions, then, are
3052based on the spontaneity of thought, as sensuous intuitions are on the
3053receptivity of impressions. Now, the understanding cannot make any
3054other use of these conceptions than to judge by means of them. As no
3055representation, except an intuition, relates immediately to its object,
3056a conception never relates immediately to an object, but only to
3057some other representation thereof, be that an intuition or itself a
3058conception. A judgement, therefore, is the mediate cognition of an
3059object, consequently the representation of a representation of it. In
3060every judgement there is a conception which applies to, and is valid for
3061many other conceptions, and which among these comprehends also a given
3062representation, this last being immediately connected with an object.
3063For example, in the judgement--"All bodies are divisible," our
3064conception of divisible applies to various other conceptions; among
3065these, however, it is here particularly applied to the conception of
3066body, and this conception of body relates to certain phenomena which
3067occur to us. These objects, therefore, are mediately represented by the
3068conception of divisibility. All judgements, accordingly, are functions
3069of unity in our representations, inasmuch as, instead of an immediate, a
3070higher representation, which comprises this and various others, is used
3071for our cognition of the object, and thereby many possible cognitions
3072are collected into one. But we can reduce all acts of the understanding
3073to judgements, so that understanding may be represented as the faculty
3074of judging. For it is, according to what has been said above, a faculty
3075of thought. Now thought is cognition by means of conceptions. But
3076conceptions, as predicates of possible judgements, relate to some
3077representation of a yet undetermined object. Thus the conception of body
3078indicates something--for example, metal--which can be cognized by means
3079of that conception. It is therefore a conception, for the reason alone
3080that other representations are contained under it, by means of which
3081it can relate to objects. It is therefore the predicate to a possible
3082judgement; for example: "Every metal is a body." All the functions of
3083the understanding therefore can be discovered, when we can completely
3084exhibit the functions of unity in judgements. And that this may be
3085effected very easily, the following section will show.
3086
3087
3088SS 5. SECTION II. Of the Logical Function of the Understanding in
3089Judgements.
3090
3091If we abstract all the content of a judgement, and consider only the
3092intellectual form thereof, we find that the function of thought in a
3093judgement can be brought under four heads, of which each contains three
3094momenta. These may be conveniently represented in the following table:
3095
3096 1
3097 Quantity of judgements
3098 Universal
3099 Particular
3100 Singular
3101
3102 2 3
3103 Quality Relation
3104 Affirmative Categorical
3105 Negative Hypothetical
3106 Infinite Disjunctive
3107
3108 4
3109 Modality
3110 Problematical
3111 Assertorical
3112 Apodeictical
3113
3114
3115As this division appears to differ in some, though not essential points,
3116from the usual technique of logicians, the following observations,
3117for the prevention of otherwise possible misunderstanding, will not be
3118without their use.
3119
31201. Logicians say, with justice, that in the use of judgements in
3121syllogisms, singular judgements may be treated like universal ones.
3122For, precisely because a singular judgement has no extent at all, its
3123predicate cannot refer to a part of that which is contained in the
3124conception of the subject and be excluded from the rest. The predicate
3125is valid for the whole conception just as if it were a general
3126conception, and had extent, to the whole of which the predicate applied.
3127On the other hand, let us compare a singular with a general judgement,
3128merely as a cognition, in regard to quantity. The singular judgement
3129relates to the general one, as unity to infinity, and is therefore in
3130itself essentially different. Thus, if we estimate a singular judgement
3131(judicium singulare) not merely according to its intrinsic validity as a
3132judgement, but also as a cognition generally, according to its quantity
3133in comparison with that of other cognitions, it is then entirely
3134different from a general judgement (judicium commune), and in a complete
3135table of the momenta of thought deserves a separate place--though,
3136indeed, this would not be necessary in a logic limited merely to the
3137consideration of the use of judgements in reference to each other.
3138
31392. In like manner, in transcendental logic, infinite must be
3140distinguished from affirmative judgements, although in general logic
3141they are rightly enough classed under affirmative. General logic
3142abstracts all content of the predicate (though it be negative), and
3143only considers whether the said predicate be affirmed or denied of the
3144subject. But transcendental logic considers also the worth or content of
3145this logical affirmation--an affirmation by means of a merely negative
3146predicate, and inquires how much the sum total of our cognition gains
3147by this affirmation. For example, if I say of the soul, "It is not
3148mortal"--by this negative judgement I should at least ward off error.
3149Now, by the proposition, "The soul is not mortal," I have, in respect of
3150the logical form, really affirmed, inasmuch as I thereby place the soul
3151in the unlimited sphere of immortal beings. Now, because of the whole
3152sphere of possible existences, the mortal occupies one part, and the
3153immortal the other, neither more nor less is affirmed by the proposition
3154than that the soul is one among the infinite multitude of things
3155which remain over, when I take away the whole mortal part. But by this
3156proceeding we accomplish only this much, that the infinite sphere of
3157all possible existences is in so far limited that the mortal is excluded
3158from it, and the soul is placed in the remaining part of the extent
3159of this sphere. But this part remains, notwithstanding this exception,
3160infinite, and more and more parts may be taken away from the whole
3161sphere, without in the slightest degree thereby augmenting or
3162affirmatively determining our conception of the soul. These judgements,
3163therefore, infinite in respect of their logical extent, are, in
3164respect of the content of their cognition, merely limitative; and are
3165consequently entitled to a place in our transcendental table of all
3166the momenta of thought in judgements, because the function of the
3167understanding exercised by them may perhaps be of importance in the
3168field of its pure a priori cognition.
3169
31703. All relations of thought in judgements are those (a) of the predicate
3171to the subject; (b) of the principle to its consequence; (c) of the
3172divided cognition and all the members of the division to each other. In
3173the first of these three classes, we consider only two conceptions; in
3174the second, two judgements; in the third, several judgements in relation
3175to each other. The hypothetical proposition, "If perfect justice exists,
3176the obstinately wicked are punished," contains properly the relation to
3177each other of two propositions, namely, "Perfect justice exists," and
3178"The obstinately wicked are punished." Whether these propositions are in
3179themselves true is a question not here decided. Nothing is cogitated
3180by means of this judgement except a certain consequence. Finally, the
3181disjunctive judgement contains a relation of two or more propositions to
3182each other--a relation not of consequence, but of logical opposition, in
3183so far as the sphere of the one proposition excludes that of the other.
3184But it contains at the same time a relation of community, in so far as
3185all the propositions taken together fill up the sphere of the cognition.
3186The disjunctive judgement contains, therefore, the relation of the parts
3187of the whole sphere of a cognition, since the sphere of each part is a
3188complemental part of the sphere of the other, each contributing to
3189form the sum total of the divided cognition. Take, for example, the
3190proposition, "The world exists either through blind chance, or through
3191internal necessity, or through an external cause." Each of these
3192propositions embraces a part of the sphere of our possible cognition
3193as to the existence of a world; all of them taken together, the whole
3194sphere. To take the cognition out of one of these spheres, is equivalent
3195to placing it in one of the others; and, on the other hand, to place
3196it in one sphere is equivalent to taking it out of the rest. There is,
3197therefore, in a disjunctive judgement a certain community of cognitions,
3198which consists in this, that they mutually exclude each other, yet
3199thereby determine, as a whole, the true cognition, inasmuch as, taken
3200together, they make up the complete content of a particular given
3201cognition. And this is all that I find necessary, for the sake of what
3202follows, to remark in this place.
3203
32044. The modality of judgements is a quite peculiar function, with this
3205distinguishing characteristic, that it contributes nothing to the
3206content of a judgement (for besides quantity, quality, and relation,
3207there is nothing more that constitutes the content of a judgement), but
3208concerns itself only with the value of the copula in relation to thought
3209in general. Problematical judgements are those in which the affirmation
3210or negation is accepted as merely possible (ad libitum). In the
3211assertorical, we regard the proposition as real (true); in the
3212apodeictical, we look on it as necessary.* Thus the two judgements
3213(antecedens et consequens), the relation of which constitutes a
3214hypothetical judgement, likewise those (the members of the division) in
3215whose reciprocity the disjunctive consists, are only problematical. In
3216the example above given the proposition, "There exists perfect justice,"
3217is not stated assertorically, but as an ad libitum judgement, which
3218someone may choose to adopt, and the consequence alone is assertorical.
3219Hence such judgements may be obviously false, and yet, taken
3220problematically, be conditions of our cognition of the truth. Thus
3221the proposition, "The world exists only by blind chance," is in the
3222disjunctive judgement of problematical import only: that is to say, one
3223may accept it for the moment, and it helps us (like the indication of
3224the wrong road among all the roads that one can take) to find out the
3225true proposition. The problematical proposition is, therefore, that
3226which expresses only logical possibility (which is not objective);
3227that is, it expresses a free choice to admit the validity of such a
3228proposition--a merely arbitrary reception of it into the understanding.
3229The assertorical speaks of logical reality or truth; as, for example,
3230in a hypothetical syllogism, the antecedens presents itself in a
3231problematical form in the major, in an assertorical form in the minor,
3232and it shows that the proposition is in harmony with the laws of the
3233understanding. The apodeictical proposition cogitates the assertorical
3234as determined by these very laws of the understanding, consequently as
3235affirming a priori, and in this manner it expresses logical
3236necessity. Now because all is here gradually incorporated with the
3237understanding--inasmuch as in the first place we judge problematically;
3238then accept assertorically our judgement as true; lastly, affirm it as
3239inseparably united with the understanding, that is, as necessary and
3240apodeictical--we may safely reckon these three functions of modality as
3241so many momenta of thought.
3242
3243 [*Footnote: Just as if thought were in the first instance a
3244 function of the understanding; in the second, of judgement;
3245 in the third, of reason. A remark which will be explained in
3246 the sequel.]
3247
3248
3249SS 6. SECTION III. Of the Pure Conceptions of the Understanding, or Categories.
3250
3251General logic, as has been repeatedly said, makes abstraction of all
3252content of cognition, and expects to receive representations from some
3253other quarter, in order, by means of analysis, to convert them into
3254conceptions. On the contrary, transcendental logic has lying before
3255it the manifold content of a priori sensibility, which transcendental
3256aesthetic presents to it in order to give matter to the pure conceptions
3257of the understanding, without which transcendental logic would have no
3258content, and be therefore utterly void. Now space and time contain an
3259infinite diversity of determinations of pure a priori intuition, but are
3260nevertheless the condition of the mind's receptivity, under which alone
3261it can obtain representations of objects, and which, consequently, must
3262always affect the conception of these objects. But the spontaneity of
3263thought requires that this diversity be examined after a certain manner,
3264received into the mind, and connected, in order afterwards to form a
3265cognition out of it. This Process I call synthesis.
3266
3267By the word synthesis, in its most general signification, I understand
3268the process of joining different representations to each other and of
3269comprehending their diversity in one cognition. This synthesis is pure
3270when the diversity is not given empirically but a priori (as that in
3271space and time). Our representations must be given previously to any
3272analysis of them; and no conceptions can arise, quoad their content,
3273analytically. But the synthesis of a diversity (be it given a priori or
3274empirically) is the first requisite for the production of a cognition,
3275which in its beginning, indeed, may be crude and confused, and therefore
3276in need of analysis--still, synthesis is that by which alone the
3277elements of our cognitions are collected and united into a certain
3278content, consequently it is the first thing on which we must fix our
3279attention, if we wish to investigate the origin of our knowledge.
3280
3281Synthesis, generally speaking, is, as we shall afterwards see, the mere
3282operation of the imagination--a blind but indispensable function of the
3283soul, without which we should have no cognition whatever, but of the
3284working of which we are seldom even conscious. But to reduce this
3285synthesis to conceptions is a function of the understanding, by means of
3286which we attain to cognition, in the proper meaning of the term.
3287
3288Pure synthesis, represented generally, gives us the pure conception of
3289the understanding. But by this pure synthesis, I mean that which rests
3290upon a basis of a priori synthetical unity. Thus, our numeration (and
3291this is more observable in large numbers) is a synthesis according to
3292conceptions, because it takes place according to a common basis of unity
3293(for example, the decade). By means of this conception, therefore, the
3294unity in the synthesis of the manifold becomes necessary.
3295
3296By means of analysis different representations are brought under one
3297conception--an operation of which general logic treats. On the other
3298hand, the duty of transcendental logic is to reduce to conceptions, not
3299representations, but the pure synthesis of representations. The first
3300thing which must be given to us for the sake of the a priori cognition
3301of all objects, is the diversity of the pure intuition; the synthesis
3302of this diversity by means of the imagination is the second; but this
3303gives, as yet, no cognition. The conceptions which give unity to this
3304pure synthesis, and which consist solely in the representation of
3305this necessary synthetical unity, furnish the third requisite for
3306the cognition of an object, and these conceptions are given by the
3307understanding.
3308
3309The same function which gives unity to the different representation in
3310a judgement, gives also unity to the mere synthesis of different
3311representations in an intuition; and this unity we call the pure
3312conception of the understanding. Thus, the same understanding, and by
3313the same operations, whereby in conceptions, by means of analytical
3314unity, it produced the logical form of a judgement, introduces, by means
3315of the synthetical unity of the manifold in intuition, a transcendental
3316content into its representations, on which account they are called pure
3317conceptions of the understanding, and they apply a priori to objects, a
3318result not within the power of general logic.
3319
3320In this manner, there arise exactly so many pure conceptions of the
3321understanding, applying a priori to objects of intuition in general, as
3322there are logical functions in all possible judgements. For there is no
3323other function or faculty existing in the understanding besides those
3324enumerated in that table. These conceptions we shall, with Aristotle,
3325call categories, our purpose being originally identical with his,
3326notwithstanding the great difference in the execution.
3327
3328 TABLE OF THE CATEGORIES
3329
3330 1 2
3331
3332 Of Quantity Of Quality
3333 Unity Reality
3334 Plurality Negation
3335 Totality Limitation
3336
3337 3
3338 Of Relation
3339 Of Inherence and Subsistence (substantia et accidens)
3340 Of Causality and Dependence (cause and effect)
3341 Of Community (reciprocity between the agent and patient)
3342
3343 4
3344 Of Modality
3345 Possibility--Impossibility
3346 Existence--Non-existence
3347 Necessity--Contingence
3348
3349
3350This, then, is a catalogue of all the originally pure conceptions of
3351the synthesis which the understanding contains a priori, and these
3352conceptions alone entitle it to be called a pure understanding; inasmuch
3353as only by them it can render the manifold of intuition conceivable,
3354in other words, think an object of intuition. This division is made
3355systematically from a common principle, namely the faculty of judgement
3356(which is just the same as the power of thought), and has not arisen
3357rhapsodically from a search at haphazard after pure conceptions,
3358respecting the full number of which we never could be certain, inasmuch
3359as we employ induction alone in our search, without considering that in
3360this way we can never understand wherefore precisely these conceptions,
3361and none others, abide in the pure understanding. It was a design worthy
3362of an acute thinker like Aristotle, to search for these fundamental
3363conceptions. Destitute, however, of any guiding principle, he picked
3364them up just as they occurred to him, and at first hunted out ten, which
3365he called categories (predicaments). Afterwards be believed that he
3366had discovered five others, which were added under the name of post
3367predicaments. But his catalogue still remained defective. Besides,
3368there are to be found among them some of the modes of pure sensibility
3369(quando, ubi, situs, also prius, simul), and likewise an empirical
3370conception (motus)--which can by no means belong to this genealogical
3371register of the pure understanding. Moreover, there are deduced
3372conceptions (actio, passio) enumerated among the original conceptions,
3373and, of the latter, some are entirely wanting.
3374
3375With regard to these, it is to be remarked, that the categories, as the
3376true primitive conceptions of the pure understanding, have also their
3377pure deduced conceptions, which, in a complete system of transcendental
3378philosophy, must by no means be passed over; though in a merely critical
3379essay we must be contented with the simple mention of the fact.
3380
3381Let it be allowed me to call these pure, but deduced conceptions of
3382the understanding, the predicables of the pure understanding, in
3383contradistinction to predicaments. If we are in possession of the
3384original and primitive, the deduced and subsidiary conceptions can
3385easily be added, and the genealogical tree of the understanding
3386completely delineated. As my present aim is not to set forth a complete
3387system, but merely the principles of one, I reserve this task for
3388another time. It may be easily executed by any one who will refer to the
3389ontological manuals, and subordinate to the category of causality,
3390for example, the predicables of force, action, passion; to that of
3391community, those of presence and resistance; to the categories of
3392modality, those of origination, extinction, change; and so with the
3393rest. The categories combined with the modes of pure sensibility, or
3394with one another, afford a great number of deduced a priori conceptions;
3395a complete enumeration of which would be a useful and not unpleasant,
3396but in this place a perfectly dispensable, occupation.
3397
3398I purposely omit the definitions of the categories in this treatise.
3399I shall analyse these conceptions only so far as is necessary for the
3400doctrine of method, which is to form a part of this critique. In
3401a system of pure reason, definitions of them would be with justice
3402demanded of me, but to give them here would only bide from our view
3403the main aim of our investigation, at the same time raising doubts and
3404objections, the consideration of which, without injustice to our main
3405purpose, may be very well postponed till another opportunity. Meanwhile,
3406it ought to be sufficiently clear, from the little we have already said
3407on this subject, that the formation of a complete vocabulary of pure
3408conceptions, accompanied by all the requisite explanations, is not only
3409a possible, but an easy undertaking. The compartments already exist;
3410it is only necessary to fill them up; and a systematic topic like the
3411present, indicates with perfect precision the proper place to which each
3412conception belongs, while it readily points out any that have not yet
3413been filled up.
3414
3415
3416
3417SS 7.
3418
3419Our table of the categories suggests considerations of some importance,
3420which may perhaps have significant results in regard to the scientific
3421form of all rational cognitions. For, that this table is useful in the
3422theoretical part of philosophy, nay, indispensable for the sketching
3423of the complete plan of a science, so far as that science rests upon
3424conceptions a priori, and for dividing it mathematically, according to
3425fixed principles, is most manifest from the fact that it contains all
3426the elementary conceptions of the understanding, nay, even the form of a
3427system of these in the understanding itself, and consequently indicates
3428all the momenta, and also the internal arrangement of a projected
3429speculative science, as I have elsewhere shown. [Footnote: In the
3430Metaphysical Principles of Natural Science.] Here follow some of these
3431observations.
3432
3433I. This table, which contains four classes of conceptions of the
3434understanding, may, in the first instance, be divided into two classes,
3435the first of which relates to objects of intuition--pure as well as
3436empirical; the second, to the existence of these objects, either in
3437relation to one another, or to the understanding.
3438
3439The former of these classes of categories I would entitle the
3440mathematical, and the latter the dynamical categories. The former, as we
3441see, has no correlates; these are only to be found in the second
3442class. This difference must have a ground in the nature of the human
3443understanding.
3444
3445II. The number of the categories in each class is always the same,
3446namely, three--a fact which also demands some consideration, because
3447in all other cases division a priori through conceptions is necessarily
3448dichotomy. It is to be added, that the third category in each triad
3449always arises from the combination of the second with the first.
3450
3451Thus totality is nothing else but plurality contemplated as unity;
3452limitation is merely reality conjoined with negation; community is the
3453causality of a substance, reciprocally determining, and determined by
3454other substances; and finally, necessity is nothing but existence,
3455which is given through the possibility itself. Let it not be supposed,
3456however, that the third category is merely a deduced, and not a
3457primitive conception of the pure understanding. For the conjunction of
3458the first and second, in order to produce the third conception, requires
3459a particular function of the understanding, which is by no means
3460identical with those which are exercised in the first and second. Thus,
3461the conception of a number (which belongs to the category of totality)
3462is not always possible, where the conceptions of multitude and unity
3463exist (for example, in the representation of the infinite). Or, if I
3464conjoin the conception of a cause with that of a substance, it does not
3465follow that the conception of influence, that is, how one substance can
3466be the cause of something in another substance, will be understood from
3467that. Thus it is evident that a particular act of the understanding is
3468here necessary; and so in the other instances.
3469
3470III. With respect to one category, namely, that of community, which is
3471found in the third class, it is not so easy as with the others to
3472detect its accordance with the form of the disjunctive judgement which
3473corresponds to it in the table of the logical functions.
3474
3475In order to assure ourselves of this accordance, we must observe that in
3476every disjunctive judgement, the sphere of the judgement (that is,
3477the complex of all that is contained in it) is represented as a whole
3478divided into parts; and, since one part cannot be contained in the
3479other, they are cogitated as co-ordinated with, not subordinated to each
3480other, so that they do not determine each other unilaterally, as in a
3481linear series, but reciprocally, as in an aggregate--(if one member of
3482the division is posited, all the rest are excluded; and conversely).
3483
3484Now a like connection is cogitated in a whole of things; for one thing
3485is not subordinated, as effect, to another as cause of its existence,
3486but, on the contrary, is co-ordinated contemporaneously and
3487reciprocally, as a cause in relation to the determination of the others
3488(for example, in a body--the parts of which mutually attract and repel
3489each other). And this is an entirely different kind of connection from
3490that which we find in the mere relation of the cause to the effect (the
3491principle to the consequence), for in such a connection the consequence
3492does not in its turn determine the principle, and therefore does not
3493constitute, with the latter, a whole--just as the Creator does not with
3494the world make up a whole. The process of understanding by which it
3495represents to itself the sphere of a divided conception, is employed
3496also when we think of a thing as divisible; and in the same manner as
3497the members of the division in the former exclude one another, and yet
3498are connected in one sphere, so the understanding represents to itself
3499the parts of the latter, as having--each of them--an existence (as
3500substances), independently of the others, and yet as united in one
3501whole.
3502
3503
3504
3505SS 8.
3506
3507In the transcendental philosophy of the ancients there exists one more
3508leading division, which contains pure conceptions of the understanding,
3509and which, although not numbered among the categories, ought, according
3510to them, as conceptions a priori, to be valid of objects. But in this
3511case they would augment the number of the categories; which cannot
3512be. These are set forth in the proposition, so renowned among the
3513schoolmen--"Quodlibet ens est UNUM, VERUM, BONUM." Now, though the
3514inferences from this principle were mere tautological propositions,
3515and though it is allowed only by courtesy to retain a place in modern
3516metaphysics, yet a thought which maintained itself for such a length
3517of time, however empty it seems to be, deserves an investigation of its
3518origin, and justifies the conjecture that it must be grounded in some
3519law of the understanding, which, as is often the case, has only been
3520erroneously interpreted. These pretended transcendental predicates are,
3521in fact, nothing but logical requisites and criteria of all cognition
3522of objects, and they employ, as the basis for this cognition, the
3523categories of quantity, namely, unity, plurality, and totality. But
3524these, which must be taken as material conditions, that is, as belonging
3525to the possibility of things themselves, they employed merely in a
3526formal signification, as belonging to the logical requisites of all
3527cognition, and yet most unguardedly changed these criteria of thought
3528into properties of objects, as things in themselves. Now, in every
3529cognition of an object, there is unity of conception, which may be
3530called qualitative unity, so far as by this term we understand only the
3531unity in our connection of the manifold; for example, unity of the theme
3532in a play, an oration, or a story. Secondly, there is truth in respect
3533of the deductions from it. The more true deductions we have from a given
3534conception, the more criteria of its objective reality. This we might
3535call the qualitative plurality of characteristic marks, which belong
3536to a conception as to a common foundation, but are not cogitated as a
3537quantity in it. Thirdly, there is perfection--which consists in this,
3538that the plurality falls back upon the unity of the conception, and
3539accords completely with that conception and with no other. This we may
3540denominate qualitative completeness. Hence it is evident that these
3541logical criteria of the possibility of cognition are merely the three
3542categories of quantity modified and transformed to suit an unauthorized
3543manner of applying them. That is to say, the three categories, in
3544which the unity in the production of the quantum must be homogeneous
3545throughout, are transformed solely with a view to the connection of
3546heterogeneous parts of cognition in one act of consciousness, by
3547means of the quality of the cognition, which is the principle of that
3548connection. Thus the criterion of the possibility of a conception
3549(not of its object) is the definition of it, in which the unity of the
3550conception, the truth of all that may be immediately deduced from it,
3551and finally, the completeness of what has been thus deduced, constitute
3552the requisites for the reproduction of the whole conception. Thus also,
3553the criterion or test of an hypothesis is the intelligibility of the
3554received principle of explanation, or its unity (without help from any
3555subsidiary hypothesis)--the truth of our deductions from it (consistency
3556with each other and with experience)--and lastly, the completeness of
3557the principle of the explanation of these deductions, which refer
3558to neither more nor less than what was admitted in the hypothesis,
3559restoring analytically and a posteriori, what was cogitated
3560synthetically and a priori. By the conceptions, therefore, of unity,
3561truth, and perfection, we have made no addition to the transcendental
3562table of the categories, which is complete without them. We have, on
3563the contrary, merely employed the three categories of quantity, setting
3564aside their application to objects of experience, as general logical
3565laws of the consistency of cognition with itself.
3566
3567
3568
3569CHAPTER II Of the Deduction of the Pure Conceptions of the
3570Understanding.
3571
3572
3573SS 9. SECTION I Of the Principles of a Transcendental Deduction in
3574general.
3575
3576Teachers of jurisprudence, when speaking of rights and claims,
3577distinguish in a cause the question of right (quid juris) from the
3578question of fact (quid facti), and while they demand proof of both, they
3579give to the proof of the former, which goes to establish right or claim
3580in law, the name of deduction. Now we make use of a great number of
3581empirical conceptions, without opposition from any one; and consider
3582ourselves, even without any attempt at deduction, justified in attaching
3583to them a sense, and a supposititious signification, because we have
3584always experience at hand to demonstrate their objective reality. There
3585exist also, however, usurped conceptions, such as fortune, fate, which
3586circulate with almost universal indulgence, and yet are occasionally
3587challenged by the question, "quid juris?" In such cases, we have great
3588difficulty in discovering any deduction for these terms, inasmuch as we
3589cannot produce any manifest ground of right, either from experience or
3590from reason, on which the claim to employ them can be founded.
3591
3592Among the many conceptions, which make up the very variegated web of
3593human cognition, some are destined for pure use a priori, independent
3594of all experience; and their title to be so employed always requires
3595a deduction, inasmuch as, to justify such use of them, proofs from
3596experience are not sufficient; but it is necessary to know how these
3597conceptions can apply to objects without being derived from experience.
3598I term, therefore, an examination of the manner in which conceptions can
3599apply a priori to objects, the transcendental deduction of conceptions,
3600and I distinguish it from the empirical deduction, which indicates the
3601mode in which conception is obtained through experience and reflection
3602thereon; consequently, does not concern itself with the right, but only
3603with the fact of our obtaining conceptions in such and such a manner. We
3604have already seen that we are in possession of two perfectly different
3605kinds of conceptions, which nevertheless agree with each other in this,
3606that they both apply to objects completely a priori. These are
3607the conceptions of space and time as forms of sensibility, and the
3608categories as pure conceptions of the understanding. To attempt an
3609empirical deduction of either of these classes would be labour in vain,
3610because the distinguishing characteristic of their nature consists in
3611this, that they apply to their objects, without having borrowed anything
3612from experience towards the representation of them. Consequently, if
3613a deduction of these conceptions is necessary, it must always be
3614transcendental.
3615
3616Meanwhile, with respect to these conceptions, as with respect to all our
3617cognition, we certainly may discover in experience, if not the principle
3618of their possibility, yet the occasioning causes of their production. It
3619will be found that the impressions of sense give the first occasion
3620for bringing into action the whole faculty of cognition, and for the
3621production of experience, which contains two very dissimilar elements,
3622namely, a matter for cognition, given by the senses, and a certain form
3623for the arrangement of this matter, arising out of the inner fountain
3624of pure intuition and thought; and these, on occasion given by sensuous
3625impressions, are called into exercise and produce conceptions. Such
3626an investigation into the first efforts of our faculty of cognition to
3627mount from particular perceptions to general conceptions is undoubtedly
3628of great utility; and we have to thank the celebrated Locke for having
3629first opened the way for this inquiry. But a deduction of the pure a
3630priori conceptions of course never can be made in this way, seeing that,
3631in regard to their future employment, which must be entirely independent
3632of experience, they must have a far different certificate of birth
3633to show from that of a descent from experience. This attempted
3634physiological derivation, which cannot properly be called deduction,
3635because it relates merely to a quaestio facti, I shall entitle an
3636explanation of the possession of a pure cognition. It is therefore
3637manifest that there can only be a transcendental deduction of these
3638conceptions and by no means an empirical one; also, that all attempts
3639at an empirical deduction, in regard to pure a priori conceptions, are
3640vain, and can only be made by one who does not understand the altogether
3641peculiar nature of these cognitions.
3642
3643But although it is admitted that the only possible deduction of pure
3644a priori cognition is a transcendental deduction, it is not, for
3645that reason, perfectly manifest that such a deduction is absolutely
3646necessary. We have already traced to their sources the conceptions of
3647space and time, by means of a transcendental deduction, and we have
3648explained and determined their objective validity a priori. Geometry,
3649nevertheless, advances steadily and securely in the province of pure
3650a priori cognitions, without needing to ask from philosophy any
3651certificate as to the pure and legitimate origin of its fundamental
3652conception of space. But the use of the conception in this science
3653extends only to the external world of sense, the pure form of the
3654intuition of which is space; and in this world, therefore, all
3655geometrical cognition, because it is founded upon a priori intuition,
3656possesses immediate evidence, and the objects of this cognition are
3657given a priori (as regards their form) in intuition by and through the
3658cognition itself. With the pure conceptions of understanding, on the
3659contrary, commences the absolute necessity of seeking a transcendental
3660deduction, not only of these conceptions themselves, but likewise of
3661space, because, inasmuch as they make affirmations concerning objects
3662not by means of the predicates of intuition and sensibility, but of pure
3663thought a priori, they apply to objects without any of the conditions
3664of sensibility. Besides, not being founded on experience, they are not
3665presented with any object in a priori intuition upon which, antecedently
3666to experience, they might base their synthesis. Hence results, not only
3667doubt as to the objective validity and proper limits of their use, but
3668that even our conception of space is rendered equivocal; inasmuch as we
3669are very ready with the aid of the categories, to carry the use of this
3670conception beyond the conditions of sensuous intuition--and, for this
3671reason, we have already found a transcendental deduction of it needful.
3672The reader, then, must be quite convinced of the absolute necessity of
3673a transcendental deduction, before taking a single step in the field of
3674pure reason; because otherwise he goes to work blindly, and after he
3675has wondered about in all directions, returns to the state of utter
3676ignorance from which he started. He ought, moreover, clearly to
3677recognize beforehand the unavoidable difficulties in his undertaking,
3678so that he may not afterwards complain of the obscurity in which the
3679subject itself is deeply involved, or become too soon impatient of
3680the obstacles in his path; because we have a choice of only two
3681things--either at once to give up all pretensions to knowledge
3682beyond the limits of possible experience, or to bring this critical
3683investigation to completion.
3684
3685We have been able, with very little trouble, to make it comprehensible
3686how the conceptions of space and time, although a priori cognitions,
3687must necessarily apply to external objects, and render a synthetical
3688cognition of these possible, independently of all experience. For
3689inasmuch as only by means of such pure form of sensibility an object can
3690appear to us, that is, be an object of empirical intuition, space and
3691time are pure intuitions, which contain a priori the condition of the
3692possibility of objects as phenomena, and an a priori synthesis in these
3693intuitions possesses objective validity.
3694
3695On the other hand, the categories of the understanding do not represent
3696the conditions under which objects are given to us in intuition; objects
3697can consequently appear to us without necessarily connecting themselves
3698with these, and consequently without any necessity binding on the
3699understanding to contain a priori the conditions of these objects. Thus
3700we find ourselves involved in a difficulty which did not present itself
3701in the sphere of sensibility, that is to say, we cannot discover how the
3702subjective conditions of thought can have objective validity, in other
3703words, can become conditions of the possibility of all cognition of
3704objects; for phenomena may certainly be given to us in intuition without
3705any help from the functions of the understanding. Let us take, for
3706example, the conception of cause, which indicates a peculiar kind of
3707synthesis, namely, that with something, A, something entirely different,
3708B, is connected according to a law. It is not a priori manifest why
3709phenomena should contain anything of this kind (we are of course
3710debarred from appealing for proof to experience, for the objective
3711validity of this conception must be demonstrated a priori), and it hence
3712remains doubtful a priori, whether such a conception be not quite void
3713and without any corresponding object among phenomena. For that objects
3714of sensuous intuition must correspond to the formal conditions of
3715sensibility existing a priori in the mind is quite evident, from the
3716fact that without these they could not be objects for us; but that they
3717must also correspond to the conditions which understanding requires for
3718the synthetical unity of thought is an assertion, the grounds for
3719which are not so easily to be discovered. For phenomena might be so
3720constituted as not to correspond to the conditions of the unity of
3721thought; and all things might lie in such confusion that, for example,
3722nothing could be met with in the sphere of phenomena to suggest a law of
3723synthesis, and so correspond to the conception of cause and effect;
3724so that this conception would be quite void, null, and without
3725significance. Phenomena would nevertheless continue to present objects
3726to our intuition; for mere intuition does not in any respect stand in
3727need of the functions of thought.
3728
3729If we thought to free ourselves from the labour of these investigations
3730by saying: "Experience is constantly offering us examples of the
3731relation of cause and effect in phenomena, and presents us with abundant
3732opportunity of abstracting the conception of cause, and so at the same
3733time of corroborating the objective validity of this conception"; we
3734should in this case be overlooking the fact, that the conception of
3735cause cannot arise in this way at all; that, on the contrary, it must
3736either have an a priori basis in the understanding, or be rejected as a
3737mere chimera. For this conception demands that something, A, should
3738be of such a nature that something else, B, should follow from it
3739necessarily, and according to an absolutely universal law. We may
3740certainly collect from phenomena a law, according to which this or that
3741usually happens, but the element of necessity is not to be found in it.
3742Hence it is evident that to the synthesis of cause and effect belongs a
3743dignity, which is utterly wanting in any empirical synthesis; for it is
3744no mere mechanical synthesis, by means of addition, but a dynamical one;
3745that is to say, the effect is not to be cogitated as merely annexed to
3746the cause, but as posited by and through the cause, and resulting from
3747it. The strict universality of this law never can be a characteristic
3748of empirical laws, which obtain through induction only a comparative
3749universality, that is, an extended range of practical application. But
3750the pure conceptions of the understanding would entirely lose all their
3751peculiar character, if we treated them merely as the productions of
3752experience.
3753
3754
3755
3756SS 10. Transition to the Transcendental Deduction of the Categories.
3757
3758There are only two possible ways in which synthetical representation and
3759its objects can coincide with and relate necessarily to each other,
3760and, as it were, meet together. Either the object alone makes the
3761representation possible, or the representation alone makes the object
3762possible. In the former case, the relation between them is only
3763empirical, and an a priori representation is impossible. And this is the
3764case with phenomena, as regards that in them which is referable to mere
3765sensation. In the latter case--although representation alone (for of its
3766causality, by means of the will, we do not here speak) does not produce
3767the object as to its existence, it must nevertheless be a priori
3768determinative in regard to the object, if it is only by means of the
3769representation that we can cognize anything as an object. Now there
3770are only two conditions of the possibility of a cognition of objects;
3771firstly, intuition, by means of which the object, though only as
3772phenomenon, is given; secondly, conception, by means of which the object
3773which corresponds to this intuition is thought. But it is evident from
3774what has been said on aesthetic that the first condition, under which
3775alone objects can be intuited, must in fact exist, as a formal basis for
3776them, a priori in the mind. With this formal condition of sensibility,
3777therefore, all phenomena necessarily correspond, because it is
3778only through it that they can be phenomena at all; that is, can be
3779empirically intuited and given. Now the question is whether there do
3780not exist, a priori in the mind, conceptions of understanding also, as
3781conditions under which alone something, if not intuited, is yet thought
3782as object. If this question be answered in the affirmative, it follows
3783that all empirical cognition of objects is necessarily conformable to
3784such conceptions, since, if they are not presupposed, it is impossible
3785that anything can be an object of experience. Now all experience
3786contains, besides the intuition of the senses through which an object
3787is given, a conception also of an object that is given in intuition.
3788Accordingly, conceptions of objects in general must lie as a
3789priori conditions at the foundation of all empirical cognition; and
3790consequently, the objective validity of the categories, as a priori
3791conceptions, will rest upon this, that experience (as far as regards the
3792form of thought) is possible only by their means. For in that case they
3793apply necessarily and a priori to objects of experience, because only
3794through them can an object of experience be thought.
3795
3796The whole aim of the transcendental deduction of all a priori
3797conceptions is to show that these conceptions are a priori conditions
3798of the possibility of all experience. Conceptions which afford us the
3799objective foundation of the possibility of experience are for that very
3800reason necessary. But the analysis of the experiences in which they are
3801met with is not deduction, but only an illustration of them, because
3802from experience they could never derive the attribute of necessity.
3803Without their original applicability and relation to all possible
3804experience, in which all objects of cognition present themselves, the
3805relation of the categories to objects, of whatever nature, would be
3806quite incomprehensible.
3807
3808The celebrated Locke, for want of due reflection on these points, and
3809because he met with pure conceptions of the understanding in experience,
3810sought also to deduce them from experience, and yet proceeded so
3811inconsequently as to attempt, with their aid, to arrive it cognitions
3812which lie far beyond the limits of all experience. David Hume perceived
3813that, to render this possible, it was necessary that the conceptions
3814should have an a priori origin. But as he could not explain how it was
3815possible that conceptions which are not connected with each other in the
3816understanding must nevertheless be thought as necessarily connected in
3817the object--and it never occurred to him that the understanding itself
3818might, perhaps, by means of these conceptions, be the author of the
3819experience in which its objects were presented to it--he was forced
3820to drive these conceptions from experience, that is, from a subjective
3821necessity arising from repeated association of experiences erroneously
3822considered to be objective--in one word, from habit. But he proceeded
3823with perfect consequence and declared it to be impossible, with such
3824conceptions and the principles arising from them, to overstep the limits
3825of experience. The empirical derivation, however, which both of these
3826philosophers attributed to these conceptions, cannot possibly be
3827reconciled with the fact that we do possess scientific a priori
3828cognitions, namely, those of pure mathematics and general physics.
3829
3830The former of these two celebrated men opened a wide door to
3831extravagance--(for if reason has once undoubted right on its side,
3832it will not allow itself to be confined to set limits, by vague
3833recommendations of moderation); the latter gave himself up entirely
3834to scepticism--a natural consequence, after having discovered, as he
3835thought, that the faculty of cognition was not trustworthy. We now
3836intend to make a trial whether it be not possible safely to conduct
3837reason between these two rocks, to assign her determinate limits, and
3838yet leave open for her the entire sphere of her legitimate activity.
3839
3840I shall merely premise an explanation of what the categories are. They
3841are conceptions of an object in general, by means of which its intuition
3842is contemplated as determined in relation to one of the logical
3843functions of judgement. The following will make this plain. The function
3844of the categorical judgement is that of the relation of subject to
3845predicate; for example, in the proposition: "All bodies are divisible."
3846But in regard to the merely logical use of the understanding, it still
3847remains undetermined to which Of these two conceptions belongs the
3848function Of subject and to which that of predicate. For we could also
3849say: "Some divisible is a body." But the category of substance, when
3850the conception of a body is brought under it, determines that; and its
3851empirical intuition in experience must be contemplated always as subject
3852and never as mere predicate. And so with all the other categories.
3853
3854
3855
3856SS 11. SECTION II Transcendental Deduction of the pure Conceptions of
3857the Understanding.
3858
3859Of the Possibility of a Conjunction of the manifold representations
3860given by Sense.
3861
3862The manifold content in our representations can be given in an intuition
3863which is merely sensuous--in other words, is nothing but susceptibility;
3864and the form of this intuition can exist a priori in our faculty of
3865representation, without being anything else but the mode in which the
3866subject is affected. But the conjunction (conjunctio) of a manifold in
3867intuition never can be given us by the senses; it cannot therefore
3868be contained in the pure form of sensuous intuition, for it is a
3869spontaneous act of the faculty of representation. And as we must, to
3870distinguish it from sensibility, entitle this faculty understanding; so
3871all conjunction whether conscious or unconscious, be it of the manifold
3872in intuition, sensuous or non-sensuous, or of several conceptions--is
3873an act of the understanding. To this act we shall give the general
3874appellation of synthesis, thereby to indicate, at the same time, that
3875we cannot represent anything as conjoined in the object without having
3876previously conjoined it ourselves. Of all mental notions, that of
3877conjunction is the only one which cannot be given through objects, but
3878can be originated only by the subject itself, because it is an act of
3879its purely spontaneous activity. The reader will easily enough perceive
3880that the possibility of conjunction must be grounded in the very nature
3881of this act, and that it must be equally valid for all conjunction, and
3882that analysis, which appears to be its contrary, must, nevertheless,
3883always presuppose it; for where the understanding has not previously
3884conjoined, it cannot dissect or analyse, because only as conjoined by
3885it, must that which is to be analysed have been given to our faculty of
3886representation.
3887
3888But the conception of conjunction includes, besides the conception of
3889the manifold and of the synthesis of it, that of the unity of it also.
3890Conjunction is the representation of the synthetical unity of the
3891manifold.* This idea of unity, therefore, cannot arise out of that of
3892conjunction; much rather does that idea, by combining itself with the
3893representation of the manifold, render the conception of conjunction
3894possible. This unity, which a priori precedes all conceptions of
3895conjunction, is not the category of unity (SS 6); for all the categories
3896are based upon logical functions of judgement, and in these functions we
3897already have conjunction, and consequently unity of given conceptions.
3898It is therefore evident that the category of unity presupposes
3899conjunction. We must therefore look still higher for this unity (as
3900qualitative, SS 8), in that, namely, which contains the ground of the
3901unity of diverse conceptions in judgements, the ground, consequently, of
3902the possibility of the existence of the understanding, even in regard to
3903its logical use.
3904
3905 [*Footnote: Whether the representations are in themselves
3906 identical, and consequently whether one can be thought
3907 analytically by means of and through the other, is a
3908 question which we need not at present consider. Our
3909 Consciousness of the one, when we speak of the manifold, is
3910 always distinguishable from our consciousness of the other;
3911 and it is only respecting the synthesis of this (possible)
3912 consciousness that we here treat.]
3913
3914
3915
3916SS 12. Of the Originally Synthetical Unity of Apperception.
3917
3918The "I think" must accompany all my representations, for otherwise
3919something would be represented in me which could not be thought; in
3920other words, the representation would either be impossible, or at least
3921be, in relation to me, nothing. That representation which can be given
3922previously to all thought is called intuition. All the diversity or
3923manifold content of intuition, has, therefore, a necessary relation to
3924the "I think," in the subject in which this diversity is found. But this
3925representation, "I think," is an act of spontaneity; that is to say,
3926it cannot be regarded as belonging to mere sensibility. I call it pure
3927apperception, in order to distinguish it from empirical; or primitive
3928apperception, because it is self-consciousness which, whilst it gives
3929birth to the representation "I think," must necessarily be capable of
3930accompanying all our representations. It is in all acts of consciousness
3931one and the same, and unaccompanied by it, no representation can exist
3932for me. The unity of this apperception I call the transcendental unity
3933of self-consciousness, in order to indicate the possibility of a priori
3934cognition arising from it. For the manifold representations which are
3935given in an intuition would not all of them be my representations,
3936if they did not all belong to one self-consciousness, that is, as my
3937representations (even although I am not conscious of them as such), they
3938must conform to the condition under which alone they can exist together
3939in a common self-consciousness, because otherwise they would not all
3940without exception belong to me. From this primitive conjunction follow
3941many important results.
3942
3943For example, this universal identity of the apperception of the manifold
3944given in intuition contains a synthesis of representations and is
3945possible only by means of the consciousness of this synthesis. For the
3946empirical consciousness which accompanies different representations
3947is in itself fragmentary and disunited, and without relation to the
3948identity of the subject. This relation, then, does not exist because I
3949accompany every representation with consciousness, but because I join
3950one representation to another, and am conscious of the synthesis of
3951them. Consequently, only because I can connect a variety of given
3952representations in one consciousness, is it possible that I
3953can represent to myself the identity of consciousness in these
3954representations; in other words, the analytical unity of apperception
3955is possible only under the presupposition of a synthetical unity.* The
3956thought, "These representations given in intuition belong all of them
3957to me," is accordingly just the same as, "I unite them in one
3958self-consciousness, or can at least so unite them"; and although
3959this thought is not itself the consciousness of the synthesis of
3960representations, it presupposes the possibility of it; that is to
3961say, for the reason alone that I can comprehend the variety of my
3962representations in one consciousness, do I call them my representations,
3963for otherwise I must have as many-coloured and various a self as are
3964the representations of which I am conscious. Synthetical unity of the
3965manifold in intuitions, as given a priori, is therefore the foundation
3966of the identity of apperception itself, which antecedes a priori all
3967determinate thought. But the conjunction of representations into a
3968conception is not to be found in objects themselves, nor can it be,
3969as it were, borrowed from them and taken up into the understanding by
3970perception, but it is on the contrary an operation of the understanding
3971itself, which is nothing more than the faculty of conjoining a priori
3972and of bringing the variety of given representations under the unity of
3973apperception. This principle is the highest in all human cognition.
3974
3975 [*Footnote: All general conceptions--as such--depend, for
3976 their existence, on the analytical unity of consciousness.
3977 For example, when I think of red in general, I thereby think
3978 to myself a property which (as a characteristic mark) can be
3979 discovered somewhere, or can be united with other
3980 representations; consequently, it is only by means of a
3981 forethought possible synthetical unity that I can think to
3982 myself the analytical. A representation which is cogitated
3983 as common to different representations, is regarded as
3984 belonging to such as, besides this common representation,
3985 contain something different; consequently it must be
3986 previously thought in synthetical unity with other although
3987 only possible representations, before I can think in it the
3988 analytical unity of consciousness which makes it a conceptas
3989 communis. And thus the synthetical unity of apperception is
3990 the highest point with which we must connect every operation
3991 of the understanding, even the whole of logic, and after it
3992 our transcendental philosophy; indeed, this faculty is the
3993 understanding itself.]
3994
3995
3996This fundamental principle of the necessary unity of apperception is
3997indeed an identical, and therefore analytical, proposition; but it
3998nevertheless explains the necessity for a synthesis of the manifold
3999given in an intuition, without which the identity of self-consciousness
4000would be incogitable. For the ego, as a simple representation, presents
4001us with no manifold content; only in intuition, which is quite different
4002from the representation ego, can it be given us, and by means of
4003conjunction it is cogitated in one self-consciousness. An understanding,
4004in which all the manifold should be given by means of consciousness
4005itself, would be intuitive; our understanding can only think and must
4006look for its intuition to sense. I am, therefore, conscious of my
4007identical self, in relation to all the variety of representations given
4008to me in an intuition, because I call all of them my representations. In
4009other words, I am conscious myself of a necessary a priori synthesis of
4010my representations, which is called the original synthetical unity of
4011apperception, under which rank all the representations presented to me,
4012but that only by means of a synthesis.
4013
4014
4015
4016SS 13. The Principle of the Synthetical Unity of Apperception is the
4017highest Principle of all exercise of the Understanding.
4018
4019The supreme principle of the possibility of all intuition in relation to
4020sensibility was, according to our transcendental aesthetic, that all the
4021manifold in intuition be subject to the formal conditions of space and
4022time. The supreme principle of the possibility of it in relation to the
4023understanding is that all the manifold in it be subject to conditions
4024of the originally synthetical unity or apperception.* To the former
4025of these two principles are subject all the various representations of
4026intuition, in so far as they are given to us; to the latter, in so far
4027as they must be capable of conjunction in one consciousness; for
4028without this nothing can be thought or cognized, because the given
4029representations would not have in common the act Of the apperception "I
4030think" and therefore could not be connected in one self-consciousness.
4031
4032 [*Footnote: Space and time, and all portions thereof, are
4033 intuitions; consequently are, with a manifold for their
4034 content, single representations. (See the Transcendental
4035 Aesthetic.) Consequently, they are not pure conceptions, by
4036 means of which the same consciousness is found in a great
4037 number of representations; but, on the contrary, they are
4038 many representations contained in one, the consciousness of
4039 which is, so to speak, compounded. The unity of
4040 consciousness is nevertheless synthetical and, therefore,
4041 primitive. From this peculiar character of consciousness
4042 follow many important consequences. (See SS 21.)]
4043
4044
4045Understanding is, to speak generally, the faculty Of cognitions. These
4046consist in the determined relation of given representation to an object.
4047But an object is that, in the conception of which the manifold in a
4048given intuition is united. Now all union of representations requires
4049unity of consciousness in the synthesis of them. Consequently, it is
4050the unity of consciousness alone that constitutes the possibility of
4051representations relating to an object, and therefore of their objective
4052validity, and of their becoming cognitions, and consequently, the
4053possibility of the existence of the understanding itself.
4054
4055The first pure cognition of understanding, then, upon which is founded
4056all its other exercise, and which is at the same time perfectly
4057independent of all conditions of mere sensuous intuition, is the
4058principle of the original synthetical unity of apperception. Thus the
4059mere form of external sensuous intuition, namely, space, affords us,
4060per se, no cognition; it merely contributes the manifold in a priori
4061intuition to a possible cognition. But, in order to cognize something
4062in space (for example, a line), I must draw it, and thus produce
4063synthetically a determined conjunction of the given manifold, so that
4064the unity of this act is at the same time the unity of consciousness
4065(in the conception of a line), and by this means alone is an object (a
4066determinate space) cognized. The synthetical unity of consciousness
4067is, therefore, an objective condition of all cognition, which I do
4068not merely require in order to cognize an object, but to which every
4069intuition must necessarily be subject, in order to become an object for
4070me; because in any other way, and without this synthesis, the manifold
4071in intuition could not be united in one consciousness.
4072
4073This proposition is, as already said, itself analytical, although it
4074constitutes the synthetical unity, the condition of all thought; for
4075it states nothing more than that all my representations in any given
4076intuition must be subject to the condition which alone enables me to
4077connect them, as my representation with the identical self, and so to
4078unite them synthetically in one apperception, by means of the general
4079expression, "I think."
4080
4081But this principle is not to be regarded as a principle for every
4082possible understanding, but only for the understanding by means of whose
4083pure apperception in the thought I am, no manifold content is given. The
4084understanding or mind which contained the manifold in intuition, in and
4085through the act itself of its own self-consciousness, in other words, an
4086understanding by and in the representation of which the objects of
4087the representation should at the same time exist, would not require a
4088special act of synthesis of the manifold as the condition of the unity
4089of its consciousness, an act of which the human understanding, which
4090thinks only and cannot intuite, has absolute need. But this principle is
4091the first principle of all the operations of our understanding, so that
4092we cannot form the least conception of any other possible understanding,
4093either of one such as should be itself intuition, or possess a sensuous
4094intuition, but with forms different from those of space and time.
4095
4096
4097
4098SS 14. What Objective Unity of Self-consciousness is.
4099
4100It is by means of the transcendental unity of apperception that all
4101the manifold, given in an intuition is united into a conception of
4102the object. On this account it is called objective, and must be
4103distinguished from the subjective unity of consciousness, which is a
4104determination of the internal sense, by means of which the said manifold
4105in intuition is given empirically to be so united. Whether I can be
4106empirically conscious of the manifold as coexistent or as successive,
4107depends upon circumstances, or empirical conditions. Hence the empirical
4108unity of consciousness by means of association of representations,
4109itself relates to a phenomenal world and is wholly contingent. On the
4110contrary, the pure form of intuition in time, merely as an intuition,
4111which contains a given manifold, is subject to the original unity of
4112consciousness, and that solely by means of the necessary relation of
4113the manifold in intuition to the "I think," consequently by means of
4114the pure synthesis of the understanding, which lies a priori at the
4115foundation of all empirical synthesis. The transcendental unity of
4116apperception is alone objectively valid; the empirical which we do not
4117consider in this essay, and which is merely a unity deduced from the
4118former under given conditions in concreto, possesses only subjective
4119validity. One person connects the notion conveyed in a word with one
4120thing, another with another thing; and the unity of consciousness in
4121that which is empirical, is, in relation to that which is given by
4122experience, not necessarily and universally valid.
4123
4124
4125
4126SS 15. The Logical Form of all Judgements consists in the Objective
4127Unity of Apperception of the Conceptions contained therein.
4128
4129I could never satisfy myself with the definition which logicians give of
4130a judgement. It is, according to them, the representation of a relation
4131between two conceptions. I shall not dwell here on the faultiness of
4132this definition, in that it suits only for categorical and not for
4133hypothetical or disjunctive judgements, these latter containing a
4134relation not of conceptions but of judgements themselves--a blunder from
4135which many evil results have followed.* It is more important for our
4136present purpose to observe, that this definition does not determine in
4137what the said relation consists.
4138
4139 [*Footnote: The tedious doctrine of the four syllogistic
4140 figures concerns only categorical syllogisms; and although
4141 it is nothing more than an artifice by surreptitiously
4142 introducing immediate conclusions (consequentiae immediatae)
4143 among the premises of a pure syllogism, to give ism' give
4144 rise to an appearance of more modes of drawing a conclusion
4145 than that in the first figure, the artifice would not have
4146 had much success, had not its authors succeeded in bringing
4147 categorical judgements into exclusive respect, as those to
4148 which all others must be referred--a doctrine, however,
4149 which, according to SS 5, is utterly false.]
4150
4151
4152But if I investigate more closely the relation of given cognitions in
4153every judgement, and distinguish it, as belonging to the understanding,
4154from the relation which is produced according to laws of the
4155reproductive imagination (which has only subjective validity), I find
4156that judgement is nothing but the mode of bringing given cognitions
4157under the objective unit of apperception. This is plain from our use
4158of the term of relation is in judgements, in order to distinguish the
4159objective unity of given representations from the subjective unity.
4160For this term indicates the relation of these representations to the
4161original apperception, and also their necessary unity, even although the
4162judgement is empirical, therefore contingent, as in the judgement: "All
4163bodies are heavy." I do not mean by this, that these representations
4164do necessarily belong to each other in empirical intuition, but that by
4165means of the necessary unity of appreciation they belong to each other
4166in the synthesis of intuitions, that is to say, they belong to each
4167other according to principles of the objective determination of all
4168our representations, in so far as cognition can arise from them,
4169these principles being all deduced from the main principle of the
4170transcendental unity of apperception. In this way alone can there arise
4171from this relation a judgement, that is, a relation which has objective
4172validity, and is perfectly distinct from that relation of the very same
4173representations which has only subjective validity--a relation, to wit,
4174which is produced according to laws of association. According to these
4175laws, I could only say: "When I hold in my hand or carry a body, I feel
4176an impression of weight"; but I could not say: "It, the body, is
4177heavy"; for this is tantamount to saying both these representations
4178are conjoined in the object, that is, without distinction as to the
4179condition of the subject, and do not merely stand together in my
4180perception, however frequently the perceptive act may be repeated.
4181
4182
4183
4184SS 16. All Sensuous Intuitions are subject to the Categories, as
4185Conditions under which alone the manifold Content of them can be united
4186in one Consciousness.
4187
4188The manifold content given in a sensuous intuition comes necessarily
4189under the original synthetical unity of apperception, because thereby
4190alone is the unity of intuition possible (SS 13). But that act of the
4191understanding, by which the manifold content of given representations
4192(whether intuitions or conceptions) is brought under one apperception,
4193is the logical function of judgements (SS 15). All the manifold,
4194therefore, in so far as it is given in one empirical intuition, is
4195determined in relation to one of the logical functions of judgement, by
4196means of which it is brought into union in one consciousness. Now the
4197categories are nothing else than these functions of judgement so far as
4198the manifold in a given intuition is determined in relation to them
4199(SS 9). Consequently, the manifold in a given intuition is necessarily
4200subject to the categories of the understanding.
4201
4202
4203
4204SS 17. Observation.
4205
4206The manifold in an intuition, which I call mine, is represented by means
4207of the synthesis of the understanding, as belonging to the necessary
4208unity of self-consciousness, and this takes place by means of the
4209category.* The category indicates accordingly that the empirical
4210consciousness of a given manifold in an intuition is subject to a
4211pure self-consciousness a priori, in the same manner as an empirical
4212intuition is subject to a pure sensuous intuition, which is also
4213a priori. In the above proposition, then, lies the beginning of a
4214deduction of the pure conceptions of the understanding. Now, as the
4215categories have their origin in the understanding alone, independently
4216of sensibility, I must in my deduction make abstraction of the mode in
4217which the manifold of an empirical intuition is given, in order to
4218fix my attention exclusively on the unity which is brought by the
4219understanding into the intuition by means of the category. In what
4220follows (SS 22), it will be shown, from the mode in which the empirical
4221intuition is given in the faculty of sensibility, that the unity which
4222belongs to it is no other than that which the category (according to SS
422316) imposes on the manifold in a given intuition, and thus, its a
4224priori validity in regard to all objects of sense being established, the
4225purpose of our deduction will be fully attained.
4226
4227 [*Footnote: The proof of this rests on the represented unity
4228 of intuition, by means of which an object is given, and
4229 which always includes in itself a synthesis of the manifold
4230 to be intuited, and also the relation of this latter to
4231 unity of apperception.]
4232
4233
4234But there is one thing in the above demonstration of which I could not
4235make abstraction, namely, that the manifold to be intuited must be given
4236previously to the synthesis of the understanding, and independently of
4237it. How this takes place remains here undetermined. For if I cogitate
4238an understanding which was itself intuitive (as, for example, a divine
4239understanding which should not represent given objects, but by whose
4240representation the objects themselves should be given or produced), the
4241categories would possess no significance in relation to such a faculty
4242of cognition. They are merely rules for an understanding, whose whole
4243power consists in thought, that is, in the act of submitting the
4244synthesis of the manifold which is presented to it in intuition from
4245a very different quarter, to the unity of apperception; a faculty,
4246therefore, which cognizes nothing per se, but only connects and
4247arranges the material of cognition, the intuition, namely, which must
4248be presented to it by means of the object. But to show reasons for this
4249peculiar character of our understandings, that it produces unity of
4250apperception a priori only by means of categories, and a certain kind
4251and number thereof, is as impossible as to explain why we are endowed
4252with precisely so many functions of judgement and no more, or why time
4253and space are the only forms of our intuition.
4254
4255
4256
4257SS 18. In Cognition, its Application to Objects of Experience is
4258the only legitimate use of the Category.
4259
4260To think an object and to cognize an object are by no means the same
4261thing. In cognition there are two elements: firstly, the conception,
4262whereby an object is cogitated (the category); and, secondly, the
4263intuition, whereby the object is given. For supposing that to the
4264conception a corresponding intuition could not be given, it would
4265still be a thought as regards its form, but without any object, and no
4266cognition of anything would be possible by means of it, inasmuch as, so
4267far as I knew, there existed and could exist nothing to which my
4268thought could be applied. Now all intuition possible to us is sensuous;
4269consequently, our thought of an object by means of a pure conception of
4270the understanding, can become cognition for us only in so far as this
4271conception is applied to objects of the senses. Sensuous intuition is
4272either pure intuition (space and time) or empirical intuition--of that
4273which is immediately represented in space and time by means of sensation
4274as real. Through the determination of pure intuition we obtain a priori
4275cognitions of objects, as in mathematics, but only as regards their form
4276as phenomena; whether there can exist things which must be intuited
4277in this form is not thereby established. All mathematical conceptions,
4278therefore, are not per se cognition, except in so far as we presuppose
4279that there exist things which can only be represented conformably to the
4280form of our pure sensuous intuition. But things in space and time
4281are given only in so far as they are perceptions (representations
4282accompanied with sensation), therefore only by empirical representation.
4283Consequently the pure conceptions of the understanding, even when
4284they are applied to intuitions a priori (as in mathematics), produce
4285cognition only in so far as these (and therefore the conceptions of the
4286understanding by means of them) can be applied to empirical intuitions.
4287Consequently the categories do not, even by means of pure intuition
4288afford us any cognition of things; they can only do so in so far as they
4289can be applied to empirical intuition. That is to say, the categories
4290serve only to render empirical cognition possible. But this is what
4291we call experience. Consequently, in cognition, their application to
4292objects of experience is the only legitimate use of the categories.
4293
4294
4295
4296SS 19.
4297
4298The foregoing proposition is of the utmost importance, for it determines
4299the limits of the exercise of the pure conceptions of the understanding
4300in regard to objects, just as transcendental aesthetic determined the
4301limits of the exercise of the pure form of our sensuous intuition.
4302Space and time, as conditions of the possibility of the presentation
4303of objects to us, are valid no further than for objects of sense,
4304consequently, only for experience. Beyond these limits they represent
4305to us nothing, for they belong only to sense, and have no reality apart
4306from it. The pure conceptions of the understanding are free from this
4307limitation, and extend to objects of intuition in general, be the
4308intuition like or unlike to ours, provided only it be sensuous, and not
4309intellectual. But this extension of conceptions beyond the range of our
4310intuition is of no advantage; for they are then mere empty conceptions
4311of objects, as to the possibility or impossibility of the existence of
4312which they furnish us with no means of discovery. They are mere forms
4313of thought, without objective reality, because we have no intuition to
4314which the synthetical unity of apperception, which alone the categories
4315contain, could be applied, for the purpose of determining an object. Our
4316sensuous and empirical intuition can alone give them significance and
4317meaning.
4318
4319If, then, we suppose an object of a non-sensuous intuition to be given
4320we can in that case represent it by all those predicates which are
4321implied in the presupposition that nothing appertaining to sensuous
4322intuition belongs to it; for example, that it is not extended, or in
4323space; that its duration is not time; that in it no change (the effect
4324of the determinations in time) is to be met with, and so on. But it
4325is no proper knowledge if I merely indicate what the intuition of the
4326object is not, without being able to say what is contained in it, for I
4327have not shown the possibility of an object to which my pure conception
4328of understanding could be applicable, because I have not been able to
4329furnish any intuition corresponding to it, but am only able to say that
4330our intuition is not valid for it. But the most important point is
4331this, that to a something of this kind not one category can be found
4332applicable. Take, for example, the conception of substance, that is,
4333something that can exist as subject, but never as mere predicate; in
4334regard to this conception I am quite ignorant whether there can really
4335be anything to correspond to such a determination of thought, if
4336empirical intuition did not afford me the occasion for its application.
4337But of this more in the sequel.
4338
4339
4340
4341SS 20. Of the Application of the Categories to Objects of the
4342Senses in general.
4343
4344The pure conceptions of the understanding apply to objects of intuition
4345in general, through the understanding alone, whether the intuition be
4346our own or some other, provided only it be sensuous, but are, for
4347this very reason, mere forms of thought, by means of which alone no
4348determined object can be cognized. The synthesis or conjunction of the
4349manifold in these conceptions relates, we have said, only to the unity
4350of apperception, and is for this reason the ground of the possibility
4351of a priori cognition, in so far as this cognition is dependent on the
4352understanding. This synthesis is, therefore, not merely transcendental,
4353but also purely intellectual. But because a certain form of sensuous
4354intuition exists in the mind a priori which rests on the receptivity
4355of the representative faculty (sensibility), the understanding, as a
4356spontaneity, is able to determine the internal sense by means of the
4357diversity of given representations, conformably to the synthetical
4358unity of apperception, and thus to cogitate the synthetical unity of
4359the apperception of the manifold of sensuous intuition a priori, as the
4360condition to which must necessarily be submitted all objects of human
4361intuition. And in this manner the categories as mere forms of thought
4362receive objective reality, that is, application to objects which are
4363given to us in intuition, but that only as phenomena, for it is only of
4364phenomena that we are capable of a priori intuition.
4365
4366This synthesis of the manifold of sensuous intuition, which is possible
4367and necessary a priori, may be called figurative (synthesis speciosa),
4368in contradistinction to that which is cogitated in the mere category
4369in regard to the manifold of an intuition in general, and is
4370called connection or conjunction of the understanding (synthesis
4371intellectualis). Both are transcendental, not merely because they
4372themselves precede a priori all experience, but also because they form
4373the basis for the possibility of other cognition a priori.
4374
4375But the figurative synthesis, when it has relation only to the
4376originally synthetical unity of apperception, that is to the
4377transcendental unity cogitated in the categories, must, to be
4378distinguished from the purely intellectual conjunction, be entitled the
4379transcendental synthesis of imagination. Imagination is the faculty of
4380representing an object even without its presence in intuition. Now, as
4381all our intuition is sensuous, imagination, by reason of the subjective
4382condition under which alone it can give a corresponding intuition to the
4383conceptions of the understanding, belongs to sensibility. But in so far
4384as the synthesis of the imagination is an act of spontaneity, which is
4385determinative, and not, like sense, merely determinable, and which is
4386consequently able to determine sense a priori, according to its form,
4387conformably to the unity of apperception, in so far is the imagination
4388a faculty of determining sensibility a priori, and its synthesis of
4389intuitions according to the categories must be the transcendental
4390synthesis of the imagination. It is an operation of the understanding on
4391sensibility, and the first application of the understanding to objects
4392of possible intuition, and at the same time the basis for the
4393exercise of the other functions of that faculty. As figurative, it is
4394distinguished from the merely intellectual synthesis, which is produced
4395by the understanding alone, without the aid of imagination. Now, in
4396so far as imagination is spontaneity, I sometimes call it also the
4397productive imagination, and distinguish it from the reproductive, the
4398synthesis of which is subject entirely to empirical laws, those of
4399association, namely, and which, therefore, contributes nothing to the
4400explanation of the possibility of a priori cognition, and for this
4401reason belongs not to transcendental philosophy, but to psychology.
4402
4403We have now arrived at the proper place for explaining the paradox which
4404must have struck every one in our exposition of the internal sense (SS
44056), namely--how this sense represents us to our own consciousness, only
4406as we appear to ourselves, not as we are in ourselves, because, to wit,
4407we intuite ourselves only as we are inwardly affected. Now this appears
4408to be contradictory, inasmuch as we thus stand in a passive relation
4409to ourselves; and therefore in the systems of psychology, the internal
4410sense is commonly held to be one with the faculty of apperception, while
4411we, on the contrary, carefully distinguish them.
4412
4413That which determines the internal sense is the understanding, and its
4414original power of conjoining the manifold of intuition, that is, of
4415bringing this under an apperception (upon which rests the possibility
4416of the understanding itself). Now, as the human understanding is not in
4417itself a faculty of intuition, and is unable to exercise such a power,
4418in order to conjoin, as it were, the manifold of its own intuition, the
4419synthesis of understanding is, considered per se, nothing but the unity
4420of action, of which, as such, it is self-conscious, even apart from
4421sensibility, by which, moreover, it is able to determine our internal
4422sense in respect of the manifold which may be presented to it
4423according to the form of sensuous intuition. Thus, under the name of a
4424transcendental synthesis of imagination, the understanding exercises an
4425activity upon the passive subject, whose faculty it is; and so we
4426are right in saying that the internal sense is affected thereby.
4427Apperception and its synthetical unity are by no means one and the
4428same with the internal sense. The former, as the source of all our
4429synthetical conjunction, applies, under the name of the categories, to
4430the manifold of intuition in general, prior to all sensuous intuition of
4431objects. The internal sense, on the contrary, contains merely the form
4432of intuition, but without any synthetical conjunction of the manifold
4433therein, and consequently does not contain any determined intuition,
4434which is possible only through consciousness of the determination of
4435the manifold by the transcendental act of the imagination (synthetical
4436influence of the understanding on the internal sense), which I have
4437named figurative synthesis.
4438
4439This we can indeed always perceive in ourselves. We cannot cogitate a
4440geometrical line without drawing it in thought, nor a circle without
4441describing it, nor represent the three dimensions of space without
4442drawing three lines from the same point perpendicular to one another. We
4443cannot even cogitate time, unless, in drawing a straight line (which is
4444to serve as the external figurative representation of time), we fix
4445our attention on the act of the synthesis of the manifold, whereby we
4446determine successively the internal sense, and thus attend also to the
4447succession of this determination. Motion as an act of the subject (not
4448as a determination of an object),* consequently the synthesis of the
4449manifold in space, if we make abstraction of space and attend merely to
4450the act by which we determine the internal sense according to its form,
4451is that which produces the conception of succession. The understanding,
4452therefore, does by no means find in the internal sense any such
4453synthesis of the manifold, but produces it, in that it affects this
4454sense. At the same time, how "I who think" is distinct from the "I"
4455which intuites itself (other modes of intuition being cogitable as at
4456least possible), and yet one and the same with this latter as the same
4457subject; how, therefore, I am able to say: "I, as an intelligence and
4458thinking subject, cognize myself as an object thought, so far as I am,
4459moreover, given to myself in intuition--only, like other phenomena, not
4460as I am in myself, and as considered by the understanding, but merely as
4461I appear"--is a question that has in it neither more nor less difficulty
4462than the question--"How can I be an object to myself?" or this--"How I
4463can be an object of my own intuition and internal perceptions?" But that
4464such must be the fact, if we admit that space is merely a pure form
4465of the phenomena of external sense, can be clearly proved by the
4466consideration that we cannot represent time, which is not an object of
4467external intuition, in any other way than under the image of a line,
4468which we draw in thought, a mode of representation without which we
4469could not cognize the unity of its dimension, and also that we are
4470necessitated to take our determination of periods of time, or of points
4471of time, for all our internal perceptions from the changes which
4472we perceive in outward things. It follows that we must arrange the
4473determinations of the internal sense, as phenomena in time, exactly in
4474the same manner as we arrange those of the external senses in space. And
4475consequently, if we grant, respecting this latter, that by means of them
4476we know objects only in so far as we are affected externally, we must
4477also confess, with regard to the internal sense, that by means of it we
4478intuite ourselves only as we are internally affected by ourselves; in
4479other words, as regards internal intuition, we cognize our own subject
4480only as phenomenon, and not as it is in itself.**
4481
4482
4483 [*Footnote: Motion of an object in space does not belong to
4484 a pure science, consequently not to geometry; because, that
4485 a thing is movable cannot be known a priori, but only from
4486 experience. But motion, considered as the description of a
4487 space, is a pure act of the successive synthesis of the
4488 manifold in external intuition by means of productive
4489 imagination, and belongs not only to geometry, but even to
4490 transcendental philosophy.]
4491
4492
4493 [**Footnote: I do not see why so much difficulty should be
4494 found in admitting that our internal sense is affected by
4495 ourselves. Every act of attention exemplifies it. In such an
4496 act the understanding determines the internal sense by the
4497 synthetical conjunction which it cogitates, conformably to
4498 the internal intuition which corresponds to the manifold in
4499 the synthesis of the understanding. How much the mind is
4500 usually affected thereby every one will be able to perceive
4501 in himself.]
4502
4503
4504
4505SS 21.
4506
4507On the other hand, in the transcendental synthesis of the manifold
4508content of representations, consequently in the synthetical unity of
4509apperception, I am conscious of myself, not as I appear to myself,
4510nor as I am in myself, but only that "I am." This representation is a
4511thought, not an intuition. Now, as in order to cognize ourselves, in
4512addition to the act of thinking, which subjects the manifold of every
4513possible intuition to the unity of apperception, there is necessary a
4514determinate mode of intuition, whereby this manifold is given; although
4515my own existence is certainly not mere phenomenon (much less mere
4516illusion), the determination of my existence* Can only take place
4517conformably to the form of the internal sense, according to the
4518particular mode in which the manifold which I conjoin is given in
4519internal intuition, and I have therefore no knowledge of myself as I am,
4520but merely as I appear to myself. The consciousness of self is thus
4521very far from a knowledge of self, in which I do not use the categories,
4522whereby I cogitate an object, by means of the conjunction of the
4523manifold in one apperception. In the same way as I require, for the sake
4524of the cognition of an object distinct from myself, not only the thought
4525of an object in general (in the category), but also an intuition
4526by which to determine that general conception, in the same way do I
4527require, in order to the cognition of myself, not only the consciousness
4528of myself or the thought that I think myself, but in addition an
4529intuition of the manifold in myself, by which to determine this thought.
4530It is true that I exist as an intelligence which is conscious only of
4531its faculty of conjunction or synthesis, but subjected in relation to
4532the manifold which this intelligence has to conjoin to a limitative
4533conjunction called the internal sense. My intelligence (that is, I) can
4534render that conjunction or synthesis perceptible only according to
4535the relations of time, which are quite beyond the proper sphere of the
4536conceptions of the understanding and consequently cognize itself in
4537respect to an intuition (which cannot possibly be intellectual, nor
4538given by the understanding), only as it appears to itself, and not as it
4539would cognize itself, if its intuition were intellectual.
4540
4541
4542 [*Footnote: The "I think" expresses the act of determining
4543 my own existence. My existence is thus already given by the
4544 act of consciousness; but the mode in which I must determine
4545 my existence, that is, the mode in which I must place the
4546 manifold belonging to my existence, is not thereby given.
4547 For this purpose intuition of self is required, and this
4548 intuition possesses a form given a priori, namely, time,
4549 which is sensuous, and belongs to our receptivity of the
4550 determinable. Now, as I do not possess another intuition of
4551 self which gives the determining in me (of the spontaneity
4552 of which I am conscious), prior to the act of determination,
4553 in the same manner as time gives the determinable, it is
4554 clear that I am unable to determine my own existence as that
4555 of a spontaneous being, but I am only able to represent to
4556 myself the spontaneity of my thought, that is, of my
4557 determination, and my existence remains ever determinable in
4558 a purely sensuous manner, that is to say, like the existence
4559 of a phenomenon. But it is because of this spontaneity that
4560 I call myself an intelligence.]
4561
4562
4563
4564SS 22. Transcendental Deduction of the universally possible employment
4565in experience of the Pure Conceptions of the Understanding.
4566
4567In the metaphysical deduction, the a priori origin of categories was
4568proved by their complete accordance with the general logical of thought;
4569in the transcendental deduction was exhibited the possibility of the
4570categories as a priori cognitions of objects of an intuition in general
4571(SS 16 and 17).At present we are about to explain the possibility of
4572cognizing, a priori, by means of the categories, all objects which can
4573possibly be presented to our senses, not, indeed, according to the form
4574of their intuition, but according to the laws of their conjunction or
4575synthesis, and thus, as it were, of prescribing laws to nature and even
4576of rendering nature possible. For if the categories were inadequate
4577to this task, it would not be evident to us why everything that is
4578presented to our senses must be subject to those laws which have an a
4579priori origin in the understanding itself.
4580
4581I premise that by the term synthesis of apprehension I understand
4582the combination of the manifold in an empirical intuition, whereby
4583perception, that is, empirical consciousness of the intuition (as
4584phenomenon), is possible.
4585
4586We have a priori forms of the external and internal sensuous intuition
4587in the representations of space and time, and to these must the
4588synthesis of apprehension of the manifold in a phenomenon be always
4589comformable, because the synthesis itself can only take place according
4590to these forms. But space and time are not merely forms of sensuous
4591intuition, but intuitions themselves (which contain a manifold), and
4592therefore contain a priori the determination of the unity of this
4593manifold.* (See the Transcendent Aesthetic.) Therefore is unity of the
4594synthesis of the manifold without or within us, consequently also a
4595conjunction to which all that is to be represented as determined in
4596space or time must correspond, given a priori along with (not in) these
4597intuitions, as the condition of the synthesis of all apprehension
4598of them. But this synthetical unity can be no other than that of the
4599conjunction of the manifold of a given intuition in general, in a
4600primitive act of consciousness, according to the categories, but applied
4601to our sensuous intuition. Consequently all synthesis, whereby alone
4602is even perception possible, is subject to the categories. And,
4603as experience is cognition by means of conjoined perceptions, the
4604categories are conditions of the possibility of experience and are
4605therefore valid a priori for all objects of experience.
4606
4607 [*Footnote: Space represented as an object (as geometry
4608 really requires it to be) contains more than the mere form
4609 of the intuition; namely, a combination of the manifold
4610 given according to the form of sensibility into a
4611 representation that can be intuited; so that the form of the
4612 intuition gives us merely the manifold, but the formal
4613 intuition gives unity of representation. In the aesthetic, I
4614 regarded this unity as belonging entirely to sensibility,
4615 for the purpose of indicating that it antecedes all
4616 conceptions, although it presupposes a synthesis which does
4617 not belong to sense, through which alone, however, all our
4618 conceptions of space and time are possible. For as by means
4619 of this unity alone (the understanding determining the
4620 sensibility) space and time are given as intuitions, it
4621 follows that the unity of this intuition a priori belongs to
4622 space and time, and not to the conception of the
4623 understanding (SS 20).]
4624
4625
4626When, then, for example, I make the empirical intuition of a house by
4627apprehension of the manifold contained therein into a perception, the
4628necessary unity of space and of my external sensuous intuition lies
4629at the foundation of this act, and I, as it were, draw the form of the
4630house conformably to this synthetical unity of the manifold in space.
4631But this very synthetical unity remains, even when I abstract the form
4632of space, and has its seat in the understanding, and is in fact the
4633category of the synthesis of the homogeneous in an intuition; that is
4634to say, the category of quantity, to which the aforesaid synthesis of
4635apprehension, that is, the perception, must be completely conformable.*
4636
4637 [*Footnote: In this manner it is proved, that the synthesis
4638 of apprehension, which is empirical, must necessarily be
4639 conformable to the synthesis of apperception, which is
4640 intellectual, and contained a priori in the category. It is
4641 one and the same spontaneity which at one time, under the
4642 name of imagination, at another under that of understanding,
4643 produces conjunction in the manifold of intuition.]
4644
4645
4646To take another example, when I perceive the freezing of water, I
4647apprehend two states (fluidity and solidity), which, as such, stand
4648toward each other mutually in a relation of time. But in the time, which
4649I place as an internal intuition, at the foundation of this phenomenon,
4650I represent to myself synthetical unity of the manifold, without which
4651the aforesaid relation could not be given in an intuition as determined
4652(in regard to the succession of time). Now this synthetical unity,
4653as the a priori condition under which I conjoin the manifold of an
4654intuition, is, if I make abstraction of the permanent form of my
4655internal intuition (that is to say, of time), the category of cause, by
4656means of which, when applied to my sensibility, I determine everything
4657that occurs according to relations of time. Consequently apprehension in
4658such an event, and the event itself, as far as regards the possibility
4659of its perception, stands under the conception of the relation of cause
4660and effect: and so in all other cases.
4661
4662Categories are conceptions which prescribe laws a priori to phenomena,
4663consequently to nature as the complex of all phenomena (natura
4664materialiter spectata). And now the question arises--inasmuch as these
4665categories are not derived from nature, and do not regulate themselves
4666according to her as their model (for in that case they would be
4667empirical)--how it is conceivable that nature must regulate herself
4668according to them, in other words, how the categories can determine a
4669priori the synthesis of the manifold of nature, and yet not derive their
4670origin from her. The following is the solution of this enigma.
4671
4672It is not in the least more difficult to conceive how the laws of the
4673phenomena of nature must harmonize with the understanding and with its a
4674priori form--that is, its faculty of conjoining the manifold--than it
4675is to understand how the phenomena themselves must correspond with the
4676a priori form of our sensuous intuition. For laws do not exist in the
4677phenomena any more than the phenomena exist as things in themselves.
4678Laws do not exist except by relation to the subject in which the
4679phenomena inhere, in so far as it possesses understanding, just as
4680phenomena have no existence except by relation to the same existing
4681subject in so far as it has senses. To things as things in themselves,
4682conformability to law must necessarily belong independently of an
4683understanding to cognize them. But phenomena are only representations
4684of things which are utterly unknown in respect to what they are in
4685themselves. But as mere representations, they stand under no law of
4686conjunction except that which the conjoining faculty prescribes. Now
4687that which conjoins the manifold of sensuous intuition is imagination,
4688a mental act to which understanding contributes unity of intellectual
4689synthesis, and sensibility, manifoldness of apprehension. Now as all
4690possible perception depends on the synthesis of apprehension, and this
4691empirical synthesis itself on the transcendental, consequently on the
4692categories, it is evident that all possible perceptions, and therefore
4693everything that can attain to empirical consciousness, that is, all
4694phenomena of nature, must, as regards their conjunction, be subject to
4695the categories. And nature (considered merely as nature in general)
4696is dependent on them, as the original ground of her necessary
4697conformability to law (as natura formaliter spectata). But the pure
4698faculty (of the understanding) of prescribing laws a priori to phenomena
4699by means of mere categories, is not competent to enounce other or more
4700laws than those on which a nature in general, as a conformability to law
4701of phenomena of space and time, depends. Particular laws, inasmuch
4702as they concern empirically determined phenomena, cannot be entirely
4703deduced from pure laws, although they all stand under them. Experience
4704must be superadded in order to know these particular laws; but in regard
4705to experience in general, and everything that can be cognized as an
4706object thereof, these a priori laws are our only rule and guide.
4707
4708
4709
4710SS 23. Result of this Deduction of the Conceptions of the
4711Understanding.
4712
4713We cannot think any object except by means of the categories; we cannot
4714cognize any thought except by means of intuitions corresponding to these
4715conceptions. Now all our intuitions are sensuous, and our cognition,
4716in so far as the object of it is given, is empirical. But empirical
4717cognition is experience; consequently no a priori cognition is possible
4718for us, except of objects of possible experience.*
4719
4720 [Footnote: Lest my readers should stumble at this assertion,
4721 and the conclusions that may be too rashly drawn from it, I
4722 must remind them that the categories in the act of thought
4723 are by no means limited by the conditions of our sensuous
4724 intuition, but have an unbounded sphere of action. It is
4725 only the cognition of the object of thought, the determining
4726 of the object, which requires intuition. In the absence of
4727 intuition, our thought of an object may still have true and
4728 useful consequences in regard to the exercise of reason by
4729 the subject. But as this exercise of reason is not always
4730 directed on the determination of the object, in other words,
4731 on cognition thereof, but also on the determination of the
4732 subject and its volition, I do not intend to treat of it in
4733 this place.]
4734
4735
4736But this cognition, which is limited to objects of experience, is not
4737for that reason derived entirely, from, experience, but--and this
4738is asserted of the pure intuitions and the pure conceptions of the
4739understanding--there are, unquestionably, elements of cognition, which
4740exist in the mind a priori. Now there are only two ways in which a
4741necessary harmony of experience with the conceptions of its objects can
4742be cogitated. Either experience makes these conceptions possible, or
4743the conceptions make experience possible. The former of these statements
4744will not bold good with respect to the categories (nor in regard to pure
4745sensuous intuition), for they are a priori conceptions, and therefore
4746independent of experience. The assertion of an empirical origin would
4747attribute to them a sort of generatio aequivoca. Consequently, nothing
4748remains but to adopt the second alternative (which presents us with a
4749system, as it were, of the epigenesis of pure reason), namely, that on
4750the part of the understanding the categories do contain the grounds of
4751the possibility of all experience. But with respect to the questions
4752how they make experience possible, and what are the principles of the
4753possibility thereof with which they present us in their application to
4754phenomena, the following section on the transcendental exercise of the
4755faculty of judgement will inform the reader.
4756
4757It is quite possible that someone may propose a species of
4758preformation-system of pure reason--a middle way between the two--to
4759wit, that the categories are neither innate and first a priori
4760principles of cognition, nor derived from experience, but are merely
4761subjective aptitudes for thought implanted in us contemporaneously with
4762our existence, which were so ordered and disposed by our Creator,
4763that their exercise perfectly harmonizes with the laws of nature which
4764regulate experience. Now, not to mention that with such an hypothesis
4765it is impossible to say at what point we must stop in the employment of
4766predetermined aptitudes, the fact that the categories would in this case
4767entirely lose that character of necessity which is essentially involved
4768in the very conception of them, is a conclusive objection to it. The
4769conception of cause, for example, which expresses the necessity of an
4770effect under a presupposed condition, would be false, if it rested only
4771upon such an arbitrary subjective necessity of uniting certain empirical
4772representations according to such a rule of relation. I could not then
4773say--"The effect is connected with its cause in the object (that is,
4774necessarily)," but only, "I am so constituted that I can think this
4775representation as so connected, and not otherwise." Now this is just
4776what the sceptic wants. For in this case, all our knowledge, depending
4777on the supposed objective validity of our judgement, is nothing but
4778mere illusion; nor would there be wanting people who would deny any such
4779subjective necessity in respect to themselves, though they must feel it.
4780At all events, we could not dispute with any one on that which merely
4781depends on the manner in which his subject is organized.
4782
4783
4784
4785Short view of the above Deduction.
4786
4787The foregoing deduction is an exposition of the pure conceptions of the
4788understanding (and with them of all theoretical a priori cognition), as
4789principles of the possibility of experience, but of experience as
4790the determination of all phenomena in space and time in general--of
4791experience, finally, from the principle of the original synthetical
4792unity of apperception, as the form of the understanding in relation to
4793time and space as original forms of sensibility.
4794
4795I consider the division by paragraphs to be necessary only up to this
4796point, because we had to treat of the elementary conceptions. As we
4797now proceed to the exposition of the employment of these, I shall not
4798designate the chapters in this manner any further.
4799
4800
4801
4802BOOK II.
4803
4804Analytic of Principles.
4805
4806General logic is constructed upon a plan which coincides exactly
4807with the division of the higher faculties of cognition. These are,
4808understanding, judgement, and reason. This science, accordingly, treats
4809in its analytic of conceptions, judgements, and conclusions in exact
4810correspondence with the functions and order of those mental powers which
4811we include generally under the generic denomination of understanding.
4812
4813As this merely formal logic makes abstraction of all content of
4814cognition, whether pure or empirical, and occupies itself with the mere
4815form of thought (discursive cognition), it must contain in its analytic
4816a canon for reason. For the form of reason has its law, which, without
4817taking into consideration the particular nature of the cognition about
4818which it is employed, can be discovered a priori, by the simple analysis
4819of the action of reason into its momenta.
4820
4821Transcendental logic, limited as it is to a determinate content, that of
4822pure a priori cognitions, to wit, cannot imitate general logic in this
4823division. For it is evident that the transcendental employment of reason
4824is not objectively valid, and therefore does not belong to the logic
4825of truth (that is, to analytic), but as a logic of illusion, occupies
4826a particular department in the scholastic system under the name of
4827transcendental dialectic.
4828
4829Understanding and judgement accordingly possess in transcendental logic
4830a canon of objectively valid, and therefore true exercise, and are
4831comprehended in the analytical department of that logic. But reason,
4832in her endeavours to arrive by a priori means at some true statement
4833concerning objects and to extend cognition beyond the bounds of possible
4834experience, is altogether dialectic, and her illusory assertions cannot
4835be constructed into a canon such as an analytic ought to contain.
4836
4837Accordingly, the analytic of principles will be merely a canon for
4838the faculty of judgement, for the instruction of this faculty in its
4839application to phenomena of the pure conceptions of the understanding,
4840which contain the necessary condition for the establishment of a priori
4841laws. On this account, although the subject of the following chapters is
4842the especial principles of understanding, I shall make use of the
4843term Doctrine of the faculty of judgement, in order to define more
4844particularly my present purpose.
4845
4846
4847
4848INTRODUCTION. Of the Transcendental Faculty of judgement in General.
4849
4850If understanding in general be defined as the faculty of laws or rules,
4851the faculty of judgement may be termed the faculty of subsumption under
4852these rules; that is, of distinguishing whether this or that does or
4853does not stand under a given rule (casus datae legis). General logic
4854contains no directions or precepts for the faculty of judgement, nor
4855can it contain any such. For as it makes abstraction of all content of
4856cognition, no duty is left for it, except that of exposing analytically
4857the mere form of cognition in conceptions, judgements, and conclusions,
4858and of thereby establishing formal rules for all exercise of the
4859understanding. Now if this logic wished to give some general direction
4860how we should subsume under these rules, that is, how we should
4861distinguish whether this or that did or did not stand under them, this
4862again could not be done otherwise than by means of a rule. But this
4863rule, precisely because it is a rule, requires for itself direction from
4864the faculty of judgement. Thus, it is evident that the understanding
4865is capable of being instructed by rules, but that the judgement is a
4866peculiar talent, which does not, and cannot require tuition, but
4867only exercise. This faculty is therefore the specific quality of the
4868so-called mother wit, the want of which no scholastic discipline can
4869compensate.
4870
4871For although education may furnish, and, as it were, engraft upon a
4872limited understanding rules borrowed from other minds, yet the power of
4873employing these rules correctly must belong to the pupil himself; and no
4874rule which we can prescribe to him with this purpose is, in the absence
4875or deficiency of this gift of nature, secure from misuse.* A physician
4876therefore, a judge or a statesman, may have in his head many admirable
4877pathological, juridical, or political rules, in a degree that may enable
4878him to be a profound teacher in his particular science, and yet in the
4879application of these rules he may very possibly blunder--either because
4880he is wanting in natural judgement (though not in understanding) and,
4881whilst he can comprehend the general in abstracto, cannot distinguish
4882whether a particular case in concreto ought to rank under the former; or
4883because his faculty of judgement has not been sufficiently exercised by
4884examples and real practice. Indeed, the grand and only use of examples,
4885is to sharpen the judgement. For as regards the correctness and
4886precision of the insight of the understanding, examples are commonly
4887injurious rather than otherwise, because, as casus in terminis they
4888seldom adequately fulfil the conditions of the rule. Besides, they often
4889weaken the power of our understanding to apprehend rules or laws
4890in their universality, independently of particular circumstances of
4891experience; and hence, accustom us to employ them more as formulae than
4892as principles. Examples are thus the go-cart of the judgement, which
4893he who is naturally deficient in that faculty cannot afford to dispense
4894with.
4895
4896 [*Footnote: Deficiency in judgement is properly that which
4897 is called stupidity; and for such a failing we know no
4898 remedy. A dull or narrow-minded person, to whom nothing is
4899 wanting but a proper degree of understanding, may be
4900 improved by tuition, even so far as to deserve the epithet
4901 of learned. But as such persons frequently labour under a
4902 deficiency in the faculty of judgement, it is not uncommon
4903 to find men extremely learned who in the application of
4904 their science betray a lamentable degree this irremediable
4905 want.]
4906
4907
4908But although general logic cannot give directions to the faculty of
4909judgement, the case is very different as regards transcendental logic,
4910insomuch that it appears to be the especial duty of the latter to secure
4911and direct, by means of determinate rules, the faculty of judgement in
4912the employment of the pure understanding. For, as a doctrine, that is,
4913as an endeavour to enlarge the sphere of the understanding in regard to
4914pure a priori cognitions, philosophy is worse than useless, since from
4915all the attempts hitherto made, little or no ground has been gained.
4916But, as a critique, in order to guard against the mistakes of the
4917faculty of judgement (lapsus judicii) in the employment of the few pure
4918conceptions of the understanding which we possess, although its use is
4919in this case purely negative, philosophy is called upon to apply all its
4920acuteness and penetration.
4921
4922But transcendental philosophy has this peculiarity, that besides
4923indicating the rule, or rather the general condition for rules, which is
4924given in the pure conception of the understanding, it can, at the same
4925time, indicate a priori the case to which the rule must be applied.
4926The cause of the superiority which, in this respect, transcendental
4927philosophy possesses above all other sciences except mathematics, lies
4928in this: it treats of conceptions which must relate a priori to their
4929objects, whose objective validity consequently cannot be demonstrated a
4930posteriori, and is, at the same time, under the obligation of presenting
4931in general but sufficient tests, the conditions under which objects can
4932be given in harmony with those conceptions; otherwise they would be
4933mere logical forms, without content, and not pure conceptions of the
4934understanding.
4935
4936Our transcendental doctrine of the faculty of judgement will contain
4937two chapters. The first will treat of the sensuous condition under which
4938alone pure conceptions of the understanding can be employed--that is, of
4939the schematism of the pure understanding. The second will treat of those
4940synthetical judgements which are derived a priori from pure conceptions
4941of the understanding under those conditions, and which lie a priori at
4942the foundation of all other cognitions, that is to say, it will treat of
4943the principles of the pure understanding.
4944
4945
4946
4947TRANSCENDENTAL DOCTRINE OF THE FACULTY OF JUDGEMENT OR, ANALYTIC OF
4948PRINCIPLES.
4949
4950CHAPTER I. Of the Schematism at of the Pure Conceptions of the
4951Understanding.
4952
4953In all subsumptions of an object under a conception, the representation
4954of the object must be homogeneous with the conception; in other words,
4955the conception must contain that which is represented in the object to
4956be subsumed under it. For this is the meaning of the expression: "An
4957object is contained under a conception." Thus the empirical conception
4958of a plate is homogeneous with the pure geometrical conception of a
4959circle, inasmuch as the roundness which is cogitated in the former is
4960intuited in the latter.
4961
4962But pure conceptions of the understanding, when compared with empirical
4963intuitions, or even with sensuous intuitions in general, are quite
4964heterogeneous, and never can be discovered in any intuition. How then
4965is the subsumption of the latter under the former, and consequently
4966the application of the categories to phenomena, possible?--For it is
4967impossible to say, for example: "Causality can be intuited through the
4968senses and is contained in the phenomenon."--This natural and important
4969question forms the real cause of the necessity of a transcendental
4970doctrine of the faculty of judgement, with the purpose, to wit, of
4971showing how pure conceptions of the understanding can be applied to
4972phenomena. In all other sciences, where the conceptions by which the
4973object is thought in the general are not so different and heterogeneous
4974from those which represent the object in concreto--as it is given, it
4975is quite unnecessary to institute any special inquiries concerning the
4976application of the former to the latter.
4977
4978Now it is quite clear that there must be some third thing, which on the
4979one side is homogeneous with the category, and with the phenomenon on
4980the other, and so makes the application of the former to the latter
4981possible. This mediating representation must be pure (without any
4982empirical content), and yet must on the one side be intellectual, on the
4983other sensuous. Such a representation is the transcendental schema.
4984
4985The conception of the understanding contains pure synthetical unity of
4986the manifold in general. Time, as the formal condition of the
4987manifold of the internal sense, consequently of the conjunction of all
4988representations, contains a priori a manifold in the pure intuition. Now
4989a transcendental determination of time is so far homogeneous with the
4990category, which constitutes the unity thereof, that it is universal and
4991rests upon a rule a priori. On the other hand, it is so far homogeneous
4992with the phenomenon, inasmuch as time is contained in every empirical
4993representation of the manifold. Thus an application of the category to
4994phenomena becomes possible, by means of the transcendental determination
4995of time, which, as the schema of the conceptions of the understanding,
4996mediates the subsumption of the latter under the former.
4997
4998After what has been proved in our deduction of the categories, no
4999one, it is to be hoped, can hesitate as to the proper decision of
5000the question, whether the employment of these pure conceptions of the
5001understanding ought to be merely empirical or also transcendental;
5002in other words, whether the categories, as conditions of a possible
5003experience, relate a priori solely to phenomena, or whether, as
5004conditions of the possibility of things in general, their application
5005can be extended to objects as things in themselves. For we have
5006there seen that conceptions are quite impossible, and utterly without
5007signification, unless either to them, or at least to the elements of
5008which they consist, an object be given; and that, consequently, they
5009cannot possibly apply to objects as things in themselves without regard
5010to the question whether and how these may be given to us; and, further,
5011that the only manner in which objects can be given to us is by means of
5012the modification of our sensibility; and, finally, that pure a priori
5013conceptions, in addition to the function of the understanding in the
5014category, must contain a priori formal conditions of sensibility (of the
5015internal sense, namely), which again contain the general condition under
5016which alone the category can be applied to any object. This formal
5017and pure condition of sensibility, to which the conception of the
5018understanding is restricted in its employment, we shall name the
5019schema of the conception of the understanding, and the procedure of the
5020understanding with these schemata we shall call the schematism of the
5021pure understanding.
5022
5023The schema is, in itself, always a mere product of the imagination. But,
5024as the synthesis of imagination has for its aim no single intuition, but
5025merely unity in the determination of sensibility, the schema is clearly
5026distinguishable from the image. Thus, if I place five points one after
5027another.... this is an image of the number five. On the other hand, if
5028I only think a number in general, which may be either five or a hundred,
5029this thought is rather the representation of a method of representing in
5030an image a sum (e.g., a thousand) in conformity with a conception, than
5031the image itself, an image which I should find some little difficulty in
5032reviewing, and comparing with the conception. Now this representation
5033of a general procedure of the imagination to present its image to a
5034conception, I call the schema of this conception.
5035
5036In truth, it is not images of objects, but schemata, which lie at the
5037foundation of our pure sensuous conceptions. No image could ever be
5038adequate to our conception of a triangle in general. For the generalness
5039of the conception it never could attain to, as this includes under
5040itself all triangles, whether right-angled, acute-angled, etc., whilst
5041the image would always be limited to a single part of this sphere. The
5042schema of the triangle can exist nowhere else than in thought, and it
5043indicates a rule of the synthesis of the imagination in regard to pure
5044figures in space. Still less is an object of experience, or an image
5045of the object, ever to the empirical conception. On the contrary, the
5046conception always relates immediately to the schema of the imagination,
5047as a rule for the determination of our intuition, in conformity with a
5048certain general conception. The conception of a dog indicates a
5049rule, according to which my imagination can delineate the figure of a
5050four-footed animal in general, without being limited to any particular
5051individual form which experience presents to me, or indeed to any
5052possible image that I can represent to myself in concreto. This
5053schematism of our understanding in regard to phenomena and their mere
5054form, is an art, hidden in the depths of the human soul, whose true
5055modes of action we shall only with difficulty discover and unveil. Thus
5056much only can we say: "The image is a product of the empirical faculty
5057of the productive imagination--the schema of sensuous conceptions (of
5058figures in space, for example) is a product, and, as it were, a monogram
5059of the pure imagination a priori, whereby and according to which
5060images first become possible, which, however, can be connected with the
5061conception only mediately by means of the schema which they indicate,
5062and are in themselves never fully adequate to it." On the other hand,
5063the schema of a pure conception of the understanding is something that
5064cannot be reduced into any image--it is nothing else than the pure
5065synthesis expressed by the category, conformably, to a rule of unity
5066according to conceptions. It is a transcendental product of the
5067imagination, a product which concerns the determination of the internal
5068sense, according to conditions of its form (time) in respect to all
5069representations, in so far as these representations must be conjoined a
5070priori in one conception, conformably to the unity of apperception.
5071
5072Without entering upon a dry and tedious analysis of the essential
5073requisites of transcendental schemata of the pure conceptions of the
5074understanding, we shall rather proceed at once to give an explanation
5075of them according to the order of the categories, and in connection
5076therewith.
5077
5078For the external sense the pure image of all quantities (quantorum) is
5079space; the pure image of all objects of sense in general, is time.
5080But the pure schema of quantity (quantitatis) as a conception of
5081the understanding, is number, a representation which comprehends the
5082successive addition of one to one (homogeneous quantities). Thus, number
5083is nothing else than the unity of the synthesis of the manifold in
5084a homogeneous intuition, by means of my generating time itself in my
5085apprehension of the intuition.
5086
5087Reality, in the pure conception of the understanding, is that which
5088corresponds to a sensation in general; that, consequently, the
5089conception of which indicates a being (in time). Negation is that the
5090conception of which represents a not-being (in time). The opposition of
5091these two consists therefore in the difference of one and the same
5092time, as a time filled or a time empty. Now as time is only the form of
5093intuition, consequently of objects as phenomena, that which in objects
5094corresponds to sensation is the transcendental matter of all objects
5095as things in themselves (Sachheit, reality). Now every sensation has
5096a degree or quantity by which it can fill time, that is to say, the
5097internal sense in respect of the representation of an object, more or
5098less, until it vanishes into nothing (= 0 = negatio). Thus there is
5099a relation and connection between reality and negation, or rather a
5100transition from the former to the latter, which makes every reality
5101representable to us as a quantum; and the schema of a reality as the
5102quantity of something in so far as it fills time, is exactly this
5103continuous and uniform generation of the reality in time, as we descend
5104in time from the sensation which has a certain degree, down to the
5105vanishing thereof, or gradually ascend from negation to the quantity
5106thereof.
5107
5108The schema of substance is the permanence of the real in time; that is,
5109the representation of it as a substratum of the empirical determination
5110of time; a substratum which therefore remains, whilst all else changes.
5111(Time passes not, but in it passes the existence of the changeable. To
5112time, therefore, which is itself unchangeable and permanent, corresponds
5113that which in the phenomenon is unchangeable in existence, that is,
5114substance, and it is only by it that the succession and coexistence of
5115phenomena can be determined in regard to time.)
5116
5117The schema of cause and of the causality of a thing is the real which,
5118when posited, is always followed by something else. It consists,
5119therefore, in the succession of the manifold, in so far as that
5120succession is subjected to a rule.
5121
5122The schema of community (reciprocity of action and reaction), or the
5123reciprocal causality of substances in respect of their accidents, is the
5124coexistence of the determinations of the one with those of the other,
5125according to a general rule.
5126
5127The schema of possibility is the accordance of the synthesis of
5128different representations with the conditions of time in general (as,
5129for example, opposites cannot exist together at the same time in
5130the same thing, but only after each other), and is therefore the
5131determination of the representation of a thing at any time.
5132
5133The schema of reality is existence in a determined time.
5134
5135The schema of necessity is the existence of an object in all time.
5136
5137It is clear, from all this, that the schema of the category of quantity
5138contains and represents the generation (synthesis) of time itself, in
5139the successive apprehension of an object; the schema of quality the
5140synthesis of sensation with the representation of time, or the filling
5141up of time; the schema of relation the relation of perceptions to each
5142other in all time (that is, according to a rule of the determination
5143of time): and finally, the schema of modality and its categories, time
5144itself, as the correlative of the determination of an object--whether it
5145does belong to time, and how. The schemata, therefore, are nothing but
5146a priori determinations of time according to rules, and these, in regard
5147to all possible objects, following the arrangement of the categories,
5148relate to the series in time, the content in time, the order in time,
5149and finally, to the complex or totality in time.
5150
5151Hence it is apparent that the schematism of the understanding, by means
5152of the transcendental synthesis of the imagination, amounts to nothing
5153else than the unity of the manifold of intuition in the internal
5154sense, and thus indirectly to the unity of apperception, as a function
5155corresponding to the internal sense (a receptivity). Thus, the schemata
5156of the pure conceptions of the understanding are the true and only
5157conditions whereby our understanding receives an application to objects,
5158and consequently significance. Finally, therefore, the categories are
5159only capable of empirical use, inasmuch as they serve merely to subject
5160phenomena to the universal rules of synthesis, by means of an a priori
5161necessary unity (on account of the necessary union of all consciousness
5162in one original apperception); and so to render them susceptible of a
5163complete connection in one experience. But within this whole of possible
5164experience lie all our cognitions, and in the universal relation to this
5165experience consists transcendental truth, which antecedes all empirical
5166truth, and renders the latter possible.
5167
5168It is, however, evident at first sight, that although the schemata of
5169sensibility are the sole agents in realizing the categories, they do,
5170nevertheless, also restrict them, that is, they limit the categories
5171by conditions which lie beyond the sphere of understanding--namely, in
5172sensibility. Hence the schema is properly only the phenomenon, or the
5173sensuous conception of an object in harmony with the category. (Numerus
5174est quantitas phaenomenon--sensatio realitas phaenomenon; constans
5175et perdurabile rerum substantia phaenomenon--aeternitas, necessitas,
5176phaenomena, etc.) Now, if we remove a restrictive condition, we thereby
5177amplify, it appears, the formerly limited conception. In this way, the
5178categories in their pure signification, free from all conditions of
5179sensibility, ought to be valid of things as they are, and not, as the
5180schemata represent them, merely as they appear; and consequently the
5181categories must have a significance far more extended, and wholly
5182independent of all schemata. In truth, there does always remain to the
5183pure conceptions of the understanding, after abstracting every sensuous
5184condition, a value and significance, which is, however, merely logical.
5185But in this case, no object is given them, and therefore they have no
5186meaning sufficient to afford us a conception of an object. The notion
5187of substance, for example, if we leave out the sensuous determination
5188of permanence, would mean nothing more than a something which can be
5189cogitated as subject, without the possibility of becoming a predicate to
5190anything else. Of this representation I can make nothing, inasmuch as
5191it does not indicate to me what determinations the thing possesses which
5192must thus be valid as premier subject. Consequently, the categories,
5193without schemata are merely functions of the understanding for the
5194production of conceptions, but do not represent any object. This
5195significance they derive from sensibility, which at the same time
5196realizes the understanding and restricts it.
5197
5198
5199
5200CHAPTER II. System of all Principles of the Pure Understanding.
5201
5202In the foregoing chapter we have merely considered the general
5203conditions under which alone the transcendental faculty of judgement
5204is justified in using the pure conceptions of the understanding for
5205synthetical judgements. Our duty at present is to exhibit in systematic
5206connection those judgements which the understanding really produces a
5207priori. For this purpose, our table of the categories will certainly
5208afford us the natural and safe guidance. For it is precisely the
5209categories whose application to possible experience must constitute all
5210pure a priori cognition of the understanding; and the relation of which
5211to sensibility will, on that very account, present us with a complete
5212and systematic catalogue of all the transcendental principles of the use
5213of the understanding.
5214
5215Principles a priori are so called, not merely because they contain
5216in themselves the grounds of other judgements, but also because they
5217themselves are not grounded in higher and more general cognitions. This
5218peculiarity, however, does not raise them altogether above the need of
5219a proof. For although there could be found no higher cognition, and
5220therefore no objective proof, and although such a principle rather
5221serves as the foundation for all cognition of the object, this by no
5222means hinders us from drawing a proof from the subjective sources of the
5223possibility of the cognition of an object. Such a proof is necessary,
5224moreover, because without it the principle might be liable to the
5225imputation of being a mere gratuitous assertion.
5226
5227In the second place, we shall limit our investigations to those
5228principles which relate to the categories. For as to the principles
5229of transcendental aesthetic, according to which space and time are
5230the conditions of the possibility of things as phenomena, as also the
5231restriction of these principles, namely, that they cannot be applied to
5232objects as things in themselves--these, of course, do not fall within
5233the scope of our present inquiry. In like manner, the principles of
5234mathematical science form no part of this system, because they are
5235all drawn from intuition, and not from the pure conception of the
5236understanding. The possibility of these principles, however, will
5237necessarily be considered here, inasmuch as they are synthetical
5238judgements a priori, not indeed for the purpose of proving their
5239accuracy and apodeictic certainty, which is unnecessary, but merely to
5240render conceivable and deduce the possibility of such evident a priori
5241cognitions.
5242
5243But we shall have also to speak of the principle of analytical
5244judgements, in opposition to synthetical judgements, which is the proper
5245subject of our inquiries, because this very opposition will free the
5246theory of the latter from all ambiguity, and place it clearly before our
5247eyes in its true nature.
5248
5249
5250
5251SYSTEM OF THE PRINCIPLES OF THE PURE UNDERSTANDING.
5252
5253SECTION I. Of the Supreme Principle of all Analytical Judgements.
5254
5255Whatever may be the content of our cognition, and in whatever manner
5256our cognition may be related to its object, the universal, although only
5257negative conditions of all our judgements is that they do not contradict
5258themselves; otherwise these judgements are in themselves (even without
5259respect to the object) nothing. But although there may exist no
5260contradiction in our judgement, it may nevertheless connect conceptions
5261in such a manner that they do not correspond to the object, or without
5262any grounds either a priori or a posteriori for arriving at such a
5263judgement, and thus, without being self-contradictory, a judgement may
5264nevertheless be either false or groundless.
5265
5266Now, the proposition: "No subject can have a predicate that contradicts
5267it," is called the principle of contradiction, and is a universal but
5268purely negative criterion of all truth. But it belongs to logic alone,
5269because it is valid of cognitions, merely as cognitions and without
5270respect to their content, and declares that the contradiction entirely
5271nullifies them. We can also, however, make a positive use of this
5272principle, that is, not merely to banish falsehood and error (in so far
5273as it rests upon contradiction), but also for the cognition of truth.
5274For if the judgement is analytical, be it affirmative or negative,
5275its truth must always be recognizable by means of the principle of
5276contradiction. For the contrary of that which lies and is cogitated
5277as conception in the cognition of the object will be always properly
5278negatived, but the conception itself must always be affirmed of the
5279object, inasmuch as the contrary thereof would be in contradiction to
5280the object.
5281
5282We must therefore hold the principle of contradiction to be the
5283universal and fully sufficient Principle of all analytical cognition.
5284But as a sufficient criterion of truth, it has no further utility or
5285authority. For the fact that no cognition can be at variance with this
5286principle without nullifying itself, constitutes this principle the sine
5287qua non, but not the determining ground of the truth of our cognition.
5288As our business at present is properly with the synthetical part of our
5289knowledge only, we shall always be on our guard not to transgress this
5290inviolable principle; but at the same time not to expect from it any
5291direct assistance in the establishment of the truth of any synthetical
5292proposition.
5293
5294There exists, however, a formula of this celebrated principle--a
5295principle merely formal and entirely without content--which contains a
5296synthesis that has been inadvertently and quite unnecessarily mixed up
5297with it. It is this: "It is impossible for a thing to be and not to be
5298at the same time." Not to mention the superfluousness of the addition of
5299the word impossible to indicate the apodeictic certainty, which ought to
5300be self-evident from the proposition itself, the proposition is affected
5301by the condition of time, and as it were says: "A thing = A, which is
5302something = B, cannot at the same time be non-B." But both, B as well
5303as non-B, may quite well exist in succession. For example, a man who is
5304young cannot at the same time be old; but the same man can very well
5305be at one time young, and at another not young, that is, old. Now the
5306principle of contradiction as a merely logical proposition must not
5307by any means limit its application merely to relations of time, and
5308consequently a formula like the preceding is quite foreign to its
5309true purpose. The misunderstanding arises in this way. We first of all
5310separate a predicate of a thing from the conception of the thing, and
5311afterwards connect with this predicate its opposite, and hence do
5312not establish any contradiction with the subject, but only with its
5313predicate, which has been conjoined with the subject synthetically--a
5314contradiction, moreover, which obtains only when the first and second
5315predicate are affirmed in the same time. If I say: "A man who is
5316ignorant is not learned," the condition "at the same time" must be
5317added, for he who is at one time ignorant, may at another be learned.
5318But if I say: "No ignorant man is a learned man," the proposition is
5319analytical, because the characteristic ignorance is now a constituent
5320part of the conception of the subject; and in this case the
5321negative proposition is evident immediately from the proposition of
5322contradiction, without the necessity of adding the condition "the
5323same time." This is the reason why I have altered the formula of this
5324principle--an alteration which shows very clearly the nature of an
5325analytical proposition.
5326
5327
5328
5329SECTION II. Of the Supreme Principle of all Synthetical Judgements.
5330
5331The explanation of the possibility of synthetical judgements is a task
5332with which general logic has nothing to do; indeed she needs not even
5333be acquainted with its name. But in transcendental logic it is the most
5334important matter to be dealt with--indeed the only one, if the question
5335is of the possibility of synthetical judgements a priori, the conditions
5336and extent of their validity. For when this question is fully decided,
5337it can reach its aim with perfect ease, the determination, to wit, of
5338the extent and limits of the pure understanding.
5339
5340In an analytical judgement I do not go beyond the given conception,
5341in order to arrive at some decision respecting it. If the judgement is
5342affirmative, I predicate of the conception only that which was already
5343cogitated in it; if negative, I merely exclude from the conception its
5344contrary. But in synthetical judgements, I must go beyond the given
5345conception, in order to cogitate, in relation with it, something quite
5346different from that which was cogitated in it, a relation which is
5347consequently never one either of identity or contradiction, and by means
5348of which the truth or error of the judgement cannot be discerned merely
5349from the judgement itself.
5350
5351Granted, then, that we must go out beyond a given conception, in order
5352to compare it synthetically with another, a third thing is necessary, in
5353which alone the synthesis of two conceptions can originate. Now what
5354is this tertium quid that is to be the medium of all synthetical
5355judgements? It is only a complex in which all our representations are
5356contained, the internal sense to wit, and its form a priori, time.
5357
5358The synthesis of our representations rests upon the imagination; their
5359synthetical unity (which is requisite to a judgement), upon the unity
5360of apperception. In this, therefore, is to be sought the possibility of
5361synthetical judgements, and as all three contain the sources of a priori
5362representations, the possibility of pure synthetical judgements also;
5363nay, they are necessary upon these grounds, if we are to possess
5364a knowledge of objects, which rests solely upon the synthesis of
5365representations.
5366
5367If a cognition is to have objective reality, that is, to relate to an
5368object, and possess sense and meaning in respect to it, it is necessary
5369that the object be given in some way or another. Without this, our
5370conceptions are empty, and we may indeed have thought by means of them,
5371but by such thinking we have not, in fact, cognized anything, we have
5372merely played with representation. To give an object, if this expression
5373be understood in the sense of "to present" the object, not mediately
5374but immediately in intuition, means nothing else than to apply the
5375representation of it to experience, be that experience real or only
5376possible. Space and time themselves, pure as these conceptions are from
5377all that is empirical, and certain as it is that they are represented
5378fully a priori in the mind, would be completely without objective
5379validity, and without sense and significance, if their necessary use
5380in the objects of experience were not shown. Nay, the representation
5381of them is a mere schema, that always relates to the reproductive
5382imagination, which calls up the objects of experience, without which
5383they have no meaning. And so it is with all conceptions without
5384distinction.
5385
5386The possibility of experience is, then, that which gives objective
5387reality to all our a priori cognitions. Now experience depends upon the
5388synthetical unity of phenomena, that is, upon a synthesis according to
5389conceptions of the object of phenomena in general, a synthesis without
5390which experience never could become knowledge, but would be merely a
5391rhapsody of perceptions, never fitting together into any connected text,
5392according to rules of a thoroughly united (possible) consciousness, and
5393therefore never subjected to the transcendental and necessary unity
5394of apperception. Experience has therefore for a foundation, a priori
5395principles of its form, that is to say, general rules of unity in
5396the synthesis of phenomena, the objective reality of which rules, as
5397necessary conditions even of the possibility of experience can
5398which rules, as necessary conditions--even of the possibility of
5399experience--can always be shown in experience. But apart from this
5400relation, a priori synthetical propositions are absolutely impossible,
5401because they have no third term, that is, no pure object, in which the
5402synthetical unity can exhibit the objective reality of its conceptions.
5403
5404Although, then, respecting space, or the forms which productive
5405imagination describes therein, we do cognize much a priori in
5406synthetical judgements, and are really in no need of experience for this
5407purpose, such knowledge would nevertheless amount to nothing but a busy
5408trifling with a mere chimera, were not space to be considered as the
5409condition of the phenomena which constitute the material of external
5410experience. Hence those pure synthetical judgements do relate, though
5411but mediately, to possible experience, or rather to the possibility of
5412experience, and upon that alone is founded the objective validity of
5413their synthesis.
5414
5415While then, on the one hand, experience, as empirical synthesis, is
5416the only possible mode of cognition which gives reality to all other
5417synthesis; on the other hand, this latter synthesis, as cognition a
5418priori, possesses truth, that is, accordance with its object, only in
5419so far as it contains nothing more than what is necessary to the
5420synthetical unity of experience.
5421
5422Accordingly, the supreme principle of all synthetical judgements is:
5423"Every object is subject to the necessary conditions of the synthetical
5424unity of the manifold of intuition in a possible experience."
5425
5426A priori synthetical judgements are possible when we apply the formal
5427conditions of the a priori intuition, the synthesis of the imagination,
5428and the necessary unity of that synthesis in a transcendental
5429apperception, to a possible cognition of experience, and say: "The
5430conditions of the possibility of experience in general are at the same
5431time conditions of the possibility of the objects of experience, and
5432have, for that reason, objective validity in an a priori synthetical
5433judgement."
5434
5435
5436
5437SECTION III. Systematic Representation of all Synthetical Principles of
5438the Pure Understanding.
5439
5440That principles exist at all is to be ascribed solely to the pure
5441understanding, which is not only the faculty of rules in regard to that
5442which happens, but is even the source of principles according to which
5443everything that can be presented to us as an object is necessarily
5444subject to rules, because without such rules we never could attain
5445to cognition of an object. Even the laws of nature, if they are
5446contemplated as principles of the empirical use of the understanding,
5447possess also a characteristic of necessity, and we may therefore at
5448least expect them to be determined upon grounds which are valid a
5449priori and antecedent to all experience. But all laws of nature, without
5450distinction, are subject to higher principles of the understanding,
5451inasmuch as the former are merely applications of the latter to
5452particular cases of experience. These higher principles alone therefore
5453give the conception, which contains the necessary condition, and, as it
5454were, the exponent of a rule; experience, on the other hand, gives the
5455case which comes under the rule.
5456
5457There is no danger of our mistaking merely empirical principles for
5458principles of the pure understanding, or conversely; for the character
5459of necessity, according to conceptions which distinguish the latter,
5460and the absence of this in every empirical proposition, how extensively
5461valid soever it may be, is a perfect safeguard against confounding
5462them. There are, however, pure principles a priori, which nevertheless I
5463should not ascribe to the pure understanding--for this reason, that they
5464are not derived from pure conceptions, but (although by the mediation
5465of the understanding) from pure intuitions. But understanding is the
5466faculty of conceptions. Such principles mathematical science possesses,
5467but their application to experience, consequently their objective
5468validity, nay the possibility of such a priori synthetical cognitions
5469(the deduction thereof) rests entirely upon the pure understanding.
5470
5471On this account, I shall not reckon among my principles those of
5472mathematics; though I shall include those upon the possibility and
5473objective validity a priori, of principles of the mathematical science,
5474which, consequently, are to be looked upon as the principle of these,
5475and which proceed from conceptions to intuition, and not from intuition
5476to conceptions.
5477
5478In the application of the pure conceptions of the understanding to
5479possible experience, the employment of their synthesis is either
5480mathematical or dynamical, for it is directed partly on the intuition
5481alone, partly on the existence of a phenomenon. But the a priori
5482conditions of intuition are in relation to a possible experience
5483absolutely necessary, those of the existence of objects of a possible
5484empirical intuition are in themselves contingent. Hence the principles
5485of the mathematical use of the categories will possess a character of
5486absolute necessity, that is, will be apodeictic; those, on the other
5487hand, of the dynamical use, the character of an a priori necessity
5488indeed, but only under the condition of empirical thought in an
5489experience, therefore only mediately and indirectly. Consequently
5490they will not possess that immediate evidence which is peculiar to the
5491former, although their application to experience does not, for that
5492reason, lose its truth and certitude. But of this point we shall be
5493better able to judge at the conclusion of this system of principles.
5494
5495The table of the categories is naturally our guide to the table of
5496principles, because these are nothing else than rules for the objective
5497employment of the former. Accordingly, all principles of the pure
5498understanding are:
5499
5500 1
5501 Axioms
5502 of Intuition
5503
5504 2 3
5505 Anticipations Analogies
5506 of Perception of Experience
5507 4
5508 Postulates of
5509 Empirical Thought
5510 in general
5511
5512
5513These appellations I have chosen advisedly, in order that we might
5514not lose sight of the distinctions in respect of the evidence and the
5515employment of these principles. It will, however, soon appear that--a
5516fact which concerns both the evidence of these principles, and the
5517a priori determination of phenomena--according to the categories of
5518quantity and quality (if we attend merely to the form of these), the
5519principles of these categories are distinguishable from those of the two
5520others, in as much as the former are possessed of an intuitive, but
5521the latter of a merely discursive, though in both instances a complete,
5522certitude. I shall therefore call the former mathematical, and the
5523latter dynamical principles.* It must be observed, however, that by
5524these terms I mean just as little in the one case the principles of
5525mathematics as those of general (physical) dynamics in the other. I have
5526here in view merely the principles of the pure understanding, in
5527their application to the internal sense (without distinction of the
5528representations given therein), by means of which the sciences of
5529mathematics and dynamics become possible. Accordingly, I have named
5530these principles rather with reference to their application than their
5531content; and I shall now proceed to consider them in the order in which
5532they stand in the table.
5533
5534
5535 [*Footnote: All combination (conjunctio) is either
5536 composition (compositio) or connection (nexus). The former
5537 is the synthesis of a manifold, the parts of which do not
5538 necessarily belong to each other. For example, the two
5539 triangles into which a square is divided by a diagonal, do
5540 not necessarily belong to each other, and of this kind is
5541 the synthesis of the homogeneous in everything that can be
5542 mathematically considered. This synthesis can be divided
5543 into those of aggregation and coalition, the former of which
5544 is applied to extensive, the latter to intensive quantities.
5545 The second sort of combination (nexus) is the synthesis of a
5546 manifold, in so far as its parts do belong necessarily to
5547 each other; for example, the accident to a substance, or the
5548 effect to the cause. Consequently it is a synthesis of that
5549 which though heterogeneous, is represented as connected a
5550 priori. This combination--not an arbitrary one--I entitle
5551 dynamical because it concerns the connection of the
5552 existence of the manifold. This, again, may be divided into
5553 the physical synthesis, of the phenomena divided among each
5554 other, and the metaphysical synthesis, or the connection of
5555 phenomena a priori in the faculty of cognition.]
5556
5557
5558
55591. AXIOMS OF INTUITION.
5560
5561
5562The principle of these is: All Intuitions are Extensive Quantities.
5563
5564
5565PROOF.
5566
5567
5568All phenomena contain, as regards their form, an intuition in space and
5569time, which lies a priori at the foundation of all without exception.
5570Phenomena, therefore, cannot be apprehended, that is, received into
5571empirical consciousness otherwise than through the synthesis of a
5572manifold, through which the representations of a determinate space
5573or time are generated; that is to say, through the composition of the
5574homogeneous and the consciousness of the synthetical unity of this
5575manifold (homogeneous). Now the consciousness of a homogeneous manifold
5576in intuition, in so far as thereby the representation of an object
5577is rendered possible, is the conception of a quantity (quanti).
5578Consequently, even the perception of an object as phenomenon is possible
5579only through the same synthetical unity of the manifold of the given
5580sensuous intuition, through which the unity of the composition of the
5581homogeneous manifold in the conception of a quantity is cogitated;
5582that is to say, all phenomena are quantities, and extensive quantities,
5583because as intuitions in space or time they must be represented by
5584means of the same synthesis through which space and time themselves are
5585determined.
5586
5587An extensive quantity I call that wherein the representation of the
5588parts renders possible (and therefore necessarily antecedes) the
5589representation of the whole. I cannot represent to myself any line,
5590however small, without drawing it in thought, that is, without
5591generating from a point all its parts one after another, and in this
5592way alone producing this intuition. Precisely the same is the case with
5593every, even the smallest, portion of time. I cogitate therein only the
5594successive progress from one moment to another, and hence, by means of
5595the different portions of time and the addition of them, a determinate
5596quantity of time is produced. As the pure intuition in all phenomena
5597is either time or space, so is every phenomenon in its character of
5598intuition an extensive quantity, inasmuch as it can only be cognized
5599in our apprehension by successive synthesis (from part to part). All
5600phenomena are, accordingly, to be considered as aggregates, that is, as
5601a collection of previously given parts; which is not the case with
5602every sort of quantities, but only with those which are represented and
5603apprehended by us as extensive.
5604
5605On this successive synthesis of the productive imagination, in the
5606generation of figures, is founded the mathematics of extension, or
5607geometry, with its axioms, which express the conditions of sensuous
5608intuition a priori, under which alone the schema of a pure conception of
5609external intuition can exist; for example, "be tween two points only one
5610straight line is possible," "two straight lines cannot enclose a space,"
5611etc. These are the axioms which properly relate only to quantities
5612(quanta) as such.
5613
5614But, as regards the quantity of a thing (quantitas), that is to say, the
5615answer to the question: "How large is this or that object?" although, in
5616respect to this question, we have various propositions synthetical and
5617immediately certain (indemonstrabilia); we have, in the proper sense of
5618the term, no axioms. For example, the propositions: "If equals be added
5619to equals, the wholes are equal"; "If equals be taken from equals,
5620the remainders are equal"; are analytical, because I am immediately
5621conscious of the identity of the production of the one quantity with
5622the production of the other; whereas axioms must be a priori synthetical
5623propositions. On the other hand, the self-evident propositions as to the
5624relation of numbers, are certainly synthetical but not universal, like
5625those of geometry, and for this reason cannot be called axioms, but
5626numerical formulae. That 7 + 5 = 12 is not an analytical proposition.
5627For neither in the representation of seven, nor of five, nor of the
5628composition of the two numbers, do I cogitate the number twelve.
5629(Whether I cogitate the number in the addition of both, is not at
5630present the question; for in the case of an analytical proposition,
5631the only point is whether I really cogitate the predicate in the
5632representation of the subject.) But although the proposition is
5633synthetical, it is nevertheless only a singular proposition. In so far
5634as regard is here had merely to the synthesis of the homogeneous (the
5635units), it cannot take place except in one manner, although our use
5636of these numbers is afterwards general. If I say: "A triangle can
5637be constructed with three lines, any two of which taken together are
5638greater than the third," I exercise merely the pure function of the
5639productive imagination, which may draw the lines longer or shorter and
5640construct the angles at its pleasure. On the contrary, the number seven
5641is possible only in one manner, and so is likewise the number twelve,
5642which results from the synthesis of seven and five. Such propositions,
5643then, cannot be termed axioms (for in that case we should have an
5644infinity of these), but numerical formulae.
5645
5646This transcendental principle of the mathematics of phenomena greatly
5647enlarges our a priori cognition. For it is by this principle alone that
5648pure mathematics is rendered applicable in all its precision to objects
5649of experience, and without it the validity of this application would not
5650be so self-evident; on the contrary, contradictions and confusions have
5651often arisen on this very point. Phenomena are not things in themselves.
5652Empirical intuition is possible only through pure intuition (of space
5653and time); consequently, what geometry affirms of the latter, is
5654indisputably valid of the former. All evasions, such as the statement
5655that objects of sense do not conform to the rules of construction in
5656space (for example, to the rule of the infinite divisibility of lines or
5657angles), must fall to the ground. For, if these objections hold good, we
5658deny to space, and with it to all mathematics, objective validity, and
5659no longer know wherefore, and how far, mathematics can be applied to
5660phenomena. The synthesis of spaces and times as the essential form of
5661all intuition, is that which renders possible the apprehension of a
5662phenomenon, and therefore every external experience, consequently all
5663cognition of the objects of experience; and whatever mathematics in its
5664pure use proves of the former, must necessarily hold good of the latter.
5665All objections are but the chicaneries of an ill-instructed reason,
5666which erroneously thinks to liberate the objects of sense from the
5667formal conditions of our sensibility, and represents these, although
5668mere phenomena, as things in themselves, presented as such to our
5669understanding. But in this case, no a priori synthetical cognition of
5670them could be possible, consequently not through pure conceptions of
5671space and the science which determines these conceptions, that is to
5672say, geometry, would itself be impossible.
5673
5674
5675
56762. ANTICIPATIONS OF PERCEPTION.
5677
5678The principle of these is: In all phenomena the Real, that which is an
5679object of sensation, has Intensive Quantity, that is, has a Degree.
5680
5681
5682PROOF.
5683
5684
5685Perception is empirical consciousness, that is to say, a consciousness
5686which contains an element of sensation. Phenomena as objects of
5687perception are not pure, that is, merely formal intuitions, like space
5688and time, for they cannot be perceived in themselves.
5689
5690
5691 [Footnote: They can be perceived only as phenomena, and some
5692 part of them must always belong to the non-ego; whereas pure
5693 intuitions are entirely the products of the mind itself, and
5694 as such are coguized IN THEMSELVES.--Tr]
5695
5696
5697They contain, then, over and above the intuition, the materials for an
5698object (through which is represented something existing in space or
5699time), that is to say, they contain the real of sensation, as a
5700representation merely subjective, which gives us merely the
5701consciousness that the subject is affected, and which we refer to some
5702external object. Now, a gradual transition from empirical consciousness
5703to pure consciousness is possible, inasmuch as the real in this
5704consciousness entirely vanishes, and there remains a merely formal
5705consciousness (a priori) of the manifold in time and space; consequently
5706there is possible a synthesis also of the production of the quantity of
5707a sensation from its commencement, that is, from the pure intuition = 0
5708onwards up to a certain quantity of the sensation. Now as sensation in
5709itself is not an objective representation, and in it is to be found
5710neither the intuition of space nor of time, it cannot possess any
5711extensive quantity, and yet there does belong to it a quantity (and that
5712by means of its apprehension, in which empirical consciousness can
5713within a certain time rise from nothing = 0 up to its given amount),
5714consequently an intensive quantity. And thus we must ascribe intensive
5715quantity, that is, a degree of influence on sense to all objects of
5716perception, in so far as this perception contains sensation.
5717
5718All cognition, by means of which I am enabled to cognize and determine
5719a priori what belongs to empirical cognition, may be called an
5720anticipation; and without doubt this is the sense in which Epicurus
5721employed his expression prholepsis. But as there is in phenomena
5722something which is never cognized a priori, which on this account
5723constitutes the proper difference between pure and empirical cognition,
5724that is to say, sensation (as the matter of perception), it follows,
5725that sensation is just that element in cognition which cannot be at
5726all anticipated. On the other hand, we might very well term the pure
5727determinations in space and time, as well in regard to figure as to
5728quantity, anticipations of phenomena, because they represent a priori
5729that which may always be given a posteriori in experience. But suppose
5730that in every sensation, as sensation in general, without any particular
5731sensation being thought of, there existed something which could be
5732cognized a priori, this would deserve to be called anticipation in a
5733special sense--special, because it may seem surprising to forestall
5734experience, in that which concerns the matter of experience, and which
5735we can only derive from itself. Yet such really is the case here.
5736
5737Apprehension*, by means of sensation alone, fills only one moment, that
5738is, if I do not take into consideration a succession of many sensations.
5739As that in the phenomenon, the apprehension of which is not a successive
5740synthesis advancing from parts to an entire representation, sensation
5741has therefore no extensive quantity; the want of sensation in a moment
5742of time would represent it as empty, consequently = 0. That which in
5743the empirical intuition corresponds to sensation is reality (realitas
5744phaenomenon); that which corresponds to the absence of it, negation = 0.
5745Now every sensation is capable of a diminution, so that it can decrease,
5746and thus gradually disappear. Therefore, between reality in a phenomenon
5747and negation, there exists a continuous concatenation of many possible
5748intermediate sensations, the difference of which from each other is
5749always smaller than that between the given sensation and zero, or
5750complete negation. That is to say, the real in a phenomenon has always a
5751quantity, which however is not discoverable in apprehension, inasmuch as
5752apprehension take place by means of mere sensation in one instant, and
5753not by the successive synthesis of many sensations, and therefore does
5754not progress from parts to the whole. Consequently, it has a quantity,
5755but not an extensive quantity.
5756
5757
5758 [*Footnote: Apprehension is the Kantian word for preception, in the
5759largest sense in which we employ that term. It is the genus which
5760includes under i, as species, perception proper and sensation
5761proper--Tr]
5762
5763
5764Now that quantity which is apprehended only as unity, and in which
5765plurality can be represented only by approximation to negation = O,
5766I term intensive quantity. Consequently, reality in a phenomenon has
5767intensive quantity, that is, a degree. If we consider this reality as
5768cause (be it of sensation or of another reality in the phenomenon, for
5769example, a change), we call the degree of reality in its character of
5770cause a momentum, for example, the momentum of weight; and for this
5771reason, that the degree only indicates that quantity the apprehension of
5772which is not successive, but instantaneous. This, however, I touch upon
5773only in passing, for with causality I have at present nothing to do.
5774
5775Accordingly, every sensation, consequently every reality in phenomena,
5776however small it may be, has a degree, that is, an intensive quantity,
5777which may always be lessened, and between reality and negation there
5778exists a continuous connection of possible realities, and possible
5779smaller perceptions. Every colour--for example, red--has a degree,
5780which, be it ever so small, is never the smallest, and so is it always
5781with heat, the momentum of weight, etc.
5782
5783This property of quantities, according to which no part of them is the
5784smallest possible (no part simple), is called their continuity. Space
5785and time are quanta continua, because no part of them can be
5786given, without enclosing it within boundaries (points and moments),
5787consequently, this given part is itself a space or a time. Space,
5788therefore, consists only of spaces, and time of times. Points and
5789moments are only boundaries, that is, the mere places or positions of
5790their limitation. But places always presuppose intuitions which are to
5791limit or determine them; and we cannot conceive either space or time
5792composed of constituent parts which are given before space or time.
5793Such quantities may also be called flowing, because synthesis (of the
5794productive imagination) in the production of these quantities is a
5795progression in time, the continuity of which we are accustomed to
5796indicate by the expression flowing.
5797
5798All phenomena, then, are continuous quantities, in respect both to
5799intuition and mere perception (sensation, and with it reality). In the
5800former case they are extensive quantities; in the latter, intensive.
5801When the synthesis of the manifold of a phenomenon is interrupted, there
5802results merely an aggregate of several phenomena, and not properly a
5803phenomenon as a quantity, which is not produced by the mere continuation
5804of the productive synthesis of a certain kind, but by the repetition of
5805a synthesis always ceasing. For example, if I call thirteen dollars a
5806sum or quantity of money, I employ the term quite correctly, inasmuch as
5807I understand by thirteen dollars the value of a mark in standard silver,
5808which is, to be sure, a continuous quantity, in which no part is the
5809smallest, but every part might constitute a piece of money, which would
5810contain material for still smaller pieces. If, however, by the words
5811thirteen dollars I understand so many coins (be their value in silver
5812what it may), it would be quite erroneous to use the expression a
5813quantity of dollars; on the contrary, I must call them aggregate, that
5814is, a number of coins. And as in every number we must have unity as the
5815foundation, so a phenomenon taken as unity is a quantity, and as such
5816always a continuous quantity (quantum continuum).
5817
5818Now, seeing all phenomena, whether considered as extensive or intensive,
5819are continuous quantities, the proposition: "All change (transition of a
5820thing from one state into another) is continuous," might be proved here
5821easily, and with mathematical evidence, were it not that the causality
5822of a change lies, entirely beyond the bounds of a transcendental
5823philosophy, and presupposes empirical principles. For of the possibility
5824of a cause which changes the condition of things, that is, which
5825determines them to the contrary to a certain given state, the
5826understanding gives us a priori no knowledge; not merely because it has
5827no insight into the possibility of it (for such insight is absent in
5828several a priori cognitions), but because the notion of change concerns
5829only certain determinations of phenomena, which experience alone can
5830acquaint us with, while their cause lies in the unchangeable. But seeing
5831that we have nothing which we could here employ but the pure fundamental
5832conceptions of all possible experience, among which of course nothing
5833empirical can be admitted, we dare not, without injuring the unity of
5834our system, anticipate general physical science, which is built upon
5835certain fundamental experiences.
5836
5837Nevertheless, we are in no want of proofs of the great influence
5838which the principle above developed exercises in the anticipation of
5839perceptions, and even in supplying the want of them, so far as to shield
5840us against the false conclusions which otherwise we might rashly draw.
5841
5842If all reality in perception has a degree, between which and negation
5843there is an endless sequence of ever smaller degrees, and if,
5844nevertheless, every sense must have a determinate degree of receptivity
5845for sensations; no perception, and consequently no experience is
5846possible, which can prove, either immediately or mediately, an entire
5847absence of all reality in a phenomenon; in other words, it is impossible
5848ever to draw from experience a proof of the existence of empty space or
5849of empty time. For in the first place, an entire absence of reality in
5850a sensuous intuition cannot of course be an object of perception;
5851secondly, such absence cannot be deduced from the contemplation of any
5852single phenomenon, and the difference of the degrees in its reality; nor
5853ought it ever to be admitted in explanation of any phenomenon. For if
5854even the complete intuition of a determinate space or time is thoroughly
5855real, that is, if no part thereof is empty, yet because every reality
5856has its degree, which, with the extensive quantity of the phenomenon
5857unchanged, can diminish through endless gradations down to nothing (the
5858void), there must be infinitely graduated degrees, with which space or
5859time is filled, and the intensive quantity in different phenomena may
5860be smaller or greater, although the extensive quantity of the intuition
5861remains equal and unaltered.
5862
5863We shall give an example of this. Almost all natural philosophers,
5864remarking a great difference in the quantity of the matter of different
5865kinds in bodies with the same volume (partly on account of the momentum
5866of gravity or weight, partly on account of the momentum of resistance
5867to other bodies in motion), conclude unanimously that this volume
5868(extensive quantity of the phenomenon) must be void in all bodies,
5869although in different proportion. But who would suspect that these for
5870the most part mathematical and mechanical inquirers into nature should
5871ground this conclusion solely on a metaphysical hypothesis--a sort of
5872hypothesis which they profess to disparage and avoid? Yet this they
5873do, in assuming that the real in space (I must not here call it
5874impenetrability or weight, because these are empirical conceptions)
5875is always identical, and can only be distinguished according to its
5876extensive quantity, that is, multiplicity. Now to this presupposition,
5877for which they can have no ground in experience, and which consequently
5878is merely metaphysical, I oppose a transcendental demonstration, which
5879it is true will not explain the difference in the filling up of spaces,
5880but which nevertheless completely does away with the supposed necessity
5881of the above-mentioned presupposition that we cannot explain the said
5882difference otherwise than by the hypothesis of empty spaces. This
5883demonstration, moreover, has the merit of setting the understanding
5884at liberty to conceive this distinction in a different manner, if the
5885explanation of the fact requires any such hypothesis. For we perceive
5886that although two equal spaces may be completely filled by matters
5887altogether different, so that in neither of them is there left a single
5888point wherein matter is not present, nevertheless, every reality has its
5889degree (of resistance or of weight), which, without diminution of the
5890extensive quantity, can become less and less ad infinitum, before it
5891passes into nothingness and disappears. Thus an expansion which fills
5892a space--for example, caloric, or any other reality in the phenomenal
5893world--can decrease in its degrees to infinity, yet without leaving the
5894smallest part of the space empty; on the contrary, filling it with those
5895lesser degrees as completely as another phenomenon could with greater.
5896My intention here is by no means to maintain that this is really
5897the case with the difference of matters, in regard to their
5898specific gravity; I wish only to prove, from a principle of the pure
5899understanding, that the nature of our perceptions makes such a mode of
5900explanation possible, and that it is erroneous to regard the real in
5901a phenomenon as equal quoad its degree, and different only quoad its
5902aggregation and extensive quantity, and this, too, on the pretended
5903authority of an a priori principle of the understanding.
5904
5905Nevertheless, this principle of the anticipation of perception must
5906somewhat startle an inquirer whom initiation into transcendental
5907philosophy has rendered cautious. We must naturally entertain some
5908doubt whether or not the understanding can enounce any such synthetical
5909proposition as that respecting the degree of all reality in phenomena,
5910and consequently the possibility of the internal difference of sensation
5911itself--abstraction being made of its empirical quality. Thus it is a
5912question not unworthy of solution: "How the understanding can pronounce
5913synthetically and a priori respecting phenomena, and thus anticipate
5914these, even in that which is peculiarly and merely empirical, that,
5915namely, which concerns sensation itself?"
5916
5917The quality of sensation is in all cases merely empirical, and cannot
5918be represented a priori (for example, colours, taste, etc.). But the
5919real--that which corresponds to sensation--in opposition to negation =
59200, only represents something the conception of which in itself contains
5921a being (ein seyn), and signifies nothing but the synthesis in an
5922empirical consciousness. That is to say, the empirical consciousness in
5923the internal sense can be raised from 0 to every higher degree, so that
5924the very same extensive quantity of intuition, an illuminated surface,
5925for example, excites as great a sensation as an aggregate of many other
5926surfaces less illuminated. We can therefore make complete abstraction
5927of the extensive quantity of a phenomenon, and represent to ourselves
5928in the mere sensation in a certain momentum, a synthesis of homogeneous
5929ascension from 0 up to the given empirical consciousness, All sensations
5930therefore as such are given only a posteriori, but this property
5931thereof, namely, that they have a degree, can be known a priori. It
5932is worthy of remark, that in respect to quantities in general, we can
5933cognize a priori only a single quality, namely, continuity; but in
5934respect to all quality (the real in phenomena), we cannot cognize a
5935priori anything more than the intensive quantity thereof, namely, that
5936they have a degree. All else is left to experience.
5937
5938
5939
59403. ANALOGIES OF EXPERIENCE.
5941
5942The principle of these is: Experience is possible only through the
5943representation of a necessary connection of Perceptions.
5944
5945
5946PROOF.
5947
5948
5949Experience is an empirical cognition; that is to say, a cognition
5950which determines an object by means of perceptions. It is therefore a
5951synthesis of perceptions, a synthesis which is not itself contained in
5952perception, but which contains the synthetical unity of the manifold of
5953perception in a consciousness; and this unity constitutes the essential
5954of our cognition of objects of the senses, that is, of experience (not
5955merely of intuition or sensation). Now in experience our perceptions
5956come together contingently, so that no character of necessity in their
5957connection appears, or can appear from the perceptions themselves,
5958because apprehension is only a placing together of the manifold of
5959empirical intuition, and no representation of a necessity in the
5960connected existence of the phenomena which apprehension brings together,
5961is to be discovered therein. But as experience is a cognition of objects
5962by means of perceptions, it follows that the relation of the existence
5963of the existence of the manifold must be represented in experience not
5964as it is put together in time, but as it is objectively in time. And as
5965time itself cannot be perceived, the determination of the existence of
5966objects in time can only take place by means of their connection in
5967time in general, consequently only by means of a priori connecting
5968conceptions. Now as these conceptions always possess the character of
5969necessity, experience is possible only by means of a representation of
5970the necessary connection of perception.
5971
5972The three modi of time are permanence, succession, and coexistence.
5973Accordingly, there are three rules of all relations of time in
5974phenomena, according to which the existence of every phenomenon is
5975determined in respect of the unity of all time, and these antecede all
5976experience and render it possible.
5977
5978The general principle of all three analogies rests on the necessary
5979unity of apperception in relation to all possible empirical
5980consciousness (perception) at every time, consequently, as this unity
5981lies a priori at the foundation of all mental operations, the principle
5982rests on the synthetical unity of all phenomena according to their
5983relation in time. For the original apperception relates to our internal
5984sense (the complex of all representations), and indeed relates a priori
5985to its form, that is to say, the relation of the manifold empirical
5986consciousness in time. Now this manifold must be combined in original
5987apperception according to relations of time--a necessity imposed by the
5988a priori transcendental unity of apperception, to which is subjected all
5989that can belong to my (i.e., my own) cognition, and therefore all that
5990can become an object for me. This synthetical and a priori determined
5991unity in relation of perceptions in time is therefore the rule: "All
5992empirical determinations of time must be subject to rules of the general
5993determination of time"; and the analogies of experience, of which we are
5994now about to treat, must be rules of this nature.
5995
5996These principles have this peculiarity, that they do not concern
5997phenomena, and the synthesis of the empirical intuition thereof, but
5998merely the existence of phenomena and their relation to each other in
5999regard to this existence. Now the mode in which we apprehend a thing in
6000a phenomenon can be determined a priori in such a manner that the rule
6001of its synthesis can give, that is to say, can produce this a priori
6002intuition in every empirical example. But the existence of phenomena
6003cannot be known a priori, and although we could arrive by this path at
6004a conclusion of the fact of some existence, we could not cognize that
6005existence determinately, that is to say, we should be incapable of
6006anticipating in what respect the empirical intuition of it would be
6007distinguishable from that of others.
6008
6009The two principles above mentioned, which I called mathematical, in
6010consideration of the fact of their authorizing the application of
6011mathematic phenomena, relate to these phenomena only in regard to their
6012possibility, and instruct us how phenomena, as far as regards their
6013intuition or the real in their perception, can be generated according
6014to the rules of a mathematical synthesis. Consequently, numerical
6015quantities, and with them the determination of a phenomenon as a
6016quantity, can be employed in the one case as well as in the other. Thus,
6017for example, out of 200,000 illuminations by the moon, I might compose
6018and give a priori, that is construct, the degree of our sensations
6019of the sun-light.* We may therefore entitle these two principles
6020constitutive.
6021
6022
6023 [*Footnote: Kant's meaning is: The two principles enunciated under
6024the heads of "Axioms of Intuition," and "Anticipations of Perception,"
6025authorize the application to phenomena of determinations of size and
6026number, that is of mathematic. For example, I may compute the light of
6027the sun, and say that its quantity is a certain number of times
6028greater than that of the moon. In the same way, heat is measured by the
6029comparison of its different effects on water, &c., and on mercury in a
6030thermometer.--Tr]
6031
6032
6033The case is very different with those principles whose province it is to
6034subject the existence of phenomena to rules a priori. For as existence
6035does not admit of being constructed, it is clear that they must only
6036concern the relations of existence and be merely regulative principles.
6037In this case, therefore, neither axioms nor anticipations are to be
6038thought of. Thus, if a perception is given us, in a certain relation of
6039time to other (although undetermined) perceptions, we cannot then say
6040a priori, what and how great (in quantity) the other perception
6041necessarily connected with the former is, but only how it is connected,
6042quoad its existence, in this given modus of time. Analogies in
6043philosophy mean something very different from that which they represent
6044in mathematics. In the latter they are formulae, which enounce the
6045equality of two relations of quantity, and are always constitutive, so
6046that if two terms of the proportion are given, the third is also
6047given, that is, can be constructed by the aid of these formulae. But in
6048philosophy, analogy is not the equality of two quantitative but of two
6049qualitative relations. In this case, from three given terms, I can
6050give a priori and cognize the relation to a fourth member, but not this
6051fourth term itself, although I certainly possess a rule to guide me in
6052the search for this fourth term in experience, and a mark to assist me
6053in discovering it. An analogy of experience is therefore only a rule
6054according to which unity of experience must arise out of perceptions in
6055respect to objects (phenomena) not as a constitutive, but merely as
6056a regulative principle. The same holds good also of the postulates of
6057empirical thought in general, which relate to the synthesis of mere
6058intuition (which concerns the form of phenomena), the synthesis of
6059perception (which concerns the matter of phenomena), and the synthesis
6060of experience (which concerns the relation of these perceptions). For
6061they are only regulative principles, and clearly distinguishable from
6062the mathematical, which are constitutive, not indeed in regard to the
6063certainty which both possess a priori, but in the mode of evidence
6064thereof, consequently also in the manner of demonstration.
6065
6066But what has been observed of all synthetical propositions, and must
6067be particularly remarked in this place, is this, that these
6068analogies possess significance and validity, not as principles of the
6069transcendental, but only as principles of the empirical use of the
6070understanding, and their truth can therefore be proved only as such, and
6071that consequently the phenomena must not be subjoined directly under the
6072categories, but only under their schemata. For if the objects to which
6073those principles must be applied were things in themselves, it would
6074be quite impossible to cognize aught concerning them synthetically a
6075priori. But they are nothing but phenomena; a complete knowledge
6076of which--a knowledge to which all principles a priori must at
6077last relate--is the only possible experience. It follows that these
6078principles can have nothing else for their aim than the conditions of
6079the empirical cognition in the unity of synthesis of phenomena. But this
6080synthesis is cogitated only in the schema of the pure conception of the
6081understanding, of whose unity, as that of a synthesis in general, the
6082category contains the function unrestricted by any sensuous condition.
6083These principles will therefore authorize us to connect phenomena
6084according to an analogy, with the logical and universal unity of
6085conceptions, and consequently to employ the categories in the principles
6086themselves; but in the application of them to experience, we shall use
6087only their schemata, as the key to their proper application, instead of
6088the categories, or rather the latter as restricting conditions, under
6089the title of "formulae" of the former.
6090
6091
6092
6093A. FIRST ANALOGY.
6094
6095Principle of the Permanence of Substance.
6096
6097In all changes of phenomena, substance is permanent, and the quantum
6098thereof in nature is neither increased nor diminished.
6099
6100
6101PROOF.
6102
6103
6104All phenomena exist in time, wherein alone as substratum, that is, as
6105the permanent form of the internal intuition, coexistence and succession
6106can be represented. Consequently time, in which all changes of phenomena
6107must be cogitated, remains and changes not, because it is that in which
6108succession and coexistence can be represented only as determinations
6109thereof. Now, time in itself cannot be an object of perception. It
6110follows that in objects of perception, that is, in phenomena, there must
6111be found a substratum which represents time in general, and in which
6112all change or coexistence can be perceived by means of the relation of
6113phenomena to it. But the substratum of all reality, that is, of all that
6114pertains to the existence of things, is substance; all that pertains
6115to existence can be cogitated only as a determination of substance.
6116Consequently, the permanent, in relation to which alone can all
6117relations of time in phenomena be determined, is substance in the
6118world of phenomena, that is, the real in phenomena, that which, as the
6119substratum of all change, remains ever the same. Accordingly, as this
6120cannot change in existence, its quantity in nature can neither be
6121increased nor diminished.
6122
6123Our apprehension of the manifold in a phenomenon is always successive,
6124is Consequently always changing. By it alone we could, therefore,
6125never determine whether this manifold, as an object of experience, is
6126coexistent or successive, unless it had for a foundation something fixed
6127and permanent, of the existence of which all succession and coexistence
6128are nothing but so many modes (modi of time). Only in the permanent,
6129then, are relations of time possible (for simultaneity and succession
6130are the only relations in time); that is to say, the permanent is the
6131substratum of our empirical representation of time itself, in which
6132alone all determination of time is possible. Permanence is, in fact,
6133just another expression for time, as the abiding correlate of all
6134existence of phenomena, and of all change, and of all coexistence. For
6135change does not affect time itself, but only the phenomena in time (just
6136as coexistence cannot be regarded as a modus of time itself, seeing
6137that in time no parts are coexistent, but all successive). If we were
6138to attribute succession to time itself, we should be obliged to cogitate
6139another time, in which this succession would be possible. It is only
6140by means of the permanent that existence in different parts of the
6141successive series of time receives a quantity, which we entitle
6142duration. For in mere succession, existence is perpetually vanishing and
6143recommencing, and therefore never has even the least quantity. Without
6144the permanent, then, no relation in time is possible. Now, time in
6145itself is not an object of perception; consequently the permanent in
6146phenomena must be regarded as the substratum of all determination of
6147time, and consequently also as the condition of the possibility of
6148all synthetical unity of perceptions, that is, of experience; and all
6149existence and all change in time can only be regarded as a mode in
6150the existence of that which abides unchangeably. Therefore, in all
6151phenomena, the permanent is the object in itself, that is, the substance
6152(phenomenon); but all that changes or can change belongs only to the
6153mode of the existence of this substance or substances, consequently to
6154its determinations.
6155
6156I find that in all ages not only the philosopher, but even the common
6157understanding, has preposited this permanence as a substratum of all
6158change in phenomena; indeed, I am compelled to believe that they
6159will always accept this as an indubitable fact. Only the philosopher
6160expresses himself in a more precise and definite manner, when he says:
6161"In all changes in the world, the substance remains, and the accidents
6162alone are changeable." But of this decidedly synthetical proposition, I
6163nowhere meet with even an attempt at proof; nay, it very rarely has the
6164good fortune to stand, as it deserves to do, at the head of the pure and
6165entirely a priori laws of nature. In truth, the statement that substance
6166is permanent, is tautological. For this very permanence is the ground
6167on which we apply the category of substance to the phenomenon; and
6168we should have been obliged to prove that in all phenomena there is
6169something permanent, of the existence of which the changeable is nothing
6170but a determination. But because a proof of this nature cannot be
6171dogmatical, that is, cannot be drawn from conceptions, inasmuch as it
6172concerns a synthetical proposition a priori, and as philosophers never
6173reflected that such propositions are valid only in relation to possible
6174experience, and therefore cannot be proved except by means of a
6175deduction of the possibility of experience, it is no wonder that while
6176it has served as the foundation of all experience (for we feel the need
6177of it in empirical cognition), it has never been supported by proof.
6178
6179A philosopher was asked: "What is the weight of smoke?" He answered:
6180"Subtract from the weight of the burnt wood the weight of the remaining
6181ashes, and you will have the weight of the smoke." Thus he presumed it
6182to be incontrovertible that even in fire the matter (substance) does not
6183perish, but that only the form of it undergoes a change. In like manner
6184was the saying: "From nothing comes nothing," only another inference
6185from the principle or permanence, or rather of the ever-abiding
6186existence of the true subject in phenomena. For if that in the
6187phenomenon which we call substance is to be the proper substratum of all
6188determination of time, it follows that all existence in past as well as
6189in future time, must be determinable by means of it alone. Hence we are
6190entitled to apply the term substance to a phenomenon, only because we
6191suppose its existence in all time, a notion which the word permanence
6192does not fully express, as it seems rather to be referable to future
6193time. However, the internal necessity perpetually to be, is inseparably
6194connected with the necessity always to have been, and so the expression
6195may stand as it is. "Gigni de nihilo nihil; in nihilum nil posse
6196reverti,"* are two propositions which the ancients never parted, and
6197which people nowadays sometimes mistakenly disjoin, because they imagine
6198that the propositions apply to objects as things in themselves, and that
6199the former might be inimical to the dependence (even in respect of its
6200substance also) of the world upon a supreme cause. But this apprehension
6201is entirely needless, for the question in this case is only of phenomena
6202in the sphere of experience, the unity of which never could be possible,
6203if we admitted the possibility that new things (in respect of their
6204substance) should arise. For in that case, we should lose altogether
6205that which alone can represent the unity of time, to wit, the identity
6206of the substratum, as that through which alone all change possesses
6207complete and thorough unity. This permanence is, however, nothing but
6208the manner in which we represent to ourselves the existence of things in
6209the phenomenal world.
6210
6211
6212 [*Footnote: Persius, Satirae, iii.83-84.]
6213
6214
6215The determinations of a substance, which are only particular modes of
6216its existence, are called accidents. They are always real, because they
6217concern the existence of substance (negations are only determinations,
6218which express the non-existence of something in the substance). Now,
6219if to this real in the substance we ascribe a particular existence (for
6220example, to motion as an accident of matter), this existence is called
6221inherence, in contradistinction to the existence of substance, which we
6222call subsistence. But hence arise many misconceptions, and it would be a
6223more accurate and just mode of expression to designate the accident
6224only as the mode in which the existence of a substance is positively
6225determined. Meanwhile, by reason of the conditions of the logical
6226exercise of our understanding, it is impossible to avoid separating,
6227as it were, that which in the existence of a substance is subject to
6228change, whilst the substance remains, and regarding it in relation to
6229that which is properly permanent and radical. On this account, this
6230category of substance stands under the title of relation, rather because
6231it is the condition thereof than because it contains in itself any
6232relation.
6233
6234Now, upon this notion of permanence rests the proper notion of the
6235conception change. Origin and extinction are not changes of that which
6236originates or becomes extinct. Change is but a mode of existence, which
6237follows on another mode of existence of the same object; hence all that
6238changes is permanent, and only the condition thereof changes. Now since
6239this mutation affects only determinations, which can have a beginning
6240or an end, we may say, employing an expression which seems somewhat
6241paradoxical: "Only the permanent (substance) is subject to change; the
6242mutable suffers no change, but rather alternation, that is, when certain
6243determinations cease, others begin."
6244
6245Change, when, cannot be perceived by us except in substances, and origin
6246or extinction in an absolute sense, that does not concern merely a
6247determination of the permanent, cannot be a possible perception, for
6248it is this very notion of the permanent which renders possible the
6249representation of a transition from one state into another, and from
6250non-being to being, which, consequently, can be empirically cognized
6251only as alternating determinations of that which is permanent. Grant
6252that a thing absolutely begins to be; we must then have a point of time
6253in which it was not. But how and by what can we fix and determine
6254this point of time, unless by that which already exists? For a void
6255time--preceding--is not an object of perception; but if we connect this
6256beginning with objects which existed previously, and which continue to
6257exist till the object in question in question begins to be, then the
6258latter can only be a determination of the former as the permanent. The
6259same holds good of the notion of extinction, for this presupposes the
6260empirical representation of a time, in which a phenomenon no longer
6261exists.
6262
6263Substances (in the world of phenomena) are the substratum of all
6264determinations of time. The beginning of some, and the ceasing to be of
6265other substances, would utterly do away with the only condition of the
6266empirical unity of time; and in that case phenomena would relate to two
6267different times, in which, side by side, existence would pass; which is
6268absurd. For there is only one time in which all different times must be
6269placed, not as coexistent, but as successive.
6270
6271Accordingly, permanence is a necessary condition under which alone
6272phenomena, as things or objects, are determinable in a possible
6273experience. But as regards the empirical criterion of this necessary
6274permanence, and with it of the substantiality of phenomena, we shall
6275find sufficient opportunity to speak in the sequel.
6276
6277
6278
6279B. SECOND ANALOGY.
6280
6281Principle of the Succession of Time According to the Law of Causality.
6282All changes take place according to the law of the connection of Cause
6283and Effect.
6284
6285
6286PROOF.
6287
6288
6289(That all phenomena in the succession of time are only changes, that
6290is, a successive being and non-being of the determinations of substance,
6291which is permanent; consequently that a being of substance itself which
6292follows on the non-being thereof, or a non-being of substance which
6293follows on the being thereof, in other words, that the origin or
6294extinction of substance itself, is impossible--all this has been fully
6295established in treating of the foregoing principle. This principle
6296might have been expressed as follows: "All alteration (succession) of
6297phenomena is merely change"; for the changes of substance are not origin
6298or extinction, because the conception of change presupposes the same
6299subject as existing with two opposite determinations, and consequently
6300as permanent. After this premonition, we shall proceed to the proof.)
6301
6302I perceive that phenomena succeed one another, that is to say, a state
6303of things exists at one time, the opposite of which existed in a former
6304state. In this case, then, I really connect together two perceptions in
6305time. Now connection is not an operation of mere sense and intuition,
6306but is the product of a synthetical faculty of imagination, which
6307determines the internal sense in respect of a relation of time. But
6308imagination can connect these two states in two ways, so that either the
6309one or the other may antecede in time; for time in itself cannot be an
6310object of perception, and what in an object precedes and what follows
6311cannot be empirically determined in relation to it. I am only conscious,
6312then, that my imagination places one state before and the other after;
6313not that the one state antecedes the other in the object. In other
6314words, the objective relation of the successive phenomena remains
6315quite undetermined by means of mere perception. Now in order that this
6316relation may be cognized as determined, the relation between the two
6317states must be so cogitated that it is thereby determined as necessary,
6318which of them must be placed before and which after, and not conversely.
6319But the conception which carries with it a necessity of synthetical
6320unity, can be none other than a pure conception of the understanding
6321which does not lie in mere perception; and in this case it is the
6322conception of "the relation of cause and effect," the former of which
6323determines the latter in time, as its necessary consequence, and not as
6324something which might possibly antecede (or which might in some cases
6325not be perceived to follow). It follows that it is only because we
6326subject the sequence of phenomena, and consequently all change, to the
6327law of causality, that experience itself, that is, empirical cognition
6328of phenomena, becomes possible; and consequently, that phenomena
6329themselves, as objects of experience, are possible only by virtue of
6330this law.
6331
6332Our apprehension of the manifold of phenomena is always successive. The
6333representations of parts succeed one another. Whether they succeed one
6334another in the object also, is a second point for reflection, which
6335was not contained in the former. Now we may certainly give the name of
6336object to everything, even to every representation, so far as we are
6337conscious thereof; but what this word may mean in the case of phenomena,
6338not merely in so far as they (as representations) are objects, but only
6339in so far as they indicate an object, is a question requiring deeper
6340consideration. In so far as they, regarded merely as representations,
6341are at the same time objects of consciousness, they are not to be
6342distinguished from apprehension, that is, reception into the synthesis
6343of imagination, and we must therefore say: "The manifold of phenomena is
6344always produced successively in the mind." If phenomena were things in
6345themselves, no man would be able to conjecture from the succession of
6346our representations how this manifold is connected in the object; for
6347we have to do only with our representations. How things may be in
6348themselves, without regard to the representations through which they
6349affect us, is utterly beyond the sphere of our cognition. Now although
6350phenomena are not things in themselves, and are nevertheless the only
6351thing given to us to be cognized, it is my duty to show what sort of
6352connection in time belongs to the manifold in phenomena themselves,
6353while the representation of this manifold in apprehension is always
6354successive. For example, the apprehension of the manifold in the
6355phenomenon of a house which stands before me, is successive. Now
6356comes the question whether the manifold of this house is in itself
6357successive--which no one will be at all willing to grant. But, so
6358soon as I raise my conception of an object to the transcendental
6359signification thereof, I find that the house is not a thing in itself,
6360but only a phenomenon, that is, a representation, the transcendental
6361object of which remains utterly unknown. What then am I to understand
6362by the question: "How can the manifold be connected in the phenomenon
6363itself--not considered as a thing in itself, but merely as a
6364phenomenon?" Here that which lies in my successive apprehension is
6365regarded as representation, whilst the phenomenon which is given
6366me, notwithstanding that it is nothing more than a complex of these
6367representations, is regarded as the object thereof, with which my
6368conception, drawn from the representations of apprehension, must
6369harmonize. It is very soon seen that, as accordance of the cognition
6370with its object constitutes truth, the question now before us can
6371only relate to the formal conditions of empirical truth; and that the
6372phenomenon, in opposition to the representations of apprehension, can
6373only be distinguished therefrom as the object of them, if it is subject
6374to a rule which distinguishes it from every other apprehension, and
6375which renders necessary a mode of connection of the manifold. That in
6376the phenomenon which contains the condition of this necessary rule of
6377apprehension, is the object.
6378
6379Let us now proceed to our task. That something happens, that is to say,
6380that something or some state exists which before was not, cannot be
6381empirically perceived, unless a phenomenon precedes, which does not
6382contain in itself this state. For a reality which should follow upon
6383a void time, in other words, a beginning, which no state of things
6384precedes, can just as little be apprehended as the void time itself.
6385Every apprehension of an event is therefore a perception which follows
6386upon another perception. But as this is the case with all synthesis
6387of apprehension, as I have shown above in the example of a house, my
6388apprehension of an event is not yet sufficiently distinguished from
6389other apprehensions. But I remark also that if in a phenomenon which
6390contains an occurrence, I call the antecedent state of my perception,
6391A, and the following state, B, the perception B can only follow A in
6392apprehension, and the perception A cannot follow B, but only precede
6393it. For example, I see a ship float down the stream of a river. My
6394perception of its place lower down follows upon my perception of its
6395place higher up the course of the river, and it is impossible that,
6396in the apprehension of this phenomenon, the vessel should be perceived
6397first below and afterwards higher up the stream. Here, therefore, the
6398order in the sequence of perceptions in apprehension is determined;
6399and by this order apprehension is regulated. In the former example, my
6400perceptions in the apprehension of a house might begin at the roof and
6401end at the foundation, or vice versa; or I might apprehend the manifold
6402in this empirical intuition, by going from left to right, and from right
6403to left. Accordingly, in the series of these perceptions, there was no
6404determined order, which necessitated my beginning at a certain point, in
6405order empirically to connect the manifold. But this rule is always to be
6406met with in the perception of that which happens, and it makes the order
6407of the successive perceptions in the apprehension of such a phenomenon
6408necessary.
6409
6410I must, therefore, in the present case, deduce the subjective sequence
6411of apprehension from the objective sequence of phenomena, for
6412otherwise the former is quite undetermined, and one phenomenon is not
6413distinguishable from another. The former alone proves nothing as to the
6414connection of the manifold in an object, for it is quite arbitrary.
6415The latter must consist in the order of the manifold in a phenomenon,
6416according to which order the apprehension of one thing (that which
6417happens) follows that of another thing (which precedes), in conformity
6418with a rule. In this way alone can I be authorized to say of the
6419phenomenon itself, and not merely of my own apprehension, that a certain
6420order or sequence is to be found therein. That is, in other words, I
6421cannot arrange my apprehension otherwise than in this order.
6422
6423In conformity with this rule, then, it is necessary that in that which
6424antecedes an event there be found the condition of a rule, according to
6425which in this event follows always and necessarily; but I cannot reverse
6426this and go back from the event, and determine (by apprehension) that
6427which antecedes it. For no phenomenon goes back from the succeeding
6428point of time to the preceding point, although it does certainly relate
6429to a preceding point of time; from a given time, on the other hand,
6430there is always a necessary progression to the determined succeeding
6431time. Therefore, because there certainly is something that follows, I
6432must of necessity connect it with something else, which antecedes, and
6433upon which it follows, in conformity with a rule, that is necessarily,
6434so that the event, as conditioned, affords certain indication of a
6435condition, and this condition determines the event.
6436
6437Let us suppose that nothing precedes an event, upon which this event
6438must follow in conformity with a rule. All sequence of perception
6439would then exist only in apprehension, that is to say, would be merely
6440subjective, and it could not thereby be objectively determined what
6441thing ought to precede, and what ought to follow in perception. In such
6442a case, we should have nothing but a play of representations, which
6443would possess no application to any object. That is to say, it would
6444not be possible through perception to distinguish one phenomenon from
6445another, as regards relations of time; because the succession in the act
6446of apprehension would always be of the same sort, and therefore there
6447would be nothing in the phenomenon to determine the succession, and to
6448render a certain sequence objectively necessary. And, in this case, I
6449cannot say that two states in a phenomenon follow one upon the other,
6450but only that one apprehension follows upon another. But this is merely
6451subjective, and does not determine an object, and consequently cannot be
6452held to be cognition of an object--not even in the phenomenal world.
6453
6454Accordingly, when we know in experience that something happens, we
6455always presuppose that something precedes, whereupon it follows in
6456conformity with a rule. For otherwise I could not say of the object that
6457it follows; because the mere succession in my apprehension, if it be
6458not determined by a rule in relation to something preceding, does not
6459authorize succession in the object. Only, therefore, in reference to
6460a rule, according to which phenomena are determined in their sequence,
6461that is, as they happen, by the preceding state, can I make my
6462subjective synthesis (of apprehension) objective, and it is only under
6463this presupposition that even the experience of an event is possible.
6464
6465No doubt it appears as if this were in thorough contradiction to all
6466the notions which people have hitherto entertained in regard to the
6467procedure of the human understanding. According to these opinions, it
6468is by means of the perception and comparison of similar consequences
6469following upon certain antecedent phenomena that the understanding is
6470led to the discovery of a rule, according to which certain events always
6471follow certain phenomena, and it is only by this process that we attain
6472to the conception of cause. Upon such a basis, it is clear that this
6473conception must be merely empirical, and the rule which it furnishes
6474us with--"Everything that happens must have a cause"--would be just as
6475contingent as experience itself. The universality and necessity of the
6476rule or law would be perfectly spurious attributes of it. Indeed, it
6477could not possess universal validity, inasmuch as it would not in this
6478case be a priori, but founded on deduction. But the same is the case
6479with this law as with other pure a priori representations (e.g., space
6480and time), which we can draw in perfect clearness and completeness from
6481experience, only because we had already placed them therein, and by that
6482means, and by that alone, had rendered experience possible. Indeed,
6483the logical clearness of this representation of a rule, determining
6484the series of events, is possible only when we have made use thereof in
6485experience. Nevertheless, the recognition of this rule, as a condition
6486of the synthetical unity of phenomena in time, was the ground of
6487experience itself and consequently preceded it a priori.
6488
6489It is now our duty to show by an example that we never, even in
6490experience, attribute to an object the notion of succession or effect
6491(of an event--that is, the happening of something that did not
6492exist before), and distinguish it from the subjective succession of
6493apprehension, unless when a rule lies at the foundation, which compels
6494us to observe this order of perception in preference to any other, and
6495that, indeed, it is this necessity which first renders possible the
6496representation of a succession in the object.
6497
6498We have representations within us, of which also we can be conscious.
6499But, however widely extended, however accurate and thoroughgoing this
6500consciousness may be, these representations are still nothing more than
6501representations, that is, internal determinations of the mind in this or
6502that relation of time. Now how happens it that to these representations
6503we should set an object, or that, in addition to their subjective
6504reality, as modifications, we should still further attribute to them
6505a certain unknown objective reality? It is clear that objective
6506significancy cannot consist in a relation to another representation
6507(of that which we desire to term object), for in that case the question
6508again arises: "How does this other representation go out of itself, and
6509obtain objective significancy over and above the subjective, which
6510is proper to it, as a determination of a state of mind?" If we try to
6511discover what sort of new property the relation to an object gives to
6512our subjective representations, and what new importance they thereby
6513receive, we shall find that this relation has no other effect than
6514that of rendering necessary the connection of our representations in a
6515certain manner, and of subjecting them to a rule; and that conversely,
6516it is only because a certain order is necessary in the relations of time
6517of our representations, that objective significancy is ascribed to them.
6518
6519In the synthesis of phenomena, the manifold of our representations is
6520always successive. Now hereby is not represented an object, for by means
6521of this succession, which is common to all apprehension, no one thing is
6522distinguished from another. But so soon as I perceive or assume that in
6523this succession there is a relation to a state antecedent, from which
6524the representation follows in accordance with a rule, so soon do I
6525represent something as an event, or as a thing that happens; in other
6526words, I cognize an object to which I must assign a certain determinate
6527position in time, which cannot be altered, because of the preceding
6528state in the object. When, therefore, I perceive that something happens,
6529there is contained in this representation, in the first place, the fact,
6530that something antecedes; because, it is only in relation to this that
6531the phenomenon obtains its proper relation of time, in other words,
6532exists after an antecedent time, in which it did not exist. But it can
6533receive its determined place in time only by the presupposition
6534that something existed in the foregoing state, upon which it follows
6535inevitably and always, that is, in conformity with a rule. From all this
6536it is evident that, in the first place, I cannot reverse the order
6537of succession, and make that which happens precede that upon which
6538it follows; and that, in the second place, if the antecedent state be
6539posited, a certain determinate event inevitably and necessarily
6540follows. Hence it follows that there exists a certain order in our
6541representations, whereby the present gives a sure indication of some
6542previously existing state, as a correlate, though still undetermined, of
6543the existing event which is given--a correlate which itself relates to
6544the event as its consequence, conditions it, and connects it necessarily
6545with itself in the series of time.
6546
6547If then it be admitted as a necessary law of sensibility, and
6548consequently a formal condition of all perception, that the preceding
6549necessarily determines the succeeding time (inasmuch as I cannot arrive
6550at the succeeding except through the preceding), it must likewise be an
6551indispensable law of empirical representation of the series of time
6552that the phenomena of the past determine all phenomena in the succeeding
6553time, and that the latter, as events, cannot take place, except in so
6554far as the former determine their existence in time, that is to say,
6555establish it according to a rule. For it is of course only in phenomena
6556that we can empirically cognize this continuity in the connection of
6557times.
6558
6559For all experience and for the possibility of experience, understanding
6560is indispensable, and the first step which it takes in this sphere is
6561not to render the representation of objects clear, but to render the
6562representation of an object in general, possible. It does this by
6563applying the order of time to phenomena, and their existence. In other
6564words, it assigns to each phenomenon, as a consequence, a place in
6565relation to preceding phenomena, determined a priori in time, without
6566which it could not harmonize with time itself, which determines a place
6567a priori to all its parts. This determination of place cannot be derived
6568from the relation of phenomena to absolute time (for it is not an
6569object of perception); but, on the contrary, phenomena must reciprocally
6570determine the places in time of one another, and render these necessary
6571in the order of time. In other words, whatever follows or happens, must
6572follow in conformity with a universal rule upon that which was contained
6573in the foregoing state. Hence arises a series of phenomena, which, by
6574means of the understanding, produces and renders necessary exactly
6575the same order and continuous connection in the series of our possible
6576perceptions, as is found a priori in the form of internal intuition
6577(time), in which all our perceptions must have place.
6578
6579That something happens, then, is a perception which belongs to a
6580possible experience, which becomes real only because I look upon the
6581phenomenon as determined in regard to its place in time, consequently as
6582an object, which can always be found by means of a rule in the connected
6583series of my perceptions. But this rule of the determination of a thing
6584according to succession in time is as follows: "In what precedes may be
6585found the condition, under which an event always (that is, necessarily)
6586follows." From all this it is obvious that the principle of cause and
6587effect is the principle of possible experience, that is, of objective
6588cognition of phenomena, in regard to their relations in the succession
6589of time.
6590
6591The proof of this fundamental proposition rests entirely on the
6592following momenta of argument. To all empirical cognition belongs the
6593synthesis of the manifold by the imagination, a synthesis which is
6594always successive, that is, in which the representations therein always
6595follow one another. But the order of succession in imagination is not
6596determined, and the series of successive representations may be taken
6597retrogressively as well as progressively. But if this synthesis is a
6598synthesis of apprehension (of the manifold of a given phenomenon), then
6599the order is determined in the object, or to speak more accurately,
6600there is therein an order of successive synthesis which determines an
6601object, and according to which something necessarily precedes, and
6602when this is posited, something else necessarily follows. If, then,
6603my perception is to contain the cognition of an event, that is, of
6604something which really happens, it must be an empirical judgement,
6605wherein we think that the succession is determined; that is, it
6606presupposes another phenomenon, upon which this event follows
6607necessarily, or in conformity with a rule. If, on the contrary, when I
6608posited the antecedent, the event did not necessarily follow, I should
6609be obliged to consider it merely as a subjective play of my imagination,
6610and if in this I represented to myself anything as objective, I must
6611look upon it as a mere dream. Thus, the relation of phenomena (as
6612possible perceptions), according to which that which happens is, as
6613to its existence, necessarily determined in time by something which
6614antecedes, in conformity with a rule--in other words, the relation of
6615cause and effect--is the condition of the objective validity of
6616our empirical judgements in regard to the sequence of perceptions,
6617consequently of their empirical truth, and therefore of experience. The
6618principle of the relation of causality in the succession of phenomena is
6619therefore valid for all objects of experience, because it is itself the
6620ground of the possibility of experience.
6621
6622Here, however, a difficulty arises, which must be resolved. The
6623principle of the connection of causality among phenomena is limited in
6624our formula to the succession thereof, although in practice we find that
6625the principle applies also when the phenomena exist together in the same
6626time, and that cause and effect may be simultaneous. For example, there
6627is heat in a room, which does not exist in the open air. I look about
6628for the cause, and find it to be the fire, Now the fire as the cause is
6629simultaneous with its effect, the heat of the room. In this case, then,
6630there is no succession as regards time, between cause and effect, but
6631they are simultaneous; and still the law holds good. The greater part of
6632operating causes in nature are simultaneous with their effects, and
6633the succession in time of the latter is produced only because the cause
6634cannot achieve the total of its effect in one moment. But at the
6635moment when the effect first arises, it is always simultaneous with the
6636causality of its cause, because, if the cause had but a moment before
6637ceased to be, the effect could not have arisen. Here it must be
6638specially remembered that we must consider the order of time and not the
6639lapse thereof. The relation remains, even though no time has elapsed.
6640The time between the causality of the cause and its immediate effect may
6641entirely vanish, and the cause and effect be thus simultaneous, but the
6642relation of the one to the other remains always determinable according
6643to time. If, for example, I consider a leaden ball, which lies upon a
6644cushion and makes a hollow in it, as a cause, then it is simultaneous
6645with the effect. But I distinguish the two through the relation of time
6646of the dynamical connection of both. For if I lay the ball upon the
6647cushion, then the hollow follows upon the before smooth surface; but
6648supposing the cushion has, from some cause or another, a hollow, there
6649does not thereupon follow a leaden ball.
6650
6651Thus, the law of succession of time is in all instances the only
6652empirical criterion of effect in relation to the causality of the
6653antecedent cause. The glass is the cause of the rising of the
6654water above its horizontal surface, although the two phenomena are
6655contemporaneous. For, as soon as I draw some water with the glass from
6656a larger vessel, an effect follows thereupon, namely, the change of the
6657horizontal state which the water had in the large vessel into a concave,
6658which it assumes in the glass.
6659
6660This conception of causality leads us to the conception of action; that
6661of action, to the conception of force; and through it, to the conception
6662of substance. As I do not wish this critical essay, the sole purpose of
6663which is to treat of the sources of our synthetical cognition a priori,
6664to be crowded with analyses which merely explain, but do not enlarge
6665the sphere of our conceptions, I reserve the detailed explanation of the
6666above conceptions for a future system of pure reason. Such an analysis,
6667indeed, executed with great particularity, may already be found in
6668well-known works on this subject. But I cannot at present refrain from
6669making a few remarks on the empirical criterion of a substance, in so
6670far as it seems to be more evident and more easily recognized through
6671the conception of action than through that of the permanence of a
6672phenomenon.
6673
6674Where action (consequently activity and force) exists, substance also
6675must exist, and in it alone must be sought the seat of that fruitful
6676source of phenomena. Very well. But if we are called upon to explain
6677what we mean by substance, and wish to avoid the vice of reasoning in
6678a circle, the answer is by no means so easy. How shall we conclude
6679immediately from the action to the permanence of that which acts, this
6680being nevertheless an essential and peculiar criterion of substance
6681(phenomenon)? But after what has been said above, the solution of
6682this question becomes easy enough, although by the common mode
6683of procedure--merely analysing our conceptions--it would be quite
6684impossible. The conception of action indicates the relation of the
6685subject of causality to the effect. Now because all effect consists
6686in that which happens, therefore in the changeable, the last subject
6687thereof is the permanent, as the substratum of all that changes, that
6688is, substance. For according to the principle of causality, actions are
6689always the first ground of all change in phenomena and, consequently,
6690cannot be a property of a subject which itself changes, because if this
6691were the case, other actions and another subject would be necessary to
6692determine this change. From all this it results that action alone, as
6693an empirical criterion, is a sufficient proof of the presence of
6694substantiality, without any necessity on my part of endeavouring to
6695discover the permanence of substance by a comparison. Besides, by this
6696mode of induction we could not attain to the completeness which the
6697magnitude and strict universality of the conception requires. For that
6698the primary subject of the causality of all arising and passing away,
6699all origin and extinction, cannot itself (in the sphere of phenomena)
6700arise and pass away, is a sound and safe conclusion, a conclusion which
6701leads us to the conception of empirical necessity and permanence
6702in existence, and consequently to the conception of a substance as
6703phenomenon.
6704
6705When something happens, the mere fact of the occurrence, without
6706regard to that which occurs, is an object requiring investigation.
6707The transition from the non-being of a state into the existence of it,
6708supposing that this state contains no quality which previously existed
6709in the phenomenon, is a fact of itself demanding inquiry. Such an event,
6710as has been shown in No. A, does not concern substance (for substance
6711does not thus originate), but its condition or state. It is therefore
6712only change, and not origin from nothing. If this origin be regarded as
6713the effect of a foreign cause, it is termed creation, which cannot be
6714admitted as an event among phenomena, because the very possibility of
6715it would annihilate the unity of experience. If, however, I regard all
6716things not as phenomena, but as things in themselves and objects of
6717understanding alone, they, although substances, may be considered as
6718dependent, in respect of their existence, on a foreign cause. But this
6719would require a very different meaning in the words, a meaning which
6720could not apply to phenomena as objects of possible experience.
6721
6722How a thing can be changed, how it is possible that upon one state
6723existing in one point of time, an opposite state should follow in
6724another point of time--of this we have not the smallest conception a
6725priori. There is requisite for this the knowledge of real powers, which
6726can only be given empirically; for example, knowledge of moving forces,
6727or, in other words, of certain successive phenomena (as movements) which
6728indicate the presence of such forces. But the form of every change,
6729the condition under which alone it can take place as the coming into
6730existence of another state (be the content of the change, that is, the
6731state which is changed, what it may), and consequently the succession of
6732the states themselves can very well be considered a priori, in relation
6733to the law of causality and the conditions of time.*
6734
6735
6736 [*Footnote: It must be remarked that I do not speak of the change of
6737certain relations, but of the change of the state. Thus, when a body
6738moves in a uniform manner, it does not change its state (of motion); but
6739only when all motion increases or decreases.]
6740
6741
6742When a substance passes from one state, a, into another state, b,
6743the point of time in which the latter exists is different from, and
6744subsequent to that in which the former existed. In like manner, the
6745second state, as reality (in the phenomenon), differs from the first, in
6746which the reality of the second did not exist, as b from zero. That is
6747to say, if the state, b, differs from the state, a, only in respect to
6748quantity, the change is a coming into existence of b -a, which in the
6749former state did not exist, and in relation to which that state is = O.
6750
6751Now the question arises how a thing passes from one state = a, into
6752another state = b. Between two moments there is always a certain time,
6753and between two states existing in these moments there is always a
6754difference having a certain quantity (for all parts of phenomena are in
6755their turn quantities). Consequently, every transition from one state
6756into another is always effected in a time contained between two moments,
6757of which the first determines the state which leaves, and the second
6758determines the state into the thing passes. The thing leaves, and the
6759second determines the state into which the thing Both moments, then, are
6760limitations of the time of a change, consequently of the intermediate
6761state between both, and as such they belong to the total of the change.
6762Now every change has a cause, which evidences its causality in the whole
6763time during which the charge takes place. The cause, therefore, does not
6764produce the change all at once or in one moment, but in a time, so that,
6765as the time gradually increases from the commencing instant, a, to its
6766completion at b, in like manner also, the quantity of the reality (b -
6767a) is generated through the lesser degrees which are contained between
6768the first and last. All change is therefore possible only through a
6769continuous action of the causality, which, in so far as it is uniform,
6770we call a momentum. The change does not consist of these momenta, but is
6771generated or produced by them as their effect.
6772
6773Such is the law of the continuity of all change, the ground of which is
6774that neither time itself nor any phenomenon in time consists of parts
6775which are the smallest possible, but that, notwithstanding, the state
6776of a thing passes in the process of a change through all these parts, as
6777elements, to its second state. There is no smallest degree of reality
6778in a phenomenon, just as there is no smallest degree in the quantity of
6779time; and so the new state of reality grows up out of the former state,
6780through all the infinite degrees thereof, the differences of which one
6781from another, taken all together, are less than the difference between o
6782and a.
6783
6784It is not our business to inquire here into the utility of this
6785principle in the investigation of nature. But how such a proposition,
6786which appears so greatly to extend our knowledge of nature, is possible
6787completely a priori, is indeed a question which deserves investigation,
6788although the first view seems to demonstrate the truth and reality of
6789the principle, and the question, how it is possible, may be considered
6790superfluous. For there are so many groundless pretensions to the
6791enlargement of our knowledge by pure reason that we must take it as a
6792general rule to be mistrustful of all such, and without a thoroughgoing
6793and radical deduction, to believe nothing of the sort even on the
6794clearest dogmatical evidence.
6795
6796Every addition to our empirical knowledge, and every advance made in
6797the exercise of our perception, is nothing more than an extension of the
6798determination of the internal sense, that is to say, a progression
6799in time, be objects themselves what they may, phenomena, or pure
6800intuitions. This progression in time determines everything, and is
6801itself determined by nothing else. That is to say, the parts of the
6802progression exist only in time, and by means of the synthesis thereof,
6803and are not given antecedently to it. For this reason, every transition
6804in perception to anything which follows upon another in time, is a
6805determination of time by means of the production of this perception.
6806And as this determination of time is, always and in all its parts, a
6807quantity, the perception produced is to be considered as a quantity
6808which proceeds through all its degrees--no one of which is the smallest
6809possible--from zero up to its determined degree. From this we perceive
6810the possibility of cognizing a priori a law of changes--a law,
6811however, which concerns their form merely. We merely anticipate our own
6812apprehension, the formal condition of which, inasmuch as it is itself to
6813be found in the mind antecedently to all given phenomena, must certainly
6814be capable of being cognized a priori.
6815
6816Thus, as time contains the sensuous condition a priori of the
6817possibility of a continuous progression of that which exists to
6818that which follows it, the understanding, by virtue of the unity of
6819apperception, contains the condition a priori of the possibility of a
6820continuous determination of the position in time of all phenomena, and
6821this by means of the series of causes and effects, the former of which
6822necessitate the sequence of the latter, and thereby render universally
6823and for all time, and by consequence, objectively, valid the empirical
6824cognition of the relations of time.
6825
6826
6827
6828C. THIRD ANALOGY.
6829
6830Principle of Coexistence, According to the Law of Reciprocity or
6831Community.
6832
6833All substances, in so far as they can be perceived in space at the same
6834time, exist in a state of complete reciprocity of action.
6835
6836
6837PROOF.
6838
6839
6840Things are coexistent, when in empirical intuition the perception of the
6841one can follow upon the perception of the other, and vice versa--which
6842cannot occur in the succession of phenomena, as we have shown in the
6843explanation of the second principle. Thus I can perceive the moon and
6844then the earth, or conversely, first the earth and then the moon; and
6845for the reason that my perceptions of these objects can reciprocally
6846follow each other, I say, they exist contemporaneously. Now coexistence
6847is the existence of the manifold in the same time. But time itself is
6848not an object of perception; and therefore we cannot conclude from the
6849fact that things are placed in the same time, the other fact, that
6850the perception of these things can follow each other reciprocally. The
6851synthesis of the imagination in apprehension would only present to us
6852each of these perceptions as present in the subject when the other is
6853not present, and contrariwise; but would not show that the objects are
6854coexistent, that is to say, that, if the one exists, the other also
6855exists in the same time, and that this is necessarily so, in order that
6856the perceptions may be capable of following each other reciprocally.
6857It follows that a conception of the understanding or category of the
6858reciprocal sequence of the determinations of phenomena (existing, as
6859they do, apart from each other, and yet contemporaneously), is requisite
6860to justify us in saying that the reciprocal succession of perceptions
6861has its foundation in the object, and to enable us to represent
6862coexistence as objective. But that relation of substances in which
6863the one contains determinations the ground of which is in the other
6864substance, is the relation of influence. And, when this influence is
6865reciprocal, it is the relation of community or reciprocity. Consequently
6866the coexistence of substances in space cannot be cognized in experience
6867otherwise than under the precondition of their reciprocal action. This
6868is therefore the condition of the possibility of things themselves as
6869objects of experience.
6870
6871Things are coexistent, in so far as they exist in one and the same time.
6872But how can we know that they exist in one and the same time? Only
6873by observing that the order in the synthesis of apprehension of the
6874manifold is arbitrary and a matter of indifference, that is to say, that
6875it can proceed from A, through B, C, D, to E, or contrariwise from E
6876to A. For if they were successive in time (and in the order, let
6877us suppose, which begins with A), it is quite impossible for the
6878apprehension in perception to begin with E and go backwards to A,
6879inasmuch as A belongs to past time and, therefore, cannot be an object
6880of apprehension.
6881
6882Let us assume that in a number of substances considered as phenomena
6883each is completely isolated, that is, that no one acts upon another.
6884Then I say that the coexistence of these cannot be an object of
6885possible perception and that the existence of one cannot, by any mode of
6886empirical synthesis, lead us to the existence of another. For we imagine
6887them in this case to be separated by a completely void space, and thus
6888perception, which proceeds from the one to the other in time, would
6889indeed determine their existence by means of a following perception, but
6890would be quite unable to distinguish whether the one phenomenon follows
6891objectively upon the first, or is coexistent with it.
6892
6893Besides the mere fact of existence, then, there must be something by
6894means of which A determines the position of B in time and, conversely,
6895B the position of A; because only under this condition can substances
6896be empirically represented as existing contemporaneously. Now that alone
6897determines the position of another thing in time which is the cause of
6898it or of its determinations. Consequently every substance (inasmuch
6899as it can have succession predicated of it only in respect of its
6900determinations) must contain the causality of certain determinations in
6901another substance, and at the same time the effects of the causality of
6902the other in itself. That is to say, substances must stand (mediately or
6903immediately) in dynamical community with each other, if coexistence is
6904to be cognized in any possible experience. But, in regard to objects of
6905experience, that is absolutely necessary without which the experience of
6906these objects would itself be impossible. Consequently it is absolutely
6907necessary that all substances in the world of phenomena, in so far
6908as they are coexistent, stand in a relation of complete community of
6909reciprocal action to each other.
6910
6911The word community has in our language [Footnote: German] two meanings,
6912and contains the two notions conveyed in the Latin communio and
6913commercium. We employ it in this place in the latter sense--that of a
6914dynamical community, without which even the community of place (communio
6915spatii) could not be empirically cognized. In our experiences it is easy
6916to observe that it is only the continuous influences in all parts of
6917space that can conduct our senses from one object to another; that the
6918light which plays between our eyes and the heavenly bodies produces a
6919mediating community between them and us, and thereby evidences their
6920coexistence with us; that we cannot empirically change our position
6921(perceive this change), unless the existence of matter throughout the
6922whole of space rendered possible the perception of the positions we
6923occupy; and that this perception can prove the contemporaneous existence
6924of these places only through their reciprocal influence, and thereby
6925also the coexistence of even the most remote objects--although in this
6926case the proof is only mediate. Without community, every perception (of
6927a phenomenon in space) is separated from every other and isolated, and
6928the chain of empirical representations, that is, of experience, must,
6929with the appearance of a new object, begin entirely de novo, without the
6930least connection with preceding representations, and without standing
6931towards these even in the relation of time. My intention here is by no
6932means to combat the notion of empty space; for it may exist where our
6933perceptions cannot exist, inasmuch as they cannot reach thereto, and
6934where, therefore, no empirical perception of coexistence takes place.
6935But in this case it is not an object of possible experience.
6936
6937The following remarks may be useful in the way of explanation. In the
6938mind, all phenomena, as contents of a possible experience, must exist in
6939community (communio) of apperception or consciousness, and in so far as
6940it is requisite that objects be represented as coexistent and connected,
6941in so far must they reciprocally determine the position in time of each
6942other and thereby constitute a whole. If this subjective community is
6943to rest upon an objective basis, or to be applied to substances as
6944phenomena, the perception of one substance must render possible the
6945perception of another, and conversely. For otherwise succession, which
6946is always found in perceptions as apprehensions, would be predicated of
6947external objects, and their representation of their coexistence be thus
6948impossible. But this is a reciprocal influence, that is to say, a
6949real community (commercium) of substances, without which therefore the
6950empirical relation of coexistence would be a notion beyond the reach of
6951our minds. By virtue of this commercium, phenomena, in so far as
6952they are apart from, and nevertheless in connection with each other,
6953constitute a compositum reale. Such composita are possible in many
6954different ways. The three dynamical relations then, from which all
6955others spring, are those of inherence, consequence, and composition.
6956
6957These, then, are the three analogies of experience. They are nothing
6958more than principles of the determination of the existence of phenomena
6959in time, according to the three modi of this determination; to wit, the
6960relation to time itself as a quantity (the quantity of existence, that
6961is, duration), the relation in time as a series or succession, finally,
6962the relation in time as the complex of all existence (simultaneity).
6963This unity of determination in regard to time is thoroughly dynamical;
6964that is to say, time is not considered as that in which experience
6965determines immediately to every existence its position; for this is
6966impossible, inasmuch as absolute time is not an object of perception,
6967by means of which phenomena can be connected with each other. On
6968the contrary, the rule of the understanding, through which alone
6969the existence of phenomena can receive synthetical unity as regards
6970relations of time, determines for every phenomenon its position in time,
6971and consequently a priori, and with validity for all and every time.
6972
6973By nature, in the empirical sense of the word, we understand the
6974totality of phenomena connected, in respect of their existence,
6975according to necessary rules, that is, laws. There are therefore certain
6976laws (which are moreover a priori) which make nature possible; and all
6977empirical laws can exist only by means of experience, and by virtue of
6978those primitive laws through which experience itself becomes possible.
6979The purpose of the analogies is therefore to represent to us the unity
6980of nature in the connection of all phenomena under certain exponents,
6981the only business of which is to express the relation of time (in so far
6982as it contains all existence in itself) to the unity of apperception,
6983which can exist in synthesis only according to rules. The combined
6984expression of all is this: "All phenomena exist in one nature, and
6985must so exist, inasmuch as without this a priori unity, no unity of
6986experience, and consequently no determination of objects in experience,
6987is possible."
6988
6989As regards the mode of proof which we have employed in treating of these
6990transcendental laws of nature, and the peculiar character of we must
6991make one remark, which will at the same time be important as a guide
6992in every other attempt to demonstrate the truth of intellectual and
6993likewise synthetical propositions a priori. Had we endeavoured to prove
6994these analogies dogmatically, that is, from conceptions; that is to say,
6995had we employed this method in attempting to show that everything which
6996exists, exists only in that which is permanent--that every thing or
6997event presupposes the existence of something in a preceding state,
6998upon which it follows in conformity with a rule--lastly, that in the
6999manifold, which is coexistent, the states coexist in connection with
7000each other according to a rule, all our labour would have been utterly
7001in vain. For more conceptions of things, analyse them as we may, cannot
7002enable us to conclude from the existence of one object to the existence
7003of another. What other course was left for us to pursue? This only, to
7004demonstrate the possibility of experience as a cognition in which
7005at last all objects must be capable of being presented to us, if the
7006representation of them is to possess any objective reality. Now in this
7007third, this mediating term, the essential form of which consists in
7008the synthetical unity of the apperception of all phenomena, we found
7009a priori conditions of the universal and necessary determination as
7010to time of all existences in the world of phenomena, without which the
7011empirical determination thereof as to time would itself be impossible,
7012and we also discovered rules of synthetical unity a priori, by means of
7013which we could anticipate experience. For want of this method, and from
7014the fancy that it was possible to discover a dogmatical proof of the
7015synthetical propositions which are requisite in the empirical employment
7016of the understanding, has it happened that a proof of the principle of
7017sufficient reason has been so often attempted, and always in vain.
7018The other two analogies nobody has ever thought of, although they have
7019always been silently employed by the mind,* because the guiding thread
7020furnished by the categories was wanting, the guide which alone can
7021enable us to discover every hiatus, both in the system of conceptions
7022and of principles.
7023
7024
7025 [*Footnote: The unity of the universe, in which all phenomena to be
7026connected, is evidently a mere consequence of the admitted principle
7027of the community of all substances which are coexistent. For were
7028substances isolated, they could not as parts constitute a whole, and
7029were their connection (reciprocal action of the manifold) not necessary
7030from the very fact of coexistence, we could not conclude from the fact
7031of the latter as a merely ideal relation to the former as a real one. We
7032have, however, shown in its place that community is the proper ground
7033of the possibility of an empirical cognition of coexistence, and that
7034we may therefore properly reason from the latter to the former as its
7035condition.]
7036
7037
7038
70394. THE POSTULATES OF EMPIRICAL THOUGHT.
7040
70411. That which agrees with the formal conditions (intuition and
7042conception) of experience, is possible.
7043
70442. That which coheres with the material conditions of experience
7045(sensation), is real.
7046
70473. That whose coherence with the real is determined according to
7048universal conditions of experience is (exists) necessary.
7049
7050
7051Explanation.
7052
7053
7054The categories of modality possess this peculiarity, that they do not in
7055the least determine the object, or enlarge the conception to which they
7056are annexed as predicates, but only express its relation to the faculty
7057of cognition. Though my conception of a thing is in itself complete, I
7058am still entitled to ask whether the object of it is merely possible,
7059or whether it is also real, or, if the latter, whether it is also
7060necessary. But hereby the object itself is not more definitely
7061determined in thought, but the question is only in what relation it,
7062including all its determinations, stands to the understanding and its
7063employment in experience, to the empirical faculty of judgement, and to
7064the reason of its application to experience.
7065
7066For this very reason, too, the categories of modality are nothing
7067more than explanations of the conceptions of possibility, reality, and
7068necessity, as employed in experience, and at the same time, restrictions
7069of all the categories to empirical use alone, not authorizing the
7070transcendental employment of them. For if they are to have something
7071more than a merely logical significance, and to be something more than
7072a mere analytical expression of the form of thought, and to have a
7073relation to things and their possibility, reality, or necessity, they
7074must concern possible experience and its synthetical unity, in which
7075alone objects of cognition can be given.
7076
7077The postulate of the possibility of things requires also, that the
7078conception of the things agree with the formal conditions of our
7079experience in general. But this, that is to say, the objective form of
7080experience, contains all the kinds of synthesis which are requisite for
7081the cognition of objects. A conception which contains a synthesis
7082must be regarded as empty and, without reference to an object, if its
7083synthesis does not belong to experience--either as borrowed from it,
7084and in this case it is called an empirical conception, or such as is the
7085ground and a priori condition of experience (its form), and in this
7086case it is a pure conception, a conception which nevertheless belongs to
7087experience, inasmuch as its object can be found in this alone. For where
7088shall we find the criterion or character of the possibility of an object
7089which is cogitated by means of an a priori synthetical conception, if
7090not in the synthesis which constitutes the form of empirical cognition
7091of objects? That in such a conception no contradiction exists is indeed
7092a necessary logical condition, but very far from being sufficient
7093to establish the objective reality of the conception, that is, the
7094possibility of such an object as is thought in the conception. Thus, in
7095the conception of a figure which is contained within two straight lines,
7096there is no contradiction, for the conceptions of two straight lines and
7097of their junction contain no negation of a figure. The impossibility in
7098such a case does not rest upon the conception in itself, but upon the
7099construction of it in space, that is to say, upon the conditions of
7100space and its determinations. But these have themselves objective
7101reality, that is, they apply to possible things, because they contain a
7102priori the form of experience in general.
7103
7104And now we shall proceed to point out the extensive utility and
7105influence of this postulate of possibility. When I represent to myself a
7106thing that is permanent, so that everything in it which changes belongs
7107merely to its state or condition, from such a conception alone I never
7108can cognize that such a thing is possible. Or, if I represent to myself
7109something which is so constituted that, when it is posited,
7110something else follows always and infallibly, my thought contains no
7111self-contradiction; but whether such a property as causality is to
7112be found in any possible thing, my thought alone affords no means
7113of judging. Finally, I can represent to myself different things
7114(substances) which are so constituted that the state or condition of one
7115causes a change in the state of the other, and reciprocally; but whether
7116such a relation is a property of things cannot be perceived from these
7117conceptions, which contain a merely arbitrary synthesis. Only from the
7118fact, therefore, that these conceptions express a priori the relations
7119of perceptions in every experience, do we know that they possess
7120objective reality, that is, transcendental truth; and that independent
7121of experience, though not independent of all relation to form of an
7122experience in general and its synthetical unity, in which alone objects
7123can be empirically cognized.
7124
7125But when we fashion to ourselves new conceptions of substances, forces,
7126action, and reaction, from the material presented to us by perception,
7127without following the example of experience in their connection, we
7128create mere chimeras, of the possibility of which we cannot discover any
7129criterion, because we have not taken experience for our instructress,
7130though we have borrowed the conceptions from her. Such fictitious
7131conceptions derive their character of possibility not, like the
7132categories, a priori, as conceptions on which all experience depends,
7133but only, a posteriori, as conceptions given by means of experience
7134itself, and their possibility must either be cognized a posteriori
7135and empirically, or it cannot be cognized at all. A substance which is
7136permanently present in space, yet without filling it (like that tertium
7137quid between matter and the thinking subject which some have tried to
7138introduce into metaphysics), or a peculiar fundamental power of the mind
7139of intuiting the future by anticipation (instead of merely inferring
7140from past and present events), or, finally, a power of the mind to place
7141itself in community of thought with other men, however distant they may
7142be--these are conceptions the possibility of which has no ground to rest
7143upon. For they are not based upon experience and its known laws; and,
7144without experience, they are a merely arbitrary conjunction of thoughts,
7145which, though containing no internal contradiction, has no claim to
7146objective reality, neither, consequently, to the possibility of such an
7147object as is thought in these conceptions. As far as concerns reality,
7148it is self-evident that we cannot cogitate such a possibility in
7149concreto without the aid of experience; because reality is concerned
7150only with sensation, as the matter of experience, and not with the form
7151of thought, with which we can no doubt indulge in shaping fancies.
7152
7153But I pass by everything which derives its possibility from reality in
7154experience, and I purpose treating here merely of the possibility of
7155things by means of a priori conceptions. I maintain, then, that the
7156possibility of things is not derived from such conceptions per se, but
7157only when considered as formal and objective conditions of an experience
7158in general.
7159
7160It seems, indeed, as if the possibility of a triangle could be cognized
7161from the conception of it alone (which is certainly independent of
7162experience); for we can certainly give to the conception a corresponding
7163object completely a priori, that is to say, we can construct it. But as
7164a triangle is only the form of an object, it must remain a mere product
7165of the imagination, and the possibility of the existence of an object
7166corresponding to it must remain doubtful, unless we can discover some
7167other ground, unless we know that the figure can be cogitated under the
7168conditions upon which all objects of experience rest. Now, the facts
7169that space is a formal condition a priori of external experience,
7170that the formative synthesis, by which we construct a triangle in
7171imagination, is the very same as that we employ in the apprehension of a
7172phenomenon for the purpose of making an empirical conception of it, are
7173what alone connect the notion of the possibility of such a thing, with
7174the conception of it. In the same manner, the possibility of continuous
7175quantities, indeed of quantities in general, for the conceptions of them
7176are without exception synthetical, is never evident from the conceptions
7177in themselves, but only when they are considered as the formal
7178conditions of the determination of objects in experience. And where,
7179indeed, should we look for objects to correspond to our conceptions, if
7180not in experience, by which alone objects are presented to us? It is,
7181however, true that without antecedent experience we can cognize and
7182characterize the possibility of things, relatively to the formal
7183conditions, under which something is determined in experience as an
7184object, consequently, completely a priori. But still this is possible
7185only in relation to experience and within its limits.
7186
7187The postulate concerning the cognition of the reality of things requires
7188perception, consequently conscious sensation, not indeed immediately,
7189that is, of the object itself, whose existence is to be cognized, but
7190still that the object have some connection with a real perception, in
7191accordance with the analogies of experience, which exhibit all kinds of
7192real connection in experience.
7193
7194From the mere conception of a thing it is impossible to conclude its
7195existence. For, let the conception be ever so complete, and containing
7196a statement of all the determinations of the thing, the existence of
7197it has nothing to do with all this, but only with thew question whether
7198such a thing is given, so that the perception of it can in every case
7199precede the conception. For the fact that the conception of it precedes
7200the perception, merely indicates the possibility of its existence; it
7201is perception which presents matter to the conception, that is the sole
7202criterion of reality. Prior to the perception of the thing, however, and
7203therefore comparatively a priori, we are able to cognize its existence,
7204provided it stands in connection with some perceptions according to the
7205principles of the empirical conjunction of these, that is, in conformity
7206with the analogies of perception. For, in this case, the existence
7207of the supposed thing is connected with our perception in a possible
7208experience, and we are able, with the guidance of these analogies, to
7209reason in the series of possible perceptions from a thing which we do
7210really perceive to the thing we do not perceive. Thus, we cognize
7211the existence of a magnetic matter penetrating all bodies from the
7212perception of the attraction of the steel-filings by the magnet,
7213although the constitution of our organs renders an immediate perception
7214of this matter impossible for us. For, according to the laws of
7215sensibility and the connected context of our perceptions, we should
7216in an experience come also on an immediate empirical intuition of
7217this matter, if our senses were more acute--but this obtuseness has
7218no influence upon and cannot alter the form of possible experience in
7219general. Our knowledge of the existence of things reaches as far as our
7220perceptions, and what may be inferred from them according to empirical
7221laws, extend. If we do not set out from experience, or do not proceed
7222according to the laws of the empirical connection of phenomena, our
7223pretensions to discover the existence of a thing which we do not
7224immediately perceive are vain. Idealism, however, brings forward
7225powerful objections to these rules for proving existence mediately. This
7226is, therefore, the proper place for its refutation.
7227
7228
7229
7230REFUTATION OF IDEALISM.
7231
7232Idealism--I mean material idealism--is the theory which declares the
7233existence of objects in space without us to be either () doubtful
7234and indemonstrable, or (2) false and impossible. The first is the
7235problematical idealism of Descartes, who admits the undoubted certainty
7236of only one empirical assertion (assertio), to wit, "I am." The second
7237is the dogmatical idealism of Berkeley, who maintains that space,
7238together with all the objects of which it is the inseparable condition,
7239is a thing which is in itself impossible, and that consequently the
7240objects in space are mere products of the imagination. The dogmatical
7241theory of idealism is unavoidable, if we regard space as a property
7242of things in themselves; for in that case it is, with all to which it
7243serves as condition, a nonentity. But the foundation for this kind of
7244idealism we have already destroyed in the transcendental aesthetic.
7245Problematical idealism, which makes no such assertion, but only alleges
7246our incapacity to prove the existence of anything besides ourselves by
7247means of immediate experience, is a theory rational and evidencing a
7248thorough and philosophical mode of thinking, for it observes the rule
7249not to form a decisive judgement before sufficient proof be shown. The
7250desired proof must therefore demonstrate that we have experience of
7251external things, and not mere fancies. For this purpose, we must prove,
7252that our internal and, to Descartes, indubitable experience is itself
7253possible only under the previous assumption of external experience.
7254
7255
7256
7257THEOREM.
7258
7259The simple but empirically determined consciousness of my own existence
7260proves the existence of external objects in space.
7261
7262
7263PROOF
7264
7265
7266I am conscious of my own existence as determined in time. All
7267determination in regard to time presupposes the existence of something
7268permanent in perception. But this permanent something cannot be
7269something in me, for the very reason that my existence in time is itself
7270determined by this permanent something. It follows that the perception
7271of this permanent existence is possible only through a thing without
7272me and not through the mere representation of a thing without me.
7273Consequently, the determination of my existence in time is possible only
7274through the existence of real things external to me. Now, consciousness
7275in time is necessarily connected with the consciousness of the
7276possibility of this determination in time. Hence it follows that
7277consciousness in time is necessarily connected also with the existence
7278of things without me, inasmuch as the existence of these things is the
7279condition of determination in time. That is to say, the consciousness of
7280my own existence is at the same time an immediate consciousness of the
7281existence of other things without me.
7282
7283Remark I. The reader will observe, that in the foregoing proof the game
7284which idealism plays is retorted upon itself, and with more justice.
7285It assumed that the only immediate experience is internal and that from
7286this we can only infer the existence of external things. But, as
7287always happens, when we reason from given effects to determined causes,
7288idealism has reasoned with too much haste and uncertainty, for it
7289is quite possible that the cause of our representations may lie in
7290ourselves, and that we ascribe it falsely to external things. But our
7291proof shows that external experience is properly immediate,* that only
7292by virtue of it--not, indeed, the consciousness of our own existence,
7293but certainly the determination of our existence in time, that is,
7294internal experience--is possible. It is true, that the representation
7295"I am," which is the expression of the consciousness which can accompany
7296all my thoughts, is that which immediately includes the existence of a
7297subject. But in this representation we cannot find any knowledge of the
7298subject, and therefore also no empirical knowledge, that is, experience.
7299For experience contains, in addition to the thought of something
7300existing, intuition, and in this case it must be internal intuition,
7301that is, time, in relation to which the subject must be determined.
7302But the existence of external things is absolutely requisite for this
7303purpose, so that it follows that internal experience is itself possible
7304only mediately and through external experience.
7305
7306
7307 [*Footnote: The immediate consciousness of the existence of external
7308things is, in the preceding theorem, not presupposed, but proved, by the
7309possibility of this consciousness understood by us or not. The question
7310as to the possibility of it would stand thus: "Have we an internal
7311sense, but no external sense, and is our belief in external perception
7312a mere delusion?" But it is evident that, in order merely to fancy to
7313ourselves anything as external, that is, to present it to the sense in
7314intuition we must already possess an external sense, and must thereby
7315distinguish immediately the mere receptivity of an external intuition
7316from the spontaneity which characterizes every act of imagination. For
7317merely to imagine also an external sense, would annihilate the faculty
7318of intuition itself which is to be determined by the imagination.]
7319
7320
7321Remark II. Now with this view all empirical use of our faculty of
7322cognition in the determination of time is in perfect accordance.
7323Its truth is supported by the fact that it is possible to perceive a
7324determination of time only by means of a change in external relations
7325(motion) to the permanent in space (for example, we become aware of the
7326sun's motion by observing the changes of his relation to the objects
7327of this earth). But this is not all. We find that we possess nothing
7328permanent that can correspond and be submitted to the conception of a
7329substance as intuition, except matter. This idea of permanence is not
7330itself derived from external experience, but is an a priori necessary
7331condition of all determination of time, consequently also of the
7332internal sense in reference to our own existence, and that through
7333the existence of external things. In the representation "I," the
7334consciousness of myself is not an intuition, but a merely intellectual
7335representation produced by the spontaneous activity of a thinking
7336subject. It follows, that this "I" has not any predicate of intuition,
7337which, in its character of permanence, could serve as correlate to
7338the determination of time in the internal sense--in the same way as
7339impenetrability is the correlate of matter as an empirical intuition.
7340
7341Remark III. From the fact that the existence of external things is a
7342necessary condition of the possibility of a determined consciousness
7343of ourselves, it does not follow that every intuitive representation
7344of external things involves the existence of these things, for their
7345representations may very well be the mere products of the imagination
7346(in dreams as well as in madness); though, indeed, these are themselves
7347created by the reproduction of previous external perceptions, which,
7348as has been shown, are possible only through the reality of external
7349objects. The sole aim of our remarks has, however, been to prove
7350that internal experience in general is possible only through external
7351experience in general. Whether this or that supposed experience be
7352purely imaginary must be discovered from its particular determinations
7353and by comparing these with the criteria of all real experience.
7354
7355Finally, as regards the third postulate, it applies to material
7356necessity in existence, and not to merely formal and logical necessity
7357in the connection of conceptions. Now as we cannot cognize completely
7358a priori the existence of any object of sense, though we can do so
7359comparatively a priori, that is, relatively to some other previously
7360given existence--a cognition, however, which can only be of such an
7361existence as must be contained in the complex of experience, of which
7362the previously given perception is a part--the necessity of existence
7363can never be cognized from conceptions, but always, on the contrary,
7364from its connection with that which is an object of perception. But the
7365only existence cognized, under the condition of other given phenomena,
7366as necessary, is the existence of effects from given causes in
7367conformity with the laws of causality. It is consequently not the
7368necessity of the existence of things (as substances), but the necessity
7369of the state of things that we cognize, and that not immediately, but by
7370means of the existence of other states given in perception, according
7371to empirical laws of causality. Hence it follows that the criterion of
7372necessity is to be found only in the law of possible experience--that
7373everything which happens is determined a priori in the phenomenon by
7374its cause. Thus we cognize only the necessity of effects in nature, the
7375causes of which are given us. Moreover, the criterion of necessity
7376in existence possesses no application beyond the field of possible
7377experience, and even in this it is not valid of the existence of things
7378as substances, because these can never be considered as empirical
7379effects, or as something that happens and has a beginning. Necessity,
7380therefore, regards only the relations of phenomena according to the
7381dynamical law of causality, and the possibility grounded thereon, of
7382reasoning from some given existence (of a cause) a priori to another
7383existence (of an effect). "Everything that happens is hypothetically
7384necessary," is a principle which subjects the changes that take place in
7385the world to a law, that is, to a rule of necessary existence, without
7386which nature herself could not possibly exist. Hence the proposition,
7387"Nothing happens by blind chance (in mundo non datur casus)," is an
7388a priori law of nature. The case is the same with the proposition,
7389"Necessity in nature is not blind," that is, it is conditioned,
7390consequently intelligible necessity (non datur fatum). Both laws subject
7391the play of change to "a nature of things (as phenomena)," or, which
7392is the same thing, to the unity of the understanding, and through
7393the understanding alone can changes belong to an experience, as the
7394synthetical unity of phenomena. Both belong to the class of dynamical
7395principles. The former is properly a consequence of the principle of
7396causality--one of the analogies of experience. The latter belongs to the
7397principles of modality, which to the determination of causality adds the
7398conception of necessity, which is itself, however, subject to a rule of
7399the understanding. The principle of continuity forbids any leap in the
7400series of phenomena regarded as changes (in mundo non datur saltus); and
7401likewise, in the complex of all empirical intuitions in space, any
7402break or hiatus between two phenomena (non datur hiatus)--for we can so
7403express the principle, that experience can admit nothing which proves
7404the existence of a vacuum, or which even admits it as a part of an
7405empirical synthesis. For, as regards a vacuum or void, which we may
7406cogitate as out and beyond the field of possible experience (the world),
7407such a question cannot come before the tribunal of mere understanding,
7408which decides only upon questions that concern the employment of given
7409phenomena for the construction of empirical cognition. It is rather a
7410problem for ideal reason, which passes beyond the sphere of a possible
7411experience and aims at forming a judgement of that which surrounds and
7412circumscribes it, and the proper place for the consideration of it is
7413the transcendental dialectic. These four propositions, "In mundo non
7414datur hiatus, non datur saltus, non datur casus, non datur fatum," as
7415well as all principles of transcendental origin, we could very easily
7416exhibit in their proper order, that is, in conformity with the order
7417of the categories, and assign to each its proper place. But the already
7418practised reader will do this for himself, or discover the clue to such
7419an arrangement. But the combined result of all is simply this, to admit
7420into the empirical synthesis nothing which might cause a break in or
7421be foreign to the understanding and the continuous connection of all
7422phenomena, that is, the unity of the conceptions of the understanding.
7423For in the understanding alone is the unity of experience, in which all
7424perceptions must have their assigned place, possible.
7425
7426Whether the field of possibility be greater than that of reality,
7427and whether the field of the latter be itself greater than that of
7428necessity, are interesting enough questions, and quite capable
7429of synthetic solution, questions, however, which come under the
7430jurisdiction of reason alone. For they are tantamount to asking whether
7431all things as phenomena do without exception belong to the complex and
7432connected whole of a single experience, of which every given
7433perception is a part which therefore cannot be conjoined with any
7434other phenomena--or, whether my perceptions can belong to more than one
7435possible experience? The understanding gives to experience, according
7436to the subjective and formal conditions, of sensibility as well as of
7437apperception, the rules which alone make this experience possible.
7438Other forms of intuition besides those of space and time, other forms of
7439understanding besides the discursive forms of thought, or of cognition
7440by means of conceptions, we can neither imagine nor make intelligible
7441to ourselves; and even if we could, they would still not belong to
7442experience, which is the only mode of cognition by which objects are
7443presented to us. Whether other perceptions besides those which belong
7444to the total of our possible experience, and consequently whether some
7445other sphere of matter exists, the understanding has no power to decide,
7446its proper occupation being with the synthesis of that which is given.
7447Moreover, the poverty of the usual arguments which go to prove the
7448existence of a vast sphere of possibility, of which all that is real
7449(every object of experience) is but a small part, is very remarkable.
7450"All real is possible"; from this follows naturally, according to the
7451logical laws of conversion, the particular proposition: "Some possible
7452is real." Now this seems to be equivalent to: "Much is possible that is
7453not real." No doubt it does seem as if we ought to consider the sum of
7454the possible to be greater than that of the real, from the fact that
7455something must be added to the former to constitute the latter. But this
7456notion of adding to the possible is absurd. For that which is not in
7457the sum of the possible, and consequently requires to be added to it,
7458is manifestly impossible. In addition to accordance with the formal
7459conditions of experience, the understanding requires a connection with
7460some perception; but that which is connected with this perception is
7461real, even although it is not immediately perceived. But that another
7462series of phenomena, in complete coherence with that which is given
7463in perception, consequently more than one all-embracing experience is
7464possible, is an inference which cannot be concluded from the data given
7465us by experience, and still less without any data at all. That which is
7466possible only under conditions which are themselves merely possible, is
7467not possible in any respect. And yet we can find no more certain ground
7468on which to base the discussion of the question whether the sphere of
7469possibility is wider than that of experience.
7470
7471I have merely mentioned these questions, that in treating of the
7472conception of the understanding, there might be no omission of anything
7473that, in the common opinion, belongs to them. In reality, however, the
7474notion of absolute possibility (possibility which is valid in every
7475respect) is not a mere conception of the understanding, which can be
7476employed empirically, but belongs to reason alone, which passes the
7477bounds of all empirical use of the understanding. We have, therefore,
7478contented ourselves with a merely critical remark, leaving the subject
7479to be explained in the sequel.
7480
7481Before concluding this fourth section, and at the same time the system
7482of all principles of the pure understanding, it seems proper to
7483mention the reasons which induced me to term the principles of modality
7484postulates. This expression I do not here use in the sense which some
7485more recent philosophers, contrary to its meaning with mathematicians,
7486to whom the word properly belongs, attach to it--that of a proposition,
7487namely, immediately certain, requiring neither deduction nor proof. For
7488if, in the case of synthetical propositions, however evident they may
7489be, we accord to them without deduction, and merely on the strength
7490of their own pretensions, unqualified belief, all critique of the
7491understanding is entirely lost; and, as there is no want of bold
7492pretensions, which the common belief (though for the philosopher this is
7493no credential) does not reject, the understanding lies exposed to every
7494delusion and conceit, without the power of refusing its assent to those
7495assertions, which, though illegitimate, demand acceptance as veritable
7496axioms. When, therefore, to the conception of a thing an a priori
7497determination is synthetically added, such a proposition must obtain, if
7498not a proof, at least a deduction of the legitimacy of its assertion.
7499
7500The principles of modality are, however, not objectively synthetical,
7501for the predicates of possibility, reality, and necessity do not in
7502the least augment the conception of that of which they are affirmed,
7503inasmuch as they contribute nothing to the representation of the object.
7504But as they are, nevertheless, always synthetical, they are so merely
7505subjectively. That is to say, they have a reflective power, and apply
7506to the conception of a thing, of which, in other respects, they affirm
7507nothing, the faculty of cognition in which the conception originates
7508and has its seat. So that if the conception merely agree with the formal
7509conditions of experience, its object is called possible; if it is in
7510connection with perception, and determined thereby, the object is real;
7511if it is determined according to conceptions by means of the connection
7512of perceptions, the object is called necessary. The principles of
7513modality therefore predicate of a conception nothing more than the
7514procedure of the faculty of cognition which generated it. Now a
7515postulate in mathematics is a practical proposition which contains
7516nothing but the synthesis by which we present an object to ourselves,
7517and produce the conception of it, for example--"With a given line,
7518to describe a circle upon a plane, from a given point"; and such a
7519proposition does not admit of proof, because the procedure, which it
7520requires, is exactly that by which alone it is possible to generate the
7521conception of such a figure. With the same right, accordingly, can we
7522postulate the principles of modality, because they do not augment* the
7523conception of a thing but merely indicate the manner in which it is
7524connected with the faculty of cognition.
7525
7526
7527 [*Footnote: When I think the reality of a thing, I do really think more
7528than the possibility, but not in the thing; for that can never contain
7529more in reality than was contained in its complete possibility. But
7530while the notion of possibility is merely the notion of a position of
7531thing in relation to the understanding (its empirical use), reality is
7532the conjunction of the thing with perception.]
7533
7534
7535
7536GENERAL REMARK ON THE SYSTEM OF PRINCIPLES.
7537
7538It is very remarkable that we cannot perceive the possibility of a thing
7539from the category alone, but must always have an intuition, by which
7540to make evident the objective reality of the pure conception of the
7541understanding. Take, for example, the categories of relation. How (1)
7542a thing can exist only as a subject, and not as a mere determination of
7543other things, that is, can be substance; or how (2), because something
7544exists, some other thing must exist, consequently how a thing can be a
7545cause; or how (3), when several things exist, from the fact that one
7546of these things exists, some consequence to the others follows,
7547and reciprocally, and in this way a community of substances can be
7548possible--are questions whose solution cannot be obtained from mere
7549conceptions. The very same is the case with the other categories; for
7550example, how a thing can be of the same sort with many others, that is,
7551can be a quantity, and so on. So long as we have not intuition we cannot
7552know whether we do really think an object by the categories, and where
7553an object can anywhere be found to cohere with them, and thus the truth
7554is established, that the categories are not in themselves cognitions,
7555but mere forms of thought for the construction of cognitions from given
7556intuitions. For the same reason is it true that from categories alone
7557no synthetical proposition can be made. For example: "In every existence
7558there is substance," that is, something that can exist only as a subject
7559and not as mere predicate; or, "Everything is a quantity"--to construct
7560propositions such as these, we require something to enable us to go out
7561beyond the given conception and connect another with it. For the same
7562reason the attempt to prove a synthetical proposition by means of mere
7563conceptions, for example: "Everything that exists contingently has a
7564cause," has never succeeded. We could never get further than proving
7565that, without this relation to conceptions, we could not conceive the
7566existence of the contingent, that is, could not a priori through the
7567understanding cognize the existence of such a thing; but it does not
7568hence follow that this is also the condition of the possibility of the
7569thing itself that is said to be contingent. If, accordingly; we look
7570back to our proof of the principle of causality, we shall find that we
7571were able to prove it as valid only of objects of possible experience,
7572and, indeed, only as itself the principle of the possibility of
7573experience, Consequently of the cognition of an object given in
7574empirical intuition, and not from mere conceptions. That, however,
7575the proposition: "Everything that is contingent must have a cause," is
7576evident to every one merely from conceptions, is not to be denied. But
7577in this case the conception of the contingent is cogitated as involving
7578not the category of modality (as that the non-existence of which can
7579be conceived) but that of relation (as that which can exist only as
7580the consequence of something else), and so it is really an identical
7581proposition: "That which can exist only as a consequence, has a cause."
7582In fact, when we have to give examples of contingent existence, we
7583always refer to changes, and not merely to the possibility of conceiving
7584the opposite.* But change is an event, which, as such, is possible only
7585through a cause, and considered per se its non-existence is therefore
7586possible, and we become cognizant of its contingency from the fact that
7587it can exist only as the effect of a cause. Hence, if a thing is assumed
7588to be contingent, it is an analytical proposition to say, it has a
7589cause.
7590
7591
7592 [*Footnote: We can easily conceive the non-existence of matter; but the
7593ancients did not thence infer its contingency. But even the alternation
7594of the existence and non-existence of a given state in a thing, in
7595which all change consists, by no means proves the contingency of that
7596state--the ground of proof being the reality of its opposite. For
7597example, a body is in a state of rest after motion, but we cannot infer
7598the contingency of the motion from the fact that the former is the
7599opposite of the latter. For this opposite is merely a logical and not a
7600real opposite to the other. If we wish to demonstrate the contingency of
7601the motion, what we ought to prove is that, instead of the motion which
7602took place in the preceding point of time, it was possible for the body
7603to have been then in rest, not, that it is afterwards in rest; for in
7604this case, both opposites are perfectly consistent with each other.]
7605
7606
7607But it is still more remarkable that, to understand the possibility of
7608things according to the categories and thus to demonstrate the objective
7609reality of the latter, we require not merely intuitions, but external
7610intuitions. If, for example, we take the pure conceptions of relation,
7611we find that (1) for the purpose of presenting to the conception of
7612substance something permanent in intuition corresponding thereto and
7613thus of demonstrating the objective reality of this conception, we
7614require an intuition (of matter) in space, because space alone is
7615permanent and determines things as such, while time, and with it all
7616that is in the internal sense, is in a state of continual flow; (2)
7617in order to represent change as the intuition corresponding to the
7618conception of causality, we require the representation of motion as
7619change in space; in fact, it is through it alone that changes, the
7620possibility of which no pure understanding can perceive, are capable
7621of being intuited. Change is the connection of determinations
7622contradictorily opposed to each other in the existence of one and the
7623same thing. Now, how it is possible that out of a given state one
7624quite opposite to it in the same thing should follow, reason without
7625an example can not only not conceive, but cannot even make intelligible
7626without intuition; and this intuition is the motion of a point in space;
7627the existence of which in different spaces (as a consequence of opposite
7628determinations) alone makes the intuition of change possible. For, in
7629order to make even internal change cognitable, we require to represent
7630time, as the form of the internal sense, figuratively by a line, and the
7631internal change by the drawing of that line (motion), and consequently
7632are obliged to employ external intuition to be able to represent the
7633successive existence of ourselves in different states. The proper ground
7634of this fact is that all change to be perceived as change presupposes
7635something permanent in intuition, while in the internal sense no
7636permanent intuition is to be found. Lastly, the objective possibility
7637of the category of community cannot be conceived by mere reason, and
7638consequently its objective reality cannot be demonstrated without an
7639intuition, and that external in space. For how can we conceive the
7640possibility of community, that is, when several substances exist, that
7641some effect on the existence of the one follows from the existence
7642of the other, and reciprocally, and therefore that, because something
7643exists in the latter, something else must exist in the former, which
7644could not be understood from its own existence alone? For this is the
7645very essence of community--which is inconceivable as a property of
7646things which are perfectly isolated. Hence, Leibnitz, in attributing to
7647the substances of the world--as cogitated by the understanding alone--a
7648community, required the mediating aid of a divinity; for, from their
7649existence, such a property seemed to him with justice inconceivable. But
7650we can very easily conceive the possibility of community (of
7651substances as phenomena) if we represent them to ourselves as in space,
7652consequently in external intuition. For external intuition contains
7653in itself a priori formal external relations, as the conditions of the
7654possibility of the real relations of action and reaction, and
7655therefore of the possibility of community. With the same ease can it
7656be demonstrated, that the possibility of things as quantities, and
7657consequently the objective reality of the category of quantity, can be
7658grounded only in external intuition, and that by its means alone is the
7659notion of quantity appropriated by the internal sense. But I must avoid
7660prolixity, and leave the task of illustrating this by examples to the
7661reader's own reflection.
7662
7663The above remarks are of the greatest importance, not only for the
7664confirmation of our previous confutation of idealism, but still more
7665when the subject of self-cognition by mere internal consciousness
7666and the determination of our own nature without the aid of external
7667empirical intuitions is under discussion, for the indication of the
7668grounds of the possibility of such a cognition.
7669
7670The result of the whole of this part of the analytic of principles is,
7671therefore: "All principles of the pure understanding are nothing more
7672than a priori principles of the possibility of experience, and to
7673experience alone do all a priori synthetical propositions apply and
7674relate"; indeed, their possibility itself rests entirely on this
7675relation.
7676
7677
7678
7679CHAPTER III Of the Ground of the Division of all Objects into Phenomena
7680and Noumena.
7681
7682We have now not only traversed the region of the pure understanding and
7683carefully surveyed every part of it, but we have also measured it, and
7684assigned to everything therein its proper place. But this land is an
7685island, and enclosed by nature herself within unchangeable limits. It is
7686the land of truth (an attractive word), surrounded by a wide and stormy
7687ocean, the region of illusion, where many a fog-bank, many an iceberg,
7688seems to the mariner, on his voyage of discovery, a new country, and,
7689while constantly deluding him with vain hopes, engages him in dangerous
7690adventures, from which he never can desist, and which yet he never can
7691bring to a termination. But before venturing upon this sea, in order
7692to explore it in its whole extent, and to arrive at a certainty whether
7693anything is to be discovered there, it will not be without advantage if
7694we cast our eyes upon the chart of the land that we are about to
7695leave, and to ask ourselves, firstly, whether we cannot rest perfectly
7696contented with what it contains, or whether we must not of necessity
7697be contented with it, if we can find nowhere else a solid foundation to
7698build upon; and, secondly, by what title we possess this land itself,
7699and how we hold it secure against all hostile claims? Although, in the
7700course of our analytic, we have already given sufficient answers to
7701these questions, yet a summary recapitulation of these solutions may
7702be useful in strengthening our conviction, by uniting in one point the
7703momenta of the arguments.
7704
7705We have seen that everything which the understanding draws from itself,
7706without borrowing from experience, it nevertheless possesses only
7707for the behoof and use of experience. The principles of the pure
7708understanding, whether constitutive a priori (as the mathematical
7709principles), or merely regulative (as the dynamical), contain nothing
7710but the pure schema, as it were, of possible experience. For experience
7711possesses its unity from the synthetical unity which the understanding,
7712originally and from itself, imparts to the synthesis of the imagination
7713in relation to apperception, and in a priori relation to and agreement
7714with which phenomena, as data for a possible cognition, must stand. But
7715although these rules of the understanding are not only a priori true,
7716but the very source of all truth, that is, of the accordance of our
7717cognition with objects, and on this ground, that they contain the basis
7718of the possibility of experience, as the ensemble of all cognition, it
7719seems to us not enough to propound what is true--we desire also to
7720be told what we want to know. If, then, we learn nothing more by this
7721critical examination than what we should have practised in the merely
7722empirical use of the understanding, without any such subtle inquiry,
7723the presumption is that the advantage we reap from it is not worth
7724the labour bestowed upon it. It may certainly be answered that no rash
7725curiosity is more prejudicial to the enlargement of our knowledge than
7726that which must know beforehand the utility of this or that piece
7727of information which we seek, before we have entered on the needful
7728investigations, and before one could form the least conception of its
7729utility, even though it were placed before our eyes. But there is
7730one advantage in such transcendental inquiries which can be made
7731comprehensible to the dullest and most reluctant learner--this, namely,
7732that the understanding which is occupied merely with empirical exercise,
7733and does not reflect on the sources of its own cognition, may exercise
7734its functions very well and very successfully, but is quite unable to do
7735one thing, and that of very great importance, to determine, namely,
7736the bounds that limit its employment, and to know what lies within
7737or without its own sphere. This purpose can be obtained only by
7738such profound investigations as we have instituted. But if it cannot
7739distinguish whether certain questions lie within its horizon or not, it
7740can never be sure either as to its claims or possessions, but must lay
7741its account with many humiliating corrections, when it transgresses, as
7742it unavoidably will, the limits of its own territory, and loses itself
7743in fanciful opinions and blinding illusions.
7744
7745That the understanding, therefore, cannot make of its a priori
7746principles, or even of its conceptions, other than an empirical use,
7747is a proposition which leads to the most important results. A
7748transcendental use is made of a conception in a fundamental proposition
7749or principle, when it is referred to things in general and considered
7750as things in themselves; an empirical use, when it is referred merely to
7751phenomena, that is, to objects of a possible experience. That the latter
7752use of a conception is the only admissible one is evident from the
7753reasons following. For every conception are requisite, firstly, the
7754logical form of a conception (of thought) general; and, secondly, the
7755possibility of presenting to this an object to which it may apply.
7756Failing this latter, it has no sense, and utterly void of content,
7757although it may contain the logical function for constructing a
7758conception from certain data. Now, object cannot be given to a
7759conception otherwise than by intuition, and, even if a pure intuition
7760antecedent to the object is a priori possible, this pure intuition can
7761itself obtain objective validity only from empirical intuition, of which
7762it is itself but the form. All conceptions, therefore, and with them
7763all principles, however high the degree of their a priori possibility,
7764relate to empirical intuitions, that is, to data towards a possible
7765experience. Without this they possess no objective validity, but are
7766mere play of imagination or of understanding with images or notions. Let
7767us take, for example, the conceptions of mathematics, and first in its
7768pure intuitions. "Space has three dimensions"--"Between two points there
7769can be only one straight line," etc. Although all these principles,
7770and the representation of the object with which this science occupies
7771itself, are generated in the mind entirely a priori, they would
7772nevertheless have no significance if we were not always able to exhibit
7773their significance in and by means of phenomena (empirical objects).
7774Hence it is requisite that an abstract conception be made sensuous,
7775that is, that an object corresponding to it in intuition be forthcoming,
7776otherwise the conception remains, as we say, without sense, that
7777is, without meaning. Mathematics fulfils this requirement by the
7778construction of the figure, which is a phenomenon evident to the senses.
7779The same science finds support and significance in number; this in its
7780turn finds it in the fingers, or in counters, or in lines and points.
7781The conception itself is always produced a priori, together with the
7782synthetical principles or formulas from such conceptions; but the proper
7783employment of them, and their application to objects, can exist nowhere
7784but in experience, the possibility of which, as regards its form, they
7785contain a priori.
7786
7787That this is also the case with all of the categories and the principles
7788based upon them is evident from the fact that we cannot render
7789intelligible the possibility of an object corresponding to them without
7790having recourse to the conditions of sensibility, consequently, to the
7791form of phenomena, to which, as their only proper objects, their use
7792must therefore be confined, inasmuch as, if this condition is removed,
7793all significance, that is, all relation to an object, disappears, and
7794no example can be found to make it comprehensible what sort of things we
7795ought to think under such conceptions.
7796
7797The conception of quantity cannot be explained except by saying that
7798it is the determination of a thing whereby it can be cogitated how
7799many times one is placed in it. But this "how many times" is based upon
7800successive repetition, consequently upon time and the synthesis of the
7801homogeneous therein. Reality, in contradistinction to negation, can be
7802explained only by cogitating a time which is either filled therewith or
7803is void. If I leave out the notion of permanence (which is existence in
7804all time), there remains in the conception of substance nothing but the
7805logical notion of subject, a notion of which I endeavour to realize by
7806representing to myself something that can exist only as a subject. But
7807not only am I perfectly ignorant of any conditions under which this
7808logical prerogative can belong to a thing, I can make nothing out of
7809the notion, and draw no inference from it, because no object to which
7810to apply the conception is determined, and we consequently do not know
7811whether it has any meaning at all. In like manner, if I leave out the
7812notion of time, in which something follows upon some other thing in
7813conformity with a rule, I can find nothing in the pure category, except
7814that there is a something of such a sort that from it a conclusion may
7815be drawn as to the existence of some other thing. But in this case
7816it would not only be impossible to distinguish between a cause and an
7817effect, but, as this power to draw conclusions requires conditions of
7818which I am quite ignorant, the conception is not determined as to the
7819mode in which it ought to apply to an object. The so-called principle:
7820"Everything that is contingent has a cause," comes with a gravity and
7821self-assumed authority that seems to require no support from without.
7822But, I ask, what is meant by contingent? The answer is that the
7823non-existence of which is possible. But I should like very well to know
7824by what means this possibility of non-existence is to be cognized, if we
7825do not represent to ourselves a succession in the series of phenomena,
7826and in this succession an existence which follows a non-existence, or
7827conversely, consequently, change. For to say, that the non-existence
7828of a thing is not self-contradictory is a lame appeal to a logical
7829condition, which is no doubt a necessary condition of the existence of
7830the conception, but is far from being sufficient for the real objective
7831possibility of non-existence. I can annihilate in thought every existing
7832substance without self-contradiction, but I cannot infer from this their
7833objective contingency in existence, that is to say, the possibility of
7834their non-existence in itself. As regards the category of community,
7835it may easily be inferred that, as the pure categories of substance and
7836causality are incapable of a definition and explanation sufficient to
7837determine their object without the aid of intuition, the category
7838of reciprocal causality in the relation of substances to each other
7839(commercium) is just as little susceptible thereof. Possibility,
7840existence, and necessity nobody has ever yet been able to explain
7841without being guilty of manifest tautology, when the definition has been
7842drawn entirely from the pure understanding. For the substitution of the
7843logical possibility of the conception--the condition of which is that
7844it be not self-contradictory, for the transcendental possibility of
7845things--the condition of which is that there be an object corresponding
7846to the conception, is a trick which can only deceive the inexperienced.*
7847
7848
7849 [*Footnote: In one word, to none of these conceptions belongs a
7850corresponding object, and consequently their real possibility cannot
7851be demonstrated, if we take away sensuous intuition--the only intuition
7852which we possess--and there then remains nothing but the logical
7853possibility, that is, the fact that the conception or thought is
7854possible--which, however, is not the question; what we want to know
7855being, whether it relates to an object and thus possesses any meaning.]
7856
7857
7858It follows incontestably, that the pure conceptions of the understanding
7859are incapable of transcendental, and must always be of empirical use
7860alone, and that the principles of the pure understanding relate only
7861to the general conditions of a possible experience, to objects of the
7862senses, and never to things in general, apart from the mode in which we
7863intuite them.
7864
7865Transcendental analytic has accordingly this important result, to wit,
7866that the understanding is competent' effect nothing a priori, except the
7867anticipation of the form of a possible experience in general, and that,
7868as that which is not phenomenon cannot be an object of experience, it
7869can never overstep the limits of sensibility, within which alone
7870objects are presented to us. Its principles are merely principles of
7871the exposition of phenomena, and the proud name of an ontology, which
7872professes to present synthetical cognitions a priori of things in
7873general in a systematic doctrine, must give place to the modest title of
7874analytic of the pure understanding.
7875
7876Thought is the act of referring a given intuition to an object. If
7877the mode of this intuition is unknown to us, the object is merely
7878transcendental, and the conception of the understanding is employed only
7879transcendentally, that is, to produce unity in the thought of a manifold
7880in general. Now a pure category, in which all conditions of sensuous
7881intuition--as the only intuition we possess--are abstracted, does not
7882determine an object, but merely expresses the thought of an object in
7883general, according to different modes. Now, to employ a conception, the
7884function of judgement is required, by which an object is subsumed under
7885the conception, consequently the at least formal condition, under which
7886something can be given in intuition. Failing this condition of judgement
7887(schema), subsumption is impossible; for there is in such a case
7888nothing given, which may be subsumed under the conception. The merely
7889transcendental use of the categories is therefore, in fact, no use at
7890all and has no determined, or even, as regards its form, determinable
7891object. Hence it follows that the pure category is incompetent to
7892establish a synthetical a priori principle, and that the principles of
7893the pure understanding are only of empirical and never of transcendental
7894use, and that beyond the sphere of possible experience no synthetical a
7895priori principles are possible.
7896
7897It may be advisable, therefore, to express ourselves thus. The pure
7898categories, apart from the formal conditions of sensibility, have
7899a merely transcendental meaning, but are nevertheless not of
7900transcendental use, because this is in itself impossible, inasmuch as
7901all the conditions of any employment or use of them (in judgements) are
7902absent, to wit, the formal conditions of the subsumption of an object
7903under these conceptions. As, therefore, in the character of pure
7904categories, they must be employed empirically, and cannot be employed
7905transcendentally, they are of no use at all, when separated from
7906sensibility, that is, they cannot be applied to an object. They are
7907merely the pure form of the employment of the understanding in respect
7908of objects in general and of thought, without its being at the same time
7909possible to think or to determine any object by their means. But there
7910lurks at the foundation of this subject an illusion which it is very
7911difficult to avoid. The categories are not based, as regards their
7912origin, upon sensibility, like the forms of intuition, space, and time;
7913they seem, therefore, to be capable of an application beyond the sphere
7914of sensuous objects. But this is not the case. They are nothing but mere
7915forms of thought, which contain only the logical faculty of uniting a
7916priori in consciousness the manifold given in intuition. Apart, then,
7917from the only intuition possible for us, they have still less meaning
7918than the pure sensuous forms, space and time, for through them an object
7919is at least given, while a mode of connection of the manifold, when the
7920intuition which alone gives the manifold is wanting, has no meaning at
7921all. At the same time, when we designate certain objects as phenomena or
7922sensuous existences, thus distinguishing our mode of intuiting them from
7923their own nature as things in themselves, it is evident that by this
7924very distinction we as it were place the latter, considered in this
7925their own nature, although we do not so intuite them, in opposition to
7926the former, or, on the other hand, we do so place other possible
7927things, which are not objects of our senses, but are cogitated by the
7928understanding alone, and call them intelligible existences (noumena).
7929Now the question arises whether the pure conceptions of our
7930understanding do possess significance in respect of these latter, and
7931may possibly be a mode of cognizing them.
7932
7933But we are met at the very commencement with an ambiguity, which may
7934easily occasion great misapprehension. The understanding, when it terms
7935an object in a certain relation phenomenon, at the same time forms out
7936of this relation a representation or notion of an object in itself, and
7937hence believes that it can form also conceptions of such objects. Now as
7938the understanding possesses no other fundamental conceptions besides the
7939categories, it takes for granted that an object considered as a thing
7940in itself must be capable of being thought by means of these pure
7941conceptions, and is thereby led to hold the perfectly undetermined
7942conception of an intelligible existence, a something out of the sphere
7943of our sensibility, for a determinate conception of an existence which
7944we can cognize in some way or other by means of the understanding.
7945
7946If, by the term noumenon, we understand a thing so far as it is not an
7947object of our sensuous intuition, thus making abstraction of our mode of
7948intuiting it, this is a noumenon in the negative sense of the word. But
7949if we understand by it an object of a non-sensuous intuition, we in this
7950case assume a peculiar mode of intuition, an intellectual intuition, to
7951wit, which does not, however, belong to us, of the very possibility of
7952which we have no notion--and this is a noumenon in the positive sense.
7953
7954The doctrine of sensibility is also the doctrine of noumena in the
7955negative sense, that is, of things which the understanding is obliged to
7956cogitate apart from any relation to our mode of intuition, consequently
7957not as mere phenomena, but as things in themselves. But the
7958understanding at the same time comprehends that it cannot employ its
7959categories for the consideration of things in themselves, because these
7960possess significance only in relation to the unity of intuitions in
7961space and time, and that they are competent to determine this unity by
7962means of general a priori connecting conceptions only on account of the
7963pure ideality of space and time. Where this unity of time is not to be
7964met with, as is the case with noumena, the whole use, indeed the whole
7965meaning of the categories is entirely lost, for even the possibility of
7966things to correspond to the categories is in this case incomprehensible.
7967On this point, I need only refer the reader to what I have said at the
7968commencement of the General Remark appended to the foregoing chapter.
7969Now, the possibility of a thing can never be proved from the fact that
7970the conception of it is not self-contradictory, but only by means of
7971an intuition corresponding to the conception. If, therefore, we wish to
7972apply the categories to objects which cannot be regarded as phenomena,
7973we must have an intuition different from the sensuous, and in this case
7974the objects would be a noumena in the positive sense of the word. Now,
7975as such an intuition, that is, an intellectual intuition, is no part of
7976our faculty of cognition, it is absolutely impossible for the categories
7977to possess any application beyond the limits of experience. It may be
7978true that there are intelligible existences to which our faculty of
7979sensuous intuition has no relation, and cannot be applied, but our
7980conceptions of the understanding, as mere forms of thought for our
7981sensuous intuition, do not extend to these. What, therefore, we call
7982noumenon must be understood by us as such in a negative sense.
7983
7984If I take away from an empirical intuition all thought (by means of the
7985categories), there remains no cognition of any object; for by means of
7986mere intuition nothing is cogitated, and, from the existence of such
7987or such an affection of sensibility in me, it does not follow that this
7988affection or representation has any relation to an object without
7989me. But if I take away all intuition, there still remains the form of
7990thought, that is, the mode of determining an object for the manifold
7991of a possible intuition. Thus the categories do in some measure really
7992extend further than sensuous intuition, inasmuch as they think objects
7993in general, without regard to the mode (of sensibility) in which
7994these objects are given. But they do not for this reason apply to and
7995determine a wider sphere of objects, because we cannot assume that such
7996can be given, without presupposing the possibility of another than
7997the sensuous mode of intuition, a supposition we are not justified in
7998making.
7999
8000I call a conception problematical which contains in itself no
8001contradiction, and which is connected with other cognitions as a
8002limitation of given conceptions, but whose objective reality cannot
8003be cognized in any manner. The conception of a noumenon, that is, of a
8004thing which must be cogitated not as an object of sense, but as a
8005thing in itself (solely through the pure understanding), is not
8006self-contradictory, for we are not entitled to maintain that sensibility
8007is the only possible mode of intuition. Nay, further, this conception is
8008necessary to restrain sensuous intuition within the bounds of phenomena,
8009and thus to limit the objective validity of sensuous cognition; for
8010things in themselves, which lie beyond its province, are called noumena
8011for the very purpose of indicating that this cognition does not extend
8012its application to all that the understanding thinks. But, after all,
8013the possibility of such noumena is quite incomprehensible, and beyond
8014the sphere of phenomena, all is for us a mere void; that is to say,
8015we possess an understanding whose province does problematically extend
8016beyond this sphere, but we do not possess an intuition, indeed, not even
8017the conception of a possible intuition, by means of which objects beyond
8018the region of sensibility could be given us, and in reference to which
8019the understanding might be employed assertorically. The conception of a
8020noumenon is therefore merely a limitative conception and therefore only
8021of negative use. But it is not an arbitrary or fictitious notion, but
8022is connected with the limitation of sensibility, without, however, being
8023capable of presenting us with any positive datum beyond this sphere.
8024
8025The division of objects into phenomena and noumena, and of the
8026world into a mundus sensibilis and intelligibilis is therefore quite
8027inadmissible in a positive sense, although conceptions do certainly
8028admit of such a division; for the class of noumena have no determinate
8029object corresponding to them, and cannot therefore possess objective
8030validity. If we abandon the senses, how can it be made conceivable
8031that the categories (which are the only conceptions that could serve as
8032conceptions for noumena) have any sense or meaning at all, inasmuch
8033as something more than the mere unity of thought, namely, a possible
8034intuition, is requisite for their application to an object? The
8035conception of a noumenon, considered as merely problematical, is,
8036however, not only admissible, but, as a limitative conception of
8037sensibility, absolutely necessary. But, in this case, a noumenon is not
8038a particular intelligible object for our understanding; on the contrary,
8039the kind of understanding to which it could belong is itself a problem,
8040for we cannot form the most distant conception of the possibility of an
8041understanding which should cognize an object, not discursively by
8042means of categories, but intuitively in a non-sensuous intuition. Our
8043understanding attains in this way a sort of negative extension. That is
8044to say, it is not limited by, but rather limits, sensibility, by giving
8045the name of noumena to things, not considered as phenomena, but as
8046things in themselves. But it at the same time prescribes limits to
8047itself, for it confesses itself unable to cognize these by means of the
8048categories, and hence is compelled to cogitate them merely as an unknown
8049something.
8050
8051I find, however, in the writings of modern authors, an entirely
8052different use of the expressions, mundus sensibilis and intelligibilis,
8053which quite departs from the meaning of the ancients--an acceptation in
8054which, indeed, there is to be found no difficulty, but which at the same
8055time depends on mere verbal quibbling. According to this meaning,
8056some have chosen to call the complex of phenomena, in so far as it is
8057intuited, mundus sensibilis, but in so far as the connection thereof is
8058cogitated according to general laws of thought, mundus intelligibilis.
8059Astronomy, in so far as we mean by the word the mere observation of the
8060starry heaven, may represent the former; a system of astronomy, such as
8061the Copernican or Newtonian, the latter. But such twisting of words is a
8062mere sophistical subterfuge, to avoid a difficult question, by modifying
8063its meaning to suit our own convenience. To be sure, understanding and
8064reason are employed in the cognition of phenomena; but the question is,
8065whether these can be applied when the object is not a phenomenon and in
8066this sense we regard it if it is cogitated as given to the understanding
8067alone, and not to the senses. The question therefore is whether, over
8068and above the empirical use of the understanding, a transcendental use
8069is possible, which applies to the noumenon as an object. This question
8070we have answered in the negative.
8071
8072When therefore we say, the senses represent objects as they appear, the
8073understanding as they are, the latter statement must not be understood
8074in a transcendental, but only in an empirical signification, that is,
8075as they must be represented in the complete connection of phenomena, and
8076not according to what they may be, apart from their relation to possible
8077experience, consequently not as objects of the pure understanding. For
8078this must ever remain unknown to us. Nay, it is also quite unknown to us
8079whether any such transcendental or extraordinary cognition is possible
8080under any circumstances, at least, whether it is possible by means of
8081our categories. Understanding and sensibility, with us, can determine
8082objects only in conjunction. If we separate them, we have intuitions
8083without conceptions, or conceptions without intuitions; in both cases,
8084representations, which we cannot apply to any determinate object.
8085
8086If, after all our inquiries and explanations, any one still hesitates to
8087abandon the mere transcendental use of the categories, let him attempt
8088to construct with them a synthetical proposition. It would, of course,
8089be unnecessary for this purpose to construct an analytical proposition,
8090for that does not extend the sphere of the understanding, but, being
8091concerned only about what is cogitated in the conception itself, it
8092leaves it quite undecided whether the conception has any relation to
8093objects, or merely indicates the unity of thought--complete abstraction
8094being made of the modi in which an object may be given: in such a
8095proposition, it is sufficient for the understanding to know what lies
8096in the conception--to what it applies is to it indifferent. The attempt
8097must therefore be made with a synthetical and so-called transcendental
8098principle, for example: "Everything that exists, exists as substance,"
8099or, "Everything that is contingent exists as an effect of some other
8100thing, viz., of its cause." Now I ask, whence can the understanding draw
8101these synthetical propositions, when the conceptions contained therein
8102do not relate to possible experience but to things in themselves
8103(noumena)? Where is to be found the third term, which is always
8104requisite PURE site in a synthetical proposition, which may connect
8105in the same proposition conceptions which have no logical (analytical)
8106connection with each other? The proposition never will be demonstrated,
8107nay, more, the possibility of any such pure assertion never can
8108be shown, without making reference to the empirical use of the
8109understanding, and thus, ipso facto, completely renouncing pure
8110and non-sensuous judgement. Thus the conception of pure and merely
8111intelligible objects is completely void of all principles of its
8112application, because we cannot imagine any mode in which they might be
8113given, and the problematical thought which leaves a place open for
8114them serves only, like a void space, to limit the use of empirical
8115principles, without containing at the same time any other object of
8116cognition beyond their sphere.
8117
8118
8119
8120APPENDIX.
8121
8122Of the Equivocal Nature or Amphiboly of the Conceptions of Reflection
8123from the Confusion of the Transcendental with the Empirical use of the
8124Understanding.
8125
8126Reflection (reflexio) is not occupied about objects themselves, for the
8127purpose of directly obtaining conceptions of them, but is that state of
8128the mind in which we set ourselves to discover the subjective conditions
8129under which we obtain conceptions. It is the consciousness of the
8130relation of given representations to the different sources or faculties
8131of cognition, by which alone their relation to each other can be
8132rightly determined. The first question which occurs in considering our
8133representations is to what faculty of cognition do they belong? To the
8134understanding or to the senses? Many judgements are admitted to be true
8135from mere habit or inclination; but, because reflection neither precedes
8136nor follows, it is held to be a judgement that has its origin in the
8137understanding. All judgements do not require examination, that is,
8138investigation into the grounds of their truth. For, when they are
8139immediately certain (for example: "Between two points there can be only
8140one straight line"), no better or less mediate test of their truth can
8141be found than that which they themselves contain and express. But
8142all judgement, nay, all comparisons require reflection, that is, a
8143distinction of the faculty of cognition to which the given conceptions
8144belong. The act whereby I compare my representations with the faculty of
8145cognition which originates them, and whereby I distinguish whether they
8146are compared with each other as belonging to the pure understanding
8147or to sensuous intuition, I term transcendental reflection. Now, the
8148relations in which conceptions can stand to each other are those of
8149identity and difference, agreement and opposition, of the internal and
8150external, finally, of the determinable and the determining (matter
8151and form). The proper determination of these relations rests on the
8152question, to what faculty of cognition they subjectively belong, whether
8153to sensibility or understanding? For, on the manner in which we solve
8154this question depends the manner in which we must cogitate these
8155relations.
8156
8157Before constructing any objective judgement, we compare the conceptions
8158that are to be placed in the judgement, and observe whether there exists
8159identity (of many representations in one conception), if a general
8160judgement is to be constructed, or difference, if a particular; whether
8161there is agreement when affirmative; and opposition when negative
8162judgements are to be constructed, and so on. For this reason we ought
8163to call these conceptions, conceptions of comparison (conceptus
8164comparationis). But as, when the question is not as to the logical form,
8165but as to the content of conceptions, that is to say, whether the things
8166themselves are identical or different, in agreement or opposition,
8167and so on, the things can have a twofold relation to our faculty
8168of cognition, to wit, a relation either to sensibility or to the
8169understanding, and as on this relation depends their relation to
8170each other, transcendental reflection, that is, the relation of given
8171representations to one or the other faculty of cognition, can alone
8172determine this latter relation. Thus we shall not be able to discover
8173whether the things are identical or different, in agreement or
8174opposition, etc., from the mere conception of the things by means
8175of comparison (comparatio), but only by distinguishing the mode
8176of cognition to which they belong, in other words, by means of
8177transcendental reflection. We may, therefore, with justice say, that
8178logical reflection is mere comparison, for in it no account is taken of
8179the faculty of cognition to which the given conceptions belong, and
8180they are consequently, as far as regards their origin, to be treated
8181as homogeneous; while transcendental reflection (which applies to the
8182objects themselves) contains the ground of the possibility of objective
8183comparison of representations with each other, and is therefore very
8184different from the former, because the faculties of cognition to which
8185they belong are not even the same. Transcendental reflection is a duty
8186which no one can neglect who wishes to establish an a priori judgement
8187upon things. We shall now proceed to fulfil this duty, and thereby throw
8188not a little light on the question as to the determination of the proper
8189business of the understanding.
8190
81911. Identity and Difference. When an object is presented to us several
8192times, but always with the same internal determinations (qualitas et
8193quantitas), it, if an object of pure understanding, is always the same,
8194not several things, but only one thing (numerica identitas); but if a
8195phenomenon, we do not concern ourselves with comparing the conception of
8196the thing with the conception of some other, but, although they may be
8197in this respect perfectly the same, the difference of place at the same
8198time is a sufficient ground for asserting the numerical difference of
8199these objects (of sense). Thus, in the case of two drops of water, we
8200may make complete abstraction of all internal difference (quality and
8201quantity), and, the fact that they are intuited at the same time in
8202different places, is sufficient to justify us in holding them to
8203be numerically different. Leibnitz regarded phenomena as things in
8204themselves, consequently as intelligibilia, that is, objects of pure
8205understanding (although, on account of the confused nature of their
8206representations, he gave them the name of phenomena), and in this
8207case his principle of the indiscernible (principium identatis
8208indiscernibilium) is not to be impugned. But, as phenomena are objects
8209of sensibility, and, as the understanding, in respect of them, must be
8210employed empirically and not purely or transcendentally, plurality
8211and numerical difference are given by space itself as the condition of
8212external phenomena. For one part of space, although it may be perfectly
8213similar and equal to another part, is still without it, and for this
8214reason alone is different from the latter, which is added to it in order
8215to make up a greater space. It follows that this must hold good of
8216all things that are in the different parts of space at the same time,
8217however similar and equal one may be to another.
8218
82192. Agreement and Opposition. When reality is represented by the pure
8220understanding (realitas noumenon), opposition between realities is
8221incogitable--such a relation, that is, that when these realities are
8222connected in one subject, they annihilate the effects of each other and
8223may be represented in the formula 3 -3 = 0. On the other hand, the
8224real in a phenomenon (realitas phaenomenon) may very well be in mutual
8225opposition, and, when united in the same subject, the one may completely
8226or in part annihilate the effect or consequence of the other; as in the
8227case of two moving forces in the same straight line drawing or
8228impelling a point in opposite directions, or in the case of a pleasure
8229counterbalancing a certain amount of pain.
8230
82313. The Internal and External. In an object of the pure understanding,
8232only that is internal which has no relation (as regards its existence)
8233to anything different from itself. On the other hand, the internal
8234determinations of a substantia phaenomenon in space are nothing
8235but relations, and it is itself nothing more than a complex of mere
8236relations. Substance in space we are cognizant of only through forces
8237operative in it, either drawing others towards itself (attraction),
8238or preventing others from forcing into itself (repulsion and
8239impenetrability). We know no other properties that make up the
8240conception of substance phenomenal in space, and which we term matter.
8241On the other hand, as an object of the pure understanding, every
8242substance must have internal determination and forces. But what other
8243internal attributes of such an object can I think than those which my
8244internal sense presents to me? That, to wit, which in either itself
8245thought, or something analogous to it. Hence Leibnitz, who looked upon
8246things as noumena, after denying them everything like external relation,
8247and therefore also composition or combination, declared that all
8248substances, even the component parts of matter, were simple substances
8249with powers of representation, in one word, monads.
8250
82514. Matter and Form. These two conceptions lie at the foundation of all
8252other reflection, so inseparably are they connected with every mode of
8253exercising the understanding. The former denotes the determinable in
8254general, the second its determination, both in a transcendental sense,
8255abstraction being made of every difference in that which is given, and
8256of the mode in which it is determined. Logicians formerly termed the
8257universal, matter, the specific difference of this or that part of
8258the universal, form. In a judgement one may call the given conceptions
8259logical matter (for the judgement), the relation of these to each other
8260(by means of the copula), the form of the judgement. In an object, the
8261composite parts thereof (essentialia) are the matter; the mode in which
8262they are connected in the object, the form. In respect to things
8263in general, unlimited reality was regarded as the matter of all
8264possibility, the limitation thereof (negation) as the form, by which
8265one thing is distinguished from another according to transcendental
8266conceptions. The understanding demands that something be given (at least
8267in the conception), in order to be able to determine it in a certain
8268manner. Hence, in a conception of the pure understanding, the matter
8269precedes the form, and for this reason Leibnitz first assumed the
8270existence of things (monads) and of an internal power of representation
8271in them, in order to found upon this their external relation and the
8272community their state (that is, of their representations). Hence, with
8273him, space and time were possible--the former through the relation of
8274substances, the latter through the connection of their determinations
8275with each other, as causes and effects. And so would it really be,
8276if the pure understanding were capable of an immediate application
8277to objects, and if space and time were determinations of things in
8278themselves. But being merely sensuous intuitions, in which we determine
8279all objects solely as phenomena, the form of intuition (as a subjective
8280property of sensibility) must antecede all matter (sensations),
8281consequently space and time must antecede all phenomena and all data
8282of experience, and rather make experience itself possible. But the
8283intellectual philosopher could not endure that the form should precede
8284the things themselves and determine their possibility; an objection
8285perfectly correct, if we assume that we intuite things as they are,
8286although with confused representation. But as sensuous intuition is a
8287peculiar subjective condition, which is a priori at the foundation of
8288all perception, and the form of which is primitive, the form must be
8289given per se, and so far from matter (or the things themselves which
8290appear) lying at the foundation of experience (as we must conclude,
8291if we judge by mere conceptions), the very possibility of itself
8292presupposes, on the contrary, a given formal intuition (space and time).
8293
8294
8295
8296REMARK ON THE AMPHIBOLY OF THE CONCEPTIONS OF REFLECTION.
8297
8298Let me be allowed to term the position which we assign to a
8299conception either in the sensibility or in the pure understanding, the
8300transcendental place. In this manner, the appointment of the position
8301which must be taken by each conception according to the difference
8302in its use, and the directions for determining this place to all
8303conceptions according to rules, would be a transcendental topic, a
8304doctrine which would thoroughly shield us from the surreptitious devices
8305of the pure understanding and the delusions which thence arise, as it
8306would always distinguish to what faculty of cognition each conception
8307properly belonged. Every conception, every title, under which many
8308cognitions rank together, may be called a logical place. Upon this is
8309based the logical topic of Aristotle, of which teachers and rhetoricians
8310could avail themselves, in order, under certain titles of thought, to
8311observe what would best suit the matter they had to treat, and thus
8312enable themselves to quibble and talk with fluency and an appearance of
8313profundity.
8314
8315Transcendental topic, on the contrary, contains nothing more than the
8316above-mentioned four titles of all comparison and distinction, which
8317differ from categories in this respect, that they do not represent the
8318object according to that which constitutes its conception (quantity,
8319reality), but set forth merely the comparison of representations, which
8320precedes our conceptions of things. But this comparison requires a
8321previous reflection, that is, a determination of the place to which the
8322representations of the things which are compared belong, whether,
8323to wit, they are cogitated by the pure understanding, or given by
8324sensibility.
8325
8326Conceptions may be logically compared without the trouble of inquiring
8327to what faculty their objects belong, whether as noumena, to the
8328understanding, or as phenomena, to sensibility. If, however, we wish to
8329employ these conceptions in respect of objects, previous transcendental
8330reflection is necessary. Without this reflection I should make a very
8331unsafe use of these conceptions, and construct pretended synthetical
8332propositions which critical reason cannot acknowledge and which
8333are based solely upon a transcendental amphiboly, that is, upon a
8334substitution of an object of pure understanding for a phenomenon.
8335
8336For want of this doctrine of transcendental topic, and consequently
8337deceived by the amphiboly of the conceptions of reflection, the
8338celebrated Leibnitz constructed an intellectual system of the world,
8339or rather, believed himself competent to cognize the internal nature of
8340things, by comparing all objects merely with the understanding and the
8341abstract formal conceptions of thought. Our table of the conceptions of
8342reflection gives us the unexpected advantage of being able to exhibit
8343the distinctive peculiarities of his system in all its parts, and at the
8344same time of exposing the fundamental principle of this peculiar mode of
8345thought, which rested upon naught but a misconception. He compared all
8346things with each other merely by means of conceptions, and naturally
8347found no other differences than those by which the understanding
8348distinguishes its pure conceptions one from another. The conditions
8349of sensuous intuition, which contain in themselves their own means of
8350distinction, he did not look upon as primitive, because sensibility
8351was to him but a confused mode of representation and not any particular
8352source of representations. A phenomenon was for him the representation
8353of the thing in itself, although distinguished from cognition by the
8354understanding only in respect of the logical form--the former with its
8355usual want of analysis containing, according to him, a certain mixture
8356of collateral representations in its conception of a thing, which it is
8357the duty of the understanding to separate and distinguish. In one word,
8358Leibnitz intellectualized phenomena, just as Locke, in his system
8359of noogony (if I may be allowed to make use of such expressions),
8360sensualized the conceptions of the understanding, that is to say,
8361declared them to be nothing more than empirical or abstract conceptions
8362of reflection. Instead of seeking in the understanding and sensibility
8363two different sources of representations, which, however, can present us
8364with objective judgements of things only in conjunction, each of
8365these great men recognized but one of these faculties, which, in their
8366opinion, applied immediately to things in themselves, the other having
8367no duty but that of confusing or arranging the representations of the
8368former.
8369
8370Accordingly, the objects of sense were compared by Leibnitz as things in
8371general merely in the understanding.
8372
83731st. He compares them in regard to their identity or difference--as
8374judged by the understanding. As, therefore, he considered merely the
8375conceptions of objects, and not their position in intuition, in
8376which alone objects can be given, and left quite out of sight the
8377transcendental locale of these conceptions--whether, that is, their
8378object ought to be classed among phenomena, or among things in
8379themselves, it was to be expected that he should extend the application
8380of the principle of indiscernibles, which is valid solely of conceptions
8381of things in general, to objects of sense (mundus phaenomenon), and that
8382he should believe that he had thereby contributed in no small degree to
8383extend our knowledge of nature. In truth, if I cognize in all its inner
8384determinations a drop of water as a thing in itself, I cannot look upon
8385one drop as different from another, if the conception of the one is
8386completely identical with that of the other. But if it is a phenomenon
8387in space, it has a place not merely in the understanding (among
8388conceptions), but also in sensuous external intuition (in space), and in
8389this case, the physical locale is a matter of indifference in regard to
8390the internal determinations of things, and one place, B, may contain
8391a thing which is perfectly similar and equal to another in a place, A,
8392just as well as if the two things were in every respect different from
8393each other. Difference of place without any other conditions, makes the
8394plurality and distinction of objects as phenomena, not only possible in
8395itself, but even necessary. Consequently, the above so-called law is not
8396a law of nature. It is merely an analytical rule for the comparison of
8397things by means of mere conceptions.
8398
83992nd. The principle: "Realities (as simple affirmations) never logically
8400contradict each other," is a proposition perfectly true respecting the
8401relation of conceptions, but, whether as regards nature, or things in
8402themselves (of which we have not the slightest conception), is without
8403any the least meaning. For real opposition, in which A -B is = 0,
8404exists everywhere, an opposition, that is, in which one reality united
8405with another in the same subject annihilates the effects of the other--a
8406fact which is constantly brought before our eyes by the different
8407antagonistic actions and operations in nature, which, nevertheless, as
8408depending on real forces, must be called realitates phaenomena. General
8409mechanics can even present us with the empirical condition of this
8410opposition in an a priori rule, as it directs its attention to the
8411opposition in the direction of forces--a condition of which the
8412transcendental conception of reality can tell us nothing. Although M.
8413Leibnitz did not announce this proposition with precisely the pomp of
8414a new principle, he yet employed it for the establishment of
8415new propositions, and his followers introduced it into their
8416Leibnitzio-Wolfian system of philosophy. According to this principle,
8417for example, all evils are but consequences of the limited nature of
8418created beings, that is, negations, because these are the only opposite
8419of reality. (In the mere conception of a thing in general this is really
8420the case, but not in things as phenomena.) In like manner, the upholders
8421of this system deem it not only possible, but natural also, to connect
8422and unite all reality in one being, because they acknowledge no other
8423sort of opposition than that of contradiction (by which the conception
8424itself of a thing is annihilated), and find themselves unable to
8425conceive an opposition of reciprocal destruction, so to speak, in which
8426one real cause destroys the effect of another, and the conditions of
8427whose representation we meet with only in sensibility.
8428
84293rd. The Leibnitzian monadology has really no better foundation than on
8430this philosopher's mode of falsely representing the difference of
8431the internal and external solely in relation to the understanding.
8432Substances, in general, must have something inward, which is therefore
8433free from external relations, consequently from that of composition
8434also. The simple--that which can be represented by a unit--is therefore
8435the foundation of that which is internal in things in themselves. The
8436internal state of substances cannot therefore consist in place, shape,
8437contact, or motion, determinations which are all external relations,
8438and we can ascribe to them no other than that whereby we internally
8439determine our faculty of sense itself, that is to say, the state of
8440representation. Thus, then, were constructed the monads, which were to
8441form the elements of the universe, the active force of which consists in
8442representation, the effects of this force being thus entirely confined
8443to themselves.
8444
8445For the same reason, his view of the possible community of substances
8446could not represent it but as a predetermined harmony, and by no means
8447as a physical influence. For inasmuch as everything is occupied only
8448internally, that is, with its own representations, the state of the
8449representations of one substance could not stand in active and living
8450connection with that of another, but some third cause operating on all
8451without exception was necessary to make the different states correspond
8452with one another. And this did not happen by means of assistance applied
8453in each particular case (systema assistentiae), but through the unity of
8454the idea of a cause occupied and connected with all substances, in which
8455they necessarily receive, according to the Leibnitzian school, their
8456existence and permanence, consequently also reciprocal correspondence,
8457according to universal laws.
8458
84594th. This philosopher's celebrated doctrine of space and time, in which
8460he intellectualized these forms of sensibility, originated in the same
8461delusion of transcendental reflection. If I attempt to represent by the
8462mere understanding, the external relations of things, I can do so only
8463by employing the conception of their reciprocal action, and if I wish
8464to connect one state of the same thing with another state, I must avail
8465myself of the notion of the order of cause and effect. And thus Leibnitz
8466regarded space as a certain order in the community of substances, and
8467time as the dynamical sequence of their states. That which space and
8468time possess proper to themselves and independent of things, he ascribed
8469to a necessary confusion in our conceptions of them, whereby that which
8470is a mere form of dynamical relations is held to be a self-existent
8471intuition, antecedent even to things themselves. Thus space and time
8472were the intelligible form of the connection of things (substances and
8473their states) in themselves. But things were intelligible substances
8474(substantiae noumena). At the same time, he made these conceptions valid
8475of phenomena, because he did not allow to sensibility a peculiar mode of
8476intuition, but sought all, even the empirical representation of objects,
8477in the understanding, and left to sense naught but the despicable task
8478of confusing and disarranging the representations of the former.
8479
8480But even if we could frame any synthetical proposition concerning things
8481in themselves by means of the pure understanding (which is impossible),
8482it could not apply to phenomena, which do not represent things in
8483themselves. In such a case I should be obliged in transcendental
8484reflection to compare my conceptions only under the conditions of
8485sensibility, and so space and time would not be determinations of things
8486in themselves, but of phenomena. What things may be in themselves, I
8487know not and need not know, because a thing is never presented to me
8488otherwise than as a phenomenon.
8489
8490I must adopt the same mode of procedure with the other conceptions
8491of reflection. Matter is substantia phaenomenon. That in it which is
8492internal I seek to discover in all parts of space which it occupies, and
8493in all the functions and operations it performs, and which are indeed
8494never anything but phenomena of the external sense. I cannot therefore
8495find anything that is absolutely, but only what is comparatively
8496internal, and which itself consists of external relations. The
8497absolutely internal in matter, and as it should be according to the pure
8498understanding, is a mere chimera, for matter is not an object for
8499the pure understanding. But the transcendental object, which is the
8500foundation of the phenomenon which we call matter, is a mere nescio
8501quid, the nature of which we could not understand, even though someone
8502were found able to tell us. For we can understand nothing that does not
8503bring with it something in intuition corresponding to the expressions
8504employed. If, by the complaint of being unable to perceive the internal
8505nature of things, it is meant that we do not comprehend by the pure
8506understanding what the things which appear to us may be in themselves,
8507it is a silly and unreasonable complaint; for those who talk thus really
8508desire that we should be able to cognize, consequently to intuite,
8509things without senses, and therefore wish that we possessed a faculty
8510of cognition perfectly different from the human faculty, not merely in
8511degree, but even as regards intuition and the mode thereof, so that thus
8512we should not be men, but belong to a class of beings, the possibility
8513of whose existence, much less their nature and constitution, we have
8514no means of cognizing. By observation and analysis of phenomena we
8515penetrate into the interior of nature, and no one can say what progress
8516this knowledge may make in time. But those transcendental questions
8517which pass beyond the limits of nature, we could never answer, even
8518although all nature were laid open to us, because we have not the power
8519of observing our own mind with any other intuition than that of our
8520internal sense. For herein lies the mystery of the origin and source
8521of our faculty of sensibility. Its application to an object, and the
8522transcendental ground of this unity of subjective and objective, lie too
8523deeply concealed for us, who cognize ourselves only through the
8524internal sense, consequently as phenomena, to be able to discover in our
8525existence anything but phenomena, the non-sensuous cause of which we at
8526the same time earnestly desire to penetrate to.
8527
8528The great utility of this critique of conclusions arrived at by the
8529processes of mere reflection consists in its clear demonstration of the
8530nullity of all conclusions respecting objects which are compared
8531with each other in the understanding alone, while it at the same time
8532confirms what we particularly insisted on, namely, that, although
8533phenomena are not included as things in themselves among the objects of
8534the pure understanding, they are nevertheless the only things by which
8535our cognition can possess objective reality, that is to say, which give
8536us intuitions to correspond with our conceptions.
8537
8538When we reflect in a purely logical manner, we do nothing more than
8539compare conceptions in our understanding, to discover whether both have
8540the same content, whether they are self-contradictory or not, whether
8541anything is contained in either conception, which of the two is given,
8542and which is merely a mode of thinking that given. But if I apply these
8543conceptions to an object in general (in the transcendental sense),
8544without first determining whether it is an object of sensuous or
8545intellectual intuition, certain limitations present themselves, which
8546forbid us to pass beyond the conceptions and render all empirical use
8547of them impossible. And thus these limitations prove that the
8548representation of an object as a thing in general is not only
8549insufficient, but, without sensuous determination and independently of
8550empirical conditions, self-contradictory; that we must therefore make
8551abstraction of all objects, as in logic, or, admitting them, must think
8552them under conditions of sensuous intuition; that, consequently, the
8553intelligible requires an altogether peculiar intuition, which we do not
8554possess, and in the absence of which it is for us nothing; while, on the
8555other hand phenomena cannot be objects in themselves. For, when I merely
8556think things in general, the difference in their external relations
8557cannot constitute a difference in the things themselves; on the
8558contrary, the former presupposes the latter, and if the conception of
8559one of two things is not internally different from that of the other,
8560I am merely thinking the same thing in different relations. Further,
8561by the addition of one affirmation (reality) to the other, the positive
8562therein is really augmented, and nothing is abstracted or withdrawn
8563from it; hence the real in things cannot be in contradiction with or
8564opposition to itself--and so on.
8565
8566The true use of the conceptions of reflection in the employment of the
8567understanding has, as we have shown, been so misconceived by Leibnitz,
8568one of the most acute philosophers of either ancient or modern times,
8569that he has been misled into the construction of a baseless system of
8570intellectual cognition, which professes to determine its objects without
8571the intervention of the senses. For this reason, the exposition of the
8572cause of the amphiboly of these conceptions, as the origin of these
8573false principles, is of great utility in determining with certainty the
8574proper limits of the understanding.
8575
8576It is right to say whatever is affirmed or denied of the whole of a
8577conception can be affirmed or denied of any part of it (dictum de omni
8578et nullo); but it would be absurd so to alter this logical proposition
8579as to say whatever is not contained in a general conception is likewise
8580not contained in the particular conceptions which rank under it; for
8581the latter are particular conceptions, for the very reason that
8582their content is greater than that which is cogitated in the general
8583conception. And yet the whole intellectual system of Leibnitz is based
8584upon this false principle, and with it must necessarily fall to the
8585ground, together with all the ambiguous principles in reference to the
8586employment of the understanding which have thence originated.
8587
8588Leibnitz's principle of the identity of indiscernibles or
8589indistinguishables is really based on the presupposition that, if in the
8590conception of a thing a certain distinction is not to be found, it is
8591also not to be met with in things themselves; that, consequently,
8592all things are completely identical (numero eadem) which are not
8593distinguishable from each other (as to quality or quantity) in our
8594conceptions of them. But, as in the mere conception of anything
8595abstraction has been made of many necessary conditions of intuition,
8596that of which abstraction has been made is rashly held to be
8597non-existent, and nothing is attributed to the thing but what is
8598contained in its conception.
8599
8600The conception of a cubic foot of space, however I may think it, is
8601in itself completely identical. But two cubic feet in space are
8602nevertheless distinct from each other from the sole fact of their being
8603in different places (they are numero diversa); and these places are
8604conditions of intuition, wherein the object of this conception is
8605given, and which do not belong to the conception, but to the faculty of
8606sensibility. In like manner, there is in the conception of a thing no
8607contradiction when a negative is not connected with an affirmative;
8608and merely affirmative conceptions cannot, in conjunction, produce any
8609negation. But in sensuous intuition, wherein reality (take for example,
8610motion) is given, we find conditions (opposite directions)--of which
8611abstraction has been made in the conception of motion in general--which
8612render possible a contradiction or opposition (not indeed of a logical
8613kind)--and which from pure positives produce zero = 0. We are therefore
8614not justified in saying that all reality is in perfect agreement
8615and harmony, because no contradiction is discoverable among its
8616conceptions.* According to mere conceptions, that which is internal
8617is the substratum of all relations or external determinations. When,
8618therefore, I abstract all conditions of intuition, and confine myself
8619solely to the conception of a thing in general, I can make abstraction
8620of all external relations, and there must nevertheless remain a
8621conception of that which indicates no relation, but merely internal
8622determinations. Now it seems to follow that in everything (substance)
8623there is something which is absolutely internal and which antecedes all
8624external determinations, inasmuch as it renders them possible; and
8625that therefore this substratum is something which does not contain any
8626external relations and is consequently simple (for corporeal things are
8627never anything but relations, at least of their parts external to
8628each other); and, inasmuch as we know of no other absolutely internal
8629determinations than those of the internal sense, this substratum is not
8630only simple, but also, analogously with our internal sense, determined
8631through representations, that is to say, all things are properly monads,
8632or simple beings endowed with the power of representation. Now all this
8633would be perfectly correct, if the conception of a thing were the
8634only necessary condition of the presentation of objects of external
8635intuition. It is, on the contrary, manifest that a permanent phenomenon
8636in space (impenetrable extension) can contain mere relations, and
8637nothing that is absolutely internal, and yet be the primary substratum
8638of all external perception. By mere conceptions I cannot think anything
8639external, without, at the same time, thinking something internal, for
8640the reason that conceptions of relations presuppose given things, and
8641without these are impossible. But, as an intuition there is something
8642(that is, space, which, with all it contains, consists of purely formal,
8643or, indeed, real relations) which is not found in the mere conception of
8644a thing in general, and this presents to us the substratum which could
8645not be cognized through conceptions alone, I cannot say: because a thing
8646cannot be represented by mere conceptions without something absolutely
8647internal, there is also, in the things themselves which are contained
8648under these conceptions, and in their intuition nothing external to
8649which something absolutely internal does not serve as the foundation.
8650For, when we have made abstraction of all the conditions of intuition,
8651there certainly remains in the mere conception nothing but the internal
8652in general, through which alone the external is possible. But this
8653necessity, which is grounded upon abstraction alone, does not obtain in
8654the case of things themselves, in so far as they are given in intuition
8655with such determinations as express mere relations, without having
8656anything internal as their foundation; for they are not things of a
8657thing of which we can neither for they are not things in themselves, but
8658only phenomena. What we cognize in matter is nothing but relations (what
8659we call its internal determinations are but comparatively internal). But
8660there are some self-subsistent and permanent, through which a determined
8661object is given. That I, when abstraction is made of these relations,
8662have nothing more to think, does not destroy the conception of a thing
8663as phenomenon, nor the conception of an object in abstracto, but it does
8664away with the possibility of an object that is determinable according to
8665mere conceptions, that is, of a noumenon. It is certainly startling to
8666hear that a thing consists solely of relations; but this thing is simply
8667a phenomenon, and cannot be cogitated by means of the mere categories:
8668it does itself consist in the mere relation of something in general to
8669the senses. In the same way, we cannot cogitate relations of things in
8670abstracto, if we commence with conceptions alone, in any other manner
8671than that one is the cause of determinations in the other; for that is
8672itself the conception of the understanding or category of relation. But,
8673as in this case we make abstraction of all intuition, we lose altogether
8674the mode in which the manifold determines to each of its parts its
8675place, that is, the form of sensibility (space); and yet this mode
8676antecedes all empirical causality.
8677
8678
8679
8680 [*Footnote: If any one wishes here to have recourse to the usual
8681subterfuge, and to say, that at least realitates noumena cannot be in
8682opposition to each other, it will be requisite for him to adduce an
8683example of this pure and non-sensuous reality, that it may be understood
8684whether the notion represents something or nothing. But an example
8685cannot be found except in experience, which never presents to us
8686anything more than phenomena; and thus the proposition means nothing
8687more than that the conception which contains only affirmatives does not
8688contain anything negative--a proposition nobody ever doubted.]
8689
8690
8691
8692If by intelligible objects we understand things which can be thought
8693by means of the pure categories, without the need of the schemata of
8694sensibility, such objects are impossible. For the condition of the
8695objective use of all our conceptions of understanding is the mode of
8696our sensuous intuition, whereby objects are given; and, if we make
8697abstraction of the latter, the former can have no relation to an object.
8698And even if we should suppose a different kind of intuition from our
8699own, still our functions of thought would have no use or signification
8700in respect thereof. But if we understand by the term, objects of a
8701non-sensuous intuition, in respect of which our categories are not
8702valid, and of which we can accordingly have no knowledge (neither
8703intuition nor conception), in this merely negative sense noumena must be
8704admitted. For this is no more than saying that our mode of intuition is
8705not applicable to all things, but only to objects of our senses,
8706that consequently its objective validity is limited, and that room is
8707therefore left for another kind of intuition, and thus also for things
8708that may be objects of it. But in this sense the conception of a
8709noumenon is problematical, that is to say, it is the notion of that it
8710that it is possible, nor that it is impossible, inasmuch as we do not
8711know of any mode of intuition besides the sensuous, or of any other sort
8712of conceptions than the categories--a mode of intuition and a kind of
8713conception neither of which is applicable to a non-sensuous object. We
8714are on this account incompetent to extend the sphere of our objects
8715of thought beyond the conditions of our sensibility, and to assume the
8716existence of objects of pure thought, that is, of noumena, inasmuch as
8717these have no true positive signification. For it must be confessed
8718of the categories that they are not of themselves sufficient for the
8719cognition of things in themselves and, without the data of sensibility,
8720are mere subjective forms of the unity of the understanding. Thought is
8721certainly not a product of the senses, and in so far is not limited by
8722them, but it does not therefore follow that it may be employed purely
8723and without the intervention of sensibility, for it would then be
8724without reference to an object. And we cannot call a noumenon an object
8725of pure thought; for the representation thereof is but the problematical
8726conception of an object for a perfectly different intuition and
8727a perfectly different understanding from ours, both of which are
8728consequently themselves problematical. The conception of a noumenon is
8729therefore not the conception of an object, but merely a problematical
8730conception inseparably connected with the limitation of our sensibility.
8731That is to say, this conception contains the answer to the question:
8732"Are there objects quite unconnected with, and independent of, our
8733intuition?"--a question to which only an indeterminate answer can be
8734given. That answer is: "Inasmuch as sensuous intuition does not apply
8735to all things without distinction, there remains room for other and
8736different objects." The existence of these problematical objects
8737is therefore not absolutely denied, in the absence of a determinate
8738conception of them, but, as no category is valid in respect of them,
8739neither must they be admitted as objects for our understanding.
8740
8741Understanding accordingly limits sensibility, without at the same time
8742enlarging its own field. While, moreover, it forbids sensibility to
8743apply its forms and modes to things in themselves and restricts it
8744to the sphere of phenomena, it cogitates an object in itself, only,
8745however, as a transcendental object, which is the cause of a phenomenon
8746(consequently not itself a phenomenon), and which cannot be thought
8747either as a quantity or as reality, or as substance (because these
8748conceptions always require sensuous forms in which to determine an
8749object)--an object, therefore, of which we are quite unable to say
8750whether it can be met with in ourselves or out of us, whether it would
8751be annihilated together with sensibility, or, if this were taken away,
8752would continue to exist. If we wish to call this object a noumenon,
8753because the representation of it is non-sensuous, we are at liberty
8754to do so. But as we can apply to it none of the conceptions of our
8755understanding, the representation is for us quite void, and is available
8756only for the indication of the limits of our sensuous intuition, thereby
8757leaving at the same time an empty space, which we are competent to
8758fill by the aid neither of possible experience, nor of the pure
8759understanding.
8760
8761The critique of the pure understanding, accordingly, does not permit us
8762to create for ourselves a new field of objects beyond those which are
8763presented to us as phenomena, and to stray into intelligible worlds;
8764nay, it does not even allow us to endeavour to form so much as a
8765conception of them. The specious error which leads to this--and which is
8766a perfectly excusable one--lies in the fact that the employment of the
8767understanding, contrary to its proper purpose and destination, is made
8768transcendental, and objects, that is, possible intuitions, are made to
8769regulate themselves according to conceptions, instead of the conceptions
8770arranging themselves according to the intuitions, on which alone their
8771own objective validity rests. Now the reason of this again is that
8772apperception, and with it thought, antecedes all possible determinate
8773arrangement of representations. Accordingly we think something in
8774general and determine it on the one hand sensuously, but, on the other,
8775distinguish the general and in abstracto represented object from this
8776particular mode of intuiting it. In this case there remains a mode of
8777determining the object by mere thought, which is really but a logical
8778form without content, which, however, seems to us to be a mode of
8779the existence of the object in itself (noumenon), without regard to
8780intuition which is limited to our senses.
8781
8782Before ending this transcendental analytic, we must make an addition,
8783which, although in itself of no particular importance, seems to be
8784necessary to the completeness of the system. The highest conception,
8785with which a transcendental philosophy commonly begins, is the division
8786into possible and impossible. But as all division presupposes a divided
8787conception, a still higher one must exist, and this is the conception of
8788an object in general--problematically understood and without its being
8789decided whether it is something or nothing. As the categories are the
8790only conceptions which apply to objects in general, the distinguishing
8791of an object, whether it is something or nothing, must proceed according
8792to the order and direction of the categories.
8793
87941. To the categories of quantity, that is, the conceptions of all, many,
8795and one, the conception which annihilates all, that is, the conception
8796of none, is opposed. And thus the object of a conception, to which no
8797intuition can be found to correspond, is = nothing. That is, it is a
8798conception without an object (ens rationis), like noumena, which cannot
8799be considered possible in the sphere of reality, though they must not
8800therefore be held to be impossible--or like certain new fundamental
8801forces in matter, the existence of which is cogitable without
8802contradiction, though, as examples from experience are not forthcoming,
8803they must not be regarded as possible.
8804
88052. Reality is something; negation is nothing, that is, a conception of
8806the absence of an object, as cold, a shadow (nihil privativum).
8807
88083. The mere form of intuition, without substance, is in itself no
8809object, but the merely formal condition of an object (as phenomenon),
8810as pure space and pure time. These are certainly something, as forms
8811of intuition, but are not themselves objects which are intuited (ens
8812imaginarium).
8813
88144. The object of a conception which is self-contradictory, is nothing,
8815because the conception is nothing--is impossible, as a figure composed
8816of two straight lines (nihil negativum).
8817
8818The table of this division of the conception of nothing (the
8819corresponding division of the conception of something does not require
8820special description) must therefore be arranged as follows:
8821
8822
8823 NOTHING
8824 AS
8825
8826 1
8827 As Empty Conception
8828 without object,
8829 ens rationis
8830 2 3
8831 Empty object of Empty intuition
8832 a conception, without object,
8833 nihil privativum ens imaginarium
8834 4
8835 Empty object
8836 without conception,
8837 nihil negativum
8838
8839
8840We see that the ens rationis is distinguished from the nihil negativum
8841or pure nothing by the consideration that the former must not be
8842reckoned among possibilities, because it is a mere fiction--though
8843not self-contradictory, while the latter is completely opposed to
8844all possibility, inasmuch as the conception annihilates itself. Both,
8845however, are empty conceptions. On the other hand, the nihil privativum
8846and ens imaginarium are empty data for conceptions. If light be not
8847given to the senses, we cannot represent to ourselves darkness, and if
8848extended objects are not perceived, we cannot represent space. Neither
8849the negation, nor the mere form of intuition can, without something
8850real, be an object.
8851
8852
8853
8854TRANSCENDENTAL LOGIC. SECOND DIVISION.
8855
8856TRANSCENDENTAL DIALECTIC. INTRODUCTION.
8857
8858I. Of Transcendental Illusory Appearance.
8859
8860We termed dialectic in general a logic of appearance. This does not
8861signify a doctrine of probability; for probability is truth, only
8862cognized upon insufficient grounds, and though the information it gives
8863us is imperfect, it is not therefore deceitful. Hence it must not be
8864separated from the analytical part of logic. Still less must phenomenon
8865and appearance be held to be identical. For truth or illusory appearance
8866does not reside in the object, in so far as it is intuited, but in the
8867judgement upon the object, in so far as it is thought. It is, therefore,
8868quite correct to say that the senses do not err, not because they always
8869judge correctly, but because they do not judge at all. Hence truth and
8870error, consequently also, illusory appearance as the cause of error, are
8871only to be found in a judgement, that is, in the relation of an object
8872to our understanding. In a cognition which completely harmonizes with
8873the laws of the understanding, no error can exist. In a representation
8874of the senses--as not containing any judgement--there is also no error.
8875But no power of nature can of itself deviate from its own laws. Hence
8876neither the understanding per se (without the influence of another
8877cause), nor the senses per se, would fall into error; the former could
8878not, because, if it acts only according to its own laws, the effect (the
8879judgement) must necessarily accord with these laws. But in accordance
8880with the laws of the understanding consists the formal element in all
8881truth. In the senses there is no judgement--neither a true nor a false
8882one. But, as we have no source of cognition besides these two, it
8883follows that error is caused solely by the unobserved influence of
8884the sensibility upon the understanding. And thus it happens that the
8885subjective grounds of a judgement and are confounded with the objective,
8886and cause them to deviate from their proper determination,* just as a
8887body in motion would always of itself proceed in a straight line, but
8888if another impetus gives to it a different direction, it will then
8889start off into a curvilinear line of motion. To distinguish the peculiar
8890action of the understanding from the power which mingles with it, it is
8891necessary to consider an erroneous judgement as the diagonal between two
8892forces, that determine the judgement in two different directions, which,
8893as it were, form an angle, and to resolve this composite operation into
8894the simple ones of the understanding and the sensibility. In pure
8895a priori judgements this must be done by means of transcendental
8896reflection, whereby, as has been already shown, each representation
8897has its place appointed in the corresponding faculty of cognition, and
8898consequently the influence of the one faculty upon the other is made
8899apparent.
8900
8901
8902 [*Footnote: Sensibility, subjected to the understanding, as the object
8903upon which the understanding employs its functions, is the source of
8904real cognitions. But, in so far as it exercises an influence upon the
8905action of the understanding and determines it to judgement, sensibility
8906is itself the cause of error.]
8907
8908
8909It is not at present our business to treat of empirical illusory
8910appearance (for example, optical illusion), which occurs in the
8911empirical application of otherwise correct rules of the understanding,
8912and in which the judgement is misled by the influence of imagination.
8913Our purpose is to speak of transcendental illusory appearance, which
8914influences principles--that are not even applied to experience, for in
8915this case we should possess a sure test of their correctness--but which
8916leads us, in disregard of all the warnings of criticism, completely
8917beyond the empirical employment of the categories and deludes us with
8918the chimera of an extension of the sphere of the pure understanding.
8919We shall term those principles the application of which is confined
8920entirely within the limits of possible experience, immanent; those,
8921on the other hand, which transgress these limits, we shall call
8922transcendent principles. But by these latter I do not understand
8923principles of the transcendental use or misuse of the categories, which
8924is in reality a mere fault of the judgement when not under due restraint
8925from criticism, and therefore not paying sufficient attention to the
8926limits of the sphere in which the pure understanding is allowed to
8927exercise its functions; but real principles which exhort us to break
8928down all those barriers, and to lay claim to a perfectly new field of
8929cognition, which recognizes no line of demarcation. Thus transcendental
8930and transcendent are not identical terms. The principles of the
8931pure understanding, which we have already propounded, ought to be
8932of empirical and not of transcendental use, that is, they are not
8933applicable to any object beyond the sphere of experience. A principle
8934which removes these limits, nay, which authorizes us to overstep them,
8935is called transcendent. If our criticism can succeed in exposing the
8936illusion in these pretended principles, those which are limited in their
8937employment to the sphere of experience may be called, in opposition to
8938the others, immanent principles of the pure understanding.
8939
8940Logical illusion, which consists merely in the imitation of the form of
8941reason (the illusion in sophistical syllogisms), arises entirely from
8942a want of due attention to logical rules. So soon as the attention
8943is awakened to the case before us, this illusion totally disappears.
8944Transcendental illusion, on the contrary, does not cease to exist, even
8945after it has been exposed, and its nothingness clearly perceived by
8946means of transcendental criticism. Take, for example, the illusion in
8947the proposition: "The world must have a beginning in time." The cause of
8948this is as follows. In our reason, subjectively considered as a faculty
8949of human cognition, there exist fundamental rules and maxims of its
8950exercise, which have completely the appearance of objective principles.
8951Now from this cause it happens that the subjective necessity of a
8952certain connection of our conceptions, is regarded as an objective
8953necessity of the determination of things in themselves. This illusion it
8954is impossible to avoid, just as we cannot avoid perceiving that the sea
8955appears to be higher at a distance than it is near the shore, because we
8956see the former by means of higher rays than the latter, or, which is a
8957still stronger case, as even the astronomer cannot prevent himself from
8958seeing the moon larger at its rising than some time afterwards, although
8959he is not deceived by this illusion.
8960
8961Transcendental dialectic will therefore content itself with exposing
8962the illusory appearance in transcendental judgements, and guarding us
8963against it; but to make it, as in the case of logical illusion, entirely
8964disappear and cease to be illusion is utterly beyond its power. For we
8965have here to do with a natural and unavoidable illusion, which rests
8966upon subjective principles and imposes these upon us as objective, while
8967logical dialectic, in the detection of sophisms, has to do merely with
8968an error in the logical consequence of the propositions, or with an
8969artificially constructed illusion, in imitation of the natural error.
8970There is, therefore, a natural and unavoidable dialectic of pure
8971reason--not that in which the bungler, from want of the requisite
8972knowledge, involves himself, nor that which the sophist devises for the
8973purpose of misleading, but that which is an inseparable adjunct of human
8974reason, and which, even after its illusions have been exposed, does not
8975cease to deceive, and continually to lead reason into momentary errors,
8976which it becomes necessary continually to remove.
8977
8978
8979
8980II. Of Pure Reason as the Seat of Transcendental Illusory Appearance.
8981
8982A. OF REASON IN GENERAL.
8983
8984All our knowledge begins with sense, proceeds thence to understanding,
8985and ends with reason, beyond which nothing higher can be discovered in
8986the human mind for elaborating the matter of intuition and subjecting it
8987to the highest unity of thought. At this stage of our inquiry it is my
8988duty to give an explanation of this, the highest faculty of cognition,
8989and I confess I find myself here in some difficulty. Of reason, as of
8990the understanding, there is a merely formal, that is, logical use, in
8991which it makes abstraction of all content of cognition; but there is
8992also a real use, inasmuch as it contains in itself the source of certain
8993conceptions and principles, which it does not borrow either from the
8994senses or the understanding. The former faculty has been long defined by
8995logicians as the faculty of mediate conclusion in contradistinction to
8996immediate conclusions (consequentiae immediatae); but the nature of the
8997latter, which itself generates conceptions, is not to be understood
8998from this definition. Now as a division of reason into a logical and
8999a transcendental faculty presents itself here, it becomes necessary to
9000seek for a higher conception of this source of cognition which shall
9001comprehend both conceptions. In this we may expect, according to the
9002analogy of the conceptions of the understanding, that the logical
9003conception will give us the key to the transcendental, and that the
9004table of the functions of the former will present us with the clue to
9005the conceptions of reason.
9006
9007In the former part of our transcendental logic, we defined the
9008understanding to be the faculty of rules; reason may be distinguished
9009from understanding as the faculty of principles.
9010
9011The term principle is ambiguous, and commonly signifies merely a
9012cognition that may be employed as a principle, although it is not in
9013itself, and as regards its proper origin, entitled to the distinction.
9014Every general proposition, even if derived from experience by the
9015process of induction, may serve as the major in a syllogism; but it is
9016not for that reason a principle. Mathematical axioms (for example, there
9017can be only one straight line between two points) are general a priori
9018cognitions, and are therefore rightly denominated principles, relatively
9019to the cases which can be subsumed under them. But I cannot for
9020this reason say that I cognize this property of a straight line from
9021principles--I cognize it only in pure intuition.
9022
9023Cognition from principles, then, is that cognition in which I cognize
9024the particular in the general by means of conceptions. Thus every
9025syllogism is a form of the deduction of a cognition from a principle.
9026For the major always gives a conception, through which everything that
9027is subsumed under the condition thereof is cognized according to a
9028principle. Now as every general cognition may serve as the major in a
9029syllogism, and the understanding presents us with such general a
9030priori propositions, they may be termed principles, in respect of their
9031possible use.
9032
9033But if we consider these principles of the pure understanding in
9034relation to their origin, we shall find them to be anything rather
9035than cognitions from conceptions. For they would not even be possible
9036a priori, if we could not rely on the assistance of pure intuition (in
9037mathematics), or on that of the conditions of a possible experience.
9038That everything that happens has a cause, cannot be concluded from the
9039general conception of that which happens; on the contrary the principle
9040of causality instructs us as to the mode of obtaining from that which
9041happens a determinate empirical conception.
9042
9043Synthetical cognitions from conceptions the understanding cannot supply,
9044and they alone are entitled to be called principles. At the same time,
9045all general propositions may be termed comparative principles.
9046
9047It has been a long-cherished wish--that (who knows how late), may one
9048day, be happily accomplished--that the principles of the endless variety
9049of civil laws should be investigated and exposed; for in this way alone
9050can we find the secret of simplifying legislation. But in this case,
9051laws are nothing more than limitations of our freedom upon conditions
9052under which it subsists in perfect harmony with itself; they
9053consequently have for their object that which is completely our own
9054work, and of which we ourselves may be the cause by means of these
9055conceptions. But how objects as things in themselves--how the nature of
9056things is subordinated to principles and is to be determined, according
9057to conceptions, is a question which it seems well nigh impossible to
9058answer. Be this, however, as it may--for on this point our investigation
9059is yet to be made--it is at least manifest from what we have said that
9060cognition from principles is something very different from cognition by
9061means of the understanding, which may indeed precede other cognitions
9062in the form of a principle, but in itself--in so far as it is
9063synthetical--is neither based upon mere thought, nor contains a general
9064proposition drawn from conceptions alone.
9065
9066The understanding may be a faculty for the production of unity of
9067phenomena by virtue of rules; the reason is a faculty for the production
9068of unity of rules (of the understanding) under principles. Reason,
9069therefore, never applies directly to experience, or to any sensuous
9070object; its object is, on the contrary, the understanding, to the
9071manifold cognition of which it gives a unity a priori by means of
9072conceptions--a unity which may be called rational unity, and which is
9073of a nature very different from that of the unity produced by the
9074understanding.
9075
9076The above is the general conception of the faculty of reason, in so
9077far as it has been possible to make it comprehensible in the absence of
9078examples. These will be given in the sequel.
9079
9080
9081
9082B. OF THE LOGICAL USE OF REASON.
9083
9084A distinction is commonly made between that which is immediately
9085cognized and that which is inferred or concluded. That in a figure
9086which is bounded by three straight lines there are three angles, is an
9087immediate cognition; but that these angles are together equal to two
9088right angles, is an inference or conclusion. Now, as we are constantly
9089employing this mode of thought and have thus become quite accustomed to
9090it, we no longer remark the above distinction, and, as in the case of
9091the so-called deceptions of sense, consider as immediately perceived,
9092what has really been inferred. In every reasoning or syllogism, there
9093is a fundamental proposition, afterwards a second drawn from it, and
9094finally the conclusion, which connects the truth in the first with the
9095truth in the second--and that infallibly. If the judgement concluded
9096is so contained in the first proposition that it can be deduced from
9097it without the meditation of a third notion, the conclusion is called
9098immediate (consequentia immediata); I prefer the term conclusion of
9099the understanding. But if, in addition to the fundamental cognition, a
9100second judgement is necessary for the production of the conclusion, it
9101is called a conclusion of the reason. In the proposition: All men are
9102mortal, are contained the propositions: Some men are mortal, Nothing
9103that is not mortal is a man, and these are therefore immediate
9104conclusions from the first. On the other hand, the proposition: all the
9105learned are mortal, is not contained in the main proposition (for the
9106conception of a learned man does not occur in it), and it can be deduced
9107from the main proposition only by means of a mediating judgement.
9108
9109In every syllogism I first cogitate a rule (the major) by means of
9110the understanding. In the next place I subsume a cognition under the
9111condition of the rule (and this is the minor) by means of the judgement.
9112And finally I determine my cognition by means of the predicate of the
9113rule (this is the conclusio), consequently, I determine it a priori
9114by means of the reason. The relations, therefore, which the major
9115proposition, as the rule, represents between a cognition and its
9116condition, constitute the different kinds of syllogisms. These are just
9117threefold--analogously with all judgements, in so far as they differ
9118in the mode of expressing the relation of a cognition in the
9119understanding--namely, categorical, hypothetical, and disjunctive.
9120
9121When as often happens, the conclusion is a judgement which may follow
9122from other given judgements, through which a perfectly different object
9123is cogitated, I endeavour to discover in the understanding whether the
9124assertion in this conclusion does not stand under certain conditions
9125according to a general rule. If I find such a condition, and if the
9126object mentioned in the conclusion can be subsumed under the given
9127condition, then this conclusion follows from a rule which is also valid
9128for other objects of cognition. From this we see that reason endeavours
9129to subject the great variety of the cognitions of the understanding to
9130the smallest possible number of principles (general conditions), and
9131thus to produce in it the highest unity.
9132
9133
9134
9135C. OF THE PURE USE OF REASON.
9136
9137Can we isolate reason, and, if so, is it in this case a peculiar source
9138of conceptions and judgements which spring from it alone, and through
9139which it can be applied to objects; or is it merely a subordinate
9140faculty, whose duty it is to give a certain form to given cognitions--a
9141form which is called logical, and through which the cognitions of the
9142understanding are subordinated to each other, and lower rules to higher
9143(those, to wit, whose condition comprises in its sphere the condition
9144of the others), in so far as this can be done by comparison? This is the
9145question which we have at present to answer. Manifold variety of rules
9146and unity of principles is a requirement of reason, for the purpose of
9147bringing the understanding into complete accordance with itself, just as
9148understanding subjects the manifold content of intuition to conceptions,
9149and thereby introduces connection into it. But this principle prescribes
9150no law to objects, and does not contain any ground of the possibility of
9151cognizing or of determining them as such, but is merely a subjective
9152law for the proper arrangement of the content of the understanding.
9153The purpose of this law is, by a comparison of the conceptions of the
9154understanding, to reduce them to the smallest possible number, although,
9155at the same time, it does not justify us in demanding from objects
9156themselves such a uniformity as might contribute to the convenience and
9157the enlargement of the sphere of the understanding, or in expecting that
9158it will itself thus receive from them objective validity. In one word,
9159the question is: "does reason in itself, that is, does pure reason
9160contain a priori synthetical principles and rules, and what are those
9161principles?"
9162
9163The formal and logical procedure of reason in syllogisms gives
9164us sufficient information in regard to the ground on which the
9165transcendental principle of reason in its pure synthetical cognition
9166will rest.
9167
91681. Reason, as observed in the syllogistic process, is not applicable to
9169intuitions, for the purpose of subjecting them to rules--for this is the
9170province of the understanding with its categories--but to conceptions
9171and judgements. If pure reason does apply to objects and the intuition
9172of them, it does so not immediately, but mediately--through the
9173understanding and its judgements, which have a direct relation to
9174the senses and their intuition, for the purpose of determining their
9175objects. The unity of reason is therefore not the unity of a possible
9176experience, but is essentially different from this unity, which is that
9177of the understanding. That everything which happens has a cause, is not
9178a principle cognized and prescribed by reason. This principle makes the
9179unity of experience possible and borrows nothing from reason, which,
9180without a reference to possible experience, could never have produced by
9181means of mere conceptions any such synthetical unity.
9182
91832. Reason, in its logical use, endeavours to discover the general
9184condition of its judgement (the conclusion), and a syllogism is itself
9185nothing but a judgement by means of the subsumption of its condition
9186under a general rule (the major). Now as this rule may itself be
9187subjected to the same process of reason, and thus the condition of the
9188condition be sought (by means of a prosyllogism) as long as the process
9189can be continued, it is very manifest that the peculiar principle of
9190reason in its logical use is to find for the conditioned cognition of
9191the understanding the unconditioned whereby the unity of the former is
9192completed.
9193
9194But this logical maxim cannot be a principle of pure reason, unless we
9195admit that, if the conditioned is given, the whole series of conditions
9196subordinated to one another--a series which is consequently itself
9197unconditioned--is also given, that is, contained in the object and its
9198connection.
9199
9200But this principle of pure reason is evidently synthetical; for,
9201analytically, the conditioned certainly relates to some condition, but
9202not to the unconditioned. From this principle also there must originate
9203different synthetical propositions, of which the pure understanding is
9204perfectly ignorant, for it has to do only with objects of a possible
9205experience, the cognition and synthesis of which is always conditioned.
9206The unconditioned, if it does really exist, must be especially
9207considered in regard to the determinations which distinguish it from
9208whatever is conditioned, and will thus afford us material for many a
9209priori synthetical propositions.
9210
9211The principles resulting from this highest principle of pure reason
9212will, however, be transcendent in relation to phenomena, that is to
9213say, it will be impossible to make any adequate empirical use of this
9214principle. It is therefore completely different from all principles of
9215the understanding, the use made of which is entirely immanent, their
9216object and purpose being merely the possibility of experience. Now our
9217duty in the transcendental dialectic is as follows. To discover whether
9218the principle that the series of conditions (in the synthesis of
9219phenomena, or of thought in general) extends to the unconditioned is
9220objectively true, or not; what consequences result therefrom affecting
9221the empirical use of the understanding, or rather whether there exists
9222any such objectively valid proposition of reason, and whether it is not,
9223on the contrary, a merely logical precept which directs us to ascend
9224perpetually to still higher conditions, to approach completeness in the
9225series of them, and thus to introduce into our cognition the highest
9226possible unity of reason. We must ascertain, I say, whether this
9227requirement of reason has not been regarded, by a misunderstanding, as
9228a transcendental principle of pure reason, which postulates a thorough
9229completeness in the series of conditions in objects themselves. We
9230must show, moreover, the misconceptions and illusions that intrude into
9231syllogisms, the major proposition of which pure reason has supplied--a
9232proposition which has perhaps more of the character of a petitio than
9233of a postulatum--and that proceed from experience upwards to its
9234conditions. The solution of these problems is our task in transcendental
9235dialectic, which we are about to expose even at its source, that lies
9236deep in human reason. We shall divide it into two parts, the first of
9237which will treat of the transcendent conceptions of pure reason, the
9238second of transcendent and dialectical syllogisms.
9239
9240
9241
9242BOOK I. -- OF THE CONCEPTIONS OF PURE REASON.
9243
9244The conceptions of pure reason--we do not here speak of the possibility
9245of them--are not obtained by reflection, but by inference or
9246conclusion. The conceptions of understanding are also cogitated a priori
9247antecedently to experience, and render it possible; but they contain
9248nothing but the unity of reflection upon phenomena, in so far as these
9249must necessarily belong to a possible empirical consciousness. Through
9250them alone are cognition and the determination of an object possible. It
9251is from them, accordingly, that we receive material for reasoning, and
9252antecedently to them we possess no a priori conceptions of objects from
9253which they might be deduced, On the other hand, the sole basis of
9254their objective reality consists in the necessity imposed on them, as
9255containing the intellectual form of all experience, of restricting their
9256application and influence to the sphere of experience.
9257
9258But the term, conception of reason, or rational conception, itself
9259indicates that it does not confine itself within the limits of
9260experience, because its object-matter is a cognition, of which every
9261empirical cognition is but a part--nay, the whole of possible experience
9262may be itself but a part of it--a cognition to which no actual
9263experience ever fully attains, although it does always pertain to it.
9264The aim of rational conceptions is the comprehension, as that of the
9265conceptions of understanding is the understanding of perceptions.
9266If they contain the unconditioned, they relate to that to which all
9267experience is subordinate, but which is never itself an object of
9268experience--that towards which reason tends in all its conclusions from
9269experience, and by the standard of which it estimates the degree
9270of their empirical use, but which is never itself an element in an
9271empirical synthesis. If, notwithstanding, such conceptions possess
9272objective validity, they may be called conceptus ratiocinati
9273(conceptions legitimately concluded); in cases where they do not,
9274they have been admitted on account of having the appearance of
9275being correctly concluded, and may be called conceptus ratiocinantes
9276(sophistical conceptions). But as this can only be sufficiently
9277demonstrated in that part of our treatise which relates to the
9278dialectical conclusions of reason, we shall omit any consideration of
9279it in this place. As we called the pure conceptions of the understanding
9280categories, we shall also distinguish those of pure reason by a new name
9281and call them transcendental ideas. These terms, however, we must in the
9282first place explain and justify.
9283
9284
9285
9286SECTION I--Of Ideas in General.
9287
9288Despite the great wealth of words which European languages possess, the
9289thinker finds himself often at a loss for an expression exactly suited
9290to his conception, for want of which he is unable to make himself
9291intelligible either to others or to himself. To coin new words is a
9292pretension to legislation in language which is seldom successful; and,
9293before recourse is taken to so desperate an expedient, it is advisable
9294to examine the dead and learned languages, with the hope and the
9295probability that we may there meet with some adequate expression of the
9296notion we have in our minds. In this case, even if the original meaning
9297of the word has become somewhat uncertain, from carelessness or want of
9298caution on the part of the authors of it, it is always better to adhere
9299to and confirm its proper meaning--even although it may be doubtful
9300whether it was formerly used in exactly this sense--than to make our
9301labour vain by want of sufficient care to render ourselves intelligible.
9302
9303For this reason, when it happens that there exists only a single word to
9304express a certain conception, and this word, in its usual acceptation,
9305is thoroughly adequate to the conception, the accurate distinction of
9306which from related conceptions is of great importance, we ought not to
9307employ the expression improvidently, or, for the sake of variety and
9308elegance of style, use it as a synonym for other cognate words. It
9309is our duty, on the contrary, carefully to preserve its peculiar
9310signification, as otherwise it easily happens that when the attention of
9311the reader is no longer particularly attracted to the expression, and it
9312is lost amid the multitude of other words of very different import, the
9313thought which it conveyed, and which it alone conveyed, is lost with it.
9314
9315Plato employed the expression idea in a way that plainly showed he meant
9316by it something which is never derived from the senses, but which
9317far transcends even the conceptions of the understanding (with which
9318Aristotle occupied himself), inasmuch as in experience nothing perfectly
9319corresponding to them could be found. Ideas are, according to him,
9320archetypes of things themselves, and not merely keys to possible
9321experiences, like the categories. In his view they flow from the
9322highest reason, by which they have been imparted to human reason, which,
9323however, exists no longer in its original state, but is obliged with
9324great labour to recall by reminiscence--which is called philosophy--the
9325old but now sadly obscured ideas. I will not here enter upon any
9326literary investigation of the sense which this sublime philosopher
9327attached to this expression. I shall content myself with remarking that
9328it is nothing unusual, in common conversation as well as in written
9329works, by comparing the thoughts which an author has delivered upon a
9330subject, to understand him better than he understood himself inasmuch
9331as he may not have sufficiently determined his conception, and thus have
9332sometimes spoken, nay even thought, in opposition to his own opinions.
9333
9334Plato perceived very clearly that our faculty of cognition has the
9335feeling of a much higher vocation than that of merely spelling out
9336phenomena according to synthetical unity, for the purpose of being able
9337to read them as experience, and that our reason naturally raises itself
9338to cognitions far too elevated to admit of the possibility of an
9339object given by experience corresponding to them--cognitions which are
9340nevertheless real, and are not mere phantoms of the brain.
9341
9342This philosopher found his ideas especially in all that is practical,*
9343that is, which rests upon freedom, which in its turn ranks under
9344cognitions that are the peculiar product of reason. He who would derive
9345from experience the conceptions of virtue, who would make (as many
9346have really done) that, which at best can but serve as an imperfectly
9347illustrative example, a model for or the formation of a perfectly
9348adequate idea on the subject, would in fact transform virtue into a
9349nonentity changeable according to time and circumstance and utterly
9350incapable of being employed as a rule. On the contrary, every one is
9351conscious that, when any one is held up to him as a model of virtue, he
9352compares this so-called model with the true original which he possesses
9353in his own mind and values him according to this standard. But this
9354standard is the idea of virtue, in relation to which all possible
9355objects of experience are indeed serviceable as examples--proofs of
9356the practicability in a certain degree of that which the conception of
9357virtue demands--but certainly not as archetypes. That the actions of
9358man will never be in perfect accordance with all the requirements of the
9359pure ideas of reason, does not prove the thought to be chimerical. For
9360only through this idea are all judgements as to moral merit or demerit
9361possible; it consequently lies at the foundation of every approach to
9362moral perfection, however far removed from it the obstacles in human
9363nature--indeterminable as to degree--may keep us.
9364
9365
9366 [*Footnote: He certainly extended the application of his conception
9367to speculative cognitions also, provided they were given pure and
9368completely a priori, nay, even to mathematics, although this science
9369cannot possess an object otherwhere than in Possible experience.
9370I cannot follow him in this, and as little can I follow him in his
9371mystical deduction of these ideas, or in his hypostatization of them;
9372although, in truth, the elevated and exaggerated language which he
9373employed in describing them is quite capable of an interpretation more
9374subdued and more in accordance with fact and the nature of things.]
9375
9376
9377The Platonic Republic has become proverbial as an example--and a
9378striking one--of imaginary perfection, such as can exist only in the
9379brain of the idle thinker; and Brucker ridicules the philosopher
9380for maintaining that a prince can never govern well, unless he is
9381participant in the ideas. But we should do better to follow up this
9382thought and, where this admirable thinker leaves us without assistance,
9383employ new efforts to place it in clearer light, rather than carelessly
9384fling it aside as useless, under the very miserable and pernicious
9385pretext of impracticability. A constitution of the greatest possible
9386human freedom according to laws, by which the liberty of every
9387individual can consist with the liberty of every other (not of the
9388greatest possible happiness, for this follows necessarily from the
9389former), is, to say the least, a necessary idea, which must be placed
9390at the foundation not only of the first plan of the constitution of a
9391state, but of all its laws. And, in this, it not necessary at the outset
9392to take account of the obstacles which lie in our way--obstacles which
9393perhaps do not necessarily arise from the character of human nature, but
9394rather from the previous neglect of true ideas in legislation. For there
9395is nothing more pernicious and more unworthy of a philosopher, than the
9396vulgar appeal to a so-called adverse experience, which indeed would not
9397have existed, if those institutions had been established at the proper
9398time and in accordance with ideas; while, instead of this, conceptions,
9399crude for the very reason that they have been drawn from experience,
9400have marred and frustrated all our better views and intentions. The more
9401legislation and government are in harmony with this idea, the more rare
9402do punishments become and thus it is quite reasonable to maintain,
9403as Plato did, that in a perfect state no punishments at all would be
9404necessary. Now although a perfect state may never exist, the idea is
9405not on that account the less just, which holds up this maximum as the
9406archetype or standard of a constitution, in order to bring legislative
9407government always nearer and nearer to the greatest possible perfection.
9408For at what precise degree human nature must stop in its progress, and
9409how wide must be the chasm which must necessarily exist between the
9410idea and its realization, are problems which no one can or ought to
9411determine--and for this reason, that it is the destination of freedom to
9412overstep all assigned limits between itself and the idea.
9413
9414But not only in that wherein human reason is a real causal agent and
9415where ideas are operative causes (of actions and their objects), that is
9416to say, in the region of ethics, but also in regard to nature herself,
9417Plato saw clear proofs of an origin from ideas. A plant, and animal,
9418the regular order of nature--probably also the disposition of the whole
9419universe--give manifest evidence that they are possible only by means
9420of and according to ideas; that, indeed, no one creature, under the
9421individual conditions of its existence, perfectly harmonizes with the
9422idea of the most perfect of its kind--just as little as man with
9423the idea of humanity, which nevertheless he bears in his soul as the
9424archetypal standard of his actions; that, notwithstanding, these ideas
9425are in the highest sense individually, unchangeably, and completely
9426determined, and are the original causes of things; and that the totality
9427of connected objects in the universe is alone fully adequate to that
9428idea. Setting aside the exaggerations of expression in the writings of
9429this philosopher, the mental power exhibited in this ascent from the
9430ectypal mode of regarding the physical world to the architectonic
9431connection thereof according to ends, that is, ideas, is an effort which
9432deserves imitation and claims respect. But as regards the principles of
9433ethics, of legislation, and of religion, spheres in which ideas
9434alone render experience possible, although they never attain to full
9435expression therein, he has vindicated for himself a position of peculiar
9436merit, which is not appreciated only because it is judged by the very
9437empirical rules, the validity of which as principles is destroyed by
9438ideas. For as regards nature, experience presents us with rules and is
9439the source of truth, but in relation to ethical laws experience is the
9440parent of illusion, and it is in the highest degree reprehensible to
9441limit or to deduce the laws which dictate what I ought to do, from what
9442is done.
9443
9444We must, however, omit the consideration of these important subjects,
9445the development of which is in reality the peculiar duty and dignity of
9446philosophy, and confine ourselves for the present to the more humble but
9447not less useful task of preparing a firm foundation for those majestic
9448edifices of moral science. For this foundation has been hitherto
9449insecure from the many subterranean passages which reason in its
9450confident but vain search for treasures has made in all directions.
9451Our present duty is to make ourselves perfectly acquainted with the
9452transcendental use made of pure reason, its principles and ideas, that
9453we may be able properly to determine and value its influence and real
9454worth. But before bringing these introductory remarks to a close, I
9455beg those who really have philosophy at heart--and their number is but
9456small--if they shall find themselves convinced by the considerations
9457following as well as by those above, to exert themselves to preserve to
9458the expression idea its original signification, and to take care that
9459it be not lost among those other expressions by which all sorts of
9460representations are loosely designated--that the interests of science
9461may not thereby suffer. We are in no want of words to denominate
9462adequately every mode of representation, without the necessity of
9463encroaching upon terms which are proper to others. The following is
9464a graduated list of them. The genus is representation in general
9465(representatio). Under it stands representation with consciousness
9466(perceptio). A perception which relates solely to the subject as a
9467modification of its state, is a sensation (sensatio), an objective
9468perception is a cognition (cognitio). A cognition is either an intuition
9469or a conception (intuitus vel conceptus). The former has an immediate
9470relation to the object and is singular and individual; the latter has
9471but a mediate relation, by means of a characteristic mark which may be
9472common to several things. A conception is either empirical or pure. A
9473pure conception, in so far as it has its origin in the understanding
9474alone, and is not the conception of a pure sensuous image, is
9475called notio. A conception formed from notions, which transcends the
9476possibility of experience, is an idea, or a conception of reason. To
9477one who has accustomed himself to these distinctions, it must be quite
9478intolerable to hear the representation of the colour red called an idea.
9479It ought not even to be called a notion or conception of understanding.
9480
9481
9482
9483SECTION II. Of Transcendental Ideas.
9484
9485Transcendental analytic showed us how the mere logical form of
9486our cognition can contain the origin of pure conceptions a priori,
9487conceptions which represent objects antecedently to all experience, or
9488rather, indicate the synthetical unity which alone renders possible an
9489empirical cognition of objects. The form of judgements--converted into a
9490conception of the synthesis of intuitions--produced the categories
9491which direct the employment of the understanding in experience. This
9492consideration warrants us to expect that the form of syllogisms, when
9493applied to synthetical unity of intuitions, following the rule of the
9494categories, will contain the origin of particular a priori conceptions,
9495which we may call pure conceptions of reason or transcendental ideas,
9496and which will determine the use of the understanding in the totality of
9497experience according to principles.
9498
9499The function of reason in arguments consists in the universality of
9500a cognition according to conceptions, and the syllogism itself is
9501a judgement which is determined a priori in the whole extent of its
9502condition. The proposition: "Caius is mortal," is one which may be
9503obtained from experience by the aid of the understanding alone; but my
9504wish is to find a conception which contains the condition under which
9505the predicate of this judgement is given--in this case, the conception
9506of man--and after subsuming under this condition, taken in its whole
9507extent (all men are mortal), I determine according to it the cognition
9508of the object thought, and say: "Caius is mortal."
9509
9510Hence, in the conclusion of a syllogism we restrict a predicate to a
9511certain object, after having thought it in the major in its whole extent
9512under a certain condition. This complete quantity of the extent in
9513relation to such a condition is called universality (universalitas). To
9514this corresponds totality (universitas) of conditions in the synthesis
9515of intuitions. The transcendental conception of reason is therefore
9516nothing else than the conception of the totality of the conditions of
9517a given conditioned. Now as the unconditioned alone renders possible
9518totality of conditions, and, conversely, the totality of conditions is
9519itself always unconditioned; a pure rational conception in general
9520can be defined and explained by means of the conception of the
9521unconditioned, in so far as it contains a basis for the synthesis of the
9522conditioned.
9523
9524To the number of modes of relation which the understanding cogitates by
9525means of the categories, the number of pure rational conceptions will
9526correspond. We must therefore seek for, first, an unconditioned of
9527the categorical synthesis in a subject; secondly, of the hypothetical
9528synthesis of the members of a series; thirdly, of the disjunctive
9529synthesis of parts in a system.
9530
9531There are exactly the same number of modes of syllogisms, each of which
9532proceeds through prosyllogisms to the unconditioned--one to the subject
9533which cannot be employed as predicate, another to the presupposition
9534which supposes nothing higher than itself, and the third to an aggregate
9535of the members of the complete division of a conception. Hence the pure
9536rational conceptions of totality in the synthesis of conditions have a
9537necessary foundation in the nature of human reason--at least as modes of
9538elevating the unity of the understanding to the unconditioned. They
9539may have no valid application, corresponding to their transcendental
9540employment, in concreto, and be thus of no greater utility than
9541to direct the understanding how, while extending them as widely
9542as possible, to maintain its exercise and application in perfect
9543consistence and harmony.
9544
9545But, while speaking here of the totality of conditions and of the
9546unconditioned as the common title of all conceptions of reason, we again
9547light upon an expression which we find it impossible to dispense with,
9548and which nevertheless, owing to the ambiguity attaching to it from long
9549abuse, we cannot employ with safety. The word absolute is one of the few
9550words which, in its original signification, was perfectly adequate to
9551the conception it was intended to convey--a conception which no other
9552word in the same language exactly suits, and the loss--or, which is
9553the same thing, the incautious and loose employment--of which must
9554be followed by the loss of the conception itself. And, as it is a
9555conception which occupies much of the attention of reason, its loss
9556would be greatly to the detriment of all transcendental philosophy. The
9557word absolute is at present frequently used to denote that something can
9558be predicated of a thing considered in itself and intrinsically. In this
9559sense absolutely possible would signify that which is possible in itself
9560(interne)--which is, in fact, the least that one can predicate of an
9561object. On the other hand, it is sometimes employed to indicate that
9562a thing is valid in all respects--for example, absolute sovereignty.
9563Absolutely possible would in this sense signify that which is possible
9564in all relations and in every respect; and this is the most that can be
9565predicated of the possibility of a thing. Now these significations do
9566in truth frequently coincide. Thus, for example, that which is
9567intrinsically impossible, is also impossible in all relations, that is,
9568absolutely impossible. But in most cases they differ from each other
9569toto caelo, and I can by no means conclude that, because a thing is in
9570itself possible, it is also possible in all relations, and therefore
9571absolutely. Nay, more, I shall in the sequel show that absolute
9572necessity does not by any means depend on internal necessity, and
9573that, therefore, it must not be considered as synonymous with it. Of an
9574opposite which is intrinsically impossible, we may affirm that it is in
9575all respects impossible, and that, consequently, the thing itself, of
9576which this is the opposite, is absolutely necessary; but I cannot reason
9577conversely and say, the opposite of that which is absolutely necessary
9578is intrinsically impossible, that is, that the absolute necessity of
9579things is an internal necessity. For this internal necessity is in
9580certain cases a mere empty word with which the least conception cannot
9581be connected, while the conception of the necessity of a thing in all
9582relations possesses very peculiar determinations. Now as the loss of a
9583conception of great utility in speculative science cannot be a matter of
9584indifference to the philosopher, I trust that the proper determination
9585and careful preservation of the expression on which the conception
9586depends will likewise be not indifferent to him.
9587
9588In this enlarged signification, then, shall I employ the word absolute,
9589in opposition to that which is valid only in some particular respect;
9590for the latter is restricted by conditions, the former is valid without
9591any restriction whatever.
9592
9593Now the transcendental conception of reason has for its object nothing
9594else than absolute totality in the synthesis of conditions and does not
9595rest satisfied till it has attained to the absolutely, that is, in all
9596respects and relations, unconditioned. For pure reason leaves to the
9597understanding everything that immediately relates to the object of
9598intuition or rather to their synthesis in imagination. The former
9599restricts itself to the absolute totality in the employment of
9600the conceptions of the understanding and aims at carrying out the
9601synthetical unity which is cogitated in the category, even to the
9602unconditioned. This unity may hence be called the rational unity of
9603phenomena, as the other, which the category expresses, may be termed the
9604unity of the understanding. Reason, therefore, has an immediate relation
9605to the use of the understanding, not indeed in so far as the latter
9606contains the ground of possible experience (for the conception of the
9607absolute totality of conditions is not a conception that can be employed
9608in experience, because no experience is unconditioned), but solely
9609for the purpose of directing it to a certain unity, of which the
9610understanding has no conception, and the aim of which is to collect into
9611an absolute whole all acts of the understanding. Hence the objective
9612employment of the pure conceptions of reason is always transcendent,
9613while that of the pure conceptions of the understanding must, according
9614to their nature, be always immanent, inasmuch as they are limited to
9615possible experience.
9616
9617I understand by idea a necessary conception of reason, to which
9618no corresponding object can be discovered in the world of sense.
9619Accordingly, the pure conceptions of reason at present under
9620consideration are transcendental ideas. They are conceptions of pure
9621reason, for they regard all empirical cognition as determined by means
9622of an absolute totality of conditions. They are not mere fictions, but
9623natural and necessary products of reason, and have hence a necessary
9624relation to the whole sphere of the exercise of the understanding.
9625And, finally, they are transcendent, and overstep the limits of all
9626experiences, in which, consequently, no object can ever be presented
9627that would be perfectly adequate to a transcendental idea. When we use
9628the word idea, we say, as regards its object (an object of the pure
9629understanding), a great deal, but as regards its subject (that is, in
9630respect of its reality under conditions of experience), exceedingly
9631little, because the idea, as the conception of a maximum, can never be
9632completely and adequately presented in concreto. Now, as in the merely
9633speculative employment of reason the latter is properly the sole aim,
9634and as in this case the approximation to a conception, which is never
9635attained in practice, is the same thing as if the conception were
9636non-existent--it is commonly said of the conception of this kind, "it is
9637only an idea." So we might very well say, "the absolute totality of all
9638phenomena is only an idea," for, as we never can present an adequate
9639representation of it, it remains for us a problem incapable of solution.
9640On the other hand, as in the practical use of the understanding we have
9641only to do with action and practice according to rules, an idea of pure
9642reason can always be given really in concreto, although only partially,
9643nay, it is the indispensable condition of all practical employment of
9644reason. The practice or execution of the idea is always limited
9645and defective, but nevertheless within indeterminable boundaries,
9646consequently always under the influence of the conception of an absolute
9647perfection. And thus the practical idea is always in the highest degree
9648fruitful, and in relation to real actions indispensably necessary.
9649In the idea, pure reason possesses even causality and the power of
9650producing that which its conception contains. Hence we cannot say of
9651wisdom, in a disparaging way, "it is only an idea." For, for the very
9652reason that it is the idea of the necessary unity of all possible aims,
9653it must be for all practical exertions and endeavours the primitive
9654condition and rule--a rule which, if not constitutive, is at least
9655limitative.
9656
9657Now, although we must say of the transcendental conceptions of reason,
9658"they are only ideas," we must not, on this account, look upon them as
9659superfluous and nugatory. For, although no object can be determined by
9660them, they can be of great utility, unobserved and at the basis of
9661the edifice of the understanding, as the canon for its extended and
9662self-consistent exercise--a canon which, indeed, does not enable it to
9663cognize more in an object than it would cognize by the help of its own
9664conceptions, but which guides it more securely in its cognition. Not
9665to mention that they perhaps render possible a transition from our
9666conceptions of nature and the non-ego to the practical conceptions, and
9667thus produce for even ethical ideas keeping, so to speak, and connection
9668with the speculative cognitions of reason. The explication of all this
9669must be looked for in the sequel.
9670
9671But setting aside, in conformity with our original purpose, the
9672consideration of the practical ideas, we proceed to contemplate reason
9673in its speculative use alone, nay, in a still more restricted sphere, to
9674wit, in the transcendental use; and here must strike into the same path
9675which we followed in our deduction of the categories. That is to say, we
9676shall consider the logical form of the cognition of reason, that we
9677may see whether reason may not be thereby a source of conceptions which
9678enables us to regard objects in themselves as determined synthetically a
9679priori, in relation to one or other of the functions of reason.
9680
9681Reason, considered as the faculty of a certain logical form of
9682cognition, is the faculty of conclusion, that is, of mediate
9683judgement--by means of the subsumption of the condition of a possible
9684judgement under the condition of a given judgement. The given judgement
9685is the general rule (major). The subsumption of the condition of another
9686possible judgement under the condition of the rule is the minor. The
9687actual judgement, which enounces the assertion of the rule in the
9688subsumed case, is the conclusion (conclusio). The rule predicates
9689something generally under a certain condition. The condition of the rule
9690is satisfied in some particular case. It follows that what was valid
9691in general under that condition must also be considered as valid in the
9692particular case which satisfies this condition. It is very plain that
9693reason attains to a cognition, by means of acts of the understanding
9694which constitute a series of conditions. When I arrive at the
9695proposition, "All bodies are changeable," by beginning with the more
9696remote cognition (in which the conception of body does not appear, but
9697which nevertheless contains the condition of that conception), "All
9698compound is changeable," by proceeding from this to a less remote
9699cognition, which stands under the condition of the former, "Bodies are
9700compound," and hence to a third, which at length connects for me the
9701remote cognition (changeable) with the one before me, "Consequently,
9702bodies are changeable"--I have arrived at a cognition (conclusion)
9703through a series of conditions (premisses). Now every series, whose
9704exponent (of the categorical or hypothetical judgement) is given, can be
9705continued; consequently the same procedure of reason conducts us to the
9706ratiocinatio polysyllogistica, which is a series of syllogisms, that can
9707be continued either on the side of the conditions (per prosyllogismos)
9708or of the conditioned (per episyllogismos) to an indefinite extent.
9709
9710But we very soon perceive that the chain or series of prosyllogisms,
9711that is, of deduced cognitions on the side of the grounds or conditions
9712of a given cognition, in other words, the ascending series of syllogisms
9713must have a very different relation to the faculty of reason from that
9714of the descending series, that is, the progressive procedure of reason
9715on the side of the conditioned by means of episyllogisms. For, as in
9716the former case the cognition (conclusio) is given only as conditioned,
9717reason can attain to this cognition only under the presupposition that
9718all the members of the series on the side of the conditions are
9719given (totality in the series of premisses), because only under this
9720supposition is the judgement we may be considering possible a priori;
9721while on the side of the conditioned or the inferences, only an
9722incomplete and becoming, and not a presupposed or given series,
9723consequently only a potential progression, is cogitated. Hence, when
9724a cognition is contemplated as conditioned, reason is compelled to
9725consider the series of conditions in an ascending line as completed and
9726given in their totality. But if the very same condition is considered
9727at the same time as the condition of other cognitions, which together
9728constitute a series of inferences or consequences in a descending
9729line, reason may preserve a perfect indifference, as to how far this
9730progression may extend a parte posteriori, and whether the totality of
9731this series is possible, because it stands in no need of such a series
9732for the purpose of arriving at the conclusion before it, inasmuch as
9733this conclusion is sufficiently guaranteed and determined on grounds a
9734parte priori. It may be the case, that upon the side of the conditions
9735the series of premisses has a first or highest condition, or it may
9736not possess this, and so be a parte priori unlimited; but it must,
9737nevertheless, contain totality of conditions, even admitting that we
9738never could succeed in completely apprehending it; and the whole series
9739must be unconditionally true, if the conditioned, which is considered
9740as an inference resulting from it, is to be held as true. This is a
9741requirement of reason, which announces its cognition as determined a
9742priori and as necessary, either in itself--and in this case it needs no
9743grounds to rest upon--or, if it is deduced, as a member of a series of
9744grounds, which is itself unconditionally true.
9745
9746
9747
9748SECTION III. System of Transcendental Ideas.
9749
9750We are not at present engaged with a logical dialectic, which makes
9751complete abstraction of the content of cognition and aims only at
9752unveiling the illusory appearance in the form of syllogisms. Our subject
9753is transcendental dialectic, which must contain, completely a priori,
9754the origin of certain cognitions drawn from pure reason, and the origin
9755of certain deduced conceptions, the object of which cannot be given
9756empirically and which therefore lie beyond the sphere of the faculty
9757of understanding. We have observed, from the natural relation which
9758the transcendental use of our cognition, in syllogisms as well as in
9759judgements, must have to the logical, that there are three kinds of
9760dialectical arguments, corresponding to the three modes of conclusion,
9761by which reason attains to cognitions on principles; and that in all
9762it is the business of reason to ascend from the conditioned synthesis,
9763beyond which the understanding never proceeds, to the unconditioned
9764which the understanding never can reach.
9765
9766Now the most general relations which can exist in our representations
9767are: 1st, the relation to the subject; 2nd, the relation to objects,
9768either as phenomena, or as objects of thought in general. If we connect
9769this subdivision with the main division, all the relations of our
9770representations, of which we can form either a conception or an idea,
9771are threefold: 1. The relation to the subject; 2. The relation to the
9772manifold of the object as a phenomenon; 3. The relation to all things in
9773general.
9774
9775Now all pure conceptions have to do in general with the synthetical
9776unity of representations; conceptions of pure reason (transcendental
9777ideas), on the other hand, with the unconditional synthetical unity
9778of all conditions. It follows that all transcendental ideas arrange
9779themselves in three classes, the first of which contains the absolute
9780(unconditioned) unity of the thinking subject, the second the absolute
9781unity of the series of the conditions of a phenomenon, the third the
9782absolute unity of the condition of all objects of thought in general.
9783
9784The thinking subject is the object-matter of Psychology; the sum total
9785of all phenomena (the world) is the object-matter of Cosmology; and the
9786thing which contains the highest condition of the possibility of all
9787that is cogitable (the being of all beings) is the object-matter of all
9788Theology. Thus pure reason presents us with the idea of a transcendental
9789doctrine of the soul (psychologia rationalis), of a transcendental
9790science of the world (cosmologia rationalis), and finally of
9791a transcendental doctrine of God (theologia transcendentalis).
9792Understanding cannot originate even the outline of any of these
9793sciences, even when connected with the highest logical use of reason,
9794that is, all cogitable syllogisms--for the purpose of proceeding from
9795one object (phenomenon) to all others, even to the utmost limits of
9796the empirical synthesis. They are, on the contrary, pure and genuine
9797products, or problems, of pure reason.
9798
9799What modi of the pure conceptions of reason these transcendental ideas
9800are will be fully exposed in the following chapter. They follow
9801the guiding thread of the categories. For pure reason never relates
9802immediately to objects, but to the conceptions of these contained in the
9803understanding. In like manner, it will be made manifest in the detailed
9804explanation of these ideas--how reason, merely through the synthetical
9805use of the same function which it employs in a categorical syllogism,
9806necessarily attains to the conception of the absolute unity of the
9807thinking subject--how the logical procedure in hypothetical ideas
9808necessarily produces the idea of the absolutely unconditioned in a
9809series of given conditions, and finally--how the mere form of the
9810disjunctive syllogism involves the highest conception of a being of
9811all beings: a thought which at first sight seems in the highest degree
9812paradoxical.
9813
9814An objective deduction, such as we were able to present in the case of
9815the categories, is impossible as regards these transcendental ideas. For
9816they have, in truth, no relation to any object, in experience, for the
9817very reason that they are only ideas. But a subjective deduction of them
9818from the nature of our reason is possible, and has been given in the
9819present chapter.
9820
9821It is easy to perceive that the sole aim of pure reason is the absolute
9822totality of the synthesis on the side of the conditions, and that it
9823does not concern itself with the absolute completeness on the Part of
9824the conditioned. For of the former alone does she stand in need, in
9825order to preposit the whole series of conditions, and thus present them
9826to the understanding a priori. But if we once have a completely (and
9827unconditionally) given condition, there is no further necessity,
9828in proceeding with the series, for a conception of reason; for the
9829understanding takes of itself every step downward, from the condition
9830to the conditioned. Thus the transcendental ideas are available only for
9831ascending in the series of conditions, till we reach the unconditioned,
9832that is, principles. As regards descending to the conditioned, on the
9833other hand, we find that there is a widely extensive logical use which
9834reason makes of the laws of the understanding, but that a transcendental
9835use thereof is impossible; and that when we form an idea of the absolute
9836totality of such a synthesis, for example, of the whole series of
9837all future changes in the world, this idea is a mere ens rationis, an
9838arbitrary fiction of thought, and not a necessary presupposition of
9839reason. For the possibility of the conditioned presupposes the totality
9840of its conditions, but not of its consequences. Consequently, this
9841conception is not a transcendental idea--and it is with these alone that
9842we are at present occupied.
9843
9844Finally, it is obvious that there exists among the transcendental ideas
9845a certain connection and unity, and that pure reason, by means of them,
9846collects all its cognitions into one system. From the cognition of self
9847to the cognition of the world, and through these to the supreme being,
9848the progression is so natural, that it seems to resemble the logical
9849march of reason from the premisses to the conclusion.* Now whether there
9850lies unobserved at the foundation of these ideas an analogy of the
9851same kind as exists between the logical and transcendental procedure of
9852reason, is another of those questions, the answer to which we must not
9853expect till we arrive at a more advanced stage in our inquiries. In this
9854cursory and preliminary view, we have, meanwhile, reached our aim. For
9855we have dispelled the ambiguity which attached to the transcendental
9856conceptions of reason, from their being commonly mixed up with
9857other conceptions in the systems of philosophers, and not properly
9858distinguished from the conceptions of the understanding; we have exposed
9859their origin and, thereby, at the same time their determinate number,
9860and presented them in a systematic connection, and have thus marked out
9861and enclosed a definite sphere for pure reason.
9862
9863
9864
9865 [*Footnote: The science of Metaphysics has for the proper object of its
9866inquiries only three grand ideas: GOD, FREEDOM, and IMMORTALITY, and it
9867aims at showing, that the second conception, conjoined with the first,
9868must lead to the third, as a necessary conclusion. All the other
9869subjects with which it occupies itself, are merely means for the
9870attainment and realization of these ideas. It does not require these
9871ideas for the construction of a science of nature, but, on the contrary,
9872for the purpose of passing beyond the sphere of nature. A complete
9873insight into and comprehension of them would render Theology, Ethics,
9874and, through the conjunction of both, Religion, solely dependent on the
9875speculative faculty of reason. In a systematic representation of these
9876ideas the above-mentioned arrangement--the synthetical one--would be the
9877most suitable; but in the investigation which must necessarily precede
9878it, the analytical, which reverses this arrangement, would be better
9879adapted to our purpose, as in it we should proceed from that which
9880experience immediately presents to us--psychology, to cosmology, and
9881thence to theology.]
9882
9883
9884
9885BOOK II.-- OF THE DIALECTICAL PROCEDURE OF PURE REASON.
9886
9887It may be said that the object of a merely transcendental idea is
9888something of which we have no conception, although the idea may be a
9889necessary product of reason according to its original laws. For, in
9890fact, a conception of an object that is adequate to the idea given
9891by reason, is impossible. For such an object must be capable of being
9892presented and intuited in a Possible experience. But we should express
9893our meaning better, and with less risk of being misunderstood, if
9894we said that we can have no knowledge of an object, which perfectly
9895corresponds to an idea, although we may possess a problematical
9896conception thereof.
9897
9898Now the transcendental (subjective) reality at least of the pure
9899conceptions of reason rests upon the fact that we are led to such ideas
9900by a necessary procedure of reason. There must therefore be syllogisms
9901which contain no empirical premisses, and by means of which we conclude
9902from something that we do know, to something of which we do not even
9903possess a conception, to which we, nevertheless, by an unavoidable
9904illusion, ascribe objective reality. Such arguments are, as regards
9905their result, rather to be termed sophisms than syllogisms, although
9906indeed, as regards their origin, they are very well entitled to the
9907latter name, inasmuch as they are not fictions or accidental products of
9908reason, but are necessitated by its very nature. They are sophisms, not
9909of men, but of pure reason herself, from which the Wisest cannot free
9910himself. After long labour he may be able to guard against the error,
9911but he can never be thoroughly rid of the illusion which continually
9912mocks and misleads him.
9913
9914Of these dialectical arguments there are three kinds, corresponding to
9915the number of the ideas which their conclusions present. In the argument
9916or syllogism of the first class, I conclude, from the transcendental
9917conception of the subject contains no manifold, the absolute unity
9918of the subject itself, of which I cannot in this manner attain to a
9919conception. This dialectical argument I shall call the transcendental
9920paralogism. The second class of sophistical arguments is occupied with
9921the transcendental conception of the absolute totality of the series of
9922conditions for a given phenomenon, and I conclude, from the fact that
9923I have always a self-contradictory conception of the unconditioned
9924synthetical unity of the series upon one side, the truth of the opposite
9925unity, of which I have nevertheless no conception. The condition of
9926reason in these dialectical arguments, I shall term the antinomy of pure
9927reason. Finally, according to the third kind of sophistical argument,
9928I conclude, from the totality of the conditions of thinking objects in
9929general, in so far as they can be given, the absolute synthetical unity
9930of all conditions of the possibility of things in general; that is, from
9931things which I do not know in their mere transcendental conception, I
9932conclude a being of all beings which I know still less by means of a
9933transcendental conception, and of whose unconditioned necessity I can
9934form no conception whatever. This dialectical argument I shall call the
9935ideal of pure reason.
9936
9937
9938
9939CHAPTER I. Of the Paralogisms of Pure Reason.
9940
9941The logical paralogism consists in the falsity of an argument in respect
9942of its form, be the content what it may. But a transcendental paralogism
9943has a transcendental foundation, and concludes falsely, while the form
9944is correct and unexceptionable. In this manner the paralogism has
9945its foundation in the nature of human reason, and is the parent of an
9946unavoidable, though not insoluble, mental illusion.
9947
9948We now come to a conception which was not inserted in the general list
9949of transcendental conceptions, and yet must be reckoned with them,
9950but at the same time without in the least altering, or indicating a
9951deficiency in that table. This is the conception, or, if the term is
9952preferred, the judgement, "I think." But it is readily perceived that
9953this thought is as it were the vehicle of all conceptions in general,
9954and consequently of transcendental conceptions also, and that it is
9955therefore regarded as a transcendental conception, although it can
9956have no peculiar claim to be so ranked, inasmuch as its only use is to
9957indicate that all thought is accompanied by consciousness. At the same
9958time, pure as this conception is from empirical content (impressions
9959of the senses), it enables us to distinguish two different kinds of
9960objects. "I," as thinking, am an object of the internal sense, and am
9961called soul. That which is an object of the external senses is called
9962body. Thus the expression, "I," as a thinking being, designates the
9963object-matter of psychology, which may be called "the rational doctrine
9964of the soul," inasmuch as in this science I desire to know nothing of
9965the soul but what, independently of all experience (which determines me
9966in concreto), may be concluded from this conception "I," in so far as it
9967appears in all thought.
9968
9969Now, the rational doctrine of the soul is really an undertaking of
9970this kind. For if the smallest empirical element of thought, if any
9971particular perception of my internal state, were to be introduced among
9972the grounds of cognition of this science, it would not be a rational,
9973but an empirical doctrine of the soul. We have thus before us a
9974pretended science, raised upon the single proposition, "I think," whose
9975foundation or want of foundation we may very properly, and agreeably
9976with the nature of a transcendental philosophy, here examine. It
9977ought not to be objected that in this proposition, which expresses the
9978perception of one's self, an internal experience is asserted, and that
9979consequently the rational doctrine of the soul which is founded upon it,
9980is not pure, but partly founded upon an empirical principle. For this
9981internal perception is nothing more than the mere apperception, "I
9982think," which in fact renders all transcendental conceptions possible,
9983in which we say, "I think substance, cause, etc." For internal
9984experience in general and its possibility, or perception in general, and
9985its relation to other perceptions, unless some particular distinction
9986or determination thereof is empirically given, cannot be regarded as
9987empirical cognition, but as cognition of the empirical, and belongs
9988to the investigation of the possibility of every experience, which
9989is certainly transcendental. The smallest object of experience (for
9990example, only pleasure or pain), that should be included in the general
9991representation of self-consciousness, would immediately change the
9992rational into an empirical psychology.
9993
9994"I think" is therefore the only text of rational psychology, from which
9995it must develop its whole system. It is manifest that this thought, when
9996applied to an object (myself), can contain nothing but transcendental
9997predicates thereof; because the least empirical predicate would destroy
9998the purity of the science and its independence of all experience.
9999
10000But we shall have to follow here the guidance of the categories--only,
10001as in the present case a thing, "I," as thinking being, is at first
10002given, we shall--not indeed change the order of the categories as it
10003stands in the table--but begin at the category of substance, by which at
10004the a thing in itself is represented and proceeds backwards through
10005the series. The topic of the rational doctrine of the soul, from which
10006everything else it may contain must be deduced, is accordingly as
10007follows:
10008
10009
10010 1 2
10011 The Soul is SUBSTANCE As regards its quality
10012 it is SIMPLE
10013
10014 3
10015 As regards the different
10016 times in which it exists,
10017 it is numerically identical,
10018 that is UNITY, not Plurality.
10019
10020 4
10021 It is in relation to possible objects in space*
10022
10023
10024
10025 [*Footnote: The reader, who may not so easily perceive the psychological
10026sense of these expressions, taken here in their transcendental
10027abstraction, and cannot guess why the latter attribute of the soul
10028belongs to the category of existence, will find the expressions
10029sufficiently explained and justified in the sequel. I have, moreover, to
10030apologize for the Latin terms which have been employed, instead of their
10031German synonyms, contrary to the rules of correct writing. But I judged
10032it better to sacrifice elegance to perspicuity.]
10033
10034
10035From these elements originate all the conceptions of pure psychology,
10036by combination alone, without the aid of any other principle. This
10037substance, merely as an object of the internal sense, gives
10038the conception of Immateriality; as simple substance, that of
10039Incorruptibility; its identity, as intellectual substance, gives the
10040conception of Personality; all these three together, Spirituality.
10041Its relation to objects in space gives us the conception of connection
10042(commercium) with bodies. Thus it represents thinking substance as the
10043principle of life in matter, that is, as a soul (anima), and as the
10044ground of Animality; and this, limited and determined by the conception
10045of spirituality, gives us that of Immortality.
10046
10047Now to these conceptions relate four paralogisms of a transcendental
10048psychology, which is falsely held to be a science of pure reason,
10049touching the nature of our thinking being. We can, however, lay at
10050the foundation of this science nothing but the simple and in itself
10051perfectly contentless representation "I" which cannot even be called
10052a conception, but merely a consciousness which accompanies all
10053conceptions. By this "I," or "He," or "It," who or which thinks, nothing
10054more is represented than a transcendental subject of thought = x, which
10055is cognized only by means of the thoughts that are its predicates, and
10056of which, apart from these, we cannot form the least conception. Hence
10057in a perpetual circle, inasmuch as we must always employ it, in order
10058to frame any judgement respecting it. And this inconvenience we find it
10059impossible to rid ourselves of, because consciousness in itself is not
10060so much a representation distinguishing a particular object, as a form
10061of representation in general, in so far as it may be termed cognition;
10062for in and by cognition alone do I think anything.
10063
10064It must, however, appear extraordinary at first sight that the condition
10065under which I think, and which is consequently a property of my subject,
10066should be held to be likewise valid for every existence which thinks,
10067and that we can presume to base upon a seemingly empirical proposition
10068a judgement which is apodeictic and universal, to wit, that everything
10069which thinks is constituted as the voice of my consciousness declares it
10070to be, that is, as a self-conscious being. The cause of this belief is
10071to be found in the fact that we necessarily attribute to things a priori
10072all the properties which constitute conditions under which alone we
10073can cogitate them. Now I cannot obtain the least representation of
10074a thinking being by means of external experience, but solely through
10075self-consciousness. Such objects are consequently nothing more than the
10076transference of this consciousness of mine to other things which can
10077only thus be represented as thinking beings. The proposition, "I think,"
10078is, in the present case, understood in a problematical sense, not in
10079so far as it contains a perception of an existence (like the Cartesian
10080"Cogito, ergo sum"),[Footnote: "I think, therefore I am."] but in regard
10081to its mere possibility--for the purpose of discovering what properties
10082may be inferred from so simple a proposition and predicated of the
10083subject of it.
10084
10085If at the foundation of our pure rational cognition of thinking beings
10086there lay more than the mere Cogito--if we could likewise call in aid
10087observations on the play of our thoughts, and the thence derived natural
10088laws of the thinking self, there would arise an empirical psychology
10089which would be a kind of physiology of the internal sense and might
10090possibly be capable of explaining the phenomena of that sense. But it
10091could never be available for discovering those properties which do not
10092belong to possible experience (such as the quality of simplicity),
10093nor could it make any apodeictic enunciation on the nature of thinking
10094beings: it would therefore not be a rational psychology.
10095
10096Now, as the proposition "I think" (in the problematical sense) contains
10097the form of every judgement in general and is the constant accompaniment
10098of all the categories, it is manifest that conclusions are drawn from
10099it only by a transcendental employment of the understanding. This use of
10100the understanding excludes all empirical elements; and we cannot, as
10101has been shown above, have any favourable conception beforehand of
10102its procedure. We shall therefore follow with a critical eye this
10103proposition through all the predicaments of pure psychology; but we
10104shall, for brevity's sake, allow this examination to proceed in an
10105uninterrupted connection.
10106
10107Before entering on this task, however, the following general remark may
10108help to quicken our attention to this mode of argument. It is not
10109merely through my thinking that I cognize an object, but only through my
10110determining a given intuition in relation to the unity of consciousness
10111in which all thinking consists. It follows that I cognize myself, not
10112through my being conscious of myself as thinking, but only when I am
10113conscious of the intuition of myself as determined in relation to the
10114function of thought. All the modi of self-consciousness in thought
10115are hence not conceptions of objects (conceptions of the
10116understanding--categories); they are mere logical functions, which do
10117not present to thought an object to be cognized, and cannot therefore
10118present my Self as an object. Not the consciousness of the determining,
10119but only that of the determinable self, that is, of my internal
10120intuition (in so far as the manifold contained in it can be connected
10121conformably with the general condition of the unity of apperception in
10122thought), is the object.
10123
101241. In all judgements I am the determining subject of that relation which
10125constitutes a judgement. But that the I which thinks, must be considered
10126as in thought always a subject, and as a thing which cannot be a
10127predicate to thought, is an apodeictic and identical proposition. But
10128this proposition does not signify that I, as an object, am, for
10129myself, a self-subsistent being or substance. This latter
10130statement--an ambitious one--requires to be supported by data which are
10131not to be discovered in thought; and are perhaps (in so far as I
10132consider the thinking self merely as such) not to be discovered in the
10133thinking self at all.
10134
101352. That the I or Ego of apperception, and consequently in all thought,
10136is singular or simple, and cannot be resolved into a plurality of
10137subjects, and therefore indicates a logically simple subject--this is
10138self-evident from the very conception of an Ego, and is consequently an
10139analytical proposition. But this is not tantamount to declaring that
10140the thinking Ego is a simple substance--for this would be a synthetical
10141proposition. The conception of substance always relates to intuitions,
10142which with me cannot be other than sensuous, and which consequently lie
10143completely out of the sphere of the understanding and its thought:
10144but to this sphere belongs the affirmation that the Ego is simple
10145in thought. It would indeed be surprising, if the conception of
10146"substance," which in other cases requires so much labour to distinguish
10147from the other elements presented by intuition--so much trouble, too,
10148to discover whether it can be simple (as in the case of the parts of
10149matter)--should be presented immediately to me, as if by revelation, in
10150the poorest mental representation of all.
10151
101523. The proposition of the identity of my Self amidst all the manifold
10153representations of which I am conscious, is likewise a proposition lying
10154in the conceptions themselves, and is consequently analytical. But
10155this identity of the subject, of which I am conscious in all its
10156representations, does not relate to or concern the intuition of the
10157subject, by which it is given as an object. This proposition cannot
10158therefore enounce the identity of the person, by which is understood the
10159consciousness of the identity of its own substance as a thinking being
10160in all change and variation of circumstances. To prove this, we
10161should require not a mere analysis of the proposition, but synthetical
10162judgements based upon a given intuition.
10163
101644. I distinguish my own existence, as that of a thinking being, from
10165that of other things external to me--among which my body also is
10166reckoned. This is also an analytical proposition, for other things are
10167exactly those which I think as different or distinguished from myself.
10168But whether this consciousness of myself is possible without things
10169external to me; and whether therefore I can exist merely as a thinking
10170being (without being man)--cannot be known or inferred from this
10171proposition.
10172
10173Thus we have gained nothing as regards the cognition of myself as
10174object, by the analysis of the consciousness of my Self in thought. The
10175logical exposition of thought in general is mistaken for a metaphysical
10176determination of the object.
10177
10178Our Critique would be an investigation utterly superfluous, if there
10179existed a possibility of proving a priori, that all thinking beings
10180are in themselves simple substances, as such, therefore, possess
10181the inseparable attribute of personality, and are conscious of their
10182existence apart from and unconnected with matter. For we should thus
10183have taken a step beyond the world of sense, and have penetrated into
10184the sphere of noumena; and in this case the right could not be denied
10185us of extending our knowledge in this sphere, of establishing ourselves,
10186and, under a favouring star, appropriating to ourselves possessions
10187in it. For the proposition: "Every thinking being, as such, is simple
10188substance," is an a priori synthetical proposition; because in the first
10189place it goes beyond the conception which is the subject of it, and adds
10190to the mere notion of a thinking being the mode of its existence, and in
10191the second place annexes a predicate (that of simplicity) to the latter
10192conception--a predicate which it could not have discovered in the sphere
10193of experience. It would follow that a priori synthetical propositions
10194are possible and legitimate, not only, as we have maintained, in
10195relation to objects of possible experience, and as principles of the
10196possibility of this experience itself, but are applicable to things
10197in themselves--an inference which makes an end of the whole of this
10198Critique, and obliges us to fall back on the old mode of metaphysical
10199procedure. But indeed the danger is not so great, if we look a little
10200closer into the question.
10201
10202There lurks in the procedure of rational Psychology a paralogism, which
10203is represented in the following syllogism:
10204
10205That which cannot be cogitated otherwise than as subject, does not exist
10206otherwise than as subject, and is therefore substance.
10207
10208A thinking being, considered merely as such, cannot be cogitated
10209otherwise than as subject.
10210
10211Therefore it exists also as such, that is, as substance.
10212
10213In the major we speak of a being that can be cogitated generally and in
10214every relation, consequently as it may be given in intuition. But in the
10215minor we speak of the same being only in so far as it regards itself as
10216subject, relatively to thought and the unity of consciousness, but
10217not in relation to intuition, by which it is presented as an object to
10218thought. Thus the conclusion is here arrived at by a Sophisma figurae
10219dictionis.*
10220
10221
10222 [*Footnote: Thought is taken in the two premisses in two totally
10223different senses. In the major it is considered as relating and applying
10224to objects in general, consequently to objects of intuition also. In
10225the minor, we understand it as relating merely to self-consciousness. In
10226this sense, we do not cogitate an object, but merely the relation to the
10227self-consciousness of the subject, as the form of thought. In the former
10228premiss we speak of things which cannot be cogitated otherwise than as
10229subjects. In the second, we do not speak of things, but of thought (all
10230objects being abstracted), in which the Ego is always the subject of
10231consciousness. Hence the conclusion cannot be, "I cannot exist otherwise
10232than as subject"; but only "I can, in cogitating my existence, employ
10233my Ego only as the subject of the judgement." But this is an identical
10234proposition, and throws no light on the mode of my existence.]
10235
10236
10237That this famous argument is a mere paralogism, will be plain to any one
10238who will consider the general remark which precedes our exposition of
10239the principles of the pure understanding, and the section on noumena.
10240For it was there proved that the conception of a thing, which can
10241exist per se--only as a subject and never as a predicate, possesses
10242no objective reality; that is to say, we can never know whether there
10243exists any object to correspond to the conception; consequently, the
10244conception is nothing more than a conception, and from it we derive
10245no proper knowledge. If this conception is to indicate by the term
10246substance, an object that can be given, if it is to become a cognition,
10247we must have at the foundation of the cognition a permanent intuition,
10248as the indispensable condition of its objective reality. For through
10249intuition alone can an object be given. But in internal intuition
10250there is nothing permanent, for the Ego is but the consciousness of my
10251thought. If then, we appeal merely to thought, we cannot discover
10252the necessary condition of the application of the conception of
10253substance--that is, of a subject existing per se--to the subject as
10254a thinking being. And thus the conception of the simple nature of
10255substance, which is connected with the objective reality of this
10256conception, is shown to be also invalid, and to be, in fact, nothing
10257more than the logical qualitative unity of self-consciousness in
10258thought; whilst we remain perfectly ignorant whether the subject is
10259composite or not.
10260
10261
10262
10263Refutation of the Argument of Mendelssohn for the Substantiality or
10264Permanence of the Soul.
10265
10266This acute philosopher easily perceived the insufficiency of the common
10267argument which attempts to prove that the soul--it being granted that it
10268is a simple being--cannot perish by dissolution or decomposition; he
10269saw it is not impossible for it to cease to be by extinction, or
10270disappearance. He endeavoured to prove in his Phaedo, that the soul
10271cannot be annihilated, by showing that a simple being cannot cease to
10272exist. Inasmuch as, he said, a simple existence cannot diminish, nor
10273gradually lose portions of its being, and thus be by degrees reduced
10274to nothing (for it possesses no parts, and therefore no multiplicity),
10275between the moment in which it is, and the moment in which it is not,
10276no time can be discovered--which is impossible. But this philosopher
10277did not consider that, granting the soul to possess this simple nature,
10278which contains no parts external to each other and consequently no
10279extensive quantity, we cannot refuse to it any less than to any other
10280being, intensive quantity, that is, a degree of reality in regard to
10281all its faculties, nay, to all that constitutes its existence. But this
10282degree of reality can become less and less through an infinite series
10283of smaller degrees. It follows, therefore, that this supposed
10284substance--this thing, the permanence of which is not assured in any
10285other way, may, if not by decomposition, by gradual loss (remissio)
10286of its powers (consequently by elanguescence, if I may employ this
10287expression), be changed into nothing. For consciousness itself has
10288always a degree, which may be lessened.* Consequently the faculty of
10289being conscious may be diminished; and so with all other faculties. The
10290permanence of the soul, therefore, as an object of the internal sense,
10291remains undemonstrated, nay, even indemonstrable. Its permanence in
10292life is evident, per se, inasmuch as the thinking being (as man) is to
10293itself, at the same time, an object of the external senses. But this
10294does not authorize the rational psychologist to affirm, from mere
10295conceptions, its permanence beyond life.*[2]
10296
10297
10298 [*Footnote: Clearness is not, as logicians maintain, the consciousness
10299of a representation. For a certain degree of consciousness, which may
10300not, however, be sufficient for recollection, is to be met with in many
10301dim representations. For without any consciousness at all, we should not
10302be able to recognize any difference in the obscure representations we
10303connect; as we really can do with many conceptions, such as those
10304of right and justice, and those of the musician, who strikes at once
10305several notes in improvising a piece of music. But a representation is
10306clear, in which our consciousness is sufficient for the consciousness
10307of the difference of this representation from others. If we are only
10308conscious that there is a difference, but are not conscious of the
10309difference--that is, what the difference is--the representation must be
10310termed obscure. There is, consequently, an infinite series of degrees of
10311consciousness down to its entire disappearance.]
10312
10313
10314 [*[2]Footnote: There are some who think they have done enough to
10315establish a new possibility in the mode of the existence of souls, when
10316they have shown that there is no contradiction in their hypotheses on
10317this subject. Such are those who affirm the possibility of thought--of
10318which they have no other knowledge than what they derive from its use in
10319connecting empirical intuitions presented in this our human life--after
10320this life has ceased. But it is very easy to embarrass them by the
10321introduction of counter-possibilities, which rest upon quite as good a
10322foundation. Such, for example, is the possibility of the division of
10323a simple substance into several substances; and conversely, of
10324the coalition of several into one simple substance. For, although
10325divisibility presupposes composition, it does not necessarily require
10326a composition of substances, but only of the degrees (of the several
10327faculties) of one and the same substance. Now we can cogitate all
10328the powers and faculties of the soul--even that of consciousness--as
10329diminished by one half, the substance still remaining. In the same way
10330we can represent to ourselves without contradiction, this obliterated
10331half as preserved, not in the soul, but without it; and we can believe
10332that, as in this case every thing that is real in the soul, and has a
10333degree--consequently its entire existence--has been halved, a particular
10334substance would arise out of the soul. For the multiplicity, which has
10335been divided, formerly existed, but not as a multiplicity of substances,
10336but of every reality as the quantum of existence in it; and the unity of
10337substance was merely a mode of existence, which by this division alone
10338has been transformed into a plurality of subsistence. In the same manner
10339several simple substances might coalesce into one, without anything
10340being lost except the plurality of subsistence, inasmuch as the
10341one substance would contain the degree of reality of all the former
10342substances. Perhaps, indeed, the simple substances, which appear under
10343the form of matter, might (not indeed by a mechanical or chemical
10344influence upon each other, but by an unknown influence, of which the
10345former would be but the phenomenal appearance), by means of such a
10346dynamical division of the parent-souls, as intensive quantities, produce
10347other souls, while the former repaired the loss thus sustained with
10348new matter of the same sort. I am far from allowing any value to such
10349chimeras; and the principles of our analytic have clearly proved that
10350no other than an empirical use of the categories--that of substance,
10351for example--is possible. But if the rationalist is bold enough to
10352construct, on the mere authority of the faculty of thought--without any
10353intuition, whereby an object is given--a self-subsistent being, merely
10354because the unity of apperception in thought cannot allow him to believe
10355it a composite being, instead of declaring, as he ought to do, that he
10356is unable to explain the possibility of a thinking nature; what ought to
10357hinder the materialist, with as complete an independence of experience,
10358to employ the principle of the rationalist in a directly opposite
10359manner--still preserving the formal unity required by his opponent?]
10360
10361
10362If, now, we take the above propositions--as they must be accepted as
10363valid for all thinking beings in the system of rational psychology--in
10364synthetical connection, and proceed, from the category of relation,
10365with the proposition: "All thinking beings are, as such, substances,"
10366backwards through the series, till the circle is completed; we come
10367at last to their existence, of which, in this system of rational
10368psychology, substances are held to be conscious, independently of
10369external things; nay, it is asserted that, in relation to the permanence
10370which is a necessary characteristic of substance, they can of
10371themselves determine external things. It follows that idealism--at least
10372problematical idealism, is perfectly unavoidable in this rationalistic
10373system. And, if the existence of outward things is not held to be
10374requisite to the determination of the existence of a substance in time,
10375the existence of these outward things at all, is a gratuitous assumption
10376which remains without the possibility of a proof.
10377
10378But if we proceed analytically--the "I think" as a proposition
10379containing in itself an existence as given, consequently modality being
10380the principle--and dissect this proposition, in order to ascertain its
10381content, and discover whether and how this Ego determines its existence
10382in time and space without the aid of anything external; the propositions
10383of rationalistic psychology would not begin with the conception of a
10384thinking being, but with a reality, and the properties of a thinking
10385being in general would be deduced from the mode in which this reality is
10386cogitated, after everything empirical had been abstracted; as is shown
10387in the following table:
10388
10389
10390 1
10391 I think,
10392
10393 2 3
10394 as Subject, as simple Subject,
10395
10396 4
10397 as identical Subject,
10398 in every state of my thought.
10399
10400
10401Now, inasmuch as it is not determined in this second proposition,
10402whether I can exist and be cogitated only as subject, and not also as a
10403predicate of another being, the conception of a subject is here taken in
10404a merely logical sense; and it remains undetermined, whether substance
10405is to be cogitated under the conception or not. But in the third
10406proposition, the absolute unity of apperception--the simple Ego in the
10407representation to which all connection and separation, which constitute
10408thought, relate, is of itself important; even although it presents
10409us with no information about the constitution or subsistence of the
10410subject. Apperception is something real, and the simplicity of its
10411nature is given in the very fact of its possibility. Now in space there
10412is nothing real that is at the same time simple; for points, which are
10413the only simple things in space, are merely limits, but not constituent
10414parts of space. From this follows the impossibility of a definition
10415on the basis of materialism of the constitution of my Ego as a merely
10416thinking subject. But, because my existence is considered in the first
10417proposition as given, for it does not mean, "Every thinking being
10418exists" (for this would be predicating of them absolute necessity),
10419but only, "I exist thinking"; the proposition is quite empirical, and
10420contains the determinability of my existence merely in relation to my
10421representations in time. But as I require for this purpose something
10422that is permanent, such as is not given in internal intuition; the
10423mode of my existence, whether as substance or as accident, cannot
10424be determined by means of this simple self-consciousness. Thus,
10425if materialism is inadequate to explain the mode in which I exist,
10426spiritualism is likewise as insufficient; and the conclusion is that we
10427are utterly unable to attain to any knowledge of the constitution of
10428the soul, in so far as relates to the possibility of its existence apart
10429from external objects.
10430
10431And, indeed, how should it be possible, merely by the aid of the unity
10432of consciousness--which we cognize only for the reason that it is
10433indispensable to the possibility of experience--to pass the bounds of
10434experience (our existence in this life); and to extend our cognition
10435to the nature of all thinking beings by means of the empirical--but
10436in relation to every sort of intuition, perfectly undetermined--proposition,
10437"I think"?
10438
10439There does not then exist any rational psychology as a doctrine
10440furnishing any addition to our knowledge of ourselves. It is nothing
10441more than a discipline, which sets impassable limits to speculative
10442reason in this region of thought, to prevent it, on the one hand, from
10443throwing itself into the arms of a soulless materialism, and, on the
10444other, from losing itself in the mazes of a baseless spiritualism.
10445It teaches us to consider this refusal of our reason to give any
10446satisfactory answer to questions which reach beyond the limits of this
10447our human life, as a hint to abandon fruitless speculation; and to
10448direct, to a practical use, our knowledge of ourselves--which, although
10449applicable only to objects of experience, receives its principles from a
10450higher source, and regulates its procedure as if our destiny reached far
10451beyond the boundaries of experience and life.
10452
10453From all this it is evident that rational psychology has its origin in
10454a mere misunderstanding. The unity of consciousness, which lies at the
10455basis of the categories, is considered to be an intuition of the subject
10456as an object; and the category of substance is applied to the intuition.
10457But this unity is nothing more than the unity in thought, by which no
10458object is given; to which therefore the category of substance--which
10459always presupposes a given intuition--cannot be applied. Consequently,
10460the subject cannot be cognized. The subject of the categories cannot,
10461therefore, for the very reason that it cogitates these, frame any
10462conception of itself as an object of the categories; for, to
10463cogitate these, it must lay at the foundation its own pure
10464self-consciousness--the very thing that it wishes to explain and
10465describe. In like manner, the subject, in which the representation of
10466time has its basis, cannot determine, for this very reason, its own
10467existence in time. Now, if the latter is impossible, the former, as an
10468attempt to determine itself by means of the categories as a thinking
10469being in general, is no less so.*
10470
10471
10472 [*Footnote: The "I think" is, as has been already stated, an empirical
10473proposition, and contains the proposition, "I exist." But I cannot say,
10474"Everything, which thinks, exists"; for in this case the property of
10475thought would constitute all beings possessing it, necessary beings.
10476Hence my existence cannot be considered as an inference from the
10477proposition, "I think," as Descartes maintained--because in this case
10478the major premiss, "Everything, which thinks, exists," must precede--but
10479the two propositions are identical. The proposition, "I think,"
10480expresses an undetermined empirical intuition, that perception (proving
10481consequently that sensation, which must belong to sensibility, lies at
10482the foundation of this proposition); but it precedes experience, whose
10483province it is to determine an object of perception by means of the
10484categories in relation to time; and existence in this proposition is not
10485a category, as it does not apply to an undetermined given object, but
10486only to one of which we have a conception, and about which we wish to
10487know whether it does or does not exist, out of, and apart from this
10488conception. An undetermined perception signifies here merely something
10489real that has been given, only, however, to thought in general--but
10490not as a phenomenon, nor as a thing in itself (noumenon), but only
10491as something that really exists, and is designated as such in the
10492proposition, "I think." For it must be remarked that, when I call the
10493proposition, "I think," an empirical proposition, I do not thereby mean
10494that the Ego in the proposition is an empirical representation; on the
10495contrary, it is purely intellectual, because it belongs to thought in
10496general. But without some empirical representation, which presents to
10497the mind material for thought, the mental act, "I think," would not take
10498place; and the empirical is only the condition of the application or
10499employment of the pure intellectual faculty.]
10500
10501
10502Thus, then, appears the vanity of the hope of establishing a cognition
10503which is to extend its rule beyond the limits of experience--a cognition
10504which is one of the highest interests of humanity; and thus is proved
10505the futility of the attempt of speculative philosophy in this region of
10506thought. But, in this interest of thought, the severity of criticism has
10507rendered to reason a not unimportant service, by the demonstration of
10508the impossibility of making any dogmatical affirmation concerning an
10509object of experience beyond the boundaries of experience. She has thus
10510fortified reason against all affirmations of the contrary. Now, this can
10511be accomplished in only two ways. Either our proposition must be
10512proved apodeictically; or, if this is unsuccessful, the sources of this
10513inability must be sought for, and, if these are discovered to exist in
10514the natural and necessary limitation of our reason, our opponents must
10515submit to the same law of renunciation and refrain from advancing claims
10516to dogmatic assertion.
10517
10518But the right, say rather the necessity to admit a future life, upon
10519principles of the practical conjoined with the speculative use
10520of reason, has lost nothing by this renunciation; for the merely
10521speculative proof has never had any influence upon the common reason of
10522men. It stands upon the point of a hair, so that even the schools have
10523been able to preserve it from falling only by incessantly discussing
10524it and spinning it like a top; and even in their eyes it has never been
10525able to present any safe foundation for the erection of a theory.
10526The proofs which have been current among men, preserve their value
10527undiminished; nay, rather gain in clearness and unsophisticated power,
10528by the rejection of the dogmatical assumptions of speculative reason.
10529For reason is thus confined within her own peculiar province--the
10530arrangement of ends or aims, which is at the same time the arrangement
10531of nature; and, as a practical faculty, without limiting itself to the
10532latter, it is justified in extending the former, and with it our own
10533existence, beyond the boundaries of experience and life. If we turn our
10534attention to the analogy of the nature of living beings in this world,
10535in the consideration of which reason is obliged to accept as a principle
10536that no organ, no faculty, no appetite is useless, and that nothing is
10537superfluous, nothing disproportionate to its use, nothing unsuited to
10538its end; but that, on the contrary, everything is perfectly conformed to
10539its destination in life--we shall find that man, who alone is the final
10540end and aim of this order, is still the only animal that seems to be
10541excepted from it. For his natural gifts--not merely as regards the
10542talents and motives that may incite him to employ them, but especially
10543the moral law in him--stretch so far beyond all mere earthly utility and
10544advantage, that he feels himself bound to prize the mere consciousness
10545of probity, apart from all advantageous consequences--even the shadowy
10546gift of posthumous fame--above everything; and he is conscious of an
10547inward call to constitute himself, by his conduct in this world--without
10548regard to mere sublunary interests--the citizen of a better. This
10549mighty, irresistible proof--accompanied by an ever-increasing knowledge
10550of the conformability to a purpose in everything we see around us,
10551by the conviction of the boundless immensity of creation, by the
10552consciousness of a certain illimitableness in the possible extension
10553of our knowledge, and by a desire commensurate therewith--remains to
10554humanity, even after the theoretical cognition of ourselves has failed
10555to establish the necessity of an existence after death.
10556
10557
10558
10559Conclusion of the Solution of the Psychological Paralogism.
10560
10561The dialectical illusion in rational psychology arises from our
10562confounding an idea of reason (of a pure intelligence) with the
10563conception--in every respect undetermined--of a thinking being in
10564general. I cogitate myself in behalf of a possible experience, at
10565the same time making abstraction of all actual experience; and infer
10566therefrom that I can be conscious of myself apart from experience
10567and its empirical conditions. I consequently confound the possible
10568abstraction of my empirically determined existence with the supposed
10569consciousness of a possible separate existence of my thinking self;
10570and I believe that I cognize what is substantial in myself as a
10571transcendental subject, when I have nothing more in thought than the
10572unity of consciousness, which lies at the basis of all determination of
10573cognition.
10574
10575The task of explaining the community of the soul with the body does not
10576properly belong to the psychology of which we are here speaking;
10577because it proposes to prove the personality of the soul apart from this
10578communion (after death), and is therefore transcendent in the proper
10579sense of the word, although occupying itself with an object of
10580experience--only in so far, however, as it ceases to be an object of
10581experience. But a sufficient answer may be found to the question in
10582our system. The difficulty which lies in the execution of this task
10583consists, as is well known, in the presupposed heterogeneity of the
10584object of the internal sense (the soul) and the objects of the external
10585senses; inasmuch as the formal condition of the intuition of the one is
10586time, and of that of the other space also. But if we consider that both
10587kinds of objects do not differ internally, but only in so far as the
10588one appears externally to the other--consequently, that what lies at the
10589basis of phenomena, as a thing in itself, may not be heterogeneous; this
10590difficulty disappears. There then remains no other difficulty than is to
10591be found in the question--how a community of substances is possible;
10592a question which lies out of the region of psychology, and which the
10593reader, after what in our analytic has been said of primitive forces
10594and faculties, will easily judge to be also beyond the region of human
10595cognition.
10596
10597
10598
10599GENERAL REMARK
10600
10601On the Transition from Rational Psychology to Cosmology.
10602
10603The proposition, "I think," or, "I exist thinking," is an empirical
10604proposition. But such a proposition must be based on empirical
10605intuition, and the object cogitated as a phenomenon; and thus our
10606theory appears to maintain that the soul, even in thought, is merely
10607a phenomenon; and in this way our consciousness itself, in fact, abuts
10608upon nothing.
10609
10610Thought, per se, is merely the purely spontaneous logical function which
10611operates to connect the manifold of a possible intuition; and it does
10612not represent the subject of consciousness as a phenomenon--for this
10613reason alone, that it pays no attention to the question whether the
10614mode of intuiting it is sensuous or intellectual. I therefore do not
10615represent myself in thought either as I am, or as I appear to myself; I
10616merely cogitate myself as an object in general, of the mode of intuiting
10617which I make abstraction. When I represent myself as the subject of
10618thought, or as the ground of thought, these modes of representation are
10619not related to the categories of substance or of cause; for these are
10620functions of thought applicable only to our sensuous intuition. The
10621application of these categories to the Ego would, however, be necessary,
10622if I wished to make myself an object of knowledge. But I wish to be
10623conscious of myself only as thinking; in what mode my Self is given
10624in intuition, I do not consider, and it may be that I, who think, am a
10625phenomenon--although not in so far as I am a thinking being; but in
10626the consciousness of myself in mere thought I am a being, though this
10627consciousness does not present to me any property of this being as
10628material for thought.
10629
10630But the proposition, "I think," in so far as it declares, "I exist
10631thinking," is not the mere representation of a logical function.
10632It determines the subject (which is in this case an object also) in
10633relation to existence; and it cannot be given without the aid of the
10634internal sense, whose intuition presents to us an object, not as a thing
10635in itself, but always as a phenomenon. In this proposition there is
10636therefore something more to be found than the mere spontaneity of
10637thought; there is also the receptivity of intuition, that is, my thought
10638of myself applied to the empirical intuition of myself. Now, in this
10639intuition the thinking self must seek the conditions of the employment
10640of its logical functions as categories of substance, cause, and so
10641forth; not merely for the purpose of distinguishing itself as an object
10642in itself by means of the representation "I," but also for the purpose
10643of determining the mode of its existence, that is, of cognizing
10644itself as noumenon. But this is impossible, for the internal empirical
10645intuition is sensuous, and presents us with nothing but phenomenal data,
10646which do not assist the object of pure consciousness in its attempt
10647to cognize itself as a separate existence, but are useful only as
10648contributions to experience.
10649
10650But, let it be granted that we could discover, not in experience, but in
10651certain firmly-established a priori laws of the use of pure reason--laws
10652relating to our existence, authority to consider ourselves as
10653legislating a priori in relation to our own existence and as determining
10654this existence; we should, on this supposition, find ourselves possessed
10655of a spontaneity, by which our actual existence would be determinable,
10656without the aid of the conditions of empirical intuition. We should also
10657become aware that in the consciousness of our existence there was an
10658a priori content, which would serve to determine our own existence--an
10659existence only sensuously determinable--relatively, however, to a
10660certain internal faculty in relation to an intelligible world.
10661
10662But this would not give the least help to the attempts of rational
10663psychology. For this wonderful faculty, which the consciousness of
10664the moral law in me reveals, would present me with a principle of the
10665determination of my own existence which is purely intellectual--but by
10666what predicates? By none other than those which are given in sensuous
10667intuition. Thus I should find myself in the same position in rational
10668psychology which I formerly occupied, that is to say, I should
10669find myself still in need of sensuous intuitions, in order to give
10670significance to my conceptions of substance and cause, by means of which
10671alone I can possess a knowledge of myself: but these intuitions can
10672never raise me above the sphere of experience. I should be justified,
10673however, in applying these conceptions, in regard to their practical
10674use, which is always directed to objects of experience--in conformity
10675with their analogical significance when employed theoretically--to
10676freedom and its subject. At the same time, I should understand by them
10677merely the logical functions of subject and predicate, of principle and
10678consequence, in conformity with which all actions are so determined,
10679that they are capable of being explained along with the laws of nature,
10680conformably to the categories of substance and cause, although
10681they originate from a very different principle. We have made these
10682observations for the purpose of guarding against misunderstanding, to
10683which the doctrine of our intuition of self as a phenomenon is exposed.
10684We shall have occasion to perceive their utility in the sequel.
10685
10686
10687
10688CHAPTER II. The Antinomy of Pure Reason.
10689
10690We showed in the introduction to this part of our work, that all
10691transcendental illusion of pure reason arose from dialectical arguments,
10692the schema of which logic gives us in its three formal species of
10693syllogisms--just as the categories find their logical schema in the
10694four functions of all judgements. The first kind of these sophistical
10695arguments related to the unconditioned unity of the subjective
10696conditions of all representations in general (of the subject or soul),
10697in correspondence with the categorical syllogisms, the major of which,
10698as the principle, enounces the relation of a predicate to a subject.
10699The second kind of dialectical argument will therefore be concerned,
10700following the analogy with hypothetical syllogisms, with the
10701unconditioned unity of the objective conditions in the phenomenon;
10702and, in this way, the theme of the third kind to be treated of in the
10703following chapter will be the unconditioned unity of the objective
10704conditions of the possibility of objects in general.
10705
10706But it is worthy of remark that the transcendental paralogism produced
10707in the mind only a one-third illusion, in regard to the idea of the
10708subject of our thought; and the conceptions of reason gave no ground to
10709maintain the contrary proposition. The advantage is completely on the
10710side of Pneumatism; although this theory itself passes into naught, in
10711the crucible of pure reason.
10712
10713Very different is the case when we apply reason to the objective
10714synthesis of phenomena. Here, certainly, reason establishes, with much
10715plausibility, its principle of unconditioned unity; but it very soon
10716falls into such contradictions that it is compelled, in relation to
10717cosmology, to renounce its pretensions.
10718
10719For here a new phenomenon of human reason meets us--a perfectly natural
10720antithetic, which does not require to be sought for by subtle sophistry,
10721but into which reason of itself unavoidably falls. It is thereby
10722preserved, to be sure, from the slumber of a fancied conviction--which
10723a merely one-sided illusion produces; but it is at the same time
10724compelled, either, on the one hand, to abandon itself to a despairing
10725scepticism, or, on the other, to assume a dogmatical confidence and
10726obstinate persistence in certain assertions, without granting a fair
10727hearing to the other side of the question. Either is the death of a
10728sound philosophy, although the former might perhaps deserve the title of
10729the euthanasia of pure reason.
10730
10731Before entering this region of discord and confusion, which the conflict
10732of the laws of pure reason (antinomy) produces, we shall present the
10733reader with some considerations, in explanation and justification of the
10734method we intend to follow in our treatment of this subject. I term all
10735transcendental ideas, in so far as they relate to the absolute totality
10736in the synthesis of phenomena, cosmical conceptions; partly on
10737account of this unconditioned totality, on which the conception of the
10738world-whole is based--a conception, which is itself an idea--partly
10739because they relate solely to the synthesis of phenomena--the empirical
10740synthesis; while, on the other hand, the absolute totality in the
10741synthesis of the conditions of all possible things gives rise to
10742an ideal of pure reason, which is quite distinct from the cosmical
10743conception, although it stands in relation with it. Hence, as the
10744paralogisms of pure reason laid the foundation for a dialectical
10745psychology, the antinomy of pure reason will present us with the
10746transcendental principles of a pretended pure (rational) cosmology--not,
10747however, to declare it valid and to appropriate it, but--as the very
10748term of a conflict of reason sufficiently indicates, to present it as an
10749idea which cannot be reconciled with phenomena and experience.
10750
10751
10752
10753SECTION I. System of Cosmological Ideas.
10754
10755That We may be able to enumerate with systematic precision these ideas
10756according to a principle, we must remark, in the first place, that it
10757is from the understanding alone that pure and transcendental conceptions
10758take their origin; that the reason does not properly give birth to any
10759conception, but only frees the conception of the understanding from the
10760unavoidable limitation of a possible experience, and thus endeavours to
10761raise it above the empirical, though it must still be in connection with
10762it. This happens from the fact that, for a given conditioned, reason
10763demands absolute totality on the side of the conditions (to which the
10764understanding submits all phenomena), and thus makes of the category a
10765transcendental idea. This it does that it may be able to give absolute
10766completeness to the empirical synthesis, by continuing it to the
10767unconditioned (which is not to be found in experience, but only in
10768the idea). Reason requires this according to the principle: If the
10769conditioned is given the whole of the conditions, and consequently the
10770absolutely unconditioned, is also given, whereby alone the former was
10771possible. First, then, the transcendental ideas are properly nothing but
10772categories elevated to the unconditioned; and they may be arranged in
10773a table according to the titles of the latter. But, secondly, all the
10774categories are not available for this purpose, but only those in which
10775the synthesis constitutes a series--of conditions subordinated to, not
10776co-ordinated with, each other. Absolute totality is required of reason
10777only in so far as concerns the ascending series of the conditions of
10778a conditioned; not, consequently, when the question relates to
10779the descending series of consequences, or to the aggregate of the
10780co-ordinated conditions of these consequences. For, in relation to a
10781given conditioned, conditions are presupposed and considered to be given
10782along with it. On the other hand, as the consequences do not render
10783possible their conditions, but rather presuppose them--in the
10784consideration of the procession of consequences (or in the descent from
10785the given condition to the conditioned), we may be quite unconcerned
10786whether the series ceases or not; and their totality is not a necessary
10787demand of reason.
10788
10789Thus we cogitate--and necessarily--a given time completely elapsed up
10790to a given moment, although that time is not determinable by us. But
10791as regards time future, which is not the condition of arriving at the
10792present, in order to conceive it; it is quite indifferent whether we
10793consider future time as ceasing at some point, or as prolonging itself
10794to infinity. Take, for example, the series m, n, o, in which n is given
10795as conditioned in relation to m, but at the same time as the condition
10796of o, and let the series proceed upwards from the conditioned n to m (l,
10797k, i, etc.), and also downwards from the condition n to the conditioned
10798o (p, q, r, etc.)--I must presuppose the former series, to be able
10799to consider n as given, and n is according to reason (the totality of
10800conditions) possible only by means of that series. But its possibility
10801does not rest on the following series o, p, q, r, which for this
10802reason cannot be regarded as given, but only as capable of being given
10803(dabilis).
10804
10805I shall term the synthesis of the series on the side of the
10806conditions--from that nearest to the given phenomenon up to the more
10807remote--regressive; that which proceeds on the side of the conditioned,
10808from the immediate consequence to the more remote, I shall call the
10809progressive synthesis. The former proceeds in antecedentia, the latter
10810in consequentia. The cosmological ideas are therefore occupied with the
10811totality of the regressive synthesis, and proceed in antecedentia, not
10812in consequentia. When the latter takes place, it is an arbitrary and
10813not a necessary problem of pure reason; for we require, for the complete
10814understanding of what is given in a phenomenon, not the consequences
10815which succeed, but the grounds or principles which precede.
10816
10817In order to construct the table of ideas in correspondence with the
10818table of categories, we take first the two primitive quanta of all our
10819intuitions, time and space. Time is in itself a series (and the formal
10820condition of all series), and hence, in relation to a given present,
10821we must distinguish a priori in it the antecedentia as conditions
10822(time past) from the consequentia (time future). Consequently, the
10823transcendental idea of the absolute totality of the series of the
10824conditions of a given conditioned, relates merely to all past time.
10825According to the idea of reason, the whole past time, as the condition
10826of the given moment, is necessarily cogitated as given. But, as
10827regards space, there exists in it no distinction between progressus and
10828regressus; for it is an aggregate and not a series--its parts existing
10829together at the same time. I can consider a given point of time in
10830relation to past time only as conditioned, because this given moment
10831comes into existence only through the past time rather through the
10832passing of the preceding time. But as the parts of space are not
10833subordinated, but co-ordinated to each other, one part cannot be the
10834condition of the possibility of the other; and space is not in itself,
10835like time, a series. But the synthesis of the manifold parts of
10836space--(the syntheses whereby we apprehend space)--is nevertheless
10837successive; it takes place, therefore, in time, and contains a series.
10838And as in this series of aggregated spaces (for example, the feet in a
10839rood), beginning with a given portion of space, those which continue
10840to be annexed form the condition of the limits of the former--the
10841measurement of a space must also be regarded as a synthesis of the
10842series of the conditions of a given conditioned. It differs, however, in
10843this respect from that of time, that the side of the conditioned is
10844not in itself distinguishable from the side of the condition; and,
10845consequently, regressus and progressus in space seem to be identical.
10846But, inasmuch as one part of space is not given, but only limited,
10847by and through another, we must also consider every limited space
10848as conditioned, in so far as it presupposes some other space as
10849the condition of its limitation, and so on. As regards limitation,
10850therefore, our procedure in space is also a regressus, and the
10851transcendental idea of the absolute totality of the synthesis in a
10852series of conditions applies to space also; and I am entitled to demand
10853the absolute totality of the phenomenal synthesis in space as well as in
10854time. Whether my demand can be satisfied is a question to be answered in
10855the sequel.
10856
10857Secondly, the real in space--that is, matter--is conditioned. Its
10858internal conditions are its parts, and the parts of parts its remote
10859conditions; so that in this case we find a regressive synthesis, the
10860absolute totality of which is a demand of reason. But this cannot be
10861obtained otherwise than by a complete division of parts, whereby the
10862real in matter becomes either nothing or that which is not matter,
10863that is to say, the simple. Consequently we find here also a series of
10864conditions and a progress to the unconditioned.
10865
10866Thirdly, as regards the categories of a real relation between phenomena,
10867the category of substance and its accidents is not suitable for the
10868formation of a transcendental idea; that is to say, reason has no
10869ground, in regard to it, to proceed regressively with conditions. For
10870accidents (in so far as they inhere in a substance) are co-ordinated
10871with each other, and do not constitute a series. And, in relation to
10872substance, they are not properly subordinated to it, but are the mode
10873of existence of the substance itself. The conception of the substantial
10874might nevertheless seem to be an idea of the transcendental reason.
10875But, as this signifies nothing more than the conception of an object
10876in general, which subsists in so far as we cogitate in it merely a
10877transcendental subject without any predicates; and as the question here
10878is of an unconditioned in the series of phenomena--it is clear that
10879the substantial can form no member thereof. The same holds good of
10880substances in community, which are mere aggregates and do not form a
10881series. For they are not subordinated to each other as conditions of the
10882possibility of each other; which, however, may be affirmed of spaces,
10883the limits of which are never determined in themselves, but always by
10884some other space. It is, therefore, only in the category of causality
10885that we can find a series of causes to a given effect, and in which
10886we ascend from the latter, as the conditioned, to the former as the
10887conditions, and thus answer the question of reason.
10888
10889Fourthly, the conceptions of the possible, the actual, and the necessary
10890do not conduct us to any series--excepting only in so far as the
10891contingent in existence must always be regarded as conditioned, and as
10892indicating, according to a law of the understanding, a condition, under
10893which it is necessary to rise to a higher, till in the totality of the
10894series, reason arrives at unconditioned necessity.
10895
10896There are, accordingly, only four cosmological ideas, corresponding
10897with the four titles of the categories. For we can select only such as
10898necessarily furnish us with a series in the synthesis of the manifold.
10899
10900 1
10901 The absolute Completeness
10902 of the
10903 COMPOSITION
10904 of the given totality of all phenomena.
10905
10906 2
10907 The absolute Completeness
10908 of the
10909 DIVISION
10910 of given totality in a phenomenon.
10911
10912 3
10913 The absolute Completeness
10914 of the
10915 ORIGINATION
10916 of a phenomenon.
10917
10918 4
10919 The absolute Completeness
10920 of the DEPENDENCE of the EXISTENCE
10921 of what is changeable in a phenomenon.
10922
10923
10924We must here remark, in the first place, that the idea of absolute
10925totality relates to nothing but the exposition of phenomena, and
10926therefore not to the pure conception of a totality of things. Phenomena
10927are here, therefore, regarded as given, and reason requires the absolute
10928completeness of the conditions of their possibility, in so far as these
10929conditions constitute a series--consequently an absolutely (that is, in
10930every respect) complete synthesis, whereby a phenomenon can be explained
10931according to the laws of the understanding.
10932
10933Secondly, it is properly the unconditioned alone that reason seeks in
10934this serially and regressively conducted synthesis of conditions. It
10935wishes, to speak in another way, to attain to completeness in the series
10936of premisses, so as to render it unnecessary to presuppose others.
10937This unconditioned is always contained in the absolute totality of the
10938series, when we endeavour to form a representation of it in thought.
10939But this absolutely complete synthesis is itself but an idea; for it is
10940impossible, at least before hand, to know whether any such synthesis is
10941possible in the case of phenomena. When we represent all existence in
10942thought by means of pure conceptions of the understanding, without any
10943conditions of sensuous intuition, we may say with justice that for a
10944given conditioned the whole series of conditions subordinated to each
10945other is also given; for the former is only given through the latter.
10946But we find in the case of phenomena a particular limitation of the mode
10947in which conditions are given, that is, through the successive synthesis
10948of the manifold of intuition, which must be complete in the regress. Now
10949whether this completeness is sensuously possible, is a problem. But the
10950idea of it lies in the reason--be it possible or impossible to connect
10951with the idea adequate empirical conceptions. Therefore, as in the
10952absolute totality of the regressive synthesis of the manifold in a
10953phenomenon (following the guidance of the categories, which represent it
10954as a series of conditions to a given conditioned) the unconditioned is
10955necessarily contained--it being still left unascertained whether and
10956how this totality exists; reason sets out from the idea of totality,
10957although its proper and final aim is the unconditioned--of the whole
10958series, or of a part thereof.
10959
10960This unconditioned may be cogitated--either as existing only in the
10961entire series, all the members of which therefore would be
10962without exception conditioned and only the totality absolutely
10963unconditioned--and in this case the regressus is called infinite; or
10964the absolutely unconditioned is only a part of the series, to which the
10965other members are subordinated, but which Is not itself submitted to
10966any other condition.* In the former case the series is a parte priori
10967unlimited (without beginning), that is, infinite, and nevertheless
10968completely given. But the regress in it is never completed, and can only
10969be called potentially infinite. In the second case there exists a first
10970in the series. This first is called, in relation to past time, the
10971beginning of the world; in relation to space, the limit of the world; in
10972relation to the parts of a given limited whole, the simple; in relation
10973to causes, absolute spontaneity (liberty); and in relation to the
10974existence of changeable things, absolute physical necessity.
10975
10976
10977 [*Footnote: The absolute totality of the series of conditions to a given
10978conditioned is always unconditioned; because beyond it there exist no
10979other conditions, on which it might depend. But the absolute totality of
10980such a series is only an idea, or rather a problematical conception, the
10981possibility of which must be investigated--particularly in relation to
10982the mode in which the unconditioned, as the transcendental idea which is
10983the real subject of inquiry, may be contained therein.]
10984
10985
10986We possess two expressions, world and nature, which are generally
10987interchanged. The first denotes the mathematical total of all phenomena
10988and the totality of their synthesis--in its progress by means of
10989composition, as well as by division. And the world is termed nature,*
10990when it is regarded as a dynamical whole--when our attention is not
10991directed to the aggregation in space and time, for the purpose of
10992cogitating it as a quantity, but to the unity in the existence of
10993phenomena. In this case the condition of that which happens is called
10994a cause; the unconditioned causality of the cause in a phenomenon is
10995termed liberty; the conditioned cause is called in a more limited sense
10996a natural cause. The conditioned in existence is termed contingent, and
10997the unconditioned necessary. The unconditioned necessity of phenomena
10998may be called natural necessity.
10999
11000
11001 [*Footnote: Nature, understood adjective (formaliter), signifies the
11002complex of the determinations of a thing, connected according to an
11003internal principle of causality. On the other hand, we understand by
11004nature, substantive (materialiter), the sum total of phenomena, in
11005so far as they, by virtue of an internal principle of causality, are
11006connected with each other throughout. In the former sense we speak of
11007the nature of liquid matter, of fire, etc., and employ the word only
11008adjective; while, if speaking of the objects of nature, we have in our
11009minds the idea of a subsisting whole.]
11010
11011
11012The ideas which we are at present engaged in discussing I have called
11013cosmological ideas; partly because by the term world is understood the
11014entire content of all phenomena, and our ideas are directed solely to
11015the unconditioned among phenomena; partly also, because world, in the
11016transcendental sense, signifies the absolute totality of the content
11017of existing things, and we are directing our attention only to the
11018completeness of the synthesis--although, properly, only in regression.
11019In regard to the fact that these ideas are all transcendent, and,
11020although they do not transcend phenomena as regards their mode, but
11021are concerned solely with the world of sense (and not with noumena),
11022nevertheless carry their synthesis to a degree far above all possible
11023experience--it still seems to me that we can, with perfect propriety,
11024designate them cosmical conceptions. As regards the distinction between
11025the mathematically and the dynamically unconditioned which is the aim of
11026the regression of the synthesis, I should call the two former, in a
11027more limited signification, cosmical conceptions, the remaining two
11028transcendent physical conceptions. This distinction does not at present
11029seem to be of particular importance, but we shall afterwards find it to
11030be of some value.
11031
11032
11033
11034SECTION II. Antithetic of Pure Reason.
11035
11036Thetic is the term applied to every collection of dogmatical
11037propositions. By antithetic I do not understand dogmatical assertions
11038of the opposite, but the self-contradiction of seemingly dogmatical
11039cognitions (thesis cum antithesis), in none of which we can discover
11040any decided superiority. Antithetic is not, therefore, occupied with
11041one-sided statements, but is engaged in considering the contradictory
11042nature of the general cognitions of reason and its causes.
11043Transcendental antithetic is an investigation into the antinomy of pure
11044reason, its causes and result. If we employ our reason not merely in
11045the application of the principles of the understanding to objects of
11046experience, but venture with it beyond these boundaries, there arise
11047certain sophistical propositions or theorems. These assertions have
11048the following peculiarities: They can find neither confirmation
11049nor confutation in experience; and each is in itself not only
11050self-consistent, but possesses conditions of its necessity in the very
11051nature of reason--only that, unluckily, there exist just as valid and
11052necessary grounds for maintaining the contrary proposition.
11053
11054The questions which naturally arise in the consideration of this
11055dialectic of pure reason, are therefore: 1st. In what propositions is
11056pure reason unavoidably subject to an antinomy? 2nd. What are the causes
11057of this antinomy? 3rd. Whether and in what way can reason free itself
11058from this self-contradiction?
11059
11060A dialectical proposition or theorem of pure reason must, according
11061to what has been said, be distinguishable from all sophistical
11062propositions, by the fact that it is not an answer to an arbitrary
11063question, which may be raised at the mere pleasure of any person, but
11064to one which human reason must necessarily encounter in its progress. In
11065the second place, a dialectical proposition, with its opposite, does not
11066carry the appearance of a merely artificial illusion, which disappears
11067as soon as it is investigated, but a natural and unavoidable illusion,
11068which, even when we are no longer deceived by it, continues to mock us
11069and, although rendered harmless, can never be completely removed.
11070
11071This dialectical doctrine will not relate to the unity of understanding
11072in empirical conceptions, but to the unity of reason in pure ideas. The
11073conditions of this doctrine are--inasmuch as it must, as a synthesis
11074according to rules, be conformable to the understanding, and at the same
11075time as the absolute unity of the synthesis, to the reason--that, if
11076it is adequate to the unity of reason, it is too great for the
11077understanding, if according with the understanding, it is too small for
11078the reason. Hence arises a mutual opposition, which cannot be avoided,
11079do what we will.
11080
11081These sophistical assertions of dialectic open, as it were, a
11082battle-field, where that side obtains the victory which has been
11083permitted to make the attack, and he is compelled to yield who has been
11084unfortunately obliged to stand on the defensive. And hence, champions of
11085ability, whether on the right or on the wrong side, are certain to carry
11086away the crown of victory, if they only take care to have the right to
11087make the last attack, and are not obliged to sustain another onset from
11088their opponent. We can easily believe that this arena has been often
11089trampled by the feet of combatants, that many victories have been
11090obtained on both sides, but that the last victory, decisive of the
11091affair between the contending parties, was won by him who fought for the
11092right, only if his adversary was forbidden to continue the tourney. As
11093impartial umpires, we must lay aside entirely the consideration whether
11094the combatants are fighting for the right or for the wrong side, for
11095the true or for the false, and allow the combat to be first decided.
11096Perhaps, after they have wearied more than injured each other, they
11097will discover the nothingness of their cause of quarrel and part good
11098friends.
11099
11100This method of watching, or rather of originating, a conflict of
11101assertions, not for the purpose of finally deciding in favour of either
11102side, but to discover whether the object of the struggle is not a mere
11103illusion, which each strives in vain to reach, but which would be
11104no gain even when reached--this procedure, I say, may be termed the
11105sceptical method. It is thoroughly distinct from scepticism--the
11106principle of a technical and scientific ignorance, which undermines
11107the foundations of all knowledge, in order, if possible, to destroy
11108our belief and confidence therein. For the sceptical method aims at
11109certainty, by endeavouring to discover in a conflict of this kind,
11110conducted honestly and intelligently on both sides, the point
11111of misunderstanding; just as wise legislators derive, from the
11112embarrassment of judges in lawsuits, information in regard to the
11113defective and ill-defined parts of their statutes. The antinomy which
11114reveals itself in the application of laws, is for our limited wisdom
11115the best criterion of legislation. For the attention of reason, which in
11116abstract speculation does not easily become conscious of its errors, is
11117thus roused to the momenta in the determination of its principles.
11118
11119But this sceptical method is essentially peculiar to transcendental
11120philosophy, and can perhaps be dispensed with in every other field of
11121investigation. In mathematics its use would be absurd; because in it no
11122false assertions can long remain hidden, inasmuch as its demonstrations
11123must always proceed under the guidance of pure intuition, and by means
11124of an always evident synthesis. In experimental philosophy, doubt and
11125delay may be very useful; but no misunderstanding is possible, which
11126cannot be easily removed; and in experience means of solving the
11127difficulty and putting an end to the dissension must at last be found,
11128whether sooner or later. Moral philosophy can always exhibit its
11129principles, with their practical consequences, in concreto--at least in
11130possible experiences, and thus escape the mistakes and ambiguities of
11131abstraction. But transcendental propositions, which lay claim to insight
11132beyond the region of possible experience, cannot, on the one hand,
11133exhibit their abstract synthesis in any a priori intuition, nor, on the
11134other, expose a lurking error by the help of experience. Transcendental
11135reason, therefore, presents us with no other criterion than that of an
11136attempt to reconcile such assertions, and for this purpose to permit a
11137free and unrestrained conflict between them. And this we now proceed to
11138arrange.*
11139
11140
11141
11142 [*Footnote: The antinomies stand in the order of the four transcendental
11143ideas above detailed.]
11144
11145
11146
11147FIRST CONFLICT OF THE TRANSCENDENTAL IDEAS.
11148
11149THESIS.
11150
11151The world has a beginning in time, and is also limited in regard to
11152space.
11153
11154
11155PROOF.
11156
11157Granted that the world has no beginning in time; up to every given
11158moment of time, an eternity must have elapsed, and therewith passed away
11159an infinite series of successive conditions or states of things in the
11160world. Now the infinity of a series consists in the fact that it never
11161can be completed by means of a successive synthesis. It follows that an
11162infinite series already elapsed is impossible and that, consequently,
11163a beginning of the world is a necessary condition of its existence. And
11164this was the first thing to be proved.
11165
11166As regards the second, let us take the opposite for granted. In this
11167case, the world must be an infinite given total of coexistent things.
11168Now we cannot cogitate the dimensions of a quantity, which is not given
11169within certain limits of an intuition,* in any other way than by means
11170of the synthesis of its parts, and the total of such a quantity only
11171by means of a completed synthesis, or the repeated addition of unity to
11172itself. Accordingly, to cogitate the world, which fills all spaces, as
11173a whole, the successive synthesis of the parts of an infinite world must
11174be looked upon as completed, that is to say, an infinite time must be
11175regarded as having elapsed in the enumeration of all co-existing things;
11176which is impossible. For this reason an infinite aggregate of actual
11177things cannot be considered as a given whole, consequently, not as a
11178contemporaneously given whole. The world is consequently, as regards
11179extension in space, not infinite, but enclosed in limits. And this was
11180the second thing to be proved.
11181
11182
11183 [*Footnote: We may consider an undetermined quantity as a whole, when it
11184is enclosed within limits, although we cannot construct or ascertain
11185its totality by measurement, that is, by the successive synthesis of
11186its parts. For its limits of themselves determine its completeness as a
11187whole.]
11188
11189
11190
11191ANTITHESIS.
11192
11193The world has no beginning, and no limits in space, but is, in relation
11194both to time and space, infinite.
11195
11196
11197PROOF.
11198
11199For let it be granted that it has a beginning. A beginning is an
11200existence which is preceded by a time in which the thing does not exist.
11201On the above supposition, it follows that there must have been a time in
11202which the world did not exist, that is, a void time. But in a void time
11203the origination of a thing is impossible; because no part of any such
11204time contains a distinctive condition of being, in preference to that of
11205non-being (whether the supposed thing originate of itself, or by means
11206of some other cause). Consequently, many series of things may have a
11207beginning in the world, but the world itself cannot have a beginning,
11208and is, therefore, in relation to past time, infinite.
11209
11210As regards the second statement, let us first take the opposite for
11211granted--that the world is finite and limited in space; it follows that
11212it must exist in a void space, which is not limited. We should therefore
11213meet not only with a relation of things in space, but also a relation
11214of things to space. Now, as the world is an absolute whole, out of and
11215beyond which no object of intuition, and consequently no correlate to
11216which can be discovered, this relation of the world to a void space is
11217merely a relation to no object. But such a relation, and consequently
11218the limitation of the world by void space, is nothing. Consequently,
11219the world, as regards space, is not limited, that is, it is infinite in
11220regard to extension.*
11221
11222
11223
11224 [*Footnote: Space is merely the form of external intuition (formal
11225intuition), and not a real object which can be externally perceived.
11226Space, prior to all things which determine it (fill or limit it), or,
11227rather, which present an empirical intuition conformable to it, is,
11228under the title of absolute space, nothing but the mere possibility of
11229external phenomena, in so far as they either exist in themselves, or can
11230annex themselves to given intuitions. Empirical intuition is therefore
11231not a composition of phenomena and space (of perception and empty
11232intuition). The one is not the correlate of the other in a synthesis,
11233but they are vitally connected in the same empirical intuition, as
11234matter and form. If we wish to set one of these two apart from
11235the other--space from phenomena--there arise all sorts of empty
11236determinations of external intuition, which are very far from being
11237possible perceptions. For example, motion or rest of the world in an
11238infinite empty space, or a determination of the mutual relation of both,
11239cannot possibly be perceived, and is therefore merely the predicate of a
11240notional entity.]
11241
11242
11243
11244OBSERVATIONS ON THE FIRST ANTINOMY.
11245
11246ON THE THESIS.
11247
11248In bringing forward these conflicting arguments, I have not been on
11249the search for sophisms, for the purpose of availing myself of special
11250pleading, which takes advantage of the carelessness of the opposite
11251party, appeals to a misunderstood statute, and erects its unrighteous
11252claims upon an unfair interpretation. Both proofs originate fairly from
11253the nature of the case, and the advantage presented by the mistakes of
11254the dogmatists of both parties has been completely set aside.
11255
11256The thesis might also have been unfairly demonstrated, by the
11257introduction of an erroneous conception of the infinity of a given
11258quantity. A quantity is infinite, if a greater than itself cannot
11259possibly exist. The quantity is measured by the number of given
11260units--which are taken as a standard--contained in it. Now no number can
11261be the greatest, because one or more units can always be added. It follows
11262that an infinite given quantity, consequently an infinite world (both as
11263regards time and extension) is impossible. It is, therefore, limited in
11264both respects. In this manner I might have conducted my proof; but the
11265conception given in it does not agree with the true conception of an
11266infinite whole. In this there is no representation of its quantity,
11267it is not said how large it is; consequently its conception is not the
11268conception of a maximum. We cogitate in it merely its relation to an
11269arbitrarily assumed unit, in relation to which it is greater than any
11270number. Now, just as the unit which is taken is greater or smaller, the
11271infinite will be greater or smaller; but the infinity, which consists
11272merely in the relation to this given unit, must remain always the same,
11273although the absolute quantity of the whole is not thereby cognized.
11274
11275The true (transcendental) conception of infinity is: that the successive
11276synthesis of unity in the measurement of a given quantum can never be
11277completed.* Hence it follows, without possibility of mistake, that an
11278eternity of actual successive states up to a given (the present) moment
11279cannot have elapsed, and that the world must therefore have a beginning.
11280
11281
11282 [*Footnote: The quantum in this sense contains a congeries of given
11283units, which is greater than any number--and this is the mathematical
11284conception of the infinite.]
11285
11286
11287In regard to the second part of the thesis, the difficulty as to an
11288infinite and yet elapsed series disappears; for the manifold of a world
11289infinite in extension is contemporaneously given. But, in order to
11290cogitate the total of this manifold, as we cannot have the aid of limits
11291constituting by themselves this total in intuition, we are obliged to
11292give some account of our conception, which in this case cannot proceed
11293from the whole to the determined quantity of the parts, but must
11294demonstrate the possibility of a whole by means of a successive
11295synthesis of the parts. But as this synthesis must constitute a series
11296that cannot be completed, it is impossible for us to cogitate prior to
11297it, and consequently not by means of it, a totality. For the conception
11298of totality itself is in the present case the representation of a
11299completed synthesis of the parts; and this completion, and consequently
11300its conception, is impossible.
11301
11302
11303
11304ON THE ANTITHESIS.
11305
11306The proof in favour of the infinity of the cosmical succession and the
11307cosmical content is based upon the consideration that, in the opposite
11308case, a void time and a void space must constitute the limits of the
11309world. Now I am not unaware, that there are some ways of escaping this
11310conclusion. It may, for example, be alleged, that a limit to the world,
11311as regards both space and time, is quite possible, without at the same
11312time holding the existence of an absolute time before the beginning of
11313the world, or an absolute space extending beyond the actual world--which
11314is impossible. I am quite well satisfied with the latter part of this
11315opinion of the philosophers of the Leibnitzian school. Space is merely
11316the form of external intuition, but not a real object which can itself
11317be externally intuited; it is not a correlate of phenomena, it is
11318the form of phenomena itself. Space, therefore, cannot be regarded as
11319absolutely and in itself something determinative of the existence
11320of things, because it is not itself an object, but only the form of
11321possible objects. Consequently, things, as phenomena, determine space;
11322that is to say, they render it possible that, of all the possible
11323predicates of space (size and relation), certain may belong to
11324reality. But we cannot affirm the converse, that space, as something
11325self-subsistent, can determine real things in regard to size or shape,
11326for it is in itself not a real thing. Space (filled or void)* may
11327therefore be limited by phenomena, but phenomena cannot be limited by
11328an empty space without them. This is true of time also. All this being
11329granted, it is nevertheless indisputable, that we must assume these two
11330nonentities, void space without and void time before the world, if we
11331assume the existence of cosmical limits, relatively to space or time.
11332
11333
11334 [*Footnote: It is evident that what is meant here is, that empty space,
11335in so far as it is limited by phenomena--space, that is, within the
11336world--does not at least contradict transcendental principles, and may
11337therefore, as regards them, be admitted, although its possibility cannot
11338on that account be affirmed.]
11339
11340
11341For, as regards the subterfuge adopted by those who endeavour to evade
11342the consequence--that, if the world is limited as to space and time, the
11343infinite void must determine the existence of actual things in regard
11344to their dimensions--it arises solely from the fact that instead of a
11345sensuous world, an intelligible world--of which nothing is known--is
11346cogitated; instead of a real beginning (an existence, which is preceded
11347by a period in which nothing exists), an existence which presupposes no
11348other condition than that of time; and, instead of limits of extension,
11349boundaries of the universe. But the question relates to the mundus
11350phaenomenon, and its quantity; and in this case we cannot make
11351abstraction of the conditions of sensibility, without doing away with
11352the essential reality of this world itself. The world of sense, if it is
11353limited, must necessarily lie in the infinite void. If this, and with it
11354space as the a priori condition of the possibility of phenomena, is left
11355out of view, the whole world of sense disappears. In our problem is this
11356alone considered as given. The mundus intelligibilis is nothing but the
11357general conception of a world, in which abstraction has been made of
11358all conditions of intuition, and in relation to which no synthetical
11359proposition--either affirmative or negative--is possible.
11360
11361
11362
11363SECOND CONFLICT OF TRANSCENDENTAL IDEAS.
11364
11365THESIS.
11366
11367Every composite substance in the world consists of simple parts; and
11368there exists nothing that is not either itself simple, or composed of
11369simple parts.
11370
11371
11372PROOF.
11373
11374For, grant that composite substances do not consist of simple parts;
11375in this case, if all combination or composition were annihilated in
11376thought, no composite part, and (as, by the supposition, there do
11377not exist simple parts) no simple part would exist. Consequently,
11378no substance; consequently, nothing would exist. Either, then, it
11379is impossible to annihilate composition in thought; or, after such
11380annihilation, there must remain something that subsists without
11381composition, that is, something that is simple. But in the former case
11382the composite could not itself consist of substances, because with
11383substances composition is merely a contingent relation, apart from
11384which they must still exist as self-subsistent beings. Now, as this case
11385contradicts the supposition, the second must contain the truth--that the
11386substantial composite in the world consists of simple parts.
11387
11388It follows, as an immediate inference, that the things in the world are
11389all, without exception, simple beings--that composition is merely an
11390external condition pertaining to them--and that, although we never
11391can separate and isolate the elementary substances from the state of
11392composition, reason must cogitate these as the primary subjects of
11393all composition, and consequently, as prior thereto--and as simple
11394substances.
11395
11396
11397ANTITHESIS.
11398
11399No composite thing in the world consists of simple parts; and there does
11400not exist in the world any simple substance.
11401
11402
11403PROOF.
11404
11405Let it be supposed that a composite thing (as substance) consists
11406of simple parts. Inasmuch as all external relation, consequently
11407all composition of substances, is possible only in space; the space,
11408occupied by that which is composite, must consist of the same number of
11409parts as is contained in the composite. But space does not consist of
11410simple parts, but of spaces. Therefore, every part of the composite must
11411occupy a space. But the absolutely primary parts of what is composite
11412are simple. It follows that what is simple occupies a space. Now, as
11413everything real that occupies a space, contains a manifold the parts of
11414which are external to each other, and is consequently composite--and
11415a real composite, not of accidents (for these cannot exist external to
11416each other apart from substance), but of substances--it follows that the
11417simple must be a substantial composite, which is self-contradictory.
11418
11419The second proposition of the antithesis--that there exists in the
11420world nothing that is simple--is here equivalent to the following:
11421The existence of the absolutely simple cannot be demonstrated from any
11422experience or perception either external or internal; and the absolutely
11423simple is a mere idea, the objective reality of which cannot be
11424demonstrated in any possible experience; it is consequently, in the
11425exposition of phenomena, without application and object. For, let us
11426take for granted that an object may be found in experience for this
11427transcendental idea; the empirical intuition of such an object must then
11428be recognized to contain absolutely no manifold with its parts external
11429to each other, and connected into unity. Now, as we cannot reason from
11430the non-consciousness of such a manifold to the impossibility of its
11431existence in the intuition of an object, and as the proof of this
11432impossibility is necessary for the establishment and proof of absolute
11433simplicity; it follows that this simplicity cannot be inferred from any
11434perception whatever. As, therefore, an absolutely simple object cannot
11435be given in any experience, and the world of sense must be considered as
11436the sum total of all possible experiences: nothing simple exists in the
11437world.
11438
11439This second proposition in the antithesis has a more extended aim than
11440the first. The first merely banishes the simple from the intuition of
11441the composite; while the second drives it entirely out of nature. Hence
11442we were unable to demonstrate it from the conception of a given object
11443of external intuition (of the composite), but we were obliged to prove
11444it from the relation of a given object to a possible experience in
11445general.
11446
11447
11448
11449OBSERVATIONS ON THE SECOND ANTINOMY.
11450
11451THESIS.
11452
11453When I speak of a whole, which necessarily consists of simple parts, I
11454understand thereby only a substantial whole, as the true composite; that
11455is to say, I understand that contingent unity of the manifold which is
11456given as perfectly isolated (at least in thought), placed in reciprocal
11457connection, and thus constituted a unity. Space ought not to be called a
11458compositum but a totum, for its parts are possible in the whole, and not
11459the whole by means of the parts. It might perhaps be called a compositum
11460ideale, but not a compositum reale. But this is of no importance. As
11461space is not a composite of substances (and not even of real accidents),
11462if I abstract all composition therein--nothing, not even a
11463point, remains; for a point is possible only as the limit of a
11464space--consequently of a composite. Space and time, therefore, do not
11465consist of simple parts. That which belongs only to the condition or
11466state of a substance, even although it possesses a quantity (motion or
11467change, for example), likewise does not consist of simple parts. That is
11468to say, a certain degree of change does not originate from the addition
11469of many simple changes. Our inference of the simple from the composite
11470is valid only of self-subsisting things. But the accidents of a state
11471are not self-subsistent. The proof, then, for the necessity of the
11472simple, as the component part of all that is substantial and composite,
11473may prove a failure, and the whole case of this thesis be lost, if we
11474carry the proposition too far, and wish to make it valid of everything
11475that is composite without distinction--as indeed has really now and then
11476happened. Besides, I am here speaking only of the simple, in so far as
11477it is necessarily given in the composite--the latter being capable
11478of solution into the former as its component parts. The proper
11479signification of the word monas (as employed by Leibnitz) ought to
11480relate to the simple, given immediately as simple substance (for
11481example, in consciousness), and not as an element of the composite. As
11482an clement, the term atomus would be more appropriate. And as I wish to
11483prove the existence of simple substances, only in relation to, and
11484as the elements of, the composite, I might term the antithesis of the
11485second Antinomy, transcendental Atomistic. But as this word has long
11486been employed to designate a particular theory of corporeal phenomena
11487(moleculae), and thus presupposes a basis of empirical conceptions, I
11488prefer calling it the dialectical principle of Monadology.
11489
11490
11491ANTITHESIS.
11492
11493Against the assertion of the infinite subdivisibility of matter whose
11494ground of proof is purely mathematical, objections have been alleged by
11495the Monadists. These objections lay themselves open, at first sight,
11496to suspicion, from the fact that they do not recognize the clearest
11497mathematical proofs as propositions relating to the constitution of
11498space, in so far as it is really the formal condition of the possibility
11499of all matter, but regard them merely as inferences from abstract but
11500arbitrary conceptions, which cannot have any application to real things.
11501Just as if it were possible to imagine another mode of intuition than
11502that given in the primitive intuition of space; and just as if its a
11503priori determinations did not apply to everything, the existence of
11504which is possible, from the fact alone of its filling space. If we
11505listen to them, we shall find ourselves required to cogitate, in
11506addition to the mathematical point, which is simple--not, however,
11507a part, but a mere limit of space--physical points, which are indeed
11508likewise simple, but possess the peculiar property, as parts of space,
11509of filling it merely by their aggregation. I shall not repeat here the
11510common and clear refutations of this absurdity, which are to be
11511found everywhere in numbers: every one knows that it is impossible to
11512undermine the evidence of mathematics by mere discursive conceptions; I
11513shall only remark that, if in this case philosophy endeavours to gain
11514an advantage over mathematics by sophistical artifices, it is because
11515it forgets that the discussion relates solely to Phenomena and their
11516conditions. It is not sufficient to find the conception of the simple
11517for the pure conception of the composite, but we must discover for the
11518intuition of the composite (matter), the intuition of the simple. Now
11519this, according to the laws of sensibility, and consequently in the
11520case of objects of sense, is utterly impossible. In the case of a
11521whole composed of substances, which is cogitated solely by the pure
11522understanding, it may be necessary to be in possession of the simple
11523before composition is possible. But this does not hold good of the Totum
11524substantiale phaenomenon, which, as an empirical intuition in space,
11525possesses the necessary property of containing no simple part, for the
11526very reason that no part of space is simple. Meanwhile, the Monadists
11527have been subtle enough to escape from this difficulty, by presupposing
11528intuition and the dynamical relation of substances as the condition of
11529the possibility of space, instead of regarding space as the condition
11530of the possibility of the objects of external intuition, that is, of
11531bodies. Now we have a conception of bodies only as phenomena, and, as
11532such, they necessarily presuppose space as the condition of all external
11533phenomena. The evasion is therefore in vain; as, indeed, we have
11534sufficiently shown in our Aesthetic. If bodies were things in
11535themselves, the proof of the Monadists would be unexceptionable.
11536
11537The second dialectical assertion possesses the peculiarity of
11538having opposed to it a dogmatical proposition, which, among all such
11539sophistical statements, is the only one that undertakes to prove in the
11540case of an object of experience, that which is properly a transcendental
11541idea--the absolute simplicity of substance. The proposition is that the
11542object of the internal sense, the thinking Ego, is an absolute simple
11543substance. Without at present entering upon this subject--as it has been
11544considered at length in a former chapter--I shall merely remark that, if
11545something is cogitated merely as an object, without the addition of any
11546synthetical determination of its intuition--as happens in the case
11547of the bare representation, I--it is certain that no manifold and no
11548composition can be perceived in such a representation. As, moreover, the
11549predicates whereby I cogitate this object are merely intuitions of the
11550internal sense, there cannot be discovered in them anything to prove
11551the existence of a manifold whose parts are external to each other,
11552and, consequently, nothing to prove the existence of real composition.
11553Consciousness, therefore, is so constituted that, inasmuch as the
11554thinking subject is at the same time its own object, it cannot divide
11555itself--although it can divide its inhering determinations. For every
11556object in relation to itself is absolute unity. Nevertheless, if the
11557subject is regarded externally, as an object of intuition, it must, in
11558its character of phenomenon, possess the property of composition. And it
11559must always be regarded in this manner, if we wish to know whether there
11560is or is not contained in it a manifold whose parts are external to each
11561other.
11562
11563
11564
11565THIRD CONFLICT OF THE TRANSCENDENTAL IDEAS.
11566
11567THESIS.
11568
11569Causality according to the laws of nature, is not the only causality
11570operating to originate the phenomena of the world. A causality of
11571freedom is also necessary to account fully for these phenomena.
11572
11573
11574PROOF.
11575
11576Let it be supposed, that there is no other kind of causality than that
11577according to the laws of nature. Consequently, everything that happens
11578presupposes a previous condition, which it follows with absolute
11579certainty, in conformity with a rule. But this previous condition must
11580itself be something that has happened (that has arisen in time, as it
11581did not exist before), for, if it has always been in existence, its
11582consequence or effect would not thus originate for the first time,
11583but would likewise have always existed. The causality, therefore, of a
11584cause, whereby something happens, is itself a thing that has happened.
11585Now this again presupposes, in conformity with the law of nature, a
11586previous condition and its causality, and this another anterior to the
11587former, and so on. If, then, everything happens solely in accordance
11588with the laws of nature, there cannot be any real first beginning of
11589things, but only a subaltern or comparative beginning. There cannot,
11590therefore, be a completeness of series on the side of the causes which
11591originate the one from the other. But the law of nature is that
11592nothing can happen without a sufficient a priori determined cause. The
11593proposition therefore--if all causality is possible only in accordance
11594with the laws of nature--is, when stated in this unlimited and general
11595manner, self-contradictory. It follows that this cannot be the only kind
11596of causality.
11597
11598From what has been said, it follows that a causality must be admitted,
11599by means of which something happens, without its cause being determined
11600according to necessary laws by some other cause preceding. That is to
11601say, there must exist an absolute spontaneity of cause, which of itself
11602originates a series of phenomena which proceeds according to natural
11603laws--consequently transcendental freedom, without which even in the
11604course of nature the succession of phenomena on the side of causes is
11605never complete.
11606
11607
11608ANTITHESIS.
11609
11610There is no such thing as freedom, but everything in the world happens
11611solely according to the laws of nature.
11612
11613
11614PROOF.
11615
11616Granted, that there does exist freedom in the transcendental sense, as a
11617peculiar kind of causality, operating to produce events in the world--a
11618faculty, that is to say, of originating a state, and consequently a
11619series of consequences from that state. In this case, not only the
11620series originated by this spontaneity, but the determination of this
11621spontaneity itself to the production of the series, that is to say, the
11622causality itself must have an absolute commencement, such that nothing
11623can precede to determine this action according to unvarying laws. But
11624every beginning of action presupposes in the acting cause a state of
11625inaction; and a dynamically primal beginning of action presupposes a
11626state, which has no connection--as regards causality--with the preceding
11627state of the cause--which does not, that is, in any wise result from it.
11628Transcendental freedom is therefore opposed to the natural law of cause
11629and effect, and such a conjunction of successive states in effective
11630causes is destructive of the possibility of unity in experience and
11631for that reason not to be found in experience--is consequently a mere
11632fiction of thought.
11633
11634We have, therefore, nothing but nature to which we must look for
11635connection and order in cosmical events. Freedom--independence of the
11636laws of nature--is certainly a deliverance from restraint, but it is
11637also a relinquishing of the guidance of law and rule. For it cannot
11638be alleged that, instead of the laws of nature, laws of freedom may be
11639introduced into the causality of the course of nature. For, if freedom
11640were determined according to laws, it would be no longer freedom,
11641but merely nature. Nature, therefore, and transcendental freedom are
11642distinguishable as conformity to law and lawlessness. The former imposes
11643upon understanding the difficulty of seeking the origin of events ever
11644higher and higher in the series of causes, inasmuch as causality is
11645always conditioned thereby; while it compensates this labour by the
11646guarantee of a unity complete and in conformity with law. The latter, on
11647the contrary, holds out to the understanding the promise of a point
11648of rest in the chain of causes, by conducting it to an unconditioned
11649causality, which professes to have the power of spontaneous origination,
11650but which, in its own utter blindness, deprives it of the guidance of
11651rules, by which alone a completely connected experience is possible.
11652
11653
11654
11655OBSERVATIONS ON THE THIRD ANTINOMY.
11656
11657ON THE THESIS.
11658
11659The transcendental idea of freedom is far from constituting the entire
11660content of the psychological conception so termed, which is for the most
11661part empirical. It merely presents us with the conception of spontaneity
11662of action, as the proper ground for imputing freedom to the cause of a
11663certain class of objects. It is, however, the true stumbling-stone to
11664philosophy, which meets with unconquerable difficulties in the way of
11665its admitting this kind of unconditioned causality. That element in the
11666question of the freedom of the will, which has for so long a time placed
11667speculative reason in such perplexity, is properly only transcendental,
11668and concerns the question, whether there must be held to exist a faculty
11669of spontaneous origination of a series of successive things or states.
11670How such a faculty is possible is not a necessary inquiry; for in the
11671case of natural causality itself, we are obliged to content ourselves
11672with the a priori knowledge that such a causality must be presupposed,
11673although we are quite incapable of comprehending how the being of
11674one thing is possible through the being of another, but must for this
11675information look entirely to experience. Now we have demonstrated this
11676necessity of a free first beginning of a series of phenomena, only in so
11677far as it is required for the comprehension of an origin of the world,
11678all following states being regarded as a succession according to laws
11679of nature alone. But, as there has thus been proved the existence of a
11680faculty which can of itself originate a series in time--although we
11681are unable to explain how it can exist--we feel ourselves authorized to
11682admit, even in the midst of the natural course of events, a beginning,
11683as regards causality, of different successions of phenomena, and at the
11684same time to attribute to all substances a faculty of free action.
11685But we ought in this case not to allow ourselves to fall into a common
11686misunderstanding, and to suppose that, because a successive series in
11687the world can only have a comparatively first beginning--another state
11688or condition of things always preceding--an absolutely first beginning
11689of a series in the course of nature is impossible. For we are not
11690speaking here of an absolutely first beginning in relation to time, but
11691as regards causality alone. When, for example, I, completely of my own
11692free will, and independently of the necessarily determinative influence
11693of natural causes, rise from my chair, there commences with this event,
11694including its material consequences in infinitum, an absolutely
11695new series; although, in relation to time, this event is merely the
11696continuation of a preceding series. For this resolution and act of mine
11697do not form part of the succession of effects in nature, and are not
11698mere continuations of it; on the contrary, the determining causes of
11699nature cease to operate in reference to this event, which certainly
11700succeeds the acts of nature, but does not proceed from them. For
11701these reasons, the action of a free agent must be termed, in regard to
11702causality, if not in relation to time, an absolutely primal beginning of
11703a series of phenomena.
11704
11705The justification of this need of reason to rest upon a free act as
11706the first beginning of the series of natural causes is evident from
11707the fact, that all philosophers of antiquity (with the exception of the
11708Epicurean school) felt themselves obliged, when constructing a theory of
11709the motions of the universe, to accept a prime mover, that is, a freely
11710acting cause, which spontaneously and prior to all other causes evolved
11711this series of states. They always felt the need of going beyond mere
11712nature, for the purpose of making a first beginning comprehensible.
11713
11714
11715
11716ON THE ANTITHESIS.
11717
11718The assertor of the all-sufficiency of nature in regard to causality
11719(transcendental Physiocracy), in opposition to the doctrine of freedom,
11720would defend his view of the question somewhat in the following manner.
11721He would say, in answer to the sophistical arguments of the opposite
11722party: If you do not accept a mathematical first, in relation to time,
11723you have no need to seek a dynamical first, in regard to causality. Who
11724compelled you to imagine an absolutely primal condition of the world,
11725and therewith an absolute beginning of the gradually progressing
11726successions of phenomena--and, as some foundation for this fancy of
11727yours, to set bounds to unlimited nature? Inasmuch as the substances in
11728the world have always existed--at least the unity of experience renders
11729such a supposition quite necessary--there is no difficulty in believing
11730also, that the changes in the conditions of these substances have always
11731existed; and, consequently, that a first beginning, mathematical or
11732dynamical, is by no means required. The possibility of such an infinite
11733derivation, without any initial member from which all the others result,
11734is certainly quite incomprehensible. But, if you are rash enough to
11735deny the enigmatical secrets of nature for this reason, you will find
11736yourselves obliged to deny also the existence of many fundamental
11737properties of natural objects (such as fundamental forces), which you
11738can just as little comprehend; and even the possibility of so simple
11739a conception as that of change must present to you insuperable
11740difficulties. For if experience did not teach you that it was real, you
11741never could conceive a priori the possibility of this ceaseless sequence
11742of being and non-being.
11743
11744But if the existence of a transcendental faculty of freedom is
11745granted--a faculty of originating changes in the world--this faculty
11746must at least exist out of and apart from the world; although it is
11747certainly a bold assumption, that, over and above the complete content
11748of all possible intuitions, there still exists an object which cannot be
11749presented in any possible perception. But, to attribute to substances
11750in the world itself such a faculty, is quite inadmissible; for, in
11751this case; the connection of phenomena reciprocally determining and
11752determined according to general laws, which is termed nature, and along
11753with it the criteria of empirical truth, which enable us to distinguish
11754experience from mere visionary dreaming, would almost entirely
11755disappear. In proximity with such a lawless faculty of freedom, a system
11756of nature is hardly cogitable; for the laws of the latter would be
11757continually subject to the intrusive influences of the former, and
11758the course of phenomena, which would otherwise proceed regularly and
11759uniformly, would become thereby confused and disconnected.
11760
11761
11762
11763FOURTH CONFLICT OF THE TRANSCENDENTAL IDEAS.
11764
11765THESIS.
11766
11767There exists either in, or in connection with the world--either as a
11768part of it, or as the cause of it--an absolutely necessary being.
11769
11770
11771PROOF.
11772
11773The world of sense, as the sum total of all phenomena, contains a series
11774of changes. For, without such a series, the mental representation of
11775the series of time itself, as the condition of the possibility of the
11776sensuous world, could not be presented to us.* But every change stands
11777under its condition, which precedes it in time and renders it necessary.
11778Now the existence of a given condition presupposes a complete series of
11779conditions up to the absolutely unconditioned, which alone is absolutely
11780necessary. It follows that something that is absolutely necessary must
11781exist, if change exists as its consequence. But this necessary thing
11782itself belongs to the sensuous world. For suppose it to exist out of and
11783apart from it, the series of cosmical changes would receive from it a
11784beginning, and yet this necessary cause would not itself belong to
11785the world of sense. But this is impossible. For, as the beginning of a
11786series in time is determined only by that which precedes it in time, the
11787supreme condition of the beginning of a series of changes must exist
11788in the time in which this series itself did not exist; for a beginning
11789supposes a time preceding, in which the thing that begins to be was
11790not in existence. The causality of the necessary cause of changes,
11791and consequently the cause itself, must for these reasons belong
11792to time--and to phenomena, time being possible only as the form of
11793phenomena. Consequently, it cannot be cogitated as separated from the
11794world of sense--the sum total of all phenomena. There is, therefore,
11795contained in the world, something that is absolutely necessary--whether
11796it be the whole cosmical series itself, or only a part of it.
11797
11798
11799 [*Footnote: Objectively, time, as the formal condition of the
11800possibility of change, precedes all changes; but subjectively, and in
11801consciousness, the representation of time, like every other, is given
11802solely by occasion of perception.]
11803
11804
11805
11806ANTITHESIS.
11807
11808An absolutely necessary being does not exist, either in the world, or
11809out of it--as its cause.
11810
11811
11812PROOF.
11813
11814Grant that either the world itself is necessary, or that there is
11815contained in it a necessary existence. Two cases are possible. First,
11816there must either be in the series of cosmical changes a beginning,
11817which is unconditionally necessary, and therefore uncaused--which is at
11818variance with the dynamical law of the determination of all phenomena
11819in time; or, secondly, the series itself is without beginning, and,
11820although contingent and conditioned in all its parts, is nevertheless
11821absolutely necessary and unconditioned as a whole--which is
11822self-contradictory. For the existence of an aggregate cannot be
11823necessary, if no single part of it possesses necessary existence.
11824
11825Grant, on the other band, that an absolutely necessary cause exists out
11826of and apart from the world. This cause, as the highest member in the
11827series of the causes of cosmical changes, must originate or begin* the
11828existence of the latter and their series. In this case it must also
11829begin to act, and its causality would therefore belong to time, and
11830consequently to the sum total of phenomena, that is, to the world.
11831It follows that the cause cannot be out of the world; which is
11832contradictory to the hypothesis. Therefore, neither in the world, nor
11833out of it (but in causal connection with it), does there exist any
11834absolutely necessary being.
11835
11836
11837 [*Footnote: The word begin is taken in two senses. The first is
11838active--the cause being regarded as beginning a series of conditions as
11839its effect (infit). The second is passive--the causality in the cause
11840itself beginning to operate (fit). I reason here from the first to the
11841second.]
11842
11843
11844
11845OBSERVATIONS ON THE FOURTH ANTINOMY.
11846
11847ON THE THESIS.
11848
11849To demonstrate the existence of a necessary being, I cannot be permitted
11850in this place to employ any other than the cosmological argument,
11851which ascends from the conditioned in phenomena to the unconditioned in
11852conception--the unconditioned being considered the necessary condition
11853of the absolute totality of the series. The proof, from the mere idea
11854of a supreme being, belongs to another principle of reason and requires
11855separate discussion.
11856
11857The pure cosmological proof demonstrates the existence of a necessary
11858being, but at the same time leaves it quite unsettled, whether this
11859being is the world itself, or quite distinct from it. To establish
11860the truth of the latter view, principles are requisite, which are not
11861cosmological and do not proceed in the series of phenomena. We
11862should require to introduce into our proof conceptions of contingent
11863beings--regarded merely as objects of the understanding, and also
11864a principle which enables us to connect these, by means of mere
11865conceptions, with a necessary being. But the proper place for all such
11866arguments is a transcendent philosophy, which has unhappily not yet been
11867established.
11868
11869But, if we begin our proof cosmologically, by laying at the foundation
11870of it the series of phenomena, and the regress in it according to
11871empirical laws of causality, we are not at liberty to break off from
11872this mode of demonstration and to pass over to something which is not
11873itself a member of the series. The condition must be taken in exactly
11874the same signification as the relation of the conditioned to its
11875condition in the series has been taken, for the series must conduct us
11876in an unbroken regress to this supreme condition. But if this relation
11877is sensuous, and belongs to the possible empirical employment of
11878understanding, the supreme condition or cause must close the regressive
11879series according to the laws of sensibility and consequently, must
11880belong to the series of time. It follows that this necessary existence
11881must be regarded as the highest member of the cosmical series.
11882
11883Certain philosophers have, nevertheless, allowed themselves the liberty
11884of making such a saltus (metabasis eis allo gonos). From the changes
11885in the world they have concluded their empirical contingency, that
11886is, their dependence on empirically-determined causes, and they thus
11887admitted an ascending series of empirical conditions: and in this they
11888are quite right. But as they could not find in this series any primal
11889beginning or any highest member, they passed suddenly from the empirical
11890conception of contingency to the pure category, which presents us with
11891a series--not sensuous, but intellectual--whose completeness does
11892certainly rest upon the existence of an absolutely necessary cause. Nay,
11893more, this intellectual series is not tied to any sensuous conditions;
11894and is therefore free from the condition of time, which requires it
11895spontaneously to begin its causality in time. But such a procedure is
11896perfectly inadmissible, as will be made plain from what follows.
11897
11898In the pure sense of the categories, that is contingent the
11899contradictory opposite of which is possible. Now we cannot reason from
11900empirical contingency to intellectual. The opposite of that which is
11901changed--the opposite of its state--is actual at another time, and is
11902therefore possible. Consequently, it is not the contradictory opposite
11903of the former state. To be that, it is necessary that, in the same time
11904in which the preceding state existed, its opposite could have existed in
11905its place; but such a cognition is not given us in the mere phenomenon
11906of change. A body that was in motion = A, comes into a state of rest =
11907non-A. Now it cannot be concluded from the fact that a state opposite
11908to the state A follows it, that the contradictory opposite of A is
11909possible; and that A is therefore contingent. To prove this, we should
11910require to know that the state of rest could have existed in the very
11911same time in which the motion took place. Now we know nothing more than
11912that the state of rest was actual in the time that followed the state of
11913motion; consequently, that it was also possible. But motion at one time,
11914and rest at another time, are not contradictorily opposed to each other.
11915It follows from what has been said that the succession of opposite
11916determinations, that is, change, does not demonstrate the fact of
11917contingency as represented in the conceptions of the pure understanding;
11918and that it cannot, therefore, conduct us to the fact of the existence
11919of a necessary being. Change proves merely empirical contingency, that
11920is to say, that the new state could not have existed without a cause,
11921which belongs to the preceding time. This cause--even although it is
11922regarded as absolutely necessary--must be presented to us in time, and
11923must belong to the series of phenomena.
11924
11925
11926ON THE ANTITHESIS.
11927
11928The difficulties which meet us, in our attempt to rise through the
11929series of phenomena to the existence of an absolutely necessary supreme
11930cause, must not originate from our inability to establish the truth of
11931our mere conceptions of the necessary existence of a thing. That is to
11932say, our objections not be ontological, but must be directed against
11933the causal connection with a series of phenomena of a condition which is
11934itself unconditioned. In one word, they must be cosmological and relate
11935to empirical laws. We must show that the regress in the series of
11936causes (in the world of sense) cannot conclude with an empirically
11937unconditioned condition, and that the cosmological argument from the
11938contingency of the cosmical state--a contingency alleged to arise from
11939change--does not justify us in accepting a first cause, that is, a prime
11940originator of the cosmical series.
11941
11942The reader will observe in this antinomy a very remarkable contrast. The
11943very same grounds of proof which established in the thesis the existence
11944of a supreme being, demonstrated in the antithesis--and with equal
11945strictness--the non-existence of such a being. We found, first, that a
11946necessary being exists, because the whole time past contains the series
11947of all conditions, and with it, therefore, the unconditioned (the
11948necessary); secondly, that there does not exist any necessary being,
11949for the same reason, that the whole time past contains the series of
11950all conditions--which are themselves, therefore, in the aggregate,
11951conditioned. The cause of this seeming incongruity is as follows. We
11952attend, in the first argument, solely to the absolute totality of the
11953series of conditions, the one of which determines the other in time, and
11954thus arrive at a necessary unconditioned. In the second, we consider,
11955on the contrary, the contingency of everything that is determined in
11956the series of time--for every event is preceded by a time, in which the
11957condition itself must be determined as conditioned--and thus everything
11958that is unconditioned or absolutely necessary disappears. In both, the
11959mode of proof is quite in accordance with the common procedure of human
11960reason, which often falls into discord with itself, from considering an
11961object from two different points of view. Herr von Mairan regarded
11962the controversy between two celebrated astronomers, which arose from
11963a similar difficulty as to the choice of a proper standpoint, as a
11964phenomenon of sufficient importance to warrant a separate treatise
11965on the subject. The one concluded: the moon revolves on its own axis,
11966because it constantly presents the same side to the earth; the other
11967declared that the moon does not revolve on its own axis, for the same
11968reason. Both conclusions were perfectly correct, according to the point
11969of view from which the motions of the moon were considered.
11970
11971
11972
11973SECTION III. Of the Interest of Reason in these Self-contradictions.
11974
11975We have thus completely before us the dialectical procedure of the
11976cosmological ideas. No possible experience can present us with an object
11977adequate to them in extent. Nay, more, reason itself cannot cogitate
11978them as according with the general laws of experience. And yet they
11979are not arbitrary fictions of thought. On the contrary, reason, in
11980its uninterrupted progress in the empirical synthesis, is necessarily
11981conducted to them, when it endeavours to free from all conditions and
11982to comprehend in its unconditioned totality that which can only be
11983determined conditionally in accordance with the laws of experience.
11984These dialectical propositions are so many attempts to solve four
11985natural and unavoidable problems of reason. There are neither more, nor
11986can there be less, than this number, because there are no other series
11987of synthetical hypotheses, limiting a priori the empirical synthesis.
11988
11989The brilliant claims of reason striving to extend its dominion beyond
11990the limits of experience, have been represented above only in dry
11991formulae, which contain merely the grounds of its pretensions. They
11992have, besides, in conformity with the character of a transcendental
11993philosophy, been freed from every empirical element; although the full
11994splendour of the promises they hold out, and the anticipations they
11995excite, manifests itself only when in connection with empirical
11996cognitions. In the application of them, however, and in the advancing
11997enlargement of the employment of reason, while struggling to rise from
11998the region of experience and to soar to those sublime ideas, philosophy
11999discovers a value and a dignity, which, if it could but make good its
12000assertions, would raise it far above all other departments of human
12001knowledge--professing, as it does, to present a sure foundation for our
12002highest hopes and the ultimate aims of all the exertions of reason.
12003The questions: whether the world has a beginning and a limit to its
12004extension in space; whether there exists anywhere, or perhaps, in my
12005own thinking Self, an indivisible and indestructible unity--or whether
12006nothing but what is divisible and transitory exists; whether I am a free
12007agent, or, like other beings, am bound in the chains of nature and fate;
12008whether, finally, there is a supreme cause of the world, or all our
12009thought and speculation must end with nature and the order of external
12010things--are questions for the solution of which the mathematician would
12011willingly exchange his whole science; for in it there is no satisfaction
12012for the highest aspirations and most ardent desires of humanity. Nay, it
12013may even be said that the true value of mathematics--that pride of human
12014reason--consists in this: that she guides reason to the knowledge of
12015nature--in her greater as well as in her less manifestations--in her
12016beautiful order and regularity--guides her, moreover, to an insight into
12017the wonderful unity of the moving forces in the operations of nature,
12018far beyond the expectations of a philosophy building only on experience;
12019and that she thus encourages philosophy to extend the province of reason
12020beyond all experience, and at the same time provides it with the most
12021excellent materials for supporting its investigations, in so far as
12022their nature admits, by adequate and accordant intuitions.
12023
12024Unfortunately for speculation--but perhaps fortunately for the
12025practical interests of humanity--reason, in the midst of her highest
12026anticipations, finds herself hemmed in by a press of opposite and
12027contradictory conclusions, from which neither her honour nor her safety
12028will permit her to draw back. Nor can she regard these conflicting
12029trains of reasoning with indifference as mere passages at arms, still
12030less can she command peace; for in the subject of the conflict she has a
12031deep interest. There is no other course left open to her than to reflect
12032with herself upon the origin of this disunion in reason--whether it may
12033not arise from a mere misunderstanding. After such an inquiry, arrogant
12034claims would have to be given up on both sides; but the sovereignty
12035of reason over understanding and sense would be based upon a sure
12036foundation.
12037
12038We shall at present defer this radical inquiry and, in the meantime,
12039consider for a little what side in the controversy we should most
12040willingly take, if we were obliged to become partisans at all. As, in
12041this case, we leave out of sight altogether the logical criterion of
12042truth, and merely consult our own interest in reference to the question,
12043these considerations, although inadequate to settle the question of
12044right in either party, will enable us to comprehend how those who have
12045taken part in the struggle, adopt the one view rather than the other--no
12046special insight into the subject, however, having influenced their
12047choice. They will, at the same time, explain to us many other things
12048by the way--for example, the fiery zeal on the one side and the cold
12049maintenance of their cause on the other; why the one party has met with
12050the warmest approbations, and the other has always been repulsed by
12051irreconcilable prejudices.
12052
12053There is one thing, however, that determines the proper point of view,
12054from which alone this preliminary inquiry can be instituted and carried
12055on with the proper completeness--and that is the comparison of the
12056principles from which both sides, thesis and antithesis, proceed. My
12057readers would remark in the propositions of the antithesis a complete
12058uniformity in the mode of thought and a perfect unity of principle. Its
12059principle was that of pure empiricism, not only in the explication
12060of the phenomena in the world, but also in the solution of the
12061transcendental ideas, even of that of the universe itself. The
12062affirmations of the thesis, on the contrary, were based, in addition to
12063the empirical mode of explanation employed in the series of phenomena,
12064on intellectual propositions; and its principles were in so far
12065not simple. I shall term the thesis, in view of its essential
12066characteristic, the dogmatism of pure reason.
12067
12068On the side of Dogmatism, or of the thesis, therefore, in the
12069determination of the cosmological ideas, we find:
12070
120711. A practical interest, which must be very dear to every right-thinking
12072man. That the word has a beginning--that the nature of my thinking self
12073is simple, and therefore indestructible--that I am a free agent, and
12074raised above the compulsion of nature and her laws--and, finally, that
12075the entire order of things, which form the world, is dependent upon a
12076Supreme Being, from whom the whole receives unity and connection--these
12077are so many foundation-stones of morality and religion. The antithesis
12078deprives us of all these supports--or, at least, seems so to deprive us.
12079
120802. A speculative interest of reason manifests itself on this side. For,
12081if we take the transcendental ideas and employ them in the manner which
12082the thesis directs, we can exhibit completely a priori the entire
12083chain of conditions, and understand the derivation of the
12084conditioned--beginning from the unconditioned. This the antithesis does
12085not do; and for this reason does not meet with so welcome a reception.
12086For it can give no answer to our question respecting the conditions of
12087its synthesis--except such as must be supplemented by another question,
12088and so on to infinity. According to it, we must rise from a given
12089beginning to one still higher; every part conducts us to a still smaller
12090one; every event is preceded by another event which is its cause; and
12091the conditions of existence rest always upon other and still higher
12092conditions, and find neither end nor basis in some self-subsistent thing
12093as the primal being.
12094
120953. This side has also the advantage of popularity; and this constitutes
12096no small part of its claim to favour. The common understanding does not
12097find the least difficulty in the idea of the unconditioned beginning of
12098all synthesis--accustomed, as it is, rather to follow our consequences
12099than to seek for a proper basis for cognition. In the conception of an
12100absolute first, moreover--the possibility of which it does not inquire
12101into--it is highly gratified to find a firmly-established point
12102of departure for its attempts at theory; while in the restless and
12103continuous ascent from the conditioned to the condition, always with one
12104foot in the air, it can find no satisfaction.
12105
12106On the side of the antithesis, or Empiricism, in the determination of
12107the cosmological ideas:
12108
121091. We cannot discover any such practical interest arising from pure
12110principles of reason as morality and religion present. On the contrary,
12111pure empiricism seems to empty them of all their power and influence.
12112If there does not exist a Supreme Being distinct from the world--if the
12113world is without beginning, consequently without a Creator--if our wills
12114are not free, and the soul is divisible and subject to corruption just
12115like matter--the ideas and principles of morality lose all validity and
12116fall with the transcendental ideas which constituted their theoretical
12117support.
12118
121192. But empiricism, in compensation, holds out to reason, in its
12120speculative interests, certain important advantages, far exceeding any
12121that the dogmatist can promise us. For, when employed by the empiricist,
12122understanding is always upon its proper ground of investigation--the
12123field of possible experience, the laws of which it can explore, and thus
12124extend its cognition securely and with clear intelligence without being
12125stopped by limits in any direction. Here can it and ought it to find and
12126present to intuition its proper object--not only in itself, but in all
12127its relations; or, if it employ conceptions, upon this ground it can
12128always present the corresponding images in clear and unmistakable
12129intuitions. It is quite unnecessary for it to renounce the guidance of
12130nature, to attach itself to ideas, the objects of which it cannot know;
12131because, as mere intellectual entities, they cannot be presented in
12132any intuition. On the contrary, it is not even permitted to abandon
12133its proper occupation, under the pretence that it has been brought to
12134a conclusion (for it never can be), and to pass into the region of
12135idealizing reason and transcendent conceptions, which it is not required
12136to observe and explore the laws of nature, but merely to think and to
12137imagine--secure from being contradicted by facts, because they have not
12138been called as witnesses, but passed by, or perhaps subordinated to the
12139so-called higher interests and considerations of pure reason.
12140
12141Hence the empiricist will never allow himself to accept any epoch of
12142nature for the first--the absolutely primal state; he will not believe
12143that there can be limits to his outlook into her wide domains, nor pass
12144from the objects of nature, which he can satisfactorily explain by
12145means of observation and mathematical thought--which he can determine
12146synthetically in intuition, to those which neither sense nor imagination
12147can ever present in concreto; he will not concede the existence of a
12148faculty in nature, operating independently of the laws of nature--a
12149concession which would introduce uncertainty into the procedure of the
12150understanding, which is guided by necessary laws to the observation of
12151phenomena; nor, finally, will he permit himself to seek a cause beyond
12152nature, inasmuch as we know nothing but it, and from it alone receive an
12153objective basis for all our conceptions and instruction in the unvarying
12154laws of things.
12155
12156In truth, if the empirical philosopher had no other purpose in the
12157establishment of his antithesis than to check the presumption of a
12158reason which mistakes its true destination, which boasts of its insight
12159and its knowledge, just where all insight and knowledge cease to
12160exist, and regards that which is valid only in relation to a practical
12161interest, as an advancement of the speculative interests of the mind
12162(in order, when it is convenient for itself, to break the thread of our
12163physical investigations, and, under pretence of extending our cognition,
12164connect them with transcendental ideas, by means of which we really know
12165only that we know nothing)--if, I say, the empiricist rested satisfied
12166with this benefit, the principle advanced by him would be a maxim
12167recommending moderation in the pretensions of reason and modesty in its
12168affirmations, and at the same time would direct us to the right mode
12169of extending the province of the understanding, by the help of the only
12170true teacher, experience. In obedience to this advice, intellectual
12171hypotheses and faith would not be called in aid of our practical
12172interests; nor should we introduce them under the pompous titles of
12173science and insight. For speculative cognition cannot find an objective
12174basis any other where than in experience; and, when we overstep its
12175limits our synthesis, which requires ever new cognitions independent of
12176experience, has no substratum of intuition upon which to build.
12177
12178But if--as often happens--empiricism, in relation to ideas, becomes
12179itself dogmatic and boldly denies that which is above the sphere of its
12180phenomenal cognition, it falls itself into the error of intemperance--an
12181error which is here all the more reprehensible, as thereby the practical
12182interest of reason receives an irreparable injury.
12183
12184And this constitutes the opposition between Epicureanism* and Platonism.
12185
12186
12187 [*Footnote: It is, however, still a matter of doubt whether Epicurus
12188ever propounded these principles as directions for the objective
12189employment of the understanding. If, indeed, they were nothing more than
12190maxims for the speculative exercise of reason, he gives evidence therein
12191a more genuine philosophic spirit than any of the philosophers of
12192antiquity. That, in the explanation of phenomena, we must proceed as
12193if the field of inquiry had neither limits in space nor commencement
12194in time; that we must be satisfied with the teaching of experience in
12195reference to the material of which the world is posed; that we must not
12196look for any other mode of the origination of events than that which is
12197determined by the unalterable laws of nature; and finally, that we not
12198employ the hypothesis of a cause distinct from the world to account for
12199a phenomenon or for the world itself--are principles for the extension
12200of speculative philosophy, and the discovery of the true sources of the
12201principles of morals, which, however little conformed to in the present
12202day, are undoubtedly correct. At the same time, any one desirous of
12203ignoring, in mere speculation, these dogmatical propositions, need not
12204for that reason be accused of denying them.]
12205
12206
12207Both Epicurus and Plato assert more in their systems than they know. The
12208former encourages and advances science--although to the prejudice of
12209the practical; the latter presents us with excellent principles for the
12210investigation of the practical, but, in relation to everything regarding
12211which we can attain to speculative cognition, permits reason to append
12212idealistic explanations of natural phenomena, to the great injury of
12213physical investigation.
12214
122153. In regard to the third motive for the preliminary choice of a party
12216in this war of assertions, it seems very extraordinary that empiricism
12217should be utterly unpopular. We should be inclined to believe that the
12218common understanding would receive it with pleasure--promising as it
12219does to satisfy it without passing the bounds of experience and its
12220connected order; while transcendental dogmatism obliges it to rise to
12221conceptions which far surpass the intelligence and ability of the most
12222practised thinkers. But in this, in truth, is to be found its real
12223motive. For the common understanding thus finds itself in a situation
12224where not even the most learned can have the advantage of it. If it
12225understands little or nothing about these transcendental conceptions, no
12226one can boast of understanding any more; and although it may not express
12227itself in so scholastically correct a manner as others, it can busy
12228itself with reasoning and arguments without end, wandering among mere
12229ideas, about which one can always be very eloquent, because we know
12230nothing about them; while, in the observation and investigation of
12231nature, it would be forced to remain dumb and to confess its utter
12232ignorance. Thus indolence and vanity form of themselves strong
12233recommendations of these principles. Besides, although it is a hard
12234thing for a philosopher to assume a principle, of which he can give to
12235himself no reasonable account, and still more to employ conceptions, the
12236objective reality of which cannot be established, nothing is more usual
12237with the common understanding. It wants something which will allow it
12238to go to work with confidence. The difficulty of even comprehending
12239a supposition does not disquiet it, because--not knowing what
12240comprehending means--it never even thinks of the supposition it may be
12241adopting as a principle; and regards as known that with which it
12242has become familiar from constant use. And, at last, all speculative
12243interests disappear before the practical interests which it holds dear;
12244and it fancies that it understands and knows what its necessities
12245and hopes incite it to assume or to believe. Thus the empiricism of
12246transcendentally idealizing reason is robbed of all popularity; and,
12247however prejudicial it may be to the highest practical principles, there
12248is no fear that it will ever pass the limits of the schools, or acquire
12249any favour or influence in society or with the multitude.
12250
12251Human reason is by nature architectonic. That is to say, it regards all
12252cognitions as parts of a possible system, and hence accepts only such
12253principles as at least do not incapacitate a cognition to which we may
12254have attained from being placed along with others in a general system.
12255But the propositions of the antithesis are of a character which renders
12256the completion of an edifice of cognitions impossible. According to
12257these, beyond one state or epoch of the world there is always to be
12258found one more ancient; in every part always other parts themselves
12259divisible; preceding every event another, the origin of which
12260must itself be sought still higher; and everything in existence is
12261conditioned, and still not dependent on an unconditioned and primal
12262existence. As, therefore, the antithesis will not concede the existence
12263of a first beginning which might be available as a foundation, a
12264complete edifice of cognition, in the presence of such hypothesis, is
12265utterly impossible. Thus the architectonic interest of reason, which
12266requires a unity--not empirical, but a priori and rational--forms a
12267natural recommendation for the assertions of the thesis in our antinomy.
12268
12269But if any one could free himself entirely from all considerations
12270of interest, and weigh without partiality the assertions of reason,
12271attending only to their content, irrespective of the consequences which
12272follow from them; such a person, on the supposition that he knew no
12273other way out of the confusion than to settle the truth of one or
12274other of the conflicting doctrines, would live in a state of continual
12275hesitation. Today, he would feel convinced that the human will is free;
12276to-morrow, considering the indissoluble chain of nature, he would look
12277on freedom as a mere illusion and declare nature to be all-in-all. But,
12278if he were called to action, the play of the merely speculative reason
12279would disappear like the shapes of a dream, and practical interest would
12280dictate his choice of principles. But, as it well befits a reflective
12281and inquiring being to devote certain periods of time to the examination
12282of its own reason--to divest itself of all partiality, and frankly to
12283communicate its observations for the judgement and opinion of others; so
12284no one can be blamed for, much less prevented from, placing both
12285parties on their trial, with permission to end themselves, free
12286from intimidation, before intimidation, before a sworn jury of equal
12287condition with themselves--the condition of weak and fallible men.
12288
12289
12290
12291SECTION IV. Of the necessity imposed upon Pure Reason of
12292presenting a Solution of its Transcendental Problems.
12293
12294To avow an ability to solve all problems and to answer all questions
12295would be a profession certain to convict any philosopher of extravagant
12296boasting and self-conceit, and at once to destroy the confidence that
12297might otherwise have been reposed in him. There are, however, sciences
12298so constituted that every question arising within their sphere must
12299necessarily be capable of receiving an answer from the knowledge already
12300possessed, for the answer must be received from the same sources whence
12301the question arose. In such sciences it is not allowable to excuse
12302ourselves on the plea of necessary and unavoidable ignorance; a solution
12303is absolutely requisite. The rule of right and wrong must help us to the
12304knowledge of what is right or wrong in all possible cases; otherwise,
12305the idea of obligation or duty would be utterly null, for we cannot have
12306any obligation to that which we cannot know. On the other hand, in our
12307investigations of the phenomena of nature, much must remain uncertain,
12308and many questions continue insoluble; because what we know of nature
12309is far from being sufficient to explain all the phenomena that are
12310presented to our observation. Now the question is: Whether there is in
12311transcendental philosophy any question, relating to an object presented
12312to pure reason, which is unanswerable by this reason; and whether we
12313must regard the subject of the question as quite uncertain, so far as
12314our knowledge extends, and must give it a place among those subjects, of
12315which we have just so much conception as is sufficient to enable us
12316to raise a question--faculty or materials failing us, however, when we
12317attempt an answer.
12318
12319Now I maintain that, among all speculative cognition, the peculiarity of
12320transcendental philosophy is that there is no question, relating to an
12321object presented to pure reason, which is insoluble by this reason; and
12322that the profession of unavoidable ignorance--the problem being alleged
12323to be beyond the reach of our faculties--cannot free us from the
12324obligation to present a complete and satisfactory answer. For the very
12325conception which enables us to raise the question must give us the power
12326of answering it; inasmuch as the object, as in the case of right and
12327wrong, is not to be discovered out of the conception.
12328
12329But, in transcendental philosophy, it is only the cosmological questions
12330to which we can demand a satisfactory answer in relation to the
12331constitution of their object; and the philosopher is not permitted to
12332avail himself of the pretext of necessary ignorance and impenetrable
12333obscurity. These questions relate solely to the cosmological ideas. For
12334the object must be given in experience, and the question relates to the
12335adequateness of the object to an idea. If the object is transcendental
12336and therefore itself unknown; if the question, for example, is whether
12337the object--the something, the phenomenon of which (internal--in
12338ourselves) is thought--that is to say, the soul, is in itself a simple
12339being; or whether there is a cause of all things, which is absolutely
12340necessary--in such cases we are seeking for our idea an object, of which
12341we may confess that it is unknown to us, though we must not on that
12342account assert that it is impossible.* The cosmological ideas alone
12343posses the peculiarity that we can presuppose the object of them and the
12344empirical synthesis requisite for the conception of that object to be
12345given; and the question, which arises from these ideas, relates merely
12346to the progress of this synthesis, in so far as it must contain absolute
12347totality--which, however, is not empirical, as it cannot be given in any
12348experience. Now, as the question here is solely in regard to a thing as
12349the object of a possible experience and not as a thing in itself, the
12350answer to the transcendental cosmological question need not be sought
12351out of the idea, for the question does not regard an object in itself.
12352The question in relation to a possible experience is not, "What can be
12353given in an experience in concreto" but "what is contained in the idea,
12354to which the empirical synthesis must approximate." The question must
12355therefore be capable of solution from the idea alone. For the idea is
12356a creation of reason itself, which therefore cannot disclaim the
12357obligation to answer or refer us to the unknown object.
12358
12359
12360 [*Footnote: The question, "What is the constitution of a transcendental
12361object?" is unanswerable--we are unable to say what it is; but we can
12362perceive that the question itself is nothing; because it does not relate
12363to any object that can be presented to us. For this reason, we must
12364consider all the questions raised in transcendental psychology as
12365answerable and as really answered; for they relate to the transcendental
12366subject of all internal phenomena, which is not itself phenomenon and
12367consequently not given as an object, in which, moreover, none of
12368the categories--and it is to them that the question is properly
12369directed--find any conditions of its application. Here, therefore, is a
12370case where no answer is the only proper answer. For a question regarding
12371the constitution of a something which cannot be cogitated by any
12372determined predicate, being completely beyond the sphere of objects and
12373experience, is perfectly null and void.]
12374
12375
12376It is not so extraordinary, as it at first sight appears, that a science
12377should demand and expect satisfactory answers to all the questions that
12378may arise within its own sphere (questiones domesticae), although, up to
12379a certain time, these answers may not have been discovered. There are,
12380in addition to transcendental philosophy, only two pure sciences
12381of reason; the one with a speculative, the other with a practical
12382content--pure mathematics and pure ethics. Has any one ever heard
12383it alleged that, from our complete and necessary ignorance of the
12384conditions, it is uncertain what exact relation the diameter of a circle
12385bears to the circle in rational or irrational numbers? By the former
12386the sum cannot be given exactly, by the latter only approximately; and
12387therefore we decide that the impossibility of a solution of the question
12388is evident. Lambert presented us with a demonstration of this. In the
12389general principles of morals there can be nothing uncertain, for the
12390propositions are either utterly without meaning, or must originate
12391solely in our rational conceptions. On the other hand, there must be
12392in physical science an infinite number of conjectures, which can never
12393become certainties; because the phenomena of nature are not given as
12394objects dependent on our conceptions. The key to the solution of such
12395questions cannot, therefore, be found in our conceptions, or in pure
12396thought, but must lie without us and for that reason is in many cases
12397not to be discovered; and consequently a satisfactory explanation cannot
12398be expected. The questions of transcendental analytic, which relate to
12399the deduction of our pure cognition, are not to be regarded as of the
12400same kind as those mentioned above; for we are not at present treating
12401of the certainty of judgements in relation to the origin of our
12402conceptions, but only of that certainty in relation to objects.
12403
12404We cannot, therefore, escape the responsibility of at least a critical
12405solution of the questions of reason, by complaints of the limited nature
12406of our faculties, and the seemingly humble confession that it is beyond
12407the power of our reason to decide, whether the world has existed from
12408all eternity or had a beginning--whether it is infinitely extended, or
12409enclosed within certain limits--whether anything in the world is simple,
12410or whether everything must be capable of infinite divisibility--whether
12411freedom can originate phenomena, or whether everything is absolutely
12412dependent on the laws and order of nature--and, finally, whether there
12413exists a being that is completely unconditioned and necessary, or
12414whether the existence of everything is conditioned and consequently
12415dependent on something external to itself, and therefore in its own
12416nature contingent. For all these questions relate to an object, which
12417can be given nowhere else than in thought. This object is the absolutely
12418unconditioned totality of the synthesis of phenomena. If the conceptions
12419in our minds do not assist us to some certain result in regard to these
12420problems, we must not defend ourselves on the plea that the object
12421itself remains hidden from and unknown to us. For no such thing or
12422object can be given--it is not to be found out of the idea in our minds.
12423We must seek the cause of our failure in our idea itself, which is an
12424insoluble problem and in regard to which we obstinately assume that
12425there exists a real object corresponding and adequate to it. A clear
12426explanation of the dialectic which lies in our conception, will very
12427soon enable us to come to a satisfactory decision in regard to such a
12428question.
12429
12430The pretext that we are unable to arrive at certainty in regard to these
12431problems may be met with this question, which requires at least a plain
12432answer: "From what source do the ideas originate, the solution of which
12433involves you in such difficulties? Are you seeking for an explanation
12434of certain phenomena; and do you expect these ideas to give you the
12435principles or the rules of this explanation?" Let it be granted, that
12436all nature was laid open before you; that nothing was hid from your
12437senses and your consciousness. Still, you could not cognize in concreto
12438the object of your ideas in any experience. For what is demanded is not
12439only this full and complete intuition, but also a complete synthesis and
12440the consciousness of its absolute totality; and this is not possible by
12441means of any empirical cognition. It follows that your question--your
12442idea--is by no means necessary for the explanation of any phenomenon;
12443and the idea cannot have been in any sense given by the object itself.
12444For such an object can never be presented to us, because it cannot be
12445given by any possible experience. Whatever perceptions you may attain
12446to, you are still surrounded by conditions--in space, or in time--and
12447you cannot discover anything unconditioned; nor can you decide whether
12448this unconditioned is to be placed in an absolute beginning of the
12449synthesis, or in an absolute totality of the series without beginning.
12450A whole, in the empirical signification of the term, is always merely
12451comparative. The absolute whole of quantity (the universe), of division,
12452of derivation, of the condition of existence, with the question--whether
12453it is to be produced by finite or infinite synthesis, no possible
12454experience can instruct us concerning. You will not, for example, be
12455able to explain the phenomena of a body in the least degree better,
12456whether you believe it to consist of simple, or of composite parts;
12457for a simple phenomenon--and just as little an infinite series of
12458composition--can never be presented to your perception. Phenomena
12459require and admit of explanation, only in so far as the conditions of
12460that explanation are given in perception; but the sum total of that
12461which is given in phenomena, considered as an absolute whole, is itself
12462a perception--and we cannot therefore seek for explanations of this
12463whole beyond itself, in other perceptions. The explanation of this whole
12464is the proper object of the transcendental problems of pure reason.
12465
12466Although, therefore, the solution of these problems is unattainable
12467through experience, we must not permit ourselves to say that it is
12468uncertain how the object of our inquiries is constituted. For the object
12469is in our own mind and cannot be discovered in experience; and we have
12470only to take care that our thoughts are consistent with each other,
12471and to avoid falling into the amphiboly of regarding our idea as a
12472representation of an object empirically given, and therefore to be
12473cognized according to the laws of experience. A dogmatical solution is
12474therefore not only unsatisfactory but impossible. The critical solution,
12475which may be a perfectly certain one, does not consider the question
12476objectively, but proceeds by inquiring into the basis of the cognition
12477upon which the question rests.
12478
12479
12480
12481SECTION V. Sceptical Exposition of the Cosmological Problems
12482presented in the four Transcendental Ideas.
12483
12484We should be quite willing to desist from the demand of a dogmatical
12485answer to our questions, if we understood beforehand that, be the answer
12486what it may, it would only serve to increase our ignorance, to throw
12487us from one incomprehensibility into another, from one obscurity
12488into another still greater, and perhaps lead us into irreconcilable
12489contradictions. If a dogmatical affirmative or negative answer is
12490demanded, is it at all prudent to set aside the probable grounds of
12491a solution which lie before us and to take into consideration what
12492advantage we shall gain, if the answer is to favour the one side or the
12493other? If it happens that in both cases the answer is mere nonsense,
12494we have in this an irresistible summons to institute a critical
12495investigation of the question, for the purpose of discovering whether
12496it is based on a groundless presupposition and relates to an idea, the
12497falsity of which would be more easily exposed in its application and
12498consequences than in the mere representation of its content. This is the
12499great utility of the sceptical mode of treating the questions addressed
12500by pure reason to itself. By this method we easily rid ourselves of
12501the confusions of dogmatism, and establish in its place a temperate
12502criticism, which, as a genuine cathartic, will successfully remove
12503the presumptuous notions of philosophy and their consequence--the vain
12504pretension to universal science.
12505
12506If, then, I could understand the nature of a cosmological idea and
12507perceive, before I entered on the discussion of the subject at all,
12508that, whatever side of the question regarding the unconditioned of the
12509regressive synthesis of phenomena it favoured--it must either be too
12510great or too small for every conception of the understanding--I would
12511be able to comprehend how the idea, which relates to an object of
12512experience--an experience which must be adequate to and in accordance
12513with a possible conception of the understanding--must be completely void
12514and without significance, inasmuch as its object is inadequate, consider
12515it as we may. And this is actually the case with all cosmological
12516conceptions, which, for the reason above mentioned, involve reason, so
12517long as it remains attached to them, in an unavoidable antinomy. For
12518suppose:
12519
12520First, that the world has no beginning--in this case it is too large
12521for our conception; for this conception, which consists in a successive
12522regress, cannot overtake the whole eternity that has elapsed. Grant
12523that it has a beginning, it is then too small for the conception of
12524the understanding. For, as a beginning presupposes a time preceding, it
12525cannot be unconditioned; and the law of the empirical employment of the
12526understanding imposes the necessity of looking for a higher condition of
12527time; and the world is, therefore, evidently too small for this law.
12528
12529The same is the case with the double answer to the question regarding
12530the extent, in space, of the world. For, if it is infinite and
12531unlimited, it must be too large for every possible empirical conception.
12532If it is finite and limited, we have a right to ask: "What determines
12533these limits?" Void space is not a self-subsistent correlate of things,
12534and cannot be a final condition--and still less an empirical condition,
12535forming a part of a possible experience. For how can we have any
12536experience or perception of an absolute void? But the absolute totality
12537of the empirical synthesis requires that the unconditioned be an
12538empirical conception. Consequently, a finite world is too small for our
12539conception.
12540
12541Secondly, if every phenomenon (matter) in space consists of an infinite
12542number of parts, the regress of the division is always too great for our
12543conception; and if the division of space must cease with some member
12544of the division (the simple), it is too small for the idea of the
12545unconditioned. For the member at which we have discontinued our division
12546still admits a regress to many more parts contained in the object.
12547
12548Thirdly, suppose that every event in the world happens in accordance
12549with the laws of nature; the causality of a cause must itself be
12550an event and necessitates a regress to a still higher cause, and
12551consequently the unceasing prolongation of the series of conditions
12552a parte priori. Operative nature is therefore too large for every
12553conception we can form in the synthesis of cosmical events.
12554
12555If we admit the existence of spontaneously produced events, that is, of
12556free agency, we are driven, in our search for sufficient reasons, on an
12557unavoidable law of nature and are compelled to appeal to the empirical
12558law of causality, and we find that any such totality of connection in
12559our synthesis is too small for our necessary empirical conception.
12560
12561Fourthly, if we assume the existence of an absolutely necessary
12562being--whether it be the world or something in the world, or the cause
12563of the world--we must place it in a time at an infinite distance from
12564any given moment; for, otherwise, it must be dependent on some other and
12565higher existence. Such an existence is, in this case, too large for our
12566empirical conception, and unattainable by the continued regress of any
12567synthesis.
12568
12569But if we believe that everything in the world--be it condition or
12570conditioned--is contingent; every given existence is too small for our
12571conception. For in this case we are compelled to seek for some other
12572existence upon which the former depends.
12573
12574We have said that in all these cases the cosmological idea is either
12575too great or too small for the empirical regress in a synthesis, and
12576consequently for every possible conception of the understanding. Why did
12577we not express ourselves in a manner exactly the reverse of this and,
12578instead of accusing the cosmological idea of over stepping or of falling
12579short of its true aim, possible experience, say that, in the first case,
12580the empirical conception is always too small for the idea, and in the
12581second too great, and thus attach the blame of these contradictions to
12582the empirical regress? The reason is this. Possible experience can alone
12583give reality to our conceptions; without it a conception is merely an
12584idea, without truth or relation to an object. Hence a possible empirical
12585conception must be the standard by which we are to judge whether an
12586idea is anything more than an idea and fiction of thought, or whether it
12587relates to an object in the world. If we say of a thing that in
12588relation to some other thing it is too large or too small, the former is
12589considered as existing for the sake of the latter, and requiring to
12590be adapted to it. Among the trivial subjects of discussion in the old
12591schools of dialectics was this question: "If a ball cannot pass through
12592a hole, shall we say that the ball is too large or the hole too small?"
12593In this case it is indifferent what expression we employ; for we do
12594not know which exists for the sake of the other. On the other hand, we
12595cannot say: "The man is too long for his coat"; but: "The coat is too
12596short for the man."
12597
12598We are thus led to the well-founded suspicion that the cosmological
12599ideas, and all the conflicting sophistical assertions connected with
12600them, are based upon a false and fictitious conception of the mode in
12601which the object of these ideas is presented to us; and this suspicion
12602will probably direct us how to expose the illusion that has so long led
12603us astray from the truth.
12604
12605
12606
12607SECTION VI. Transcendental Idealism as the Key to theSolution
12608of Pure Cosmological Dialectic.
12609
12610In the transcendental aesthetic we proved that everything intuited in
12611space and time, all objects of a possible experience, are nothing but
12612phenomena, that is, mere representations; and that these, as
12613presented to us--as extended bodies, or as series of changes--have no
12614self-subsistent existence apart from human thought. This doctrine I
12615call Transcendental Idealism.* The realist in the transcendental
12616sense regards these modifications of our sensibility, these mere
12617representations, as things subsisting in themselves.
12618
12619
12620 [*Footnote: I have elsewhere termed this theory formal idealism, to
12621distinguish it from material idealism, which doubts or denies the
12622existence of external things. To avoid ambiguity, it seems advisable in
12623many cases to employ this term instead of that mentioned in the text.]
12624
12625
12626It would be unjust to accuse us of holding the long-decried theory of
12627empirical idealism, which, while admitting the reality of space, denies,
12628or at least doubts, the existence of bodies extended in it, and thus
12629leaves us without a sufficient criterion of reality and illusion. The
12630supporters of this theory find no difficulty in admitting the reality of
12631the phenomena of the internal sense in time; nay, they go the length
12632of maintaining that this internal experience is of itself a sufficient
12633proof of the real existence of its object as a thing in itself.
12634
12635Transcendental idealism allows that the objects of external
12636intuition--as intuited in space, and all changes in time--as represented
12637by the internal sense, are real. For, as space is the form of that
12638intuition which we call external, and, without objects in space, no
12639empirical representation could be given us, we can and ought to regard
12640extended bodies in it as real. The case is the same with representations
12641in time. But time and space, with all phenomena therein, are not in
12642themselves things. They are nothing but representations and cannot exist
12643out of and apart from the mind. Nay, the sensuous internal intuition of
12644the mind (as the object of consciousness), the determination of which
12645is represented by the succession of different states in time, is not
12646the real, proper self, as it exists in itself--not the transcendental
12647subject--but only a phenomenon, which is presented to the sensibility of
12648this, to us, unknown being. This internal phenomenon cannot be admitted
12649to be a self-subsisting thing; for its condition is time, and time
12650cannot be the condition of a thing in itself. But the empirical truth
12651of phenomena in space and time is guaranteed beyond the possibility of
12652doubt, and sufficiently distinguished from the illusion of dreams
12653or fancy--although both have a proper and thorough connection in an
12654experience according to empirical laws. The objects of experience then
12655are not things in themselves, but are given only in experience, and have
12656no existence apart from and independently of experience. That there may
12657be inhabitants in the moon, although no one has ever observed them, must
12658certainly be admitted; but this assertion means only, that we may in the
12659possible progress of experience discover them at some future time. For
12660that which stands in connection with a perception according to the
12661laws of the progress of experience is real. They are therefore really
12662existent, if they stand in empirical connection with my actual or real
12663consciousness, although they are not in themselves real, that is, apart
12664from the progress of experience.
12665
12666There is nothing actually given--we can be conscious of nothing as
12667real, except a perception and the empirical progression from it to other
12668possible perceptions. For phenomena, as mere representations, are real
12669only in perception; and perception is, in fact, nothing but the reality
12670of an empirical representation, that is, a phenomenon. To call a
12671phenomenon a real thing prior to perception means either that we must
12672meet with this phenomenon in the progress of experience, or it means
12673nothing at all. For I can say only of a thing in itself that it exists
12674without relation to the senses and experience. But we are speaking here
12675merely of phenomena in space and time, both of which are determinations
12676of sensibility, and not of things in themselves. It follows that
12677phenomena are not things in themselves, but are mere representations,
12678which if not given in us--in perception--are non-existent.
12679
12680The faculty of sensuous intuition is properly a receptivity--a capacity
12681of being affected in a certain manner by representations, the relation
12682of which to each other is a pure intuition of space and time--the pure
12683forms of sensibility. These representations, in so far as they are
12684connected and determinable in this relation (in space and time)
12685according to laws of the unity of experience, are called objects. The
12686non-sensuous cause of these representations is completely unknown to us
12687and hence cannot be intuited as an object. For such an object could not
12688be represented either in space or in time; and without these conditions
12689intuition or representation is impossible. We may, at the same time,
12690term the non-sensuous cause of phenomena the transcendental object--but
12691merely as a mental correlate to sensibility, considered as a
12692receptivity. To this transcendental object we may attribute the whole
12693connection and extent of our possible perceptions, and say that it is
12694given and exists in itself prior to all experience. But the phenomena,
12695corresponding to it, are not given as things in themselves, but in
12696experience alone. For they are mere representations, receiving from
12697perceptions alone significance and relation to a real object, under
12698the condition that this or that perception--indicating an object--is in
12699complete connection with all others in accordance with the rules of the
12700unity of experience. Thus we can say: "The things that really existed
12701in past time are given in the transcendental object of experience." But
12702these are to me real objects, only in so far as I can represent to my
12703own mind, that a regressive series of possible perceptions--following
12704the indications of history, or the footsteps of cause and effect--in
12705accordance with empirical laws--that, in one word, the course of the
12706world conducts us to an elapsed series of time as the condition of the
12707present time. This series in past time is represented as real, not in
12708itself, but only in connection with a possible experience. Thus, when
12709I say that certain events occurred in past time, I merely assert the
12710possibility of prolonging the chain of experience, from the present
12711perception, upwards to the conditions that determine it according to
12712time.
12713
12714If I represent to myself all objects existing in all space and time, I
12715do not thereby place these in space and time prior to all experience; on
12716the contrary, such a representation is nothing more than the notion of
12717a possible experience, in its absolute completeness. In experience alone
12718are those objects, which are nothing but representations, given. But,
12719when I say they existed prior to my experience, this means only that
12720I must begin with the perception present to me and follow the track
12721indicated until I discover them in some part or region of experience.
12722The cause of the empirical condition of this progression--and
12723consequently at what member therein I must stop, and at what point
12724in the regress I am to find this member--is transcendental, and hence
12725necessarily incognizable. But with this we have not to do; our concern
12726is only with the law of progression in experience, in which objects,
12727that is, phenomena, are given. It is a matter of indifference, whether
12728I say, "I may in the progress of experience discover stars, at a hundred
12729times greater distance than the most distant of those now visible," or,
12730"Stars at this distance may be met in space, although no one has,
12731or ever will discover them." For, if they are given as things in
12732themselves, without any relation to possible experience, they are for me
12733non-existent, consequently, are not objects, for they are not contained
12734in the regressive series of experience. But, if these phenomena must be
12735employed in the construction or support of the cosmological idea of an
12736absolute whole, and when we are discussing a question that oversteps the
12737limits of possible experience, the proper distinction of the different
12738theories of the reality of sensuous objects is of great importance,
12739in order to avoid the illusion which must necessarily arise from the
12740misinterpretation of our empirical conceptions.
12741
12742
12743
12744SECTION VII. Critical Solution of the Cosmological Problem.
12745
12746The antinomy of pure reason is based upon the following dialectical
12747argument: "If that which is conditioned is given, the whole series
12748of its conditions is also given; but sensuous objects are given as
12749conditioned; consequently..." This syllogism, the major of which seems
12750so natural and evident, introduces as many cosmological ideas as there
12751are different kinds of conditions in the synthesis of phenomena, in
12752so far as these conditions constitute a series. These ideas require
12753absolute totality in the series, and thus place reason in inextricable
12754embarrassment. Before proceeding to expose the fallacy in this
12755dialectical argument, it will be necessary to have a correct
12756understanding of certain conceptions that appear in it.
12757
12758In the first place, the following proposition is evident, and
12759indubitably certain: "If the conditioned is given, a regress in the
12760series of all its conditions is thereby imperatively required." For the
12761very conception of a conditioned is a conception of something related
12762to a condition, and, if this condition is itself conditioned, to
12763another condition--and so on through all the members of the series.
12764This proposition is, therefore, analytical and has nothing to fear
12765from transcendental criticism. It is a logical postulate of reason:
12766to pursue, as far as possible, the connection of a conception with its
12767conditions.
12768
12769If, in the second place, both the conditioned and the condition are
12770things in themselves, and if the former is given, not only is the
12771regress to the latter requisite, but the latter is really given with
12772the former. Now, as this is true of all the members of the series, the
12773entire series of conditions, and with them the unconditioned, is at the
12774same time given in the very fact of the conditioned, the existence of
12775which is possible only in and through that series, being given. In
12776this case, the synthesis of the conditioned with its condition, is a
12777synthesis of the understanding merely, which represents things as they
12778are, without regarding whether and how we can cognize them. But if
12779I have to do with phenomena, which, in their character of mere
12780representations, are not given, if I do not attain to a cognition of
12781them (in other words, to themselves, for they are nothing more than
12782empirical cognitions), I am not entitled to say: "If the conditioned
12783is given, all its conditions (as phenomena) are also given." I cannot,
12784therefore, from the fact of a conditioned being given, infer the
12785absolute totality of the series of its conditions. For phenomena are
12786nothing but an empirical synthesis in apprehension or perception, and
12787are therefore given only in it. Now, in speaking of phenomena it does
12788not follow that, if the conditioned is given, the synthesis which
12789constitutes its empirical condition is also thereby given and
12790presupposed; such a synthesis can be established only by an actual
12791regress in the series of conditions. But we are entitled to say in this
12792case that a regress to the conditions of a conditioned, in other
12793words, that a continuous empirical synthesis is enjoined; that, if the
12794conditions are not given, they are at least required; and that we are
12795certain to discover the conditions in this regress.
12796
12797We can now see that the major, in the above cosmological syllogism,
12798takes the conditioned in the transcendental signification which it has
12799in the pure category, while the minor speaks of it in the empirical
12800signification which it has in the category as applied to phenomena.
12801There is, therefore, a dialectical fallacy in the syllogism--a sophisma
12802figurae dictionis. But this fallacy is not a consciously devised one,
12803but a perfectly natural illusion of the common reason of man. For,
12804when a thing is given as conditioned, we presuppose in the major its
12805conditions and their series, unperceived, as it were, and unseen;
12806because this is nothing more than the logical requirement of complete
12807and satisfactory premisses for a given conclusion. In this case, time
12808is altogether left out in the connection of the conditioned with
12809the condition; they are supposed to be given in themselves, and
12810contemporaneously. It is, moreover, just as natural to regard phenomena
12811(in the minor) as things in themselves and as objects presented to the
12812pure understanding, as in the major, in which complete abstraction was
12813made of all conditions of intuition. But it is under these conditions
12814alone that objects are given. Now we overlooked a remarkable distinction
12815between the conceptions. The synthesis of the conditioned with its
12816condition, and the complete series of the latter (in the major) are not
12817limited by time, and do not contain the conception of succession. On the
12818contrary, the empirical synthesis and the series of conditions in the
12819phenomenal world--subsumed in the minor--are necessarily successive and
12820given in time alone. It follows that I cannot presuppose in the minor,
12821as I did in the major, the absolute totality of the synthesis and of
12822the series therein represented; for in the major all the members of the
12823series are given as things in themselves--without any limitations or
12824conditions of time, while in the minor they are possible only in and
12825through a successive regress, which cannot exist, except it be actually
12826carried into execution in the world of phenomena.
12827
12828After this proof of the viciousness of the argument commonly employed
12829in maintaining cosmological assertions, both parties may now be justly
12830dismissed, as advancing claims without grounds or title. But the process
12831has not been ended by convincing them that one or both were in the
12832wrong and had maintained an assertion which was without valid grounds
12833of proof. Nothing seems to be clearer than that, if one maintains: "The
12834world has a beginning," and another: "The world has no beginning," one
12835of the two must be right. But it is likewise clear that, if the evidence
12836on both sides is equal, it is impossible to discover on what side the
12837truth lies; and the controversy continues, although the parties have
12838been recommended to peace before the tribunal of reason. There remains,
12839then, no other means of settling the question than to convince the
12840parties, who refute each other with such conclusiveness and ability,
12841that they are disputing about nothing, and that a transcendental
12842illusion has been mocking them with visions of reality where there is
12843none. The mode of adjusting a dispute which cannot be decided upon its
12844own merits, we shall now proceed to lay before our readers.
12845
12846Zeno of Elea, a subtle dialectician, was severely reprimanded by Plato
12847as a sophist, who, merely from the base motive of exhibiting his
12848skill in discussion, maintained and subverted the same proposition by
12849arguments as powerful and convincing on the one side as on the other. He
12850maintained, for example, that God (who was probably nothing more, in his
12851view, than the world) is neither finite nor infinite, neither in motion
12852nor in rest, neither similar nor dissimilar to any other thing. It
12853seemed to those philosophers who criticized his mode of discussion
12854that his purpose was to deny completely both of two self-contradictory
12855propositions--which is absurd. But I cannot believe that there is any
12856justice in this accusation. The first of these propositions I shall
12857presently consider in a more detailed manner. With regard to the others,
12858if by the word of God he understood merely the Universe, his meaning
12859must have been--that it cannot be permanently present in one place--that
12860is, at rest--nor be capable of changing its place--that is, of
12861moving--because all places are in the universe, and the universe itself
12862is, therefore, in no place. Again, if the universe contains in itself
12863everything that exists, it cannot be similar or dissimilar to any other
12864thing, because there is, in fact, no other thing with which it can be
12865compared. If two opposite judgements presuppose a contingent impossible,
12866or arbitrary condition, both--in spite of their opposition (which is,
12867however, not properly or really a contradiction)--fall away; because the
12868condition, which ensured the validity of both, has itself disappeared.
12869
12870If we say: "Everybody has either a good or a bad smell," we have omitted
12871a third possible judgement--it has no smell at all; and thus both
12872conflicting statements may be false. If we say: "It is either
12873good-smelling or not good-smelling (vel suaveolens vel non-suaveolens),"
12874both judgements are contradictorily opposed; and the contradictory
12875opposite of the former judgement--some bodies are not
12876good-smelling--embraces also those bodies which have no smell at all. In
12877the preceding pair of opposed judgements (per disparata), the contingent
12878condition of the conception of body (smell) attached to both conflicting
12879statements, instead of having been omitted in the latter, which is
12880consequently not the contradictory opposite of the former.
12881
12882If, accordingly, we say: "The world is either infinite in extension, or
12883it is not infinite (non est infinitus)"; and if the former proposition
12884is false, its contradictory opposite--the world is not infinite--must
12885be true. And thus I should deny the existence of an infinite, without,
12886however affirming the existence of a finite world. But if we construct
12887our proposition thus: "The world is either infinite or finite
12888(non-infinite)," both statements may be false. For, in this case, we
12889consider the world as per se determined in regard to quantity, and
12890while, in the one judgement, we deny its infinite and consequently,
12891perhaps, its independent existence; in the other, we append to the
12892world, regarded as a thing in itself, a certain determination--that
12893of finitude; and the latter may be false as well as the former, if the
12894world is not given as a thing in itself, and thus neither as finite nor
12895as infinite in quantity. This kind of opposition I may be allowed to
12896term dialectical; that of contradictories may be called analytical
12897opposition. Thus then, of two dialectically opposed judgements both may
12898be false, from the fact, that the one is not a mere contradictory of
12899the other, but actually enounces more than is requisite for a full and
12900complete contradiction.
12901
12902When we regard the two propositions--"The world is infinite in
12903quantity," and, "The world is finite in quantity," as contradictory
12904opposites, we are assuming that the world--the complete series of
12905phenomena--is a thing in itself. For it remains as a permanent quantity,
12906whether I deny the infinite or the finite regress in the series of
12907its phenomena. But if we dismiss this assumption--this transcendental
12908illusion--and deny that it is a thing in itself, the contradictory
12909opposition is metamorphosed into a merely dialectical one; and the
12910world, as not existing in itself--independently of the regressive series
12911of my representations--exists in like manner neither as a whole which is
12912infinite nor as a whole which is finite in itself. The universe exists
12913for me only in the empirical regress of the series of phenomena and not
12914per se. If, then, it is always conditioned, it is never completely or as
12915a whole; and it is, therefore, not an unconditioned whole and does not
12916exist as such, either with an infinite, or with a finite quantity.
12917
12918What we have here said of the first cosmological idea--that of the
12919absolute totality of quantity in phenomena--applies also to the
12920others. The series of conditions is discoverable only in the regressive
12921synthesis itself, and not in the phenomenon considered as a thing in
12922itself--given prior to all regress. Hence I am compelled to say: "The
12923aggregate of parts in a given phenomenon is in itself neither finite nor
12924infinite; and these parts are given only in the regressive synthesis
12925of decomposition--a synthesis which is never given in absolute
12926completeness, either as finite, or as infinite." The same is the case
12927with the series of subordinated causes, or of the conditioned up to the
12928unconditioned and necessary existence, which can never be regarded as in
12929itself, ind in its totality, either as finite or as infinite; because,
12930as a series of subordinate representations, it subsists only in the
12931dynamical regress and cannot be regarded as existing previously to this
12932regress, or as a self-subsistent series of things.
12933
12934Thus the antinomy of pure reason in its cosmological ideas disappears.
12935For the above demonstration has established the fact that it is merely
12936the product of a dialectical and illusory opposition, which arises from
12937the application of the idea of absolute totality--admissible only as a
12938condition of things in themselves--to phenomena, which exist only in
12939our representations, and--when constituting a series--in a successive
12940regress. This antinomy of reason may, however, be really profitable to
12941our speculative interests, not in the way of contributing any dogmatical
12942addition, but as presenting to us another material support in our
12943critical investigations. For it furnishes us with an indirect proof
12944of the transcendental ideality of phenomena, if our minds were
12945not completely satisfied with the direct proof set forth in the
12946Trancendental Aesthetic. The proof would proceed in the following
12947dilemma. If the world is a whole existing in itself, it must be either
12948finite or infinite. But it is neither finite nor infinite--as has been
12949shown, on the one side, by the thesis, on the other, by the antithesis.
12950Therefore the world--the content of all phenomena--is not a whole
12951existing in itself. It follows that phenomena are nothing, apart
12952from our representations. And this is what we mean by transcendental
12953ideality.
12954
12955This remark is of some importance. It enables us to see that the proofs
12956of the fourfold antinomy are not mere sophistries--are not fallacious,
12957but grounded on the nature of reason, and valid--under the supposition
12958that phenomena are things in themselves. The opposition of the
12959judgements which follow makes it evident that a fallacy lay in the
12960initial supposition, and thus helps us to discover the true constitution
12961of objects of sense. This transcendental dialectic does not favour
12962scepticism, although it presents us with a triumphant demonstration of
12963the advantages of the sceptical method, the great utility of which is
12964apparent in the antinomy, where the arguments of reason were allowed to
12965confront each other in undiminished force. And although the result of
12966these conflicts of reason is not what we expected--although we have
12967obtained no positive dogmatical addition to metaphysical science--we
12968have still reaped a great advantage in the correction of our judgements
12969on these subjects of thought.
12970
12971
12972SECTION VIII. Regulative Principle of Pure Reason in relation to the
12973Cosmological Ideas.
12974
12975The cosmological principle of totality could not give us any certain
12976knowledge in regard to the maximum in the series of conditions in the
12977world of sense, considered as a thing in itself. The actual regress in
12978the series is the only means of approaching this maximum. This principle
12979of pure reason, therefore, may still be considered as valid--not as an
12980axiom enabling us to cogitate totality in the object as actual, but as
12981a problem for the understanding, which requires it to institute and
12982to continue, in conformity with the idea of totality in the mind, the
12983regress in the series of the conditions of a given conditioned. For in
12984the world of sense, that is, in space and time, every condition which
12985we discover in our investigation of phenomena is itself conditioned;
12986because sensuous objects are not things in themselves (in which case an
12987absolutely unconditioned might be reached in the progress of cognition),
12988but are merely empirical representations the conditions of which must
12989always be found in intuition. The principle of reason is therefore
12990properly a mere rule--prescribing a regress in the series of conditions
12991for given phenomena, and prohibiting any pause or rest on an absolutely
12992unconditioned. It is, therefore, not a principle of the possibility
12993of experience or of the empirical cognition of sensuous
12994objects--consequently not a principle of the understanding; for every
12995experience is confined within certain proper limits determined by the
12996given intuition. Still less is it a constitutive principle of reason
12997authorizing us to extend our conception of the sensuous world beyond all
12998possible experience. It is merely a principle for the enlargement and
12999extension of experience as far as is possible for human faculties. It
13000forbids us to consider any empirical limits as absolute. It is, hence, a
13001principle of reason, which, as a rule, dictates how we ought to proceed
13002in our empirical regress, but is unable to anticipate or indicate prior
13003to the empirical regress what is given in the object itself. I have
13004termed it for this reason a regulative principle of reason; while the
13005principle of the absolute totality of the series of conditions,
13006as existing in itself and given in the object, is a constitutive
13007cosmological principle. This distinction will at once demonstrate the
13008falsehood of the constitutive principle, and prevent us from attributing
13009(by a transcendental subreptio) objective reality to an idea, which is
13010valid only as a rule.
13011
13012In order to understand the proper meaning of this rule of pure reason,
13013we must notice first that it cannot tell us what the object is, but only
13014how the empirical regress is to be proceeded with in order to attain to
13015the complete conception of the object. If it gave us any information in
13016respect to the former statement, it would be a constitutive principle--a
13017principle impossible from the nature of pure reason. It will not
13018therefore enable us to establish any such conclusions as: "The series
13019of conditions for a given conditioned is in itself finite," or, "It is
13020infinite." For, in this case, we should be cogitating in the mere idea
13021of absolute totality, an object which is not and cannot be given in
13022experience; inasmuch as we should be attributing a reality objective and
13023independent of the empirical synthesis, to a series of phenomena. This
13024idea of reason cannot then be regarded as valid--except as a rule for
13025the regressive synthesis in the series of conditions, according to
13026which we must proceed from the conditioned, through all intermediate and
13027subordinate conditions, up to the unconditioned; although this goal is
13028unattained and unattainable. For the absolutely unconditioned cannot be
13029discovered in the sphere of experience.
13030
13031We now proceed to determine clearly our notion of a synthesis which
13032can never be complete. There are two terms commonly employed for this
13033purpose. These terms are regarded as expressions of different and
13034distinguishable notions, although the ground of the distinction has
13035never been clearly exposed. The term employed by the mathematicians
13036is progressus in infinitum. The philosophers prefer the expression
13037progressus in indefinitum. Without detaining the reader with an
13038examination of the reasons for such a distinction, or with remarks
13039on the right or wrong use of the terms, I shall endeavour clearly to
13040determine these conceptions, so far as is necessary for the purpose in
13041this Critique.
13042
13043We may, with propriety, say of a straight line, that it may be produced
13044to infinity. In this case the distinction between a progressus in
13045infinitum and a progressus in indefinitum is a mere piece of subtlety.
13046For, although when we say, "Produce a straight line," it is more correct
13047to say in indefinitum than in infinitum; because the former means,
13048"Produce it as far as you please," the second, "You must not cease to
13049produce it"; the expression in infinitum is, when we are speaking of the
13050power to do it, perfectly correct, for we can always make it longer if
13051we please--on to infinity. And this remark holds good in all cases, when
13052we speak of a progressus, that is, an advancement from the condition to
13053the conditioned; this possible advancement always proceeds to infinity.
13054We may proceed from a given pair in the descending line of generation
13055from father to son, and cogitate a never-ending line of descendants from
13056it. For in such a case reason does not demand absolute totality in the
13057series, because it does not presuppose it as a condition and as given
13058(datum), but merely as conditioned, and as capable of being given
13059(dabile).
13060
13061Very different is the case with the problem: "How far the regress, which
13062ascends from the given conditioned to the conditions, must extend";
13063whether I can say: "It is a regress in infinitum," or only "in
13064indefinitum"; and whether, for example, setting out from the human
13065beings at present alive in the world, I may ascend in the series of
13066their ancestors, in infinitum--mr whether all that can be said is, that
13067so far as I have proceeded, I have discovered no empirical ground for
13068considering the series limited, so that I am justified, and indeed,
13069compelled to search for ancestors still further back, although I am not
13070obliged by the idea of reason to presuppose them.
13071
13072My answer to this question is: "If the series is given in empirical
13073intuition as a whole, the regress in the series of its internal
13074conditions proceeds in infinitum; but, if only one member of the series
13075is given, from which the regress is to proceed to absolute totality, the
13076regress is possible only in indefinitum." For example, the division of
13077a portion of matter given within certain limits--of a body, that
13078is--proceeds in infinitum. For, as the condition of this whole is its
13079part, and the condition of the part a part of the part, and so on, and
13080as in this regress of decomposition an unconditioned indivisible member
13081of the series of conditions is not to be found; there are no reasons
13082or grounds in experience for stopping in the division, but, on the
13083contrary, the more remote members of the division are actually and
13084empirically given prior to this division. That is to say, the division
13085proceeds to infinity. On the other hand, the series of ancestors of
13086any given human being is not given, in its absolute totality, in any
13087experience, and yet the regress proceeds from every genealogical member
13088of this series to one still higher, and does not meet with any empirical
13089limit presenting an absolutely unconditioned member of the series.
13090But as the members of such a series are not contained in the empirical
13091intuition of the whole, prior to the regress, this regress does not
13092proceed to infinity, but only in indefinitum, that is, we are called
13093upon to discover other and higher members, which are themselves always
13094conditioned.
13095
13096In neither case--the regressus in infinitum, nor the regressus in
13097indefinitum, is the series of conditions to be considered as actually
13098infinite in the object itself. This might be true of things in
13099themselves, but it cannot be asserted of phenomena, which, as conditions
13100of each other, are only given in the empirical regress itself. Hence,
13101the question no longer is, "What is the quantity of this series of
13102conditions in itself--is it finite or infinite?" for it is nothing in
13103itself; but, "How is the empirical regress to be commenced, and how
13104far ought we to proceed with it?" And here a signal distinction in
13105the application of this rule becomes apparent. If the whole is given
13106empirically, it is possible to recede in the series of its internal
13107conditions to infinity. But if the whole is not given, and can only
13108be given by and through the empirical regress, I can only say: "It
13109is possible to infinity, to proceed to still higher conditions in
13110the series." In the first case, I am justified in asserting that more
13111members are empirically given in the object than I attain to in the
13112regress (of decomposition). In the second case, I am justified only in
13113saying, that I can always proceed further in the regress, because no
13114member of the series is given as absolutely conditioned, and thus
13115a higher member is possible, and an inquiry with regard to it is
13116necessary. In the one case it is necessary to find other members of the
13117series, in the other it is necessary to inquire for others, inasmuch as
13118experience presents no absolute limitation of the regress. For, either
13119you do not possess a perception which absolutely limits your empirical
13120regress, and in this case the regress cannot be regarded as complete;
13121or, you do possess such a limitative perception, in which case it is not
13122a part of your series (for that which limits must be distinct from
13123that which is limited by it), and it is incumbent you to continue your
13124regress up to this condition, and so on.
13125
13126These remarks will be placed in their proper light by their application
13127in the following section.
13128
13129
13130
13131SECTION IX. Of the Empirical Use of the Regulative Principle of
13132Reason with regard to the Cosmological Ideas.
13133
13134We have shown that no transcendental use can be made either of the
13135conceptions of reason or of understanding. We have shown, likewise, that
13136the demand of absolute totality in the series of conditions in the world
13137of sense arises from a transcendental employment of reason, resting on
13138the opinion that phenomena are to be regarded as things in themselves.
13139It follows that we are not required to answer the question respecting
13140the absolute quantity of a series--whether it is in itself limited or
13141unlimited. We are only called upon to determine how far we must proceed
13142in the empirical regress from condition to condition, in order to
13143discover, in conformity with the rule of reason, a full and correct
13144answer to the questions proposed by reason itself.
13145
13146This principle of reason is hence valid only as a rule for the extension
13147of a possible experience--its invalidity as a principle constitutive of
13148phenomena in themselves having been sufficiently demonstrated. And thus,
13149too, the antinomial conflict of reason with itself is completely put an
13150end to; inasmuch as we have not only presented a critical solution of
13151the fallacy lurking in the opposite statements of reason, but have shown
13152the true meaning of the ideas which gave rise to these statements. The
13153dialectical principle of reason has, therefore, been changed into a
13154doctrinal principle. But in fact, if this principle, in the subjective
13155signification which we have shown to be its only true sense, may be
13156guaranteed as a principle of the unceasing extension of the employment
13157of our understanding, its influence and value are just as great as if
13158it were an axiom for the a priori determination of objects. For such
13159an axiom could not exert a stronger influence on the extension and
13160rectification of our knowledge, otherwise than by procuring for the
13161principles of the understanding the most widely expanded employment in
13162the field of experience.
13163
13164
13165
13166I. Solution of the Cosmological Idea of the Totality of the
13167Composition of Phenomena in the Universe.
13168
13169Here, as well as in the case of the other cosmological problems, the
13170ground of the regulative principle of reason is the proposition that
13171in our empirical regress no experience of an absolute limit, and
13172consequently no experience of a condition, which is itself absolutely
13173unconditioned, is discoverable. And the truth of this proposition itself
13174rests upon the consideration that such an experience must represent
13175to us phenomena as limited by nothing or the mere void, on which our
13176continued regress by means of perception must abut--which is impossible.
13177
13178Now this proposition, which declares that every condition attained in
13179the empirical regress must itself be considered empirically conditioned,
13180contains the rule in terminis, which requires me, to whatever extent
13181I may have proceeded in the ascending series, always to look for some
13182higher member in the series--whether this member is to become known to
13183me through experience, or not.
13184
13185Nothing further is necessary, then, for the solution of the first
13186cosmological problem, than to decide, whether, in the regress to the
13187unconditioned quantity of the universe (as regards space and time),
13188this never limited ascent ought to be called a regressus in infinitum or
13189indefinitum.
13190
13191The general representation which we form in our minds of the series of
13192all past states or conditions of the world, or of all the things which
13193at present exist in it, is itself nothing more than a possible empirical
13194regress, which is cogitated--although in an undetermined manner--in the
13195mind, and which gives rise to the conception of a series of conditions
13196for a given object.* Now I have a conception of the universe, but not
13197an intuition--that is, not an intuition of it as a whole. Thus I cannot
13198infer the magnitude of the regress from the quantity or magnitude of the
13199world, and determine the former by means of the latter; on the contrary,
13200I must first of all form a conception of the quantity or magnitude
13201of the world from the magnitude of the empirical regress. But of this
13202regress I know nothing more than that I ought to proceed from every
13203given member of the series of conditions to one still higher. But the
13204quantity of the universe is not thereby determined, and we cannot affirm
13205that this regress proceeds in infinitum. Such an affirmation would
13206anticipate the members of the series which have not yet been reached,
13207and represent the number of them as beyond the grasp of any empirical
13208synthesis; it would consequently determine the cosmical quantity
13209prior to the regress (although only in a negative manner)--which is
13210impossible. For the world is not given in its totality in any intuition:
13211consequently, its quantity cannot be given prior to the regress. It
13212follows that we are unable to make any declaration respecting the
13213cosmical quantity in itself--not even that the regress in it is a
13214regress in infinitum; we must only endeavour to attain to a conception
13215of the quantity of the universe, in conformity with the rule which
13216determines the empirical regress in it. But this rule merely requires
13217us never to admit an absolute limit to our series--how far soever we may
13218have proceeded in it, but always, on the contrary, to subordinate every
13219phenomenon to some other as its condition, and consequently to proceed
13220to this higher phenomenon. Such a regress is, therefore, the regressus
13221in indefinitum, which, as not determining a quantity in the object, is
13222clearly distinguishable from the regressus in infinitum.
13223
13224
13225 [*Footnote: The cosmical series can neither be greater nor smaller than
13226the possible empirical regress, upon which its conception is based. And
13227as this regress cannot be a determinate infinite regress, still less a
13228determinate finite (absolutely limited), it is evident that we cannot
13229regard the world as either finite or infinite, because the regress,
13230which gives us the representation of the world, is neither finite nor
13231infinite.]
13232
13233
13234It follows from what we have said that we are not justified in declaring
13235the world to be infinite in space, or as regards past time. For this
13236conception of an infinite given quantity is empirical; but we cannot
13237apply the conception of an infinite quantity to the world as an object
13238of the senses. I cannot say, "The regress from a given perception to
13239everything limited either in space or time, proceeds in infinitum," for
13240this presupposes an infinite cosmical quantity; neither can I say, "It
13241is finite," for an absolute limit is likewise impossible in experience.
13242It follows that I am not entitled to make any assertion at all
13243respecting the whole object of experience--the world of sense; I must
13244limit my declarations to the rule according to which experience or
13245empirical knowledge is to be attained.
13246
13247To the question, therefore, respecting the cosmical quantity, the first
13248and negative answer is: "The world has no beginning in time, and no
13249absolute limit in space."
13250
13251For, in the contrary case, it would be limited by a void time on the
13252one hand, and by a void space on the other. Now, since the world, as a
13253phenomenon, cannot be thus limited in itself for a phenomenon is not a
13254thing in itself; it must be possible for us to have a perception of this
13255limitation by a void time and a void space. But such a perception--such
13256an experience is impossible; because it has no content. Consequently,
13257an absolute cosmical limit is empirically, and therefore absolutely,
13258impossible.*
13259
13260
13261 [*Footnote: The reader will remark that the proof presented above is
13262very different from the dogmatical demonstration given in the antithesis
13263of the first antinomy. In that demonstration, it was taken for granted
13264that the world is a thing in itself--given in its totality prior to
13265all regress, and a determined position in space and time was denied to
13266it--if it was not considered as occupying all time and all space. Hence
13267our conclusion differed from that given above; for we inferred in the
13268antithesis the actual infinity of the world.]
13269
13270
13271From this follows the affirmative answer: "The regress in the series
13272of phenomena--as a determination of the cosmical quantity, proceeds in
13273indefinitum." This is equivalent to saying: "The world of sense has no
13274absolute quantity, but the empirical regress (through which alone the
13275world of sense is presented to us on the side of its conditions) rests
13276upon a rule, which requires it to proceed from every member of the
13277series, as conditioned, to one still more remote (whether through
13278personal experience, or by means of history, or the chain of cause and
13279effect), and not to cease at any point in this extension of the possible
13280empirical employment of the understanding." And this is the proper and
13281only use which reason can make of its principles.
13282
13283The above rule does not prescribe an unceasing regress in one kind of
13284phenomena. It does not, for example, forbid us, in our ascent from an
13285individual human being through the line of his ancestors, to expect that
13286we shall discover at some point of the regress a primeval pair, or to
13287admit, in the series of heavenly bodies, a sun at the farthest possible
13288distance from some centre. All that it demands is a perpetual progress
13289from phenomena to phenomena, even although an actual perception is not
13290presented by them (as in the case of our perceptions being so weak
13291as that we are unable to become conscious of them), since they,
13292nevertheless, belong to possible experience.
13293
13294Every beginning is in time, and all limits to extension are in space.
13295But space and time are in the world of sense. Consequently phenomena
13296in the world are conditionally limited, but the world itself is not
13297limited, either conditionally or unconditionally.
13298
13299For this reason, and because neither the world nor the cosmical series
13300of conditions to a given conditioned can be completely given, our
13301conception of the cosmical quantity is given only in and through the
13302regress and not prior to it--in a collective intuition. But the regress
13303itself is really nothing more than the determining of the cosmical
13304quantity, and cannot therefore give us any determined conception of
13305it--still less a conception of a quantity which is, in relation to a
13306certain standard, infinite. The regress does not, therefore, proceed to
13307infinity (an infinity given), but only to an indefinite extent, for or
13308the of presenting to us a quantity--realized only in and through the
13309regress itself.
13310
13311
13312
13313II. Solution of the Cosmological Idea of the Totality of the Division
13314of a Whole given in Intuition.
13315
13316When I divide a whole which is given in intuition, I proceed from a
13317conditioned to its conditions. The division of the parts of the whole
13318(subdivisio or decompositio) is a regress in the series of these
13319conditions. The absolute totality of this series would be actually
13320attained and given to the mind, if the regress could arrive at simple
13321parts. But if all the parts in a continuous decomposition are themselves
13322divisible, the division, that is to say, the regress, proceeds from the
13323conditioned to its conditions in infinitum; because the conditions (the
13324parts) are themselves contained in the conditioned, and, as the latter
13325is given in a limited intuition, the former are all given along with it.
13326This regress cannot, therefore, be called a regressus in indefinitum, as
13327happened in the case of the preceding cosmological idea, the regress
13328in which proceeded from the conditioned to the conditions not given
13329contemporaneously and along with it, but discoverable only through the
13330empirical regress. We are not, however, entitled to affirm of a whole
13331of this kind, which is divisible in infinitum, that it consists of an
13332infinite number of parts. For, although all the parts are contained in
13333the intuition of the whole, the whole division is not contained therein.
13334The division is contained only in the progressing decomposition--in the
13335regress itself, which is the condition of the possibility and actuality
13336of the series. Now, as this regress is infinite, all the members
13337(parts) to which it attains must be contained in the given whole as an
13338aggregate. But the complete series of division is not contained therein.
13339For this series, being infinite in succession and always incomplete,
13340cannot represent an infinite number of members, and still less a
13341composition of these members into a whole.
13342
13343To apply this remark to space. Every limited part of space presented to
13344intuition is a whole, the parts of which are always spaces--to whatever
13345extent subdivided. Every limited space is hence divisible to infinity.
13346
13347Let us again apply the remark to an external phenomenon enclosed in
13348limits, that is, a body. The divisibility of a body rests upon the
13349divisibility of space, which is the condition of the possibility of the
13350body as an extended whole. A body is consequently divisible to infinity,
13351though it does not, for that reason, consist of an infinite number of
13352parts.
13353
13354It certainly seems that, as a body must be cogitated as substance
13355in space, the law of divisibility would not be applicable to it as
13356substance. For we may and ought to grant, in the case of space, that
13357division or decomposition, to any extent, never can utterly annihilate
13358composition (that is to say, the smallest part of space must still
13359consist of spaces); otherwise space would entirely cease to
13360exist--which is impossible. But, the assertion on the other band that
13361when all composition in matter is annihilated in thought, nothing
13362remains, does not seem to harmonize with the conception of substance,
13363which must be properly the subject of all composition and must remain,
13364even after the conjunction of its attributes in space--which constituted
13365a body--is annihilated in thought. But this is not the case with
13366substance in the phenomenal world, which is not a thing in itself
13367cogitated by the pure category. Phenomenal substance is not an absolute
13368subject; it is merely a permanent sensuous image, and nothing more than
13369an intuition, in which the unconditioned is not to be found.
13370
13371But, although this rule of progress to infinity is legitimate and
13372applicable to the subdivision of a phenomenon, as a mere occupation or
13373filling of space, it is not applicable to a whole consisting of a number
13374of distinct parts and constituting a quantum discretum--that is to say,
13375an organized body. It cannot be admitted that every part in an organized
13376whole is itself organized, and that, in analysing it to infinity, we
13377must always meet with organized parts; although we may allow that the
13378parts of the matter which we decompose in infinitum, may be organized.
13379For the infinity of the division of a phenomenon in space rests
13380altogether on the fact that the divisibility of a phenomenon is given
13381only in and through this infinity, that is, an undetermined number of
13382parts is given, while the parts themselves are given and determined only
13383in and through the subdivision; in a word, the infinity of the division
13384necessarily presupposes that the whole is not already divided in se.
13385Hence our division determines a number of parts in the whole--a number
13386which extends just as far as the actual regress in the division; while,
13387on the other hand, the very notion of a body organized to infinity
13388represents the whole as already and in itself divided. We expect,
13389therefore, to find in it a determinate, but at the same time, infinite,
13390number of parts--which is self-contradictory. For we should thus have a
13391whole containing a series of members which could not be completed in
13392any regress--which is infinite, and at the same time complete in an
13393organized composite. Infinite divisibility is applicable only to a
13394quantum continuum, and is based entirely on the infinite divisibility
13395of space, But in a quantum discretum the multitude of parts or units is
13396always determined, and hence always equal to some number. To what extent
13397a body may be organized, experience alone can inform us; and although,
13398so far as our experience of this or that body has extended, we may not
13399have discovered any inorganic part, such parts must exist in possible
13400experience. But how far the transcendental division of a phenomenon
13401must extend, we cannot know from experience--it is a question which
13402experience cannot answer; it is answered only by the principle of reason
13403which forbids us to consider the empirical regress, in the analysis of
13404extended body, as ever absolutely complete.
13405
13406
13407
13408Concluding Remark on the Solution of the Transcendental Mathematical
13409Ideas--and Introductory to the Solution of the Dynamical Ideas.
13410
13411We presented the antinomy of pure reason in a tabular form, and we
13412endeavoured to show the ground of this self-contradiction on the part
13413of reason, and the only means of bringing it to a conclusion--namely, by
13414declaring both contradictory statements to be false. We represented
13415in these antinomies the conditions of phenomena as belonging to the
13416conditioned according to relations of space and time--which is the
13417usual supposition of the common understanding. In this respect, all
13418dialectical representations of totality, in the series of conditions
13419to a given conditioned, were perfectly homogeneous. The condition was
13420always a member of the series along with the conditioned, and thus the
13421homogeneity of the whole series was assured. In this case the regress
13422could never be cogitated as complete; or, if this was the case, a member
13423really conditioned was falsely regarded as a primal member, consequently
13424as unconditioned. In such an antinomy, therefore, we did not consider
13425the object, that is, the conditioned, but the series of conditions
13426belonging to the object, and the magnitude of that series. And thus
13427arose the difficulty--a difficulty not to be settled by any decision
13428regarding the claims of the two parties, but simply by cutting the
13429knot--by declaring the series proposed by reason to be either too long
13430or too short for the understanding, which could in neither case make its
13431conceptions adequate with the ideas.
13432
13433But we have overlooked, up to this point, an essential difference
13434existing between the conceptions of the understanding which reason
13435endeavours to raise to the rank of ideas--two of these indicating a
13436mathematical, and two a dynamical synthesis of phenomena. Hitherto, it
13437was necessary to signalize this distinction; for, just as in our general
13438representation of all transcendental ideas, we considered them under
13439phenomenal conditions, so, in the two mathematical ideas, our discussion
13440is concerned solely with an object in the world of phenomena. But as
13441we are now about to proceed to the consideration of the dynamical
13442conceptions of the understanding, and their adequateness with ideas, we
13443must not lose sight of this distinction. We shall find that it opens up
13444to us an entirely new view of the conflict in which reason is involved.
13445For, while in the first two antinomies, both parties were dismissed, on
13446the ground of having advanced statements based upon false hypothesis; in
13447the present case the hope appears of discovering a hypothesis which may
13448be consistent with the demands of reason, and, the judge completing the
13449statement of the grounds of claim, which both parties had left in an
13450unsatisfactory state, the question may be settled on its own merits,
13451not by dismissing the claimants, but by a comparison of the arguments on
13452both sides. If we consider merely their extension, and whether they are
13453adequate with ideas, the series of conditions may be regarded as all
13454homogeneous. But the conception of the understanding which lies at the
13455basis of these ideas, contains either a synthesis of the homogeneous
13456(presupposed in every quantity--in its composition as well as in its
13457division) or of the heterogeneous, which is the case in the dynamical
13458synthesis of cause and effect, as well as of the necessary and the
13459contingent.
13460
13461Thus it happens that in the mathematical series of phenomena no other
13462than a sensuous condition is admissible--a condition which is itself a
13463member of the series; while the dynamical series of sensuous conditions
13464admits a heterogeneous condition, which is not a member of the series,
13465but, as purely intelligible, lies out of and beyond it. And thus reason
13466is satisfied, and an unconditioned placed at the head of the series
13467of phenomena, without introducing confusion into or discontinuing it,
13468contrary to the principles of the understanding.
13469
13470Now, from the fact that the dynamical ideas admit a condition of
13471phenomena which does not form a part of the series of phenomena, arises
13472a result which we should not have expected from an antinomy. In former
13473cases, the result was that both contradictory dialectical statements
13474were declared to be false. In the present case, we find the conditioned
13475in the dynamical series connected with an empirically unconditioned,
13476but non-sensuous condition; and thus satisfaction is done to the
13477understanding on the one hand and to the reason on the other.* While,
13478moreover, the dialectical arguments for unconditioned totality in mere
13479phenomena fall to the ground, both propositions of reason may be shown
13480to be true in their proper signification. This could not happen in
13481the case of the cosmological ideas which demanded a mathematically
13482unconditioned unity; for no condition could be placed at the head of
13483the series of phenomena, except one which was itself a phenomenon and
13484consequently a member of the series.
13485
13486
13487 [*Footnote: For the understanding cannot admit among phenomena a
13488condition which is itself empirically unconditioned. But if it is
13489possible to cogitate an intelligible condition--one which is not a
13490member of the series of phenomena--for a conditioned phenomenon, without
13491breaking the series of empirical conditions, such a condition may be
13492admissible as empirically unconditioned, and the empirical regress
13493continue regular, unceasing, and intact.]
13494
13495
13496
13497III. Solution of the Cosmological Idea of the Totality of the
13498Deduction of Cosmical Events from their Causes.
13499
13500There are only two modes of causality cogitable--the causality of nature
13501or of freedom. The first is the conjunction of a particular state with
13502another preceding it in the world of sense, the former following the
13503latter by virtue of a law. Now, as the causality of phenomena is
13504subject to conditions of time, and the preceding state, if it had always
13505existed, could not have produced an effect which would make its first
13506appearance at a particular time, the causality of a cause must itself be
13507an effect--must itself have begun to be, and therefore, according to the
13508principle of the understanding, itself requires a cause.
13509
13510We must understand, on the contrary, by the term freedom, in the
13511cosmological sense, a faculty of the spontaneous origination of a state;
13512the causality of which, therefore, is not subordinated to another cause
13513determining it in time. Freedom is in this sense a pure transcendental
13514idea, which, in the first place, contains no empirical element; the
13515object of which, in the second place, cannot be given or determined in
13516any experience, because it is a universal law of the very possibility
13517of experience, that everything which happens must have a cause, that
13518consequently the causality of a cause, being itself something that has
13519happened, must also have a cause. In this view of the case, the whole
13520field of experience, how far soever it may extend, contains nothing that
13521is not subject to the laws of nature. But, as we cannot by this means
13522attain to an absolute totality of conditions in reference to the series
13523of causes and effects, reason creates the idea of a spontaneity, which
13524can begin to act of itself, and without any external cause determining
13525it to action, according to the natural law of causality.
13526
13527It is especially remarkable that the practical conception of freedom
13528is based upon the transcendental idea, and that the question of
13529the possibility of the former is difficult only as it involves the
13530consideration of the truth of the latter. Freedom, in the practical
13531sense, is the independence of the will of coercion by sensuous impulses.
13532A will is sensuous, in so far as it is pathologically affected (by
13533sensuous impulses); it is termed animal (arbitrium brutum), when it is
13534pathologically necessitated. The human will is certainly an arbitrium
13535sensitivum, not brutum, but liberum; because sensuousness does not
13536necessitate its action, a faculty existing in man of self-determination,
13537independently of all sensuous coercion.
13538
13539It is plain that, if all causality in the world of sense were
13540natural--and natural only--every event would be determined by another
13541according to necessary laws, and that, consequently, phenomena, in
13542so far as they determine the will, must necessitate every action as a
13543natural effect from themselves; and thus all practical freedom would
13544fall to the ground with the transcendental idea. For the latter
13545presupposes that although a certain thing has not happened, it ought to
13546have happened, and that, consequently, its phenomenal cause was not so
13547powerful and determinative as to exclude the causality of our will--a
13548causality capable of producing effects independently of and even in
13549opposition to the power of natural causes, and capable, consequently, of
13550spontaneously originating a series of events.
13551
13552Here, too, we find it to be the case, as we generally found in the
13553self-contradictions and perplexities of a reason which strives to pass
13554the bounds of possible experience, that the problem is properly not
13555physiological, but transcendental. The question of the possibility
13556of freedom does indeed concern psychology; but, as it rests upon
13557dialectical arguments of pure reason, its solution must engage the
13558attention of transcendental philosophy. Before attempting this solution,
13559a task which transcendental philosophy cannot decline, it will
13560be advisable to make a remark with regard to its procedure in the
13561settlement of the question.
13562
13563If phenomena were things in themselves, and time and space forms of the
13564existence of things, condition and conditioned would always be members
13565of the same series; and thus would arise in the present case the
13566antinomy common to all transcendental ideas--that their series is either
13567too great or too small for the understanding. The dynamical ideas, which
13568we are about to discuss in this and the following section, possess the
13569peculiarity of relating to an object, not considered as a quantity, but
13570as an existence; and thus, in the discussion of the present question,
13571we may make abstraction of the quantity of the series of conditions,
13572and consider merely the dynamical relation of the condition to the
13573conditioned. The question, then, suggests itself, whether freedom is
13574possible; and, if it is, whether it can consist with the universality
13575of the natural law of causality; and, consequently, whether we enounce a
13576proper disjunctive proposition when we say: "Every effect must have its
13577origin either in nature or in freedom," or whether both cannot exist
13578together in the same event in different relations. The principle of
13579an unbroken connection between all events in the phenomenal world, in
13580accordance with the unchangeable laws of nature, is a well-established
13581principle of transcendental analytic which admits of no exception. The
13582question, therefore, is: "Whether an effect, determined according to
13583the laws of nature, can at the same time be produced by a free agent, or
13584whether freedom and nature mutually exclude each other?" And here, the
13585common but fallacious hypothesis of the absolute reality of phenomena
13586manifests its injurious influence in embarrassing the procedure
13587of reason. For if phenomena are things in themselves, freedom is
13588impossible. In this case, nature is the complete and all-sufficient
13589cause of every event; and condition and conditioned, cause and effect
13590are contained in the same series, and necessitated by the same law. If,
13591on the contrary, phenomena are held to be, as they are in fact, nothing
13592more than mere representations, connected with each other in accordance
13593with empirical laws, they must have a ground which is not phenomenal.
13594But the causality of such an intelligible cause is not determined or
13595determinable by phenomena; although its effects, as phenomena, must be
13596determined by other phenomenal existences. This cause and its causality
13597exist therefore out of and apart from the series of phenomena; while
13598its effects do exist and are discoverable in the series of empirical
13599conditions. Such an effect may therefore be considered to be free in
13600relation to its intelligible cause, and necessary in relation to the
13601phenomena from which it is a necessary consequence--a distinction which,
13602stated in this perfectly general and abstract manner, must appear in
13603the highest degree subtle and obscure. The sequel will explain. It is
13604sufficient, at present, to remark that, as the complete and unbroken
13605connection of phenomena is an unalterable law of nature, freedom is
13606impossible--on the supposition that phenomena are absolutely real. Hence
13607those philosophers who adhere to the common opinion on this subject can
13608never succeed in reconciling the ideas of nature and freedom.
13609
13610
13611
13612Possibility of Freedom in Harmony with the Universal Law of Natural
13613Necessity.
13614
13615That element in a sensuous object which is not itself sensuous, I may be
13616allowed to term intelligible. If, accordingly, an object which must be
13617regarded as a sensuous phenomenon possesses a faculty which is not an
13618object of sensuous intuition, but by means of which it is capable of
13619being the cause of phenomena, the causality of an object or existence of
13620this kind may be regarded from two different points of view. It may be
13621considered to be intelligible, as regards its action--the action of
13622a thing which is a thing in itself, and sensuous, as regards its
13623effects--the effects of a phenomenon belonging to the sensuous world. We
13624should accordingly, have to form both an empirical and an intellectual
13625conception of the causality of such a faculty or power--both, however,
13626having reference to the same effect. This twofold manner of cogitating
13627a power residing in a sensuous object does not run counter to any of
13628the conceptions which we ought to form of the world of phenomena or of
13629a possible experience. Phenomena--not being things in themselves--must
13630have a transcendental object as a foundation, which determines them as
13631mere representations; and there seems to be no reason why we should not
13632ascribe to this transcendental object, in addition to the property of
13633self-phenomenization, a causality whose effects are to be met with in
13634the world of phenomena, although it is not itself a phenomenon. But
13635every effective cause must possess a character, that is to say, a law of
13636its causality, without which it would cease to be a cause. In the above
13637case, then, every sensuous object would possess an empirical character,
13638which guaranteed that its actions, as phenomena, stand in complete and
13639harmonious connection, conformably to unvarying natural laws, with all
13640other phenomena, and can be deduced from these, as conditions, and that
13641they do thus, in connection with these, constitute a series in the order
13642of nature. This sensuous object must, in the second place, possess an
13643intelligible character, which guarantees it to be the cause of those
13644actions, as phenomena, although it is not itself a phenomenon nor
13645subordinate to the conditions of the world of sense. The former may
13646be termed the character of the thing as a phenomenon, the latter the
13647character of the thing as a thing in itself.
13648
13649Now this active subject would, in its character of intelligible subject,
13650be subordinate to no conditions of time, for time is only a condition
13651of phenomena, and not of things in themselves. No action would begin or
13652cease to be in this subject; it would consequently be free from the law
13653of all determination of time--the law of change, namely, that everything
13654which happens must have a cause in the phenomena of a preceding
13655state. In one word, the causality of the subject, in so far as it is
13656intelligible, would not form part of the series of empirical conditions
13657which determine and necessitate an event in the world of sense. Again,
13658this intelligible character of a thing cannot be immediately cognized,
13659because we can perceive nothing but phenomena, but it must be capable of
13660being cogitated in harmony with the empirical character; for we always
13661find ourselves compelled to place, in thought, a transcendental object
13662at the basis of phenomena although we can never know what this object is
13663in itself.
13664
13665In virtue of its empirical character, this subject would at the same
13666time be subordinate to all the empirical laws of causality, and, as a
13667phenomenon and member of the sensuous world, its effects would have
13668to be accounted for by a reference to preceding phenomena. Eternal
13669phenomena must be capable of influencing it; and its actions, in
13670accordance with natural laws, must explain to us how its empirical
13671character, that is, the law of its causality, is to be cognized in and
13672by means of experience. In a word, all requisites for a complete and
13673necessary determination of these actions must be presented to us by
13674experience.
13675
13676In virtue of its intelligible character, on the other hand (although we
13677possess only a general conception of this character), the subject
13678must be regarded as free from all sensuous influences, and from
13679all phenomenal determination. Moreover, as nothing happens in this
13680subject--for it is a noumenon, and there does not consequently exist in
13681it any change, demanding the dynamical determination of time, and for
13682the same reason no connection with phenomena as causes--this active
13683existence must in its actions be free from and independent of natural
13684necessity, for or necessity exists only in the world of phenomena. It
13685would be quite correct to say that it originates or begins its effects
13686in the world of sense from itself, although the action productive of
13687these effects does not begin in itself. We should not be in this case
13688affirming that these sensuous effects began to exist of themselves,
13689because they are always determined by prior empirical conditions--by
13690virtue of the empirical character, which is the phenomenon of the
13691intelligible character--and are possible only as constituting a
13692continuation of the series of natural causes. And thus nature and
13693freedom, each in the complete and absolute signification of these terms,
13694can exist, without contradiction or disagreement, in the same action.
13695
13696
13697
13698Exposition of the Cosmological Idea of Freedom in Harmony with the
13699Universal Law of Natural Necessity.
13700
13701I have thought it advisable to lay before the reader at first merely
13702a sketch of the solution of this transcendental problem, in order to
13703enable him to form with greater ease a clear conception of the course
13704which reason must adopt in the solution. I shall now proceed to exhibit
13705the several momenta of this solution, and to consider them in their
13706order.
13707
13708The natural law that everything which happens must have a cause, that
13709the causality of this cause, that is, the action of the cause (which
13710cannot always have existed, but must be itself an event, for it precedes
13711in time some effect which it has originated), must have itself a
13712phenomenal cause, by which it is determined and, and, consequently, all
13713events are empirically determined in an order of nature--this law, I
13714say, which lies at the foundation of the possibility of experience,
13715and of a connected system of phenomena or nature is a law of the
13716understanding, from which no departure, and to which no exception, can
13717be admitted. For to except even a single phenomenon from its operation
13718is to exclude it from the sphere of possible experience and thus to
13719admit it to be a mere fiction of thought or phantom of the brain.
13720
13721Thus we are obliged to acknowledge the existence of a chain of causes,
13722in which, however, absolute totality cannot be found. But we need
13723not detain ourselves with this question, for it has already been
13724sufficiently answered in our discussion of the antinomies into which
13725reason falls, when it attempts to reach the unconditioned in the series
13726of phenomena. If we permit ourselves to be deceived by the illusion of
13727transcendental idealism, we shall find that neither nature nor freedom
13728exists. Now the question is: "Whether, admitting the existence of
13729natural necessity in the world of phenomena, it is possible to consider
13730an effect as at the same time an effect of nature and an effect of
13731freedom--or, whether these two modes of causality are contradictory and
13732incompatible?"
13733
13734No phenomenal cause can absolutely and of itself begin a series. Every
13735action, in so far as it is productive of an event, is itself an event or
13736occurrence, and presupposes another preceding state, in which its cause
13737existed. Thus everything that happens is but a continuation of a series,
13738and an absolute beginning is impossible in the sensuous world. The
13739actions of natural causes are, accordingly, themselves effects, and
13740presuppose causes preceding them in time. A primal action which forms an
13741absolute beginning, is beyond the causal power of phenomena.
13742
13743Now, is it absolutely necessary that, granting that all effects are
13744phenomena, the causality of the cause of these effects must also be a
13745phenomenon and belong to the empirical world? Is it not rather possible
13746that, although every effect in the phenomenal world must be connected
13747with an empirical cause, according to the universal law of nature, this
13748empirical causality may be itself the effect of a non-empirical and
13749intelligible causality--its connection with natural causes remaining
13750nevertheless intact? Such a causality would be considered, in reference
13751to phenomena, as the primal action of a cause, which is in so far,
13752therefore, not phenomenal, but, by reason of this faculty or power,
13753intelligible; although it must, at the same time, as a link in the chain
13754of nature, be regarded as belonging to the sensuous world.
13755
13756A belief in the reciprocal causality of phenomena is necessary, if
13757we are required to look for and to present the natural conditions of
13758natural events, that is to say, their causes. This being admitted as
13759unexceptionably valid, the requirements of the understanding, which
13760recognizes nothing but nature in the region of phenomena, are satisfied,
13761and our physical explanations of physical phenomena may proceed in their
13762regular course, without hindrance and without opposition. But it is no
13763stumbling-block in the way, even assuming the idea to be a pure fiction,
13764to admit that there are some natural causes in the possession of a
13765faculty which is not empirical, but intelligible, inasmuch as it is not
13766determined to action by empirical conditions, but purely and solely upon
13767grounds brought forward by the understanding--this action being still,
13768when the cause is phenomenized, in perfect accordance with the laws of
13769empirical causality. Thus the acting subject, as a causal phenomenon,
13770would continue to preserve a complete connection with nature and
13771natural conditions; and the phenomenon only of the subject (with all
13772its phenomenal causality) would contain certain conditions, which, if we
13773ascend from the empirical to the transcendental object, must necessarily
13774be regarded as intelligible. For, if we attend, in our inquiries with
13775regard to causes in the world of phenomena, to the directions of nature
13776alone, we need not trouble ourselves about the relation in which the
13777transcendental subject, which is completely unknown to us, stands to
13778these phenomena and their connection in nature. The intelligible ground
13779of phenomena in this subject does not concern empirical questions. It
13780has to do only with pure thought; and, although the effects of this
13781thought and action of the pure understanding are discoverable in
13782phenomena, these phenomena must nevertheless be capable of a full and
13783complete explanation, upon purely physical grounds and in accordance
13784with natural laws. And in this case we attend solely to their empirical
13785and omit all consideration of their intelligible character (which is the
13786transcendental cause of the former) as completely unknown, except in so
13787far as it is exhibited by the latter as its empirical symbol. Now let us
13788apply this to experience. Man is a phenomenon of the sensuous world and,
13789at the same time, therefore, a natural cause, the causality of which
13790must be regulated by empirical laws. As such, he must possess an
13791empirical character, like all other natural phenomena. We remark this
13792empirical character in his actions, which reveal the presence of certain
13793powers and faculties. If we consider inanimate or merely animal nature,
13794we can discover no reason for ascribing to ourselves any other than a
13795faculty which is determined in a purely sensuous manner. But man, to
13796whom nature reveals herself only through sense, cognizes himself not
13797only by his senses, but also through pure apperception; and this in
13798actions and internal determinations, which he cannot regard as sensuous
13799impressions. He is thus to himself, on the one hand, a phenomenon,
13800but on the other hand, in respect of certain faculties, a purely
13801intelligible object--intelligible, because its action cannot be ascribed
13802to sensuous receptivity. These faculties are understanding and reason.
13803The latter, especially, is in a peculiar manner distinct from all
13804empirically-conditioned faculties, for it employs ideas alone in the
13805consideration of its objects, and by means of these determines the
13806understanding, which then proceeds to make an empirical use of its
13807own conceptions, which, like the ideas of reason, are pure and
13808non-empirical.
13809
13810That reason possesses the faculty of causality, or that at least we are
13811compelled so to represent it, is evident from the imperatives, which in
13812the sphere of the practical we impose on many of our executive powers.
13813The words I ought express a species of necessity, and imply a connection
13814with grounds which nature does not and cannot present to the mind of
13815man. Understanding knows nothing in nature but that which is, or has
13816been, or will be. It would be absurd to say that anything in nature
13817ought to be other than it is in the relations of time in which it
13818stands; indeed, the ought, when we consider merely the course of nature,
13819has neither application nor meaning. The question, "What ought to happen
13820in the sphere of nature?" is just as absurd as the question, "What ought
13821to be the properties of a circle?" All that we are entitled to ask is,
13822"What takes place in nature?" or, in the latter case, "What are the
13823properties of a circle?"
13824
13825But the idea of an ought or of duty indicates a possible action, the
13826ground of which is a pure conception; while the ground of a merely
13827natural action is, on the contrary, always a phenomenon. This action
13828must certainly be possible under physical conditions, if it is
13829prescribed by the moral imperative ought; but these physical or natural
13830conditions do not concern the determination of the will itself, they
13831relate to its effects alone, and the consequences of the effect in the
13832world of phenomena. Whatever number of motives nature may present to
13833my will, whatever sensuous impulses--the moral ought it is beyond their
13834power to produce. They may produce a volition, which, so far from
13835being necessary, is always conditioned--a volition to which the ought
13836enunciated by reason, sets an aim and a standard, gives permission or
13837prohibition. Be the object what it may, purely sensuous--as pleasure,
13838or presented by pure reason--as good, reason will not yield to grounds
13839which have an empirical origin. Reason will not follow the order
13840of things presented by experience, but, with perfect spontaneity,
13841rearranges them according to ideas, with which it compels empirical
13842conditions to agree. It declares, in the name of these ideas, certain
13843actions to be necessary which nevertheless have not taken place and
13844which perhaps never will take place; and yet presupposes that it
13845possesses the faculty of causality in relation to these actions. For,
13846in the absence of this supposition, it could not expect its ideas to
13847produce certain effects in the world of experience.
13848
13849Now, let us stop here and admit it to be at least possible that reason
13850does stand in a really causal relation to phenomena. In this case it
13851must--pure reason as it is--exhibit an empirical character. For every
13852cause supposes a rule, according to which certain phenomena follow as
13853effects from the cause, and every rule requires uniformity in these
13854effects; and this is the proper ground of the conception of a cause--as
13855a faculty or power. Now this conception (of a cause) may be termed the
13856empirical character of reason; and this character is a permanent one,
13857while the effects produced appear, in conformity with the various
13858conditions which accompany and partly limit them, in various forms.
13859
13860Thus the volition of every man has an empirical character, which is
13861nothing more than the causality of his reason, in so far as its effects
13862in the phenomenal world manifest the presence of a rule, according to
13863which we are enabled to examine, in their several kinds and degrees, the
13864actions of this causality and the rational grounds for these actions,
13865and in this way to decide upon the subjective principles of the
13866volition. Now we learn what this empirical character is only from
13867phenomenal effects, and from the rule of these which is presented by
13868experience; and for this reason all the actions of man in the world
13869of phenomena are determined by his empirical character, and the
13870co-operative causes of nature. If, then, we could investigate all the
13871phenomena of human volition to their lowest foundation in the mind,
13872there would be no action which we could not anticipate with certainty,
13873and recognize to be absolutely necessary from its preceding conditions.
13874So far as relates to this empirical character, therefore, there can be
13875no freedom; and it is only in the light of this character that we can
13876consider the human will, when we confine ourselves to simple observation
13877and, as is the case in anthropology, institute a physiological
13878investigation of the motive causes of human actions.
13879
13880But when we consider the same actions in relation to reason--not for the
13881purpose of explaining their origin, that is, in relation to speculative
13882reason, but to practical reason, as the producing cause of these
13883actions--we shall discover a rule and an order very different from those
13884of nature and experience. For the declaration of this mental faculty may
13885be that what has and could not but take place in the course of nature,
13886ought not to have taken place. Sometimes, too, we discover, or believe
13887that we discover, that the ideas of reason did actually stand in a
13888causal relation to certain actions of man; and that these actions have
13889taken place because they were determined, not by empirical causes, but
13890by the act of the will upon grounds of reason.
13891
13892Now, granting that reason stands in a causal relation to phenomena; can
13893an action of reason be called free, when we know that, sensuously, in
13894its empirical character, it is completely determined and absolutely
13895necessary? But this empirical character is itself determined by the
13896intelligible character. The latter we cannot cognize; we can only
13897indicate it by means of phenomena, which enable us to have an immediate
13898cognition only of the empirical character.* An action, then, in so far
13899as it is to be ascribed to an intelligible cause, does not result from
13900it in accordance with empirical laws. That is to say, not the conditions
13901of pure reason, but only their effects in the internal sense, precede
13902the act. Pure reason, as a purely intelligible faculty, is not subject
13903to the conditions of time. The causality of reason in its intelligible
13904character does not begin to be; it does not make its appearance at a
13905certain time, for the purpose of producing an effect. If this were not
13906the case, the causality of reason would be subservient to the natural
13907law of phenomena, which determines them according to time, and as a
13908series of causes and effects in time; it would consequently cease to
13909be freedom and become a part of nature. We are therefore justified in
13910saying: "If reason stands in a causal relation to phenomena, it is a
13911faculty which originates the sensuous condition of an empirical
13912series of effects." For the condition, which resides in the reason, is
13913non-sensuous, and therefore cannot be originated, or begin to be. And
13914thus we find--what we could not discover in any empirical series--a
13915condition of a successive series of events itself empirically
13916unconditioned. For, in the present case, the condition stands out of and
13917beyond the series of phenomena--it is intelligible, and it
13918consequently cannot be subjected to any sensuous condition, or to any
13919time-determination by a preceding cause.
13920
13921
13922 [*Footnote: The real morality of actions--their merit or demerit, and
13923even that of our own conduct, is completely unknown to us. Our estimates
13924can relate only to their empirical character. How much is the result
13925of the action of free will, how much is to be ascribed to nature and
13926to blameless error, or to a happy constitution of temperament (merito
13927fortunae), no one can discover, nor, for this reason, determine with
13928perfect justice.]
13929
13930
13931But, in another respect, the same cause belongs also to the series
13932of phenomena. Man is himself a phenomenon. His will has an empirical
13933character, which is the empirical cause of all his actions. There is
13934no condition--determining man and his volition in conformity with this
13935character--which does not itself form part of the series of effects
13936in nature, and is subject to their law--the law according to which an
13937empirically undetermined cause of an event in time cannot exist.
13938For this reason no given action can have an absolute and spontaneous
13939origination, all actions being phenomena, and belonging to the world of
13940experience. But it cannot be said of reason, that the state in which it
13941determines the will is always preceded by some other state determining
13942it. For reason is not a phenomenon, and therefore not subject to
13943sensuous conditions; and, consequently, even in relation to its
13944causality, the sequence or conditions of time do not influence reason,
13945nor can the dynamical law of nature, which determines the sequence of
13946time according to certain rules, be applied to it.
13947
13948Reason is consequently the permanent condition of all actions of the
13949human will. Each of these is determined in the empirical character of
13950the man, even before it has taken place. The intelligible character, of
13951which the former is but the sensuous schema, knows no before or after;
13952and every action, irrespective of the time-relation in which it stands
13953with other phenomena, is the immediate effect of the intelligible
13954character of pure reason, which, consequently, enjoys freedom of
13955action, and is not dynamically determined either by internal or external
13956preceding conditions. This freedom must not be described, in a merely
13957negative manner, as independence of empirical conditions, for in this
13958case the faculty of reason would cease to be a cause of phenomena; but
13959it must be regarded, positively, as a faculty which can spontaneously
13960originate a series of events. At the same time, it must not be supposed
13961that any beginning can take place in reason; on the contrary, reason,
13962as the unconditioned condition of all action of the will, admits of no
13963time-conditions, although its effect does really begin in a series of
13964phenomena--a beginning which is not, however, absolutely primal.
13965
13966I shall illustrate this regulative principle of reason by an example,
13967from its employment in the world of experience; proved it cannot be by
13968any amount of experience, or by any number of facts, for such arguments
13969cannot establish the truth of transcendental propositions. Let us take a
13970voluntary action--for example, a falsehood--by means of which a man
13971has introduced a certain degree of confusion into the social life
13972of humanity, which is judged according to the motives from which it
13973originated, and the blame of which and of the evil consequences arising
13974from it, is imputed to the offender. We at first proceed to examine the
13975empirical character of the offence, and for this purpose we endeavour
13976to penetrate to the sources of that character, such as a defective
13977education, bad company, a shameless and wicked disposition, frivolity,
13978and want of reflection--not forgetting also the occasioning causes which
13979prevailed at the moment of the transgression. In this the procedure is
13980exactly the same as that pursued in the investigation of the series of
13981causes which determine a given physical effect. Now, although we believe
13982the action to have been determined by all these circumstances, we do
13983not the less blame the offender. We do not blame him for his unhappy
13984disposition, nor for the circumstances which influenced him, nay, not
13985even for his former course of life; for we presuppose that all these
13986considerations may be set aside, that the series of preceding conditions
13987may be regarded as having never existed, and that the action may
13988be considered as completely unconditioned in relation to any state
13989preceding, just as if the agent commenced with it an entirely new series
13990of effects. Our blame of the offender is grounded upon a law of reason,
13991which requires us to regard this faculty as a cause, which could have
13992and ought to have otherwise determined the behaviour of the culprit,
13993independently of all empirical conditions. This causality of reason we
13994do not regard as a co-operating agency, but as complete in itself. It
13995matters not whether the sensuous impulses favoured or opposed the
13996action of this causality, the offence is estimated according to its
13997intelligible character--the offender is decidedly worthy of blame, the
13998moment he utters a falsehood. It follows that we regard reason, in
13999spite of the empirical conditions of the act, as completely free, and
14000therefore, therefore, as in the present case, culpable.
14001
14002The above judgement is complete evidence that we are accustomed to think
14003that reason is not affected by sensuous conditions, that in it no change
14004takes place--although its phenomena, in other words, the mode in
14005which it appears in its effects, are subject to change--that in it no
14006preceding state determines the following, and, consequently, that
14007it does not form a member of the series of sensuous conditions which
14008necessitate phenomena according to natural laws. Reason is present and
14009the same in all human actions and at all times; but it does not itself
14010exist in time, and therefore does not enter upon any state in which it
14011did not formerly exist. It is, relatively to new states or conditions,
14012determining, but not determinable. Hence we cannot ask: "Why did not
14013reason determine itself in a different manner?" The question ought to
14014be thus stated: "Why did not reason employ its power of causality
14015to determine certain phenomena in a different manner?" But this is
14016a question which admits of no answer. For a different intelligible
14017character would have exhibited a different empirical character; and,
14018when we say that, in spite of the course which his whole former life has
14019taken, the offender could have refrained from uttering the falsehood,
14020this means merely that the act was subject to the power and
14021authority--permissive or prohibitive--of reason. Now, reason is not
14022subject in its causality to any conditions of phenomena or of time; and
14023a difference in time may produce a difference in the relation of
14024phenomena to each other--for these are not things and therefore not
14025causes in themselves--but it cannot produce any difference in the
14026relation in which the action stands to the faculty of reason.
14027
14028Thus, then, in our investigation into free actions and the causal power
14029which produced them, we arrive at an intelligible cause, beyond which,
14030however, we cannot go; although we can recognize that it is free, that
14031is, independent of all sensuous conditions, and that, in this way, it
14032may be the sensuously unconditioned condition of phenomena. But for what
14033reason the intelligible character generates such and such phenomena
14034and exhibits such and such an empirical character under certain
14035circumstances, it is beyond the power of our reason to decide. The
14036question is as much above the power and the sphere of reason as the
14037following would be: "Why does the transcendental object of our external
14038sensuous intuition allow of no other form than that of intuition in
14039space?" But the problem, which we were called upon to solve, does not
14040require us to entertain any such questions. The problem was merely
14041this--whether freedom and natural necessity can exist without opposition
14042in the same action. To this question we have given a sufficient
14043answer; for we have shown that, as the former stands in a relation to a
14044different kind of condition from those of the latter, the law of the one
14045does not affect the law of the other and that, consequently, both can
14046exist together in independence of and without interference with each
14047other.
14048
14049The reader must be careful to remark that my intention in the above
14050remarks has not been to prove the actual existence of freedom, as a
14051faculty in which resides the cause of certain sensuous phenomena. For,
14052not to mention that such an argument would not have a transcendental
14053character, nor have been limited to the discussion of pure
14054conceptions--all attempts at inferring from experience what cannot be
14055cogitated in accordance with its laws, must ever be unsuccessful. Nay,
14056more, I have not even aimed at demonstrating the possibility of freedom;
14057for this too would have been a vain endeavour, inasmuch as it is beyond
14058the power of the mind to cognize the possibility of a reality or of a
14059causal power by the aid of mere a priori conceptions. Freedom has been
14060considered in the foregoing remarks only as a transcendental idea, by
14061means of which reason aims at originating a series of conditions in
14062the world of phenomena with the help of that which is sensuously
14063unconditioned, involving itself, however, in an antinomy with the laws
14064which itself prescribes for the conduct of the understanding. That this
14065antinomy is based upon a mere illusion, and that nature and freedom are
14066at least not opposed--this was the only thing in our power to prove, and
14067the question which it was our task to solve.
14068
14069
14070
14071IV. Solution of the Cosmological Idea of the Totality of the
14072Dependence of Phenomenal Existences.
14073
14074In the preceding remarks, we considered the changes in the world of
14075sense as constituting a dynamical series, in which each member is
14076subordinated to another--as its cause. Our present purpose is to avail
14077ourselves of this series of states or conditions as a guide to
14078an existence which may be the highest condition of all changeable
14079phenomena, that is, to a necessary being. Our endeavour to reach,
14080not the unconditioned causality, but the unconditioned existence, of
14081substance. The series before us is therefore a series of conceptions,
14082and not of intuitions (in which the one intuition is the condition of
14083the other).
14084
14085But it is evident that, as all phenomena are subject to change and
14086conditioned in their existence, the series of dependent existences
14087cannot embrace an unconditioned member, the existence of which would
14088be absolutely necessary. It follows that, if phenomena were things
14089in themselves, and--as an immediate consequence from this
14090supposition--condition and conditioned belonged to the same series of
14091phenomena, the existence of a necessary being, as the condition of the
14092existence of sensuous phenomena, would be perfectly impossible.
14093
14094An important distinction, however, exists between the dynamical and the
14095mathematical regress. The latter is engaged solely with the combination
14096of parts into a whole, or with the division of a whole into its parts;
14097and therefore are the conditions of its series parts of the series,
14098and to be consequently regarded as homogeneous, and for this reason, as
14099consisting, without exception, of phenomena. If the former regress, on
14100the contrary, the aim of which is not to establish the possibility of
14101an unconditioned whole consisting of given parts, or of an unconditioned
14102part of a given whole, but to demonstrate the possibility of the
14103deduction of a certain state from its cause, or of the contingent
14104existence of substance from that which exists necessarily, it is not
14105requisite that the condition should form part of an empirical series
14106along with the conditioned.
14107
14108In the case of the apparent antinomy with which we are at present
14109dealing, there exists a way of escape from the difficulty; for it is
14110not impossible that both of the contradictory statements may be true
14111in different relations. All sensuous phenomena may be contingent, and
14112consequently possess only an empirically conditioned existence, and yet
14113there may also exist a non-empirical condition of the whole series,
14114or, in other words, a necessary being. For this necessary being, as an
14115intelligible condition, would not form a member--not even the highest
14116member--of the series; the whole world of sense would be left in its
14117empirically determined existence uninterfered with and uninfluenced.
14118This would also form a ground of distinction between the modes of
14119solution employed for the third and fourth antinomies. For, while in the
14120consideration of freedom in the former antinomy, the thing itself--the
14121cause (substantia phaenomenon)--was regarded as belonging to the series
14122of conditions, and only its causality to the intelligible world--we are
14123obliged in the present case to cogitate this necessary being as purely
14124intelligible and as existing entirely apart from the world of sense
14125(as an ens extramundanum); for otherwise it would be subject to the
14126phenomenal law of contingency and dependence.
14127
14128In relation to the present problem, therefore, the regulative principle
14129of reason is that everything in the sensuous world possesses an
14130empirically conditioned existence--that no property of the sensuous
14131world possesses unconditioned necessity--that we are bound to expect,
14132and, so far as is possible, to seek for the empirical condition of every
14133member in the series of conditions--and that there is no sufficient
14134reason to justify us in deducing any existence from a condition which
14135lies out of and beyond the empirical series, or in regarding any
14136existence as independent and self-subsistent; although this should not
14137prevent us from recognizing the possibility of the whole series being
14138based upon a being which is intelligible, and for this reason free from
14139all empirical conditions.
14140
14141But it has been far from my intention, in these remarks, to prove the
14142existence of this unconditioned and necessary being, or even to evidence
14143the possibility of a purely intelligible condition of the existence or
14144all sensuous phenomena. As bounds were set to reason, to prevent it from
14145leaving the guiding thread of empirical conditions and losing itself in
14146transcendent theories which are incapable of concrete presentation; so
14147it was my purpose, on the other band, to set bounds to the law of the
14148purely empirical understanding, and to protest against any attempts
14149on its part at deciding on the possibility of things, or declaring the
14150existence of the intelligible to be impossible, merely on the ground
14151that it is not available for the explanation and exposition of
14152phenomena. It has been shown, at the same time, that the contingency
14153of all the phenomena of nature and their empirical conditions is quite
14154consistent with the arbitrary hypothesis of a necessary, although purely
14155intelligible condition, that no real contradiction exists between them
14156and that, consequently, both may be true. The existence of such an
14157absolutely necessary being may be impossible; but this can never be
14158demonstrated from the universal contingency and dependence of sensuous
14159phenomena, nor from the principle which forbids us to discontinue the
14160series at some member of it, or to seek for its cause in some sphere
14161of existence beyond the world of nature. Reason goes its way in the
14162empirical world, and follows, too, its peculiar path in the sphere of
14163the transcendental.
14164
14165The sensuous world contains nothing but phenomena, which are mere
14166representations, and always sensuously conditioned; things in themselves
14167are not, and cannot be, objects to us. It is not to be wondered at,
14168therefore, that we are not justified in leaping from some member of
14169an empirical series beyond the world of sense, as if empirical
14170representations were things in themselves, existing apart from their
14171transcendental ground in the human mind, and the cause of whose
14172existence may be sought out of the empirical series. This would
14173certainly be the case with contingent things; but it cannot be with mere
14174representations of things, the contingency of which is itself merely a
14175phenomenon and can relate to no other regress than that which determines
14176phenomena, that is, the empirical. But to cogitate an intelligible
14177ground of phenomena, as free, moreover, from the contingency of the
14178latter, conflicts neither with the unlimited nature of the empirical
14179regress, nor with the complete contingency of phenomena. And the
14180demonstration of this was the only thing necessary for the solution of
14181this apparent antinomy. For if the condition of every conditioned--as
14182regards its existence--is sensuous, and for this reason a part of
14183the same series, it must be itself conditioned, as was shown in the
14184antithesis of the fourth antinomy. The embarrassments into which a
14185reason, which postulates the unconditioned, necessarily falls, must,
14186therefore, continue to exist; or the unconditioned must be placed in the
14187sphere of the intelligible. In this way, its necessity does not require,
14188nor does it even permit, the presence of an empirical condition: and it
14189is, consequently, unconditionally necessary.
14190
14191The empirical employment of reason is not affected by the assumption
14192of a purely intelligible being; it continues its operations on the
14193principle of the contingency of all phenomena, proceeding from empirical
14194conditions to still higher and higher conditions, themselves empirical.
14195Just as little does this regulative principle exclude the assumption
14196of an intelligible cause, when the question regards merely the pure
14197employment of reason--in relation to ends or aims. For, in this case, an
14198intelligible cause signifies merely the transcendental and to us unknown
14199ground of the possibility of sensuous phenomena, and its existence,
14200necessary and independent of all sensuous conditions, is not
14201inconsistent with the contingency of phenomena, or with the unlimited
14202possibility of regress which exists in the series of empirical
14203conditions.
14204
14205
14206
14207Concluding Remarks on the Antinomy of Pure Reason.
14208
14209So long as the object of our rational conceptions is the
14210totality of conditions in the world of phenomena, and the
14211satisfaction, from this source, of the requirements of reason,
14212so long are our ideas transcendental and cosmological.
14213But when we set the unconditioned--which is the aim of
14214all our inquiries--in a sphere which lies out of the
14215world of sense and possible experience, our ideas become transcendent.
14216They are then not merely serviceable towards the completion of the
14217exercise of reason (which remains an idea, never executed, but always
14218to be pursued); they detach themselves completely from experience and
14219construct for themselves objects, the material of which has not been
14220presented by experience, and the objective reality of which is not based
14221upon the completion of the empirical series, but upon pure a priori
14222conceptions. The intelligible object of these transcendent ideas may
14223be conceded, as a transcendental object. But we cannot cogitate it as
14224a thing determinable by certain distinct predicates relating to its
14225internal nature, for it has no connection with empirical conceptions;
14226nor are we justified in affirming the existence of any such object.
14227It is, consequently, a mere product of the mind alone. Of all the
14228cosmological ideas, however, it is that occasioning the fourth antinomy
14229which compels us to venture upon this step. For the existence of
14230phenomena, always conditioned and never self-subsistent, requires us
14231to look for an object different from phenomena--an intelligible object,
14232with which all contingency must cease. But, as we have allowed ourselves
14233to assume the existence of a self-subsistent reality out of the field
14234of experience, and are therefore obliged to regard phenomena as merely a
14235contingent mode of representing intelligible objects employed by beings
14236which are themselves intelligences--no other course remains for us than
14237to follow analogy and employ the same mode in forming some conception
14238of intelligible things, of which we have not the least knowledge, which
14239nature taught us to use in the formation of empirical conceptions.
14240Experience made us acquainted with the contingent. But we are at present
14241engaged in the discussion of things which are not objects of experience;
14242and must, therefore, deduce our knowledge of them from that which is
14243necessary absolutely and in itself, that is, from pure conceptions.
14244Hence the first step which we take out of the world of sense obliges
14245us to begin our system of new cognition with the investigation of
14246a necessary being, and to deduce from our conceptions of it all our
14247conceptions of intelligible things. This we propose to attempt in the
14248following chapter.
14249
14250
14251
14252CHAPTER III. The Ideal of Pure Reason.
14253
14254SECTION I. Of the Ideal in General.
14255
14256We have seen that pure conceptions do not present objects to the mind,
14257except under sensuous conditions; because the conditions of objective
14258reality do not exist in these conceptions, which contain, in fact,
14259nothing but the mere form of thought. They may, however, when applied to
14260phenomena, be presented in concreto; for it is phenomena that present to
14261them the materials for the formation of empirical conceptions, which
14262are nothing more than concrete forms of the conceptions of the
14263understanding. But ideas are still further removed from objective
14264reality than categories; for no phenomenon can ever present them to the
14265human mind in concreto. They contain a certain perfection, attainable
14266by no possible empirical cognition; and they give to reason a systematic
14267unity, to which the unity of experience attempts to approximate, but can
14268never completely attain.
14269
14270But still further removed than the idea from objective reality is the
14271Ideal, by which term I understand the idea, not in concreto, but in
14272individuo--as an individual thing, determinable or determined by the
14273idea alone. The idea of humanity in its complete perfection supposes not
14274only the advancement of all the powers and faculties, which constitute
14275our conception of human nature, to a complete attainment of their
14276final aims, but also everything which is requisite for the complete
14277determination of the idea; for of all contradictory predicates, only
14278one can conform with the idea of the perfect man. What I have termed
14279an ideal was in Plato's philosophy an idea of the divine mind--an
14280individual object present to its pure intuition, the most perfect of
14281every kind of possible beings, and the archetype of all phenomenal
14282existences.
14283
14284Without rising to these speculative heights, we are bound to confess
14285that human reason contains not only ideas, but ideals, which possess,
14286not, like those of Plato, creative, but certainly practical power--as
14287regulative principles, and form the basis of the perfectibility of
14288certain actions. Moral conceptions are not perfectly pure conceptions of
14289reason, because an empirical element--of pleasure or pain--lies at the
14290foundation of them. In relation, however, to the principle, whereby
14291reason sets bounds to a freedom which is in itself without law, and
14292consequently when we attend merely to their form, they may be considered
14293as pure conceptions of reason. Virtue and wisdom in their perfect purity
14294are ideas. But the wise man of the Stoics is an ideal, that is to say, a
14295human being existing only in thought and in complete conformity with the
14296idea of wisdom. As the idea provides a rule, so the ideal serves as an
14297archetype for the perfect and complete determination of the copy. Thus
14298the conduct of this wise and divine man serves us as a standard of
14299action, with which we may compare and judge ourselves, which may help
14300us to reform ourselves, although the perfection it demands can never be
14301attained by us. Although we cannot concede objective reality to these
14302ideals, they are not to be considered as chimeras; on the contrary,
14303they provide reason with a standard, which enables it to estimate, by
14304comparison, the degree of incompleteness in the objects presented to
14305it. But to aim at realizing the ideal in an example in the world of
14306experience--to describe, for instance, the character of the perfectly
14307wise man in a romance--is impracticable. Nay more, there is something
14308absurd in the attempt; and the result must be little edifying, as
14309the natural limitations, which are continually breaking in upon the
14310perfection and completeness of the idea, destroy the illusion in the
14311story and throw an air of suspicion even on what is good in the idea,
14312which hence appears fictitious and unreal.
14313
14314Such is the constitution of the ideal of reason, which is always based
14315upon determinate conceptions, and serves as a rule and a model for
14316limitation or of criticism. Very different is the nature of the ideals
14317of the imagination. Of these it is impossible to present an intelligible
14318conception; they are a kind of monogram, drawn according to no
14319determinate rule, and forming rather a vague picture--the production of
14320many diverse experiences--than a determinate image. Such are the ideals
14321which painters and physiognomists profess to have in their minds, and
14322which can serve neither as a model for production nor as a standard for
14323appreciation. They may be termed, though improperly, sensuous ideals, as
14324they are declared to be models of certain possible empirical intuitions.
14325They cannot, however, furnish rules or standards for explanation or
14326examination.
14327
14328In its ideals, reason aims at complete and perfect determination
14329according to a priori rules; and hence it cogitates an object, which
14330must be completely determinable in conformity with principles, although
14331all empirical conditions are absent, and the conception of the object is
14332on this account transcendent.
14333
14334
14335
14336SECTION II. Of the Transcendental Ideal (Prototypon Trancendentale).
14337
14338Every conception is, in relation to that which is not contained in
14339it, undetermined and subject to the principle of determinability. This
14340principle is that, of every two contradictorily opposed predicates, only
14341one can belong to a conception. It is a purely logical principle, itself
14342based upon the principle of contradiction; inasmuch as it makes complete
14343abstraction of the content and attends merely to the logical form of the
14344cognition.
14345
14346But again, everything, as regards its possibility, is also subject to
14347the principle of complete determination, according to which one of all
14348the possible contradictory predicates of things must belong to it.
14349This principle is not based merely upon that of contradiction; for,
14350in addition to the relation between two contradictory predicates,
14351it regards everything as standing in a relation to the sum of
14352possibilities, as the sum total of all predicates of things, and, while
14353presupposing this sum as an a priori condition, presents to the mind
14354everything as receiving the possibility of its individual existence from
14355the relation it bears to, and the share it possesses in, the aforesaid
14356sum of possibilities.* The principle of complete determination relates
14357the content and not to the logical form. It is the principle of the
14358synthesis of all the predicates which are required to constitute the
14359complete conception of a thing, and not a mere principle analytical
14360representation, which enounces that one of two contradictory predicates
14361must belong to a conception. It contains, moreover, a transcendental
14362presupposition--that, namely, of the material for all possibility, which
14363must contain a priori the data for this or that particular possibility.
14364
14365
14366 [*Footnote: Thus this principle declares everything to possess a
14367relation to a common correlate--the sum-total of possibility, which, if
14368discovered to exist in the idea of one individual thing, would establish
14369the affinity of all possible things, from the identity of the ground of
14370their complete determination. The determinability of every conception
14371is subordinate to the universality (Allgemeinheit, universalitas) of
14372the principle of excluded middle; the determination of a thing to the
14373totality (Allheit, universitas) of all possible predicates.]
14374
14375
14376The proposition, everything which exists is completely determined, means
14377not only that one of every pair of given contradictory attributes, but
14378that one of all possible attributes, is always predicable of the thing;
14379in it the predicates are not merely compared logically with each other,
14380but the thing itself is transcendentally compared with the sum-total of
14381all possible predicates. The proposition is equivalent to saying: "To
14382attain to a complete knowledge of a thing, it is necessary to possess a
14383knowledge of everything that is possible, and to determine it thereby in
14384a positive or negative manner." The conception of complete determination
14385is consequently a conception which cannot be presented in its totality
14386in concreto, and is therefore based upon an idea, which has its seat in
14387the reason--the faculty which prescribes to the understanding the laws
14388of its harmonious and perfect exercise.
14389
14390Now, although this idea of the sum-total of all possibility, in so far
14391as it forms the condition of the complete determination of everything,
14392is itself undetermined in relation to the predicates which may
14393constitute this sum-total, and we cogitate in it merely the sum-total of
14394all possible predicates--we nevertheless find, upon closer examination,
14395that this idea, as a primitive conception of the mind, excludes a
14396large number of predicates--those deduced and those irreconcilable with
14397others, and that it is evolved as a conception completely determined a
14398priori. Thus it becomes the conception of an individual object, which
14399is completely determined by and through the mere idea, and must
14400consequently be termed an ideal of pure reason.
14401
14402When we consider all possible predicates, not merely logically, but
14403transcendentally, that is to say, with reference to the content which
14404may be cogitated as existing in them a priori, we shall find that
14405some indicate a being, others merely a non-being. The logical negation
14406expressed in the word not does not properly belong to a conception, but
14407only to the relation of one conception to another in a judgement, and is
14408consequently quite insufficient to present to the mind the content of a
14409conception. The expression not mortal does not indicate that a non-being
14410is cogitated in the object; it does not concern the content at all. A
14411transcendental negation, on the contrary, indicates non-being in itself,
14412and is opposed to transcendental affirmation, the conception of which
14413of itself expresses a being. Hence this affirmation indicates a reality,
14414because in and through it objects are considered to be something--to be
14415things; while the opposite negation, on the other band, indicates a
14416mere want, or privation, or absence, and, where such negations alone
14417are attached to a representation, the non-existence of anything
14418corresponding to the representation.
14419
14420Now a negation cannot be cogitated as determined, without cogitating at
14421the same time the opposite affirmation. The man born blind has not the
14422least notion of darkness, because he has none of light; the vagabond
14423knows nothing of poverty, because he has never known what it is to be in
14424comfort;* the ignorant man has no conception of his ignorance, because
14425he has no conception of knowledge. All conceptions of negatives are
14426accordingly derived or deduced conceptions; and realities contain the
14427data, and, so to speak, the material or transcendental content of the
14428possibility and complete determination of all things.
14429
14430
14431 [*Footnote: The investigations and calculations of astronomers have
14432taught us much that is wonderful; but the most important lesson we have
14433received from them is the discovery of the abyss of our ignorance in
14434relation to the universe--an ignorance the magnitude of which reason,
14435without the information thus derived, could never have conceived.
14436This discovery of our deficiencies must produce a great change in the
14437determination of the aims of human reason.]
14438
14439
14440If, therefore, a transcendental substratum lies at the foundation of the
14441complete determination of things--a substratum which is to form the fund
14442from which all possible predicates of things are to be supplied, this
14443substratum cannot be anything else than the idea of a sum-total of
14444reality (omnitudo realitatis). In this view, negations are nothing but
14445limitations--a term which could not, with propriety, be applied to
14446them, if the unlimited (the all) did not form the true basis of our
14447conception.
14448
14449This conception of a sum-total of reality is the conception of a thing
14450in itself, regarded as completely determined; and the conception of an
14451ens realissimum is the conception of an individual being, inasmuch as
14452it is determined by that predicate of all possible contradictory
14453predicates, which indicates and belongs to being. It is, therefore, a
14454transcendental ideal which forms the basis of the complete determination
14455of everything that exists, and is the highest material condition of
14456its possibility--a condition on which must rest the cogitation of all
14457objects with respect to their content. Nay, more, this ideal is the only
14458proper ideal of which the human mind is capable; because in this case
14459alone a general conception of a thing is completely determined by and
14460through itself, and cognized as the representation of an individuum.
14461
14462The logical determination of a conception is based upon a disjunctive
14463syllogism, the major of which contains the logical division of the
14464extent of a general conception, the minor limits this extent to a
14465certain part, while the conclusion determines the conception by this
14466part. The general conception of a reality cannot be divided a priori,
14467because, without the aid of experience, we cannot know any determinate
14468kinds of reality, standing under the former as the genus. The
14469transcendental principle of the complete determination of all things is
14470therefore merely the representation of the sum-total of all reality; it
14471is not a conception which is the genus of all predicates under
14472itself, but one which comprehends them all within itself. The complete
14473determination of a thing is consequently based upon the limitation of
14474this total of reality, so much being predicated of the thing, while all
14475that remains over is excluded--a procedure which is in exact agreement
14476with that of the disjunctive syllogism and the determination of the
14477objects in the conclusion by one of the members of the division.
14478It follows that reason, in laying the transcendental ideal at the
14479foundation of its determination of all possible things, takes a course
14480in exact analogy with that which it pursues in disjunctive syllogisms--a
14481proposition which formed the basis of the systematic division of all
14482transcendental ideas, according to which they are produced in complete
14483parallelism with the three modes of syllogistic reasoning employed by
14484the human mind.
14485
14486It is self-evident that reason, in cogitating the necessary complete
14487determination of things, does not presuppose the existence of a being
14488corresponding to its ideal, but merely the idea of the ideal--for
14489the purpose of deducing from the unconditional totality of complete
14490determination, The ideal is therefore the prototype of all things,
14491which, as defective copies (ectypa), receive from it the material of
14492their possibility, and approximate to it more or less, though it is
14493impossible that they can ever attain to its perfection.
14494
14495The possibility of things must therefore be regarded as derived--except
14496that of the thing which contains in itself all reality, which must be
14497considered to be primitive and original. For all negations--and they
14498are the only predicates by means of which all other things can be
14499distinguished from the ens realissimum--are mere limitations of a
14500greater and a higher--nay, the highest reality; and they consequently
14501presuppose this reality, and are, as regards their content, derived from
14502it. The manifold nature of things is only an infinitely various mode of
14503limiting the conception of the highest reality, which is their common
14504substratum; just as all figures are possible only as different modes of
14505limiting infinite space. The object of the ideal of reason--an object
14506existing only in reason itself--is also termed the primal being (ens
14507originarium); as having no existence superior to him, the supreme being
14508(ens summum); and as being the condition of all other beings, which rank
14509under it, the being of all beings (ens entium). But none of these terms
14510indicate the objective relation of an actually existing object to
14511other things, but merely that of an idea to conceptions; and all our
14512investigations into this subject still leave us in perfect uncertainty
14513with regard to the existence of this being.
14514
14515A primal being cannot be said to consist of many other beings with an
14516existence which is derivative, for the latter presuppose the former, and
14517therefore cannot be constitutive parts of it. It follows that the ideal
14518of the primal being must be cogitated as simple.
14519
14520The deduction of the possibility of all other things from this primal
14521being cannot, strictly speaking, be considered as a limitation, or as a
14522kind of division of its reality; for this would be regarding the primal
14523being as a mere aggregate--which has been shown to be impossible,
14524although it was so represented in our first rough sketch. The highest
14525reality must be regarded rather as the ground than as the sum-total
14526of the possibility of all things, and the manifold nature of things be
14527based, not upon the limitation of the primal being itself, but upon the
14528complete series of effects which flow from it. And thus all our powers
14529of sense, as well as all phenomenal reality, phenomenal reality, may be
14530with propriety regarded as belonging to this series of effects,
14531while they could not have formed parts of the idea, considered as an
14532aggregate. Pursuing this track, and hypostatizing this idea, we shall
14533find ourselves authorized to determine our notion of the Supreme Being
14534by means of the mere conception of a highest reality, as one, simple,
14535all-sufficient, eternal, and so on--in one word, to determine it in its
14536unconditioned completeness by the aid of every possible predicate.
14537The conception of such a being is the conception of God in its
14538transcendental sense, and thus the ideal of pure reason is the
14539object-matter of a transcendental theology.
14540
14541But, by such an employment of the transcendental idea, we should be over
14542stepping the limits of its validity and purpose. For reason placed
14543it, as the conception of all reality, at the basis of the complete
14544determination of things, without requiring that this conception be
14545regarded as the conception of an objective existence. Such an existence
14546would be purely fictitious, and the hypostatizing of the content of
14547the idea into an ideal, as an individual being, is a step perfectly
14548unauthorized. Nay, more, we are not even called upon to assume the
14549possibility of such an hypothesis, as none of the deductions drawn
14550from such an ideal would affect the complete determination of things in
14551general--for the sake of which alone is the idea necessary.
14552
14553It is not sufficient to circumscribe the procedure and the dialectic
14554of reason; we must also endeavour to discover the sources of this
14555dialectic, that we may have it in our power to give a rational
14556explanation of this illusion, as a phenomenon of the human mind. For
14557the ideal, of which we are at present speaking, is based, not upon an
14558arbitrary, but upon a natural, idea. The question hence arises: How
14559happens it that reason regards the possibility of all things as deduced
14560from a single possibility, that, to wit, of the highest reality, and
14561presupposes this as existing in an individual and primal being?
14562
14563The answer is ready; it is at once presented by the procedure of
14564transcendental analytic. The possibility of sensuous objects is a
14565relation of these objects to thought, in which something (the empirical
14566form) may be cogitated a priori; while that which constitutes the
14567matter--the reality of the phenomenon (that element which corresponds to
14568sensation)--must be given from without, as otherwise it could not even
14569be cogitated by, nor could its possibility be presentable to the mind.
14570Now, a sensuous object is completely determined, when it has been
14571compared with all phenomenal predicates, and represented by means of
14572these either positively or negatively. But, as that which constitutes
14573the thing itself--the real in a phenomenon, must be given, and that, in
14574which the real of all phenomena is given, is experience, one, sole, and
14575all-embracing--the material of the possibility of all sensuous objects
14576must be presupposed as given in a whole, and it is upon the limitation
14577of this whole that the possibility of all empirical objects, their
14578distinction from each other and their complete determination, are based.
14579Now, no other objects are presented to us besides sensuous objects, and
14580these can be given only in connection with a possible experience; it
14581follows that a thing is not an object to us, unless it presupposes
14582the whole or sum-total of empirical reality as the condition of
14583its possibility. Now, a natural illusion leads us to consider this
14584principle, which is valid only of sensuous objects, as valid with regard
14585to things in general. And thus we are induced to hold the empirical
14586principle of our conceptions of the possibility of things, as phenomena,
14587by leaving out this limitative condition, to be a transcendental
14588principle of the possibility of things in general.
14589
14590We proceed afterwards to hypostatize this idea of the sum-total of all
14591reality, by changing the distributive unity of the empirical exercise
14592of the understanding into the collective unity of an empirical whole--a
14593dialectical illusion, and by cogitating this whole or sum of experience
14594as an individual thing, containing in itself all empirical reality.
14595This individual thing or being is then, by means of the above-mentioned
14596transcendental subreption, substituted for our notion of a thing which
14597stands at the head of the possibility of all things, the real conditions
14598of whose complete determination it presents.*
14599
14600
14601 [*Footnote: This ideal of the ens realissimum--although merely a
14602mental representation--is first objectivized, that is, has an objective
14603existence attributed to it, then hypostatized, and finally, by the
14604natural progress of reason to the completion of unity, personified, as
14605we shall show presently. For the regulative unity of experience is not
14606based upon phenomena themselves, but upon the connection of the variety
14607of phenomena by the understanding in a consciousness, and thus the unity
14608of the supreme reality and the complete determinability of all things,
14609seem to reside in a supreme understanding, and, consequently, in a
14610conscious intelligence.]
14611
14612
14613
14614SECTION III. Of the Arguments employed by Speculative Reason in
14615Proof of the Existence of a Supreme Being.
14616
14617Notwithstanding the pressing necessity which reason feels, to form some
14618presupposition that shall serve the understanding as a proper basis
14619for the complete determination of its conceptions, the idealistic and
14620factitious nature of such a presupposition is too evident to allow
14621reason for a moment to persuade itself into a belief of the objective
14622existence of a mere creation of its own thought. But there are other
14623considerations which compel reason to seek out some resting place in the
14624regress from the conditioned to the unconditioned, which is not given
14625as an actual existence from the mere conception of it, although it
14626alone can give completeness to the series of conditions. And this is
14627the natural course of every human reason, even of the most uneducated,
14628although the path at first entered it does not always continue to
14629follow. It does not begin from conceptions, but from common experience,
14630and requires a basis in actual existence. But this basis is insecure,
14631unless it rests upon the immovable rock of the absolutely necessary. And
14632this foundation is itself unworthy of trust, if it leave under and above
14633it empty space, if it do not fill all, and leave no room for a why or a
14634wherefore, if it be not, in one word, infinite in its reality.
14635
14636If we admit the existence of some one thing, whatever it may be, we must
14637also admit that there is something which exists necessarily. For what is
14638contingent exists only under the condition of some other thing, which
14639is its cause; and from this we must go on to conclude the existence of a
14640cause which is not contingent, and which consequently exists necessarily
14641and unconditionally. Such is the argument by which reason justifies its
14642advances towards a primal being.
14643
14644Now reason looks round for the conception of a being that may be
14645admitted, without inconsistency, to be worthy of the attribute of
14646absolute necessity, not for the purpose of inferring a priori, from
14647the conception of such a being, its objective existence (for if reason
14648allowed itself to take this course, it would not require a basis in
14649given and actual existence, but merely the support of pure conceptions),
14650but for the purpose of discovering, among all our conceptions of
14651possible things, that conception which possesses no element inconsistent
14652with the idea of absolute necessity. For that there must be some
14653absolutely necessary existence, it regards as a truth already
14654established. Now, if it can remove every existence incapable of
14655supporting the attribute of absolute necessity, excepting one--this
14656must be the absolutely necessary being, whether its necessity is
14657comprehensible by us, that is, deducible from the conception of it
14658alone, or not.
14659
14660Now that, the conception of which contains a therefore to every
14661wherefore, which is not defective in any respect whatever, which is
14662all-sufficient as a condition, seems to be the being of which we can
14663justly predicate absolute necessity--for this reason, that, possessing
14664the conditions of all that is possible, it does not and cannot itself
14665require any condition. And thus it satisfies, in one respect at least,
14666the requirements of the conception of absolute necessity. In this
14667view, it is superior to all other conceptions, which, as deficient and
14668incomplete, do not possess the characteristic of independence of all
14669higher conditions. It is true that we cannot infer from this that what
14670does not contain in itself the supreme and complete condition--the
14671condition of all other things--must possess only a conditioned
14672existence; but as little can we assert the contrary, for this supposed
14673being does not possess the only characteristic which can enable reason
14674to cognize by means of an a priori conception the unconditioned and
14675necessary nature of its existence.
14676
14677The conception of an ens realissimum is that which best agrees with
14678the conception of an unconditioned and necessary being. The former
14679conception does not satisfy all the requirements of the latter; but
14680we have no choice, we are obliged to adhere to it, for we find that we
14681cannot do without the existence of a necessary being; and even although
14682we admit it, we find it out of our power to discover in the whole sphere
14683of possibility any being that can advance well-grounded claims to such a
14684distinction.
14685
14686The following is, therefore, the natural course of human reason. It
14687begins by persuading itself of the existence of some necessary being. In
14688this being it recognizes the characteristics of unconditioned existence.
14689It then seeks the conception of that which is independent of all
14690conditions, and finds it in that which is itself the sufficient
14691condition of all other things--in other words, in that which contains
14692all reality. But the unlimited all is an absolute unity, and is
14693conceived by the mind as a being one and supreme; and thus reason
14694concludes that the Supreme Being, as the primal basis of all things,
14695possesses an existence which is absolutely necessary.
14696
14697This conception must be regarded as in some degree satisfactory, if we
14698admit the existence of a necessary being, and consider that there exists
14699a necessity for a definite and final answer to these questions. In such
14700a case, we cannot make a better choice, or rather we have no choice at
14701all, but feel ourselves obliged to declare in favour of the absolute
14702unity of complete reality, as the highest source of the possibility
14703of things. But if there exists no motive for coming to a definite
14704conclusion, and we may leave the question unanswered till we have fully
14705weighed both sides--in other words, when we are merely called upon to
14706decide how much we happen to know about the question, and how much we
14707merely flatter ourselves that we know--the above conclusion does not
14708appear to be so great advantage, but, on the contrary, seems defective
14709in the grounds upon which it is supported.
14710
14711For, admitting the truth of all that has been said, that, namely, the
14712inference from a given existence (my own, for example) to the existence
14713of an unconditioned and necessary being is valid and unassailable;
14714that, in the second place, we must consider a being which contains all
14715reality, and consequently all the conditions of other things, to
14716be absolutely unconditioned; and admitting too, that we have thus
14717discovered the conception of a thing to which may be attributed, without
14718inconsistency, absolute necessity--it does not follow from all this that
14719the conception of a limited being, in which the supreme reality does not
14720reside, is therefore incompatible with the idea of absolute necessity.
14721For, although I do not discover the element of the unconditioned in the
14722conception of such a being--an element which is manifestly existent in
14723the sum-total of all conditions--I am not entitled to conclude that its
14724existence is therefore conditioned; just as I am not entitled to affirm,
14725in a hypothetical syllogism, that where a certain condition does not
14726exist (in the present, completeness, as far as pure conceptions are
14727concerned), the conditioned does not exist either. On the contrary,
14728we are free to consider all limited beings as likewise unconditionally
14729necessary, although we are unable to infer this from the general
14730conception which we have of them. Thus conducted, this argument is
14731incapable of giving us the least notion of the properties of a necessary
14732being, and must be in every respect without result.
14733
14734This argument continues, however, to possess a weight and an authority,
14735which, in spite of its objective insufficiency, it has never been
14736divested of. For, granting that certain responsibilities lie upon us,
14737which, as based on the ideas of reason, deserve to be respected and
14738submitted to, although they are incapable of a real or practical
14739application to our nature, or, in other words, would be responsibilities
14740without motives, except upon the supposition of a Supreme Being to give
14741effect and influence to the practical laws: in such a case we should be
14742bound to obey our conceptions, which, although objectively insufficient,
14743do, according to the standard of reason, preponderate over and are
14744superior to any claims that may be advanced from any other quarter.
14745The equilibrium of doubt would in this case be destroyed by a practical
14746addition; indeed, Reason would be compelled to condemn herself, if she
14747refused to comply with the demands of the judgement, no superior to
14748which we know--however defective her understanding of the grounds of
14749these demands might be.
14750
14751This argument, although in fact transcendental, inasmuch as it rests
14752upon the intrinsic insufficiency of the contingent, is so simple and
14753natural, that the commonest understanding can appreciate its value.
14754We see things around us change, arise, and pass away; they, or their
14755condition, must therefore have a cause. The same demand must again be
14756made of the cause itself--as a datum of experience. Now it is natural
14757that we should place the highest causality just where we place supreme
14758causality, in that being, which contains the conditions of all possible
14759effects, and the conception of which is so simple as that of an
14760all-embracing reality. This highest cause, then, we regard as absolutely
14761necessary, because we find it absolutely necessary to rise to it, and
14762do not discover any reason for proceeding beyond it. Thus, among all
14763nations, through the darkest polytheism glimmer some faint sparks of
14764monotheism, to which these idolaters have been led, not from reflection
14765and profound thought, but by the study and natural progress of the
14766common understanding.
14767
14768There are only three modes of proving the existence of a Deity, on the
14769grounds of speculative reason.
14770
14771All the paths conducting to this end begin either from determinate
14772experience and the peculiar constitution of the world of sense, and
14773rise, according to the laws of causality, from it to the highest
14774cause existing apart from the world--or from a purely indeterminate
14775experience, that is, some empirical existence--or abstraction is made of
14776all experience, and the existence of a supreme cause is concluded from a
14777priori conceptions alone. The first is the physico-theological argument,
14778the second the cosmological, the third the ontological. More there are
14779not, and more there cannot be.
14780
14781I shall show it is as unsuccessful on the one path--the empirical--as on
14782the other--the transcendental, and that it stretches its wings in vain,
14783to soar beyond the world of sense by the mere might of speculative
14784thought. As regards the order in which we must discuss those arguments,
14785it will be exactly the reverse of that in which reason, in the progress
14786of its development, attains to them--the order in which they are
14787placed above. For it will be made manifest to the reader that, although
14788experience presents the occasion and the starting-point, it is the
14789transcendental idea of reason which guides it in its pilgrimage and
14790is the goal of all its struggles. I shall therefore begin with an
14791examination of the transcendental argument, and afterwards inquire what
14792additional strength has accrued to this mode of proof from the addition
14793of the empirical element.
14794
14795
14796
14797SECTION IV. Of the Impossibility of an Ontological Proof of the
14798Existence of God.
14799
14800It is evident from what has been said that the conception of an
14801absolutely necessary being is a mere idea, the objective reality of
14802which is far from being established by the mere fact that it is a
14803need of reason. On the contrary, this idea serves merely to indicate a
14804certain unattainable perfection, and rather limits the operations
14805than, by the presentation of new objects, extends the sphere of the
14806understanding. But a strange anomaly meets us at the very threshold;
14807for the inference from a given existence in general to an absolutely
14808necessary existence seems to be correct and unavoidable, while the
14809conditions of the understanding refuse to aid us in forming any
14810conception of such a being.
14811
14812Philosophers have always talked of an absolutely necessary being,
14813and have nevertheless declined to take the trouble of conceiving
14814whether--and how--a being of this nature is even cogitable, not to
14815mention that its existence is actually demonstrable. A verbal definition
14816of the conception is certainly easy enough: it is something the
14817non-existence of which is impossible. But does this definition throw
14818any light upon the conditions which render it impossible to cogitate the
14819non-existence of a thing--conditions which we wish to ascertain, that we
14820may discover whether we think anything in the conception of such a
14821being or not? For the mere fact that I throw away, by means of the word
14822unconditioned, all the conditions which the understanding habitually
14823requires in order to regard anything as necessary, is very far from
14824making clear whether by means of the conception of the unconditionally
14825necessary I think of something, or really of nothing at all.
14826
14827Nay, more, this chance-conception, now become so current, many have
14828endeavoured to explain by examples which seemed to render any inquiries
14829regarding its intelligibility quite needless. Every geometrical
14830proposition--a triangle has three angles--it was said, is absolutely
14831necessary; and thus people talked of an object which lay out of the
14832sphere of our understanding as if it were perfectly plain what the
14833conception of such a being meant.
14834
14835All the examples adduced have been drawn, without exception, from
14836judgements, and not from things. But the unconditioned necessity of
14837a judgement does not form the absolute necessity of a thing. On the
14838contrary, the absolute necessity of a judgement is only a conditioned
14839necessity of a thing, or of the predicate in a judgement. The
14840proposition above-mentioned does not enounce that three angles
14841necessarily exist, but, upon condition that a triangle exists, three
14842angles must necessarily exist--in it. And thus this logical necessity
14843has been the source of the greatest delusions. Having formed an a
14844priori conception of a thing, the content of which was made to embrace
14845existence, we believed ourselves safe in concluding that, because
14846existence belongs necessarily to the object of the conception (that is,
14847under the condition of my positing this thing as given), the existence
14848of the thing is also posited necessarily, and that it is therefore
14849absolutely necessary--merely because its existence has been cogitated in
14850the conception.
14851
14852If, in an identical judgement, I annihilate the predicate in thought,
14853and retain the subject, a contradiction is the result; and hence I say,
14854the former belongs necessarily to the latter. But if I suppress both
14855subject and predicate in thought, no contradiction arises; for there is
14856nothing at all, and therefore no means of forming a contradiction. To
14857suppose the existence of a triangle and not that of its three angles,
14858is self-contradictory; but to suppose the non-existence of both triangle
14859and angles is perfectly admissible. And so is it with the conception of
14860an absolutely necessary being. Annihilate its existence in thought, and
14861you annihilate the thing itself with all its predicates; how then can
14862there be any room for contradiction? Externally, there is nothing
14863to give rise to a contradiction, for a thing cannot be necessary
14864externally; nor internally, for, by the annihilation or suppression of
14865the thing itself, its internal properties are also annihilated. God is
14866omnipotent--that is a necessary judgement. His omnipotence cannot be
14867denied, if the existence of a Deity is posited--the existence, that is,
14868of an infinite being, the two conceptions being identical. But when you
14869say, God does not exist, neither omnipotence nor any other predicate
14870is affirmed; they must all disappear with the subject, and in this
14871judgement there cannot exist the least self-contradiction.
14872
14873You have thus seen that when the predicate of a judgement is annihilated
14874in thought along with the subject, no internal contradiction can arise,
14875be the predicate what it may. There is no possibility of evading the
14876conclusion--you find yourselves compelled to declare: There are certain
14877subjects which cannot be annihilated in thought. But this is
14878nothing more than saying: There exist subjects which are absolutely
14879necessary--the very hypothesis which you are called upon to establish.
14880For I find myself unable to form the slightest conception of a thing
14881which when annihilated in thought with all its predicates, leaves
14882behind a contradiction; and contradiction is the only criterion of
14883impossibility in the sphere of pure a priori conceptions.
14884
14885Against these general considerations, the justice of which no one can
14886dispute, one argument is adduced, which is regarded as furnishing a
14887satisfactory demonstration from the fact. It is affirmed that there is
14888one and only one conception, in which the non-being or annihilation of
14889the object is self-contradictory, and this is the conception of an ens
14890realissimum. It possesses, you say, all reality, and you feel yourselves
14891justified in admitting the possibility of such a being. (This I am
14892willing to grant for the present, although the existence of a conception
14893which is not self-contradictory is far from being sufficient to prove
14894the possibility of an object.)* Now the notion of all reality embraces
14895in it that of existence; the notion of existence lies, therefore, in
14896the conception of this possible thing. If this thing is annihilated
14897in thought, the internal possibility of the thing is also annihilated,
14898which is self-contradictory.
14899
14900
14901 [*Footnote: A conception is always possible, if it is not
14902self-contradictory. This is the logical criterion of possibility,
14903distinguishing the object of such a conception from the nihil negativum.
14904But it may be, notwithstanding, an empty conception, unless the
14905objective reality of this synthesis, but which it is generated, is
14906demonstrated; and a proof of this kind must be based upon principles
14907of possible experience, and not upon the principle of analysis or
14908contradiction. This remark may be serviceable as a warning against
14909concluding, from the possibility of a conception--which is logical--the
14910possibility of a thing--which is real.]
14911
14912
14913I answer: It is absurd to introduce--under whatever term disguised--into
14914the conception of a thing, which is to be cogitated solely in reference
14915to its possibility, the conception of its existence. If this is
14916admitted, you will have apparently gained the day, but in reality have
14917enounced nothing but a mere tautology. I ask, is the proposition,
14918this or that thing (which I am admitting to be possible) exists, an
14919analytical or a synthetical proposition? If the former, there is no
14920addition made to the subject of your thought by the affirmation of its
14921existence; but then the conception in your minds is identical with
14922the thing itself, or you have supposed the existence of a thing to
14923be possible, and then inferred its existence from its internal
14924possibility--which is but a miserable tautology. The word reality in the
14925conception of the thing, and the word existence in the conception of the
14926predicate, will not help you out of the difficulty. For, supposing you
14927were to term all positing of a thing reality, you have thereby posited
14928the thing with all its predicates in the conception of the subject
14929and assumed its actual existence, and this you merely repeat in the
14930predicate. But if you confess, as every reasonable person must, that
14931every existential proposition is synthetical, how can it be
14932maintained that the predicate of existence cannot be denied without
14933contradiction?--a property which is the characteristic of analytical
14934propositions, alone.
14935
14936I should have a reasonable hope of putting an end for ever to this
14937sophistical mode of argumentation, by a strict definition of the
14938conception of existence, did not my own experience teach me that the
14939illusion arising from our confounding a logical with a real predicate (a
14940predicate which aids in the determination of a thing) resists almost all
14941the endeavours of explanation and illustration. A logical predicate may
14942be what you please, even the subject may be predicated of itself;
14943for logic pays no regard to the content of a judgement. But the
14944determination of a conception is a predicate, which adds to and enlarges
14945the conception. It must not, therefore, be contained in the conception.
14946
14947Being is evidently not a real predicate, that is, a conception of
14948something which is added to the conception of some other thing. It is
14949merely the positing of a thing, or of certain determinations in it.
14950Logically, it is merely the copula of a judgement. The proposition, God
14951is omnipotent, contains two conceptions, which have a certain object or
14952content; the word is, is no additional predicate--it merely indicates
14953the relation of the predicate to the subject. Now, if I take the subject
14954(God) with all its predicates (omnipotence being one), and say: God is,
14955or, There is a God, I add no new predicate to the conception of God,
14956I merely posit or affirm the existence of the subject with all its
14957predicates--I posit the object in relation to my conception. The content
14958of both is the same; and there is no addition made to the conception,
14959which expresses merely the possibility of the object, by my cogitating
14960the object--in the expression, it is--as absolutely given or existing.
14961Thus the real contains no more than the possible. A hundred real dollars
14962contain no more than a hundred possible dollars. For, as the latter
14963indicate the conception, and the former the object, on the supposition
14964that the content of the former was greater than that of the latter, my
14965conception would not be an expression of the whole object, and would
14966consequently be an inadequate conception of it. But in reckoning my
14967wealth there may be said to be more in a hundred real dollars than in a
14968hundred possible dollars--that is, in the mere conception of them.
14969For the real object--the dollars--is not analytically contained in my
14970conception, but forms a synthetical addition to my conception (which
14971is merely a determination of my mental state), although this objective
14972reality--this existence--apart from my conceptions, does not in the
14973least degree increase the aforesaid hundred dollars.
14974
14975By whatever and by whatever number of predicates--even to the complete
14976determination of it--I may cogitate a thing, I do not in the least
14977augment the object of my conception by the addition of the statement:
14978This thing exists. Otherwise, not exactly the same, but something more
14979than what was cogitated in my conception, would exist, and I could not
14980affirm that the exact object of my conception had real existence. If I
14981cogitate a thing as containing all modes of reality except one, the mode
14982of reality which is absent is not added to the conception of the thing
14983by the affirmation that the thing exists; on the contrary, the thing
14984exists--if it exist at all--with the same defect as that cogitated in
14985its conception; otherwise not that which was cogitated, but something
14986different, exists. Now, if I cogitate a being as the highest reality,
14987without defect or imperfection, the question still remains--whether this
14988being exists or not? For, although no element is wanting in the possible
14989real content of my conception, there is a defect in its relation to my
14990mental state, that is, I am ignorant whether the cognition of the object
14991indicated by the conception is possible a posteriori. And here the cause
14992of the present difficulty becomes apparent. If the question regarded an
14993object of sense merely, it would be impossible for me to confound the
14994conception with the existence of a thing. For the conception merely
14995enables me to cogitate an object as according with the general
14996conditions of experience; while the existence of the object permits me
14997to cogitate it as contained in the sphere of actual experience. At the
14998same time, this connection with the world of experience does not in the
14999least augment the conception, although a possible perception has been
15000added to the experience of the mind. But if we cogitate existence by the
15001pure category alone, it is not to be wondered at, that we should find
15002ourselves unable to present any criterion sufficient to distinguish it
15003from mere possibility.
15004
15005Whatever be the content of our conception of an object, it is necessary
15006to go beyond it, if we wish to predicate existence of the object. In the
15007case of sensuous objects, this is attained by their connection according
15008to empirical laws with some one of my perceptions; but there is no means
15009of cognizing the existence of objects of pure thought, because it must
15010be cognized completely a priori. But all our knowledge of existence (be
15011it immediately by perception, or by inferences connecting some object
15012with a perception) belongs entirely to the sphere of experience--which
15013is in perfect unity with itself; and although an existence out of
15014this sphere cannot be absolutely declared to be impossible, it is a
15015hypothesis the truth of which we have no means of ascertaining.
15016
15017The notion of a Supreme Being is in many respects a highly useful idea;
15018but for the very reason that it is an idea, it is incapable of enlarging
15019our cognition with regard to the existence of things. It is not even
15020sufficient to instruct us as to the possibility of a being which we
15021do not know to exist. The analytical criterion of possibility, which
15022consists in the absence of contradiction in propositions, cannot
15023be denied it. But the connection of real properties in a thing is a
15024synthesis of the possibility of which an a priori judgement cannot be
15025formed, because these realities are not presented to us specifically;
15026and even if this were to happen, a judgement would still be impossible,
15027because the criterion of the possibility of synthetical cognitions must
15028be sought for in the world of experience, to which the object of an idea
15029cannot belong. And thus the celebrated Leibnitz has utterly failed in
15030his attempt to establish upon a priori grounds the possibility of this
15031sublime ideal being.
15032
15033The celebrated ontological or Cartesian argument for the existence of
15034a Supreme Being is therefore insufficient; and we may as well hope
15035to increase our stock of knowledge by the aid of mere ideas, as the
15036merchant to augment his wealth by the addition of noughts to his cash
15037account.
15038
15039
15040
15041SECTION V. Of the Impossibility of a Cosmological Proof of the
15042Existence of God.
15043
15044It was by no means a natural course of proceeding, but, on the contrary,
15045an invention entirely due to the subtlety of the schools, to attempt
15046to draw from a mere idea a proof of the existence of an object
15047corresponding to it. Such a course would never have been pursued,
15048were it not for that need of reason which requires it to suppose the
15049existence of a necessary being as a basis for the empirical regress, and
15050that, as this necessity must be unconditioned and a priori, reason is
15051bound to discover a conception which shall satisfy, if possible, this
15052requirement, and enable us to attain to the a priori cognition of such
15053a being. This conception was thought to be found in the idea of an ens
15054realissimum, and thus this idea was employed for the attainment of a
15055better defined knowledge of a necessary being, of the existence of
15056which we were convinced, or persuaded, on other grounds. Thus reason was
15057seduced from her natural courage; and, instead of concluding with the
15058conception of an ens realissimum, an attempt was made to begin with it,
15059for the purpose of inferring from it that idea of a necessary existence
15060which it was in fact called in to complete. Thus arose that unfortunate
15061ontological argument, which neither satisfies the healthy common sense
15062of humanity, nor sustains the scientific examination of the philosopher.
15063
15064The cosmological proof, which we are about to examine, retains the
15065connection between absolute necessity and the highest reality; but,
15066instead of reasoning from this highest reality to a necessary existence,
15067like the preceding argument, it concludes from the given unconditioned
15068necessity of some being its unlimited reality. The track it pursues,
15069whether rational or sophistical, is at least natural, and not only goes
15070far to persuade the common understanding, but shows itself deserving of
15071respect from the speculative intellect; while it contains, at the
15072same time, the outlines of all the arguments employed in natural
15073theology--arguments which always have been, and still will be, in
15074use and authority. These, however adorned, and hid under whatever
15075embellishments of rhetoric and sentiment, are at bottom identical
15076with the arguments we are at present to discuss. This proof, termed by
15077Leibnitz the argumentum a contingentia mundi, I shall now lay before the
15078reader, and subject to a strict examination.
15079
15080It is framed in the following manner: If something exists, an
15081absolutely necessary being must likewise exist. Now I, at least, exist.
15082Consequently, there exists an absolutely necessary being. The minor
15083contains an experience, the major reasons from a general experience to
15084the existence of a necessary being.* Thus this argument really begins at
15085experience, and is not completely a priori, or ontological. The
15086object of all possible experience being the world, it is called the
15087cosmological proof. It contains no reference to any peculiar property
15088of sensuous objects, by which this world of sense might be distinguished
15089from other possible worlds; and in this respect it differs from the
15090physico-theological proof, which is based upon the consideration of the
15091peculiar constitution of our sensuous world.
15092
15093
15094 [*Footnote: This inference is too well known to require more detailed
15095discussion. It is based upon the spurious transcendental law of
15096causality, that everything which is contingent has a cause, which, if
15097itself contingent, must also have a cause; and so on, till the series of
15098subordinated causes must end with an absolutely necessary cause, without
15099which it would not possess completeness.]
15100
15101
15102The proof proceeds thus: A necessary being can be determined only in one
15103way, that is, it can be determined by only one of all possible opposed
15104predicates; consequently, it must be completely determined in and by its
15105conception. But there is only a single conception of a thing possible,
15106which completely determines the thing a priori: that is, the conception
15107of the ens realissimum. It follows that the conception of the ens
15108realissimum is the only conception by and in which we can cogitate a
15109necessary being. Consequently, a Supreme Being necessarily exists.
15110
15111In this cosmological argument are assembled so many sophistical
15112propositions that speculative reason seems to have exerted in it all
15113her dialectical skill to produce a transcendental illusion of the most
15114extreme character. We shall postpone an investigation of this argument
15115for the present, and confine ourselves to exposing the stratagem by
15116which it imposes upon us an old argument in a new dress, and appeals
15117to the agreement of two witnesses, the one with the credentials of pure
15118reason, and the other with those of empiricism; while, in fact, it is
15119only the former who has changed his dress and voice, for the purpose
15120of passing himself off for an additional witness. That it may possess
15121a secure foundation, it bases its conclusions upon experience, and thus
15122appears to be completely distinct from the ontological argument, which
15123places its confidence entirely in pure a priori conceptions. But this
15124experience merely aids reason in making one step--to the existence of a
15125necessary being. What the properties of this being are cannot be learned
15126from experience; and therefore reason abandons it altogether, and
15127pursues its inquiries in the sphere of pure conception, for the purpose
15128of discovering what the properties of an absolutely necessary being
15129ought to be, that is, what among all possible things contain the
15130conditions (requisita) of absolute necessity. Reason believes that
15131it has discovered these requisites in the conception of an ens
15132realissimum--and in it alone, and hence concludes: The ens realissimum
15133is an absolutely necessary being. But it is evident that reason has
15134here presupposed that the conception of an ens realissimum is perfectly
15135adequate to the conception of a being of absolute necessity, that is,
15136that we may infer the existence of the latter from that of the former--a
15137proposition which formed the basis of the ontological argument, and
15138which is now employed in the support of the cosmological argument,
15139contrary to the wish and professions of its inventors. For the existence
15140of an absolutely necessary being is given in conceptions alone. But if
15141I say: "The conception of the ens realissimum is a conception of this
15142kind, and in fact the only conception which is adequate to our idea of a
15143necessary being," I am obliged to admit, that the latter may be inferred
15144from the former. Thus it is properly the ontological argument which
15145figures in the cosmological, and constitutes the whole strength of the
15146latter; while the spurious basis of experience has been of no further
15147use than to conduct us to the conception of absolute necessity, being
15148utterly insufficient to demonstrate the presence of this attribute in
15149any determinate existence or thing. For when we propose to ourselves
15150an aim of this character, we must abandon the sphere of experience, and
15151rise to that of pure conceptions, which we examine with the purpose of
15152discovering whether any one contains the conditions of the possibility
15153of an absolutely necessary being. But if the possibility of such a being
15154is thus demonstrated, its existence is also proved; for we may then
15155assert that, of all possible beings there is one which possesses
15156the attribute of necessity--in other words, this being possesses an
15157absolutely necessary existence.
15158
15159All illusions in an argument are more easily detected when they are
15160presented in the formal manner employed by the schools, which we now
15161proceed to do.
15162
15163If the proposition: "Every absolutely necessary being is likewise an ens
15164realissimum," is correct (and it is this which constitutes the nervus
15165probandi of the cosmological argument), it must, like all affirmative
15166judgements, be capable of conversion--the conversio per accidens, at
15167least. It follows, then, that some entia realissima are absolutely
15168necessary beings. But no ens realissimum is in any respect different
15169from another, and what is valid of some is valid of all. In this present
15170case, therefore, I may employ simple conversion, and say: "Every ens
15171realissimum is a necessary being." But as this proposition is determined
15172a priori by the conceptions contained in it, the mere conception of
15173an ens realissimum must possess the additional attribute of absolute
15174necessity. But this is exactly what was maintained in the ontological
15175argument, and not recognized by the cosmological, although it formed the
15176real ground of its disguised and illusory reasoning.
15177
15178Thus the second mode employed by speculative reason of demonstrating the
15179existence of a Supreme Being, is not only, like the first, illusory
15180and inadequate, but possesses the additional blemish of an ignoratio
15181elenchi--professing to conduct us by a new road to the desired goal, but
15182bringing us back, after a short circuit, to the old path which we had
15183deserted at its call.
15184
15185I mentioned above that this cosmological argument contains a perfect
15186nest of dialectical assumptions, which transcendental criticism does not
15187find it difficult to expose and to dissipate. I shall merely enumerate
15188these, leaving it to the reader, who must by this time be well practised
15189in such matters, to investigate the fallacies residing therein.
15190
15191The following fallacies, for example, are discoverable in this mode of
15192proof: 1. The transcendental principle: "Everything that is contingent
15193must have a cause"--a principle without significance, except in the
15194sensuous world. For the purely intellectual conception of the contingent
15195cannot produce any synthetical proposition, like that of causality,
15196which is itself without significance or distinguishing characteristic
15197except in the phenomenal world. But in the present case it is employed
15198to help us beyond the limits of its sphere. 2. "From the impossibility
15199of an infinite ascending series of causes in the world of sense a first
15200cause is inferred"; a conclusion which the principles of the employment
15201of reason do not justify even in the sphere of experience, and still
15202less when an attempt is made to pass the limits of this sphere. 3.
15203Reason allows itself to be satisfied upon insufficient grounds, with
15204regard to the completion of this series. It removes all conditions
15205(without which, however, no conception of Necessity can take place);
15206and, as after this it is beyond our power to form any other conceptions,
15207it accepts this as a completion of the conception it wishes to form of
15208the series. 4. The logical possibility of a conception of the total
15209of reality (the criterion of this possibility being the absence of
15210contradiction) is confounded with the transcendental, which requires a
15211principle of the practicability of such a synthesis--a principle which
15212again refers us to the world of experience. And so on.
15213
15214The aim of the cosmological argument is to avoid the necessity
15215of proving the existence of a necessary being priori from mere
15216conceptions--a proof which must be ontological, and of which we feel
15217ourselves quite incapable. With this purpose, we reason from an
15218actual existence--an experience in general, to an absolutely necessary
15219condition of that existence. It is in this case unnecessary to
15220demonstrate its possibility. For after having proved that it exists, the
15221question regarding its possibility is superfluous. Now, when we wish to
15222define more strictly the nature of this necessary being, we do not look
15223out for some being the conception of which would enable us to comprehend
15224the necessity of its being--for if we could do this, an empirical
15225presupposition would be unnecessary; no, we try to discover merely the
15226negative condition (conditio sine qua non), without which a being would
15227not be absolutely necessary. Now this would be perfectly admissible in
15228every sort of reasoning, from a consequence to its principle; but in
15229the present case it unfortunately happens that the condition of absolute
15230necessity can be discovered in but a single being, the conception of
15231which must consequently contain all that is requisite for demonstrating
15232the presence of absolute necessity, and thus entitle me to infer this
15233absolute necessity a priori. That is, it must be possible to reason
15234conversely, and say: The thing, to which the conception of the highest
15235reality belongs, is absolutely necessary. But if I cannot reason
15236thus--and I cannot, unless I believe in the sufficiency of the
15237ontological argument--I find insurmountable obstacles in my new path,
15238and am really no farther than the point from which I set out. The
15239conception of a Supreme Being satisfies all questions a priori regarding
15240the internal determinations of a thing, and is for this reason an ideal
15241without equal or parallel, the general conception of it indicating it
15242as at the same time an ens individuum among all possible things. But the
15243conception does not satisfy the question regarding its existence--which
15244was the purpose of all our inquiries; and, although the existence of a
15245necessary being were admitted, we should find it impossible to answer
15246the question: What of all things in the world must be regarded as such?
15247
15248It is certainly allowable to admit the existence of an all-sufficient
15249being--a cause of all possible effects--for the purpose of enabling
15250reason to introduce unity into its mode and grounds of explanation with
15251regard to phenomena. But to assert that such a being necessarily exists,
15252is no longer the modest enunciation of an admissible hypothesis, but
15253the boldest declaration of an apodeictic certainty; for the cognition of
15254that which is absolutely necessary must itself possess that character.
15255
15256The aim of the transcendental ideal formed by the mind is either to
15257discover a conception which shall harmonize with the idea of absolute
15258necessity, or a conception which shall contain that idea. If the one
15259is possible, so is the other; for reason recognizes that alone as
15260absolutely necessary which is necessary from its conception. But both
15261attempts are equally beyond our power--we find it impossible to satisfy
15262the understanding upon this point, and as impossible to induce it to
15263remain at rest in relation to this incapacity.
15264
15265Unconditioned necessity, which, as the ultimate support and stay of
15266all existing things, is an indispensable requirement of the mind, is an
15267abyss on the verge of which human reason trembles in dismay. Even the
15268idea of eternity, terrible and sublime as it is, as depicted by Haller,
15269does not produce upon the mental vision such a feeling of awe and
15270terror; for, although it measures the duration of things, it does not
15271support them. We cannot bear, nor can we rid ourselves of the
15272thought that a being, which we regard as the greatest of all possible
15273existences, should say to himself: I am from eternity to eternity;
15274beside me there is nothing, except that which exists by my will; whence
15275then am I? Here all sinks away from under us; and the greatest, as the
15276smallest, perfection, hovers without stay or footing in presence of the
15277speculative reason, which finds it as easy to part with the one as with
15278the other.
15279
15280Many physical powers, which evidence their existence by their effects,
15281are perfectly inscrutable in their nature; they elude all our powers
15282of observation. The transcendental object which forms the basis of
15283phenomena, and, in connection with it, the reason why our sensibility
15284possesses this rather than that particular kind of conditions, are and
15285must ever remain hidden from our mental vision; the fact is there, the
15286reason of the fact we cannot see. But an ideal of pure reason cannot
15287be termed mysterious or inscrutable, because the only credential of
15288its reality is the need of it felt by reason, for the purpose of giving
15289completeness to the world of synthetical unity. An ideal is not even
15290given as a cogitable object, and therefore cannot be inscrutable; on
15291the contrary, it must, as a mere idea, be based on the constitution of
15292reason itself, and on this account must be capable of explanation and
15293solution. For the very essence of reason consists in its ability to
15294give an account, of all our conceptions, opinions, and assertions--upon
15295objective, or, when they happen to be illusory and fallacious, upon
15296subjective grounds.
15297
15298
15299
15300Detection and Explanation of the Dialectical Illusion in all
15301Transcendental Arguments for the Existence of a Necessary Being.
15302
15303Both of the above arguments are transcendental; in other words, they do
15304not proceed upon empirical principles. For, although the cosmological
15305argument professed to lay a basis of experience for its edifice
15306of reasoning, it did not ground its procedure upon the peculiar
15307constitution of experience, but upon pure principles of reason--in
15308relation to an existence given by empirical consciousness; utterly
15309abandoning its guidance, however, for the purpose of supporting its
15310assertions entirely upon pure conceptions. Now what is the cause,
15311in these transcendental arguments, of the dialectical, but natural,
15312illusion, which connects the conceptions of necessity and supreme
15313reality, and hypostatizes that which cannot be anything but an idea?
15314What is the cause of this unavoidable step on the part of reason, of
15315admitting that some one among all existing things must be necessary,
15316while it falls back from the assertion of the existence of such a being
15317as from an abyss? And how does reason proceed to explain this anomaly
15318to itself, and from the wavering condition of a timid and reluctant
15319approbation--always again withdrawn--arrive at a calm and settled
15320insight into its cause?
15321
15322It is something very remarkable that, on the supposition that something
15323exists, I cannot avoid the inference that something exists necessarily.
15324Upon this perfectly natural--but not on that account reliable--inference
15325does the cosmological argument rest. But, let me form any conception
15326whatever of a thing, I find that I cannot cogitate the existence of
15327the thing as absolutely necessary, and that nothing prevents me--be the
15328thing or being what it may--from cogitating its non-existence. I may
15329thus be obliged to admit that all existing things have a necessary
15330basis, while I cannot cogitate any single or individual thing as
15331necessary. In other words, I can never complete the regress through the
15332conditions of existence, without admitting the existence of a necessary
15333being; but, on the other hand, I cannot make a commencement from this
15334being.
15335
15336If I must cogitate something as existing necessarily as the basis of
15337existing things, and yet am not permitted to cogitate any individual
15338thing as in itself necessary, the inevitable inference is that necessity
15339and contingency are not properties of things themselves--otherwise an
15340internal contradiction would result; that consequently neither of
15341these principles are objective, but merely subjective principles
15342of reason--the one requiring us to seek for a necessary ground
15343for everything that exists, that is, to be satisfied with no other
15344explanation than that which is complete a priori, the other forbidding
15345us ever to hope for the attainment of this completeness, that is, to
15346regard no member of the empirical world as unconditioned. In this
15347mode of viewing them, both principles, in their purely heuristic and
15348regulative character, and as concerning merely the formal interest of
15349reason, are quite consistent with each other. The one says: "You must
15350philosophize upon nature," as if there existed a necessary primal basis
15351of all existing things, solely for the purpose of introducing systematic
15352unity into your knowledge, by pursuing an idea of this character--a
15353foundation which is arbitrarily admitted to be ultimate; while the
15354other warns you to consider no individual determination, concerning
15355the existence of things, as such an ultimate foundation, that is,
15356as absolutely necessary, but to keep the way always open for further
15357progress in the deduction, and to treat every determination as
15358determined by some other. But if all that we perceive must be regarded
15359as conditionally necessary, it is impossible that anything which is
15360empirically given should be absolutely necessary.
15361
15362It follows from this that you must accept the absolutely necessary
15363as out of and beyond the world, inasmuch as it is useful only as a
15364principle of the highest possible unity in experience, and you cannot
15365discover any such necessary existence in the would, the second rule
15366requiring you to regard all empirical causes of unity as themselves
15367deduced.
15368
15369The philosophers of antiquity regarded all the forms of nature as
15370contingent; while matter was considered by them, in accordance with the
15371judgement of the common reason of mankind, as primal and necessary.
15372But if they had regarded matter, not relatively--as the substratum of
15373phenomena, but absolutely and in itself--as an independent existence,
15374this idea of absolute necessity would have immediately disappeared. For
15375there is nothing absolutely connecting reason with such an existence;
15376on the contrary, it can annihilate it in thought, always and without
15377self-contradiction. But in thought alone lay the idea of absolute
15378necessity. A regulative principle must, therefore, have been at
15379the foundation of this opinion. In fact, extension and
15380impenetrability--which together constitute our conception of
15381matter--form the supreme empirical principle of the unity of phenomena,
15382and this principle, in so far as it is empirically unconditioned,
15383possesses the property of a regulative principle. But, as every
15384determination of matter which constitutes what is real in it--and
15385consequently impenetrability--is an effect, which must have a cause, and
15386is for this reason always derived, the notion of matter cannot harmonize
15387with the idea of a necessary being, in its character of the principle of
15388all derived unity. For every one of its real properties, being derived,
15389must be only conditionally necessary, and can therefore be annihilated
15390in thought; and thus the whole existence of matter can be so annihilated
15391or suppressed. If this were not the case, we should have found in the
15392world of phenomena the highest ground or condition of unity--which is
15393impossible, according to the second regulative principle. It follows
15394that matter, and, in general, all that forms part of the world of sense,
15395cannot be a necessary primal being, nor even a principle of empirical
15396unity, but that this being or principle must have its place assigned
15397without the world. And, in this way, we can proceed in perfect
15398confidence to deduce the phenomena of the world and their existence from
15399other phenomena, just as if there existed no necessary being; and we
15400can at the same time, strive without ceasing towards the attainment of
15401completeness for our deduction, just as if such a being--the supreme
15402condition of all existences--were presupposed by the mind.
15403
15404These remarks will have made it evident to the reader that the ideal of
15405the Supreme Being, far from being an enouncement of the existence of a
15406being in itself necessary, is nothing more than a regulative principle
15407of reason, requiring us to regard all connection existing between
15408phenomena as if it had its origin from an all-sufficient necessary
15409cause, and basing upon this the rule of a systematic and necessary unity
15410in the explanation of phenomena. We cannot, at the same time, avoid
15411regarding, by a transcendental subreptio, this formal principle as
15412constitutive, and hypostatizing this unity. Precisely similar is the
15413case with our notion of space. Space is the primal condition of all
15414forms, which are properly just so many different limitations of it; and
15415thus, although it is merely a principle of sensibility, we cannot help
15416regarding it as an absolutely necessary and self-subsistent thing--as
15417an object given a priori in itself. In the same way, it is quite natural
15418that, as the systematic unity of nature cannot be established as a
15419principle for the empirical employment of reason, unless it is based
15420upon the idea of an ens realissimum, as the supreme cause, we should
15421regard this idea as a real object, and this object, in its character
15422of supreme condition, as absolutely necessary, and that in this way a
15423regulative should be transformed into a constitutive principle. This
15424interchange becomes evident when I regard this supreme being, which,
15425relatively to the world, was absolutely (unconditionally) necessary,
15426as a thing per se. In this case, I find it impossible to represent this
15427necessity in or by any conception, and it exists merely in my own mind,
15428as the formal condition of thought, but not as a material and hypostatic
15429condition of existence.
15430
15431
15432
15433SECTION VI. Of the Impossibility of a Physico-Theological Proof.
15434
15435If, then, neither a pure conception nor the general experience of an
15436existing being can provide a sufficient basis for the proof of the
15437existence of the Deity, we can make the attempt by the only other
15438mode--that of grounding our argument upon a determinate experience of
15439the phenomena of the present world, their constitution and disposition,
15440and discover whether we can thus attain to a sound conviction of
15441the existence of a Supreme Being. This argument we shall term the
15442physico-theological argument. If it is shown to be insufficient,
15443speculative reason cannot present us with any satisfactory proof of the
15444existence of a being corresponding to our transcendental idea.
15445
15446It is evident from the remarks that have been made in the preceding
15447sections, that an answer to this question will be far from being
15448difficult or unconvincing. For how can any experience be adequate
15449with an idea? The very essence of an idea consists in the fact that no
15450experience can ever be discovered congruent or adequate with it. The
15451transcendental idea of a necessary and all-sufficient being is so
15452immeasurably great, so high above all that is empirical, which is always
15453conditioned, that we hope in vain to find materials in the sphere of
15454experience sufficiently ample for our conception, and in vain seek the
15455unconditioned among things that are conditioned, while examples, nay,
15456even guidance is denied us by the laws of empirical synthesis.
15457
15458If the Supreme Being forms a link in the chain of empirical conditions,
15459it must be a member of the empirical series, and, like the lower members
15460which it precedes, have its origin in some higher member of the series.
15461If, on the other hand, we disengage it from the chain, and cogitate it
15462as an intelligible being, apart from the series of natural causes--how
15463shall reason bridge the abyss that separates the latter from the former?
15464All laws respecting the regress from effects to causes, all synthetical
15465additions to our knowledge relate solely to possible experience and
15466the objects of the sensuous world, and, apart from them, are without
15467significance.
15468
15469The world around us opens before our view so magnificent a spectacle of
15470order, variety, beauty, and conformity to ends, that whether we pursue
15471our observations into the infinity of space in the one direction, or
15472into its illimitable divisions in the other, whether we regard the world
15473in its greatest or its least manifestations--even after we have attained
15474to the highest summit of knowledge which our weak minds can reach, we
15475find that language in the presence of wonders so inconceivable has lost
15476its force, and number its power to reckon, nay, even thought fails to
15477conceive adequately, and our conception of the whole dissolves into an
15478astonishment without power of expression--all the more eloquent that it
15479is dumb. Everywhere around us we observe a chain of causes and effects,
15480of means and ends, of death and birth; and, as nothing has entered
15481of itself into the condition in which we find it, we are constantly
15482referred to some other thing, which itself suggests the same inquiry
15483regarding its cause, and thus the universe must sink into the abyss
15484of nothingness, unless we admit that, besides this infinite chain
15485of contingencies, there exists something that is primal and
15486self-subsistent--something which, as the cause of this phenomenal world,
15487secures its continuance and preservation.
15488
15489This highest cause--what magnitude shall we attribute to it? Of the
15490content of the world we are ignorant; still less can we estimate its
15491magnitude by comparison with the sphere of the possible. But this
15492supreme cause being a necessity of the human mind, what is there to
15493prevent us from attributing to it such a degree of perfection as to
15494place it above the sphere of all that is possible? This we can easily
15495do, although only by the aid of the faint outline of an abstract
15496conception, by representing this being to ourselves as containing
15497in itself, as an individual substance, all possible perfection--a
15498conception which satisfies that requirement of reason which demands
15499parsimony in principles, which is free from self-contradiction, which
15500even contributes to the extension of the employment of reason in
15501experience, by means of the guidance afforded by this idea to order and
15502system, and which in no respect conflicts with any law of experience.
15503
15504This argument always deserves to be mentioned with respect. It is the
15505oldest, the clearest, and that most in conformity with the common reason
15506of humanity. It animates the study of nature, as it itself derives its
15507existence and draws ever new strength from that source. It introduces
15508aims and ends into a sphere in which our observation could not of itself
15509have discovered them, and extends our knowledge of nature, by directing
15510our attention to a unity, the principle of which lies beyond nature.
15511This knowledge of nature again reacts upon this idea--its cause; and
15512thus our belief in a divine author of the universe rises to the power of
15513an irresistible conviction.
15514
15515For these reasons it would be utterly hopeless to attempt to rob this
15516argument of the authority it has always enjoyed. The mind, unceasingly
15517elevated by these considerations, which, although empirical, are so
15518remarkably powerful, and continually adding to their force, will
15519not suffer itself to be depressed by the doubts suggested by subtle
15520speculation; it tears itself out of this state of uncertainty, the
15521moment it casts a look upon the wondrous forms of nature and the majesty
15522of the universe, and rises from height to height, from condition to
15523condition, till it has elevated itself to the supreme and unconditioned
15524author of all.
15525
15526But although we have nothing to object to the reasonableness and utility
15527of this procedure, but have rather to commend and encourage it,
15528we cannot approve of the claims which this argument advances to
15529demonstrative certainty and to a reception upon its own merits, apart
15530from favour or support by other arguments. Nor can it injure the cause
15531of morality to endeavour to lower the tone of the arrogant sophist, and
15532to teach him that modesty and moderation which are the properties of a
15533belief that brings calm and content into the mind, without
15534prescribing to it an unworthy subjection. I maintain, then, that the
15535physico-theological argument is insufficient of itself to prove
15536the existence of a Supreme Being, that it must entrust this to the
15537ontological argument--to which it serves merely as an introduction, and
15538that, consequently, this argument contains the only possible ground of
15539proof (possessed by speculative reason) for the existence of this being.
15540
15541The chief momenta in the physico-theological argument are as follow:
155421. We observe in the world manifest signs of an arrangement full of
15543purpose, executed with great wisdom, and argument in whole of a
15544content indescribably various, and of an extent without limits. 2. This
15545arrangement of means and ends is entirely foreign to the things existing
15546in the world--it belongs to them merely as a contingent attribute;
15547in other words, the nature of different things could not of itself,
15548whatever means were employed, harmoniously tend towards certain
15549purposes, were they not chosen and directed for these purposes by a
15550rational and disposing principle, in accordance with certain fundamental
15551ideas. 3. There exists, therefore, a sublime and wise cause (or
15552several), which is not merely a blind, all-powerful nature, producing
15553the beings and events which fill the world in unconscious fecundity, but
15554a free and intelligent cause of the world. 4. The unity of this cause
15555may be inferred from the unity of the reciprocal relation existing
15556between the parts of the world, as portions of an artistic edifice--an
15557inference which all our observation favours, and all principles of
15558analogy support.
15559
15560In the above argument, it is inferred from the analogy of certain
15561products of nature with those of human art, when it compels Nature to
15562bend herself to its purposes, as in the case of a house, a ship, or
15563a watch, that the same kind of causality--namely, understanding
15564and will--resides in nature. It is also declared that the internal
15565possibility of this freely-acting nature (which is the source of all
15566art, and perhaps also of human reason) is derivable from another and
15567superhuman art--a conclusion which would perhaps be found incapable of
15568standing the test of subtle transcendental criticism. But to neither of
15569these opinions shall we at present object. We shall only remark that
15570it must be confessed that, if we are to discuss the subject of cause
15571at all, we cannot proceed more securely than with the guidance of the
15572analogy subsisting between nature and such products of design--these
15573being the only products whose causes and modes of organization are
15574completely known to us. Reason would be unable to satisfy her own
15575requirements, if she passed from a causality which she does know, to
15576obscure and indemonstrable principles of explanation which she does not
15577know.
15578
15579According to the physico-theological argument, the connection and
15580harmony existing in the world evidence the contingency of the form
15581merely, but not of the matter, that is, of the substance of the world.
15582To establish the truth of the latter opinion, it would be necessary to
15583prove that all things would be in themselves incapable of this harmony
15584and order, unless they were, even as regards their substance, the
15585product of a supreme wisdom. But this would require very different
15586grounds of proof from those presented by the analogy with human art.
15587This proof can at most, therefore, demonstrate the existence of an
15588architect of the world, whose efforts are limited by the capabilities of
15589the material with which he works, but not of a creator of the world, to
15590whom all things are subject. Thus this argument is utterly insufficient
15591for the task before us--a demonstration of the existence of an
15592all-sufficient being. If we wish to prove the contingency of matter,
15593we must have recourse to a transcendental argument, which the
15594physico-theological was constructed expressly to avoid.
15595
15596We infer, from the order and design visible in the universe, as a
15597disposition of a thoroughly contingent character, the existence of a
15598cause proportionate thereto. The conception of this cause must contain
15599certain determinate qualities, and it must therefore be regarded as the
15600conception of a being which possesses all power, wisdom, and so on, in
15601one word, all perfection--the conception, that is, of an all-sufficient
15602being. For the predicates of very great, astonishing, or immeasurable
15603power and excellence, give us no determinate conception of the thing,
15604nor do they inform us what the thing may be in itself. They merely
15605indicate the relation existing between the magnitude of the object and
15606the observer, who compares it with himself and with his own power of
15607comprehension, and are mere expressions of praise and reverence,
15608by which the object is either magnified, or the observing subject
15609depreciated in relation to the object. Where we have to do with the
15610magnitude (of the perfection) of a thing, we can discover no determinate
15611conception, except that which comprehends all possible perfection or
15612completeness, and it is only the total (omnitudo) of reality which is
15613completely determined in and through its conception alone.
15614
15615Now it cannot be expected that any one will be bold enough to declare
15616that he has a perfect insight into the relation which the magnitude
15617of the world he contemplates bears (in its extent as well as in its
15618content) to omnipotence, into that of the order and design in the
15619world to the highest wisdom, and that of the unity of the world to
15620the absolute unity of a Supreme Being. Physico-theology is therefore
15621incapable of presenting a determinate conception of a supreme cause of
15622the world, and is therefore insufficient as a principle of theology--a
15623theology which is itself to be the basis of religion.
15624
15625The attainment of absolute totality is completely impossible on the
15626path of empiricism. And yet this is the path pursued in the
15627physico-theological argument. What means shall we employ to bridge the
15628abyss?
15629
15630After elevating ourselves to admiration of the magnitude of the power,
15631wisdom, and other attributes of the author of the world, and finding we
15632can advance no further, we leave the argument on empirical grounds,
15633and proceed to infer the contingency of the world from the order and
15634conformity to aims that are observable in it. From this contingency we
15635infer, by the help of transcendental conceptions alone, the existence of
15636something absolutely necessary; and, still advancing, proceed from
15637the conception of the absolute necessity of the first cause to the
15638completely determined or determining conception thereof--the conception
15639of an all-embracing reality. Thus the physico-theological, failing
15640in its undertaking, recurs in its embarrassment to the cosmological
15641argument; and, as this is merely the ontological argument in disguise,
15642it executes its design solely by the aid of pure reason, although it at
15643first professed to have no connection with this faculty and to base its
15644entire procedure upon experience alone.
15645
15646The physico-theologians have therefore no reason to regard with such
15647contempt the transcendental mode of argument, and to look down upon
15648it, with the conceit of clear-sighted observers of nature, as the
15649brain-cobweb of obscure speculatists. For, if they reflect upon and
15650examine their own arguments, they will find that, after following for
15651some time the path of nature and experience, and discovering themselves
15652no nearer their object, they suddenly leave this path and pass into the
15653region of pure possibility, where they hope to reach upon the wings of
15654ideas what had eluded all their empirical investigations. Gaining, as
15655they think, a firm footing after this immense leap, they extend their
15656determinate conception--into the possession of which they have come,
15657they know not how--over the whole sphere of creation, and explain their
15658ideal, which is entirely a product of pure reason, by illustrations
15659drawn from experience--though in a degree miserably unworthy of the
15660grandeur of the object, while they refuse to acknowledge that they have
15661arrived at this cognition or hypothesis by a very different road from
15662that of experience.
15663
15664Thus the physico-theological is based upon the cosmological, and this
15665upon the ontological proof of the existence of a Supreme Being; and as
15666besides these three there is no other path open to speculative reason,
15667the ontological proof, on the ground of pure conceptions of reason, is
15668the only possible one, if any proof of a proposition so far transcending
15669the empirical exercise of the understanding is possible at all.
15670
15671
15672
15673SECTION VII. Critique of all Theology based upon Speculative
15674Principles of Reason.
15675
15676If by the term theology I understand the cognition of a primal being,
15677that cognition is based either upon reason alone (theologia rationalis)
15678or upon revelation (theologia revelata). The former cogitates its
15679object either by means of pure transcendental conceptions, as an ens
15680originarium, realissimum, ens entium, and is termed transcendental
15681theology; or, by means of a conception derived from the nature of our
15682own mind, as a supreme intelligence, and must then be entitled natural
15683theology. The person who believes in a transcendental theology alone,
15684is termed a deist; he who acknowledges the possibility of a natural
15685theology also, a theist. The former admits that we can cognize by pure
15686reason alone the existence of a Supreme Being, but at the same time
15687maintains that our conception of this being is purely transcendental,
15688and that all we can say of it is that it possesses all reality, without
15689being able to define it more closely. The second asserts that reason
15690is capable of presenting us, from the analogy with nature, with a more
15691definite conception of this being, and that its operations, as the cause
15692of all things, are the results of intelligence and free will. The former
15693regards the Supreme Being as the cause of the world--whether by the
15694necessity of his nature, or as a free agent, is left undetermined; the
15695latter considers this being as the author of the world.
15696
15697Transcendental theology aims either at inferring the existence of a
15698Supreme Being from a general experience, without any closer reference
15699to the world to which this experience belongs, and in this case it is
15700called cosmotheology; or it endeavours to cognize the existence of such
15701a being, through mere conceptions, without the aid of experience, and is
15702then termed ontotheology.
15703
15704Natural theology infers the attributes and the existence of an author of
15705the world, from the constitution of, the order and unity observable
15706in, the world, in which two modes of causality must be admitted to
15707exist--those of nature and freedom. Thus it rises from this world to a
15708supreme intelligence, either as the principle of all natural, or of
15709all moral order and perfection. In the former case it is termed
15710physico-theology, in the latter, ethical or moral-theology.*
15711
15712
15713 [*Footnote: Not theological ethics; for this science contains ethical
15714laws, which presuppose the existence of a Supreme Governor of the world;
15715while moral-theology, on the contrary, is the expression of a conviction
15716of the existence of a Supreme Being, founded upon ethical laws.]
15717
15718
15719As we are wont to understand by the term God not merely an eternal
15720nature, the operations of which are insensate and blind, but a Supreme
15721Being, who is the free and intelligent author of all things, and as it
15722is this latter view alone that can be of interest to humanity, we might,
15723in strict rigour, deny to the deist any belief in God at all, and
15724regard him merely as a maintainer of the existence of a primal being or
15725thing--the supreme cause of all other things. But, as no one ought to be
15726blamed, merely because he does not feel himself justified in maintaining
15727a certain opinion, as if he altogether denied its truth and asserted
15728the opposite, it is more correct--as it is less harsh--to say, the deist
15729believes in a God, the theist in a living God (summa intelligentia). We
15730shall now proceed to investigate the sources of all these attempts of
15731reason to establish the existence of a Supreme Being.
15732
15733It may be sufficient in this place to define theoretical knowledge or
15734cognition as knowledge of that which is, and practical knowledge as
15735knowledge of that which ought to be. In this view, the theoretical
15736employment of reason is that by which I cognize a priori (as necessary)
15737that something is, while the practical is that by which I cognize a
15738priori what ought to happen. Now, if it is an indubitably certain,
15739though at the same time an entirely conditioned truth, that something
15740is, or ought to happen, either a certain determinate condition of this
15741truth is absolutely necessary, or such a condition may be arbitrarily
15742presupposed. In the former case the condition is postulated (per
15743thesin), in the latter supposed (per hypothesin). There are certain
15744practical laws--those of morality--which are absolutely necessary. Now,
15745if these laws necessarily presuppose the existence of some being, as the
15746condition of the possibility of their obligatory power, this being must
15747be postulated, because the conditioned, from which we reason to this
15748determinate condition, is itself cognized a priori as absolutely
15749necessary. We shall at some future time show that the moral laws
15750not merely presuppose the existence of a Supreme Being, but also, as
15751themselves absolutely necessary in a different relation, demand
15752or postulate it--although only from a practical point of view. The
15753discussion of this argument we postpone for the present.
15754
15755When the question relates merely to that which is, not to that which
15756ought to be, the conditioned which is presented in experience is
15757always cogitated as contingent. For this reason its condition cannot be
15758regarded as absolutely necessary, but merely as relatively necessary,
15759or rather as needful; the condition is in itself and a priori a mere
15760arbitrary presupposition in aid of the cognition, by reason, of the
15761conditioned. If, then, we are to possess a theoretical cognition of
15762the absolute necessity of a thing, we cannot attain to this cognition
15763otherwise than a priori by means of conceptions; while it is impossible
15764in this way to cognize the existence of a cause which bears any relation
15765to an existence given in experience.
15766
15767Theoretical cognition is speculative when it relates to an object
15768or certain conceptions of an object which is not given and cannot be
15769discovered by means of experience. It is opposed to the cognition of
15770nature, which concerns only those objects or predicates which can be
15771presented in a possible experience.
15772
15773The principle that everything which happens (the empirically contingent)
15774must have a cause, is a principle of the cognition of nature, but not of
15775speculative cognition. For, if we change it into an abstract principle,
15776and deprive it of its reference to experience and the empirical, we
15777shall find that it cannot with justice be regarded any longer as a
15778synthetical proposition, and that it is impossible to discover any
15779mode of transition from that which exists to something entirely
15780different--termed cause. Nay, more, the conception of a cause likewise
15781that of the contingent--loses, in this speculative mode of employing
15782it, all significance, for its objective reality and meaning are
15783comprehensible from experience alone.
15784
15785When from the existence of the universe and the things in it the
15786existence of a cause of the universe is inferred, reason is proceeding
15787not in the natural, but in the speculative method. For the principle of
15788the former enounces, not that things themselves or substances, but only
15789that which happens or their states--as empirically contingent, have
15790a cause: the assertion that the existence of substance itself is
15791contingent is not justified by experience, it is the assertion of a
15792reason employing its principles in a speculative manner. If, again, I
15793infer from the form of the universe, from the way in which all things
15794are connected and act and react upon each other, the existence of
15795a cause entirely distinct from the universe--this would again be a
15796judgement of purely speculative reason; because the object in this
15797case--the cause--can never be an object of possible experience. In both
15798these cases the principle of causality, which is valid only in the field
15799of experience--useless and even meaningless beyond this region, would be
15800diverted from its proper destination.
15801
15802Now I maintain that all attempts of reason to establish a theology
15803by the aid of speculation alone are fruitless, that the principles of
15804reason as applied to nature do not conduct us to any theological truths,
15805and, consequently, that a rational theology can have no existence,
15806unless it is founded upon the laws of morality. For all synthetical
15807principles of the understanding are valid only as immanent in
15808experience; while the cognition of a Supreme Being necessitates their
15809being employed transcendentally, and of this the understanding is quite
15810incapable. If the empirical law of causality is to conduct us to
15811a Supreme Being, this being must belong to the chain of empirical
15812objects--in which case it would be, like all phenomena, itself
15813conditioned. If the possibility of passing the limits of experience be
15814admitted, by means of the dynamical law of the relation of an effect to
15815its cause, what kind of conception shall we obtain by this procedure?
15816Certainly not the conception of a Supreme Being, because experience
15817never presents us with the greatest of all possible effects, and it is
15818only an effect of this character that could witness to the existence
15819of a corresponding cause. If, for the purpose of fully satisfying the
15820requirements of Reason, we recognize her right to assert the existence
15821of a perfect and absolutely necessary being, this can be admitted
15822only from favour, and cannot be regarded as the result or irresistible
15823demonstration. The physico-theological proof may add weight to
15824others--if other proofs there are--by connecting speculation with
15825experience; but in itself it rather prepares the mind for theological
15826cognition, and gives it a right and natural direction, than establishes
15827a sure foundation for theology.
15828
15829It is now perfectly evident that transcendental questions admit only
15830of transcendental answers--those presented a priori by pure conceptions
15831without the least empirical admixture. But the question in the present
15832case is evidently synthetical--it aims at the extension of our cognition
15833beyond the bounds of experience--it requires an assurance respecting the
15834existence of a being corresponding with the idea in our minds, to which
15835no experience can ever be adequate. Now it has been abundantly proved
15836that all a priori synthetical cognition is possible only as the
15837expression of the formal conditions of a possible experience; and that
15838the validity of all principles depends upon their immanence in the field
15839of experience, that is, their relation to objects of empirical cognition
15840or phenomena. Thus all transcendental procedure in reference to
15841speculative theology is without result.
15842
15843If any one prefers doubting the conclusiveness of the proofs of our
15844analytic to losing the persuasion of the validity of these old and
15845time honoured arguments, he at least cannot decline answering the
15846question--how he can pass the limits of all possible experience by the
15847help of mere ideas. If he talks of new arguments, or of improvements
15848upon old arguments, I request him to spare me. There is certainly no
15849great choice in this sphere of discussion, as all speculative arguments
15850must at last look for support to the ontological, and I have, therefore,
15851very little to fear from the argumentative fecundity of the dogmatical
15852defenders of a non-sensuous reason. Without looking upon myself as a
15853remarkably combative person, I shall not decline the challenge to detect
15854the fallacy and destroy the pretensions of every attempt of speculative
15855theology. And yet the hope of better fortune never deserts those who
15856are accustomed to the dogmatical mode of procedure. I shall, therefore,
15857restrict myself to the simple and equitable demand that such reasoners
15858will demonstrate, from the nature of the human mind as well as from that
15859of the other sources of knowledge, how we are to proceed to extend
15860our cognition completely a priori, and to carry it to that point where
15861experience abandons us, and no means exist of guaranteeing the objective
15862reality of our conceptions. In whatever way the understanding may have
15863attained to a conception, the existence of the object of the conception
15864cannot be discovered in it by analysis, because the cognition of the
15865existence of the object depends upon the object's being posited and
15866given in itself apart from the conception. But it is utterly impossible
15867to go beyond our conception, without the aid of experience--which
15868presents to the mind nothing but phenomena, or to attain by the help
15869of mere conceptions to a conviction of the existence of new kinds of
15870objects or supernatural beings.
15871
15872But although pure speculative reason is far from sufficient to
15873demonstrate the existence of a Supreme Being, it is of the highest
15874utility in correcting our conception of this being--on the supposition
15875that we can attain to the cognition of it by some other means--in making
15876it consistent with itself and with all other conceptions of intelligible
15877objects, clearing it from all that is incompatible with the conception
15878of an ens summun, and eliminating from it all limitations or admixtures
15879of empirical elements.
15880
15881Transcendental theology is still therefore, notwithstanding its
15882objective insufficiency, of importance in a negative respect; it is
15883useful as a test of the procedure of reason when engaged with pure
15884ideas, no other than a transcendental standard being in this case
15885admissible. For if, from a practical point of view, the hypothesis of
15886a Supreme and All-sufficient Being is to maintain its validity without
15887opposition, it must be of the highest importance to define this
15888conception in a correct and rigorous manner--as the transcendental
15889conception of a necessary being, to eliminate all phenomenal elements
15890(anthropomorphism in its most extended signification), and at the
15891same time to overflow all contradictory assertions--be they atheistic,
15892deistic, or anthropomorphic. This is of course very easy; as the same
15893arguments which demonstrated the inability of human reason to affirm
15894the existence of a Supreme Being must be alike sufficient to prove the
15895invalidity of its denial. For it is impossible to gain from the pure
15896speculation of reason demonstration that there exists no Supreme Being,
15897as the ground of all that exists, or that this being possesses none
15898of those properties which we regard as analogical with the dynamical
15899qualities of a thinking being, or that, as the anthropomorphists would
15900have us believe, it is subject to all the limitations which sensibility
15901imposes upon those intelligences which exist in the world of experience.
15902
15903A Supreme Being is, therefore, for the speculative reason, a mere ideal,
15904though a faultless one--a conception which perfects and crowns the
15905system of human cognition, but the objective reality of which can
15906neither be proved nor disproved by pure reason. If this defect is ever
15907supplied by a moral theology, the problematic transcendental theology
15908which has preceded, will have been at least serviceable as demonstrating
15909the mental necessity existing for the conception, by the complete
15910determination of it which it has furnished, and the ceaseless testing of
15911the conclusions of a reason often deceived by sense, and not always in
15912harmony with its own ideas. The attributes of necessity, infinitude,
15913unity, existence apart from the world (and not as a world soul),
15914eternity (free from conditions of time), omnipresence (free from
15915conditions of space), omnipotence, and others, are pure transcendental
15916predicates; and thus the accurate conception of a Supreme Being, which
15917every theology requires, is furnished by transcendental theology alone.
15918
15919
15920
15921APPENDIX.
15922
15923Of the Regulative Employment of the Ideas of Pure Reason.
15924
15925The result of all the dialectical attempts of pure reason not only
15926confirms the truth of what we have already proved in our Transcendental
15927Analytic, namely, that all inferences which would lead us beyond the
15928limits of experience are fallacious and groundless, but it at the same
15929time teaches us this important lesson, that human reason has a natural
15930inclination to overstep these limits, and that transcendental ideas
15931are as much the natural property of the reason as categories are of the
15932understanding. There exists this difference, however, that while the
15933categories never mislead us, outward objects being always in perfect
15934harmony therewith, ideas are the parents of irresistible illusions, the
15935severest and most subtle criticism being required to save us from the
15936fallacies which they induce.
15937
15938Whatever is grounded in the nature of our powers will be found to be in
15939harmony with the final purpose and proper employment of these powers,
15940when once we have discovered their true direction and aim. We are
15941entitled to suppose, therefore, that there exists a mode of employing
15942transcendental ideas which is proper and immanent; although, when we
15943mistake their meaning, and regard them as conceptions of actual things,
15944their mode of application is transcendent and delusive. For it is not
15945the idea itself, but only the employment of the idea in relation to
15946possible experience, that is transcendent or immanent. An idea is
15947employed transcendently, when it is applied to an object falsely
15948believed to be adequate with and to correspond to it; imminently, when
15949it is applied solely to the employment of the understanding in the
15950sphere of experience. Thus all errors of subreptio--of misapplication,
15951are to be ascribed to defects of judgement, and not to understanding or
15952reason.
15953
15954Reason never has an immediate relation to an object; it relates
15955immediately to the understanding alone. It is only through the
15956understanding that it can be employed in the field of experience. It
15957does not form conceptions of objects, it merely arranges them and gives
15958to them that unity which they are capable of possessing when the sphere
15959of their application has been extended as widely as possible. Reason
15960avails itself of the conception of the understanding for the sole
15961purpose of producing totality in the different series. This totality the
15962understanding does not concern itself with; its only occupation is the
15963connection of experiences, by which series of conditions in accordance
15964with conceptions are established. The object of reason is, therefore,
15965the understanding and its proper destination. As the latter brings unity
15966into the diversity of objects by means of its conceptions, so the former
15967brings unity into the diversity of conceptions by means of ideas; as
15968it sets the final aim of a collective unity to the operations of the
15969understanding, which without this occupies itself with a distributive
15970unity alone.
15971
15972I accordingly maintain that transcendental ideas can never be employed
15973as constitutive ideas, that they cannot be conceptions of objects, and
15974that, when thus considered, they assume a fallacious and dialectical
15975character. But, on the other hand, they are capable of an admirable and
15976indispensably necessary application to objects--as regulative ideas,
15977directing the understanding to a certain aim, the guiding lines towards
15978which all its laws follow, and in which they all meet in one point. This
15979point--though a mere idea (focus imaginarius), that is, not a point from
15980which the conceptions of the understanding do really proceed, for it
15981lies beyond the sphere of possible experience--serves, notwithstanding,
15982to give to these conceptions the greatest possible unity combined with
15983the greatest possible extension. Hence arises the natural illusion which
15984induces us to believe that these lines proceed from an object which lies
15985out of the sphere of empirical cognition, just as objects reflected in
15986a mirror appear to be behind it. But this illusion--which we may hinder
15987from imposing upon us--is necessary and unavoidable, if we desire to
15988see, not only those objects which lie before us, but those which are at
15989a great distance behind us; that is to say, when, in the present case,
15990we direct the aims of the understanding, beyond every given experience,
15991towards an extension as great as can possibly be attained.
15992
15993If we review our cognitions in their entire extent, we shall find that
15994the peculiar business of reason is to arrange them into a system, that
15995is to say, to give them connection according to a principle. This unity
15996presupposes an idea--the idea of the form of a whole (of cognition),
15997preceding the determinate cognition of the parts, and containing the
15998conditions which determine a priori to every part its place and relation
15999to the other parts of the whole system. This idea, accordingly, demands
16000complete unity in the cognition of the understanding--not the unity of
16001a contingent aggregate, but that of a system connected according to
16002necessary laws. It cannot be affirmed with propriety that this idea is a
16003conception of an object; it is merely a conception of the complete unity
16004of the conceptions of objects, in so far as this unity is available to
16005the understanding as a rule. Such conceptions of reason are not derived
16006from nature; on the contrary, we employ them for the interrogation and
16007investigation of nature, and regard our cognition as defective so long
16008as it is not adequate to them. We admit that such a thing as pure earth,
16009pure water, or pure air, is not to be discovered. And yet we require
16010these conceptions (which have their origin in the reason, so far as
16011regards their absolute purity and completeness) for the purpose of
16012determining the share which each of these natural causes has in every
16013phenomenon. Thus the different kinds of matter are all referred to
16014earths, as mere weight; to salts and inflammable bodies, as pure force;
16015and finally, to water and air, as the vehicula of the former, or the
16016machines employed by them in their operations--for the purpose of
16017explaining the chemical action and reaction of bodies in accordance with
16018the idea of a mechanism. For, although not actually so expressed, the
16019influence of such ideas of reason is very observable in the procedure of
16020natural philosophers.
16021
16022If reason is the faculty of deducing the particular from the general,
16023and if the general be certain in se and given, it is only necessary
16024that the judgement should subsume the particular under the general,
16025the particular being thus necessarily determined. I shall term this
16026the demonstrative or apodeictic employment of reason. If, however,
16027the general is admitted as problematical only, and is a mere idea,
16028the particular case is certain, but the universality of the rule which
16029applies to this particular case remains a problem. Several particular
16030cases, the certainty of which is beyond doubt, are then taken and
16031examined, for the purpose of discovering whether the rule is applicable
16032to them; and if it appears that all the particular cases which can be
16033collected follow from the rule, its universality is inferred, and at the
16034same time, all the causes which have not, or cannot be presented to our
16035observation, are concluded to be of the same character with those which
16036we have observed. This I shall term the hypothetical employment of the
16037reason.
16038
16039The hypothetical exercise of reason by the aid of ideas employed as
16040problematical conceptions is properly not constitutive. That is to say,
16041if we consider the subject strictly, the truth of the rule, which has
16042been employed as an hypothesis, does not follow from the use that is
16043made of it by reason. For how can we know all the possible cases
16044that may arise? some of which may, however, prove exceptions to
16045the universality of the rule. This employment of reason is merely
16046regulative, and its sole aim is the introduction of unity into the
16047aggregate of our particular cognitions, and thereby the approximating of
16048the rule to universality.
16049
16050The object of the hypothetical employment of reason is therefore the
16051systematic unity of cognitions; and this unity is the criterion of the
16052truth of a rule. On the other hand, this systematic unity--as a mere
16053idea--is in fact merely a unity projected, not to be regarded as given,
16054but only in the light of a problem--a problem which serves, however, as
16055a principle for the various and particular exercise of the understanding
16056in experience, directs it with regard to those cases which are not
16057presented to our observation, and introduces harmony and consistency
16058into all its operations.
16059
16060All that we can be certain of from the above considerations is that
16061this systematic unity is a logical principle, whose aim is to assist the
16062understanding, where it cannot of itself attain to rules, by means of
16063ideas, to bring all these various rules under one principle, and thus
16064to ensure the most complete consistency and connection that can be
16065attained. But the assertion that objects and the understanding by which
16066they are cognized are so constituted as to be determined to systematic
16067unity, that this may be postulated a priori, without any reference
16068to the interest of reason, and that we are justified in declaring all
16069possible cognitions--empirical and others--to possess systematic unity,
16070and to be subject to general principles from which, notwithstanding
16071their various character, they are all derivable such an assertion can
16072be founded only upon a transcendental principle of reason, which would
16073render this systematic unity not subjectively and logically--in its
16074character of a method, but objectively necessary.
16075
16076We shall illustrate this by an example. The conceptions of the
16077understanding make us acquainted, among many other kinds of unity,
16078with that of the causality of a substance, which is termed power. The
16079different phenomenal manifestations of the same substance appear at
16080first view to be so very dissimilar that we are inclined to assume
16081the existence of just as many different powers as there are different
16082effects--as, in the case of the human mind, we have feeling,
16083consciousness, imagination, memory, wit, analysis, pleasure, desire
16084and so on. Now we are required by a logical maxim to reduce these
16085differences to as small a number as possible, by comparing them and
16086discovering the hidden identity which exists. We must inquire, for
16087example, whether or not imagination (connected with consciousness),
16088memory, wit, and analysis are not merely different forms of
16089understanding and reason. The idea of a fundamental power, the existence
16090of which no effort of logic can assure us of, is the problem to be
16091solved, for the systematic representation of the existing variety of
16092powers. The logical principle of reason requires us to produce as great
16093a unity as is possible in the system of our cognitions; and the more
16094the phenomena of this and the other power are found to be identical,
16095the more probable does it become, that they are nothing but different
16096manifestations of one and the same power, which may be called,
16097relatively speaking, a fundamental power. And so with other cases.
16098
16099These relatively fundamental powers must again be compared with
16100each other, to discover, if possible, the one radical and absolutely
16101fundamental power of which they are but the manifestations. But this
16102unity is purely hypothetical. It is not maintained, that this unity does
16103really exist, but that we must, in the interest of reason, that is,
16104for the establishment of principles for the various rules presented by
16105experience, try to discover and introduce it, so far as is practicable,
16106into the sphere of our cognitions.
16107
16108But the transcendental employment of the understanding would lead us to
16109believe that this idea of a fundamental power is not problematical, but
16110that it possesses objective reality, and thus the systematic unity
16111of the various powers or forces in a substance is demanded by the
16112understanding and erected into an apodeictic or necessary principle.
16113For, without having attempted to discover the unity of the various
16114powers existing in nature, nay, even after all our attempts have failed,
16115we notwithstanding presuppose that it does exist, and may be, sooner or
16116later, discovered. And this reason does, not only, as in the case
16117above adduced, with regard to the unity of substance, but where
16118many substances, although all to a certain extent homogeneous, are
16119discoverable, as in the case of matter in general. Here also does
16120reason presuppose the existence of the systematic unity of various
16121powers--inasmuch as particular laws of nature are subordinate to general
16122laws; and parsimony in principles is not merely an economical principle
16123of reason, but an essential law of nature.
16124
16125We cannot understand, in fact, how a logical principle of unity can of
16126right exist, unless we presuppose a transcendental principle, by which
16127such a systematic unit--as a property of objects themselves--is regarded
16128as necessary a priori. For with what right can reason, in its logical
16129exercise, require us to regard the variety of forces which nature
16130displays, as in effect a disguised unity, and to deduce them from one
16131fundamental force or power, when she is free to admit that it is just
16132as possible that all forces should be different in kind, and that a
16133systematic unity is not conformable to the design of nature? In this
16134view of the case, reason would be proceeding in direct opposition to her
16135own destination, by setting as an aim an idea which entirely conflicts
16136with the procedure and arrangement of nature. Neither can we assert that
16137reason has previously inferred this unity from the contingent nature
16138of phenomena. For the law of reason which requires us to seek for this
16139unity is a necessary law, inasmuch as without it we should not possess
16140a faculty of reason, nor without reason a consistent and self-accordant
16141mode of employing the understanding, nor, in the absence of this, any
16142proper and sufficient criterion of empirical truth. In relation to this
16143criterion, therefore, we must suppose the idea of the systematic unity
16144of nature to possess objective validity and necessity.
16145
16146We find this transcendental presupposition lurking in different forms in
16147the principles of philosophers, although they have neither recognized
16148it nor confessed to themselves its presence. That the diversities of
16149individual things do not exclude identity of species, that the various
16150species must be considered as merely different determinations of a
16151few genera, and these again as divisions of still higher races, and
16152so on--that, accordingly, a certain systematic unity of all possible
16153empirical conceptions, in so far as they can be deduced from higher and
16154more general conceptions, must be sought for, is a scholastic maxim or
16155logical principle, without which reason could not be employed by us. For
16156we can infer the particular from the general, only in so far as general
16157properties of things constitute the foundation upon which the particular
16158rest.
16159
16160That the same unity exists in nature is presupposed by philosophers
16161in the well-known scholastic maxim, which forbids us unnecessarily to
16162augment the number of entities or principles (entia praeter necessitatem
16163non esse multiplicanda). This maxim asserts that nature herself assists
16164in the establishment of this unity of reason, and that the seemingly
16165infinite diversity of phenomena should not deter us from the expectation
16166of discovering beneath this diversity a unity of fundamental properties,
16167of which the aforesaid variety is but a more or less determined form.
16168This unity, although a mere idea, thinkers have found it necessary
16169rather to moderate the desire than to encourage it. It was considered
16170a great step when chemists were able to reduce all salts to two main
16171genera--acids and alkalis; and they regard this difference as itself a
16172mere variety, or different manifestation of one and the same fundamental
16173material. The different kinds of earths (stones and even metals)
16174chemists have endeavoured to reduce to three, and afterwards to two; but
16175still, not content with this advance, they cannot but think that behind
16176these diversities there lurks but one genus--nay, that even salts and
16177earths have a common principle. It might be conjectured that this is
16178merely an economical plan of reason, for the purpose of sparing itself
16179trouble, and an attempt of a purely hypothetical character, which,
16180when successful, gives an appearance of probability to the principle of
16181explanation employed by the reason. But a selfish purpose of this kind
16182is easily to be distinguished from the idea, according to which every
16183one presupposes that this unity is in accordance with the laws of
16184nature, and that reason does not in this case request, but requires,
16185although we are quite unable to determine the proper limits of this
16186unity.
16187
16188If the diversity existing in phenomena--a diversity not of form (for
16189in this they may be similar) but of content--were so great that the
16190subtlest human reason could never by comparison discover in them the
16191least similarity (which is not impossible), in this case the logical law
16192of genera would be without foundation, the conception of a genus, nay,
16193all general conceptions would be impossible, and the faculty of the
16194understanding, the exercise of which is restricted to the world
16195of conceptions, could not exist. The logical principle of genera,
16196accordingly, if it is to be applied to nature (by which I mean objects
16197presented to our senses), presupposes a transcendental principle. In
16198accordance with this principle, homogeneity is necessarily presupposed
16199in the variety of phenomena (although we are unable to determine a
16200priori the degree of this homogeneity), because without it no empirical
16201conceptions, and consequently no experience, would be possible.
16202
16203The logical principle of genera, which demands identity in phenomena, is
16204balanced by another principle--that of species, which requires variety
16205and diversity in things, notwithstanding their accordance in the same
16206genus, and directs the understanding to attend to the one no less than
16207to the other. This principle (of the faculty of distinction) acts as a
16208check upon the reason and reason exhibits in this respect a double and
16209conflicting interest--on the one hand, the interest in the extent (the
16210interest of generality) in relation to genera; on the other, that of the
16211content (the interest of individuality) in relation to the variety of
16212species. In the former case, the understanding cogitates more under its
16213conceptions, in the latter it cogitates more in them. This distinction
16214manifests itself likewise in the habits of thought peculiar to natural
16215philosophers, some of whom--the remarkably speculative heads--may be
16216said to be hostile to heterogeneity in phenomena, and have their
16217eyes always fixed on the unity of genera, while others--with a strong
16218empirical tendency--aim unceasingly at the analysis of phenomena,
16219and almost destroy in us the hope of ever being able to estimate the
16220character of these according to general principles.
16221
16222The latter mode of thought is evidently based upon a logical principle,
16223the aim of which is the systematic completeness of all cognitions.
16224This principle authorizes me, beginning at the genus, to descend to the
16225various and diverse contained under it; and in this way extension, as
16226in the former case unity, is assured to the system. For if we merely
16227examine the sphere of the conception which indicates a genus, we cannot
16228discover how far it is possible to proceed in the division of that
16229sphere; just as it is impossible, from the consideration of the space
16230occupied by matter, to determine how far we can proceed in the division
16231of it. Hence every genus must contain different species, and these again
16232different subspecies; and as each of the latter must itself contain a
16233sphere (must be of a certain extent, as a conceptus communis), reason
16234demands that no species or sub-species is to be considered as the lowest
16235possible. For a species or sub-species, being always a conception, which
16236contains only what is common to a number of different things, does not
16237completely determine any individual thing, or relate immediately to
16238it, and must consequently contain other conceptions, that is, other
16239sub-species under it. This law of specification may be thus expressed:
16240entium varietates non temere sunt minuendae.
16241
16242But it is easy to see that this logical law would likewise be without
16243sense or application, were it not based upon a transcendental law of
16244specification, which certainly does not require that the differences
16245existing phenomena should be infinite in number, for the logical
16246principle, which merely maintains the indeterminateness of the logical
16247sphere of a conception, in relation to its possible division, does not
16248authorize this statement; while it does impose upon the understanding
16249the duty of searching for subspecies to every species, and minor
16250differences in every difference. For, were there no lower conceptions,
16251neither could there be any higher. Now the understanding cognizes only
16252by means of conceptions; consequently, how far soever it may proceed
16253in division, never by mere intuition, but always by lower and lower
16254conceptions. The cognition of phenomena in their complete determination
16255(which is possible only by means of the understanding) requires an
16256unceasingly continued specification of conceptions, and a progression
16257to ever smaller differences, of which abstraction bad been made in the
16258conception of the species, and still more in that of the genus.
16259
16260This law of specification cannot be deduced from experience; it can
16261never present us with a principle of so universal an application.
16262Empirical specification very soon stops in its distinction of
16263diversities, and requires the guidance of the transcendental law, as
16264a principle of the reason--a law which imposes on us the necessity of
16265never ceasing in our search for differences, even although these may not
16266present themselves to the senses. That absorbent earths are of different
16267kinds could only be discovered by obeying the anticipatory law of
16268reason, which imposes upon the understanding the task of discovering the
16269differences existing between these earths, and supposes that nature is
16270richer in substances than our senses would indicate. The faculty of the
16271understanding belongs to us just as much under the presupposition of
16272differences in the objects of nature, as under the condition that these
16273objects are homogeneous, because we could not possess conceptions, nor
16274make any use of our understanding, were not the phenomena included under
16275these conceptions in some respects dissimilar, as well as similar, in
16276their character.
16277
16278Reason thus prepares the sphere of the understanding for the operations
16279of this faculty: 1. By the principle of the homogeneity of the diverse
16280in higher genera; 2. By the principle of the variety of the homogeneous
16281in lower species; and, to complete the systematic unity, it adds, 3.
16282A law of the affinity of all conceptions which prescribes a continuous
16283transition from one species to every other by the gradual increase of
16284diversity. We may term these the principles of the homogeneity, the
16285specification, and the continuity of forms. The latter results from the
16286union of the two former, inasmuch as we regard the systematic connection
16287as complete in thought, in the ascent to higher genera, as well as in
16288the descent to lower species. For all diversities must be related
16289to each other, as they all spring from one highest genus, descending
16290through the different gradations of a more and more extended
16291determination.
16292
16293We may illustrate the systematic unity produced by the three logical
16294principles in the following manner. Every conception may be regarded as
16295a point, which, as the standpoint of a spectator, has a certain horizon,
16296which may be said to enclose a number of things that may be viewed,
16297so to speak, from that centre. Within this horizon there must be an
16298infinite number of other points, each of which has its own horizon,
16299smaller and more circumscribed; in other words, every species contains
16300sub-species, according to the principle of specification, and the
16301logical horizon consists of smaller horizons (subspecies), but not of
16302points (individuals), which possess no extent. But different horizons
16303or genera, which include under them so many conceptions, may have one
16304common horizon, from which, as from a mid-point, they may be surveyed;
16305and we may proceed thus, till we arrive at the highest genus, or
16306universal and true horizon, which is determined by the highest
16307conception, and which contains under itself all differences and
16308varieties, as genera, species, and subspecies.
16309
16310To this highest standpoint I am conducted by the law of homogeneity,
16311as to all lower and more variously-determined conceptions by the law
16312of specification. Now as in this way there exists no void in the whole
16313extent of all possible conceptions, and as out of the sphere of these
16314the mind can discover nothing, there arises from the presupposition of
16315the universal horizon above mentioned, and its complete division, the
16316principle: Non datur vacuum formarum. This principle asserts that there
16317are not different primitive and highest genera, which stand isolated, so
16318to speak, from each other, but all the various genera are mere divisions
16319and limitations of one highest and universal genus; and hence follows
16320immediately the principle: Datur continuum formarum. This principle
16321indicates that all differences of species limit each other, and do not
16322admit of transition from one to another by a saltus, but only through
16323smaller degrees of the difference between the one species and the other.
16324In one word, there are no species or sub-species which (in the view of
16325reason) are the nearest possible to each other; intermediate species or
16326sub-species being always possible, the difference of which from each of
16327the former is always smaller than the difference existing between these.
16328
16329The first law, therefore, directs us to avoid the notion that there
16330exist different primal genera, and enounces the fact of perfect
16331homogeneity; the second imposes a check upon this tendency to unity and
16332prescribes the distinction of sub-species, before proceeding to apply
16333our general conceptions to individuals. The third unites both the
16334former, by enouncing the fact of homogeneity as existing even in the
16335most various diversity, by means of the gradual transition from one
16336species to another. Thus it indicates a relationship between the
16337different branches or species, in so far as they all spring from the
16338same stem.
16339
16340But this logical law of the continuum specierum (formarum logicarum)
16341presupposes a transcendental principle (lex continui in natura), without
16342which the understanding might be led into error, by following the
16343guidance of the former, and thus perhaps pursuing a path contrary to
16344that prescribed by nature. This law must, consequently, be based upon
16345pure transcendental, and not upon empirical, considerations. For, in the
16346latter case, it would come later than the system; whereas it is really
16347itself the parent of all that is systematic in our cognition of nature.
16348These principles are not mere hypotheses employed for the purpose
16349of experimenting upon nature; although when any such connection is
16350discovered, it forms a solid ground for regarding the hypothetical unity
16351as valid in the sphere of nature--and thus they are in this respect not
16352without their use. But we go farther, and maintain that it is manifest
16353that these principles of parsimony in fundamental causes, variety in
16354effects, and affinity in phenomena, are in accordance both with reason
16355and nature, and that they are not mere methods or plans devised for the
16356purpose of assisting us in our observation of the external world.
16357
16358But it is plain that this continuity of forms is a mere idea, to which
16359no adequate object can be discovered in experience. And this for two
16360reasons. First, because the species in nature are really divided, and
16361hence form quanta discreta; and, if the gradual progression through
16362their affinity were continuous, the intermediate members lying between
16363two given species must be infinite in number, which is impossible.
16364Secondly, because we cannot make any determinate empirical use of this
16365law, inasmuch as it does not present us with any criterion of affinity
16366which could aid us in determining how far we ought to pursue the
16367graduation of differences: it merely contains a general indication that
16368it is our duty to seek for and, if possible, to discover them.
16369
16370When we arrange these principles of systematic unity in the order
16371conformable to their employment in experience, they will stand thus:
16372Variety, Affinity, Unity, each of them, as ideas, being taken in the
16373highest degree of their completeness. Reason presupposes the existence
16374of cognitions of the understanding, which have a direct relation to
16375experience, and aims at the ideal unity of these cognitions--a unity
16376which far transcends all experience or empirical notions. The affinity
16377of the diverse, notwithstanding the differences existing between its
16378parts, has a relation to things, but a still closer one to the mere
16379properties and powers of things. For example, imperfect experience
16380may represent the orbits of the planets as circular. But we discover
16381variations from this course, and we proceed to suppose that the planets
16382revolve in a path which, if not a circle, is of a character very similar
16383to it. That is to say, the movements of those planets which do not form
16384a circle will approximate more or less to the properties of a circle,
16385and probably form an ellipse. The paths of comets exhibit still greater
16386variations, for, so far as our observation extends, they do not return
16387upon their own course in a circle or ellipse. But we proceed to the
16388conjecture that comets describe a parabola, a figure which is closely
16389allied to the ellipse. In fact, a parabola is merely an ellipse, with
16390its longer axis produced to an indefinite extent. Thus these principles
16391conduct us to a unity in the genera of the forms of these orbits, and,
16392proceeding farther, to a unity as regards the cause of the motions of
16393the heavenly bodies--that is, gravitation. But we go on extending our
16394conquests over nature, and endeavour to explain all seeming deviations
16395from these rules, and even make additions to our system which no
16396experience can ever substantiate--for example, the theory, in affinity
16397with that of ellipses, of hyperbolic paths of comets, pursuing which,
16398these bodies leave our solar system and, passing from sun to sun, unite
16399the most distant parts of the infinite universe, which is held together
16400by the same moving power.
16401
16402The most remarkable circumstance connected with these principles is that
16403they seem to be transcendental, and, although only containing ideas
16404for the guidance of the empirical exercise of reason, and although this
16405empirical employment stands to these ideas in an asymptotic relation
16406alone (to use a mathematical term), that is, continually
16407approximate, without ever being able to attain to them, they possess,
16408notwithstanding, as a priori synthetical propositions, objective
16409though undetermined validity, and are available as rules for possible
16410experience. In the elaboration of our experience, they may also be
16411employed with great advantage, as heuristic [Footnote: From the Greek,
16412eurhioko.] principles. A transcendental deduction of them cannot be
16413made; such a deduction being always impossible in the case of ideas, as
16414has been already shown.
16415
16416We distinguished, in the Transcendental Analytic, the dynamical
16417principles of the understanding, which are regulative principles of
16418intuition, from the mathematical, which are constitutive principles of
16419intuition. These dynamical laws are, however, constitutive in relation
16420to experience, inasmuch as they render the conceptions without which
16421experience could not exist possible a priori. But the principles of pure
16422reason cannot be constitutive even in regard to empirical conceptions,
16423because no sensuous schema corresponding to them can be discovered, and
16424they cannot therefore have an object in concreto. Now, if I grant that
16425they cannot be employed in the sphere of experience, as constitutive
16426principles, how shall I secure for them employment and objective
16427validity as regulative principles, and in what way can they be so
16428employed?
16429
16430The understanding is the object of reason, as sensibility is the object
16431of the understanding. The production of systematic unity in all the
16432empirical operations of the understanding is the proper occupation of
16433reason; just as it is the business of the understanding to connect the
16434various content of phenomena by means of conceptions, and subject them
16435to empirical laws. But the operations of the understanding are, without
16436the schemata of sensibility, undetermined; and, in the same manner,
16437the unity of reason is perfectly undetermined as regards the conditions
16438under which, and the extent to which, the understanding ought to carry
16439the systematic connection of its conceptions. But, although it is
16440impossible to discover in intuition a schema for the complete systematic
16441unity of all the conceptions of the understanding, there must be some
16442analogon of this schema. This analogon is the idea of the maximum of the
16443division and the connection of our cognition in one principle. For we
16444may have a determinate notion of a maximum and an absolutely perfect,
16445all the restrictive conditions which are connected with an indeterminate
16446and various content having been abstracted. Thus the idea of reason
16447is analogous with a sensuous schema, with this difference, that the
16448application of the categories to the schema of reason does not present
16449a cognition of any object (as is the case with the application of the
16450categories to sensuous schemata), but merely provides us with a rule or
16451principle for the systematic unity of the exercise of the understanding.
16452Now, as every principle which imposes upon the exercise of the
16453understanding a priori compliance with the rule of systematic unity
16454also relates, although only in an indirect manner, to an object of
16455experience, the principles of pure reason will also possess objective
16456reality and validity in relation to experience. But they will not aim at
16457determining our knowledge in regard to any empirical object; they
16458will merely indicate the procedure, following which the empirical and
16459determinate exercise of the understanding may be in complete harmony and
16460connection with itself--a result which is produced by its being brought
16461into harmony with the principle of systematic unity, so far as that is
16462possible, and deduced from it.
16463
16464I term all subjective principles, which are not derived from observation
16465of the constitution of an object, but from the interest which Reason
16466has in producing a certain completeness in her cognition of that object,
16467maxims of reason. Thus there are maxims of speculative reason, which are
16468based solely upon its speculative interest, although they appear to be
16469objective principles.
16470
16471When principles which are really regulative are regarded as
16472constitutive, and employed as objective principles, contradictions must
16473arise; but if they are considered as mere maxims, there is no room for
16474contradictions of any kind, as they then merely indicate the different
16475interests of reason, which occasion differences in the mode of thought.
16476In effect, Reason has only one single interest, and the seeming
16477contradiction existing between her maxims merely indicates a difference
16478in, and a reciprocal limitation of, the methods by which this interest
16479is satisfied.
16480
16481This reasoner has at heart the interest of diversity--in accordance
16482with the principle of specification; another, the interest of unity--in
16483accordance with the principle of aggregation. Each believes that
16484his judgement rests upon a thorough insight into the subject he is
16485examining, and yet it has been influenced solely by a greater or less
16486degree of adherence to some one of the two principles, neither of which
16487are objective, but originate solely from the interest of reason, and on
16488this account to be termed maxims rather than principles. When I observe
16489intelligent men disputing about the distinctive characteristics of men,
16490animals, or plants, and even of minerals, those on the one side assuming
16491the existence of certain national characteristics, certain well-defined
16492and hereditary distinctions of family, race, and so on, while the other
16493side maintain that nature has endowed all races of men with the same
16494faculties and dispositions, and that all differences are but the result
16495of external and accidental circumstances--I have only to consider for
16496a moment the real nature of the subject of discussion, to arrive at the
16497conclusion that it is a subject far too deep for us to judge of, and
16498that there is little probability of either party being able to speak
16499from a perfect insight into and understanding of the nature of the
16500subject itself. Both have, in reality, been struggling for the twofold
16501interest of reason; the one maintaining the one interest, the other the
16502other. But this difference between the maxims of diversity and unity
16503may easily be reconciled and adjusted; although, so long as they
16504are regarded as objective principles, they must occasion not only
16505contradictions and polemic, but place hinderances in the way of the
16506advancement of truth, until some means is discovered of reconciling
16507these conflicting interests, and bringing reason into union and harmony
16508with itself.
16509
16510The same is the case with the so-called law discovered by Leibnitz, and
16511supported with remarkable ability by Bonnet--the law of the continuous
16512gradation of created beings, which is nothing more than an inference
16513from the principle of affinity; for observation and study of the order
16514of nature could never present it to the mind as an objective truth. The
16515steps of this ladder, as they appear in experience, are too far apart
16516from each other, and the so-called petty differences between different
16517kinds of animals are in nature commonly so wide separations that no
16518confidence can be placed in such views (particularly when we reflect
16519on the great variety of things, and the ease with which we can discover
16520resemblances), and no faith in the laws which are said to express
16521the aims and purposes of nature. On the other hand, the method of
16522investigating the order of nature in the light of this principle,
16523and the maxim which requires us to regard this order--it being still
16524undetermined how far it extends--as really existing in nature, is beyond
16525doubt a legitimate and excellent principle of reason--a principle which
16526extends farther than any experience or observation of ours and which,
16527without giving us any positive knowledge of anything in the region of
16528experience, guides us to the goal of systematic unity.
16529
16530
16531
16532Of the Ultimate End of the Natural Dialectic of Human Reason.
16533
16534The ideas of pure reason cannot be, of themselves and in their own
16535nature, dialectical; it is from their misemployment alone that fallacies
16536and illusions arise. For they originate in the nature of reason itself,
16537and it is impossible that this supreme tribunal for all the rights and
16538claims of speculation should be itself undeserving of confidence and
16539promotive of error. It is to be expected, therefore, that these ideas
16540have a genuine and legitimate aim. It is true, the mob of sophists raise
16541against reason the cry of inconsistency and contradiction, and affect to
16542despise the government of that faculty, because they cannot understand
16543its constitution, while it is to its beneficial influences alone
16544that they owe the position and the intelligence which enable them to
16545criticize and to blame its procedure.
16546
16547We cannot employ an a priori conception with certainty, until we have
16548made a transcendental deduction therefore. The ideas of pure reason do
16549not admit of the same kind of deduction as the categories. But if they
16550are to possess the least objective validity, and to represent anything
16551but mere creations of thought (entia rationis ratiocinantis), a
16552deduction of them must be possible. This deduction will complete the
16553critical task imposed upon pure reason; and it is to this part Of our
16554labours that we now proceed.
16555
16556There is a great difference between a thing's being presented to the
16557mind as an object in an absolute sense, or merely as an ideal object. In
16558the former case I employ my conceptions to determine the object; in the
16559latter case nothing is present to the mind but a mere schema, which does
16560not relate directly to an object, not even in a hypothetical sense, but
16561which is useful only for the purpose of representing other objects to
16562the mind, in a mediate and indirect manner, by means of their relation
16563to the idea in the intellect. Thus I say the conception of a supreme
16564intelligence is a mere idea; that is to say, its objective reality does
16565not consist in the fact that it has an immediate relation to an object
16566(for in this sense we have no means of establishing its objective
16567validity), it is merely a schema constructed according to the necessary
16568conditions of the unity of reason--the schema of a thing in general,
16569which is useful towards the production of the highest degree of
16570systematic unity in the empirical exercise of reason, in which we deduce
16571this or that object of experience from the imaginary object of this
16572idea, as the ground or cause of the said object of experience. In this
16573way, the idea is properly a heuristic, and not an ostensive, conception;
16574it does not give us any information respecting the constitution of an
16575object, it merely indicates how, under the guidance of the idea, we
16576ought to investigate the constitution and the relations of objects in
16577the world of experience. Now, if it can be shown that the three kinds
16578of transcendental ideas (psychological, cosmological, and theological),
16579although not relating directly to any object nor determining it, do
16580nevertheless, on the supposition of the existence of an ideal object,
16581produce systematic unity in the laws of the empirical employment of
16582the reason, and extend our empirical cognition, without ever being
16583inconsistent or in opposition with it--it must be a necessary maxim
16584of reason to regulate its procedure according to these ideas. And this
16585forms the transcendental deduction of all speculative ideas, not as
16586constitutive principles of the extension of our cognition beyond the
16587limits of our experience, but as regulative principles of the systematic
16588unity of empirical cognition, which is by the aid of these ideas
16589arranged and emended within its own proper limits, to an extent
16590unattainable by the operation of the principles of the understanding
16591alone.
16592
16593I shall make this plainer. Guided by the principles involved in these
16594ideas, we must, in the first place, so connect all the phenomena,
16595actions, and feelings of the mind, as if it were a simple substance,
16596which, endowed with personal identity, possesses a permanent existence
16597(in this life at least), while its states, among which those of the
16598body are to be included as external conditions, are in continual change.
16599Secondly, in cosmology, we must investigate the conditions of all
16600natural phenomena, internal as well as external, as if they belonged to
16601a chain infinite and without any prime or supreme member, while we do
16602not, on this account, deny the existence of intelligible grounds of
16603these phenomena, although we never employ them to explain phenomena, for
16604the simple reason that they are not objects of our cognition. Thirdly,
16605in the sphere of theology, we must regard the whole system of
16606possible experience as forming an absolute, but dependent and
16607sensuously-conditioned unity, and at the same time as based upon a
16608sole, supreme, and all-sufficient ground existing apart from the world
16609itself--a ground which is a self-subsistent, primeval and creative
16610reason, in relation to which we so employ our reason in the field of
16611experience, as if all objects drew their origin from that archetype
16612of all reason. In other words, we ought not to deduce the internal
16613phenomena of the mind from a simple thinking substance, but deduce them
16614from each other under the guidance of the regulative idea of a simple
16615being; we ought not to deduce the phenomena, order, and unity of the
16616universe from a supreme intelligence, but merely draw from this idea
16617of a supremely wise cause the rules which must guide reason in its
16618connection of causes and effects.
16619
16620Now there is nothing to hinder us from admitting these ideas to possess
16621an objective and hyperbolic existence, except the cosmological ideas,
16622which lead reason into an antinomy: the psychological and theological
16623ideas are not antinomial. They contain no contradiction; and how, then,
16624can any one dispute their objective reality, since he who denies it
16625knows as little about their possibility as we who affirm? And yet,
16626when we wish to admit the existence of a thing, it is not sufficient to
16627convince ourselves that there is no positive obstacle in the way; for
16628it cannot be allowable to regard mere creations of thought, which
16629transcend, though they do not contradict, all our conceptions, as real
16630and determinate objects, solely upon the authority of a speculative
16631reason striving to compass its own aims. They cannot, therefore, be
16632admitted to be real in themselves; they can only possess a comparative
16633reality--that of a schema of the regulative principle of the systematic
16634unity of all cognition. They are to be regarded not as actual things,
16635but as in some measure analogous to them. We abstract from the object
16636of the idea all the conditions which limit the exercise of our
16637understanding, but which, on the other hand, are the sole conditions of
16638our possessing a determinate conception of any given thing. And thus we
16639cogitate a something, of the real nature of which we have not the
16640least conception, but which we represent to ourselves as standing in a
16641relation to the whole system of phenomena, analogous to that in which
16642phenomena stand to each other.
16643
16644By admitting these ideal beings, we do not really extend our cognitions
16645beyond the objects of possible experience; we extend merely the
16646empirical unity of our experience, by the aid of systematic unity, the
16647schema of which is furnished by the idea, which is therefore valid--not
16648as a constitutive, but as a regulative principle. For although we posit
16649a thing corresponding to the idea--a something, an actual existence--we
16650do not on that account aim at the extension of our cognition by means
16651of transcendent conceptions. This existence is purely ideal, and not
16652objective; it is the mere expression of the systematic unity which is to
16653be the guide of reason in the field of experience. There are no attempts
16654made at deciding what the ground of this unity may be, or what the real
16655nature of this imaginary being.
16656
16657Thus the transcendental and only determinate conception of God, which
16658is presented to us by speculative reason, is in the strictest sense
16659deistic. In other words, reason does not assure us of the objective
16660validity of the conception; it merely gives us the idea of something, on
16661which the supreme and necessary unity of all experience is based. This
16662something we cannot, following the analogy of a real substance, cogitate
16663otherwise than as the cause of all things operating in accordance with
16664rational laws, if we regard it as an individual object; although we
16665should rest contented with the idea alone as a regulative principle
16666of reason, and make no attempt at completing the sum of the conditions
16667imposed by thought. This attempt is, indeed, inconsistent with the grand
16668aim of complete systematic unity in the sphere of cognition--a unity to
16669which no bounds are set by reason.
16670
16671Hence it happens that, admitting a divine being, I can have no
16672conception of the internal possibility of its perfection, or of the
16673necessity of its existence. The only advantage of this admission is that
16674it enables me to answer all other questions relating to the contingent,
16675and to give reason the most complete satisfaction as regards the unity
16676which it aims at attaining in the world of experience. But I cannot
16677satisfy reason with regard to this hypothesis itself; and this proves
16678that it is not its intelligence and insight into the subject, but its
16679speculative interest alone which induces it to proceed from a point
16680lying far beyond the sphere of our cognition, for the purpose of being
16681able to consider all objects as parts of a systematic whole.
16682
16683Here a distinction presents itself, in regard to the way in which we may
16684cogitate a presupposition--a distinction which is somewhat subtle, but
16685of great importance in transcendental philosophy. I may have sufficient
16686grounds to admit something, or the existence of something, in a
16687relative point of view (suppositio relativa), without being justified
16688in admitting it in an absolute sense (suppositio absoluta). This
16689distinction is undoubtedly requisite, in the case of a regulative
16690principle, the necessity of which we recognize, though we are ignorant
16691of the source and cause of that necessity, and which we assume to
16692be based upon some ultimate ground, for the purpose of being able to
16693cogitate the universality of the principle in a more determinate way.
16694For example, I cogitate the existence of a being corresponding to a
16695pure transcendental idea. But I cannot admit that this being exists
16696absolutely and in itself, because all of the conceptions by which I can
16697cogitate an object in a determinate manner fall short of assuring me
16698of its existence; nay, the conditions of the objective validity of my
16699conceptions are excluded by the idea--by the very fact of its being an
16700idea. The conceptions of reality, substance, causality, nay, even that
16701of necessity in existence, have no significance out of the sphere of
16702empirical cognition, and cannot, beyond that sphere, determine any
16703object. They may, accordingly, be employed to explain the possibility of
16704things in the world of sense, but they are utterly inadequate to explain
16705the possibility of the universe itself considered as a whole; because
16706in this case the ground of explanation must lie out of and beyond the
16707world, and cannot, therefore, be an object of possible experience.
16708Now, I may admit the existence of an incomprehensible being of this
16709nature--the object of a mere idea, relatively to the world of sense;
16710although I have no ground to admit its existence absolutely and in
16711itself. For if an idea (that of a systematic and complete unity, of
16712which I shall presently speak more particularly) lies at the foundation
16713of the most extended empirical employment of reason, and if this
16714idea cannot be adequately represented in concreto, although it is
16715indispensably necessary for the approximation of empirical unity to the
16716highest possible degree--I am not only authorized, but compelled,
16717to realize this idea, that is, to posit a real object corresponding
16718thereto. But I cannot profess to know this object; it is to me merely a
16719something, to which, as the ground of systematic unity in cognition, I
16720attribute such properties as are analogous to the conceptions employed
16721by the understanding in the sphere of experience. Following the analogy
16722of the notions of reality, substance, causality, and necessity, I
16723cogitate a being, which possesses all these attributes in the highest
16724degree; and, as this idea is the offspring of my reason alone, I
16725cogitate this being as self-subsistent reason, and as the cause of the
16726universe operating by means of ideas of the greatest possible harmony
16727and unity. Thus I abstract all conditions that would limit my idea,
16728solely for the purpose of rendering systematic unity possible in the
16729world of empirical diversity, and thus securing the widest possible
16730extension for the exercise of reason in that sphere. This I am enabled
16731to do, by regarding all connections and relations in the world of sense,
16732as if they were the dispositions of a supreme reason, of which our
16733reason is but a faint image. I then proceed to cogitate this Supreme
16734Being by conceptions which have, properly, no meaning or application,
16735except in the world of sense. But as I am authorized to employ the
16736transcendental hypothesis of such a being in a relative respect
16737alone, that is, as the substratum of the greatest possible unity in
16738experience--I may attribute to a being which I regard as distinct from
16739the world, such properties as belong solely to the sphere of sense and
16740experience. For I do not desire, and am not justified in desiring, to
16741cognize this object of my idea, as it exists in itself; for I possess
16742no conceptions sufficient for or task, those of reality, substance,
16743causality, nay, even that of necessity in existence, losing all
16744significance, and becoming merely the signs of conceptions, without
16745content and without applicability, when I attempt to carry them beyond
16746the limits of the world of sense. I cogitate merely the relation of a
16747perfectly unknown being to the greatest possible systematic unity of
16748experience, solely for the purpose of employing it as the schema of the
16749regulative principle which directs reason in its empirical exercise.
16750
16751It is evident, at the first view, that we cannot presuppose the reality
16752of this transcendental object, by means of the conceptions of reality,
16753substance, causality, and so on, because these conceptions cannot be
16754applied to anything that is distinct from the world of sense. Thus
16755the supposition of a Supreme Being or cause is purely relative; it is
16756cogitated only in behalf of the systematic unity of experience; such a
16757being is but a something, of whose existence in itself we have not the
16758least conception. Thus, too, it becomes sufficiently manifest why we
16759required the idea of a necessary being in relation to objects given by
16760sense, although we can never have the least conception of this being, or
16761of its absolute necessity.
16762
16763And now we can clearly perceive the result of our transcendental
16764dialectic, and the proper aim of the ideas of pure reason--which become
16765dialectical solely from misunderstanding and inconsiderateness. Pure
16766reason is, in fact, occupied with itself, and not with any object.
16767Objects are not presented to it to be embraced in the unity of an
16768empirical conception; it is only the cognitions of the understanding
16769that are presented to it, for the purpose of receiving the unity of
16770a rational conception, that is, of being connected according to
16771a principle. The unity of reason is the unity of system; and this
16772systematic unity is not an objective principle, extending its dominion
16773over objects, but a subjective maxim, extending its authority over the
16774empirical cognition of objects. The systematic connection which reason
16775gives to the empirical employment of the understanding not only advances
16776the extension of that employment, but ensures its correctness, and thus
16777the principle of a systematic unity of this nature is also objective,
16778although only in an indefinite respect (principium vagum). It is not,
16779however, a constitutive principle, determining an object to which
16780it directly relates; it is merely a regulative principle or maxim,
16781advancing and strengthening the empirical exercise of reason, by the
16782opening up of new paths of which the understanding is ignorant, while
16783it never conflicts with the laws of its exercise in the sphere of
16784experience.
16785
16786But reason cannot cogitate this systematic unity, without at the
16787same time cogitating an object of the idea--an object that cannot be
16788presented in any experience, which contains no concrete example of a
16789complete systematic unity. This being (ens rationis ratiocinatae) is
16790therefore a mere idea and is not assumed to be a thing which is
16791real absolutely and in itself. On the contrary, it forms merely the
16792problematical foundation of the connection which the mind introduces
16793among the phenomena of the sensuous world. We look upon this connection,
16794in the light of the above-mentioned idea, as if it drew its origin from
16795the supposed being which corresponds to the idea. And yet all we aim at
16796is the possession of this idea as a secure foundation for the systematic
16797unity of experience--a unity indispensable to reason, advantageous
16798to the understanding, and promotive of the interests of empirical
16799cognition.
16800
16801We mistake the true meaning of this idea when we regard it as an
16802enouncement, or even as a hypothetical declaration of the existence of
16803a real thing, which we are to regard as the origin or ground of a
16804systematic constitution of the universe. On the contrary, it is left
16805completely undetermined what the nature or properties of this so-called
16806ground may be. The idea is merely to be adopted as a point of view,
16807from which this unity, so essential to reason and so beneficial to
16808the understanding, may be regarded as radiating. In one word, this
16809transcendental thing is merely the schema of a regulative principle, by
16810means of which Reason, so far as in her lies, extends the dominion of
16811systematic unity over the whole sphere of experience.
16812
16813The first object of an idea of this kind is the ego, considered merely
16814as a thinking nature or soul. If I wish to investigate the properties of
16815a thinking being, I must interrogate experience. But I find that I
16816can apply none of the categories to this object, the schema of these
16817categories, which is the condition of their application, being given
16818only in sensuous intuition. But I cannot thus attain to the cognition of
16819a systematic unity of all the phenomena of the internal sense. Instead,
16820therefore, of an empirical conception of what the soul really is, reason
16821takes the conception of the empirical unity of all thought, and, by
16822cogitating this unity as unconditioned and primitive, constructs the
16823rational conception or idea of a simple substance which is in itself
16824unchangeable, possessing personal identity, and in connection with other
16825real things external to it; in one word, it constructs the idea of a
16826simple self-subsistent intelligence. But the real aim of reason in this
16827procedure is the attainment of principles of systematic unity for the
16828explanation of the phenomena of the soul. That is, reason desires to
16829be able to represent all the determinations of the internal sense as
16830existing in one subject, all powers as deduced from one fundamental
16831power, all changes as mere varieties in the condition of a being which
16832is permanent and always the same, and all phenomena in space as entirely
16833different in their nature from the procedure of thought. Essential
16834simplicity (with the other attributes predicated of the ego) is regarded
16835as the mere schema of this regulative principle; it is not assumed
16836that it is the actual ground of the properties of the soul. For these
16837properties may rest upon quite different grounds, of which we are
16838completely ignorant; just as the above predicates could not give us any
16839knowledge of the soul as it is in itself, even if we regarded them as
16840valid in respect of it, inasmuch as they constitute a mere idea, which
16841cannot be represented in concreto. Nothing but good can result from
16842a psychological idea of this kind, if we only take proper care not to
16843consider it as more than an idea; that is, if we regard it as valid
16844merely in relation to the employment of reason, in the sphere of the
16845phenomena of the soul. Under the guidance of this idea, or principle,
16846no empirical laws of corporeal phenomena are called in to explain that
16847which is a phenomenon of the internal sense alone; no windy hypotheses
16848of the generation, annihilation, and palingenesis of souls are admitted.
16849Thus the consideration of this object of the internal sense is kept
16850pure, and unmixed with heterogeneous elements; while the investigation
16851of reason aims at reducing all the grounds of explanation employed
16852in this sphere of knowledge to a single principle. All this is best
16853effected, nay, cannot be effected otherwise than by means of such
16854a schema, which requires us to regard this ideal thing as an actual
16855existence. The psychological idea is, therefore, meaningless and
16856inapplicable, except as the schema of a regulative conception. For, if
16857I ask whether the soul is not really of a spiritual nature--it is
16858a question which has no meaning. From such a conception has been
16859abstracted, not merely all corporeal nature, but all nature, that is,
16860all the predicates of a possible experience; and consequently, all the
16861conditions which enable us to cogitate an object to this conception have
16862disappeared. But, if these conditions are absent, it is evident that the
16863conception is meaningless.
16864
16865The second regulative idea of speculative reason is the conception of
16866the universe. For nature is properly the only object presented to us,
16867in regard to which reason requires regulative principles. Nature is
16868twofold--thinking and corporeal nature. To cogitate the latter in regard
16869to its internal possibility, that is, to determine the application
16870of the categories to it, no idea is required--no representation which
16871transcends experience. In this sphere, therefore, an idea is impossible,
16872sensuous intuition being our only guide; while, in the sphere of
16873psychology, we require the fundamental idea (I), which contains a priori
16874a certain form of thought namely, the unity of the ego. Pure reason has,
16875therefore, nothing left but nature in general, and the completeness of
16876conditions in nature in accordance with some principle. The absolute
16877totality of the series of these conditions is an idea, which can never
16878be fully realized in the empirical exercise of reason, while it is
16879serviceable as a rule for the procedure of reason in relation to that
16880totality. It requires us, in the explanation of given phenomena (in
16881the regress or ascent in the series), to proceed as if the series were
16882infinite in itself, that is, were prolonged in indefinitum; while on
16883the other hand, where reason is regarded as itself the determining cause
16884(in the region of freedom), we are required to proceed as if we had not
16885before us an object of sense, but of the pure understanding. In this
16886latter case, the conditions do not exist in the series of phenomena, but
16887may be placed quite out of and beyond it, and the series of conditions
16888may be regarded as if it had an absolute beginning from an intelligible
16889cause. All this proves that the cosmological ideas are nothing but
16890regulative principles, and not constitutive; and that their aim is not
16891to realize an actual totality in such series. The full discussion of
16892this subject will be found in its proper place in the chapter on the
16893antinomy of pure reason.
16894
16895The third idea of pure reason, containing the hypothesis of a being
16896which is valid merely as a relative hypothesis, is that of the one and
16897all-sufficient cause of all cosmological series, in other words, the
16898idea of God. We have not the slightest ground absolutely to admit the
16899existence of an object corresponding to this idea; for what can empower
16900or authorize us to affirm the existence of a being of the highest
16901perfection--a being whose existence is absolutely necessary--merely
16902because we possess the conception of such a being? The answer is: It is
16903the existence of the world which renders this hypothesis necessary. But
16904this answer makes it perfectly evident that the idea of this being, like
16905all other speculative ideas, is essentially nothing more than a demand
16906upon reason that it shall regulate the connection which it and its
16907subordinate faculties introduce into the phenomena of the world by
16908principles of systematic unity and, consequently, that it shall regard
16909all phenomena as originating from one all-embracing being, as the
16910supreme and all-sufficient cause. From this it is plain that the only
16911aim of reason in this procedure is the establishment of its own formal
16912rule for the extension of its dominion in the world of experience; that
16913it does not aim at an extension of its cognition beyond the limits
16914of experience; and that, consequently, this idea does not contain any
16915constitutive principle.
16916
16917The highest formal unity, which is based upon ideas alone, is the unity
16918of all things--a unity in accordance with an aim or purpose; and the
16919speculative interest of reason renders it necessary to regard all order
16920in the world as if it originated from the intention and design of a
16921supreme reason. This principle unfolds to the view of reason in the
16922sphere of experience new and enlarged prospects, and invites it to
16923connect the phenomena of the world according to teleological laws,
16924and in this way to attain to the highest possible degree of systematic
16925unity. The hypothesis of a supreme intelligence, as the sole cause of
16926the universe--an intelligence which has for us no more than an ideal
16927existence--is accordingly always of the greatest service to reason.
16928Thus, if we presuppose, in relation to the figure of the earth (which
16929is round, but somewhat flattened at the poles),* or that of mountains or
16930seas, wise designs on the part of an author of the universe, we cannot
16931fail to make, by the light of this supposition, a great number of
16932interesting discoveries. If we keep to this hypothesis, as a principle
16933which is purely regulative, even error cannot be very detrimental. For,
16934in this case, error can have no more serious consequences than that,
16935where we expected to discover a teleological connection (nexus finalis),
16936only a mechanical or physical connection appears. In such a case, we
16937merely fail to find the additional form of unity we expected, but we do
16938not lose the rational unity which the mind requires in its procedure in
16939experience. But even a miscarriage of this sort cannot affect the law in
16940its general and teleological relations. For although we may convict an
16941anatomist of an error, when he connects the limb of some animal with a
16942certain purpose, it is quite impossible to prove in a single case that
16943any arrangement of nature, be it what it may, is entirely without aim or
16944design. And thus medical physiology, by the aid of a principle presented
16945to it by pure reason, extends its very limited empirical knowledge of
16946the purposes of the different parts of an organized body so far that it
16947may be asserted with the utmost confidence, and with the approbation of
16948all reflecting men, that every organ or bodily part of an animal has its
16949use and answers a certain design. Now, this is a supposition which,
16950if regarded as of a constitutive character, goes much farther than any
16951experience or observation of ours can justify. Hence it is evident that
16952it is nothing more than a regulative principle of reason, which aims
16953at the highest degree of systematic unity, by the aid of the idea of
16954a causality according to design in a supreme cause--a cause which it
16955regards as the highest intelligence.
16956
16957
16958 [*Footnote: The advantages which a circular form, in the case of the
16959earth, has over every other, are well known. But few are aware that the
16960slight flattening at the poles, which gives it the figure of a spheroid,
16961is the only cause which prevents the elevations of continents or even
16962of mountains, perhaps thrown up by some internal convulsion, from
16963continually altering the position of the axis of the earth--and that to
16964some considerable degree in a short time. The great protuberance of the
16965earth under the Equator serves to overbalance the impetus of all other
16966masses of earth, and thus to preserve the axis of the earth, so far as
16967we can observe, in its present position. And yet this wise arrangement
16968has been unthinkingly explained from the equilibrium of the formerly
16969fluid mass.]
16970
16971
16972If, however, we neglect this restriction of the idea to a purely
16973regulative influence, reason is betrayed into numerous errors. For it
16974has then left the ground of experience, in which alone are to be
16975found the criteria of truth, and has ventured into the region of the
16976incomprehensible and unsearchable, on the heights of which it loses
16977its power and collectedness, because it has completely severed its
16978connection with experience.
16979
16980The first error which arises from our employing the idea of a Supreme
16981Being as a constitutive (in repugnance to the very nature of an idea),
16982and not as a regulative principle, is the error of inactive reason
16983(ignava ratio).* We may so term every principle which requires us to
16984regard our investigations of nature as absolutely complete, and allows
16985reason to cease its inquiries, as if it had fully executed its task.
16986Thus the psychological idea of the ego, when employed as a constitutive
16987principle for the explanation of the phenomena of the soul, and for the
16988extension of our knowledge regarding this subject beyond the limits of
16989experience--even to the condition of the soul after death--is convenient
16990enough for the purposes of pure reason, but detrimental and even ruinous
16991to its interests in the sphere of nature and experience. The dogmatizing
16992spiritualist explains the unchanging unity of our personality through
16993all changes of condition from the unity of a thinking substance, the
16994interest which we take in things and events that can happen only after
16995our death, from a consciousness of the immaterial nature of our thinking
16996subject, and so on. Thus he dispenses with all empirical investigations
16997into the cause of these internal phenomena, and with all possible
16998explanations of them upon purely natural grounds; while, at the
16999dictation of a transcendent reason, he passes by the immanent sources of
17000cognition in experience, greatly to his own ease and convenience, but
17001to the sacrifice of all, genuine insight and intelligence. These
17002prejudicial consequences become still more evident, in the case of the
17003dogmatical treatment of our idea of a Supreme Intelligence, and the
17004theological system of nature (physico-theology) which is falsely based
17005upon it. For, in this case, the aims which we observe in nature, and
17006often those which we merely fancy to exist, make the investigation
17007of causes a very easy task, by directing us to refer such and such
17008phenomena immediately to the unsearchable will and counsel of the
17009Supreme Wisdom, while we ought to investigate their causes in the
17010general laws of the mechanism of matter. We are thus recommended to
17011consider the labour of reason as ended, when we have merely dispensed
17012with its employment, which is guided surely and safely only by the order
17013of nature and the series of changes in the world--which are arranged
17014according to immanent and general laws. This error may be avoided, if we
17015do not merely consider from the view-point of final aims certain parts
17016of nature, such as the division and structure of a continent, the
17017constitution and direction of certain mountain-chains, or even the
17018organization existing in the vegetable and animal kingdoms, but look
17019upon this systematic unity of nature in a perfectly general way, in
17020relation to the idea of a Supreme Intelligence. If we pursue this
17021advice, we lay as a foundation for all investigation the conformity to
17022aims of all phenomena of nature in accordance with universal laws, for
17023which no particular arrangement of nature is exempt, but only cognized
17024by us with more or less difficulty; and we possess a regulative
17025principle of the systematic unity of a teleological connection, which we
17026do not attempt to anticipate or predetermine. All that we do, and ought
17027to do, is to follow out the physico-mechanical connection in nature
17028according to general laws, with the hope of discovering, sooner or
17029later, the teleological connection also. Thus, and thus only, can the
17030principle of final unity aid in the extension of the employment
17031of reason in the sphere of experience, without being in any case
17032detrimental to its interests.
17033
17034
17035 [*Footnote: This was the term applied by the old dialecticians to a
17036sophistical argument, which ran thus: If it is your fate to die of this
17037disease, you will die, whether you employ a physician or not. Cicero
17038says that this mode of reasoning has received this appellation, because,
17039if followed, it puts an end to the employment of reason in the affairs
17040of life. For a similar reason, I have applied this designation to the
17041sophistical argument of pure reason.]
17042
17043
17044The second error which arises from the misconception of the principle
17045of systematic unity is that of perverted reason (perversa ratio, usteron
17046roteron rationis). The idea of systematic unity is available as a
17047regulative principle in the connection of phenomena according to general
17048natural laws; and, how far soever we have to travel upon the path of
17049experience to discover some fact or event, this idea requires us to
17050believe that we have approached all the more nearly to the completion of
17051its use in the sphere of nature, although that completion can never be
17052attained. But this error reverses the procedure of reason. We begin
17053by hypostatizing the principle of systematic unity, and by giving
17054an anthropomorphic determination to the conception of a Supreme
17055Intelligence, and then proceed forcibly to impose aims upon nature. Thus
17056not only does teleology, which ought to aid in the completion of unity
17057in accordance with general laws, operate to the destruction of its
17058influence, but it hinders reason from attaining its proper aim, that
17059is, the proof, upon natural grounds, of the existence of a supreme
17060intelligent cause. For, if we cannot presuppose supreme finality in
17061nature a priori, that is, as essentially belonging to nature, how can
17062we be directed to endeavour to discover this unity and, rising gradually
17063through its different degrees, to approach the supreme perfection of an
17064author of all--a perfection which is absolutely necessary, and therefore
17065cognizable a priori? The regulative principle directs us to presuppose
17066systematic unity absolutely and, consequently, as following from the
17067essential nature of things--but only as a unity of nature, not merely
17068cognized empirically, but presupposed a priori, although only in
17069an indeterminate manner. But if I insist on basing nature upon the
17070foundation of a supreme ordaining Being, the unity of nature is in
17071effect lost. For, in this case, it is quite foreign and unessential to
17072the nature of things, and cannot be cognized from the general laws of
17073nature. And thus arises a vicious circular argument, what ought to have
17074been proved having been presupposed.
17075
17076To take the regulative principle of systematic unity in nature for a
17077constitutive principle, and to hypostatize and make a cause out of that
17078which is properly the ideal ground of the consistent and harmonious
17079exercise of reason, involves reason in inextricable embarrassments. The
17080investigation of nature pursues its own path under the guidance of the
17081chain of natural causes, in accordance with the general laws of nature,
17082and ever follows the light of the idea of an author of the universe--not
17083for the purpose of deducing the finality, which it constantly pursues,
17084from this Supreme Being, but to attain to the cognition of his existence
17085from the finality which it seeks in the existence of the phenomena of
17086nature, and, if possible, in that of all things to cognize this being,
17087consequently, as absolutely necessary. Whether this latter purpose
17088succeed or not, the idea is and must always be a true one, and its
17089employment, when merely regulative, must always be accompanied by
17090truthful and beneficial results.
17091
17092Complete unity, in conformity with aims, constitutes absolute
17093perfection. But if we do not find this unity in the nature of the things
17094which go to constitute the world of experience, that is, of objective
17095cognition, consequently in the universal and necessary laws of nature,
17096how can we infer from this unity the idea of the supreme and absolutely
17097necessary perfection of a primal being, which is the origin of all
17098causality? The greatest systematic unity, and consequently teleological
17099unity, constitutes the very foundation of the possibility of the most
17100extended employment of human reason. The idea of unity is therefore
17101essentially and indissolubly connected with the nature of our reason.
17102This idea is a legislative one; and hence it is very natural that we
17103should assume the existence of a legislative reason corresponding to it,
17104from which the systematic unity of nature--the object of the operations
17105of reason--must be derived.
17106
17107In the course of our discussion of the antinomies, we stated that it is
17108always possible to answer all the questions which pure reason may raise;
17109and that the plea of the limited nature of our cognition, which is
17110unavoidable and proper in many questions regarding natural phenomena,
17111cannot in this case be admitted, because the questions raised do not
17112relate to the nature of things, but are necessarily originated by the
17113nature of reason itself, and relate to its own internal constitution. We
17114can now establish this assertion, which at first sight appeared so rash,
17115in relation to the two questions in which reason takes the greatest
17116interest, and thus complete our discussion of the dialectic of pure
17117reason.
17118
17119If, then, the question is asked, in relation to transcendental
17120theology,* first, whether there is anything distinct from the world,
17121which contains the ground of cosmical order and connection according
17122to general laws? The answer is: Certainly. For the world is a sum of
17123phenomena; there must, therefore, be some transcendental basis of these
17124phenomena, that is, a basis cogitable by the pure understanding alone.
17125If, secondly, the question is asked whether this being is substance,
17126whether it is of the greatest reality, whether it is necessary, and so
17127forth? I answer that this question is utterly without meaning. For all
17128the categories which aid me in forming a conception of an object cannot
17129be employed except in the world of sense, and are without meaning when
17130not applied to objects of actual or possible experience. Out of this
17131sphere, they are not properly conceptions, but the mere marks or indices
17132of conceptions, which we may admit, although they cannot, without the
17133help of experience, help us to understand any subject or thing. If,
17134thirdly, the question is whether we may not cogitate this being, which
17135is distinct from the world, in analogy with the objects of experience?
17136The answer is: Undoubtedly, but only as an ideal, and not as a real
17137object. That is, we must cogitate it only as an unknown substratum of
17138the systematic unity, order, and finality of the world--a unity which
17139reason must employ as the regulative principle of its investigation of
17140nature. Nay, more, we may admit into the idea certain anthropomorphic
17141elements, which are promotive of the interests of this regulative
17142principle. For it is no more than an idea, which does not relate
17143directly to a being distinct from the world, but to the regulative
17144principle of the systematic unity of the world, by means, however, of a
17145schema of this unity--the schema of a Supreme Intelligence, who is the
17146wisely-designing author of the universe. What this basis of cosmical
17147unity may be in itself, we know not--we cannot discover from the
17148idea; we merely know how we ought to employ the idea of this unity,
17149in relation to the systematic operation of reason in the sphere of
17150experience.
17151
17152
17153 [*Footnote: After what has been said of the psychological idea of
17154the ego and its proper employment as a regulative principle of the
17155operations of reason, I need not enter into details regarding the
17156transcendental illusion by which the systematic unity of all the various
17157phenomena of the internal sense is hypostatized. The procedure is in
17158this case very similar to that which has been discussed in our remarks
17159on the theological ideal.]
17160
17161
17162But, it will be asked again, can we on these grounds, admit the
17163existence of a wise and omnipotent author of the world? Without doubt;
17164and not only so, but we must assume the existence of such a being.
17165But do we thus extend the limits of our knowledge beyond the field
17166of possible experience? By no means. For we have merely presupposed a
17167something, of which we have no conception, which we do not know as it
17168is in itself; but, in relation to the systematic disposition of the
17169universe, which we must presuppose in all our observation of nature,
17170we have cogitated this unknown being in analogy with an intelligent
17171existence (an empirical conception), that is to say, we have endowed it
17172with those attributes, which, judging from the nature of our own
17173reason, may contain the ground of such a systematic unity. This idea is
17174therefore valid only relatively to the employment in experience of our
17175reason. But if we attribute to it absolute and objective validity, we
17176overlook the fact that it is merely an ideal being that we cogitate;
17177and, by setting out from a basis which is not determinable by
17178considerations drawn from experience, we place ourselves in a position
17179which incapacitates us from applying this principle to the empirical
17180employment of reason.
17181
17182But, it will be asked further, can I make any use of this conception and
17183hypothesis in my investigations into the world and nature? Yes, for this
17184very purpose was the idea established by reason as a fundamental basis.
17185But may I regard certain arrangements, which seemed to have been made in
17186conformity with some fixed aim, as the arrangements of design, and look
17187upon them as proceeding from the divine will, with the intervention,
17188however, of certain other particular arrangements disposed to that
17189end? Yes, you may do so; but at the same time you must regard it as
17190indifferent, whether it is asserted that divine wisdom has disposed all
17191things in conformity with his highest aims, or that the idea of supreme
17192wisdom is a regulative principle in the investigation of nature, and at
17193the same time a principle of the systematic unity of nature according to
17194general laws, even in those cases where we are unable to discover that
17195unity. In other words, it must be perfectly indifferent to you whether
17196you say, when you have discovered this unity: God has wisely willed
17197it so; or: Nature has wisely arranged this. For it was nothing but the
17198systematic unity, which reason requires as a basis for the investigation
17199of nature, that justified you in accepting the idea of a supreme
17200intelligence as a schema for a regulative principle; and, the farther
17201you advance in the discovery of design and finality, the more certain
17202the validity of your idea. But, as the whole aim of this regulative
17203principle was the discovery of a necessary and systematic unity in
17204nature, we have, in so far as we attain this, to attribute our success
17205to the idea of a Supreme Being; while, at the same time, we cannot,
17206without involving ourselves in contradictions, overlook the general
17207laws of nature, as it was in reference to them alone that this idea was
17208employed. We cannot, I say, overlook the general laws of nature, and
17209regard this conformity to aims observable in nature as contingent or
17210hyperphysical in its origin; inasmuch as there is no ground which can
17211justify us in the admission of a being with such properties distinct
17212from and above nature. All that we are authorized to assert is that
17213this idea may be employed as a principle, and that the properties of
17214the being which is assumed to correspond to it may be regarded as
17215systematically connected in analogy with the causal determination of
17216phenomena.
17217
17218For the same reasons we are justified in introducing into the idea of
17219the supreme cause other anthropomorphic elements (for without these we
17220could not predicate anything of it); we may regard it as allowable
17221to cogitate this cause as a being with understanding, the feelings of
17222pleasure and displeasure, and faculties of desire and will corresponding
17223to these. At the same time, we may attribute to this being infinite
17224perfection--a perfection which necessarily transcends that which our
17225knowledge of the order and design in the world authorize us to predicate
17226of it. For the regulative law of systematic unity requires us to study
17227nature on the supposition that systematic and final unity in infinitum
17228is everywhere discoverable, even in the highest diversity. For, although
17229we may discover little of this cosmical perfection, it belongs to the
17230legislative prerogative of reason to require us always to seek for
17231and to expect it; while it must always be beneficial to institute all
17232inquiries into nature in accordance with this principle. But it is
17233evident that, by this idea of a supreme author of all, which I place as
17234the foundation of all inquiries into nature, I do not mean to assert
17235the existence of such a being, or that I have any knowledge of its
17236existence; and, consequently, I do not really deduce anything from the
17237existence of this being, but merely from its idea, that is to say, from
17238the nature of things in this world, in accordance with this idea. A
17239certain dim consciousness of the true use of this idea seems to have
17240dictated to the philosophers of all times the moderate language used
17241by them regarding the cause of the world. We find them employing
17242the expressions wisdom and care of nature, and divine wisdom, as
17243synonymous--nay, in purely speculative discussions, preferring the
17244former, because it does not carry the appearance of greater pretensions
17245than such as we are entitled to make, and at the same time directs
17246reason to its proper field of action--nature and her phenomena.
17247
17248Thus, pure reason, which at first seemed to promise us nothing less
17249than the extension of our cognition beyond the limits of experience,
17250is found, when thoroughly examined, to contain nothing but regulative
17251principles, the virtue and function of which is to introduce into our
17252cognition a higher degree of unity than the understanding could of
17253itself. These principles, by placing the goal of all our struggles at
17254so great a distance, realize for us the most thorough connection
17255between the different parts of our cognition, and the highest degree of
17256systematic unity. But, on the other hand, if misunderstood and employed
17257as constitutive principles of transcendent cognition, they become the
17258parents of illusions and contradictions, while pretending to introduce
17259us to new regions of knowledge.
17260
17261Thus all human cognition begins with intuitions, proceeds from thence to
17262conceptions, and ends with ideas. Although it possesses, in relation
17263to all three elements, a priori sources of cognition, which seemed
17264to transcend the limits of all experience, a thoroughgoing criticism
17265demonstrates that speculative reason can never, by the aid of these
17266elements, pass the bounds of possible experience, and that the proper
17267destination of this highest faculty of cognition is to employ all
17268methods, and all the principles of these methods, for the purpose of
17269penetrating into the innermost secrets of nature, by the aid of the
17270principles of unity (among all kinds of which teleological unity is
17271the highest), while it ought not to attempt to soar above the sphere of
17272experience, beyond which there lies nought for us but the void inane.
17273The critical examination, in our Transcendental Analytic, of all the
17274propositions which professed to extend cognition beyond the sphere of
17275experience, completely demonstrated that they can only conduct us to
17276a possible experience. If we were not distrustful even of the clearest
17277abstract theorems, if we were not allured by specious and inviting
17278prospects to escape from the constraining power of their evidence, we
17279might spare ourselves the laborious examination of all the dialectical
17280arguments which a transcendent reason adduces in support of its
17281pretensions; for we should know with the most complete certainty that,
17282however honest such professions might be, they are null and valueless,
17283because they relate to a kind of knowledge to which no man can by any
17284possibility attain. But, as there is no end to discussion, if we cannot
17285discover the true cause of the illusions by which even the wisest are
17286deceived, and as the analysis of all our transcendent cognition into its
17287elements is of itself of no slight value as a psychological study, while
17288it is a duty incumbent on every philosopher--it was found necessary to
17289investigate the dialectical procedure of reason in its primary sources.
17290And as the inferences of which this dialectic is the parent are not only
17291deceitful, but naturally possess a profound interest for humanity, it
17292was advisable at the same time, to give a full account of the momenta of
17293this dialectical procedure, and to deposit it in the archives of human
17294reason, as a warning to all future metaphysicians to avoid these causes
17295of speculative error.
17296
17297
17298
17299II.
17300
17301TRANSCENDENTAL DOCTRINE OF METHOD.
17302
17303If we regard the sum of the cognition of pure speculative reason as an
17304edifice, the idea of which, at least, exists in the human mind, it may
17305be said that we have in the Transcendental Doctrine of Elements examined
17306the materials and determined to what edifice these belong, and what
17307its height and stability. We have found, indeed, that, although we had
17308purposed to build for ourselves a tower which should reach to Heaven,
17309the supply of materials sufficed merely for a habitation, which was
17310spacious enough for all terrestrial purposes, and high enough to
17311enable us to survey the level plain of experience, but that the bold
17312undertaking designed necessarily failed for want of materials--not to
17313mention the confusion of tongues, which gave rise to endless disputes
17314among the labourers on the plan of the edifice, and at last scattered
17315them over all the world, each to erect a separate building for himself,
17316according to his own plans and his own inclinations. Our present task
17317relates not to the materials, but to the plan of an edifice; and, as we
17318have had sufficient warning not to venture blindly upon a design which
17319may be found to transcend our natural powers, while, at the same time,
17320we cannot give up the intention of erecting a secure abode for the mind,
17321we must proportion our design to the material which is presented to us,
17322and which is, at the same time, sufficient for all our wants.
17323
17324I understand, then, by the transcendental doctrine of method, the
17325determination of the formal conditions of a complete system of pure
17326reason. We shall accordingly have to treat of the discipline, the canon,
17327the architectonic, and, finally, the history of pure reason. This part
17328of our Critique will accomplish, from the transcendental point of view,
17329what has been usually attempted, but miserably executed, under the name
17330of practical logic. It has been badly executed, I say, because general
17331logic, not being limited to any particular kind of cognition (not
17332even to the pure cognition of the understanding) nor to any particular
17333objects, it cannot, without borrowing from other sciences, do more than
17334present merely the titles or signs of possible methods and the technical
17335expressions, which are employed in the systematic parts of all sciences;
17336and thus the pupil is made acquainted with names, the meaning and
17337application of which he is to learn only at some future time.
17338
17339
17340
17341CHAPTER I. The Discipline of Pure Reason.
17342
17343Negative judgements--those which are so not merely as regards their
17344logical form, but in respect of their content--are not commonly held in
17345especial respect. They are, on the contrary, regarded as jealous enemies
17346of our insatiable desire for knowledge; and it almost requires an
17347apology to induce us to tolerate, much less to prize and to respect
17348them.
17349
17350All propositions, indeed, may be logically expressed in a negative form;
17351but, in relation to the content of our cognition, the peculiar province
17352of negative judgements is solely to prevent error. For this reason, too,
17353negative propositions, which are framed for the purpose of correcting
17354false cognitions where error is absolutely impossible, are undoubtedly
17355true, but inane and senseless; that is, they are in reality purposeless
17356and, for this reason, often very ridiculous. Such is the proposition
17357of the schoolman that Alexander could not have subdued any countries
17358without an army.
17359
17360But where the limits of our possible cognition are very much contracted,
17361the attraction to new fields of knowledge great, the illusions to
17362which the mind is subject of the most deceptive character, and the
17363evil consequences of error of no inconsiderable magnitude--the negative
17364element in knowledge, which is useful only to guard us against error,
17365is of far more importance than much of that positive instruction which
17366makes additions to the sum of our knowledge. The restraint which is
17367employed to repress, and finally to extirpate the constant inclination
17368to depart from certain rules, is termed discipline. It is distinguished
17369from culture, which aims at the formation of a certain degree of skill,
17370without attempting to repress or to destroy any other mental power,
17371already existing. In the cultivation of a talent, which has given
17372evidence of an impulse towards self-development, discipline takes a
17373negative,* culture and doctrine a positive, part.
17374
17375
17376
17377 [*Footnote: I am well aware that, in the language of the schools, the
17378term discipline is usually employed as synonymous with instruction.
17379But there are so many cases in which it is necessary to distinguish the
17380notion of the former, as a course of corrective training, from that of
17381the latter, as the communication of knowledge, and the nature of things
17382itself demands the appropriation of the most suitable expressions for
17383this distinction, that it is my desire that the former terms should
17384never be employed in any other than a negative signification.]
17385
17386
17387
17388That natural dispositions and talents (such as imagination and wit),
17389which ask a free and unlimited development, require in many respects the
17390corrective influence of discipline, every one will readily grant. But
17391it may well appear strange that reason, whose proper duty it is to
17392prescribe rules of discipline to all the other powers of the mind,
17393should itself require this corrective. It has, in fact, hitherto
17394escaped this humiliation, only because, in presence of its magnificent
17395pretensions and high position, no one could readily suspect it to be
17396capable of substituting fancies for conceptions, and words for things.
17397
17398Reason, when employed in the field of experience, does not stand in need
17399of criticism, because its principles are subjected to the continual test
17400of empirical observations. Nor is criticism requisite in the sphere of
17401mathematics, where the conceptions of reason must always be presented
17402in concreto in pure intuition, and baseless or arbitrary assertions are
17403discovered without difficulty. But where reason is not held in a plain
17404track by the influence of empirical or of pure intuition, that is, when
17405it is employed in the transcendental sphere of pure conceptions, it
17406stands in great need of discipline, to restrain its propensity to
17407overstep the limits of possible experience and to keep it from wandering
17408into error. In fact, the utility of the philosophy of pure reason is
17409entirely of this negative character. Particular errors may be corrected
17410by particular animadversions, and the causes of these errors may be
17411eradicated by criticism. But where we find, as in the case of pure
17412reason, a complete system of illusions and fallacies, closely connected
17413with each other and depending upon grand general principles, there
17414seems to be required a peculiar and negative code of mental legislation,
17415which, under the denomination of a discipline, and founded upon the
17416nature of reason and the objects of its exercise, shall constitute a
17417system of thorough examination and testing, which no fallacy will be
17418able to withstand or escape from, under whatever disguise or concealment
17419it may lurk.
17420
17421But the reader must remark that, in this the second division of our
17422transcendental Critique the discipline of pure reason is not directed
17423to the content, but to the method of the cognition of pure reason. The
17424former task has been completed in the doctrine of elements. But there
17425is so much similarity in the mode of employing the faculty of reason,
17426whatever be the object to which it is applied, while, at the same time,
17427its employment in the transcendental sphere is so essentially different
17428in kind from every other, that, without the warning negative influence
17429of a discipline specially directed to that end, the errors are
17430unavoidable which spring from the unskillful employment of the methods
17431which are originated by reason but which are out of place in this
17432sphere.
17433
17434
17435
17436SECTION I. The Discipline of Pure Reason in the Sphere of Dogmatism.
17437
17438The science of mathematics presents the most brilliant example of the
17439extension of the sphere of pure reason without the aid of experience.
17440Examples are always contagious; and they exert an especial influence on
17441the same faculty, which naturally flatters itself that it will have
17442the same good fortune in other case as fell to its lot in one fortunate
17443instance. Hence pure reason hopes to be able to extend its empire in the
17444transcendental sphere with equal success and security, especially
17445when it applies the same method which was attended with such brilliant
17446results in the science of mathematics. It is, therefore, of the
17447highest importance for us to know whether the method of arriving at
17448demonstrative certainty, which is termed mathematical, be identical with
17449that by which we endeavour to attain the same degree of certainty in
17450philosophy, and which is termed in that science dogmatical.
17451
17452Philosophical cognition is the cognition of reason by means of
17453conceptions; mathematical cognition is cognition by means of the
17454construction of conceptions. The construction of a conception is
17455the presentation a priori of the intuition which corresponds to the
17456conception. For this purpose a non-empirical intuition is requisite,
17457which, as an intuition, is an individual object; while, as the
17458construction of a conception (a general representation), it must be seen
17459to be universally valid for all the possible intuitions which rank under
17460that conception. Thus I construct a triangle, by the presentation of the
17461object which corresponds to this conception, either by mere imagination,
17462in pure intuition, or upon paper, in empirical intuition, in both cases
17463completely a priori, without borrowing the type of that figure from any
17464experience. The individual figure drawn upon paper is empirical; but
17465it serves, notwithstanding, to indicate the conception, even in its
17466universality, because in this empirical intuition we keep our eye merely
17467on the act of the construction of the conception, and pay no attention
17468to the various modes of determining it, for example, its size, the
17469length of its sides, the size of its angles, these not in the least
17470affecting the essential character of the conception.
17471
17472Philosophical cognition, accordingly, regards the particular only in
17473the general; mathematical the general in the particular, nay, in the
17474individual. This is done, however, entirely a priori and by means of
17475pure reason, so that, as this individual figure is determined under
17476certain universal conditions of construction, the object of the
17477conception, to which this individual figure corresponds as its schema,
17478must be cogitated as universally determined.
17479
17480The essential difference of these two modes of cognition consists,
17481therefore, in this formal quality; it does not regard the difference of
17482the matter or objects of both. Those thinkers who aim at distinguishing
17483philosophy from mathematics by asserting that the former has to do with
17484quality merely, and the latter with quantity, have mistaken the effect
17485for the cause. The reason why mathematical cognition can relate only to
17486quantity is to be found in its form alone. For it is the conception of
17487quantities only that is capable of being constructed, that is, presented
17488a priori in intuition; while qualities cannot be given in any other than
17489an empirical intuition. Hence the cognition of qualities by reason is
17490possible only through conceptions. No one can find an intuition which
17491shall correspond to the conception of reality, except in experience;
17492it cannot be presented to the mind a priori and antecedently to the
17493empirical consciousness of a reality. We can form an intuition, by means
17494of the mere conception of it, of a cone, without the aid of experience;
17495but the colour of the cone we cannot know except from experience. I
17496cannot present an intuition of a cause, except in an example which
17497experience offers to me. Besides, philosophy, as well as mathematics,
17498treats of quantities; as, for example, of totality, infinity, and so
17499on. Mathematics, too, treats of the difference of lines and surfaces--as
17500spaces of different quality, of the continuity of extension--as a
17501quality thereof. But, although in such cases they have a common object,
17502the mode in which reason considers that object is very different in
17503philosophy from what it is in mathematics. The former confines itself
17504to the general conceptions; the latter can do nothing with a mere
17505conception, it hastens to intuition. In this intuition it regards the
17506conception in concreto, not empirically, but in an a priori intuition,
17507which it has constructed; and in which, all the results which follow
17508from the general conditions of the construction of the conception are in
17509all cases valid for the object of the constructed conception.
17510
17511Suppose that the conception of a triangle is given to a philosopher
17512and that he is required to discover, by the philosophical method, what
17513relation the sum of its angles bears to a right angle. He has nothing
17514before him but the conception of a figure enclosed within three right
17515lines, and, consequently, with the same number of angles. He may analyse
17516the conception of a right line, of an angle, or of the number three
17517as long as he pleases, but he will not discover any properties not
17518contained in these conceptions. But, if this question is proposed to
17519a geometrician, he at once begins by constructing a triangle. He knows
17520that two right angles are equal to the sum of all the contiguous angles
17521which proceed from one point in a straight line; and he goes on to
17522produce one side of his triangle, thus forming two adjacent angles which
17523are together equal to two right angles. He then divides the exterior of
17524these angles, by drawing a line parallel with the opposite side of the
17525triangle, and immediately perceives that he has thus got an exterior
17526adjacent angle which is equal to the interior. Proceeding in this way,
17527through a chain of inferences, and always on the ground of intuition, he
17528arrives at a clear and universally valid solution of the question.
17529
17530But mathematics does not confine itself to the construction of
17531quantities (quanta), as in the case of geometry; it occupies itself
17532with pure quantity also (quantitas), as in the case of algebra, where
17533complete abstraction is made of the properties of the object indicated
17534by the conception of quantity. In algebra, a certain method of
17535notation by signs is adopted, and these indicate the different possible
17536constructions of quantities, the extraction of roots, and so on. After
17537having thus denoted the general conception of quantities, according to
17538their different relations, the different operations by which quantity
17539or number is increased or diminished are presented in intuition in
17540accordance with general rules. Thus, when one quantity is to be divided
17541by another, the signs which denote both are placed in the form peculiar
17542to the operation of division; and thus algebra, by means of a symbolical
17543construction of quantity, just as geometry, with its ostensive or
17544geometrical construction (a construction of the objects themselves),
17545arrives at results which discursive cognition cannot hope to reach by
17546the aid of mere conceptions.
17547
17548Now, what is the cause of this difference in the fortune of the
17549philosopher and the mathematician, the former of whom follows the path
17550of conceptions, while the latter pursues that of intuitions, which he
17551represents, a priori, in correspondence with his conceptions? The cause
17552is evident from what has been already demonstrated in the introduction
17553to this Critique. We do not, in the present case, want to discover
17554analytical propositions, which may be produced merely by analysing our
17555conceptions--for in this the philosopher would have the advantage over
17556his rival; we aim at the discovery of synthetical propositions--such
17557synthetical propositions, moreover, as can be cognized a priori. I must
17558not confine myself to that which I actually cogitate in my conception
17559of a triangle, for this is nothing more than the mere definition; I
17560must try to go beyond that, and to arrive at properties which are not
17561contained in, although they belong to, the conception. Now, this is
17562impossible, unless I determine the object present to my mind according
17563to the conditions, either of empirical, or of pure, intuition. In the
17564former case, I should have an empirical proposition (arrived at by
17565actual measurement of the angles of the triangle), which would possess
17566neither universality nor necessity; but that would be of no value. In
17567the latter, I proceed by geometrical construction, by means of which I
17568collect, in a pure intuition, just as I would in an empirical intuition,
17569all the various properties which belong to the schema of a triangle
17570in general, and consequently to its conception, and thus construct
17571synthetical propositions which possess the attribute of universality.
17572
17573It would be vain to philosophize upon the triangle, that is, to reflect
17574on it discursively; I should get no further than the definition with
17575which I had been obliged to set out. There are certainly transcendental
17576synthetical propositions which are framed by means of pure conceptions,
17577and which form the peculiar distinction of philosophy; but these do not
17578relate to any particular thing, but to a thing in general, and enounce
17579the conditions under which the perception of it may become a part of
17580possible experience. But the science of mathematics has nothing to do
17581with such questions, nor with the question of existence in any fashion;
17582it is concerned merely with the properties of objects in themselves,
17583only in so far as these are connected with the conception of the
17584objects.
17585
17586In the above example, we merely attempted to show the great difference
17587which exists between the discursive employment of reason in the sphere
17588of conceptions, and its intuitive exercise by means of the construction
17589of conceptions. The question naturally arises: What is the cause which
17590necessitates this twofold exercise of reason, and how are we to discover
17591whether it is the philosophical or the mathematical method which reason
17592is pursuing in an argument?
17593
17594All our knowledge relates, finally, to possible intuitions, for it
17595is these alone that present objects to the mind. An a priori or
17596non-empirical conception contains either a pure intuition--and in this
17597case it can be constructed; or it contains nothing but the synthesis of
17598possible intuitions, which are not given a priori. In this latter case,
17599it may help us to form synthetical a priori judgements, but only in the
17600discursive method, by conceptions, not in the intuitive, by means of the
17601construction of conceptions.
17602
17603The only a priori intuition is that of the pure form of phenomena--space
17604and time. A conception of space and time as quanta may be presented
17605a priori in intuition, that is, constructed, either alone with their
17606quality (figure), or as pure quantity (the mere synthesis of the
17607homogeneous), by means of number. But the matter of phenomena, by which
17608things are given in space and time, can be presented only in perception,
17609a posteriori. The only conception which represents a priori this
17610empirical content of phenomena is the conception of a thing in general;
17611and the a priori synthetical cognition of this conception can give
17612us nothing more than the rule for the synthesis of that which may be
17613contained in the corresponding a posteriori perception; it is utterly
17614inadequate to present an a priori intuition of the real object, which
17615must necessarily be empirical.
17616
17617Synthetical propositions, which relate to things in general, an a priori
17618intuition of which is impossible, are transcendental. For this
17619reason transcendental propositions cannot be framed by means of the
17620construction of conceptions; they are a priori, and based entirely on
17621conceptions themselves. They contain merely the rule, by which we are to
17622seek in the world of perception or experience the synthetical unity
17623of that which cannot be intuited a priori. But they are incompetent
17624to present any of the conceptions which appear in them in an a priori
17625intuition; these can be given only a posteriori, in experience, which,
17626however, is itself possible only through these synthetical principles.
17627
17628If we are to form a synthetical judgement regarding a conception, we
17629must go beyond it, to the intuition in which it is given. If we keep
17630to what is contained in the conception, the judgement is merely
17631analytical--it is merely an explanation of what we have cogitated in the
17632conception. But I can pass from the conception to the pure or empirical
17633intuition which corresponds to it. I can proceed to examine my
17634conception in concreto, and to cognize, either a priori or a posterio,
17635what I find in the object of the conception. The former--a priori
17636cognition--is rational-mathematical cognition by means of the
17637construction of the conception; the latter--a posteriori cognition--is
17638purely empirical cognition, which does not possess the attributes of
17639necessity and universality. Thus I may analyse the conception I have
17640of gold; but I gain no new information from this analysis, I merely
17641enumerate the different properties which I had connected with the notion
17642indicated by the word. My knowledge has gained in logical clearness
17643and arrangement, but no addition has been made to it. But if I take the
17644matter which is indicated by this name, and submit it to the examination
17645of my senses, I am enabled to form several synthetical--although still
17646empirical--propositions. The mathematical conception of a triangle I
17647should construct, that is, present a priori in intuition, and in
17648this way attain to rational-synthetical cognition. But when the
17649transcendental conception of reality, or substance, or power is
17650presented to my mind, I find that it does not relate to or indicate
17651either an empirical or pure intuition, but that it indicates merely the
17652synthesis of empirical intuitions, which cannot of course be given
17653a priori. The synthesis in such a conception cannot proceed a
17654priori--without the aid of experience--to the intuition which
17655corresponds to the conception; and, for this reason, none of these
17656conceptions can produce a determinative synthetical proposition, they
17657can never present more than a principle of the synthesis* of possible
17658empirical intuitions. A transcendental proposition is, therefore, a
17659synthetical cognition of reason by means of pure conceptions and the
17660discursive method, and it renders possible all synthetical unity in
17661empirical cognition, though it cannot present us with any intuition a
17662priori.
17663
17664
17665
17666 [*Footnote: In the case of the conception of cause, I do really go
17667beyond the empirical conception of an event--but not to the
17668intuition which presents this conception in concreto, but only to the
17669time-conditions, which may be found in experience to correspond to
17670the conception. My procedure is, therefore, strictly according to
17671conceptions; I cannot in a case of this kind employ the construction of
17672conceptions, because the conception is merely a rule for the synthesis
17673of perceptions, which are not pure intuitions, and which, therefore,
17674cannot be given a priori.]
17675
17676
17677
17678There is thus a twofold exercise of reason. Both modes have the
17679properties of universality and an a priori origin in common, but are,
17680in their procedure, of widely different character. The reason of this is
17681that in the world of phenomena, in which alone objects are presented to
17682our minds, there are two main elements--the form of intuition (space and
17683time), which can be cognized and determined completely a priori, and the
17684matter or content--that which is presented in space and time, and which,
17685consequently, contains a something--an existence corresponding to our
17686powers of sensation. As regards the latter, which can never be given in
17687a determinate mode except by experience, there are no a priori notions
17688which relate to it, except the undetermined conceptions of the synthesis
17689of possible sensations, in so far as these belong (in a possible
17690experience) to the unity of consciousness. As regards the former, we
17691can determine our conceptions a priori in intuition, inasmuch as we are
17692ourselves the creators of the objects of the conceptions in space and
17693time--these objects being regarded simply as quanta. In the one case,
17694reason proceeds according to conceptions and can do nothing more than
17695subject phenomena to these--which can only be determined empirically,
17696that is, a posteriori--in conformity, however, with those conceptions as
17697the rules of all empirical synthesis. In the other case, reason proceeds
17698by the construction of conceptions; and, as these conceptions relate
17699to an a priori intuition, they may be given and determined in pure
17700intuition a priori, and without the aid of empirical data. The
17701examination and consideration of everything that exists in space
17702or time--whether it is a quantum or not, in how far the particular
17703something (which fills space or time) is a primary substratum, or a mere
17704determination of some other existence, whether it relates to anything
17705else--either as cause or effect, whether its existence is isolated or in
17706reciprocal connection with and dependence upon others, the possibility
17707of this existence, its reality and necessity or opposites--all these
17708form part of the cognition of reason on the ground of conceptions, and
17709this cognition is termed philosophical. But to determine a priori an
17710intuition in space (its figure), to divide time into periods, or merely
17711to cognize the quantity of an intuition in space and time, and to
17712determine it by number--all this is an operation of reason by means of
17713the construction of conceptions, and is called mathematical.
17714
17715The success which attends the efforts of reason in the sphere of
17716mathematics naturally fosters the expectation that the same good fortune
17717will be its lot, if it applies the mathematical method in other regions
17718of mental endeavour besides that of quantities. Its success is thus
17719great, because it can support all its conceptions by a priori intuitions
17720and, in this way, make itself a master, as it were, over nature; while
17721pure philosophy, with its a priori discursive conceptions, bungles
17722about in the world of nature, and cannot accredit or show any a priori
17723evidence of the reality of these conceptions. Masters in the science of
17724mathematics are confident of the success of this method; indeed, it is a
17725common persuasion that it is capable of being applied to any subject of
17726human thought. They have hardly ever reflected or philosophized on
17727their favourite science--a task of great difficulty; and the specific
17728difference between the two modes of employing the faculty of reason
17729has never entered their thoughts. Rules current in the field of common
17730experience, and which common sense stamps everywhere with its approval,
17731are regarded by them as axiomatic. From what source the conceptions of
17732space and time, with which (as the only primitive quanta) they have
17733to deal, enter their minds, is a question which they do not trouble
17734themselves to answer; and they think it just as unnecessary to examine
17735into the origin of the pure conceptions of the understanding and the
17736extent of their validity. All they have to do with them is to employ
17737them. In all this they are perfectly right, if they do not overstep the
17738limits of the sphere of nature. But they pass, unconsciously, from the
17739world of sense to the insecure ground of pure transcendental conceptions
17740(instabilis tellus, innabilis unda), where they can neither stand nor
17741swim, and where the tracks of their footsteps are obliterated by time;
17742while the march of mathematics is pursued on a broad and magnificent
17743highway, which the latest posterity shall frequent without fear of
17744danger or impediment.
17745
17746As we have taken upon us the task of determining, clearly and certainly,
17747the limits of pure reason in the sphere of transcendentalism, and as
17748the efforts of reason in this direction are persisted in, even after the
17749plainest and most expressive warnings, hope still beckoning us past the
17750limits of experience into the splendours of the intellectual world--it
17751becomes necessary to cut away the last anchor of this fallacious and
17752fantastic hope. We shall, accordingly, show that the mathematical
17753method is unattended in the sphere of philosophy by the least
17754advantage--except, perhaps, that it more plainly exhibits its own
17755inadequacy--that geometry and philosophy are two quite different things,
17756although they go band in hand in hand in the field of natural science,
17757and, consequently, that the procedure of the one can never be imitated
17758by the other.
17759
17760The evidence of mathematics rests upon definitions, axioms, and
17761demonstrations. I shall be satisfied with showing that none of these
17762forms can be employed or imitated in philosophy in the sense in which
17763they are understood by mathematicians; and that the geometrician, if
17764he employs his method in philosophy, will succeed only in building
17765card-castles, while the employment of the philosophical method in
17766mathematics can result in nothing but mere verbiage. The essential
17767business of philosophy, indeed, is to mark out the limits of the
17768science; and even the mathematician, unless his talent is naturally
17769circumscribed and limited to this particular department of knowledge,
17770cannot turn a deaf ear to the warnings of philosophy, or set himself
17771above its direction.
17772
17773I. Of Definitions. A definition is, as the term itself indicates, the
17774representation, upon primary grounds, of the complete conception of
17775a thing within its own limits.* Accordingly, an empirical conception
17776cannot be defined, it can only be explained. For, as there are in such
17777a conception only a certain number of marks or signs, which denote a
17778certain class of sensuous objects, we can never be sure that we do not
17779cogitate under the word which indicates the same object, at one time
17780a greater, at another a smaller number of signs. Thus, one person may
17781cogitate in his conception of gold, in addition to its properties of
17782weight, colour, malleability, that of resisting rust, while another
17783person may be ignorant of this quality. We employ certain signs only so
17784long as we require them for the sake of distinction; new observations
17785abstract some and add new ones, so that an empirical conception never
17786remains within permanent limits. It is, in fact, useless to define a
17787conception of this kind. If, for example, we are speaking of water and
17788its properties, we do not stop at what we actually think by the word
17789water, but proceed to observation and experiment; and the word, with
17790the few signs attached to it, is more properly a designation than a
17791conception of the thing. A definition in this case would evidently be
17792nothing more than a determination of the word. In the second place, no
17793a priori conception, such as those of substance, cause, right, fitness,
17794and so on, can be defined. For I can never be sure, that the clear
17795representation of a given conception (which is given in a confused
17796state) has been fully developed, until I know that the representation
17797is adequate with its object. But, inasmuch as the conception, as it is
17798presented to the mind, may contain a number of obscure representations,
17799which we do not observe in our analysis, although we employ them in our
17800application of the conception, I can never be sure that my analysis is
17801complete, while examples may make this probable, although they can never
17802demonstrate the fact. Instead of the word definition, I should rather
17803employ the term exposition--a more modest expression, which the critic
17804may accept without surrendering his doubts as to the completeness of the
17805analysis of any such conception. As, therefore, neither empirical nor a
17806priori conceptions are capable of definition, we have to see whether the
17807only other kind of conceptions--arbitrary conceptions--can be subjected
17808to this mental operation. Such a conception can always be defined; for
17809I must know thoroughly what I wished to cogitate in it, as it was I who
17810created it, and it was not given to my mind either by the nature of my
17811understanding or by experience. At the same time, I cannot say that, by
17812such a definition, I have defined a real object. If the conception is
17813based upon empirical conditions, if, for example, I have a conception of
17814a clock for a ship, this arbitrary conception does not assure me of the
17815existence or even of the possibility of the object. My definition of
17816such a conception would with more propriety be termed a declaration of
17817a project than a definition of an object. There are no other conceptions
17818which can bear definition, except those which contain an arbitrary
17819synthesis, which can be constructed a priori. Consequently, the science
17820of mathematics alone possesses definitions. For the object here thought
17821is presented a priori in intuition; and thus it can never contain more
17822or less than the conception, because the conception of the object has
17823been given by the definition--and primarily, that is, without deriving
17824the definition from any other source. Philosophical definitions are,
17825therefore, merely expositions of given conceptions, while mathematical
17826definitions are constructions of conceptions originally formed by the
17827mind itself; the former are produced by analysis, the completeness of
17828which is never demonstratively certain, the latter by a synthesis. In
17829a mathematical definition the conception is formed, in a philosophical
17830definition it is only explained. From this it follows:
17831
17832
17833 [*Footnote: The definition must describe the conception completely that
17834is, omit none of the marks or signs of which it composed; within its own
17835limits, that is, it must be precise, and enumerate no more signs than
17836belong to the conception; and on primary grounds, that is to say, the
17837limitations of the bounds of the conception must not be deduced from
17838other conceptions, as in this case a proof would be necessary, and the
17839so-called definition would be incapable of taking its place at the bead
17840of all the judgements we have to form regarding an object.]
17841
17842
17843(a) That we must not imitate, in philosophy, the mathematical usage of
17844commencing with definitions--except by way of hypothesis or experiment.
17845For, as all so-called philosophical definitions are merely analyses of
17846given conceptions, these conceptions, although only in a confused form,
17847must precede the analysis; and the incomplete exposition must precede
17848the complete, so that we may be able to draw certain inferences from the
17849characteristics which an incomplete analysis has enabled us to discover,
17850before we attain to the complete exposition or definition of the
17851conception. In one word, a full and clear definition ought, in
17852philosophy, rather to form the conclusion than the commencement of our
17853labours.* In mathematics, on the contrary, we cannot have a conception
17854prior to the definition; it is the definition which gives us the
17855conception, and it must for this reason form the commencement of every
17856chain of mathematical reasoning.
17857
17858
17859 [*Footnote: Philosophy abounds in faulty definitions, especially such as
17860contain some of the elements requisite to form a complete definition.
17861If a conception could not be employed in reasoning before it had been
17862defined, it would fare ill with all philosophical thought. But,
17863as incompletely defined conceptions may always be employed without
17864detriment to truth, so far as our analysis of the elements contained in
17865them proceeds, imperfect definitions, that is, propositions which are
17866properly not definitions, but merely approximations thereto, may be used
17867with great advantage. In mathematics, definition belongs ad esse, in
17868philosophy ad melius esse. It is a difficult task to construct a proper
17869definition. Jurists are still without a complete definition of the idea
17870of right.]
17871
17872
17873(b) Mathematical definitions cannot be erroneous. For the conception is
17874given only in and through the definition, and thus it contains only what
17875has been cogitated in the definition. But although a definition cannot
17876be incorrect, as regards its content, an error may sometimes, although
17877seldom, creep into the form. This error consists in a want of precision.
17878Thus the common definition of a circle--that it is a curved line,
17879every point in which is equally distant from another point called the
17880centre--is faulty, from the fact that the determination indicated by the
17881word curved is superfluous. For there ought to be a particular theorem,
17882which may be easily proved from the definition, to the effect that every
17883line, which has all its points at equal distances from another point,
17884must be a curved line--that is, that not even the smallest part of
17885it can be straight. Analytical definitions, on the other hand, may be
17886erroneous in many respects, either by the introduction of signs which do
17887not actually exist in the conception, or by wanting in that completeness
17888which forms the essential of a definition. In the latter case, the
17889definition is necessarily defective, because we can never be fully
17890certain of the completeness of our analysis. For these reasons, the
17891method of definition employed in mathematics cannot be imitated in
17892philosophy.
17893
178942. Of Axioms. These, in so far as they are immediately certain, are a
17895priori synthetical principles. Now, one conception cannot be connected
17896synthetically and yet immediately with another; because, if we wish to
17897proceed out of and beyond a conception, a third mediating cognition is
17898necessary. And, as philosophy is a cognition of reason by the aid
17899of conceptions alone, there is to be found in it no principle which
17900deserves to be called an axiom. Mathematics, on the other hand, may
17901possess axioms, because it can always connect the predicates of an
17902object a priori, and without any mediating term, by means of the
17903construction of conceptions in intuition. Such is the case with the
17904proposition: Three points can always lie in a plane. On the other hand,
17905no synthetical principle which is based upon conceptions, can ever
17906be immediately certain (for example, the proposition: Everything that
17907happens has a cause), because I require a mediating term to connect
17908the two conceptions of event and cause--namely, the condition of
17909time-determination in an experience, and I cannot cognize any such
17910principle immediately and from conceptions alone. Discursive principles
17911are, accordingly, very different from intuitive principles or axioms.
17912The former always require deduction, which in the case of the latter
17913may be altogether dispensed with. Axioms are, for this reason, always
17914self-evident, while philosophical principles, whatever may be the degree
17915of certainty they possess, cannot lay any claim to such a distinction.
17916No synthetical proposition of pure transcendental reason can be so
17917evident, as is often rashly enough declared, as the statement, twice two
17918are four. It is true that in the Analytic I introduced into the list of
17919principles of the pure understanding, certain axioms of intuition; but
17920the principle there discussed was not itself an axiom, but served merely
17921to present the principle of the possibility of axioms in general, while
17922it was really nothing more than a principle based upon conceptions. For
17923it is one part of the duty of transcendental philosophy to establish
17924the possibility of mathematics itself. Philosophy possesses, then, no
17925axioms, and has no right to impose its a priori principles upon thought,
17926until it has established their authority and validity by a thoroughgoing
17927deduction.
17928
179293. Of Demonstrations. Only an apodeictic proof, based upon intuition,
17930can be termed a demonstration. Experience teaches us what is, but it
17931cannot convince us that it might not have been otherwise. Hence a proof
17932upon empirical grounds cannot be apodeictic. A priori conceptions, in
17933discursive cognition, can never produce intuitive certainty or evidence,
17934however certain the judgement they present may be. Mathematics alone,
17935therefore, contains demonstrations, because it does not deduce its
17936cognition from conceptions, but from the construction of conceptions,
17937that is, from intuition, which can be given a priori in accordance with
17938conceptions. The method of algebra, in equations, from which the
17939correct answer is deduced by reduction, is a kind of construction--not
17940geometrical, but by symbols--in which all conceptions, especially those
17941of the relations of quantities, are represented in intuition by signs;
17942and thus the conclusions in that science are secured from errors by the
17943fact that every proof is submitted to ocular evidence. Philosophical
17944cognition does not possess this advantage, it being required to consider
17945the general always in abstracto (by means of conceptions), while
17946mathematics can always consider it in concreto (in an individual
17947intuition), and at the same time by means of a priori representation,
17948whereby all errors are rendered manifest to the senses. The
17949former--discursive proofs--ought to be termed acroamatic proofs,
17950rather than demonstrations, as only words are employed in them, while
17951demonstrations proper, as the term itself indicates, always require a
17952reference to the intuition of the object.
17953
17954It follows from all these considerations that it is not consonant with
17955the nature of philosophy, especially in the sphere of pure reason, to
17956employ the dogmatical method, and to adorn itself with the titles and
17957insignia of mathematical science. It does not belong to that order, and
17958can only hope for a fraternal union with that science. Its attempts at
17959mathematical evidence are vain pretensions, which can only keep it back
17960from its true aim, which is to detect the illusory procedure of reason
17961when transgressing its proper limits, and by fully explaining and
17962analysing our conceptions, to conduct us from the dim regions of
17963speculation to the clear region of modest self-knowledge. Reason must
17964not, therefore, in its transcendental endeavours, look forward with such
17965confidence, as if the path it is pursuing led straight to its aim,
17966nor reckon with such security upon its premisses, as to consider it
17967unnecessary to take a step back, or to keep a strict watch for errors,
17968which, overlooked in the principles, may be detected in the arguments
17969themselves--in which case it may be requisite either to determine these
17970principles with greater strictness, or to change them entirely.
17971
17972I divide all apodeictic propositions, whether demonstrable or
17973immediately certain, into dogmata and mathemata. A direct synthetical
17974proposition, based on conceptions, is a dogma; a proposition of the same
17975kind, based on the construction of conceptions, is a mathema. Analytical
17976judgements do not teach us any more about an object than what was
17977contained in the conception we had of it; because they do not extend our
17978cognition beyond our conception of an object, they merely elucidate the
17979conception. They cannot therefore be with propriety termed dogmas. Of
17980the two kinds of a priori synthetical propositions above mentioned, only
17981those which are employed in philosophy can, according to the general
17982mode of speech, bear this name; those of arithmetic or geometry would
17983not be rightly so denominated. Thus the customary mode of speaking
17984confirms the explanation given above, and the conclusion arrived at,
17985that only those judgements which are based upon conceptions, not on the
17986construction of conceptions, can be termed dogmatical.
17987
17988Thus, pure reason, in the sphere of speculation, does not contain a
17989single direct synthetical judgement based upon conceptions. By means
17990of ideas, it is, as we have shown, incapable of producing synthetical
17991judgements, which are objectively valid; by means of the conceptions of
17992the understanding, it establishes certain indubitable principles, not,
17993however, directly on the basis of conceptions, but only indirectly by
17994means of the relation of these conceptions to something of a purely
17995contingent nature, namely, possible experience. When experience is
17996presupposed, these principles are apodeictically certain, but in
17997themselves, and directly, they cannot even be cognized a priori. Thus
17998the given conceptions of cause and event will not be sufficient for
17999the demonstration of the proposition: Every event has a cause. For this
18000reason, it is not a dogma; although from another point of view, that of
18001experience, it is capable of being proved to demonstration. The proper
18002term for such a proposition is principle, and not theorem (although
18003it does require to be proved), because it possesses the remarkable
18004peculiarity of being the condition of the possibility of its own ground
18005of proof, that is, experience, and of forming a necessary presupposition
18006in all empirical observation.
18007
18008If then, in the speculative sphere of pure reason, no dogmata are to
18009be found; all dogmatical methods, whether borrowed from mathematics,
18010or invented by philosophical thinkers, are alike inappropriate and
18011inefficient. They only serve to conceal errors and fallacies, and to
18012deceive philosophy, whose duty it is to see that reason pursues a safe
18013and straight path. A philosophical method may, however, be systematical.
18014For our reason is, subjectively considered, itself a system, and, in
18015the sphere of mere conceptions, a system of investigation according to
18016principles of unity, the material being supplied by experience alone.
18017But this is not the proper place for discussing the peculiar method
18018of transcendental philosophy, as our present task is simply to examine
18019whether our faculties are capable of erecting an edifice on the basis
18020of pure reason, and how far they may proceed with the materials at their
18021command.
18022
18023
18024
18025SECTION II. The Discipline of Pure Reason in Polemics.
18026
18027Reason must be subject, in all its operations, to criticism, which
18028must always be permitted to exercise its functions without restraint;
18029otherwise its interests are imperilled and its influence obnoxious to
18030suspicion. There is nothing, however useful, however sacred it may be,
18031that can claim exemption from the searching examination of this supreme
18032tribunal, which has no respect of persons. The very existence of reason
18033depends upon this freedom; for the voice of reason is not that of
18034a dictatorial and despotic power, it is rather like the vote of the
18035citizens of a free state, every member of which must have the privilege
18036of giving free expression to his doubts, and possess even the right of
18037veto.
18038
18039But while reason can never decline to submit itself to the tribunal of
18040criticism, it has not always cause to dread the judgement of this court.
18041Pure reason, however, when engaged in the sphere of dogmatism, is not so
18042thoroughly conscious of a strict observance of its highest laws, as to
18043appear before a higher judicial reason with perfect confidence. On the
18044contrary, it must renounce its magnificent dogmatical pretensions in
18045philosophy.
18046
18047Very different is the case when it has to defend itself, not before a
18048judge, but against an equal. If dogmatical assertions are advanced on
18049the negative side, in opposition to those made by reason on the positive
18050side, its justification kat authrhopon is complete, although the proof
18051of its propositions is kat aletheian unsatisfactory.
18052
18053By the polemic of pure reason I mean the defence of its propositions
18054made by reason, in opposition to the dogmatical counter-propositions
18055advanced by other parties. The question here is not whether its own
18056statements may not also be false; it merely regards the fact that
18057reason proves that the opposite cannot be established with demonstrative
18058certainty, nor even asserted with a higher degree of probability. Reason
18059does not hold her possessions upon sufferance; for, although she cannot
18060show a perfectly satisfactory title to them, no one can prove that she
18061is not the rightful possessor.
18062
18063It is a melancholy reflection that reason, in its highest exercise,
18064falls into an antithetic; and that the supreme tribunal for the
18065settlement of differences should not be at union with itself. It is true
18066that we had to discuss the question of an apparent antithetic, but we
18067found that it was based upon a misconception. In conformity with the
18068common prejudice, phenomena were regarded as things in themselves, and
18069thus an absolute completeness in their synthesis was required in the one
18070mode or in the other (it was shown to be impossible in both); a demand
18071entirely out of place in regard to phenomena. There was, then, no
18072real self-contradiction of reason in the propositions: The series of
18073phenomena given in themselves has an absolutely first beginning; and:
18074This series is absolutely and in itself without beginning. The two
18075propositions are perfectly consistent with each other, because phenomena
18076as phenomena are in themselves nothing, and consequently the hypothesis
18077that they are things in themselves must lead to self-contradictory
18078inferences.
18079
18080But there are cases in which a similar misunderstanding cannot be
18081provided against, and the dispute must remain unsettled. Take, for
18082example, the theistic proposition: There is a Supreme Being; and on the
18083other hand, the atheistic counter-statement: There exists no Supreme
18084Being; or, in psychology: Everything that thinks possesses the attribute
18085of absolute and permanent unity, which is utterly different from the
18086transitory unity of material phenomena; and the counter-proposition: The
18087soul is not an immaterial unity, and its nature is transitory, like that
18088of phenomena. The objects of these questions contain no heterogeneous or
18089contradictory elements, for they relate to things in themselves, and not
18090to phenomena. There would arise, indeed, a real contradiction, if reason
18091came forward with a statement on the negative side of these questions
18092alone. As regards the criticism to which the grounds of proof on the
18093affirmative side must be subjected, it may be freely admitted, without
18094necessitating the surrender of the affirmative propositions, which have,
18095at least, the interest of reason in their favour--an advantage which the
18096opposite party cannot lay claim to.
18097
18098I cannot agree with the opinion of several admirable thinkers--Sulzer
18099among the rest--that, in spite of the weakness of the arguments hitherto
18100in use, we may hope, one day, to see sufficient demonstrations of the
18101two cardinal propositions of pure reason--the existence of a Supreme
18102Being, and the immortality of the soul. I am certain, on the contrary,
18103that this will never be the case. For on what ground can reason base
18104such synthetical propositions, which do not relate to the objects
18105of experience and their internal possibility? But it is also
18106demonstratively certain that no one will ever be able to maintain the
18107contrary with the least show of probability. For, as he can attempt such
18108a proof solely upon the basis of pure reason, he is bound to prove
18109that a Supreme Being, and a thinking subject in the character of a pure
18110intelligence, are impossible. But where will he find the knowledge which
18111can enable him to enounce synthetical judgements in regard to things
18112which transcend the region of experience? We may, therefore, rest
18113assured that the opposite never will be demonstrated. We need not, then,
18114have recourse to scholastic arguments; we may always admit the truth of
18115those propositions which are consistent with the speculative interests
18116of reason in the sphere of experience, and form, moreover, the only
18117means of uniting the speculative with the practical interest. Our
18118opponent, who must not be considered here as a critic solely, we can
18119be ready to meet with a non liquet which cannot fail to disconcert him;
18120while we cannot deny his right to a similar retort, as we have on our
18121side the advantage of the support of the subjective maxim of reason,
18122and can therefore look upon all his sophistical arguments with calm
18123indifference.
18124
18125From this point of view, there is properly no antithetic of pure reason.
18126For the only arena for such a struggle would be upon the field of
18127pure theology and psychology; but on this ground there can appear no
18128combatant whom we need to fear. Ridicule and boasting can be his
18129only weapons; and these may be laughed at, as mere child's play.
18130This consideration restores to Reason her courage; for what source
18131of confidence could be found, if she, whose vocation it is to destroy
18132error, were at variance with herself and without any reasonable hope of
18133ever reaching a state of permanent repose?
18134
18135Everything in nature is good for some purpose. Even poisons are
18136serviceable; they destroy the evil effects of other poisons generated
18137in our system, and must always find a place in every complete
18138pharmacopoeia. The objections raised against the fallacies and
18139sophistries of speculative reason, are objections given by the nature
18140of this reason itself, and must therefore have a destination and
18141purpose which can only be for the good of humanity. For what purpose has
18142Providence raised many objects, in which we have the deepest interest,
18143so far above us, that we vainly try to cognize them with certainty, and
18144our powers of mental vision are rather excited than satisfied by the
18145glimpses we may chance to seize? It is very doubtful whether it is for
18146our benefit to advance bold affirmations regarding subjects involved
18147in such obscurity; perhaps it would even be detrimental to our best
18148interests. But it is undoubtedly always beneficial to leave the
18149investigating, as well as the critical reason, in perfect freedom, and
18150permit it to take charge of its own interests, which are advanced as
18151much by its limitation, as by its extension of its views, and which
18152always suffer by the interference of foreign powers forcing it, against
18153its natural tendencies, to bend to certain preconceived designs.
18154
18155Allow your opponent to say what he thinks reasonable, and combat him
18156only with the weapons of reason. Have no anxiety for the practical
18157interests of humanity--these are never imperilled in a purely
18158speculative dispute. Such a dispute serves merely to disclose the
18159antinomy of reason, which, as it has its source in the nature of
18160reason, ought to be thoroughly investigated. Reason is benefited by the
18161examination of a subject on both sides, and its judgements are corrected
18162by being limited. It is not the matter that may give occasion to
18163dispute, but the manner. For it is perfectly permissible to employ,
18164in the presence of reason, the language of a firmly rooted faith, even
18165after we have been obliged to renounce all pretensions to knowledge.
18166
18167If we were to ask the dispassionate David Hume--a philosopher endowed,
18168in a degree that few are, with a well-balanced judgement: What motive
18169induced you to spend so much labour and thought in undermining the
18170consoling and beneficial persuasion that reason is capable of assuring
18171us of the existence, and presenting us with a determinate conception
18172of a Supreme Being?--his answer would be: Nothing but the desire of
18173teaching reason to know its own powers better, and, at the same time, a
18174dislike of the procedure by which that faculty was compelled to support
18175foregone conclusions, and prevented from confessing the internal
18176weaknesses which it cannot but feel when it enters upon a rigid
18177self-examination. If, on the other hand, we were to ask Priestley--a
18178philosopher who had no taste for transcendental speculation, but was
18179entirely devoted to the principles of empiricism--what his motives were
18180for overturning those two main pillars of religion--the doctrines of
18181the freedom of the will and the immortality of the soul (in his view
18182the hope of a future life is but the expectation of the miracle of
18183resurrection)--this philosopher, himself a zealous and pious teacher of
18184religion, could give no other answer than this: I acted in the interest
18185of reason, which always suffers, when certain objects are explained
18186and judged by a reference to other supposed laws than those of material
18187nature--the only laws which we know in a determinate manner. It would be
18188unfair to decry the latter philosopher, who endeavoured to harmonize his
18189paradoxical opinions with the interests of religion, and to undervalue
18190an honest and reflecting man, because he finds himself at a loss the
18191moment he has left the field of natural science. The same grace must be
18192accorded to Hume, a man not less well-disposed, and quite as blameless
18193in his moral character, and who pushed his abstract speculations to an
18194extreme length, because, as he rightly believed, the object of them lies
18195entirely beyond the bounds of natural science, and within the sphere of
18196pure ideas.
18197
18198What is to be done to provide against the danger which seems in the
18199present case to menace the best interests of humanity? The course to be
18200pursued in reference to this subject is a perfectly plain and natural
18201one. Let each thinker pursue his own path; if he shows talent, if he
18202gives evidence of profound thought, in one word, if he shows that he
18203possesses the power of reasoning--reason is always the gainer. If you
18204have recourse to other means, if you attempt to coerce reason, if you
18205raise the cry of treason to humanity, if you excite the feelings of
18206the crowd, which can neither understand nor sympathize with such subtle
18207speculations--you will only make yourselves ridiculous. For the question
18208does not concern the advantage or disadvantage which we are expected
18209to reap from such inquiries; the question is merely how far reason can
18210advance in the field of speculation, apart from all kinds of interest,
18211and whether we may depend upon the exertions of speculative reason, or
18212must renounce all reliance on it. Instead of joining the combatants,
18213it is your part to be a tranquil spectator of the struggle--a laborious
18214struggle for the parties engaged, but attended, in its progress as
18215well as in its result, with the most advantageous consequences for
18216the interests of thought and knowledge. It is absurd to expect to be
18217enlightened by Reason, and at the same time to prescribe to her what
18218side of the question she must adopt. Moreover, reason is sufficiently
18219held in check by its own power, the limits imposed on it by its own
18220nature are sufficient; it is unnecessary for you to place over it
18221additional guards, as if its power were dangerous to the constitution of
18222the intellectual state. In the dialectic of reason there is no victory
18223gained which need in the least disturb your tranquility.
18224
18225The strife of dialectic is a necessity of reason, and we cannot but wish
18226that it had been conducted long ere this with that perfect freedom which
18227ought to be its essential condition. In this case, we should have had at
18228an earlier period a matured and profound criticism, which must have
18229put an end to all dialectical disputes, by exposing the illusions and
18230prejudices in which they originated.
18231
18232There is in human nature an unworthy propensity--a propensity which,
18233like everything that springs from nature, must in its final purpose be
18234conducive to the good of humanity--to conceal our real sentiments, and
18235to give expression only to certain received opinions, which are regarded
18236as at once safe and promotive of the common good. It is true, this
18237tendency, not only to conceal our real sentiments, but to profess those
18238which may gain us favour in the eyes of society, has not only civilized,
18239but, in a certain measure, moralized us; as no one can break through the
18240outward covering of respectability, honour, and morality, and thus
18241the seemingly-good examples which we which we see around us form an
18242excellent school for moral improvement, so long as our belief in
18243their genuineness remains unshaken. But this disposition to represent
18244ourselves as better than we are, and to utter opinions which are not
18245our own, can be nothing more than a kind of provisionary arrangement
18246of nature to lead us from the rudeness of an uncivilized state, and to
18247teach us how to assume at least the appearance and manner of the good we
18248see. But when true principles have been developed, and have obtained a
18249sure foundation in our habit of thought, this conventionalism must
18250be attacked with earnest vigour, otherwise it corrupts the heart, and
18251checks the growth of good dispositions with the mischievous weed of air
18252appearances.
18253
18254I am sorry to remark the same tendency to misrepresentation and
18255hypocrisy in the sphere of speculative discussion, where there is less
18256temptation to restrain the free expression of thought. For what can be
18257more prejudicial to the interests of intelligence than to falsify our
18258real sentiments, to conceal the doubts which we feel in regard to our
18259statements, or to maintain the validity of grounds of proof which we
18260well know to be insufficient? So long as mere personal vanity is the
18261source of these unworthy artifices--and this is generally the case
18262in speculative discussions, which are mostly destitute of practical
18263interest, and are incapable of complete demonstration--the vanity of
18264the opposite party exaggerates as much on the other side; and thus the
18265result is the same, although it is not brought about so soon as if the
18266dispute had been conducted in a sincere and upright spirit. But
18267where the mass entertains the notion that the aim of certain subtle
18268speculators is nothing less than to shake the very foundations of public
18269welfare and morality--it seems not only prudent, but even praise worthy,
18270to maintain the good cause by illusory arguments, rather than to give to
18271our supposed opponents the advantage of lowering our declarations to the
18272moderate tone of a merely practical conviction, and of compelling us to
18273confess our inability to attain to apodeictic certainty in speculative
18274subjects. But we ought to reflect that there is nothing, in the
18275world more fatal to the maintenance of a good cause than deceit,
18276misrepresentation, and falsehood. That the strictest laws of honesty
18277should be observed in the discussion of a purely speculative subject is
18278the least requirement that can be made. If we could reckon with security
18279even upon so little, the conflict of speculative reason regarding the
18280important questions of God, immortality, and freedom, would have been
18281either decided long ago, or would very soon be brought to a conclusion.
18282But, in general, the uprightness of the defence stands in an inverse
18283ratio to the goodness of the cause; and perhaps more honesty and
18284fairness are shown by those who deny than by those who uphold these
18285doctrines.
18286
18287I shall persuade myself, then, that I have readers who do not wish
18288to see a righteous cause defended by unfair arguments. Such will now
18289recognize the fact that, according to the principles of this Critique,
18290if we consider not what is, but what ought to be the case, there can be
18291really no polemic of pure reason. For how can two persons dispute about
18292a thing, the reality of which neither can present in actual or even in
18293possible experience? Each adopts the plan of meditating on his idea for
18294the purpose of drawing from the idea, if he can, what is more than the
18295idea, that is, the reality of the object which it indicates. How shall
18296they settle the dispute, since neither is able to make his assertions
18297directly comprehensible and certain, but must restrict himself to
18298attacking and confuting those of his opponent? All statements enounced
18299by pure reason transcend the conditions of possible experience, beyond
18300the sphere of which we can discover no criterion of truth, while
18301they are at the same time framed in accordance with the laws of the
18302understanding, which are applicable only to experience; and thus it is
18303the fate of all such speculative discussions that while the one party
18304attacks the weaker side of his opponent, he infallibly lays open his own
18305weaknesses.
18306
18307The critique of pure reason may be regarded as the highest tribunal
18308for all speculative disputes; for it is not involved in these disputes,
18309which have an immediate relation to certain objects and not to the laws
18310of the mind, but is instituted for the purpose of determining the rights
18311and limits of reason.
18312
18313Without the control of criticism, reason is, as it were, in a state
18314of nature, and can only establish its claims and assertions by war.
18315Criticism, on the contrary, deciding all questions according to the
18316fundamental laws of its own institution, secures to us the peace of
18317law and order, and enables us to discuss all differences in the more
18318tranquil manner of a legal process. In the former case, disputes are
18319ended by victory, which both sides may claim and which is followed by a
18320hollow armistice; in the latter, by a sentence, which, as it strikes
18321at the root of all speculative differences, ensures to all concerned a
18322lasting peace. The endless disputes of a dogmatizing reason compel us
18323to look for some mode of arriving at a settled decision by a critical
18324investigation of reason itself; just as Hobbes maintains that the state
18325of nature is a state of injustice and violence, and that we must leave
18326it and submit ourselves to the constraint of law, which indeed limits
18327individual freedom, but only that it may consist with the freedom of
18328others and with the common good of all.
18329
18330This freedom will, among other things, permit of our openly stating the
18331difficulties and doubts which we are ourselves unable to solve, without
18332being decried on that account as turbulent and dangerous citizens.
18333This privilege forms part of the native rights of human reason, which
18334recognizes no other judge than the universal reason of humanity; and
18335as this reason is the source of all progress and improvement, such a
18336privilege is to be held sacred and inviolable. It is unwise, moreover,
18337to denounce as dangerous any bold assertions against, or rash attacks
18338upon, an opinion which is held by the largest and most moral class of
18339the community; for that would be giving them an importance which they
18340do not deserve. When I hear that the freedom of the will, the hope of
18341a future life, and the existence of God have been overthrown by the
18342arguments of some able writer, I feel a strong desire to read his
18343book; for I expect that he will add to my knowledge and impart greater
18344clearness and distinctness to my views by the argumentative power shown
18345in his writings. But I am perfectly certain, even before I have opened
18346the book, that he has not succeeded in a single point, not because
18347I believe I am in possession of irrefutable demonstrations of these
18348important propositions, but because this transcendental critique, which
18349has disclosed to me the power and the limits of pure reason, has fully
18350convinced me that, as it is insufficient to establish the affirmative,
18351it is as powerless, and even more so, to assure us of the truth of
18352the negative answer to these questions. From what source does this
18353free-thinker derive his knowledge that there is, for example, no Supreme
18354Being? This proposition lies out of the field of possible experience,
18355and, therefore, beyond the limits of human cognition. But I would not
18356read at, all the answer which the dogmatical maintainer of the good
18357cause makes to his opponent, because I know well beforehand, that he
18358will merely attack the fallacious grounds of his adversary, without
18359being able to establish his own assertions. Besides, a new illusory
18360argument, in the construction of which talent and acuteness are shown,
18361is suggestive of new ideas and new trains of reasoning, and in this
18362respect the old and everyday sophistries are quite useless. Again,
18363the dogmatical opponent of religion gives employment to criticism,
18364and enables us to test and correct its principles, while there is no
18365occasion for anxiety in regard to the influence and results of his
18366reasoning.
18367
18368But, it will be said, must we not warn the youth entrusted to academical
18369care against such writings, must we not preserve them from the knowledge
18370of these dangerous assertions, until their judgement is ripened, or
18371rather until the doctrines which we wish to inculcate are so firmly
18372rooted in their minds as to withstand all attempts at instilling the
18373contrary dogmas, from whatever quarter they may come?
18374
18375If we are to confine ourselves to the dogmatical procedure in the
18376sphere of pure reason, and find ourselves unable to settle such
18377disputes otherwise than by becoming a party in them, and setting
18378counter-assertions against the statements advanced by our opponents,
18379there is certainly no plan more advisable for the moment, but, at the
18380same time, none more absurd and inefficient for the future, than this
18381retaining of the youthful mind under guardianship for a time, and thus
18382preserving it--for so long at least--from seduction into error. But
18383when, at a later period, either curiosity, or the prevalent fashion
18384of thought places such writings in their hands, will the so-called
18385convictions of their youth stand firm? The young thinker, who has in his
18386armoury none but dogmatical weapons with which to resist the attacks of
18387his opponent, and who cannot detect the latent dialectic which lies in
18388his own opinions as well as in those of the opposite party, sees the
18389advance of illusory arguments and grounds of proof which have the
18390advantage of novelty, against as illusory grounds of proof destitute
18391of this advantage, and which, perhaps, excite the suspicion that the
18392natural credulity of his youth has been abused by his instructors. He
18393thinks he can find no better means of showing that he has out grown the
18394discipline of his minority than by despising those well-meant warnings,
18395and, knowing no system of thought but that of dogmatism, he drinks deep
18396draughts of the poison that is to sap the principles in which his early
18397years were trained.
18398
18399Exactly the opposite of the system here recommended ought to be pursued
18400in academical instruction. This can only be effected, however, by a
18401thorough training in the critical investigation of pure reason. For, in
18402order to bring the principles of this critique into exercise as soon as
18403possible, and to demonstrate their perfect even in the presence of the
18404highest degree of dialectical illusion, the student ought to examine the
18405assertions made on both sides of speculative questions step by step, and
18406to test them by these principles. It cannot be a difficult task for him
18407to show the fallacies inherent in these propositions, and thus he begins
18408early to feel his own power of securing himself against the influence of
18409such sophistical arguments, which must finally lose, for him, all their
18410illusory power. And, although the same blows which overturn the edifice
18411of his opponent are as fatal to his own speculative structures, if such
18412he has wished to rear; he need not feel any sorrow in regard to this
18413seeming misfortune, as he has now before him a fair prospect into the
18414practical region in which he may reasonably hope to find a more secure
18415foundation for a rational system.
18416
18417There is, accordingly, no proper polemic in the sphere of pure reason.
18418Both parties beat the air and fight with their own shadows, as they
18419pass beyond the limits of nature, and can find no tangible point
18420of attack--no firm footing for their dogmatical conflict. Fight as
18421vigorously as they may, the shadows which they hew down, immediately
18422start up again, like the heroes in Walhalla, and renew the bloodless and
18423unceasing contest.
18424
18425But neither can we admit that there is any proper sceptical employment
18426of pure reason, such as might be based upon the principle of neutrality
18427in all speculative disputes. To excite reason against itself, to place
18428weapons in the hands of the party on the one side as well as in those of
18429the other, and to remain an undisturbed and sarcastic spectator of the
18430fierce struggle that ensues, seems, from the dogmatical point of view,
18431to be a part fitting only a malevolent disposition. But, when the
18432sophist evidences an invincible obstinacy and blindness, and a pride
18433which no criticism can moderate, there is no other practicable course
18434than to oppose to this pride and obstinacy similar feelings and
18435pretensions on the other side, equally well or ill founded, so that
18436reason, staggered by the reflections thus forced upon it, finds it
18437necessary to moderate its confidence in such pretensions and to listen
18438to the advice of criticism. But we cannot stop at these doubts, much
18439less regard the conviction of our ignorance, not only as a cure for the
18440conceit natural to dogmatism, but as the settlement of the disputes in
18441which reason is involved with itself. On the contrary, scepticism is
18442merely a means of awakening reason from its dogmatic dreams and exciting
18443it to a more careful investigation into its own powers and pretensions.
18444But, as scepticism appears to be the shortest road to a permanent peace
18445in the domain of philosophy, and as it is the track pursued by the
18446many who aim at giving a philosophical colouring to their contemptuous
18447dislike of all inquiries of this kind, I think it necessary to present
18448to my readers this mode of thought in its true light.
18449
18450
18451
18452Scepticism not a Permanent State for Human Reason.
18453
18454The consciousness of ignorance--unless this ignorance is recognized to
18455be absolutely necessary ought, instead of forming the conclusion of
18456my inquiries, to be the strongest motive to the pursuit of them. All
18457ignorance is either ignorance of things or of the limits of knowledge.
18458If my ignorance is accidental and not necessary, it must incite me, in
18459the first case, to a dogmatical inquiry regarding the objects of which I
18460am ignorant; in the second, to a critical investigation into the bounds
18461of all possible knowledge. But that my ignorance is absolutely necessary
18462and unavoidable, and that it consequently absolves from the duty of all
18463further investigation, is a fact which cannot be made out upon empirical
18464grounds--from observation--but upon critical grounds alone, that is, by
18465a thoroughgoing investigation into the primary sources of cognition. It
18466follows that the determination of the bounds of reason can be made only
18467on a priori grounds; while the empirical limitation of reason, which
18468is merely an indeterminate cognition of an ignorance that can never be
18469completely removed, can take place only a posteriori. In other words,
18470our empirical knowledge is limited by that which yet remains for us to
18471know. The former cognition of our ignorance, which is possible only on a
18472rational basis, is a science; the latter is merely a perception, and we
18473cannot say how far the inferences drawn from it may extend. If I regard
18474the earth, as it really appears to my senses, as a flat surface, I am
18475ignorant how far this surface extends. But experience teaches me that,
18476how far soever I go, I always see before me a space in which I can
18477proceed farther; and thus I know the limits--merely visual--of my actual
18478knowledge of the earth, although I am ignorant of the limits of the
18479earth itself. But if I have got so far as to know that the earth is a
18480sphere, and that its surface is spherical, I can cognize a priori and
18481determine upon principles, from my knowledge of a small part of this
18482surface--say to the extent of a degree--the diameter and circumference
18483of the earth; and although I am ignorant of the objects which this
18484surface contains, I have a perfect knowledge of its limits and extent.
18485
18486The sum of all the possible objects of our cognition seems to us to be
18487a level surface, with an apparent horizon--that which forms the limit
18488of its extent, and which has been termed by us the idea of unconditioned
18489totality. To reach this limit by empirical means is impossible, and all
18490attempts to determine it a priori according to a principle, are alike in
18491vain. But all the questions raised by pure reason relate to that which
18492lies beyond this horizon, or, at least, in its boundary line.
18493
18494The celebrated David Hume was one of those geographers of human reason
18495who believe that they have given a sufficient answer to all such
18496questions by declaring them to lie beyond the horizon of our
18497knowledge--a horizon which, however, Hume was unable to determine. His
18498attention especially was directed to the principle of causality; and he
18499remarked with perfect justice that the truth of this principle, and even
18500the objective validity of the conception of a cause, was not commonly
18501based upon clear insight, that is, upon a priori cognition. Hence
18502he concluded that this law does not derive its authority from its
18503universality and necessity, but merely from its general applicability
18504in the course of experience, and a kind of subjective necessity
18505thence arising, which he termed habit. From the inability of reason to
18506establish this principle as a necessary law for the acquisition of all
18507experience, he inferred the nullity of all the attempts of reason to
18508pass the region of the empirical.
18509
18510This procedure of subjecting the facta of reason to examination, and,
18511if necessary, to disapproval, may be termed the censura of reason. This
18512censura must inevitably lead us to doubts regarding all transcendent
18513employment of principles. But this is only the second step in our
18514inquiry. The first step in regard to the subjects of pure reason, and
18515which marks the infancy of that faculty, is that of dogmatism. The
18516second, which we have just mentioned, is that of scepticism, and it
18517gives evidence that our judgement has been improved by experience. But
18518a third step is necessary--indicative of the maturity and manhood of the
18519judgement, which now lays a firm foundation upon universal and necessary
18520principles. This is the period of criticism, in which we do not examine
18521the facta of reason, but reason itself, in the whole extent of its
18522powers, and in regard to its capability of a priori cognition; and thus
18523we determine not merely the empirical and ever-shifting bounds of our
18524knowledge, but its necessary and eternal limits. We demonstrate from
18525indubitable principles, not merely our ignorance in respect to this
18526or that subject, but in regard to all possible questions of a certain
18527class. Thus scepticism is a resting place for reason, in which it may
18528reflect on its dogmatical wanderings and gain some knowledge of the
18529region in which it happens to be, that it may pursue its way with
18530greater certainty; but it cannot be its permanent dwelling-place. It
18531must take up its abode only in the region of complete certitude, whether
18532this relates to the cognition of objects themselves, or to the limits
18533which bound all our cognition.
18534
18535Reason is not to be considered as an indefinitely extended plane, of the
18536bounds of which we have only a general knowledge; it ought rather to
18537be compared to a sphere, the radius of which may be found from the
18538curvature of its surface--that is, the nature of a priori synthetical
18539propositions--and, consequently, its circumference and extent. Beyond
18540the sphere of experience there are no objects which it can cognize; nay,
18541even questions regarding such supposititious objects relate only to the
18542subjective principles of a complete determination of the relations
18543which exist between the understanding-conceptions which lie within this
18544sphere.
18545
18546We are actually in possession of a priori synthetical cognitions, as is
18547proved by the existence of the principles of the understanding, which
18548anticipate experience. If any one cannot comprehend the possibility
18549of these principles, he may have some reason to doubt whether they
18550are really a priori; but he cannot on this account declare them to be
18551impossible, and affirm the nullity of the steps which reason may have
18552taken under their guidance. He can only say: If we perceived their
18553origin and their authenticity, we should be able to determine the
18554extent and limits of reason; but, till we can do this, all propositions
18555regarding the latter are mere random assertions. In this view, the
18556doubt respecting all dogmatical philosophy, which proceeds without the
18557guidance of criticism, is well grounded; but we cannot therefore deny
18558to reason the ability to construct a sound philosophy, when the way has
18559been prepared by a thorough critical investigation. All the conceptions
18560produced, and all the questions raised, by pure reason, do not lie in
18561the sphere of experience, but in that of reason itself, and hence they
18562must be solved, and shown to be either valid or inadmissible, by that
18563faculty. We have no right to decline the solution of such problems, on
18564the ground that the solution can be discovered only from the nature of
18565things, and under pretence of the limitation of human faculties, for
18566reason is the sole creator of all these ideas, and is therefore bound
18567either to establish their validity or to expose their illusory nature.
18568
18569The polemic of scepticism is properly directed against the dogmatist,
18570who erects a system of philosophy without having examined the
18571fundamental objective principles on which it is based, for the purpose
18572of evidencing the futility of his designs, and thus bringing him to a
18573knowledge of his own powers. But, in itself, scepticism does not give
18574us any certain information in regard to the bounds of our knowledge. All
18575unsuccessful dogmatical attempts of reason are facia, which it is always
18576useful to submit to the censure of the sceptic. But this cannot help
18577us to any decision regarding the expectations which reason cherishes of
18578better success in future endeavours; the investigations of scepticism
18579cannot, therefore, settle the dispute regarding the rights and powers of
18580human reason.
18581
18582Hume is perhaps the ablest and most ingenious of all sceptical
18583philosophers, and his writings have, undoubtedly, exerted the most
18584powerful influence in awakening reason to a thorough investigation into
18585its own powers. It will, therefore, well repay our labours to consider
18586for a little the course of reasoning which he followed and the errors
18587into which he strayed, although setting out on the path of truth and
18588certitude.
18589
18590Hume was probably aware, although he never clearly developed the notion,
18591that we proceed in judgements of a certain class beyond our conception
18592if the object. I have termed this kind of judgement synthetical. As
18593regard the manner in which I pass beyond my conception by the aid
18594of experience, no doubts can be entertained. Experience is itself a
18595synthesis of perceptions; and it employs perceptions to increment the
18596conception, which I obtain by means of another perception. But we feel
18597persuaded that we are able to proceed beyond a conception, and to extend
18598our cognition a priori. We attempt this in two ways--either, through the
18599pure understanding, in relation to that which may become an object of
18600experience, or, through pure reason, in relation to such properties of
18601things, or of the existence of things, as can never be presented in any
18602experience. This sceptical philosopher did not distinguish these
18603two kinds of judgements, as he ought to have done, but regarded this
18604augmentation of conceptions, and, if we may so express ourselves, the
18605spontaneous generation of understanding and reason, independently of the
18606impregnation of experience, as altogether impossible. The so-called a
18607priori principles of these faculties he consequently held to be invalid
18608and imaginary, and regarded them as nothing but subjective habits of
18609thought originating in experience, and therefore purely empirical
18610and contingent rules, to which we attribute a spurious necessity and
18611universality. In support of this strange assertion, he referred us to
18612the generally acknowledged principle of the relation between cause and
18613effect. No faculty of the mind can conduct us from the conception of a
18614thing to the existence of something else; and hence he believed he could
18615infer that, without experience, we possess no source from which we can
18616augment a conception, and no ground sufficient to justify us in framing
18617a judgement that is to extend our cognition a priori. That the light of
18618the sun, which shines upon a piece of wax, at the same time melts it,
18619while it hardens clay, no power of the understanding could infer from
18620the conceptions which we previously possessed of these substances;
18621much less is there any a priori law that could conduct us to such a
18622conclusion, which experience alone can certify. On the other hand, we
18623have seen in our discussion of transcendental logic, that, although we
18624can never proceed immediately beyond the content of the conception which
18625is given us, we can always cognize completely a priori--in relation,
18626however, to a third term, namely, possible experience--the law of its
18627connection with other things. For example, if I observe that a piece of
18628wax melts, I can cognize a priori that there must have been something
18629(the sun's heat) preceding, which this law; although, without the aid
18630of experience, I could not cognize a priori and in a determinate manner
18631either the cause from the effect, or the effect from the cause.
18632Hume was, therefore, wrong in inferring, from the contingency of the
18633determination according to law, the contingency of the law itself; and
18634the passing beyond the conception of a thing to possible experience
18635(which is an a priori proceeding, constituting the objective reality of
18636the conception), he confounded with our synthesis of objects in actual
18637experience, which is always, of course, empirical. Thus, too,
18638he regarded the principle of affinity, which has its seat in the
18639understanding and indicates a necessary connection, as a mere rule of
18640association, lying in the imitative faculty of imagination, which can
18641present only contingent, and not objective connections.
18642
18643The sceptical errors of this remarkably acute thinker arose principally
18644from a defect, which was common to him with the dogmatists, namely, that
18645he had never made a systematic review of all the different kinds of
18646a priori synthesis performed by the understanding. Had he done so, he
18647would have found, to take one example among many, that the principle of
18648permanence was of this character, and that it, as well as the principle
18649of causality, anticipates experience. In this way he might have been
18650able to describe the determinate limits of the a priori operations of
18651understanding and reason. But he merely declared the understanding to be
18652limited, instead of showing what its limits were; he created a
18653general mistrust in the power of our faculties, without giving us any
18654determinate knowledge of the bounds of our necessary and unavoidable
18655ignorance; he examined and condemned some of the principles of
18656the understanding, without investigating all its powers with the
18657completeness necessary to criticism. He denies, with truth, certain
18658powers to the understanding, but he goes further, and declares it to be
18659utterly inadequate to the a priori extension of knowledge, although he
18660has not fully examined all the powers which reside in the faculty; and
18661thus the fate which always overtakes scepticism meets him too. That is
18662to say, his own declarations are doubted, for his objections were based
18663upon facta, which are contingent, and not upon principles, which can
18664alone demonstrate the necessary invalidity of all dogmatical assertions.
18665
18666As Hume makes no distinction between the well-grounded claims of the
18667understanding and the dialectical pretensions of reason, against which,
18668however, his attacks are mainly directed, reason does not feel itself
18669shut out from all attempts at the extension of a priori cognition, and
18670hence it refuses, in spite of a few checks in this or that quarter, to
18671relinquish such efforts. For one naturally arms oneself to resist an
18672attack, and becomes more obstinate in the resolve to establish the
18673claims he has advanced. But a complete review of the powers of reason,
18674and the conviction thence arising that we are in possession of a limited
18675field of action, while we must admit the vanity of higher claims, puts
18676an end to all doubt and dispute, and induces reason to rest satisfied
18677with the undisturbed possession of its limited domain.
18678
18679To the uncritical dogmatist, who has not surveyed the sphere of his
18680understanding, nor determined, in accordance with principles, the limits
18681of possible cognition, who, consequently, is ignorant of his own powers,
18682and believes he will discover them by the attempts he makes in the field
18683of cognition, these attacks of scepticism are not only dangerous, but
18684destructive. For if there is one proposition in his chain of reasoning
18685which be he cannot prove, or the fallacy in which he cannot evolve in
18686accordance with a principle, suspicion falls on all his statements,
18687however plausible they may appear.
18688
18689And thus scepticism, the bane of dogmatical philosophy, conducts us to
18690a sound investigation into the understanding and the reason. When we are
18691thus far advanced, we need fear no further attacks; for the limits of
18692our domain are clearly marked out, and we can make no claims nor become
18693involved in any disputes regarding the region that lies beyond these
18694limits. Thus the sceptical procedure in philosophy does not present any
18695solution of the problems of reason, but it forms an excellent exercise
18696for its powers, awakening its circumspection, and indicating the
18697means whereby it may most fully establish its claims to its legitimate
18698possessions.
18699
18700
18701
18702SECTION III. The Discipline of Pure Reason in Hypothesis.
18703
18704This critique of reason has now taught us that all its efforts to extend
18705the bounds of knowledge, by means of pure speculation, are utterly
18706fruitless. So much the wider field, it may appear, lies open to
18707hypothesis; as, where we cannot know with certainty, we are at liberty
18708to make guesses and to form suppositions.
18709
18710Imagination may be allowed, under the strict surveillance of reason,
18711to invent suppositions; but, these must be based on something that is
18712perfectly certain--and that is the possibility of the object. If we
18713are well assured upon this point, it is allowable to have recourse to
18714supposition in regard to the reality of the object; but this supposition
18715must, unless it is utterly groundless, be connected, as its ground of
18716explanation, with that which is really given and absolutely certain.
18717Such a supposition is termed a hypothesis.
18718
18719It is beyond our power to form the least conception a priori of the
18720possibility of dynamical connection in phenomena; and the category
18721of the pure understanding will not enable us to excogitate any such
18722connection, but merely helps us to understand it, when we meet with
18723it in experience. For this reason we cannot, in accordance with the
18724categories, imagine or invent any object or any property of an object
18725not given, or that may not be given in experience, and employ it in a
18726hypothesis; otherwise, we should be basing our chain of reasoning upon
18727mere chimerical fancies, and not upon conceptions of things. Thus, we
18728have no right to assume the existence of new powers, not existing in
18729nature--for example, an understanding with a non-sensuous intuition,
18730a force of attraction without contact, or some new kind of substances
18731occupying space, and yet without the property of impenetrability--and,
18732consequently, we cannot assume that there is any other kind of community
18733among substances than that observable in experience, any kind of
18734presence than that in space, or any kind of duration than that in time.
18735In one word, the conditions of possible experience are for reason the
18736only conditions of the possibility of things; reason cannot venture
18737to form, independently of these conditions, any conceptions of things,
18738because such conceptions, although not self-contradictory, are without
18739object and without application.
18740
18741The conceptions of reason are, as we have already shown, mere ideas, and
18742do not relate to any object in any kind of experience. At the same time,
18743they do not indicate imaginary or possible objects. They are purely
18744problematical in their nature and, as aids to the heuristic exercise
18745of the faculties, form the basis of the regulative principles for the
18746systematic employment of the understanding in the field of experience.
18747If we leave this ground of experience, they become mere fictions of
18748thought, the possibility of which is quite indemonstrable; and they
18749cannot, consequently, be employed as hypotheses in the explanation of
18750real phenomena. It is quite admissible to cogitate the soul as simple,
18751for the purpose of enabling ourselves to employ the idea of a perfect
18752and necessary unity of all the faculties of the mind as the principle
18753of all our inquiries into its internal phenomena, although we cannot
18754cognize this unity in concreto. But to assume that the soul is a simple
18755substance (a transcendental conception) would be enouncing a proposition
18756which is not only indemonstrable--as many physical hypotheses are--but
18757a proposition which is purely arbitrary, and in the highest degree rash.
18758The simple is never presented in experience; and, if by substance is
18759here meant the permanent object of sensuous intuition, the possibility
18760of a simple phenomenon is perfectly inconceivable. Reason affords no
18761good grounds for admitting the existence of intelligible beings, or
18762of intelligible properties of sensuous things, although--as we have no
18763conception either of their possibility or of their impossibility--it
18764will always be out of our power to affirm dogmatically that they do not
18765exist. In the explanation of given phenomena, no other things and no
18766other grounds of explanation can be employed than those which stand
18767in connection with the given phenomena according to the known laws of
18768experience. A transcendental hypothesis, in which a mere idea of reason
18769is employed to explain the phenomena of nature, would not give us any
18770better insight into a phenomenon, as we should be trying to explain what
18771we do not sufficiently understand from known empirical principles, by
18772what we do not understand at all. The principles of such a hypothesis
18773might conduce to the satisfaction of reason, but it would not assist
18774the understanding in its application to objects. Order and conformity to
18775aims in the sphere of nature must be themselves explained upon natural
18776grounds and according to natural laws; and the wildest hypotheses, if
18777they are only physical, are here more admissible than a hyperphysical
18778hypothesis, such as that of a divine author. For such a hypothesis would
18779introduce the principle of ignava ratio, which requires us to give
18780up the search for causes that might be discovered in the course of
18781experience and to rest satisfied with a mere idea. As regards the
18782absolute totality of the grounds of explanation in the series of these
18783causes, this can be no hindrance to the understanding in the case of
18784phenomena; because, as they are to us nothing more than phenomena, we
18785have no right to look for anything like completeness in the synthesis of
18786the series of their conditions.
18787
18788Transcendental hypotheses are therefore inadmissible; and we cannot
18789use the liberty of employing, in the absence of physical, hyperphysical
18790grounds of explanation. And this for two reasons; first, because such
18791hypothesis do not advance reason, but rather stop it in its progress;
18792secondly, because this licence would render fruitless all its exertions
18793in its own proper sphere, which is that of experience. For, when the
18794explanation of natural phenomena happens to be difficult, we have
18795constantly at hand a transcendental ground of explanation, which lifts
18796us above the necessity of investigating nature; and our inquiries are
18797brought to a close, not because we have obtained all the
18798requisite knowledge, but because we abut upon a principle which is
18799incomprehensible and which, indeed, is so far back in the track of
18800thought as to contain the conception of the absolutely primal being.
18801
18802The next requisite for the admissibility of a hypothesis is its
18803sufficiency. That is, it must determine a priori the consequences
18804which are given in experience and which are supposed to follow from the
18805hypothesis itself. If we require to employ auxiliary hypotheses, the
18806suspicion naturally arises that they are mere fictions; because the
18807necessity for each of them requires the same justification as in the
18808case of the original hypothesis, and thus their testimony is invalid.
18809If we suppose the existence of an infinitely perfect cause, we possess
18810sufficient grounds for the explanation of the conformity to aims, the
18811order and the greatness which we observe in the universe; but we
18812find ourselves obliged, when we observe the evil in the world and the
18813exceptions to these laws, to employ new hypothesis in support of the
18814original one. We employ the idea of the simple nature of the human soul
18815as the foundation of all the theories we may form of its phenomena; but
18816when we meet with difficulties in our way, when we observe in the soul
18817phenomena similar to the changes which take place in matter, we require
18818to call in new auxiliary hypotheses. These may, indeed, not be false,
18819but we do not know them to be true, because the only witness to their
18820certitude is the hypothesis which they themselves have been called in to
18821explain.
18822
18823We are not discussing the above-mentioned assertions regarding the
18824immaterial unity of the soul and the existence of a Supreme Being as
18825dogmata, which certain philosophers profess to demonstrate a priori, but
18826purely as hypotheses. In the former case, the dogmatist must take care
18827that his arguments possess the apodeictic certainty of a demonstration.
18828For the assertion that the reality of such ideas is probable is as
18829absurd as a proof of the probability of a proposition in geometry. Pure
18830abstract reason, apart from all experience, can either cognize nothing
18831at all; and hence the judgements it enounces are never mere opinions,
18832they are either apodeictic certainties, or declarations that nothing can
18833be known on the subject. Opinions and probable judgements on the nature
18834of things can only be employed to explain given phenomena, or they may
18835relate to the effect, in accordance with empirical laws, of an actually
18836existing cause. In other words, we must restrict the sphere of opinion
18837to the world of experience and nature. Beyond this region opinion is
18838mere invention; unless we are groping about for the truth on a path not
18839yet fully known, and have some hopes of stumbling upon it by chance.
18840
18841But, although hypotheses are inadmissible in answers to the questions
18842of pure speculative reason, they may be employed in the defence of these
18843answers. That is to say, hypotheses are admissible in polemic, but
18844not in the sphere of dogmatism. By the defence of statements of this
18845character, I do not mean an attempt at discovering new grounds for their
18846support, but merely the refutation of the arguments of opponents. All a
18847priori synthetical propositions possess the peculiarity that, although
18848the philosopher who maintains the reality of the ideas contained in the
18849proposition is not in possession of sufficient knowledge to establish
18850the certainty of his statements, his opponent is as little able to prove
18851the truth of the opposite. This equality of fortune does not allow
18852the one party to be superior to the other in the sphere of speculative
18853cognition; and it is this sphere, accordingly, that is the proper arena
18854of these endless speculative conflicts. But we shall afterwards show
18855that, in relation to its practical exercise, Reason has the right of
18856admitting what, in the field of pure speculation, she would not be
18857justified in supposing, except upon perfectly sufficient grounds;
18858because all such suppositions destroy the necessary completeness of
18859speculation--a condition which the practical reason, however, does not
18860consider to be requisite. In this sphere, therefore, Reason is
18861mistress of a possession, her title to which she does not require to
18862prove--which, in fact, she could not do. The burden of proof accordingly
18863rests upon the opponent. But as he has just as little knowledge
18864regarding the subject discussed, and is as little able to prove the
18865non-existence of the object of an idea, as the philosopher on the other
18866side is to demonstrate its reality, it is evident that there is an
18867advantage on the side of the philosopher who maintains his proposition
18868as a practically necessary supposition (melior est conditio
18869possidentis). For he is at liberty to employ, in self-defence, the same
18870weapons as his opponent makes use of in attacking him; that is, he has a
18871right to use hypotheses not for the purpose of supporting the arguments
18872in favour of his own propositions, but to show that his opponent knows
18873no more than himself regarding the subject under 'discussion and cannot
18874boast of any speculative advantage.
18875
18876Hypotheses are, therefore, admissible in the sphere of pure reason
18877only as weapons for self-defence, and not as supports to dogmatical
18878assertions. But the opposing party we must always seek for in ourselves.
18879For speculative reason is, in the sphere of transcendentalism,
18880dialectical in its own nature. The difficulties and objections we have
18881to fear lie in ourselves. They are like old but never superannuated
18882claims; and we must seek them out, and settle them once and for ever, if
18883we are to expect a permanent peace. External tranquility is hollow and
18884unreal. The root of these contradictions, which lies in the nature of
18885human reason, must be destroyed; and this can only be done by giving it,
18886in the first instance, freedom to grow, nay, by nourishing it, that it
18887may send out shoots, and thus betray its own existence. It is our duty,
18888therefore, to try to discover new objections, to put weapons in the
18889bands of our opponent, and to grant him the most favourable position
18890in the arena that he can wish. We have nothing to fear from these
18891concessions; on the contrary, we may rather hope that we shall thus
18892make ourselves master of a possession which no one will ever venture to
18893dispute.
18894
18895The thinker requires, to be fully equipped, the hypotheses of pure
18896reason, which, although but leaden weapons (for they have not been
18897steeled in the armoury of experience), are as useful as any that can
18898be employed by his opponents. If, accordingly, we have assumed, from a
18899non-speculative point of view, the immaterial nature of the soul, and
18900are met by the objection that experience seems to prove that the growth
18901and decay of our mental faculties are mere modifications of the sensuous
18902organism--we can weaken the force of this objection by the assumption
18903that the body is nothing but the fundamental phenomenon, to which, as
18904a necessary condition, all sensibility, and consequently all thought,
18905relates in the present state of our existence; and that the separation
18906of soul and body forms the conclusion of the sensuous exercise of our
18907power of cognition and the beginning of the intellectual. The body
18908would, in this view of the question, be regarded, not as the cause of
18909thought, but merely as its restrictive condition, as promotive of the
18910sensuous and animal, but as a hindrance to the pure and spiritual life;
18911and the dependence of the animal life on the constitution of the body,
18912would not prove that the whole life of man was also dependent on the
18913state of the organism. We might go still farther, and discover new
18914objections, or carry out to their extreme consequences those which have
18915already been adduced.
18916
18917Generation, in the human race as well as among the irrational animals,
18918depends on so many accidents--of occasion, of proper sustenance, of the
18919laws enacted by the government of a country of vice even, that it is
18920difficult to believe in the eternal existence of a being whose life has
18921begun under circumstances so mean and trivial, and so entirely dependent
18922upon our own control. As regards the continuance of the existence of the
18923whole race, we need have no difficulties, for accident in single cases
18924is subject to general laws; but, in the case of each individual, it
18925would seem as if we could hardly expect so wonderful an effect from
18926causes so insignificant. But, in answer to these objections, we
18927may adduce the transcendental hypothesis that all life is properly
18928intelligible, and not subject to changes of time, and that it neither
18929began in birth, nor will end in death. We may assume that this life is
18930nothing more than a sensuous representation of pure spiritual life; that
18931the whole world of sense is but an image, hovering before the faculty of
18932cognition which we exercise in this sphere, and with no more objective
18933reality than a dream; and that if we could intuite ourselves and
18934other things as they really are, we should see ourselves in a world of
18935spiritual natures, our connection with which did not begin at our birth
18936and will not cease with the destruction of the body. And so on.
18937
18938We cannot be said to know what has been above asserted, nor do we
18939seriously maintain the truth of these assertions; and the notions
18940therein indicated are not even ideas of reason, they are purely
18941fictitious conceptions. But this hypothetical procedure is in perfect
18942conformity with the laws of reason. Our opponent mistakes the absence
18943of empirical conditions for a proof of the complete impossibility of all
18944that we have asserted; and we have to show him that he has not exhausted
18945the whole sphere of possibility and that he can as little compass that
18946sphere by the laws of experience and nature, as we can lay a secure
18947foundation for the operations of reason beyond the region of experience.
18948Such hypothetical defences against the pretensions of an opponent must
18949not be regarded as declarations of opinion. The philosopher abandons
18950them, so soon as the opposite party renounces its dogmatical conceit.
18951To maintain a simply negative position in relation to propositions which
18952rest on an insecure foundation, well befits the moderation of a true
18953philosopher; but to uphold the objections urged against an opponent as
18954proofs of the opposite statement is a proceeding just as unwarrantable
18955and arrogant as it is to attack the position of a philosopher who
18956advances affirmative propositions regarding such a subject.
18957
18958It is evident, therefore, that hypotheses, in the speculative sphere,
18959are valid, not as independent propositions, but only relatively to
18960opposite transcendent assumptions. For, to make the principles of
18961possible experience conditions of the possibility of things in general
18962is just as transcendent a procedure as to maintain the objective reality
18963of ideas which can be applied to no objects except such as lie without
18964the limits of possible experience. The judgements enounced by pure
18965reason must be necessary, or they must not be enounced at all. Reason
18966cannot trouble herself with opinions. But the hypotheses we have been
18967discussing are merely problematical judgements, which can neither be
18968confuted nor proved; while, therefore, they are not personal opinions,
18969they are indispensable as answers to objections which are liable to
18970be raised. But we must take care to confine them to this function,
18971and guard against any assumption on their part of absolute validity, a
18972proceeding which would involve reason in inextricable difficulties and
18973contradictions.
18974
18975
18976
18977SECTION IV. The Discipline of Pure Reason in Relation to Proofs.
18978
18979It is a peculiarity, which distinguishes the proofs of transcendental
18980synthetical propositions from those of all other a priori synthetical
18981cognitions, that reason, in the case of the former, does not apply its
18982conceptions directly to an object, but is first obliged to prove, a
18983priori, the objective validity of these conceptions and the possibility
18984of their syntheses. This is not merely a prudential rule, it is
18985essential to the very possibility of the proof of a transcendental
18986proposition. If I am required to pass, a priori, beyond the conception
18987of an object, I find that it is utterly impossible without the guidance
18988of something which is not contained in the conception. In mathematics,
18989it is a priori intuition that guides my synthesis; and, in this case,
18990all our conclusions may be drawn immediately from pure intuition.
18991In transcendental cognition, so long as we are dealing only with
18992conceptions of the understanding, we are guided by possible experience.
18993That is to say, a proof in the sphere of transcendental cognition does
18994not show that the given conception (that of an event, for example) leads
18995directly to another conception (that of a cause)--for this would be a
18996saltus which nothing can justify; but it shows that experience itself,
18997and consequently the object of experience, is impossible without the
18998connection indicated by these conceptions. It follows that such a
18999proof must demonstrate the possibility of arriving, synthetically and a
19000priori, at a certain knowledge of things, which was not contained in our
19001conceptions of these things. Unless we pay particular attention to this
19002requirement, our proofs, instead of pursuing the straight path indicated
19003by reason, follow the tortuous road of mere subjective association. The
19004illusory conviction, which rests upon subjective causes of association,
19005and which is considered as resulting from the perception of a real and
19006objective natural affinity, is always open to doubt and suspicion.
19007For this reason, all the attempts which have been made to prove the
19008principle of sufficient reason, have, according to the universal
19009admission of philosophers, been quite unsuccessful; and, before the
19010appearance of transcendental criticism, it was considered better, as
19011this principle could not be abandoned, to appeal boldly to the common
19012sense of mankind (a proceeding which always proves that the problem,
19013which reason ought to solve, is one in which philosophers find great
19014difficulties), rather than attempt to discover new dogmatical proofs.
19015
19016But, if the proposition to be proved is a proposition of pure reason,
19017and if I aim at passing beyond my empirical conceptions by the aid of
19018mere ideas, it is necessary that the proof should first show that such
19019a step in synthesis is possible (which it is not), before it proceeds
19020to prove the truth of the proposition itself. The so-called proof of
19021the simple nature of the soul from the unity of apperception, is a very
19022plausible one. But it contains no answer to the objection, that, as
19023the notion of absolute simplicity is not a conception which is directly
19024applicable to a perception, but is an idea which must be inferred--if
19025at all--from observation, it is by no means evident how the mere fact of
19026consciousness, which is contained in all thought, although in so far a
19027simple representation, can conduct me to the consciousness and cognition
19028of a thing which is purely a thinking substance. When I represent to my
19029mind the power of my body as in motion, my body in this thought is so
19030far absolute unity, and my representation of it is a simple one; and
19031hence I can indicate this representation by the motion of a point,
19032because I have made abstraction of the size or volume of the body. But
19033I cannot hence infer that, given merely the moving power of a body,
19034the body may be cogitated as simple substance, merely because the
19035representation in my mind takes no account of its content in space, and
19036is consequently simple. The simple, in abstraction, is very different
19037from the objectively simple; and hence the Ego, which is simple in the
19038first sense, may, in the second sense, as indicating the soul itself,
19039be a very complex conception, with a very various content. Thus it is
19040evident that in all such arguments there lurks a paralogism. We guess
19041(for without some such surmise our suspicion would not be excited
19042in reference to a proof of this character) at the presence of the
19043paralogism, by keeping ever before us a criterion of the possibility of
19044those synthetical propositions which aim at proving more than experience
19045can teach us. This criterion is obtained from the observation that such
19046proofs do not lead us directly from the subject of the proposition to
19047be proved to the required predicate, but find it necessary to presuppose
19048the possibility of extending our cognition a priori by means of ideas.
19049We must, accordingly, always use the greatest caution; we require,
19050before attempting any proof, to consider how it is possible to extend
19051the sphere of cognition by the operations of pure reason, and from
19052what source we are to derive knowledge, which is not obtained from
19053the analysis of conceptions, nor relates, by anticipation, to possible
19054experience. We shall thus spare ourselves much severe and fruitless
19055labour, by not expecting from reason what is beyond its power, or rather
19056by subjecting it to discipline, and teaching it to moderate its vehement
19057desires for the extension of the sphere of cognition.
19058
19059The first rule for our guidance is, therefore, not to attempt a
19060transcendental proof, before we have considered from what source we are
19061to derive the principles upon which the proof is to be based, and what
19062right we have to expect that our conclusions from these principles will
19063be veracious. If they are principles of the understanding, it is vain to
19064expect that we should attain by their means to ideas of pure reason;
19065for these principles are valid only in regard to objects of possible
19066experience. If they are principles of pure reason, our labour is alike
19067in vain. For the principles of reason, if employed as objective, are
19068without exception dialectical and possess no validity or truth, except
19069as regulative principles of the systematic employment of reason in
19070experience. But when such delusive proof are presented to us, it is
19071our duty to meet them with the non liquet of a matured judgement; and,
19072although we are unable to expose the particular sophism upon which the
19073proof is based, we have a right to demand a deduction of the principles
19074employed in it; and, if these principles have their origin in pure
19075reason alone, such a deduction is absolutely impossible. And thus it
19076is unnecessary that we should trouble ourselves with the exposure and
19077confutation of every sophistical illusion; we may, at once, bring all
19078dialectic, which is inexhaustible in the production of fallacies, before
19079the bar of critical reason, which tests the principles upon which all
19080dialectical procedure is based. The second peculiarity of transcendental
19081proof is that a transcendental proposition cannot rest upon more than
19082a single proof. If I am drawing conclusions, not from conceptions, but
19083from intuition corresponding to a conception, be it pure intuition, as
19084in mathematics, or empirical, as in natural science, the intuition which
19085forms the basis of my inferences presents me with materials for many
19086synthetical propositions, which I can connect in various modes, while,
19087as it is allowable to proceed from different points in the intention, I
19088can arrive by different paths at the same proposition.
19089
19090But every transcendental proposition sets out from a conception,
19091and posits the synthetical condition of the possibility of an object
19092according to this conception. There must, therefore, be but one ground
19093of proof, because it is the conception alone which determines the
19094object; and thus the proof cannot contain anything more than the
19095determination of the object according to the conception. In our
19096Transcendental Analytic, for example, we inferred the principle: Every
19097event has a cause, from the only condition of the objective possibility
19098of our conception of an event. This is that an event cannot be
19099determined in time, and consequently cannot form a part of experience,
19100unless it stands under this dynamical law. This is the only possible
19101ground of proof; for our conception of an event possesses objective
19102validity, that is, is a true conception, only because the law of
19103causality determines an object to which it can refer. Other arguments
19104in support of this principle have been attempted--such as that from the
19105contingent nature of a phenomenon; but when this argument is considered,
19106we can discover no criterion of contingency, except the fact of an
19107event--of something happening, that is to say, the existence which is
19108preceded by the non-existence of an object, and thus we fall back on the
19109very thing to be proved. If the proposition: "Every thinking being is
19110simple," is to be proved, we keep to the conception of the ego, which
19111is simple, and to which all thought has a relation. The same is the
19112case with the transcendental proof of the existence of a Deity, which is
19113based solely upon the harmony and reciprocal fitness of the conceptions
19114of an ens realissimum and a necessary being, and cannot be attempted in
19115any other manner.
19116
19117This caution serves to simplify very much the criticism of all
19118propositions of reason. When reason employs conceptions alone, only one
19119proof of its thesis is possible, if any. When, therefore, the dogmatist
19120advances with ten arguments in favour of a proposition, we may be sure
19121that not one of them is conclusive. For if he possessed one which proved
19122the proposition he brings forward to demonstration--as must always be
19123the case with the propositions of pure reason--what need is there for
19124any more? His intention can only be similar to that of the advocate who
19125had different arguments for different judges; this availing himself of
19126the weakness of those who examine his arguments, who, without going into
19127any profound investigation, adopt the view of the case which seems most
19128probable at first sight and decide according to it.
19129
19130The third rule for the guidance of pure reason in the conduct of a proof
19131is that all transcendental proofs must never be apagogic or indirect,
19132but always ostensive or direct. The direct or ostensive proof not only
19133establishes the truth of the proposition to be proved, but exposes the
19134grounds of its truth; the apagogic, on the other hand, may assure us of
19135the truth of the proposition, but it cannot enable us to comprehend
19136the grounds of its possibility. The latter is, accordingly, rather an
19137auxiliary to an argument, than a strictly philosophical and rational
19138mode of procedure. In one respect, however, they have an advantage over
19139direct proofs, from the fact that the mode of arguing by contradiction,
19140which they employ, renders our understanding of the question more
19141clear, and approximates the proof to the certainty of an intuitional
19142demonstration.
19143
19144The true reason why indirect proofs are employed in different sciences
19145is this. When the grounds upon which we seek to base a cognition are too
19146various or too profound, we try whether or not we may not discover
19147the truth of our cognition from its consequences. The modus ponens of
19148reasoning from the truth of its inferences to the truth of a proposition
19149would be admissible if all the inferences that can be drawn from it are
19150known to be true; for in this case there can be only one possible ground
19151for these inferences, and that is the true one. But this is a quite
19152impracticable procedure, as it surpasses all our powers to discover all
19153the possible inferences that can be drawn from a proposition. But this
19154mode of reasoning is employed, under favour, when we wish to prove
19155the truth of an hypothesis; in which case we admit the truth of the
19156conclusion--which is supported by analogy--that, if all the inferences
19157we have drawn and examined agree with the proposition assumed, all
19158other possible inferences will also agree with it. But, in this way, an
19159hypothesis can never be established as a demonstrated truth. The modus
19160tollens of reasoning from known inferences to the unknown proposition,
19161is not only a rigorous, but a very easy mode of proof. For, if it can
19162be shown that but one inference from a proposition is false, then the
19163proposition must itself be false. Instead, then, of examining, in an
19164ostensive argument, the whole series of the grounds on which the
19165truth of a proposition rests, we need only take the opposite of this
19166proposition, and if one inference from it be false, then must the
19167opposite be itself false; and, consequently, the proposition which we
19168wished to prove must be true.
19169
19170The apagogic method of proof is admissible only in those sciences where
19171it is impossible to mistake a subjective representation for an objective
19172cognition. Where this is possible, it is plain that the opposite of a
19173given proposition may contradict merely the subjective conditions of
19174thought, and not the objective cognition; or it may happen that both
19175propositions contradict each other only under a subjective condition,
19176which is incorrectly considered to be objective, and, as the condition
19177is itself false, both propositions may be false, and it will,
19178consequently, be impossible to conclude the truth of the one from the
19179falseness of the other.
19180
19181In mathematics such subreptions are impossible; and it is in this
19182science, accordingly, that the indirect mode of proof has its true
19183place. In the science of nature, where all assertion is based upon
19184empirical intuition, such subreptions may be guarded against by the
19185repeated comparison of observations; but this mode of proof is of little
19186value in this sphere of knowledge. But the transcendental efforts of
19187pure reason are all made in the sphere of the subjective, which is the
19188real medium of all dialectical illusion; and thus reason endeavours,
19189in its premisses, to impose upon us subjective representations for
19190objective cognitions. In the transcendental sphere of pure reason,
19191then, and in the case of synthetical propositions, it is inadmissible
19192to support a statement by disproving the counter-statement. For only
19193two cases are possible; either, the counter-statement is nothing but
19194the enouncement of the inconsistency of the opposite opinion with the
19195subjective conditions of reason, which does not affect the real case
19196(for example, we cannot comprehend the unconditioned necessity of the
19197existence of a being, and hence every speculative proof of the existence
19198of such a being must be opposed on subjective grounds, while the
19199possibility of this being in itself cannot with justice be denied); or,
19200both propositions, being dialectical in their nature, are based upon an
19201impossible conception. In this latter case the rule applies: non entis
19202nulla sunt predicata; that is to say, what we affirm and what we deny,
19203respecting such an object, are equally untrue, and the apagogic mode of
19204arriving at the truth is in this case impossible. If, for example, we
19205presuppose that the world of sense is given in itself in its totality,
19206it is false, either that it is infinite, or that it is finite and
19207limited in space. Both are false, because the hypothesis is false. For
19208the notion of phenomena (as mere representations) which are given in
19209themselves (as objects) is self-contradictory; and the infinitude of
19210this imaginary whole would, indeed, be unconditioned, but would be
19211inconsistent (as everything in the phenomenal world is conditioned)
19212with the unconditioned determination and finitude of quantities which is
19213presupposed in our conception.
19214
19215The apagogic mode of proof is the true source of those illusions which
19216have always had so strong an attraction for the admirers of dogmatical
19217philosophy. It may be compared to a champion who maintains the honour
19218and claims of the party he has adopted by offering battle to all who
19219doubt the validity of these claims and the purity of that honour; while
19220nothing can be proved in this way, except the respective strength of the
19221combatants, and the advantage, in this respect, is always on the side
19222of the attacking party. Spectators, observing that each party is
19223alternately conqueror and conquered, are led to regard the subject of
19224dispute as beyond the power of man to decide upon. But such an opinion
19225cannot be justified; and it is sufficient to apply to these reasoners
19226the remark:
19227
19228 Non defensoribus istis
19229 Tempus eget.
19230
19231Each must try to establish his assertions by a transcendental deduction
19232of the grounds of proof employed in his argument, and thus enable us to
19233see in what way the claims of reason may be supported. If an opponent
19234bases his assertions upon subjective grounds, he may be refuted with
19235ease; not, however to the advantage of the dogmatist, who likewise
19236depends upon subjective sources of cognition and is in like manner
19237driven into a corner by his opponent. But, if parties employ the direct
19238method of procedure, they will soon discover the difficulty, nay, the
19239impossibility of proving their assertions, and will be forced to appeal
19240to prescription and precedence; or they will, by the help of criticism,
19241discover with ease the dogmatical illusions by which they had been
19242mocked, and compel reason to renounce its exaggerated pretensions to
19243speculative insight and to confine itself within the limits of its
19244proper sphere--that of practical principles.
19245
19246
19247
19248CHAPTER II. The Canon of Pure Reason.
19249
19250It is a humiliating consideration for human reason that it is
19251incompetent to discover truth by means of pure speculation, but, on the
19252contrary, stands in need of discipline to check its deviations from the
19253straight path and to expose the illusions which it originates. But,
19254on the other hand, this consideration ought to elevate and to give it
19255confidence, for this discipline is exercised by itself alone, and it is
19256subject to the censure of no other power. The bounds, moreover, which it
19257is forced to set to its speculative exercise, form likewise a check upon
19258the fallacious pretensions of opponents; and thus what remains of its
19259possessions, after these exaggerated claims have been disallowed, is
19260secure from attack or usurpation. The greatest, and perhaps the only,
19261use of all philosophy of pure reason is, accordingly, of a purely
19262negative character. It is not an organon for the extension, but a
19263discipline for the determination, of the limits of its exercise; and
19264without laying claim to the discovery of new truth, it has the modest
19265merit of guarding against error.
19266
19267At the same time, there must be some source of positive cognitions which
19268belong to the domain of pure reason and which become the causes of error
19269only from our mistaking their true character, while they form the goal
19270towards which reason continually strives. How else can we account for
19271the inextinguishable desire in the human mind to find a firm footing in
19272some region beyond the limits of the world of experience? It hopes to
19273attain to the possession of a knowledge in which it has the deepest
19274interest. It enters upon the path of pure speculation; but in vain. We
19275have some reason, however, to expect that, in the only other way that
19276lies open to it--the path of practical reason--it may meet with better
19277success.
19278
19279I understand by a canon a list of the a priori principles of the proper
19280employment of certain faculties of cognition. Thus general logic, in
19281its analytical department, is a formal canon for the faculties of
19282understanding and reason. In the same way, Transcendental Analytic was
19283seen to be a canon of the pure understanding; for it alone is competent
19284to enounce true a priori synthetical cognitions. But, when no proper
19285employment of a faculty of cognition is possible, no canon can exist.
19286But the synthetical cognition of pure speculative reason is, as has been
19287shown, completely impossible. There cannot, therefore, exist any canon
19288for the speculative exercise of this faculty--for its speculative
19289exercise is entirely dialectical; and, consequently, transcendental
19290logic, in this respect, is merely a discipline, and not a canon.
19291If, then, there is any proper mode of employing the faculty of pure
19292reason--in which case there must be a canon for this faculty--this canon
19293will relate, not to the speculative, but to the practical use of reason.
19294This canon we now proceed to investigate.
19295
19296
19297
19298SECTION I. Of the Ultimate End of the Pure Use of Reason.
19299
19300There exists in the faculty of reason a natural desire to venture beyond
19301the field of experience, to attempt to reach the utmost bounds of all
19302cognition by the help of ideas alone, and not to rest satisfied until
19303it has fulfilled its course and raised the sum of its cognitions into a
19304self-subsistent systematic whole. Is the motive for this endeavour to be
19305found in its speculative, or in its practical interests alone?
19306
19307Setting aside, at present, the results of the labours of pure reason in
19308its speculative exercise, I shall merely inquire regarding the problems
19309the solution of which forms its ultimate aim, whether reached or
19310not, and in relation to which all other aims are but partial and
19311intermediate. These highest aims must, from the nature of reason,
19312possess complete unity; otherwise the highest interest of humanity could
19313not be successfully promoted.
19314
19315The transcendental speculation of reason relates to three things: the
19316freedom of the will, the immortality of the soul, and the existence of
19317God. The speculative interest which reason has in those questions is
19318very small; and, for its sake alone, we should not undertake the labour
19319of transcendental investigation--a labour full of toil and ceaseless
19320struggle. We should be loth to undertake this labour, because the
19321discoveries we might make would not be of the smallest use in the sphere
19322of concrete or physical investigation. We may find out that the will is
19323free, but this knowledge only relates to the intelligible cause of our
19324volition. As regards the phenomena or expressions of this will, that is,
19325our actions, we are bound, in obedience to an inviolable maxim, without
19326which reason cannot be employed in the sphere of experience, to explain
19327these in the same way as we explain all the other phenomena of nature,
19328that is to say, according to its unchangeable laws. We may have
19329discovered the spirituality and immortality of the soul, but we cannot
19330employ this knowledge to explain the phenomena of this life, nor the
19331peculiar nature of the future, because our conception of an incorporeal
19332nature is purely negative and does not add anything to our knowledge,
19333and the only inferences to be drawn from it are purely fictitious. If,
19334again, we prove the existence of a supreme intelligence, we should be
19335able from it to make the conformity to aims existing in the arrangement
19336of the world comprehensible; but we should not be justified in deducing
19337from it any particular arrangement or disposition, or inferring any
19338where it is not perceived. For it is a necessary rule of the speculative
19339use of reason that we must not overlook natural causes, or refuse to
19340listen to the teaching of experience, for the sake of deducing what we
19341know and perceive from something that transcends all our knowledge.
19342In one word, these three propositions are, for the speculative reason,
19343always transcendent, and cannot be employed as immanent principles in
19344relation to the objects of experience; they are, consequently, of no use
19345to us in this sphere, being but the valueless results of the severe but
19346unprofitable efforts of reason.
19347
19348If, then, the actual cognition of these three cardinal propositions is
19349perfectly useless, while Reason uses her utmost endeavours to induce us
19350to admit them, it is plain that their real value and importance relate
19351to our practical, and not to our speculative interest.
19352
19353I term all that is possible through free will, practical. But if the
19354conditions of the exercise of free volition are empirical, reason can
19355have only a regulative, and not a constitutive, influence upon it, and
19356is serviceable merely for the introduction of unity into its empirical
19357laws. In the moral philosophy of prudence, for example, the sole
19358business of reason is to bring about a union of all the ends, which
19359are aimed at by our inclinations, into one ultimate end--that of
19360happiness--and to show the agreement which should exist among the
19361means of attaining that end. In this sphere, accordingly, reason cannot
19362present to us any other than pragmatical laws of free action, for our
19363guidance towards the aims set up by the senses, and is incompetent to
19364give us laws which are pure and determined completely a priori. On the
19365other hand, pure practical laws, the ends of which have been given by
19366reason entirely a priori, and which are not empirically conditioned, but
19367are, on the contrary, absolutely imperative in their nature, would be
19368products of pure reason. Such are the moral laws; and these alone belong
19369to the sphere of the practical exercise of reason, and admit of a canon.
19370
19371All the powers of reason, in the sphere of what may be termed pure
19372philosophy, are, in fact, directed to the three above-mentioned problems
19373alone. These again have a still higher end--the answer to the question,
19374what we ought to do, if the will is free, if there is a God and a future
19375world. Now, as this problem relates to our in reference to the highest
19376aim of humanity, it is evident that the ultimate intention of nature, in
19377the constitution of our reason, has been directed to the moral alone.
19378
19379We must take care, however, in turning our attention to an object which
19380is foreign* to the sphere of transcendental philosophy, not to injure
19381the unity of our system by digressions, nor, on the other hand, to fail
19382in clearness, by saying too little on the new subject of discussion.
19383I hope to avoid both extremes, by keeping as close as possible to the
19384transcendental, and excluding all psychological, that is, empirical,
19385elements.
19386
19387
19388 [*Footnote: All practical conceptions relate to objects of pleasure and
19389pain, and consequently--in an indirect manner, at least--to objects of
19390feeling. But as feeling is not a faculty of representation, but lies
19391out of the sphere of our powers of cognition, the elements of our
19392judgements, in so far as they relate to pleasure or pain, that is, the
19393elements of our practical judgements, do not belong to transcendental
19394philosophy, which has to do with pure a priori cognitions alone.]
19395
19396
19397I have to remark, in the first place, that at present I treat of the
19398conception of freedom in the practical sense only, and set aside the
19399corresponding transcendental conception, which cannot be employed as a
19400ground of explanation in the phenomenal world, but is itself a problem
19401for pure reason. A will is purely animal (arbitrium brutum) when it is
19402determined by sensuous impulses or instincts only, that is, when it is
19403determined in a pathological manner. A will, which can be determined
19404independently of sensuous impulses, consequently by motives presented by
19405reason alone, is called a free will (arbitrium liberum); and everything
19406which is connected with this free will, either as principle or
19407consequence, is termed practical. The existence of practical freedom can
19408be proved from experience alone. For the human will is not determined
19409by that alone which immediately affects the senses; on the contrary, we
19410have the power, by calling up the notion of what is useful or hurtful in
19411a more distant relation, of overcoming the immediate impressions on
19412our sensuous faculty of desire. But these considerations of what is
19413desirable in relation to our whole state, that is, is in the end good
19414and useful, are based entirely upon reason. This faculty, accordingly,
19415enounces laws, which are imperative or objective laws of freedom and
19416which tell us what ought to take place, thus distinguishing themselves
19417from the laws of nature, which relate to that which does take place. The
19418laws of freedom or of free will are hence termed practical laws.
19419
19420Whether reason is not itself, in the actual delivery of these laws,
19421determined in its turn by other influences, and whether the action
19422which, in relation to sensuous impulses, we call free, may not, in
19423relation to higher and more remote operative causes, really form a part
19424of nature--these are questions which do not here concern us. They are
19425purely speculative questions; and all we have to do, in the practical
19426sphere, is to inquire into the rule of conduct which reason has to
19427present. Experience demonstrates to us the existence of practical
19428freedom as one of the causes which exist in nature, that is, it shows
19429the causal power of reason in the determination of the will. The idea
19430of transcendental freedom, on the contrary, requires that reason--in
19431relation to its causal power of commencing a series of phenomena--should
19432be independent of all sensuous determining causes; and thus it seems to
19433be in opposition to the law of nature and to all possible experience.
19434It therefore remains a problem for the human mind. But this problem does
19435not concern reason in its practical use; and we have, therefore, in a
19436canon of pure reason, to do with only two questions, which relate to
19437the practical interest of pure reason: Is there a God? and, Is there
19438a future life? The question of transcendental freedom is purely
19439speculative, and we may therefore set it entirely aside when we come
19440to treat of practical reason. Besides, we have already discussed this
19441subject in the antinomy of pure reason.
19442
19443
19444SECTION II. Of the Ideal of the Summum Bonum as a Determining
19445Ground of the Ultimate End of Pure Reason.
19446
19447Reason conducted us, in its speculative use, through the field of
19448experience and, as it can never find complete satisfaction in that
19449sphere, from thence to speculative ideas--which, however, in the end
19450brought us back again to experience, and thus fulfilled the purpose of
19451reason, in a manner which, though useful, was not at all in accordance
19452with our expectations. It now remains for us to consider whether pure
19453reason can be employed in a practical sphere, and whether it will here
19454conduct us to those ideas which attain the highest ends of pure reason,
19455as we have just stated them. We shall thus ascertain whether, from
19456the point of view of its practical interest, reason may not be able to
19457supply us with that which, on the speculative side, it wholly denies us.
19458
19459The whole interest of reason, speculative as well as practical, is
19460centred in the three following questions:
19461
19462 1. WHAT CAN I KNOW?
19463 2. WHAT OUGHT I TO DO?
19464 3. WHAT MAY I HOPE?
19465
19466The first question is purely speculative. We have, as I flatter myself,
19467exhausted all the replies of which it is susceptible, and have at last
19468found the reply with which reason must content itself, and with which it
19469ought to be content, so long as it pays no regard to the practical. But
19470from the two great ends to the attainment of which all these efforts of
19471pure reason were in fact directed, we remain just as far removed as if
19472we had consulted our ease and declined the task at the outset. So far,
19473then, as knowledge is concerned, thus much, at least, is established,
19474that, in regard to those two problems, it lies beyond our reach.
19475
19476The second question is purely practical. As such it may indeed fall
19477within the province of pure reason, but still it is not transcendental,
19478but moral, and consequently cannot in itself form the subject of our
19479criticism.
19480
19481The third question: If I act as I ought to do, what may I then hope?--is
19482at once practical and theoretical. The practical forms a clue to
19483the answer of the theoretical, and--in its highest form--speculative
19484question. For all hoping has happiness for its object and stands in
19485precisely the same relation to the practical and the law of morality as
19486knowing to the theoretical cognition of things and the law of nature.
19487The former arrives finally at the conclusion that something is (which
19488determines the ultimate end), because something ought to take place; the
19489latter, that something is (which operates as the highest cause), because
19490something does take place.
19491
19492Happiness is the satisfaction of all our desires; extensive, in regard
19493to their multiplicity; intensive, in regard to their degree; and
19494protensive, in regard to their duration. The practical law based on the
19495motive of happiness I term a pragmatical law (or prudential rule); but
19496that law, assuming such to exist, which has no other motive than the
19497worthiness of being happy, I term a moral or ethical law. The first
19498tells us what we have to do, if we wish to become possessed of
19499happiness; the second dictates how we ought to act, in order to deserve
19500happiness. The first is based upon empirical principles; for it is only
19501by experience that I can learn either what inclinations exist which
19502desire satisfaction, or what are the natural means of satisfying them.
19503The second takes no account of our desires or the means of satisfying
19504them, and regards only the freedom of a rational being, and the
19505necessary conditions under which alone this freedom can harmonize with
19506the distribution of happiness according to principles. This second law
19507may therefore rest upon mere ideas of pure reason, and may be cognized a
19508priori.
19509
19510I assume that there are pure moral laws which determine, entirely a
19511priori (without regard to empirical motives, that is, to happiness), the
19512conduct of a rational being, or in other words, to use which it makes of
19513its freedom, and that these laws are absolutely imperative (not merely
19514hypothetically, on the supposition of other empirical ends), and
19515therefore in all respects necessary. I am warranted in assuming this,
19516not only by the arguments of the most enlightened moralists, but by
19517the moral judgement of every man who will make the attempt to form a
19518distinct conception of such a law.
19519
19520Pure reason, then, contains, not indeed in its speculative, but in
19521its practical, or, more strictly, its moral use, principles of the
19522possibility of experience, of such actions, namely, as, in accordance
19523with ethical precepts, might be met with in the history of man. For
19524since reason commands that such actions should take place, it must
19525be possible for them to take place; and hence a particular kind of
19526systematic unity--the moral--must be possible. We have found, it is
19527true, that the systematic unity of nature could not be established
19528according to speculative principles of reason, because, while reason
19529possesses a causal power in relation to freedom, it has none in relation
19530to the whole sphere of nature; and, while moral principles of reason can
19531produce free actions, they cannot produce natural laws. It is, then, in
19532its practical, but especially in its moral use, that the principles of
19533pure reason possess objective reality.
19534
19535I call the world a moral world, in so far as it may be in accordance
19536with all the ethical laws--which, by virtue of the freedom of reasonable
19537beings, it can be, and according to the necessary laws of morality it
19538ought to be. But this world must be conceived only as an intelligible
19539world, inasmuch as abstraction is therein made of all conditions (ends),
19540and even of all impediments to morality (the weakness or pravity of
19541human nature). So far, then, it is a mere idea--though still a practical
19542idea--which may have, and ought to have, an influence on the world of
19543sense, so as to bring it as far as possible into conformity with itself.
19544The idea of a moral world has, therefore, objective reality, not as
19545referring to an object of intelligible intuition--for of such an
19546object we can form no conception whatever--but to the world of
19547sense--conceived, however, as an object of pure reason in its practical
19548use--and to a corpus mysticum of rational beings in it, in so far as the
19549liberum arbitrium of the individual is placed, under and by virtue of
19550moral laws, in complete systematic unity both with itself and with the
19551freedom of all others.
19552
19553That is the answer to the first of the two questions of pure reason
19554which relate to its practical interest: Do that which will render thee
19555worthy of happiness. The second question is this: If I conduct myself
19556so as not to be unworthy of happiness, may I hope thereby to obtain
19557happiness? In order to arrive at the solution of this question, we must
19558inquire whether the principles of pure reason, which prescribe a priori
19559the law, necessarily also connect this hope with it.
19560
19561I say, then, that just as the moral principles are necessary according
19562to reason in its practical use, so it is equally necessary according
19563to reason in its theoretical use to assume that every one has ground to
19564hope for happiness in the measure in which he has made himself worthy
19565of it in his conduct, and that therefore the system of morality is
19566inseparably (though only in the idea of pure reason) connected with that
19567of happiness.
19568
19569Now in an intelligible, that is, in the moral world, in the conception
19570of which we make abstraction of all the impediments to morality
19571(sensuous desires), such a system of happiness, connected with and
19572proportioned to morality, may be conceived as necessary, because freedom
19573of volition--partly incited, and partly restrained by moral laws--would
19574be itself the cause of general happiness; and thus rational beings,
19575under the guidance of such principles, would be themselves the authors
19576both of their own enduring welfare and that of others. But such a system
19577of self-rewarding morality is only an idea, the carrying out of which
19578depends upon the condition that every one acts as he ought; in other
19579words, that all actions of reasonable beings be such as they would be if
19580they sprung from a Supreme Will, comprehending in, or under, itself all
19581particular wills. But since the moral law is binding on each individual
19582in the use of his freedom of volition, even if others should not act
19583in conformity with this law, neither the nature of things, nor the
19584causality of actions and their relation to morality, determine how the
19585consequences of these actions will be related to happiness; and the
19586necessary connection of the hope of happiness with the unceasing
19587endeavour to become worthy of happiness, cannot be cognized by reason,
19588if we take nature alone for our guide. This connection can be hoped for
19589only on the assumption that the cause of nature is a supreme reason,
19590which governs according to moral laws.
19591
19592I term the idea of an intelligence in which the morally most perfect
19593will, united with supreme blessedness, is the cause of all happiness in
19594the world, so far as happiness stands in strict relation to morality
19595(as the worthiness of being happy), the ideal of the supreme Good. It is
19596only, then, in the ideal of the supreme original good, that pure reason
19597can find the ground of the practically necessary connection of
19598both elements of the highest derivative good, and accordingly of an
19599intelligible, that is, moral world. Now since we are necessitated by
19600reason to conceive ourselves as belonging to such a world, while the
19601senses present to us nothing but a world of phenomena, we must assume
19602the former as a consequence of our conduct in the world of sense (since
19603the world of sense gives us no hint of it), and therefore as future in
19604relation to us. Thus God and a future life are two hypotheses which,
19605according to the principles of pure reason, are inseparable from the
19606obligation which this reason imposes upon us.
19607
19608Morality per se constitutes a system. But we can form no system of
19609happiness, except in so far as it is dispensed in strict proportion to
19610morality. But this is only possible in the intelligible world, under a
19611wise author and ruler. Such a ruler, together with life in such a world,
19612which we must look upon as future, reason finds itself compelled to
19613assume; or it must regard the moral laws as idle dreams, since the
19614necessary consequence which this same reason connects with them must,
19615without this hypothesis, fall to the ground. Hence also the moral laws
19616are universally regarded as commands, which they could not be did they
19617not connect a priori adequate consequences with their dictates, and thus
19618carry with them promises and threats. But this, again, they could not
19619do, did they not reside in a necessary being, as the Supreme Good, which
19620alone can render such a teleological unity possible.
19621
19622Leibnitz termed the world, when viewed in relation to the rational
19623beings which it contains, and the moral relations in which they stand
19624to each other, under the government of the Supreme Good, the kingdom of
19625Grace, and distinguished it from the kingdom of Nature, in which these
19626rational beings live, under moral laws, indeed, but expect no other
19627consequences from their actions than such as follow according to the
19628course of nature in the world of sense. To view ourselves, therefore, as
19629in the kingdom of grace, in which all happiness awaits us, except in
19630so far as we ourselves limit our participation in it by actions which
19631render us unworthy of happiness, is a practically necessary idea of
19632reason.
19633
19634Practical laws, in so far as they are subjective grounds of actions,
19635that is, subjective principles, are termed maxims. The judgements
19636of moral according to in its purity and ultimate results are framed
19637according ideas; the observance of its laws, according to according to
19638maxims.
19639
19640The whole course of our life must be subject to moral maxims; but this
19641is impossible, unless with the moral law, which is a mere idea, reason
19642connects an efficient cause which ordains to all conduct which is in
19643conformity with the moral law an issue either in this or in another
19644life, which is in exact conformity with our highest aims. Thus, without
19645a God and without a world, invisible to us now, but hoped for, the
19646glorious ideas of morality are, indeed, objects of approbation and of
19647admiration, but cannot be the springs of purpose and action. For they do
19648not satisfy all the aims which are natural to every rational being, and
19649which are determined a priori by pure reason itself, and necessary.
19650
19651Happiness alone is, in the view of reason, far from being the complete
19652good. Reason does not approve of it (however much inclination may desire
19653it), except as united with desert. On the other hand, morality alone,
19654and with it, mere desert, is likewise far from being the complete good.
19655To make it complete, he who conducts himself in a manner not unworthy
19656of happiness, must be able to hope for the possession of happiness. Even
19657reason, unbiased by private ends, or interested considerations, cannot
19658judge otherwise, if it puts itself in the place of a being whose
19659business it is to dispense all happiness to others. For in the practical
19660idea both points are essentially combined, though in such a way
19661that participation in happiness is rendered possible by the moral
19662disposition, as its condition, and not conversely, the moral disposition
19663by the prospect of happiness. For a disposition which should require the
19664prospect of happiness as its necessary condition would not be moral, and
19665hence also would not be worthy of complete happiness--a happiness which,
19666in the view of reason, recognizes no limitation but such as arises from
19667our own immoral conduct.
19668
19669Happiness, therefore, in exact proportion with the morality of rational
19670beings (whereby they are made worthy of happiness), constitutes alone
19671the supreme good of a world into which we absolutely must transport
19672ourselves according to the commands of pure but practical reason.
19673This world is, it is true, only an intelligible world; for of such a
19674systematic unity of ends as it requires, the world of sense gives us no
19675hint. Its reality can be based on nothing else but the hypothesis of a
19676supreme original good. In it independent reason, equipped with all
19677the sufficiency of a supreme cause, founds, maintains, and fulfils the
19678universal order of things, with the most perfect teleological harmony,
19679however much this order may be hidden from us in the world of sense.
19680
19681This moral theology has the peculiar advantage, in contrast with
19682speculative theology, of leading inevitably to the conception of a sole,
19683perfect, and rational First Cause, whereof speculative theology does
19684not give us any indication on objective grounds, far less any convincing
19685evidence. For we find neither in transcendental nor in natural theology,
19686however far reason may lead us in these, any ground to warrant us in
19687assuming the existence of one only Being, which stands at the head of
19688all natural causes, and on which these are entirely dependent. On the
19689other band, if we take our stand on moral unity as a necessary law of
19690the universe, and from this point of view consider what is necessary to
19691give this law adequate efficiency and, for us, obligatory force, we
19692must come to the conclusion that there is one only supreme will, which
19693comprehends all these laws in itself. For how, under different wills,
19694should we find complete unity of ends? This will must be omnipotent,
19695that all nature and its relation to morality in the world may be
19696subject to it; omniscient, that it may have knowledge of the most secret
19697feelings and their moral worth; omnipresent, that it may be at hand to
19698supply every necessity to which the highest weal of the world may give
19699rise; eternal, that this harmony of nature and liberty may never fail;
19700and so on.
19701
19702But this systematic unity of ends in this world of intelligences--which,
19703as mere nature, is only a world of sense, but, as a system of freedom
19704of volition, may be termed an intelligible, that is, moral world (regnum
19705gratiae)--leads inevitably also to the teleological unity of all things
19706which constitute this great whole, according to universal natural
19707laws--just as the unity of the former is according to universal and
19708necessary moral laws--and unites the practical with the speculative
19709reason. The world must be represented as having originated from an idea,
19710if it is to harmonize with that use of reason without which we cannot
19711even consider ourselves as worthy of reason--namely, the moral use,
19712which rests entirely on the idea of the supreme good. Hence the
19713investigation of nature receives a teleological direction, and becomes,
19714in its widest extension, physico-theology. But this, taking its rise
19715in moral order as a unity founded on the essence of freedom, and
19716not accidentally instituted by external commands, establishes the
19717teleological view of nature on grounds which must be inseparably
19718connected with the internal possibility of things. This gives rise to
19719a transcendental theology, which takes the ideal of the highest
19720ontological perfection as a principle of systematic unity; and this
19721principle connects all things according to universal and necessary
19722natural laws, because all things have their origin in the absolute
19723necessity of the one only Primal Being.
19724
19725What use can we make of our understanding, even in respect of
19726experience, if we do not propose ends to ourselves? But the highest ends
19727are those of morality, and it is only pure reason that can give us the
19728knowledge of these. Though supplied with these, and putting ourselves
19729under their guidance, we can make no teleological use of the knowledge
19730of nature, as regards cognition, unless nature itself has established
19731teleological unity. For without this unity we should not even possess
19732reason, because we should have no school for reason, and no cultivation
19733through objects which afford the materials for its conceptions. But
19734teleological unity is a necessary unity, and founded on the essence of
19735the individual will itself. Hence this will, which is the condition of
19736the application of this unity in concreto, must be so likewise. In this
19737way the transcendental enlargement of our rational cognition would be,
19738not the cause, but merely the effect of the practical teleology which
19739pure reason imposes upon us.
19740
19741Hence, also, we find in the history of human reason that, before the
19742moral conceptions were sufficiently purified and determined, and
19743before men had attained to a perception of the systematic unity of
19744ends according to these conceptions and from necessary principles, the
19745knowledge of nature, and even a considerable amount of intellectual
19746culture in many other sciences, could produce only rude and vague
19747conceptions of the Deity, sometimes even admitting of an astonishing
19748indifference with regard to this question altogether. But the more
19749enlarged treatment of moral ideas, which was rendered necessary by
19750the extreme pure moral law of our religion, awakened the interest, and
19751thereby quickened the perceptions of reason in relation to this object.
19752In this way, and without the help either of an extended acquaintance
19753with nature, or of a reliable transcendental insight (for these have
19754been wanting in all ages), a conception of the Divine Being was arrived
19755at, which we now bold to be the correct one, not because speculative
19756reason convinces us of its correctness, but because it accords with the
19757moral principles of reason. Thus it is to pure reason, but only in its
19758practical use, that we must ascribe the merit of having connected with
19759our highest interest a cognition, of which mere speculation was able
19760only to form a conjecture, but the validity of which it was unable to
19761establish--and of having thereby rendered it, not indeed a demonstrated
19762dogma, but a hypothesis absolutely necessary to the essential ends of
19763reason.
19764
19765But if practical reason has reached this elevation, and has attained to
19766the conception of a sole Primal Being as the supreme good, it must not,
19767therefore, imagine that it has transcended the empirical conditions of
19768its application, and risen to the immediate cognition of new objects; it
19769must not presume to start from the conception which it has gained, and
19770to deduce from it the moral laws themselves. For it was these very laws,
19771the internal practical necessity of which led us to the hypothesis of an
19772independent cause, or of a wise ruler of the universe, who should give
19773them effect. Hence we are not entitled to regard them as accidental
19774and derived from the mere will of the ruler, especially as we have no
19775conception of such a will, except as formed in accordance with these
19776laws. So far, then, as practical reason has the right to conduct us,
19777we shall not look upon actions as binding on us, because they are the
19778commands of God, but we shall regard them as divine commands, because
19779we are internally bound by them. We shall study freedom under the
19780teleological unity which accords with principles of reason; we shall
19781look upon ourselves as acting in conformity with the divine will only in
19782so far as we hold sacred the moral law which reason teaches us from the
19783nature of actions themselves, and we shall believe that we can obey
19784that will only by promoting the weal of the universe in ourselves and in
19785others. Moral theology is, therefore, only of immanent use. It teaches
19786us to fulfil our destiny here in the world, by placing ourselves in
19787harmony with the general system of ends, and warns us against the
19788fanaticism, nay, the crime of depriving reason of its legislative
19789authority in the moral conduct of life, for the purpose of directly
19790connecting this authority with the idea of the Supreme Being. For this
19791would be, not an immanent, but a transcendent use of moral theology,
19792and, like the transcendent use of mere speculation, would inevitably
19793pervert and frustrate the ultimate ends of reason.
19794
19795
19796
19797SECTION III. Of Opinion, Knowledge, and Belief.
19798
19799The holding of a thing to be true is a phenomenon in our understanding
19800which may rest on objective grounds, but requires, also, subjective
19801causes in the mind of the person judging. If a judgement is valid for
19802every rational being, then its ground is objectively sufficient, and it
19803is termed a conviction. If, on the other hand, it has its ground in the
19804particular character of the subject, it is termed a persuasion.
19805
19806Persuasion is a mere illusion, the ground of the judgement, which lies
19807solely in the subject, being regarded as objective. Hence a judgement
19808of this kind has only private validity--is only valid for the individual
19809who judges, and the holding of a thing to be true in this way cannot
19810be communicated. But truth depends upon agreement with the object, and
19811consequently the judgements of all understandings, if true, must be in
19812agreement with each other (consentientia uni tertio consentiunt inter
19813se). Conviction may, therefore, be distinguished, from an external point
19814of view, from persuasion, by the possibility of communicating it and by
19815showing its validity for the reason of every man; for in this case the
19816presumption, at least, arises that the agreement of all judgements with
19817each other, in spite of the different characters of individuals, rests
19818upon the common ground of the agreement of each with the object, and
19819thus the correctness of the judgement is established.
19820
19821Persuasion, accordingly, cannot be subjectively distinguished from
19822conviction, that is, so long as the subject views its judgement simply
19823as a phenomenon of its own mind. But if we inquire whether the grounds
19824of our judgement, which are valid for us, produce the same effect on
19825the reason of others as on our own, we have then the means, though only
19826subjective means, not, indeed, of producing conviction, but of detecting
19827the merely private validity of the judgement; in other words, of
19828discovering that there is in it the element of mere persuasion.
19829
19830If we can, in addition to this, develop the subjective causes of the
19831judgement, which we have taken for its objective grounds, and thus
19832explain the deceptive judgement as a phenomenon in our mind, apart
19833altogether from the objective character of the object, we can then
19834expose the illusion and need be no longer deceived by it, although, if
19835its subjective cause lies in our nature, we cannot hope altogether to
19836escape its influence.
19837
19838I can only maintain, that is, affirm as necessarily valid for every one,
19839that which produces conviction. Persuasion I may keep for myself, if it
19840is agreeable to me; but I cannot, and ought not, to attempt to impose it
19841as binding upon others.
19842
19843Holding for true, or the subjective validity of a judgement in relation
19844to conviction (which is, at the same time, objectively valid), has the
19845three following degrees: opinion, belief, and knowledge. Opinion is a
19846consciously insufficient judgement, subjectively as well as objectively.
19847Belief is subjectively sufficient, but is recognized as being
19848objectively insufficient. Knowledge is both subjectively and objectively
19849sufficient. Subjective sufficiency is termed conviction (for myself);
19850objective sufficiency is termed certainty (for all). I need not dwell
19851longer on the explanation of such simple conceptions.
19852
19853I must never venture to be of opinion, without knowing something, at
19854least, by which my judgement, in itself merely problematical, is brought
19855into connection with the truth--which connection, although not perfect,
19856is still something more than an arbitrary fiction. Moreover, the law of
19857such a connection must be certain. For if, in relation to this law,
19858I have nothing more than opinion, my judgement is but a play of the
19859imagination, without the least relation to truth. In the judgements of
19860pure reason, opinion has no place. For, as they do not rest on empirical
19861grounds and as the sphere of pure reason is that of necessary truth
19862and a priori cognition, the principle of connection in it
19863requires universality and necessity, and consequently perfect
19864certainty--otherwise we should have no guide to the truth at all. Hence
19865it is absurd to have an opinion in pure mathematics; we must know, or
19866abstain from forming a judgement altogether. The case is the same with
19867the maxims of morality. For we must not hazard an action on the mere
19868opinion that it is allowed, but we must know it to be so. In the
19869transcendental sphere of reason, on the other hand, the term opinion
19870is too weak, while the word knowledge is too strong. From the merely
19871speculative point of view, therefore, we cannot form a judgement at
19872all. For the subjective grounds of a judgement, such as produce belief,
19873cannot be admitted in speculative inquiries, inasmuch as they cannot
19874stand without empirical support and are incapable of being communicated
19875to others in equal measure.
19876
19877But it is only from the practical point of view that a theoretically
19878insufficient judgement can be termed belief. Now the practical reference
19879is either to skill or to morality; to the former, when the end proposed
19880is arbitrary and accidental, to the latter, when it is absolutely
19881necessary.
19882
19883If we propose to ourselves any end whatever, the conditions of its
19884attainment are hypothetically necessary. The necessity is subjectively,
19885but still only comparatively, sufficient, if I am acquainted with no
19886other conditions under which the end can be attained. On the other hand,
19887it is sufficient, absolutely and for every one, if I know for certain
19888that no one can be acquainted with any other conditions under which the
19889attainment of the proposed end would be possible. In the former case my
19890supposition--my judgement with regard to certain conditions--is a merely
19891accidental belief; in the latter it is a necessary belief. The physician
19892must pursue some course in the case of a patient who is in danger, but
19893is ignorant of the nature of the disease. He observes the symptoms, and
19894concludes, according to the best of his judgement, that it is a case
19895of phthisis. His belief is, even in his own judgement, only contingent:
19896another man might, perhaps come nearer the truth. Such a belief,
19897contingent indeed, but still forming the ground of the actual use of
19898means for the attainment of certain ends, I term Pragmatical belief.
19899
19900The usual test, whether that which any one maintains is merely his
19901persuasion, or his subjective conviction at least, that is, his firm
19902belief, is a bet. It frequently happens that a man delivers his opinions
19903with so much boldness and assurance, that he appears to be under no
19904apprehension as to the possibility of his being in error. The offer of
19905a bet startles him, and makes him pause. Sometimes it turns out that
19906his persuasion may be valued at a ducat, but not at ten. For he does not
19907hesitate, perhaps, to venture a ducat, but if it is proposed to stake
19908ten, he immediately becomes aware of the possibility of his being
19909mistaken--a possibility which has hitherto escaped his observation. If
19910we imagine to ourselves that we have to stake the happiness of our whole
19911life on the truth of any proposition, our judgement drops its air of
19912triumph, we take the alarm, and discover the actual strength of our
19913belief. Thus pragmatical belief has degrees, varying in proportion to
19914the interests at stake.
19915
19916Now, in cases where we cannot enter upon any course of action in
19917reference to some object, and where, accordingly, our judgement is
19918purely theoretical, we can still represent to ourselves, in thought,
19919the possibility of a course of action, for which we suppose that we have
19920sufficient grounds, if any means existed of ascertaining the truth of
19921the matter. Thus we find in purely theoretical judgements an analogon of
19922practical judgements, to which the word belief may properly be applied,
19923and which we may term doctrinal belief. I should not hesitate to stake
19924my all on the truth of the proposition--if there were any possibility of
19925bringing it to the test of experience--that, at least, some one of the
19926planets, which we see, is inhabited. Hence I say that I have not merely
19927the opinion, but the strong belief, on the correctness of which I would
19928stake even many of the advantages of life, that there are inhabitants in
19929other worlds.
19930
19931Now we must admit that the doctrine of the existence of God belongs to
19932doctrinal belief. For, although in respect to the theoretical cognition
19933of the universe I do not require to form any theory which necessarily
19934involves this idea, as the condition of my explanation of the phenomena
19935which the universe presents, but, on the contrary, am rather bound so
19936to use my reason as if everything were mere nature, still teleological
19937unity is so important a condition of the application of my reason to
19938nature, that it is impossible for me to ignore it--especially since, in
19939addition to these considerations, abundant examples of it are supplied
19940by experience. But the sole condition, so far as my knowledge extends,
19941under which this unity can be my guide in the investigation of nature,
19942is the assumption that a supreme intelligence has ordered all things
19943according to the wisest ends. Consequently, the hypothesis of a wise
19944author of the universe is necessary for my guidance in the investigation
19945of nature--is the condition under which alone I can fulfil an end which
19946is contingent indeed, but by no means unimportant. Moreover, since
19947the result of my attempts so frequently confirms the utility of this
19948assumption, and since nothing decisive can be adduced against it, it
19949follows that it would be saying far too little to term my judgement,
19950in this case, a mere opinion, and that, even in this theoretical
19951connection, I may assert that I firmly believe in God. Still, if we use
19952words strictly, this must not be called a practical, but a doctrinal
19953belief, which the theology of nature (physico-theology) must also
19954produce in my mind. In the wisdom of a Supreme Being, and in the
19955shortness of life, so inadequate to the development of the glorious
19956powers of human nature, we may find equally sufficient grounds for a
19957doctrinal belief in the future life of the human soul.
19958
19959The expression of belief is, in such cases, an expression of modesty
19960from the objective point of view, but, at the same time, of firm
19961confidence, from the subjective. If I should venture to term this merely
19962theoretical judgement even so much as a hypothesis which I am entitled
19963to assume; a more complete conception, with regard to another world and
19964to the cause of the world, might then be justly required of me than I
19965am, in reality, able to give. For, if I assume anything, even as a mere
19966hypothesis, I must, at least, know so much of the properties of such a
19967being as will enable me, not to form the conception, but to imagine the
19968existence of it. But the word belief refers only to the guidance which
19969an idea gives me, and to its subjective influence on the conduct of
19970my reason, which forces me to hold it fast, though I may not be in a
19971position to give a speculative account of it.
19972
19973But mere doctrinal belief is, to some extent, wanting in stability.
19974We often quit our hold of it, in consequence of the difficulties which
19975occur in speculation, though in the end we inevitably return to it
19976again.
19977
19978It is quite otherwise with moral belief. For in this sphere action is
19979absolutely necessary, that is, I must act in obedience to the moral law
19980in all points. The end is here incontrovertibly established, and there
19981is only one condition possible, according to the best of my perception,
19982under which this end can harmonize with all other ends, and so have
19983practical validity--namely, the existence of a God and of a future
19984world. I know also, to a certainty, that no one can be acquainted with
19985any other conditions which conduct to the same unity of ends under the
19986moral law. But since the moral precept is, at the same time, my maxim
19987(as reason requires that it should be), I am irresistibly constrained to
19988believe in the existence of God and in a future life; and I am sure
19989that nothing can make me waver in this belief, since I should thereby
19990overthrow my moral maxims, the renunciation of which would render me
19991hateful in my own eyes.
19992
19993Thus, while all the ambitious attempts of reason to penetrate beyond the
19994limits of experience end in disappointment, there is still enough left
19995to satisfy us in a practical point of view. No one, it is true, will be
19996able to boast that he knows that there is a God and a future life; for,
19997if he knows this, he is just the man whom I have long wished to find.
19998All knowledge, regarding an object of mere reason, can be communicated;
19999and I should thus be enabled to hope that my own knowledge would
20000receive this wonderful extension, through the instrumentality of his
20001instruction. No, my conviction is not logical, but moral certainty; and
20002since it rests on subjective grounds (of the moral sentiment), I must
20003not even say: It is morally certain that there is a God, etc., but: I
20004am morally certain, that is, my belief in God and in another world is so
20005interwoven with my moral nature that I am under as little apprehension
20006of having the former torn from me as of losing the latter.
20007
20008The only point in this argument that may appear open to suspicion is
20009that this rational belief presupposes the existence of moral sentiments.
20010If we give up this assumption, and take a man who is entirely
20011indifferent with regard to moral laws, the question which reason
20012proposes, becomes then merely a problem for speculation and may, indeed,
20013be supported by strong grounds from analogy, but not by such as
20014will compel the most obstinate scepticism to give way.* But in these
20015questions no man is free from all interest. For though the want of good
20016sentiments may place him beyond the influence of moral interests, still
20017even in this case enough may be left to make him fear the existence of
20018God and a future life. For he cannot pretend to any certainty of the
20019non-existence of God and of a future life, unless--since it could only
20020be proved by mere reason, and therefore apodeictically--he is prepared
20021to establish the impossibility of both, which certainly no reasonable
20022man would undertake to do. This would be a negative belief, which could
20023not, indeed, produce morality and good sentiments, but still could
20024produce an analogon of these, by operating as a powerful restraint on
20025the outbreak of evil dispositions.
20026
20027
20028 [*Footnote: The human mind (as, I believe, every rational being must
20029of necessity do) takes a natural interest in morality, although this
20030interest is not undivided, and may not be practically in preponderance.
20031If you strengthen and increase it, you will find the reason become
20032docile, more enlightened, and more capable of uniting the speculative
20033interest with the practical. But if you do not take care at the outset,
20034or at least midway, to make men good, you will never force them into an
20035honest belief.]
20036
20037
20038But, it will be said, is this all that pure reason can effect, in
20039opening up prospects beyond the limits of experience? Nothing more than
20040two articles of belief? Common sense could have done as much as this,
20041without taking the philosophers to counsel in the matter!
20042
20043I shall not here eulogize philosophy for the benefits which the
20044laborious efforts of its criticism have conferred on human
20045reason--even granting that its merit should turn out in the end to be
20046only negative--for on this point something more will be said in the next
20047section. But, I ask, do you require that that knowledge which concerns
20048all men, should transcend the common understanding, and should only be
20049revealed to you by philosophers? The very circumstance which has called
20050forth your censure, is the best confirmation of the correctness of
20051our previous assertions, since it discloses, what could not have been
20052foreseen, that Nature is not chargeable with any partial distribution of
20053her gifts in those matters which concern all men without distinction
20054and that, in respect to the essential ends of human nature, we cannot
20055advance further with the help of the highest philosophy, than under the
20056guidance which nature has vouchsafed to the meanest understanding.
20057
20058
20059
20060CHAPTER III. The Architectonic of Pure Reason.
20061
20062By the term architectonic I mean the art of constructing a system.
20063Without systematic unity, our knowledge cannot become science; it will
20064be an aggregate, and not a system. Thus architectonic is the doctrine of
20065the scientific in cognition, and therefore necessarily forms part of our
20066methodology.
20067
20068Reason cannot permit our knowledge to remain in an unconnected and
20069rhapsodistic state, but requires that the sum of our cognitions should
20070constitute a system. It is thus alone that they can advance the ends
20071of reason. By a system I mean the unity of various cognitions under one
20072idea. This idea is the conception--given by reason--of the form of a
20073whole, in so far as the conception determines a priori not only the
20074limits of its content, but the place which each of its parts is to
20075occupy. The scientific idea contains, therefore, the end and the form of
20076the whole which is in accordance with that end. The unity of the end, to
20077which all the parts of the system relate, and through which all have a
20078relation to each other, communicates unity to the whole system, so that
20079the absence of any part can be immediately detected from our knowledge
20080of the rest; and it determines a priori the limits of the system, thus
20081excluding all contingent or arbitrary additions. The whole is thus an
20082organism (articulatio), and not an aggregate (coacervatio); it may grow
20083from within (per intussusceptionem), but it cannot increase by external
20084additions (per appositionem). It is, thus, like an animal body, the
20085growth of which does not add any limb, but, without changing their
20086proportions, makes each in its sphere stronger and more active.
20087
20088We require, for the execution of the idea of a system, a schema, that
20089is, a content and an arrangement of parts determined a priori by the
20090principle which the aim of the system prescribes. A schema which is not
20091projected in accordance with an idea, that is, from the standpoint of
20092the highest aim of reason, but merely empirically, in accordance
20093with accidental aims and purposes (the number of which cannot be
20094predetermined), can give us nothing more than technical unity. But the
20095schema which is originated from an idea (in which case reason presents
20096us with aims a priori, and does not look for them to experience), forms
20097the basis of architectonical unity. A science, in the proper acceptation
20098of that term, cannot be formed technically, that is, from observation
20099of the similarity existing between different objects, and the purely
20100contingent use we make of our knowledge in concreto with reference to
20101all kinds of arbitrary external aims; its constitution must be framed on
20102architectonical principles, that is, its parts must be shown to possess
20103an essential affinity, and be capable of being deduced from one supreme
20104and internal aim or end, which forms the condition of the possibility
20105of the scientific whole. The schema of a science must give a priori the
20106plan of it (monogramma), and the division of the whole into parts, in
20107conformity with the idea of the science; and it must also distinguish
20108this whole from all others, according to certain understood principles.
20109
20110No one will attempt to construct a science, unless he have some idea to
20111rest on as a proper basis. But, in the elaboration of the science, he
20112finds that the schema, nay, even the definition which he at first gave
20113of the science, rarely corresponds with his idea; for this idea lies,
20114like a germ, in our reason, its parts undeveloped and hid even from
20115microscopical observation. For this reason, we ought to explain and
20116define sciences, not according to the description which the originator
20117gives of them, but according to the idea which we find based in reason
20118itself, and which is suggested by the natural unity of the parts of
20119the science already accumulated. For it will of ten be found that the
20120originator of a science and even his latest successors remain attached
20121to an erroneous idea, which they cannot render clear to themselves, and
20122that they thus fail in determining the true content, the articulation or
20123systematic unity, and the limits of their science.
20124
20125It is unfortunate that, only after having occupied ourselves for a long
20126time in the collection of materials, under the guidance of an idea which
20127lies undeveloped in the mind, but not according to any definite plan of
20128arrangement--nay, only after we have spent much time and labour in the
20129technical disposition of our materials, does it become possible to view
20130the idea of a science in a clear light, and to project, according to
20131architectonical principles, a plan of the whole, in accordance with the
20132aims of reason. Systems seem, like certain worms, to be formed by a kind
20133of generatio aequivoca--by the mere confluence of conceptions, and to
20134gain completeness only with the progress of time. But the schema or
20135germ of all lies in reason; and thus is not only every system organized
20136according to its own idea, but all are united into one grand system
20137of human knowledge, of which they form members. For this reason, it is
20138possible to frame an architectonic of all human cognition, the formation
20139of which, at the present time, considering the immense materials
20140collected or to be found in the ruins of old systems, would not indeed
20141be very difficult. Our purpose at present is merely to sketch the plan
20142of the architectonic of all cognition given by pure reason; and we begin
20143from the point where the main root of human knowledge divides into two,
20144one of which is reason. By reason I understand here the whole higher
20145faculty of cognition, the rational being placed in contradistinction to
20146the empirical.
20147
20148If I make complete abstraction of the content of cognition, objectively
20149considered, all cognition is, from a subjective point of view, either
20150historical or rational. Historical cognition is cognitio ex datis,
20151rational, cognitio ex principiis. Whatever may be the original source of
20152a cognition, it is, in relation to the person who possesses it, merely
20153historical, if he knows only what has been given him from another
20154quarter, whether that knowledge was communicated by direct experience
20155or by instruction. Thus the Person who has learned a system of
20156philosophy--say the Wolfian--although he has a perfect knowledge of all
20157the principles, definitions, and arguments in that philosophy, as well
20158as of the divisions that have been made of the system, possesses really
20159no more than an historical knowledge of the Wolfian system; he knows
20160only what has been told him, his judgements are only those which he has
20161received from his teachers. Dispute the validity of a definition, and
20162he is completely at a loss to find another. He has formed his mind
20163on another's; but the imitative faculty is not the productive. His
20164knowledge has not been drawn from reason; and although, objectively
20165considered, it is rational knowledge, subjectively, it is merely
20166historical. He has learned this or that philosophy and is merely a
20167plaster cast of a living man. Rational cognitions which are objective,
20168that is, which have their source in reason, can be so termed from
20169a subjective point of view, only when they have been drawn by the
20170individual himself from the sources of reason, that is, from principles;
20171and it is in this way alone that criticism, or even the rejection of
20172what has been already learned, can spring up in the mind.
20173
20174All rational cognition is, again, based either on conceptions, or on
20175the construction of conceptions. The former is termed philosophical, the
20176latter mathematical. I have already shown the essential difference of
20177these two methods of cognition in the first chapter. A cognition may be
20178objectively philosophical and subjectively historical--as is the case
20179with the majority of scholars and those who cannot look beyond the
20180limits of their system, and who remain in a state of pupilage all their
20181lives. But it is remarkable that mathematical knowledge, when committed
20182to memory, is valid, from the subjective point of view, as rational
20183knowledge also, and that the same distinction cannot be drawn here as in
20184the case of philosophical cognition. The reason is that the only way
20185of arriving at this knowledge is through the essential principles of
20186reason, and thus it is always certain and indisputable; because reason
20187is employed in concreto--but at the same time a priori--that is, in pure
20188and, therefore, infallible intuition; and thus all causes of illusion
20189and error are excluded. Of all the a priori sciences of reason,
20190therefore, mathematics alone can be learned. Philosophy--unless it be
20191in an historical manner--cannot be learned; we can at most learn to
20192philosophize.
20193
20194Philosophy is the system of all philosophical cognition. We must use
20195this term in an objective sense, if we understand by it the archetype of
20196all attempts at philosophizing, and the standard by which all subjective
20197philosophies are to be judged. In this sense, philosophy is merely the
20198idea of a possible science, which does not exist in concreto, but
20199to which we endeavour in various ways to approximate, until we have
20200discovered the right path to pursue--a path overgrown by the errors
20201and illusions of sense--and the image we have hitherto tried in vain to
20202shape has become a perfect copy of the great prototype. Until that time,
20203we cannot learn philosophy--it does not exist; if it does, where is
20204it, who possesses it, and how shall we know it? We can only learn
20205to philosophize; in other words, we can only exercise our powers of
20206reasoning in accordance with general principles, retaining at the same
20207time, the right of investigating the sources of these principles, of
20208testing, and even of rejecting them.
20209
20210Until then, our conception of philosophy is only a scholastic
20211conception--a conception, that is, of a system of cognition which we are
20212trying to elaborate into a science; all that we at present know being
20213the systematic unity of this cognition, and consequently the logical
20214completeness of the cognition for the desired end. But there is also a
20215cosmical conception (conceptus cosmicus) of philosophy, which has always
20216formed the true basis of this term, especially when philosophy was
20217personified and presented to us in the ideal of a philosopher. In this
20218view philosophy is the science of the relation of all cognition to
20219the ultimate and essential aims of human reason (teleologia rationis
20220humanae), and the philosopher is not merely an artist--who occupies
20221himself with conceptions--but a lawgiver, legislating for human reason.
20222In this sense of the word, it would be in the highest degree arrogant to
20223assume the title of philosopher, and to pretend that we had reached the
20224perfection of the prototype which lies in the idea alone.
20225
20226The mathematician, the natural philosopher, and the logician--how far
20227soever the first may have advanced in rational, and the two latter in
20228philosophical knowledge--are merely artists, engaged in the arrangement
20229and formation of conceptions; they cannot be termed philosophers. Above
20230them all, there is the ideal teacher, who employs them as instruments
20231for the advancement of the essential aims of human reason. Him alone
20232can we call philosopher; but he nowhere exists. But the idea of his
20233legislative power resides in the mind of every man, and it alone teaches
20234us what kind of systematic unity philosophy demands in view of
20235the ultimate aims of reason. This idea is, therefore, a cosmical
20236conception.*
20237
20238
20239 [*Footnote: By a cosmical conception, I mean one in which all men
20240necessarily take an interest; the aim of a science must accordingly be
20241determined according to scholastic conceptions, if it is regarded merely
20242as a means to certain arbitrarily proposed ends.]
20243
20244
20245In view of the complete systematic unity of reason, there can only be
20246one ultimate end of all the operations of the mind. To this all other
20247aims are subordinate, and nothing more than means for its attainment.
20248This ultimate end is the destination of man, and the philosophy which
20249relates to it is termed moral philosophy. The superior position occupied
20250by moral philosophy, above all other spheres for the operations of
20251reason, sufficiently indicates the reason why the ancients always
20252included the idea--and in an especial manner--of moralist in that of
20253philosopher. Even at the present day, we call a man who appears to have
20254the power of self-government, even although his knowledge may be very
20255limited, by the name of philosopher.
20256
20257The legislation of human reason, or philosophy, has two objects--nature
20258and freedom--and thus contains not only the laws of nature, but also
20259those of ethics, at first in two separate systems, which, finally, merge
20260into one grand philosophical system of cognition. The philosophy of
20261nature relates to that which is, that of ethics to that which ought to
20262be.
20263
20264But all philosophy is either cognition on the basis of pure reason, or
20265the cognition of reason on the basis of empirical principles. The former
20266is termed pure, the latter empirical philosophy.
20267
20268The philosophy of pure reason is either propaedeutic, that is, an
20269inquiry into the powers of reason in regard to pure a priori cognition,
20270and is termed critical philosophy; or it is, secondly, the system of
20271pure reason--a science containing the systematic presentation of the
20272whole body of philosophical knowledge, true as well as illusory, given
20273by pure reason--and is called metaphysic. This name may, however, be
20274also given to the whole system of pure philosophy, critical philosophy
20275included, and may designate the investigation into the sources or
20276possibility of a priori cognition, as well as the presentation of the a
20277priori cognitions which form a system of pure philosophy--excluding, at
20278the same time, all empirical and mathematical elements.
20279
20280Metaphysic is divided into that of the speculative and that of the
20281practical use of pure reason, and is, accordingly, either the metaphysic
20282of nature, or the metaphysic of ethics. The former contains all the pure
20283rational principles--based upon conceptions alone (and thus excluding
20284mathematics)--of all theoretical cognition; the latter, the principles
20285which determine and necessitate a priori all action. Now moral
20286philosophy alone contains a code of laws--for the regulation of our
20287actions--which are deduced from principles entirely a priori. Hence the
20288metaphysic of ethics is the only pure moral philosophy, as it is not
20289based upon anthropological or other empirical considerations. The
20290metaphysic of speculative reason is what is commonly called metaphysic
20291in the more limited sense. But as pure moral philosophy properly forms a
20292part of this system of cognition, we must allow it to retain the name
20293of metaphysic, although it is not requisite that we should insist on so
20294terming it in our present discussion.
20295
20296It is of the highest importance to separate those cognitions which
20297differ from others both in kind and in origin, and to take great care
20298that they are not confounded with those with which they are generally
20299found connected. What the chemist does in the analysis of substances,
20300what the mathematician in pure mathematics, is, in a still higher
20301degree, the duty of the philosopher, that the value of each different
20302kind of cognition, and the part it takes in the operations of the mind,
20303may be clearly defined. Human reason has never wanted a metaphysic
20304of some kind, since it attained the power of thought, or rather of
20305reflection; but it has never been able to keep this sphere of thought
20306and cognition pure from all admixture of foreign elements. The idea of a
20307science of this kind is as old as speculation itself; and what mind does
20308not speculate--either in the scholastic or in the popular fashion? At
20309the same time, it must be admitted that even thinkers by profession have
20310been unable clearly to explain the distinction between the two elements
20311of our cognition--the one completely a priori, the other a posteriori;
20312and hence the proper definition of a peculiar kind of cognition, and
20313with it the just idea of a science which has so long and so deeply
20314engaged the attention of the human mind, has never been established.
20315When it was said: "Metaphysic is the science of the first principles
20316of human cognition," this definition did not signalize a peculiarity in
20317kind, but only a difference in degree; these first principles were thus
20318declared to be more general than others, but no criterion of distinction
20319from empirical principles was given. Of these some are more general,
20320and therefore higher, than others; and--as we cannot distinguish what is
20321completely a priori from that which is known to be a posteriori--where
20322shall we draw the line which is to separate the higher and so-called
20323first principles, from the lower and subordinate principles of
20324cognition? What would be said if we were asked to be satisfied with a
20325division of the epochs of the world into the earlier centuries and those
20326following them? "Does the fifth, or the tenth century belong to the
20327earlier centuries?" it would be asked. In the same way I ask: Does the
20328conception of extension belong to metaphysics? You answer, "Yes." Well,
20329that of body too? "Yes." And that of a fluid body? You stop, you are
20330unprepared to admit this; for if you do, everything will belong
20331to metaphysics. From this it is evident that the mere degree of
20332subordination--of the particular to the general--cannot determine the
20333limits of a science; and that, in the present case, we must expect to
20334find a difference in the conceptions of metaphysics both in kind and in
20335origin. The fundamental idea of metaphysics was obscured on another
20336side by the fact that this kind of a priori cognition showed a certain
20337similarity in character with the science of mathematics. Both have the
20338property in common of possessing an a priori origin; but, in the
20339one, our knowledge is based upon conceptions, in the other, on the
20340construction of conceptions. Thus a decided dissimilarity between
20341philosophical and mathematical cognition comes out--a dissimilarity
20342which was always felt, but which could not be made distinct for want
20343of an insight into the criteria of the difference. And thus it happened
20344that, as philosophers themselves failed in the proper development of the
20345idea of their science, the elaboration of the science could not
20346proceed with a definite aim, or under trustworthy guidance. Thus, too,
20347philosophers, ignorant of the path they ought to pursue and always
20348disputing with each other regarding the discoveries which each asserted
20349he had made, brought their science into disrepute with the rest of the
20350world, and finally, even among themselves.
20351
20352All pure a priori cognition forms, therefore, in view of the peculiar
20353faculty which originates it, a peculiar and distinct unity; and
20354metaphysic is the term applied to the philosophy which attempts to
20355represent that cognition in this systematic unity. The speculative part
20356of metaphysic, which has especially appropriated this appellation--that
20357which we have called the metaphysic of nature--and which considers
20358everything, as it is (not as it ought to be), by means of a priori
20359conceptions, is divided in the following manner.
20360
20361Metaphysic, in the more limited acceptation of the term, consists of two
20362parts--transcendental philosophy and the physiology of pure reason.
20363The former presents the system of all the conceptions and principles
20364belonging to the understanding and the reason, and which relate to
20365objects in general, but not to any particular given objects (Ontologia);
20366the latter has nature for its subject-matter, that is, the sum of given
20367objects--whether given to the senses, or, if we will, to some other kind
20368of intuition--and is accordingly physiology, although only rationalis.
20369But the use of the faculty of reason in this rational mode of regarding
20370nature is either physical or hyperphysical, or, more properly speaking,
20371immanent or transcendent. The former relates to nature, in so far as our
20372knowledge regarding it may be applied in experience (in concreto); the
20373latter to that connection of the objects of experience, which transcends
20374all experience. Transcendent physiology has, again, an internal and
20375an external connection with its object, both, however, transcending
20376possible experience; the former is the physiology of nature as a whole,
20377or transcendental cognition of the world, the latter of the connection
20378of the whole of nature with a being above nature, or transcendental
20379cognition of God.
20380
20381Immanent physiology, on the contrary, considers nature as the sum of
20382all sensuous objects, consequently, as it is presented to us--but still
20383according to a priori conditions, for it is under these alone that
20384nature can be presented to our minds at all. The objects of immanent
20385physiology are of two kinds: 1. Those of the external senses, or
20386corporeal nature; 2. The object of the internal sense, the soul, or, in
20387accordance with our fundamental conceptions of it, thinking nature.
20388The metaphysics of corporeal nature is called physics; but, as it must
20389contain only the principles of an a priori cognition of nature, we must
20390term it rational physics. The metaphysics of thinking nature is called
20391psychology, and for the same reason is to be regarded as merely the
20392rational cognition of the soul.
20393
20394Thus the whole system of metaphysics consists of four principal parts:
203951. Ontology; 2. Rational Physiology; 3. Rational cosmology; and 4.
20396Rational theology. The second part--that of the rational doctrine of
20397nature--may be subdivided into two, physica rationalis* and psychologia
20398rationalis.
20399
20400
20401 [*Footnote: It must not be supposed that I mean by this appellation what
20402is generally called physica general is, and which is rather mathematics
20403than a philosophy of nature. For the metaphysic of nature is completely
20404different from mathematics, nor is it so rich in results, although it
20405is of great importance as a critical test of the application of pure
20406understanding--cognition to nature. For want of its guidance, even
20407mathematicians, adopting certain common notions--which are, in fact,
20408metaphysical--have unconsciously crowded their theories of nature with
20409hypotheses, the fallacy of which becomes evident upon the application
20410of the principles of this metaphysic, without detriment, however, to the
20411employment of mathematics in this sphere of cognition.]
20412
20413
20414The fundamental idea of a philosophy of pure reason of necessity
20415dictates this division; it is, therefore, architectonical--in accordance
20416with the highest aims of reason, and not merely technical, or according
20417to certain accidentally-observed similarities existing between the
20418different parts of the whole science. For this reason, also, is the
20419division immutable and of legislative authority. But the reader may
20420observe in it a few points to which he ought to demur, and which may
20421weaken his conviction of its truth and legitimacy.
20422
20423In the first place, how can I desire an a priori cognition or metaphysic
20424of objects, in so far as they are given a posteriori? and how is
20425it possible to cognize the nature of things according to a priori
20426principles, and to attain to a rational physiology? The answer is this.
20427We take from experience nothing more than is requisite to present us
20428with an object (in general) of the external or of the internal sense;
20429in the former case, by the mere conception of matter (impenetrable and
20430inanimate extension), in the latter, by the conception of a thinking
20431being--given in the internal empirical representation, I think. As to
20432the rest, we must not employ in our metaphysic of these objects any
20433empirical principles (which add to the content of our conceptions by
20434means of experience), for the purpose of forming by their help any
20435judgements respecting these objects.
20436
20437Secondly, what place shall we assign to empirical psychology, which has
20438always been considered a part of metaphysics, and from which in our time
20439such important philosophical results have been expected, after the hope
20440of constructing an a priori system of knowledge had been abandoned? I
20441answer: It must be placed by the side of empirical physics or physics
20442proper; that is, must be regarded as forming a part of applied
20443philosophy, the a priori principles of which are contained in pure
20444philosophy, which is therefore connected, although it must not be
20445confounded, with psychology. Empirical psychology must therefore be
20446banished from the sphere of metaphysics, and is indeed excluded by
20447the very idea of that science. In conformity, however, with scholastic
20448usage, we must permit it to occupy a place in metaphysics--but only
20449as an appendix to it. We adopt this course from motives of economy;
20450as psychology is not as yet full enough to occupy our attention as
20451an independent study, while it is, at the same time, of too great
20452importance to be entirely excluded or placed where it has still less
20453affinity than it has with the subject of metaphysics. It is a stranger
20454who has been long a guest; and we make it welcome to stay, until it can
20455take up a more suitable abode in a complete system of anthropology--the
20456pendant to empirical physics.
20457
20458The above is the general idea of metaphysics, which, as more was
20459expected from it than could be looked for with justice, and as these
20460pleasant expectations were unfortunately never realized, fell into
20461general disrepute. Our Critique must have fully convinced the reader
20462that, although metaphysics cannot form the foundation of religion,
20463it must always be one of its most important bulwarks, and that human
20464reason, which naturally pursues a dialectical course, cannot do without
20465this science, which checks its tendencies towards dialectic and, by
20466elevating reason to a scientific and clear self-knowledge, prevents the
20467ravages which a lawless speculative reason would infallibly commit in
20468the sphere of morals as well as in that of religion. We may be sure,
20469therefore, whatever contempt may be thrown upon metaphysics by those who
20470judge a science not by its own nature, but according to the accidental
20471effects it may have produced, that it can never be completely abandoned,
20472that we must always return to it as to a beloved one who has been for a
20473time estranged, because the questions with which it is engaged relate
20474to the highest aims of humanity, and reason must always labour either
20475to attain to settled views in regard to these, or to destroy those which
20476others have already established.
20477
20478Metaphysic, therefore--that of nature, as well as that of ethics, but in
20479an especial manner the criticism which forms the propaedeutic to all the
20480operations of reason--forms properly that department of knowledge which
20481may be termed, in the truest sense of the word, philosophy. The path
20482which it pursues is that of science, which, when it has once been
20483discovered, is never lost, and never misleads. Mathematics, natural
20484science, the common experience of men, have a high value as means, for
20485the most part, to accidental ends--but at last also, to those which are
20486necessary and essential to the existence of humanity. But to guide them
20487to this high goal, they require the aid of rational cognition on the
20488basis of pure conceptions, which, be it termed as it may, is properly
20489nothing but metaphysics.
20490
20491For the same reason, metaphysics forms likewise the completion of the
20492culture of human reason. In this respect, it is indispensable, setting
20493aside altogether the influence which it exerts as a science. For its
20494subject-matter is the elements and highest maxims of reason, which form
20495the basis of the possibility of some sciences and of the use of all.
20496That, as a purely speculative science, it is more useful in preventing
20497error than in the extension of knowledge, does not detract from its
20498value; on the contrary, the supreme office of censor which it occupies
20499assures to it the highest authority and importance. This office it
20500administers for the purpose of securing order, harmony, and well-being
20501to science, and of directing its noble and fruitful labours to the
20502highest possible aim--the happiness of all mankind.
20503
20504
20505
20506CHAPTER IV. The History of Pure Reason.
20507
20508This title is placed here merely for the purpose of designating a
20509division of the system of pure reason of which I do not intend to treat
20510at present. I shall content myself with casting a cursory glance, from
20511a purely transcendental point of view--that of the nature of pure
20512reason--on the labours of philosophers up to the present time. They have
20513aimed at erecting an edifice of philosophy; but to my eye this edifice
20514appears to be in a very ruinous condition.
20515
20516It is very remarkable, although naturally it could not have been
20517otherwise, that, in the infancy of philosophy, the study of the nature
20518of God and the constitution of a future world formed the commencement,
20519rather than the conclusion, as we should have it, of the speculative
20520efforts of the human mind. However rude the religious conceptions
20521generated by the remains of the old manners and customs of a less
20522cultivated time, the intelligent classes were not thereby prevented from
20523devoting themselves to free inquiry into the existence and nature of
20524God; and they easily saw that there could be no surer way of pleasing
20525the invisible ruler of the world, and of attaining to happiness in
20526another world at least, than a good and honest course of life in this.
20527Thus theology and morals formed the two chief motives, or rather the
20528points of attraction in all abstract inquiries. But it was the former
20529that especially occupied the attention of speculative reason, and which
20530afterwards became so celebrated under the name of metaphysics.
20531
20532I shall not at present indicate the periods of time at which the
20533greatest changes in metaphysics took place, but shall merely give a
20534hasty sketch of the different ideas which occasioned the most important
20535revolutions in this sphere of thought. There are three different ends in
20536relation to which these revolutions have taken place.
20537
205381. In relation to the object of the cognition of reason, philosophers
20539may be divided into sensualists and intellectualists. Epicurus may be
20540regarded as the head of the former, Plato of the latter. The distinction
20541here signalized, subtle as it is, dates from the earliest times, and was
20542long maintained. The former asserted that reality resides in sensuous
20543objects alone, and that everything else is merely imaginary; the latter,
20544that the senses are the parents of illusion and that truth is to
20545be found in the understanding alone. The former did not deny to the
20546conceptions of the understanding a certain kind of reality; but with
20547them it was merely logical, with the others it was mystical. The former
20548admitted intellectual conceptions, but declared that sensuous objects
20549alone possessed real existence. The latter maintained that all real
20550objects were intelligible, and believed that the pure understanding
20551possessed a faculty of intuition apart from sense, which, in their
20552opinion, served only to confuse the ideas of the understanding.
20553
205542. In relation to the origin of the pure cognitions of reason, we find
20555one school maintaining that they are derived entirely from experience,
20556and another that they have their origin in reason alone. Aristotle may
20557be regarded as the bead of the empiricists, and Plato of the noologists.
20558Locke, the follower of Aristotle in modern times, and Leibnitz of Plato
20559(although he cannot be said to have imitated him in his mysticism),
20560have not been able to bring this question to a settled conclusion.
20561The procedure of Epicurus in his sensual system, in which he always
20562restricted his conclusions to the sphere of experience, was much more
20563consequent than that of Aristotle and Locke. The latter especially,
20564after having derived all the conceptions and principles of the mind
20565from experience, goes so far, in the employment of these conceptions and
20566principles, as to maintain that we can prove the existence of God and
20567the existence of God and the immortality of them objects lying beyond
20568the soul--both of them of possible experience--with the same force of
20569demonstration as any mathematical proposition.
20570
205713. In relation to method. Method is procedure according to principles.
20572We may divide the methods at present employed in the field of inquiry
20573into the naturalistic and the scientific. The naturalist of pure reason
20574lays it down as his principle that common reason, without the aid of
20575science--which he calls sound reason, or common sense--can give a more
20576satisfactory answer to the most important questions of metaphysics than
20577speculation is able to do. He must maintain, therefore, that we can
20578determine the content and circumference of the moon more certainly
20579by the naked eye, than by the aid of mathematical reasoning. But this
20580system is mere misology reduced to principles; and, what is the most
20581absurd thing in this doctrine, the neglect of all scientific means is
20582paraded as a peculiar method of extending our cognition. As regards
20583those who are naturalists because they know no better, they are
20584certainly not to be blamed. They follow common sense, without parading
20585their ignorance as a method which is to teach us the wonderful secret,
20586how we are to find the truth which lies at the bottom of the well of
20587Democritus.
20588
20589 Quod sapio satis est mihi, non ego curo Esse quod
20590 Arcesilas aerumnosique Solones. PERSIUS
20591 --Satirae, iii. 78-79.
20592
20593is their motto, under which they may lead a pleasant and praiseworthy
20594life, without troubling themselves with science or troubling science
20595with them.
20596
20597As regards those who wish to pursue a scientific method, they have now
20598the choice of following either the dogmatical or the sceptical, while
20599they are bound never to desert the systematic mode of procedure. When I
20600mention, in relation to the former, the celebrated Wolf, and as regards
20601the latter, David Hume, I may leave, in accordance with my present
20602intention, all others unnamed. The critical path alone is still open.
20603If my reader has been kind and patient enough to accompany me on this
20604hitherto untravelled route, he can now judge whether, if he and others
20605will contribute their exertions towards making this narrow footpath
20606a high road of thought, that which many centuries have failed to
20607accomplish may not be executed before the close of the present--namely,
20608to bring Reason to perfect contentment in regard to that which has
20609always, but without permanent results, occupied her powers and engaged
20610her ardent desire for knowledge.
20611