3 1822 01222 7831 W//////////M 'Fl j-e>3lD£C- 6 TllZ .E5 TEXT-BOOK TO KANT THE CRITIQUE OF PURE REASON. ^ «? % \A. ^V\Aw\OM0uJ^ TEXT-BOOK TO KANT THE CRITIQUE OF PUKE REASON: OTHKTIC, CATEGORIES, SCHEMATISM. TRANSLATION, REPBODUCTION, COMMENTARY. INDEX. WITH BIOGRAPHICAL SKKTUH. JAMES HUTCHISON STIRLING, LL.D. ; POREIQH MKMHKK OK THB PHILOSOPHICAL BOCIETX OF HHH.IX. NEW YORK: G. P. PUTNAM'S SONS, 27 and 29 WEST 23rd STREET. 1882. CONTENTS. Preface, BioGRArniCAL Sketch, Reproduction, Introduction, Book I. Apprehension — 1. Relation of Sense to Apprehension, 2. Relation of the Understanding to Apprehension, Book II. Judgment, Translation and Commentary : Introduction — I. Difference between Pure and Empirical Cognition, II. We do possess certain a priori Cognitions, III. Philosophy stands in need of a Science, etc., IV. Analytic and Synthetic Judgments, . V. A priori Synthetics are, etc., VI. General Proposition of Pure Reason, . VII. Idea and Divisions, Transcendental ^Esthetic — §1- §2. §3. §4. §5. Metaphysical Exposition of Space, Transcendental Exposition, Inferences, .... Metaphysical Exposition of Time, Transcendental Exposition, § 6. Inferences, PAOB ix xv 1 3 :n 44 78 T. C. .n, 115 345 117 348 119 35] 122 355 125 357 129 359 133 360 138 362 140 366 143 372 144 373 147 377 149 377 150 377 VI CONTENTS. Translation and Commentary — continued .• Transcendental iEsthetic — continued — § 7. Further Explanations, § 8. General Remarks, .... Transcendental Logic — Of Logic in General, etc., etc., . Book I. The Analytic of Notions, § 9. The Logical Function of Understanding § 10. The Categories or Pure Notions, . §11 §12 § 13. Principles of a Transcendental Deduction, § 14. Transition to Deduction, § 15. The Possibility of Conjunction, § 16. Synthetic Unity of Apperception, . § 17. The Axiom of it Ultimate Principle, § IS. Meaning of Objective Unity, . § 19. Logical Form of all Judgments, § 20. All Perceptions under the Categories, § 21. Remark, § 22. No other Function of Categories, . §23 § 24. Application of Categories, §25 § 26. Deduction of Use of Categories, § 27. Result of Deduction, The Deduction in its First Form, . Book II. The Analytic of Judgments, Transcendental Judgment, . Chap. I. The Schematism, . Chap. II. System of Primary Propositions, Sec. I. Ultimate Analytic Principle, Sec. II. Ultimate Synthetic Principle, Sec. III. System of Synthetic Primaries, 1. Axioms of Perception, . 2. Anticipations of Sense, 3. Analogies of Experience, A. Substance, B. Causality, C. Reciprocity, . TAGE T. c. 153 378 157 378 169 383 181 383 185 386 190 387 196 391 199 393 201 394 207 404 212 406 213 408 217 409 219 409 221 412 222 413 223 413 224 413 226 413 227 414 234 420 236 420 240 443 446 243 452 245 452 248 454 256 464 258 464 261 464 265 471 268 482 273 484 282 486 288 490 294 490 316 507 CONTENTS. Vll Translation and Commentary — continued: System of Synthetic Primaries — continued- 4. Postulates, . General Remark, . Appendix: I. Pen-in-Hand Analysis, II. From the Prolegomena, III. From the Logic, Index, .... PAGE T. C. 323 500 336 515 517 518 531 540 543 PREFACE. It may be desirable to premise a word in regard to the contents of this book, or even perhaps its name — "Text-book." We all feel that an effect must have a cause : that is, there is assumed to be a necessary connexion between them. Still, we take it for granted, as well, that matters of fact (which any case of causality seems really to be) are always contingent and never neces- sary : their contraries, implying no contradiction, are to be acknowledged possible ; as, for example, the sun rises and sets, but it might do neither. Hume, now, pointed out the discrepancy here, and asked, — The relation of cause and effect being matter of fact, why do we, inconsistently, assume it to be necessary? Kant, again, who took the question to himself, an- swered, — It is to understate the question, to confine it to causality ; substance and accident, action and re-action, are as necessarily connected as a cause and its effect ; and the entire interest, rather, relates to a general system— a general system of necessary con- nexion, necessary synthesis, or say synthetic necessity, even in matters of fact. This system, now, — while it is all that, properly X PREFACE. and peculiarly, is constitutive either of or with Kant (anything else, unless the categorical imperative, being either only negative and regulative, or simply a corol- lary), — is what is here exhibited, with the fullest details, and in no less than a threefold form. For such reason (indicated) it is, then, that this book is named, Text-booh to Kant The Translation has been executed with every care; and notes have been added in explanation or correc- tion of the text. Existing translations have neither been now referred to, nor at any time, indeed, either used or read (those principally in vogue, however, having years ago been sufficiently consulted, legiti- mately to warrant a general judgment). It has been the aim of the Commentary to leave as few obstacles as possible between the reader and a full understanding of the transcendental system of Kant. Perhaps the very word transcendental may henceforth carry, in general, a somewhat saner sense than seems currently in use at present. In supplement both of translation and commentary (which, reasonably, are based on the second edition 1 ), pertinent extracts will be found to have been made from the first edition of the Critique of Pure Reason, as also from the Prolegomena, early pen -in-hand sum- maries, etc. In the Index there has been a very special endea- 1 The special text that underlies this volume is (what is most gener- ally found here) that of Kant's collected works at the hand? of Rosen- kranz and Schubert : the other edition, Hartenstein's, I do not happen to have seen. I had a little old Graetz reprint (1795) of the second edition of the K. of P. R. beside me as well. PREFACE. XI vour to provide the reader with a complete referential guide. The Reproduction, that, by way of introduction, precedes the translation, will be found to cover the whole ground occupied by the rest of the book. Executed in 1862, it is all that exists of the " Exposi- tion of Kant" which is several times mentioned in the Secret of Hegel as prior to that work. Perhaps some of my friendly correspondents will — so far in response to their wishes — be pleased to see this ex- position at last. I only regret, as I write this, that Mr Lewes, who was one of the most earnest of these, is not now alive to honour it with his perusal. It is the continued interest in Kant, as well as my recent, somewhat intimate occupation with the subject, that has led to publication at length. In the Biographical Sketch, taking a day as usually spent by Kant, and filling into the distribution of it salient, characteristic expressions of his gathered from his whole works, some little has been attempted, as well in indication of the man, as of the peculiarity of his modes and subjects of thought generally. Biographical Sketch. BIOGRAPHICAL SKETCH. It is five o'clock of a winter's morning in the year 179-, and, prompt to the minute, a stiff, erect, old-soldier of a servant briskly throws up the door of a small sleeping-room, — where never the sun shone, nor a fire entered, — with the words, " Herr Professor, die Zeit ist angekommen ! " Nor does the Professor addressed neglect the call. Sitting up at once, he considerately divests himself of the carefully-calculated appli- ances of the night. The Professor is Kant, the caller his inexorable old servant Lampe, and the scene the philosopher's simple bed-closet, in his simple dwelling-house, in the remote and winter- dismal Königsberg. Once dressed, the Professor smokes his one daily pipe of tobacco, and (without eating) drinks his two daily cups of tea. Were it a matter of choice with him, he would prefer coffee ; but he finds it heating, and he relinquishes it. Neither is it as a matter of sense that he inhales the fragrance of the weed at this so early morning-hour. No ; that to him is the pain and penalty of a nauseous and confusing duty. He is but a little man, the Herr Professor, hardly more than five feet in height, small-boned, fleshless, meagre, thin, evanescent as a shadow ; and these awkwardly gulped tobacco-fumes are but the medicinal nauseant with which philosophy would clear and cleanse the mala pitaita from its poor, sunken, narrow, little chest. Not, however, that, in this case, to be always frail is ever to be actually ill. Against that, plainly, there are subtler expedients in use than even the matutinal nauseant ; for there is a cheerful red on the cheek, there is an alert eye in the head of the little philosopher. Blue, loving, and true, those eyes of his at once touch all men irresistibly into affection and respect. Not large, the head is handsome XVI BIOGRAPHICAL SKETCH. enough ; and the face under it has the look in it somehow of a simple, garrulous, half-arch, old-maidenish man. His pipe ended and his tea swallowed, Kant now prepares himself for his seven o'clock lecture — seven o'clock under the murky lights of oil or tallow in the remote, cold, dark, wintry Kcenigsberg ! Ah, but his students listen to him, for they know him. They know his honest, loyal, absolutely true nature — an absolutely true nature that is conscientiousness itself. With all his gifts and solid acquirements, too, they know him to be the most modest and retiring of mortals, and yet the man who could declare, with his heart on his lips. that " did any man propose to him, even in his last moments, a good action, him he would thank." They see him there, and they cannot choose but listen, though they must even strain the ear to catch the low voice with which he seems to think before them. He tells them this morning, in that honest ramble of unaffected talk, of the faculties by which logic is realized, and of the hostile in- fluences to which they may be exposed. If we consider our cognitions, he says, in regard of the two essentially different faculties (sense and understanding) from which they derive, we come upon the distinction between perceptions and notions (Anschauungen and Begriffe). Sense brings forward the mere stuff or material for thinking ; it is understanding disposes it and redacts it under rules or notions ; for in every cognition there must be distinguished matter and form. The distinction between aesthetic and logical cognition is identical with that between perception and notion, intuition and discourse. Symbolical cognition is not opposed to intuitive cognition (cognition through sinnliche Anscliav. ungen), but to intellectual or discursive cognition (cog- nition through Begriffe). Eeflection is the internal act (spon- taneity) that realizes a Begriff, a Gedanke ; but apprehension, again, is that other internal act (receptivity) that realizes a Wahr- nehmung, an empirische Anschauung. Our Ego has a certain duplicity : internal sense depends on an a posteriori perceptive complex ; apperception, true self-consciousness, is but a reflec- tion, a thought, a logical point, and devoid, consequently, of all matter of contents : the one is not the other ; reflection is not apprehension, nor this that. The one is pure, the other em- BIOGRAPHICAL SKETCH. XV11 pirical, etc. It is not to be supposed, however, that the Ego itself changes and differs as its states happen to change and differ ; for it is only by reference to its identity that their difference is recognised. A sound understanding, a practised judgment, a comprehensive reason, constitute the entire com- pass of the intellect. The servant, under formal orders, need only understand ; the agent of a special duty, with merely general directions to guide him, can but judge ; but the chief, who has to anticipate cases and find the rule for them, must reason. He who is deficient in wit is an obtitsum caput, a stumpfe Kopf, a blockhead. Deficiency of judgment, but with wit, is stultitia, insulsitas, Albernheit, silliness, scatterbrained- ness ; without wit, it is stupiditas, Dummheit, what charac- terizes the dolt. He who is incapable of learning, or into whom nothing can be got, — he, like an untempered blade, is hebes, dull, einfältig, a dunce. He who can only imitate is a Pinsel, a shallowpate ; while he who can originate is a Kopf He who sacrifices the substantial for the unsubstantial, as home-comfort for glitter abroad, is a Thor, a fool ; but the Thor that is offensive is a Narr, a beast 1 His lecture over, Kant now returns to his abode, and occupies himself with his studies till it wants a quarter of one. Warned by his housekeeper, he then springs up with alacrity, and hastens to deposit on the dining-table, for his own consumption at a prescribed moment, a small glass of rum carefully covered with paper. That effected, he dresses for dinner. This, in the twenty-four hours, is his only meal ; and he eats it largely (for him), and with enjoyment. It usually consists of three dishes with dessert and two bottles of wine. Kant, now, like the wise man he is, throws off the harness of the intellect, and dons the lounging-coat of the body. But he will not, like a mere brute, only crunch his bones in solitude ; he has always at least two, and sometimes five, to dine with him. He will have a rational zest to his meal in the company and discourse of his fellows. Once at table, Kant directly sets himself to put his guests at their ease : he even salutes or rallies them in the homely provincial of the district. Kant is only a Professor ; but he has lived in the houses of the distinguished : he feels up to a 1 Kant fills whole pages, in work after work, with distinctions like these. b XV111 BIOGRAPHICAL SKETCH. thing or two ; he will be quite the man of the world, and not a cuistre. The discourse at table must be light, then, and such as gently to entertain and stimulate. Kant has his own expedients, indeed, for keeping the ball up ; for he hates the mortal agony of a pause. Is it not curious, he remarks, that poltroon should be only a contraction consisting of the first syllables of the words pottex truncal its ? Or he resorts to a mild conundrum and asks, Why a woman ought to be at once like and unlike the town-clock or a snail ? Like, to be sure, as correct to the minute ; but unlike, as not proclaiming everything that happens. Like, again, as keeping to a house of her own ; but unlike, as never carrying her all on her back ! The French feel beauty, he points out again ; but it is the English who are open to the sublime. And is it not strange that, whereas the French commonly like the English, the English, on the contrary, as commonly despise the French. But it is the commercial spirit does this, and not possibly the mere rivalry of neighbours, as England is quite well aware of its own indisputable superiority. And here now the thought of other nations suggests phy- sical geography to him. He cannot help referring to some of the most interesting facts that have reached him. There is such a thing as a milk-white sea, he says ; you have that at the Moluccas. The English and Scotch differ from the High- landers, as being very delicately brought up ! Negroes are born white, all to a ring round the navel. The ibis dies the moment it quits Egypt. The lion is so noble, he will not put a paw upon a woman. He is not afraid of the crowing of a cock, but he runs away from a snake or a fire. The marrow of his bones, when dried in the sun, is so hard that you may strike a light with it. The water at the Cape is so pure that it remains sweet when brought to Europe. If you make a cup of the rhinoceros's horn, any poison will splinter it. A tree in Congo has its leaves and its bark both poisonous, but the one is the antidote to the other, take which you will first. In the Canary Islands there is the tree of life that never rots, whether in the ground or in water. There is a mussel in Italy that gives out so much light, you can see to read by it. In Languedoc there is a hot spring that hatches eggs, but its water on the fire comes much slower a-boil than ordinary BIOGRAPHICAL SKETCH. XIX water. A petrifying spring at Clermont has actually made an ordinary stone-bridge over a river. Wild beasts only eat Negroes in Gambia, and leave Europeans alone. The Negroes in America are immensely fond of dog's flesh, and all the dogs bark at them. 1 By-and-by Kant takes an opportunity to tell his guests what he knows about Swedenborg. In the end of the year 1761 Swedenborg was called to a certain Princess whose great good sense and excellent understanding made it all but im- possible for her to be deceived. She had heard many strange things told of the visions of the man, and wanted to convince herself of the truth of the matter by a trial of her own. Swedenborg came to her, and, after they had conversed to- gether for some time, she commissioned him to deliver a secret message to the spirit of one of her departed friends. In a few days Swedenborg was once more ushered into her presence. The lady said, Well ? and Swedenborg, stooping, whispered into her ear a word or two which drove the blood from her face and chilled her to the marrow. What he had whispered was true, she said, and it could have been com- municated to him only by the dead. Madame Hauteville, widow of an Envoye from Holland to the Court of Stockholm, was summoned, some time after the death of her husband, to pay an account to the goldsmith Croon, which she was morally convinced a man of her late husband's punctual and orderly ways must already long ago have settled. The sum concerned was a considerable one, and in the trouble and anxiety of the circumstances she was induced to speak to Swedenborg, who happened to be then at home. Swedenborg cheerfully undertook to carry a mes- sage to her dead husband, and bring her his answer on the point. Accordingly, three days afterwards, he presented himself at Madame Hauteville's, when it chanced that she was entertaining company. In his cold way he announced to the lady that he had seen and spoken with her departed husband, who assured him that the money had been paid, and that the receipt was then lying in a certain desk. Here- upon the lady exclaimed that that very desk had been com- 1 Almost an infinite number of such stories are to he found in Kant, and all gravely propounded ! XX BIOGRAPHICAL SKETCH. pletely cleared out and its contents thoroughly overhauled and carefully examined, but that not a vestige of any such paper had been discovered. Yes, said Swedenborg, but your husband intimates that, if you take out a certain drawer on the left side, you will see a board, which you must push away, and then you will find a concealed shelf with the Secret De- spatches of the Government on it, and beside them the receipt of Croon for payment of the plate. At these words of Sweden- borg's the widow ascended to the room where the desk in ques- tion lay, followed by her whole company, in whose presence all proved itself to be exactly as Swedenborg had described. There is the story, too, of how Swedenborg, when at Gothen- burg, saw the conflagration that was then raging at Stock- holm (some two hundred and fifty miles off, and with the whole breadth of Sweden between them), and threatening his own dwelling-house there, and of how, after watching it for some hours, he became composed at last, and said the fire had stopped at such and such a building : all which, too, proved itself true. In fact, the friend from whom the last state- ment directly came knew intimately all the best people in Gothenburg, and he spent two months there directly investigating a matter that was then recent. While, as for the first statement, it was communicated by a personal friend, who had been directly present to the whole transaction. 1 Turning to science now, Kant tells his guests of his fancies about the construction of the universe. He shows them, too, how, in process of time, the resistance of the tides must bring the rotation of the Earth to a stop, with consequent de- struction of the planet. Many evolutions and revolutions, he continues, — petrifactions, crystallizations, organizations, — must have preceded the advent of man. The thought of an affinity in things such that, in consequence of it, they must have originated either, genus after genus, from a single primitive one, or, as it were, from a single generative mother's lap, leads to ideas so monstrous that reason shudders back from them. 2 There is a wonderful power in the mind to master the body, 1 Kant actually heaps a number of circumstantial particulars together which almost seem to authenticate these wonderful stories. 2 The anticipations, as above, of Herschel and Laplace, of Thomson and Tait, are known to everybody ; but not so that assonance, let it be BIOGRAPHICAL SKETCH. XXI is the next remark, and Kant expatiates on all his thousand and one little theories about health and disease. The plan, on the whole, is, he says, not to disturb nature ; for the shortest-lived are just those who are for ever striving to set death at defiance. Health and the duty to one's self suggesting to Kant the question of temperance, he cannot help exclaiming that a man in a state of drunkenness is only a beast : the false hap- piness and spurious freedom from care which result from artificial stimulants can only end in that manner. And yet, something is to be said on the other side ; for how wine, short of intoxication, promotes good fellowship and open-hearted free communication between man and man — nay, how it has even the merit to foster virtue itself! Did not old Cato himself, according to Cicero and Horace, feed the flame of his integrity with wine ? Hume, too, liked his bottle of port with his rubber of whist ; and he actually execrates the guest who cannot forget in the morning the events of the night. It is remarked of women, priests, and Jews, that they do not get drunk; for people believe in them from the outside. Ah, one might marry were it not for that wart on the nose or that gap in the teeth ! But now the sitting has almost reached the very extremest limit that Kant endures, and, to the satisfaction of all, the company breaks up and departs. Kant, too, sallies out for his walk. Day after day it is over the same ground, and at a certain part of it he must pay toll to certain beggars, whom his own benevolence has gathered for him. Past the beggars, Kant can resign himself to the course of his own thoughts. The sight of his legs, as he walks, suggests to him his own oft-repeated original observation that white stockings do more justice to one's calves than black ones. And, so, what a thing is seeming ! How often does not studied obscurity that plays the part of depth obtain the credit it seeks ! But to save of what force it may, to Darwin. Here, too, is a curious anticipation of Bulwer. Did we take up at nights the threads of our dreams where we left them off when we awoke, it is possible that we should come to think we lived in two different worlds. And this concerning 07ie novel, we may refer to another such. Kant's milk-white sea, as lately men- tioned, reappears in The Green Hand. 12 XXII BIOGRAPHICAL SKETCH. the ship we must give the whale a barrel to play with. It does not do to be bashful, or put a great weight on what folks think of us. Just to fancy now that Hume — the fine and gentle Hume — was a great four-square man, and yet was bashful, or, in his own language, blaet (our blöde) ! he speaks very meaningly of the horror that attends the break-down of the first attempt to speak with the due assurance. 1 The season suggests now that the South Pole is colder than the North Pole : there is more land north, and the sun remains there in summer eight days longer. Strange, how, in northern lati- tudes, though no wood grows, there is plenty drifted. In Siberia they consider the devil only ; in Heaven, they say, God is too far off; but it is the devil rules here. The vul- garest smut or the stupidest practical joke will be hailed by the common man with quite as much joy as Kepler may have felt over some discovery his share in which he would not have exchanged for an entire principality. The smallest insect that springs pushes the earth back. A dog is old when a man is scarcely out of his boyhood ; and the cedars of Lebanon are but middle-aged when its firs have long since perished. Perhaps five or six thousand years are but a day in the life of the earth. Possibly, therefore, the earth was several thousand years in existence before any life appeared on it. No chance, however, or mere physical cause can produce an organism : give me all the matter in the world, and I could not make a caterpillar. There are certainly pro- visions, at the same time, in view of new conditions : birds get additional feathers when transferred to colder climates. But qucelibct natura est conservatrix sui. America, at its hottest, cannot grow a Negro, or even a Hindoo, of its own. What should show the origin of plants and animals would be a science for the gods who were there at the time. How it darkens ! Ah, day is beautiful, but night is sublime. Yes, the English — that is what newspapers do : it is scarcely possible that there can be any nation where understanding is so universal, even among the lowest classes, as is the case with England. 2 Eeturning from his walk, Kant now seats himself for the • Of course Kant is not responsible for the etymology here. - These propos are rather miscellaneous, perhaps, but there are thou- sands of others the like, in which Kant was pleased to indulge himself. BIOGRAPHICAL SKETCH. XX111 evening by the stove in his room ; and, fixing his eye on the point of a church-spire which he sees out of window, he runs over in mind — what was his work — his System. I am so far from considering metaphysic worthless or unnecessary, he thinks, that, on the contrary, ever since I have seen into its nature and the place proper to it in the circle of our knowledge, I am convinced that even the true and eternal weal of the human race depends upon it — an estimate that may appear to everybody else extravagant and wild. But it comes to this, one ought to ask : 1. Am I right in my distinction between analytic and synthetic judgments, as well as in my statement of the nature and value of the latter when a priori — that they constitute, namely, the founda- tion of metaphysic ? 2. If it be true that we cannot syn- thetically decide upon anything a priori, unless in reference to such formal conditions, whether with regard to perceptions of sense or notions of the understanding, as precede, experience, and render it possible — such experience as this of ours, that is ? 3. Whether, therefore, finally, all speculative cognition a priori possible for us, though granting necessary existence to unknown things in themselves {noumencC), avails to reach only phenomena, mere appearance to sense, and in this way leaves room for a natural dialectic, which, being understood and seen into, we are immediately at home as regards the true nature and limits at once of our knowledge and hopes ? l For, any cognition, to deserve the name metaphysical, can- not be a posteriori (matter of fact, then, and, consequently, con- tingent), but must be a priori. ISTow, there is only sense and understanding ; and the former being the special seat of the a posteriori, it must be the faculties of the understanding alone that can concern the a priori. That is, our faculties themselves will contribute to experience, even as we receive it through sense, certain elements of their own. But a notion alone never gives an actual or real cognition. We may con- 1 ceive centaurs, harpies, gorgons, chimseras, and fancy all manner of new principles or new senses ; still these are but notions and empty — they give no knowledge. Suppose we have at any time a notion only, what is it we usually do to 1 Only the above paragraph has actual words of Kant in regard : what follows is equally true to Kant, but only to be named reproduction. XXIV BIOGRAPHICAL SKETCH. realize it ? We turn to experience, we turn to the object of it, we try this object, to which the notion itself is due ; and, in this way, learning fact after fact, we fill the notion and convert it into a reality. But, this being, it would appear that, in the case of a 'priori notions, it must, once for all, be quite impossible to attain to any a priori knowledge that shall be worthy of the name. A priori, there is no turning to experience, to special sense, possible. Let us, then, have even an infinitude of a priori notions, they must, as mere notions, and without filling, prove worthless — mere possible fancies, dreams, not positive cognitions, realities. That is what the want of the special senses brings about for us. If we have a priori notions, it is an indispensable necessity that we should have a medium of sense to apply to for such material as shall impregnate them with meaning — give them sense, as we say — render them objectively real. But can a medium of sense be, by any possibility, a priori ? Why, sense is precisely what is a posteriori. Sense and the a pos- teriori are, in fact, convertible terms : they are identical. Each is the other ; the latter is the former, and the former the latter. It is manifestly impossible and absurd — an actual contradiction in terms — to speak of an a priori sense. And without it, any other a priori — even if true — must be and can only be a mere meaningless region of impalpable ghosts — ilomos vacuus et inania regno. But we have spoken of the possibility of the faculties them- selves contributing elements to, so to speak, the bolus of ex- perience ; why, then, should we not suppose this of the faculty of sense ? It is true that each of the five special senses yields, in odours, savours, colours, etc., only what is simply the a pos- teriori proper. Each odour, savour, colour, what not, is a mere sensation, a mere feeling, subjective affection, that has to be waited for — something, consequently, that can only be a pos- teriori. Still, are there no general forms which may be com- mon to them all, and, as common to them all, only possibly due to the very machinery of sense itself as machinery ? May not what we may call general sense, namely, have forms of its own, actual sensuous forms, — even, so to speak, actual material forms ? In sense, besides the mere feelings contributed by the sensations, there is such a thing as perception. We do not only BIOGRAPHICAL SKETCH. XXV feel in sense, but we also perceive — that is, we discern a manifold, a complex of units, which units we set together into the one object, and can even count. Now, that in sense that holds of perception, let sensation be as it will, may possess actual a priori forms, actual a priori objective forms — forms, that is, which add themselves objectively, even as objective units, namely, to the other units of the manifold or complex that is to be the object perceived. Any such forms must be absolutely general, and alone absolutely general: are there any such ? Why, all things whatever — that is, all special sensations whatever — are in time and space. If we want absolutely general forms, surely these are they. They are absolutely universal, absolutely unexceptive, absolutely neces- sary, absolutely infinite ; and there are no others. But time and space are themselves perceptible manifolds : time is a series of perceptibly succeeding moments ; space, though infi- nite, is as a stereoscopic whole of length, breadth, and thickness — it is a perfect aggregate of discernible units. Here, then, is a perceptive, a sensuous matter, with which we may, in two different ways, fill notions. As a priori, too, it is in a position to fill a priori notions. After all, then, it is only these latter we want now — accepting, as we do at once, time and space as a perfectly satisfactory medium of a priori sense. Well, then, it is logic shall yield us the a priori notions. Logic is an established pure science, complete in all its parts, and as pure (in that it relates to the mind only), it is also perfectly a priori. This science, in its classification of judgments, now, under quantity, quality, relation, and modality, yields us at once the entire a priori tree of the pure functions of judg- ment; and judgment is identical with the understanding, judgment is identical with apperception, judgment is identical with consciousness, with what I call the Ego, my self. Here, now, then, are all our difficulties removed, and the way completely cleared for us. A notion to have reality, meaning, or, as we say, sense, must have a sensible complex filled into it ; but then, again, a sensible complex, if to have connexion, unity, objectivity, must have a notion into which it may collapse, and become henceforth a one perceived or experi- enced entity. These stones, and edges of lime, and all the angles and colours in them, are but a disjunct, unintelligible chaos XXVI BIOGRAPHICAL SKETCH. until I have fused them all together into the single articulate entity and identity I call wall. Wall, now, is a perfectly general notion, though, of course, empirical or a posteriori in origin. Wall, too, though a notion, has become, as in amal- gamation with those sensuous materials, actually perceptive. The manifold of sensation not more than the unity of notion, mediates the perception. And this is general. There never can be formed perception, finished perception, unless a notion has added itself to the sensation ; though, of course, there may be conceived consciousness of the units of what we may call crude objectivity (time and space) without the help of a notion ; but that we should name crude perception, perception as perception, and, in that sense, perception proper. Now, with such considerations before us, it is easy to see that, possessing a priori notions and a priori matter of sense, we possess, also, all the ingredients necessary for the construc- tion of a priori objects. These a priori objects we shall call schemata; and the schemata, as resulting from the a priori action of a priori function (notions, the categories) on a priori affection (general sense, time and space), will prove so many checkers for reception into objectivity and necessity of the mere subjective contingency of our special senses in their various sensations ; which, as such, are always, evidently, our own states, our own mere feelings. And here we see the answer to Hume as regards the ques- tion of causality. What is the reason, he asked, of the neces- sary connexion we attribute to all actual examples of it ? Now, he, for his part, assured that he had only matters of fact before him, could fall on no rationale but reference to natural instinct on custom. We, for our parts, again, assured of the presence of necessity of synthesis between elements that are absolutely alien the one to the other, have become awake to this, that the rationale desiderated cannot, as Hume supposed, lie in any element of experience, but must be sought for in a source that is absolutely (not relatively) a priori. The answer, in fact, is this. The nexus of antecedent and con- sequent is a function of judgment, a function of apperception, an a priori notion, and its action on a certain modus of time is such as to determine this modus into a schema, a species, a simulacrum of the law of causality, which, reflected into all BIOGRAPHICAL SKETCH. XXV11 actual cases of cause and effect, insinuates into these that apodictic necessity which we predicate of them. But it is now ten o'clock, and the inexorable Lampe appears to put his master to bed. Kant, as his principles are, can only obey. In the little sunless, tireless bed-closet which we have already seen, he lays him down on his little bed accurately adjusted to a prescribed angle — he lays him down and in a carefully calculated position which he stoically preserves. Lampe covers him up, and wraps him in, on those strictly scientific principles which have been laid down for him. The philosopher is then left to his well-won repose. This he pur- sues, like everything else, steadily. His lips are firmly closed, and he breathes through his nostrils only. 1 Then his thoughts, they, too, must only wander in an authorized tract. He " shuts his eyes " (as his phrase is) to any too importunate thought, so that gradually such confusion of the ideas springs as is akin to dream. With realization of all these subtle calculations and appliances, then, it is not wonderful that — after due work, exercise, relaxation, and with his single meal now, perhaps, even perfectly assimilated — the wise little body, that allows itself only seven hours of sleep, should fall gently into the most natural and healthiest of slumbers — slumbers that are only disturbed at five o'clock of the following morning by the inexorable Lampe in the way we have seen. Such now has been for years, and will remain for years, the philosopher's daily course of life. Immanuel Kant was born at Königsberg, April 22, 1724, and died there February 12, 1804. His parents were pious, respectable people in humble life ; his father a saddler of the name of Cant, and by descent Scotch. It was Immanuel himself made the name Kant, in order to preclude the pro- nunciation Zant. His course in life was, on the whole, the usual one of a German student who would provide for himself by the pursuit of letters. He had the usual training at school 1 Talk of the anticipation of Laplace or of Thomson, why here is another, even more curious than of Bulwer and the Ch'een Hand. Mr Catlin startled the whole kingdom the other day hy the recommendation to keep our mouths shut . And Kant knew it all, and practised it all a hundred years ago ! XXVlll BIOGRAPHICAL SKETCH. and college. At the usual years, he was the usual house-tutor. It was only later than usual, however, that he seems to have qualified himself as the usual university teacher ; and he was no less than forty-six years of age when he came into port at last as a Professor. As teacher and professor — and he was eminently successful as either — he wrote many works, of which the principal ones are these : First of all, the works which are his work, the Three Great Kritiken, namely ; to which his Pro- legomena, his works on Morals and Law, his Metaphysic of Nature, and his Strife of the Faculties may be regarded as only supplementary. The Anthropology is very interesting, and the Logic must be read. His Eeligion within the Limits of Pure Keason has been widely influential, and is the respective corollary to his philosophy. The student has much to gain also by a perusal of his shorter essays, as his review of Herder, his Progress of Metaphysic, On Philosophy, etc. In fact, no one writing of Kant that appeared after 1781 can well be neglected. Before that date, too, there are many interesting- papers — but they are all to be found named in their order in a very useful chronological table at page 211 of the eleventh volume of the Works. It is now — 1881 — exactly a hundred years since the publi- cation of the first great Kritik ; and there can be no doubt that, at this moment, the place of Kant, as generally esti- mated, is that of greatest German philosopher, greatest modern philosopher, greatest philosopher at all with only the usual exceptions of Plato and Aristotle. Nor can there be any doubt that a like estimate will continue for some considerable time yet. Kant, in truth, was a man of a supremely active, tena- cious faculty. One might almost say that the drawing of distinctions lay in his very blood. But it must be said, too, that, in the sort of elephantine simplicity and naivete' of his countrymen, another of his characteristics is superfetation. In common with them, namely, he has the distinct drawback of seeing so very deeply through the millstone as actually to witness the gnomes of the universe at work. His reading- had, evidently, been wide and general, but not, perhaps, pre- cisely deep. His character as a man has been already, to some extent, depicted. In that respect, and every other re- spect, he was, and always will be — der ehrliche Kant. The Reproduction. THE REPRODUCTION. Introduction. State of the question : Descartes, Leibnitz, Locke, Hume. The general problem. The necessary conditions of its solution, and consequent distribution of the whole inquiry. Descartes, as Leibnitz after him, held by innate ideas. Locke controverted the term innate, and as- sumed, as well, all our ideas to originate in the sensa- tion or reflection of experience. For even reflection was to him but the mind's own further experience in manipulating the experience that was already due to sensation. If sensation were external sense, then, said Locke, reflection " might properly enough be called internal sense." " When," from these sources, " the understanding is once stored with simple ideas, it has the power to repeat, compare, and unite them, even to an almost infinite variety, and so can make at pleasure new complex ideas ; but it is not in the power of the most exalted wit, or enlarged under- standing, by any quickness or variety of thought, to invent or frame one new simple idea in the mind, not taken in by the ways aforementioned." Hume now, on the one hand, only accentuated these positions; but, on the other, he drew from them their natural consequences — consequences which Locke, for his part, had been so far from foreseeing that he had 4 TEXT-BOOK TO KANT: even reasoned, in excess of his principles, to their very opposites. It is part of these proceedings of Hume which we have now, in the first place, to see. The word idea with Locke is the Vorstellung of the Germans. It is " the most general expression for all that is present to mental consciousness ; " and it is quite as applicable to the products of sensation as to those of reflection. Hume altered this. He discrimi- nated between these products, — naming the former impressions, and only the latter ideas. Impressions (Enquiry, sec. ii.) are " all our more lively perceptions, when we hear, or see, or feel, or love, or hate, or desire, or ivill:" ideas "are the less lively perceptions of which we are conscious when we reflect on any of those sensations or movements." It is important to observe that what we generally call feelings (love, hate, desire, etc.) are here ranked with sensations as equally impres- sions. Hume will regard sense as but a single func- tion — from whatever side, whether ab intra or ab extra. Or there is to Hume, just as there was for Locke, an internal as well as an external sense; the former, moreover, even as a sense, being quite on a par with the latter. It seems to be his belief, namely, — and a belief, as seen, apparently shared by Locke, — that, to a being like man, an internal sense, calculated to take note of the successive empirical states of the inner subject, is as necessary as an external sense which shall bear to refer to an outer object. In which case, too, it must be seen that inner sense is, as a sense, to be strictly distinguished from self-con- sciousness, or the apperception of the ego. The contents of the former are all the transient states of the em- pirical subject when under sensuous feeling ; whereas those of the latter are but the simple I, a mere intel- lectual act, the bare thought /, /, /, or / that am here THE REPRODUCTION. 5 and now thinking (das "ich denke''' 1 ). It is also impor- tant to observe further that, by the term impression, Hume means only the mental state, and as simply felt, without reference to any supposed impressing stimulus: it means simply — with total suppression of the con- sideration of an agent — any cognised affection of sense, inward or outward, as such. We may state the chief points in the modified position thus: 1, Sensation is the source of all ele- ments of knowledge ; 2, There is internal as well as external sensation ; 3, Sensation externally is not more product of a sense than sensation internally; 4, What to us are the ideas of our thoughts, are, in reality, only copies of our sensible impressions. To these we may add, 5, That, for knowledge, we are shut in to our own subjective states of affection or impression : " nothing can ever be present to the mind but an image or perception — this house and that tree are nothing but perceptions in the mind " {Enquiry, sec. xii. part i.) And now the consequences of the general theory become all too manifest. If all our ideas are only copies of our impressions, then we know nothing whatever that, substantially, is not the product of sensible experience. But we have no sensible ex- perience of the hidden principles of things — we have no sensible experience, indeed, of the actual things themselves; for, by the very terms of it, sensible experience is but a consciousness of affection — affection set up in us we know not how : it is but a subjective feeling (light, sound, fragrance, etc.), and it is impossible to pass beyond it. Further, we have no sensible experience of God, or of the immortality of the soul, or (strictly) of the freedom of the will. That is, the whole business of meta- b TEXT-BOOK TO KANT : physic — metaphysic proper, metaphysic with its fore- court of ontology — is summarily sisted. But what, in that case, of experience itself? And here Hume is at once struck by the extraordinary fact that the authority of experience depends on causality, while that of causality, again, depends on experience— a circle of support which, of course, lies only in the air. These findings Hume subjects, as he believes, to a most searching scrutiny, and with no other result than — instinct apart, and so far as any intelligible principle is concerned — a relega- tion of the whole authority of both causality and experience to habit or custom. Two matters of fact that we have been in the habit of seeing; together mutually suggest each other ; and thus the terms in the whole series of experience, which we implicitly believe strung on a necessary and universal law, are really combined and held together — accreted or agglutinated, as it were— by nothing whatever but that mere customary or habitual suggestion which, in all cases, follows the simple frequency of asso- ciation. A few quotations will make clear the nature and process of Hume's thought here. (I omit the quotations, and simply extract the following passage from article, " Philosophy of Causality : Hume and Kant," page 186) : — " Hume's proceedings are these. His first reference is to this, that neither generally nor specially is causation a quality. It is impossible to point to any ' one quality which universally belongs to all beings, and gives them a title to that denomination ' (cause). Equally impossible is it to find in any particular cause any particular quality by which it is the cause it is. ' No object ever discovers, by the qualities which appear to the senses, either the causes which produced it, or the effects which will arise from it.' But if causality THE REPRODUCTION. 7 be not a quality, it can only be a relation. And this relation examined, we find all that is representative of it to be only the conjunction in time and place of the cause with the effect. We do assume, and no doubt rightly, this conjunction to be a necessary connexion; but no minutest investigation can demonstrate to the understanding the ground or reason of this necessary connexion. Depending on a matter of fact, this reason is no affair of either intuition or demonstration ; but, even as a matter of fact, this reason is not capable of being seen and understood, whether before production of the effect, or even after deliberate observation of that event. Impression as the original from which the idea of necessary connexion is copied, there is none to be found, unless simply custom from repetition of the association. If besides custom there is any- thing else to be taken into account, it is that reflection of vivacity from present impression to idea of absent object which is called belief. I hear a voice from the next room. That voice has always been conjoined in the past with a certain person. The custom of that conjunction suggests this person as the cause of this voice. The actual impression of this voice reflects its own vivacity to the person suggested. This person, suggested with all this reflected vivacity, is believed in ; or belief in the actual existence of this person, suggested by the voice through customary conjunction, is reflected into the mind from the vivacity of the actual impression. Nature, it is true, attributes to causality a tie of necessity ; but philo- sophy, for its part, can find no representative for that tie but the mere custom of repeated mental association. The tie ascribed by nature to causes and effects themselves cannot be found, so far as philosophy goes, to lie in them, but in us. It is only ' so far as causation is a natural relation, and produces a union among our ideas, that we are able to reason upon it, or draw any inference from it : ' i we infer a cause immediately from its effect ; and this inference is not only a true species of reasoning, but the strongest of all others.' Nevertheless, ' as a philosophical relation/ causation implies only ' contiguity, succession, and constant conjunction.' Hume's whole theory, point by point, is contained in what has just been said ; and the reader will have no difficulty in verifying this, whether from the Treatise or the Enquiry. 8 TEXT-BOOK TO KANT : " It is Hume's own object to refer our belief in causality to instinct. A principle so necessary to us he openly vindicates for infallible nature as against our fallible faculties. But this was for Beattie and others precisely their own conclusion. Causality was to them, too, an implanted first principle, an instinct ; and when they advanced as much as against Hume, they advanced only what Hume himself similarly advanced. Instinct here, in fact, was rather the sceptic's than the dog- matist's affair." But now is there no reply to Hume ? The word instinct, as we see, is none such. But again, whether on the one side or the other, to announce this instinct as the conclusion, is, in all cases, to leave the original difficulty precisely where it was. Let it be a fact that we cannot but think the A B of causality as necessary ; is it, then, to explain this fact, simply to name it — call it instinct? Is it not still true to say that there is here a mental inference which defies philosophy to account for it? This, at all events, we assume, for our part, to be the state of the question ; and it now belongs to us to ask, Is it final ? Is it to be confessed at last that there are no such things as fixed principles in human knowledge ? This the result to which alone he seemed to labour, surely it is to be said that Hume himself hoped against it. "Metaphysics and morals," he directly sa}'s, " form the most considerable branches of science ; mathematics and natural philosophy are not half so valuable." It is he himself, too, assures us that " the motive of blind despair can never reasonably have place in the sciences, since, however unsuccessful former attempts may have proved, there is still room to hope that the industry, good fortune, or improved sagacity of succeeding generations may reach dis- coveries unknown to former ages The like has been performed with regard to other parts of nature ; THE UEPRODUCTION. 1) and there is no room to despair of equal success in our inquiries concerning the mental powers and ceconomy, if prosecuted with equal capacity and caution." These last words are suggested by reflec- tion on the science of astronomy and the extraor- dinary perfection to which it has been brought by the genius of a single thinker, Newton alone. And, certainly, that is to be said, that such inquiries as lie now before us have never yet been guided into the highway of science. In that there is no bringing of the collaborators to agreement : the one thinks he has hit the road here, the other there ; and follow which we will, we are presently at a loss. It is not so with logic, nor mathematics, nor, since Bacon, Galilei, Torricelli, and Stahl, with natural philosophy. These, in that they have reached, each its own open high- way, are now sciences, and need only move onward. Nor is it difficult to find the reason of this. The men we have named, with others the like, "comprehended at last that reason only gets to understand what she herself planfully creates ; that she must precede, consequently, with principles of her judgment accord- ing to constant laws, and actually compel nature to answer her inquiries." Metaphysic, as said, " has not as yet had the same good fortune, nor hit the highway ; not but that she is older than all the other sciences, and would remain even if these were bodily swallowed up in the maw of an all-devouring bar- barism " (Kant, WW., ii. 669). This, then, if we would raise metaphysic into a science, is what we have to do. We must look about us for principles, and, by their aid, put reason's self to the question. Now what if these should be found precisely in those proceedings by which David Hume would seem to have brought the whole interest to a 10 TEXT-BOOK TO KANT : dead-lock ? Is it simply credible that the entire synthesis of experience should be the product of habit? The causal nexus, then, is not objective but subjective ? The effect, that is, is attributed to the cause, not because of an objective reason that will be alike for everybody and perceivable by everybody, but because of a subjective reason in consequence of a habit that has force, and can have force, only for myself? With such course of reflection before us, surely we cannot help asking, Is this true? When the sun shines upon it, I expect a stone to warm. But is this expectation only a matter of habit? When I see the shining of the sun on the stone, is it only by habit that I think of the warmth ? When I think of the arrow, I think of the bow ; but is the arrow the cause of the bow? When I think of the moon, I think of the sun ; but is the moon the cause of the sun ? When I think the letter A in the alpha- bet, I think also B, and C, and D, etc. ; and in numbers, when I think 1, I think as well 2. and 3, and 4, and 5, etc. In these cases, it is quite certain that there has been an habitual conjunction as far back as my memory carries me. But I have never considered the one the cause of the other. I feel that the nexus, in such a case, is one of custom, is one of habitual association, and that the reason for the conjunction lies in me, and not in the letters or the ciphers themselves — that it is subjective, and not objective. " Mrs Shandy," says Sterne, " could never hear the house-clock wound up, but the thoughts of some other things unavoidably popped into her head ;" but we are not for a moment led to suppose that Airs Shandy regarded the winding-up of the house-clock as the cause of those " other things." Habitual associa- tion is a nexus, then ; but it is not the nexus of THE REPRODUCTION. 11 causality. But take the principle in its absolutely general form — -take the proposition of causality itself, Every change must have a cause. What is the nature of the evidence here ? Merely subjective — an associa- tion that I feel I cannot help in consequence of habit? Or is it objective — bearing on a truth which I hold to be universal and necessary, valid for everybody, valid in itself — a truth which I intellectually perceive, and which I know everybody else will similarly perceive ? Surely the nature of the evidence, the truth, here, must be called apodictic. The universality involved is not comparative merely ; it is absolute. A change must have a cause. This is not only true because it is true, but because it must be true, and because its opposite is manifestly impossible. The necessity is rigorous; the universality is unexceptive. A change has a cause ; a straight line is the shortest possible : compare the necessity and universality of the propositions. A change may be without cause ; a straight line may not be the possible shortest : compare the absurdity and impossibility of the two contradictions. Is not the authority of the evidence in both of the original propositions equally stringent ? The law of causalit} r , then, is an apodictic truth ! But this is very strange. We have been taught to believe that apodictic evidence is confined to relations of ideas ; and here seems a truth of an apodictic nature in what is, even glaringly, a matter of fact — a mere affair of experience! But how can that be possible? Experience tells us that something is so and so, but never that something must be so and so. How, then, can experience tell us that a change must have a cause? If we be right as regards the, probable evidence of ex- perience, it is not from experience that we can possibly 1 2 TEXT-BOOK TO KANT : derive this truth. And yet it is easy to see that, with- out experience, we could never have the notion of a change. Change is an empirical fact ; it is derived wholly a posteriori: how, then, can it possibly be con- nected with evidence apodictic, necessary, universal, such as attaches to the a priori alone of mathematical science? Well indeed might Hume be startled by the fact ; and no wonder that he asked, how could apo- dictic truth attach to a manifest principle of experi- ence? His answer, habit, custom, from frequency of association, is now manifestly incompetent. But can we find a better ? What, then, if it should turn out that Hume's con- siderations are only there to bring the general interest to a crisis ? And, first of all, in regard of the answer desiderated, is causality the only empirical principle that is so situated? The very question is a flash. If there be other such principles, it is only reasonable to suppose that they, one and all of them, will have a common ground ; which ascertained, there will be a consequent advance at once to something equally new and impor- tant, something that will surely constitute one of the main pillars of human reason. Here at last we gain a glimpse of the possibility of metaphysic, — here at last, that is, we have come to the palace-gate " der Königin n aller Wissenschaften " (the queen of all the sciences), and the royal matron will no longer complain, forsaken and forlorn, like Hecuba, "Modo maxima rerum, tot generis natisque potens — nunc trahor exul, inops." 1 David Hume, then, when he brought us to the nexus of causality, shall have brought us also to the 1 Literal allusion to a sentence in the first preface to the K. of P. R. Kant appends to the quotation, with all the touching exactness of a German, " Ovid. Metam." ! THE REPRODUCTION. 13 very porch of the sanctuary; and it is for us, com- pleting the roll of all such principles, to give ourselves entrance with it into the very body of the edifice, taking it at long and last into an easy and a full pos- session. The question, then, is, Are there any other such principles as this of causality — principles, that is, at once empirical and apodictic — what we shall presume to name principles transcendental f 1 Now, where is it that this principle of causality is used — where is it generally to be found? Perhaps in the same neighbourhood we shall find others the like. But causality is one of the principles of general physics. Let us turn up the ordinary treatises on such subjects, and examine the leading propositions laid down as principles there. What is this, for example : — Through- out all its changes the original quantity of matter is neither lessened nor increased ? What are we to say to a proposition of that nature ? Here, again, some- thing is spoken of that can only be known by experi- ence, and yet an assertion is made respecting it of a strictly apodictic nature. We perceive the univer- sality, the necessity of the proposition, the moment we understand it ; or, what is the same thing, we per- ceive then the impossibility of its opposite or contra- dictory. We are conscious, too, that the nexus here is not, and cannot be, an affair of habit. Again, In all physical phenomena, action and reaction are equal. We have in this also apodictic evidence con- joined with elements of an empirical nature. Here then, now, we have at least three propositions that seem to rest on experience, and yet imply apodictic 1 Transcendent is an object beyond experience. Transcendental applies to an object that is in experience, but yet of a validity that is beyond ex- perience. Kant's cpiestion of Quid juris is addressed wholly to that validity, of which the causal nexus is an example. 14 TEXT-BOOK TO KANT : certainty. The question, consequently, comes natur- ally, How is this? Above all, how can apodictic certainty attach to elements of an empirical nature? Are we quite certain that we stated the truth when we affirmed that experience declared only that a thing is, not that it must be ? No, in that we cannot be wrong ; experience brings to us facts, and we know thus that they are, but never that they must be. Sight tells us that there is such and such a colour, and again such and such a colour, as hearing that there are such and such sounds, or smell that there are such and such odours, or taste that there are such and such flavours, or touch that there are such and such feels. But all that I perceive in these or any such circum- stances is that the facts are once for all so, without the slightest appearance of any reason being present in them to necessitate the so. The colour that is here mio-ht be the colour that is there, or the sound that was then might have been the feel that is now — and all this without the smallest contradiction. But the facts that enter the mind through any sense are all constituted in a similar manner. They all enter; the fact, then, of their being is acknowledged, but not the reason that necessitates their being, not the reason that renders it impossible for them not to be. So it is with the inner sense ; we recognise all the mind's successive empirical states : they bring with them the fact of their existence, but not the necessity of their existence. The apodictic element in the propositions in question, then, cannot be referred to experience ; it must be referred to some other source ; and the question is, To what other? Nay, is it not simply impossible that there can be such other, even in con- sequence of the conclusion which we have just seen established at the hands of both philosophers, Locke THE REPRODUCTION. 15 and Hume, that all our knowledge is due to experience alone? This it is, without doubt, that has been the obstruction to Hume, and the occasion of his resortin o- to a subjective principle of explanation, habit. But we have sufficiently seen the inadequacy of this ex- planation, and must obtain another. It is evident, then, either that all our ideas do not arise from experience, or that we possess no such thing as an apodictic truth. But, apart from all con- sideration of any proposition immediately before us, we do possess apodictic truths : the mathematics and kindred branches found on such, contain such. Nay, Hume himself admits this. Under the phrase, rela- tions of ideas, he alludes to a vast aggregate of ideas that are either intuitively or demonstratively cer- tain — that are " discoverable by the mere operation of thought" — that " would for ever retain their cer- tainty and evidence," though objects corresponding to them never existed in nature. How, then, can he have possibly reconciled himself to this duality in knowledge, and yet have believed that all knowledge was due to experience? The fact that these others were relations of ideas, and not qualities of existent objects, seems to have been latently and half-con- sciously the reason of his acquiescence without special inquiry. Perhaps, also, it appeared to him that mathematical truth was of an analytic nature, and flowed deductively, by expansion, under guidance of the principle of contradiction, from original defini- tions in which it lay from the first involved and implied. For it is to be acknowledged that the pro- ducts of all such analytic procedure are of an apo- dictic nature, and their contraries would imply a con- tradiction. This, then, it was that probably occurred to Hume, though obscurely, in explanation of the 1 6 TEXT-BOOK TO KANT : vast distinction that he observed and asserted to sub- sist between relations of ideas and matters of fact. But that, probably, was his misfortune as well. All, doubtless, would have issued very differently with him, had he but questioned the source of apodictic truth in the mathematics. So questioning, he would have been led to the consideration of such truth generally. x 1 In the Treatise Hume devotes a whole "part" to the consideration of mathematical reasoning ; and this, as usual, is not unrepresented in the re-cast of the Enquiry. No full student of his, then, can think of Hume as neglecting mathematics. Still a reader of the Enquiry alone might regard the mathematical allusions there as only casual ; and this might have been the case with Kant. I do not recollect of any direct quotation from Hume in the Kritik ; but at page 6 of the Prolegomena we have in a note what bears to be a verbal quotation from " Versuche, 4ter Theil, Seite 214, deutsche Uebers." Now the Prolegomena was published only once, Rosenkranz says, and that was in 1783. The German trans- lations of Hume mentioned in any book beside me are these : — Treatise, Jakob, 1790 ; Essays, Tennemann, 1793 ; Enquiry, Tennemann, 1793 ; Dialogues concerning Natural Religion, Schreiter, 1781 ; Enquiry, Sulzer, 1755. It thus appears that the translation necessarily used by Kant ("Versuche," etc., as above) is, so far, omitted from mention. I have not as yet been able to verify in Hume the quotation actually made by Kant from these " Versuche ; " but, supposing the latter to have used translations only, it is evidently fair to hold it probable that he was not a reader of the Treatise, and that he might, consequently, very well talk of Hume neglecting mathematics. (Quotation occurs "Essay 17.") It was no failure, then, at least to think of mathematics that led to Hume's failure to reflect on the contradiction, which lay in the apodictic validity of relations of ideas, to the conclusion that all knowledge was due to experience, which, for its part, was competent to no more than proba- bility. Whether he thought of mathematical truths being analytic only is another question, and one which I am hardly disposed to answer affirmatively. Relations of ideas, it appears to me, must have been thought of by Hume only in that jumbling sort of reference to complex ideas as inventions of the mind, which we find in Locke. We have seen this already in the quotation on our first page as concerns the power of the understanding to " make at pleasure new complex ideas even to an almost infinite variety;" and there is a passage in Hume (Enquiry, sec. v. part ii.) which, almost verbally identical with the entire citation from Locke, similarly ascribes to the mind a power of compounding ideas " to all the varieties of fiction and vision." We know, too, that Locke, in answer to the Bishop of Worcester, who objected that the idea of sub- THE REPRODUCTION. 17 We have seen, then, that there are propositions of an apodictic nature, and yet apparently of an em- pirical origin ; while there are others no less apo- dictic, but evidently independent of experience. How is this ? How is it that truths or facts occupying fields so entirely different should yet possess evidence of identical stringency ? But, for a moment, we must here divert attention to this, that propositions fairly analytic are also apo- dictic. The reason is obvious. We have no occasion, in such cases, to resort to experience for the know- ledge in question : that knowledge we attain by a mental operation, without any trial of what actual experience will teach. That is, analytic propositions are of an a priori nature, meaning by that the pro- cess of ascertainment by simple operation of the mind, though on grounds, it may be, previously stance can be derived neither from sensation nor reflection, averred "that general ideas enter the mind neither through sensation nor through reflection, but are creations or inventions of the understanding." Reid ( Works, p. 276) describes Locke's process in formation of the idea genus " till at last it becomes an abstract general idea," with powers, evidently, all its own. And the same sort of unconscious conviction is to be found in Hume again and again. Abstract ideas are to him, as complex ideas are, mere entia rationis, and their relations may, without any contradic- tion whatever to his inferences from experience, be as apodictic as they may. Relations of ideas refer, he says, always to quantity and number, and these are abstractions. Consequently, he is not at all led rightly to reflect by " the bold determinations of the abstract sciences." Even in concrete matters, " geometry assists," he says, only by giving us, through abstraction, " just dimensions." It is a great advantage to mathematical science that it can show its objects. " An oval is never mistaken for a circle," etc. ; " though there never were a true circle or triangle in nature." To Hiune there are really none such in nature, and so neither are there in nature those apodictic relations. The true circle is a fiction of the mind; and so are all its necessary qualities. He simply forgets the ellipses of the planets, and the triangles by the aid of which our mathematicians mete the heavens. Hume, in fact, like Locke, had very obscure ideas of the powers of mental abstraction; and it is at least doubtful whether the one or the other ever thought, in that connexion, either of analysis or of synthesis. B 18 TEXT-BOOK TO KANT: t established by experience, without any a posteriori reference whatever to actual trial then and there. In this there is no difficulty. The truth is developed under leading of the principle of contradiction. That is, the reason that we give for any affirmation in this situation, is that it must be so and so, else it would contradict itself; and the principle in ques- tion is simply that an idea, proposition, judgment, must not contradict itself. But if anything is true, and must be true, otherwise it would contradict itself, it is evident that we have in hand something of an apodictic nature ; and this, evidently, must be the result of all pure operations of the mind alone. As such, these operations are a ]mori, and the principle of their process is that they must not contradict them- selves ; and in this way it is evident that they must be apodictic. Illustration will make this plain. All bodies are extended. This proposition is, so to speak, an ana- lytic apodictic. The notion of extension, that is, is already contained in the notion body ; and the pro- position itself, consequently, simply must be true — true universally — else, as is evident, the notion body would be self-contradictory ; its constituent quality, namely, being at once affirmed and denied of it. As much as this results, too, entirely by operation of the mind ; or it is by such operation, and not by actual trial of some actual body then and there, that insight into the truth is attained. All bodies are divisible, is again an analytic proposition that is also a priori and apodictic ; resting for its truth on process of mind, and not on process of experiment; for the divisibility, under penalty of infringing the law of contradiction, follows from the extension which is necessarily involved in the very notion of bodies or a body. Indeed, it will be THE REPRODUCTION. 19 evident now that any complex notion whatever may be similarly expanded with consequent possibility to produce an even infinite number of propositions which, as analytic, are, however ultimately, at least proximately a priori and consequently apodictic ; for the analysis that extends the insight, or constitutes the fulcrum of the predication, is, not a trial by sense, an experiment, an experience, but an action of intellect that, without contradiction, carries its own identity throughout. Now, for us this is a most important consideration ; for it meets at once a whole host of possible objec- tions to our proceedings on the threshold. It has been our object, namely, to signalize the apodictic nature of the proposition of causality. Now, so far, an opponent might, having recourse to analysis, adduce against us a quite infinite number of apodictic propositions which, nevertheless, were, in the end, only due to experience. So it is that we have, in the first instance, eliminated all possible analytic proposi- tions. It is our desire, namely, to confine attention to propositions which, while apodictic, are at the same time, also, not analytic, but synthetic. It is this now which, as our main interest, we proceed to explain. The peculiarity of the proposition of causality, we say, is that, so to speak, it is an apodictic synthetic. This is the true universal and philosophical expression of Hume's problem, which is thus extended into a much more general reach. For the question now is not of a single proposition (causality), but of every proposition that founds on a priori synthesis. And we have already seen other such propositions besides that of causality : those of action and reaction, for example, and the permanence of substance. We say a 'priori synthesis ; for apodictic synthesis, as not pos- sibly due to experience (the evidence of which is 20 TEXT-BOOK TO KANT : always only contingent or probable), must, as we have already seen, indeed, be at least proximately a priori. But what now if it must be even ultimately a priori? The apoclictic necessity and universality that rest on analysis may be reduced to grounds in the end which are only empirical. But can this be the case with the apodictic assignments that rest on synthesis? There is, of course, a synthesis which is due to expe- rience ; but just because it is so due, it can never be apodictic. No direct evidence of experience — and a synthesis of experience is, by the very terms of it, necessarily direct — can ever be more than contingent. Any apodictic synthetic, consequently, must and can only found its peculiar validity on a principle of nexus, that is, not proximately, but^ ultimately and absolutely a priori. The full, exact, and completely general expression for Hume's problem, therefore, is at last this : How are apodictic synthetic propositions possible ? When we say all bodies are extended, we predicate one notion (extension) of another notion (body), which former notion already lay in said latter notion — as we term it, impliciter, not expliciter. But when we say, a straight line is the shortest possible between any two points, we predicate one notion of another notion where the first was not already contained in the second, whether impliciter or expliciter. Straightness is a quality, namely, and, as such, is alien to a con- sideration of quantity (shortest). Every change has its cause, again. Here the notion cause is really something quite other than the notion change. " The effect," says Hume, " is totally different from the cause, and consequently can never be discovered in it*' {Enquiry, sec. iv. part 1). The proposition of causality, therefore, is synthetic, and quite as much THE REPRODUCTION. 21 synthetic as that of the straight line. But now we have to consider, further, that all empirical proposi- tions are, if we may say so, to the very core synthetic ; the very nerve of their nexus is synthesis. We might almost see this in the very terms, experience, trial, etc. ; for they import, in their very selves, that actual sense-examination of the express matter of fact has been the means of adding to one notion as subject another notion as predicate. Take, for example, All bodies are heavy. Here heavy is by no means a pre- dicate that is involved or implied in the bare notion of body. A body having extension, but no weight, would not contradict our general notion here. Such notion, consequently, would not be self-contradictory. That all bodies are heavy, then — it is not by analysis or mental operation, it is not a priori in any degree, that we see into the truth in this case. On the con- trary, it is only in consequence of experience, of actual trial, or, what is the same thing, it is only a posteriori, that we come to express any such sentence. How synthetic propositions spring from experience, then, will now be manifest ; but it will now also be mani- fest that all such, resting for proof on experience or mere perception of sense, must be devoid of strict universality and rigorous necessity, and can possess only a comparative universality and necessity, the force of which will never exceed, logically, the experience on which it is founded. And now we can be at no loss to understand the nature of that peculiarity which distinguishes the proposition mainly before us. The attribution of an effect to a cause is a synthetic pro- position ; but, while synthetic, it is also apodictic. Being apodictic at the same time that it is not analytic, it cannot be of an origin empirical or a posteriori, but must depend at last on a mental operation, a process 22 TEXT-BOOK TO KANT : of mind which, of necessity, also, can only be a priori The other propositions which we have seen in the same connexion, permanence of matter, action and reaction, are of an identical nature; and we are again brought to the question of the origin of a priori syn- thesis, as the true general expression for the problem of Hume. This question, now, is the single question which constitutes the special and particular interest of our whole general inquiry. For said inquiry, another leading consideration, too, is this. It is quite evident, from all that has been said, that these a priori synthetics will be the result of the mind itself, of the intellectual faculties them- selves. Now, this will prove decisive of the distribu- tion and general procedure of our inquiry itself. For if the truths in question flow from the mind itself, from our intellectual powers, it is evidently by an analysis of these latter, respectively in their order, that the former will discover themselves. In this way, too, we already provide our enterprise, in the signi- ficant show of anterior probability, with a gage and guarantee of success. For this is plain : that, if know- ledge be a combination of elements, which elements are partly from without and partly from within, it will, in these latter elements, necessarily possess an a priori material — an a priori material which has been contributed by the mental faculty itself in the per- formance of its proper function on the matter presented to it for that purpose from without. It is very specially important that we should be aware of this, that the faculty concerned is an intel- lectual faculty, a cognitive faculty, the faculty by which we acquire knowledge. We are not to figure mere passive sensation here, but, on the contraiy, active perception. For only so will it be that the THE REPRODUCTION. 23 mental contribution will bring with it the force and conviction, the necessity and universality, together with the insight, of a reason. In short, we shall pre- sently learn that the faculty mainly concerned is that of judgment, and judgment is, par excellence, the faculty that discerns. The principle contributed by the mind, then, the a priori element that, in percep- tion, is added to the a posteriori element, must not be viewed as of the same nature as an instinct. This principle, this element, is not to be called, with the Scotch philosophers, an implanted first principle, an original principle of the very constitution of the mind, an innate and instinctive tendency. We are not to say, as they, that we are so constituted that we can- not think otherwise, etc. This was the answer that all the Scotch actually bawled to Hume ; and, after all, it was Hume's own ! 1 Hume, in fact, has no object unless to show that, for our expecting the future to resemble the past, we can allege not the shadow of a reason, and that it is only through an instinct we anticipate the recurrence of customary conjunctions. If Hume's instinct here differs from that of Reid and the rest in the same reference, it is only in the need of a customary conjunction to excite it. Hume does not himself so correlate his instinct with his custom or his custom with his instinct ; but perhaps it is not illegitimate to suppose as much. In that case Reid's instinct will only differ from Hume's in being direct, while Hume's, as requiring custom to call it into action, will only be indirect. The instinct, however, as an instinct, is not more blind with Hume 1 Of course I hold him not to have known Dugald Stewart when I say this for Kant. For it is a fact that everybody that in this country has come after Eeid (I do not speak of Beattie or Oswald), namely, Stewart, Brown, etc., have, on the whole, taken their causality — very absurdly — from Hume. 24 TEXT-BOOK TO KANT : than it is with Reid himself. It is to be understood, then, that our principle is by no means of this nature ; it is no blind instinctive tendency, but an act of in- tellectual insight, capable of asserting, explaining, and justifying itself by argument or reason. And thus, as Hume's call was simply for this argument or reason, we shall be able, it is to be hoped, to give a satisfac- tory answer to his question at last. Our inquiry, therefore, is capable of being explained in another manner than in that with reference to the origin of apodictic synthesis, or how apodictic syn- thetic propositions are to be conceived possible. We can say, namely, that the question it involves is, In receiving the material of sense, does the intellect, even in the act of receiving, add anything ? In that case it would be easy to understand that any object pre- sented to us must be a compound, a compound which, even as objectively, sensuously, perhaps externally there, contains in it elements quite as much from the within of intellect as from the without of sense; quite as much from the seeing activity of the one as from the blind passivity of the other. Alexander explains the (TvvoXov of Aristotle to be to ko.6 eKacrrov aicrOrjrou Kai ace ; and so, those who have treated of it have generally found it convenient, or for their interests necessary, to enlarge its contents b} 7 adding to them much foreign, though perhaps cognate, matter. And here we have in mind only universal or elementary logic ; naturally exclud- ing from view the particular logic that may precede (not but that it always comes last in time) any par- ticular whole of inquiry, as directing and guiding its general disposition and conduct. It has been the interest, then, of many writers to eke out the scanty pages of general logic with certain additions. For 32 TEXT-BOOK TO KANT : instance, many have filled out their books with pre- liminary matter bearing on psychology ; and many, again, have displayed a like industry with reference to those empirical circumstances which impede or promote the process of thought, as our passions, prejudices, etc. The latter of these references we name applied logic ; of which the business, evidently, is neither organon nor ca?w?i, but simply a cathartico?i, of thought. We say, then, that the science of logic in its purity — universal logic, general logic, elementary logic — is an absolutely complete and also an absolutely a priori science. It is complete, because, though existing now for more than two thousand years, and the constant object of consideration to the very highest intellects of each succeeding age, it has yet, since the days of Aristotle, not moved a single step whether in advance or retreat. It is a priori because it is a science purely formal ; it excludes from consideration all matter of thought whatever, and relates solely to the forms of it. The laws it establishes, the rules it prescribes, con- cern not the thing, object, or matter that is thought, but simply the general process and processes, the general forms of the mind, in thinking. These laws and rules, these processes and forms, are absolutely general and completely independent of any particular subject- matter to which they may be api^lied. Surety, then, in initiating an inquiry into the existence and nature of a priori knowledge, we are quite entitled to assume as a priori at once mind and the science of mind. Even as such science, we might say, that the science of logic must be a priori and complete ; for it depends on a unity, on the concrete, organic unity of what is itself a priori and complete, the mind. The divisions of logic, therefore, will throw the THE REPRODUCTION. 33 required light on the divisions of our subject. One great division is into the analytic and the method ; the former being a classified discussion of all the ele- ments of the subject concerned, and the latter relating to those elements as applied. This distribution we shall adopt, but not formally. Our main inquiry shall correspond to the analytic ; but only certain corollaries thence shall represent the method. Again, that part of general logic which is named dialectic, while nominally having place with us, shall be sub- stantially different. The origin of dialectic lies in this, that the formal laws of thought, while furnishing merely a negative condition of truth, and consequently adequate only to a canon in test and guidance of the disposition of its matter, are actually used as an organon of enlargement, discovery, and creation. But this is manifestly wholly incompetent to what is only a formal and negative guide. General logic can never constitute any such instrument of attainment, any such organon of knowledge, but simply, as said, a canon or standard for its correction and safety. Dia- lectic, then, in that it rests on logic as an organon, is evidently without support. What we shall substitute for this usual false dialectic, will be the consideration of an unavoidable dialectic which springs up naturally, as it were, from our unconscious application of the results of our analytic, not as a canon, which it is, but again as an organon, which it is not. For the result of our inquiry, as it is not difficult to foresee, will be veritably a counterpart to general formal logic : as the latter supplies the forms of thinking in general, the former will have to find for us, not mere forms, but the pure or a priori matter thrown by the mind into the products of sense : it will be properly named, therefore, transcendental logic, the logic of c 34 TEXT-BOOK TO KANT : what matter in the objects of experience, and experi- ence as a whole, is a priori furnished by the mind itself; and it is really no contradiction that that matter should be only formal, or consist only of forms. This transcendental logic, again, will have the same more particular divisions as general logic, which, for its part, is divided into simple apprehension, judg- ment, and reason. These, namely, are the intellectual faculties ; these are the mental powers, and all the mental powers which have to do with the procuring and extending of knowledge. And it is precisely from an analysis of these powers that we are to expect a discovery of the a priori moiety of knowledge, even knowledge perceptive. Such faculties as imagination, memory, abstraction, etc., are but implied in these, or are only other names for particular functions of these. Our inquiry, then, will, in the first instance, fall into three books under these three headings respectively ; and the further subdivisions will develop themselves as we proceed. Our first book, accordingly, will, in consideration of the pure or a priori contributions to perception, treat of simple apprehension ; our second of judg- ment ; and our third of reason. Book I. — Apprehension. 1. Relation of Sense to Apprehension. Apprehension is the faculty by which, according to logicians, notions are formed and reproduced. Our quest being only what addition to materials of sense THE REPRODUCTION. 35 is made by the mind itself in receiving and dealing with them, it is evident that, of the two operations here, formation will have much more promise for us than reproduction. The latter will have its own place, as we shall afterwards see, in the scheme of the general process for the realization of perception ; but in so far as it simply involves repetition and not formation, it is evidently not addition that we arc to look for on its part. What, then, is to have in- terest for us here is — the formation of a notion ; and that suggests two questions : 1, What are notions ? and 2, How are notions formed? What we mean by notion, is what Hume meant by his " idea " when he called it " copy of an impression." After experience of a sensible fact, we come away with a notion or idea of that fact ; which, obviously, is just this, that, having had an impression by sense, we have a notion of it by reflection. From this it would follow that a notion is confined to reflection and has no place in the impression. This, however, we shall find reason to question — impression being understood, that is, to stand for what in general we call object of sense. As regards the formation of objects, that is no con- cern of general logic, so far as it implies the taking up or aj)prehension into sense of the object which, as the original, precedes the notion, as the copy. With us, however, this is different; for it is just possible that in the very taking up of the object, the bare apprehen- sion of it, there may be an addition made to it by the faculty that takes up or apprehends, and in this very act. The faculty or faculties of sense, then, if properly omitted from a general logic, must, very certainly, have place here in the transcendental logic. We, for our part, namely, must omit no step in the general pro- 36 TEXT-BOOK TO KANT: cess of arriving at knowledge ; and it is evident at once that the sources of knowledge can be referred to two heads — sense by which objects are given, and understanding by which they are thought. We may hazard the conjecture that these are but two stems from a common root ; but we cannot as yet identify them, and must regard them apart. At all events, to an act of perception proper, we believe that both faculties must concur; and therefore it is that we have ventured to commit the apparent inconsistency of introducing questions of sense into an inquiry that guides itself by the divisions of general logic, — the rather that these questions with us will concern, not what is a posteriori, but what is a priori, even in sense. A transcendental logic is bound, in its search for a priori elements, to investigate the sensuous as well as the intellectual part of the general process of apprehension. The only faculties that are commonly spoken of as relating to sense, are sensation and perception. Of these, looking firstly at their external use, the former (special sense), plainly, is wholly of an a posteriori reference, and can relate only to a matter that must be given. Smell, taste, touch, sight, hearing, concern odours, savours, feels, light, sound ; and all of these can only come to us a p>osteriori, or from actual impression (meaning by the word impression, however, only, as Hume did, the actual sentient state without reference to any impressing cause). There is no possibility of arriving a priori at any smell, or taste, or touch, or sight, etc. : for these, come from whence they may, we have always to wait. External sensation, then, evidently contributes only what is strictly empirical. Its function is passive only ; it is a mere receptivity ; it simply takes up what is given to it. THE REPRODUCTION. 37 And yet, again, does it give what is given to it? Can a sense give what is given it ? In smell, for ex- ample, is the odour qua odour really from the object? That the object precedes this state of mine which I call odour, I readily admit ; but this odour is, after all, a state of mine ; it is a mere modification of my own feeling, — whatever may be in the object to cause it. What that may be — what quality in the object apart from my feeling, but inferred as cause of my feeling — I know not at all, and never can know. No sen- sation can give me any information but how / am affected — / myself. Of any information as regards the object I am entirely void, except that this my state is (inferentially) of its excitation. A knowledge of this my state, nevertheless, let it be as clear, distinct, and accurate as it may, is not, and never can be, a know- ledge of its state, a knowledge of it. Insight into my own self is never insight into anything else. The object, be it what it may, can only affect me ; and I can only know how, as affected, I feel. In no case of affection from something else — and all that I can know is affection from something else — can I ever get to any consciousness but of some feeling of my own. This state of the case is not peculiar to one sense (smell), but is the same in all. The taste of sweetness, for ex- ample, is mine, it is wholly a condition of my own self; I can never get out of my own self to know how that is constituted which caused it. Were a drum sentient, what could it know of any body that struck it? It could only know its own vibrations, and pre- cisely the same vibrations might be set up by a thousand different causes. Touch, too, only relates to certain feelings of pressure or resistance in me : these are conditions of mine, not of the object. The case is not one whit different with light or colour ; 38 TEXT-BOOK TO KANT : and light or colour is all that sight can see. Lastly, has my state of being when under sensation of sound any resemblance whatever to these trembling strings or to this trembling air? To know an object in itself, or as it is in itself, demands an understanding that can function directly on this object, and not one that can act only indirectly on it through a medium of sense. But such an understanding would be one not confined, like ours, to the reflecting of resultant notions : it would be one that perceived at the same time that it reflected. It would be directly present to the things themselves and as they are in themselves. Such an intellect, as an intellectuelle Anschauung, an intuitu* originarius, we may attribute to the Supreme Being, but never to man, who, con- fined to an indirect knowledge only through medium of sense, can possess no more than an intuitus cleri- vativus. This fact is fundamental in our present inquiry, and must never be lost sight of. We know only our own affections. What we call things, and know as things, are only these affections themselves, variously combined, manipulated, and placed. We assume things in themselves as antecedents, antecedent stimuli, of these affections ; but these stimuli, these antecedents, these things in themselves, we know not at all, and never — remaining as we are — can know. The affection of sense, on the assumption of things in themselves, at once reveals their presence and conceals what they are in themselves for ever; at once grants and denies access to them. The window that admits is at the same time the wall that excludes. But, in such references, it is not different with inner sense. Joy, grief, hate, scorn, are all subjective feel- ings of my own that only follow a posteriori from antecedents that precede them. Accordingly, as it THE KEPllODUCTION. 39 is only by the intervention of sense that we know the outer object, which, therefore, can never be known in itself, so it is only by the intervention of sense that Ave know the inner subject — which, similarly, there- fore, can never be known in itself. No doubt every consciousness within me must be accompanied by the further consciousness I, or it is I that am here and now thinking; but this I is but a logical copula — it is wholly without matter of contents — it is but a point, but a bare logical idea, that connects, certainly, but is itself void, or has nothing to show for itself, no- thing to exhibit in constitution of itself. Now, beside this I, the other I, the I that undergoes the succession of empirical states, the I of the inner sense, the I of empirical consciousness, is — so far as it is to be con- sidered in itself- — equally unknown, for any knowledge in its regard is only through intervention of sense, and any such knowledge is in all cases phenomenal only, and never noumenal. Sensation, then, outer and inner, must be a pos- teriori, for it is entirely passive and waits — waits for the affection that simply comes to it, it knows not how or whence. What, as contradistinguished from sensation, we call perception, however, has, while still very palpably an affair of sense, more than a passive character : it at least combines, and that is no function of mere receptivity. Accordingly, though we cannot but eliminate sensation here, it is quite possible that perception as perception may have elements for our purposes. Perception, that is, may possibly contribute to the general web of knowledge or experience a thread or threads specially its own and utterly independent of experience as regards origin. The question, then, is, Are there any pure perceptions, any non-empirical perceptions? And 40 TEXT-BOOK TO KANT : if we can but discharge from perception in general all the empirical colours, if we can remove from the general web, that is, all materials that are evidently a posteriori and from experience, the means for an answer will remain for us ; for either we shall have pure per- ceptions left, or we shall have none and just nothing at all. Now let us attempt this with outer perception, let us discharge from general outer perception any- thing of a posteriori origin, smell, colour, taste, etc., — let us attempt this, and we shall speedily find that, having withdrawn all objects whatever of an empirical nature, there remains behind — space, which we can- not withdraw, nor conceive withdrawn. In like manner, when we withdraw from inner sense all the a posteriori elements, all empirical states whatever, there remains behind — time, which Ave can neither withdraw, nor conceive withdrawn. We can conceive the removal of every element of sense, inner or outer, except space and time ; which, so to speak, are there before all other experiences, as only to receive these, and which, consequently, remain when the others disappear. What, then, are they — what are space and time ? If still entities of sense, they are mani- festly very different from all other such. Their abstrac- tion is inconceivable. Neither do they seem to have objects, as all other sensations have. Nay, they do not seem indebted to any sense for their introduction, like the others. Space is not an affair of any special sense or senses, and just as little so is time. They do not seem things, then ; nor qualities inherent in things; nor relations between things: they are en- tirely independent of things in any aspect, and would subsist though the whole universe of things were bodily taken away. All things, indeed, are finite, but they are infinite. Space is absolutely boundless ; THE KEPRODUCTION. 41 time absolutely without either beginning or end. We eannot possibly call entities differing so widely from things by the name of things. Such wwthings things ! — we cannot say so. What, then, are they ? They are pure — not sen- sations — but perceptions, pure objects, pure Anschau- ungen, which word involves both characters. They are the contributions of the faculty itself; the one attaching itself to all objects of outer, the other to all objects of inner perception, and so also through these to all objects whatever. A word or two of argument may be necessary to develop further the position assumed. And, first, they are non-empirical and quite independent of experience ; for, in regard of things, to be able to perceive, not only that they are, the one from the other, different, but also that they are in different places, is to add an element which the things them- selves manifestly do not bring, but which is, equally manifestly, simply presupposed as the fundamentally universal and necessary condition of the existence of things. And it is precisely so with time. All objects are in time ; it is not derived from them ; they are in it, as the universal and necessary condition of the very existence of them. Secondly, then, they are universal and necessar}\ But, as such, they are not possibly a posteriori; they must be, and can only be, a priori contributions of our faculties themselves. Thirdly, they are still perceptions, not notions — con- tributions of sense (general sense), not contributions of the understanding. They are each single, and have not the generality which a notion involves. A notion has many individuals under it ; but the parts of a perception are in it ; and such is the constitution 42 TEXT-BOOK TO KANT : of space and time. The parts of space and time are in each ; they are but limitations of the one single space and the one single time. Such parts do not precede their wholes, as the species precede the genus, or the individuals the species ; they are, on the contrary, even as parts, with their wholes, one. In short, time and space are, though infinite, single, each a representatio singularis, just like every other object of perception ; they resemble in no way a notion, which is always a generality, a representatio per notas communes ; they are not of a logical nature at all, but, in very truth, sensible. They are percep- tions, pure perceptions, actually pure objects — pure objects of general sense. How simple, now, the apodictic validity of all evidence that concerns the relations in geometry, for example. We simply see that the straight line is the shortest : it is a truth perceptive, it is a truth in- tuitive, as this word used to be understood, though tantamount now, for the most part, only to instinctive or immediate and at once. An intuition is evidence, and no blind trick of our original constitution itself; in fact, it is truth at a glance, and the glance should not be lost sight of. Had Hume, who understood the word, but investigated what it implied — the foundation of the mathematics, namely — he would, in all probability, not have left the general problem to us. What we perceive, then, are only phenomena, and never noamena, though we may hold the former to be gasres and smarantees of the existence of the latter. In short, both outer object and inner subject, being perceived only through sense, are, by necessary con- sequence, perceived also, not as they are in them- selves, or not as they just are, but merely as they appear. Whether we look to space or time, it is only THE REPRODUCTION. 43 our own states we know in either; and the subject of these states, as in itself and apart from these states, apart from the form or forms of sense, is no more known than the alien object or objects are known, which, in the external reference, are supposed to act on said subject, and, in that way, account for its ex- ternal states. But such being the nature of space and time, we see at once how very much mistaken the school of Leibnitz and Wolff must have been in asserting space and time to be objects of the understanding, and in attributing to us a knowledge in their regard not possibly other than obscure and confused, inasmuch as sense was in its own nature but a more obscure and confused kind of understanding. The difference between sense and understanding is generic : it is not a mere less or more of quantity; it is a difference in kind, a total difference of elemental quality. By our theory, in fact, we avoid not only the difficulties of Leibnitz and the metaphysicians, but those of the mathematicians as well. With the former, for ex- ample, it was impossible to explain how forms of the understanding, and nowise different from the form proper of the understanding but in that they were specially less perfect and more confused than it, should yet possess evidence (as in geometry) specially clear, specially perfect, — specially apodictic, in short. With the latter, again, who assume space and time to be simply objects of sense a posteriori made known to us like all other such, there cannot be any answer given why empirical elements should yet, exception- ally and contradictorily, extend to us an apodictic evidence — at the same time that we are left with two infinite wnthings, totally unlike all that we call things, staring us in the face. 44 TEXT-BOOK TO KANT: 2. Relation of the Understanding to Apprehension. Objects, then, affect the mind through sensation, and the resultant affections become disposed and arranged (but under other influences to be afterwards seen) in the two receptacula, as it were, of space and time. These two receptacula are universal forms, to the conditions of which all affections of sense must conform ; and thus it is that we are enabled, a priori, to know and predicate many peculiarities of objects, which objects can themselves be known only a posteri- ori and by experience — all those peculiarities, namely, which all objects must take on in obedience to the general forms of sense through which alone they can present themselves. Space, for example, has three dimensions, and, consequently, all objects of sensation are similarly constituted. Time, again, is only of one dimension, and therefore it is that all the variety of inner sense must present itself in conformity to this quality. All the conditions that pure science discovers in the general structure of either form, are evidently predicable of all objects that can ever come into ex- perience. So also does it become evident that, though a jiriori and independent of experience, they are there only for experience. Their use and purpose, and thefinal cause of their construction, relate to the a posteriori world that is to be given to them through sensation. They are, as it were, discs projected from within for the reception and co-ordination of the variety of par- ticulars from without; and thus it is that science can discover in them no law or principle capable of con- veying information relating to any world but that of experience. So also is it, as we can easily understand, THE REPRODUCTION. 45 that no knowledge of either time or space would have been possible without experience : it is only in actual experience that they present themselves to us, and without actual experience they could never have been known. The moment that inner sense is awakened to exercise by the possession of objects (states of empirical consciousness), the subjection of these objects to the law of time is obvious ; and this latter (time) takes up as objective and, so to speak, empirical a position as those former (the objects or states) themselves. So, also, the moment that outer sense is awakened to exercise by the possession of its objects, the subjection of these to the laws and modes of space becomes equally obvious ; and the role played by space is as much outward and real as that of the objects themselves. The objects are perceived, and time and space are perceived in connexion with them. Time and space are, as it were, the spectra projected for reception of objects, and present themselves to us only luith these objects, and as of identical nature and origin. Locke, then, had perfect reason in assuming that, de facto, our information in regard to time and space depends on experience ; at the same time that it is only our inquiry de jure, or concerning the authority asserted by them, that gives the key to their true nature. Though in sense and only known by sense, they bring with them such peculiarities as single them out from all other objects of sense whatever, and it is the investigation of these peculiarities that leads us to see that they cannot be of an a posteriori, but must be of an a priori origin. In asserting, too, that all objects of a posteriori knowledge must submit themselves to these forms, it does not follow that the special form of each individual object is also to be considered as so due. How it is 46 TEXT-BOOK TO KANT : that a mountain has this shape, and a tree that one, does not depend on space, for example, but on the object-in-itself. That object-in-itself, however, we never can know : we only know that, be its special form what it may, or, in obedience to its own transcendent or absolute nature (and transcendent is easily seen to be capable of being allowably replaced there by trail- scendental), let the special form it produces in us be what it may, that special form must still present itself as in subjection to the general laws of space. It is no objection, then, to say, This brick and that stone have each a shape of its own, which shape they cannot receive from space ; for the answer is easy. We do not say that the special empirical form is due to space; there is something in the object-in-itself that says the special empirical form shall be this only, and not another. Still the special empirical production must obey the universal conditions of space and become — but only in its own way — spatial. 1 And there is nothing really difficult in this. There are outer objects — meaning by the term at this mo- ment what we name thing s -in-themselves ; but they are wholly cut off from us — even by the very effects they produce in us. We are in presence only of these effects, or of our own resultant affections. These affections are therefore inner ; and there is no diffi- culty in the conception of their receiving further modification and development in and from the inner apparatus into which they have been received, and to which they now for ever belong. Nay, it is natural 1 It will be seen here that, as in some other cases, I have not scrupled to state and meet in the reproduction a difficulty which, as hinted in another work, I have not seen struck upon elsewhere, but which, for all that, must, I should say, have been very commonly felt by all students of Kant. The reproduction is a free one : see " Apprehension," etc. THE REPRODUCTION. 47 to suppose that new arrangements will take place on their being thus received into mental consciousness. They are in themselves, as a posteriori, wholly dis- junct, in an element of contingency ; and if there is to be such a thing as a ruled and regulated context of experience at all, in submission, namely, to neces- sary principles by which systematic arrangement and completeness will be produced, these principles must be of an a priori nature, and so bring with them an authority which experience (the a posteriori) can never bring. If experience, in short, is to be a con- nected whole, it is absolutely necessary that our contingent affections due to experience should be subject to an element of necessity ; and, these affec- tions being once for all within, that element can only come from within. Were our objects the things-in- themselves, then there could be no apodictic know- ledge in their regard possible ; for, in that case, we could have no knowledge but through experience, and such knowledge never brings, and can never bring, such authority. That is, evidently, for any information that bore on such things we should be wholly dependent on these things ; we should have to icait for them and it ; it, consequently, as a pos- teriori and en pirical, would be and could be contin- gent only. All would be as we just found it ; and, as with matters of fact now, we could say no more then than that the state of the case was so and so, but not that it must he so and so. But our objects are not the things-in-themselves ; they are not noumena but phenomena; and so long as our understanding is a discursive one, and, consequently, dependent on sense for matter of discourse, or so long as we cognise only through notions, which notions, again, can themselves only obtain filling, contents, through the information 48 TEXT-BOOK TO KANT : of senses, inner or outer, Ave never can attain — whether we look to object or to subject — to the presence of things-in -themselves. Were our objects the things- in-themselves, they would consist wholly and solely of their own elements, and would possess no ingre- dient whatever derived from us ; but they are not the things-in-themselves — they are not things in themselves at all ; they are only affections of our own within, though due, it may be, to the action on us of such things in themselves. Now affections of our own within can only receive order and arrange- ment from within. How, then, there can be an a priori element in what is an actual, objective, and, it may be, outer fact, is not difficult to see ; and it is not more difficult to see now, also, that that objective fact, so far as it is a posteriori, can only be contingent, and, as contingent, stands palpably in need of some further manipulation that shall raise it into the neces- sity and law of a consistent universe ; such manipu- lation, for its part, evidently involving such principles as are in question — principles which are also within, and attach themselves from within, but which are of an a priori origin and necessary validity. Thus it is in fact that we see time and space add themselves to the phenomena of sense, imparting (even in their own right and apart from other elements in the single realizing act) to these phenomena some such co- ordinating and subordinating conditions of necessity as are required. And thus, too, there is another reason for the phenomenal nature of the objective world which we seem to perceive around us; for, even if the so-called objects were objects in themselves so far as a posteriori sense is concerned, they could no longer be allowed to remain such, being under the necessity of subjecting themselves to the modes THE REPRODUCTION. 49 of space and time, and so of reducing themselves to mere phenomena. So far, then, as both the nature of sense in general and the forms of our sense in particular are concerned, we can know only pheno- mena, not noumena — things as to us they appear, and not as in themselves they are. Nor is there any reason that this conclusion should dissatisfy : a phenomenal existence may be as consistent as a noumenal one ; nay, as in the first instance probationary, it may have its own good ends. It does not at all follow, in fact, that objects are even illusory because, at bottom, only manifestations to sense, or only appearances. Rocks will still remain rocks to us, for all that, and as hard as ever ; fire will still really burn, and water still really drown — only, in metaphysics, things must con- sent to receive their true metaphysical expressions. All our knowledge, therefore, consists of two fac- tors, and both are subjective ; but the one, being a posteriori, empirical, sensational, is contingent, while the other, as a priori, transcendental, perceptional, is apodictic. What I call red, for example, or sweet, or loud, or smooth, is red, and sweet, and loud, and smooth only to me. It may be also red, or sweet, or loud, or smooth to you ; but we cannot know that — even though we certainly say that. We all call our feelings, that is, by the same names ; but identity of name is no clue to identity of feeling. What all other men feel to be red, I might feel to be green ; but I should still call it red, for I should have no means of knowing that I differed in feeling from other people ; and I should name it what they named it, both I and they being perfectly consistent in the use of the word for our respective feelings, however different these feelings might be in reality. In short, it comes to this : the a posteriori subjective states we 50 TEXT-BOOK TO KANT I have no means of comparing, and, consequently, can- not tell whether they are the same or not. No language can convey to you my feeling, or to me yours. Name them as we may, and name them as universally as we may, feelings are still, in point of fact, inexpressible, and consequently incomparable. But it is different with a priori elements : they are neces- sarily universal forms which are perceived, and per- ceptions can be compared, for they can be exhibited, their constitution can be submitted to process of intellectual inspection, and, consequently (with dis- charge of sensation), their fundamental conditions, principles, and laws compared. That the three angles of every possible triangle are without exception equal to two right angles, this — and the same thing can be said for every proposition due to the essential nature of the perceptive forms, or time and space — is not one thing to one man, and another thing to another : it is an affair of reason, and not of feeling ; and while all that relates to the latter is individual and incapable of comparison, all that relates to the former is universal, and consequently capable of examination for assent or rejection by all of us. Thus, then, space and time, as universal and the only universal a priori perceptive forms, are seen to possess a certain intellectual nature, and to be capable of presenting themselves in universal reason. If, then, space and time, which subject all objects to their own conditions, be themselves subject to conditions of the intellect or the understanding, all objects what- ever, outer or inner, must also (through them) subject themselves to conditions derived from the understand- ing alone. And this, indeed, seeing that our know- ledge relates only to contingent appearances, we should also naturally expect as probable and even THE REPRODUCTION. 51 necessary. For objects, at last, are obliged to relate themselves to the understanding — all objects must be understood ; and it is reasonable to suppose that, in the process of the union of objects to the understand- ing, there will be conditions. Now, we have already seen the a priori conditions of sense, and we see here the anterior probability of the existence of corre- spondent a priori conditions (in perception) for the understanding itself: the search, then, for these latter seems, as the next step here, to be presently imposed upon us. That is, we ask for those a priori condi- tions of the understanding which (if any) necessarily attach themselves as a further modifying element of perception to all objects that, as perceived, have already submitted to the conditions of time and space. Nor ought the general idea of what is essentially intellective becoming actually or empirically percep- tive to prove a perplexing one. The influence of a notion on perceptions must have manifested itself to every one. In fact, we may say at once that no per- ception is complete until a general notion has con- joined itself to the multiple of sensation. And here we may remark that perception, even as perception, is either crude and elementary or finished and com- plete. Now, crude perception is a breadth of parts, a complex of particulars, a detail of items, a multiple, a manifold, a many — just as sensation is. Perception as opposed to sensation involves more than the mere feeling of the latter: it involves, besides the appre- hension of elements into mere subjectivity, their apprehension as well into objectivity. Perception, as perception, whether crude or complete, primary or ultimate, is awareness of an object; and an object is always something that a subject conceives itself to discern as different from itself, but presented to it, 52 TEXT-BOOK TO KANT : offered to it, — as it were, for inspection, held up to it. Perception, then, is the sensible presence to conscious- ness of such discernible elements as we call objec- tive, or, indeed, just at once, objects. Pure per- ception, for example, has for objects the peculiar details of time and the peculiar details of space. So long as it is pure, it has, by way of contents, nothing else whatever. Now it may be seen at once that these details, though sensible, cannot be called sensa- tions; they have not the character of sensation, mere feeling — the feeling of light, sound, etc. : they are discernments, awarenesses ; they are to a certain ex- tent intellective and cognitive ; they are perceptive ; they are perceptions. That is what is meant by the word Anschauung. Whatever has that character in it — beyond mere sensation — of sensible discerniblc- ness, perceptivity, objectivity, is an Anschauung. But an Anschauung, a perception, is only crude and ele- mentary when, as in the first instance, the sensible details of it alone stand before consciousness. We may conceive the details of time and space always to stand elementarily thus, from the first, and in the background, sensibly before consciousness. That is crude perception. Finished perception, complete perception, is more than that : we have then an object before us, a house, a ship, a cannon-ball, a cushion, a glass, a stone, the sun, water, ice, the air. And each of these we can see to consist at once of details of sensation, as well as of details of perception ; but all combined at last into a single unity, which is at last, too, only a unity of perceptive details — the very sensations have become perceptions. This ought to make thoroughly intelligible what percep- tion is as opposed to sensation ; as well as, in oppo- sition to crude perception, what is complete percep- THE REPRODUCTION. 53 tion ; and as well the one as the other. There is a temptation to speak of complete, in contradistinction to crude perception, as perception proper. This, in view of the completeness. But then, again, in view of the character of perceptivity as perceptivity, and with that only in our eye, we might call even crude perception, perception proper. No matter of perception, then, is a simple ; or it follows from the very nature of time and space that all such matter is a plurality, a multiplicity, a detail: all objects are multiples, consist of parts, of a variety of particulars, a many of details. Indeed, there is an element of variousness in the special senses themselves; for one and the same object may owe materials to each and all of them. Perception, then, while yet in its first crude form, as before sense in the mere details of time, space, and sensation, is but itself a detail — a detached and incoherent and unconjoined many. But were it to remain such, it would be incomplete ; there would be perceptions, perceptions as it were in blur, percep- tions in the raw, but not a perception, — a formed perception, a complete and finished perception. In order to the attainment of this latter, the detail, the blur, must collapse, so to speak, into singleness; the multiple must pass into a simple ; the complexity and multiplicity must disappear into unity ; the parts must unite into a whole ; the particulars must eclipse themselves into a universal — that is, they must be thought, become notion. " For, as notions without per- ceptions are void, perceptions without notions are blind.'''' Suppose some new object be brought from abroad and put before you : you perceive it at once ; and yet you confusedly feel that you do not perceive it. You confusedly feel this, in fact, till, on a sudden light, you exclaim, It is a basket, a drum, a knife, an oar, 54 TEXT-BOOK TO KANT: a club, or whatever else. Now, if you recall that sudden light, you will find that it leaped from the collapse of the detail — the sensible many — into unity. What happened, indeed, was, that all the particulars of the perceptive detail, darkly and disconnectedly before you, sprang suddenly together into the unity and light of the universal —basket, drum, knife, oar, club. For these, so far, or though only empirical, are all univer- sale; they are all notions, or the words themselves are general terms that involve notions or represent notions. The notion under each of these words, in fact, has an infinite variety of individuals under it ; and is therefore a universal. When, one morning, the day broke, and all unexpectedly before their eyes a ship stood, what it was, was evident at a glance to Crusoe. The perceptive manifold collapsed for him at once into the unity and simplicity of the general notion, ship. But how was it with Friday? As younger and uncivilized, his eyes were presumably better than those of his master. That is, Friday saw the ship really the best of the two ; and yet he could hardly be said to see it at all. He really did not perceive it — perceive it as more than a crude and elementary perception ; he did not perceive it as a formed and finished perception. In short, what to Crusoe was an object, was to Friday only a dark and amorphous blur, a perplexing, confusing, frightening mass of de- tails, which would not collapse and become single and simple to him. It can easily be understood that this single example applies to all cases, and that we really do not perceive until by the help of a notion we also understand. Has it never happened to the reader to lie in a strange bedroom, and to puzzle himself in the morning about some distant object which he was conscious he had known perfectly well the night THE REPRODUCTION. 55 before, but which he could not put together for the life of him now ? It is an object on a shelf, peeping out of a cupboard (say) : ivhat is it? What a strange- looking object it is ! A formless detail of many per- ceptive particulars, an incomprehensible plurality of parts; but what is it? Ah! a candlestick, a family Bible, a bandbox, a general's battered hat, etc. The moment you recollect what you had recognised it to be, the moment the notion attaches itself, all is plain ; and yet you are not a bit nearer, and see (qua seeing) not a whit clearer, than before. A man, of a morn- ing, may look out of the window of a strange house, and, for full five minutes, have, to his astonishment, before his eyes a vast chaos of stones stretching over a great plain to the very verge of the horizon, which incomprehensible huge wonder will spring together at last into the very limited garden-wall he recollects to have seen the day before. This same principle it is that makes our ears so very opaque, to say so, in a foreign country. We think we should understand better did the people but speak louder ; but the real want is that of notions. The natives are able to anticipate notions — from tones, looks, gestures, and single words ; so that the whole rushes together intel- ligibly for them, even though they may not have actually heard every syllable that was enunciated. Of course, it is not to be denied that, whether for ear or eye, distinct apprehension of the sensible details is, on its side, an important factor towards readiness of perception. We see this in those who are dull of hearing, or who are short-sighted. Nevertheless, the latter, without one inch of increased propinquity, come often to perceive quite clearly and distinctly some incomprehensible blur into a familiar object, should they but stand still and wait for the notion. 56 TEXT-BOOK TO KANT : And here some one may object that the blind, with all the notions in the world, newer perceive. But that is not the case. A blind man perceives— is capable of Anschauung — quite as much as either you or I. His pipe, his knife, his loaf, is really very much the same object —Anschauung — to any blind man that ours are to us. But on this I do not dilate here. In further illustration, I add only that we see certain persons, women frequently, stutter and stammer and stumble fearfully in the attempt to pronounce some long, or for them new, word. How is this ? They have been taught to read as well as others: what causes the difficulty ? Simply the want of notions. These are mechanical, instinctive heads, that have not reduced syllables into principles of sounds ; and so a long new word is for them a wholly unconjoinable manifold of perceptive details which, with such principles, with notions, that is, would have collapsed into unity and been comprehended at once. It must now be pretty evident, then, how percep- tions without notions are blind. As for the other part, that notions without perceptions are void, we may, probably, pass that as intelligible at once. What were the notion river, for example, or the notion justice, were it incapable of being tilled and verified by perception of an actual case? Surely vacant! As for perception, once again, it may illustrate the point to reflect that the lower animals do not properly perceive. For many of these, objects are but blurs of perception in the raw to awaken aversion or desire. The dog that knows his master, doubtless, has com- bined a certain detail into a loved and feared unity ; but the principle of this unity is, after all, blind ; it is not a notion, not a universal, — though it certainly does duty for such, is a blind surrogate of such. THE REPRODUCTION. 57 The notions which we have seen instanced are em- pirical notions — basket, oar, drum, etc. But, by tak- ing objects more and more abstract, we shall perhaps arrive at such as are a priori and not empirical at all. This was our procedure with objects in search of the a priori of sense ; and it is only reasonable to try whether the same process shall succeed with us here also. Here is a nail that I picked up to-day. To me the perception is complete, for I have united the per- ceptive details, through a general notion, into the single objective reference : it is a nail. But suppose I were a Papuan or original Polynesian, and had never seen a nail, the objective reference into which the detail would collapse would no longer be a nail, but simply a piece of iron ; and the two perceptions Avould now be really quite different. Suppose, again, I had never seen iron, though acquainted with some other metals. The detail in that case would reduce itself to unity only under the notion metal. But suppose I had never seen a metal, and knew only solids and fluids, etc., the nail would be for me simply a solid. Suppose, now, I wanted to describe it and distinguish it from other solids, I should say it was blue, cubical, heavy, sharp-edged, pointed, etc. — in short, I should enumerate all the qualities in the object that presented themselves to my senses. Suppose, now, I withdraw all these qualities one by one, withdraw in thought, abstract from them, will the body wholly disappear? No, not wholly ; there will remain over the space it occupied, which, as we have seen, we cannot with- draw, because it is a pure perception, an a priori object. But, besides this pure perception that remains over, is there not as well, and similarly situated, a pure notion ? We said it is blue, and Ave remove blue. It is no longer blue. So with all the other 58 TEXT-BOOK TO KANT : qualities which we have seen in the object : and with the result, therefore, that it is no longer blue, it is no longer heavy, it is no longer hard, it is no longer cubical, etc. But what is this it/ Besides the quali- ties, we assume an it in which these qualities resided, a substrate in and through which they were thought to cohere. This it. this substrate, in tact, was the notion substance : the nail as a whole, its qualities apart, was to us a substance. But is substance an affair of sense? Is it of the nature of odours, savours, colours, Is it even oi the nature of feeling? All the various feels which the body conveys to us through touch are sensations, and, as such, abstraction can be made from them ; but still the notion substance remains behind, quite unlike all the qualities which were supposed to be grouped around it, and which it was supposed to support. It must, therefore, be a priori, and as it is not a perception, it must be a notion. There is no conclusion possible, consequently, but that it is, in some way or other, an (7 priori result of the understanding itself, of pure understanding. In this way we come to have a glimpse of the pos- sibility of <7 priori notions which shall reduce percep- tive (7 posteriori details, under subjection to the modi of space and time, into the unity of a single objective reference, and that is, into an object or objects as such. But the very idea of such is at once suggestive in regard to causality. Possibly, that is. the very question from which we start will find its answer here. The principle of causality, then, shall depend on an a priori notion. On the appearance of the cause A, the necessary and universal expectation of the effect B shall depend, not on my habit or custom of seeing A and B together (which amounts simply to empirical suggestion, and is altogether inadequate to THE REPRODUCTION. supply an absolutely necessary and universal copula), but on a law of iny understanding itself, to which A and B are reduced and submitted u rical examples, and by which they become for us, and absolute uni jec- .;-cted. It will be worth whil * - rjd a word here on the precise meaning of the ten. . just it may 1 :ed here that, after alL, the prin the new principle, : _ . e. This is true ; i :ar as said principle arises from our nstitution. je adn. to be Bobje " ; that the in reach no further than t excitation in us of empirical feelings, such fu: principles of modification and connexion must be, and can only be. within — sub ference. What we call the objects-in-thems in-themselves, are only adequate to the contir. . - which, through eye, ear, et con- scious of being set up in us. These sens "^em- selves, consequently, are henceforth in us; so that any further manipulation of them, as also iril in that point o: oeceamai - theless objects, though we knc _ of object themselves, still the term has still a mea us. and h of unavoidaljle use :i in a phenomenal of the nature . We do not call immediately due to the object- in-itself — we do not call the sensations objects. To be called objects the - sations must coalesce into single percepti - A. i _ septum that reduces into its own unity a variet] nsations. or. lookir. g I the | are forms of space and time, a varietv of per is an obje I variety, in fact, as we have seen, coll:. ; sea i -oity 60 TEXT-BOOK TO KANT : through an objective reference which is conditioned by a notion. It is combination, then, conjunction, union, that is constitutive of the object; and these processes depend on principles within us, which, there- fore, are, in origin, subjective. It is with reference, then, to the combination, that the terms object and objective come in. The distinction, in truth, turns wholly on the word reference. All that we know (as we cannot know the object-in-itself) is, in effect, sub- jective. Still all that we know has either a subjective or an objective reference. A subjective reference con- cerns only what I feel, what the particular subject empirically feels, experiences ; but an objective refer- ence is the conversion (through insights both of per- ception and understanding) of sensuous details into unities (called objects) that seem thus to separate and differentiate themselves from the subject. It is the perceptional forms (space and time) and the notions of the understanding (categories) which convert the sensuous states of the same subject to which these forms and notions belong into objects. It is also plain, too, how what is objective can, as capable of exhibi- tion or expression, be compared, while what is sub- jective must always rest individual. Our theory, then, is that objectifi cation of the sensuous details depends (must depend) on mental process within. This process, as essentially synthesis, or of a synthetic nature, can never belong to sense. Sense is passive only, it receives, it takes on only what impression is made on it ; but the understanding is active ; it reflects, it examines, it goes to work, it operates change. Union, combination, connexion, synthesis can, then, never belong to the receptivity of sense, but may and must to the spontaneity of the understanding. The understanding, now, acts through THE REPRODUCTION. 61 notions; and it will be therefore these that effect synthesis. These notions, again, as it is plainly not the receptivity of sense that gives them, must be pre- supposed as already in the mind — as already in the mind for accomplishment of the function of synthesis in question. But we have already arrived at a clear conception of all this, both as regards the reductive (or redactive) power of notions, and the actual a priori existence of some such ; e.g., substance and the prin- ciple of causality. What we seem to require now is some means of arriving at a full catalogue, at a com- plete tree of these notions. For it is evident that we must have a guarantee of completeness here, else our whole business fails. Besides a guarantee of com- pleteness, we must have also, however, one of legiti- macy or authenticity. We must possess grounds of absolute certainty in asserting that such and such are the a priori, and that such and such are all the a priori notions, that function unity of objective reference (objectification, objectivity) for all possible sensuous details, — these details, moreover, being assumed to have been (at least potentially) previously disposed in, and according to, the pure perceptive forms of space and time. Such guarantee (grounds) we might name an architectoric principle ; for it would underlie creatively the whole structure that rose from it. Such principle, then, must be the object of our special quest now. But we are supposed to have the clue and guide to our whole general quest before us, and the same clue or guide must be supposed adequate to every partial and subordinate quest. We were led, for example, by the fact of the presence of apo- dictic truths in matters apparently quite empirical to conjecture that there were two factors in all know- ledge (perception) : one empirical proper and due 62 TEXTBOOK TO KANT: a posteriori to the impression of objects-in-themselves, and another, empirical de facto, but not empirical de jure, empirical in fact, that is, but not empirical in origin and authority — due a priori, therefore, to the operations of the perceptive or cognitive faculty itself. So led, we were further led to believe that scrutiny of the faculty itself would yield to sight the peculiar additions which its operations contributed to the empirical whole, the whole of experience. An analysis of the faculty, then, was therefore suggested ; and to the divisions of this faculty, uni- versal or general logic (an admitted pure science from which all empirical elements were certainly eliminated) was adopted as guide. Under this conduct we have already advanced well as yet ; and there is no reason why we should discard it here, especially as what concerns us now is wholly the understanding (the more particular object of logic), and not sense. The function which we are examining now, the reduction or redaction of a complex of sense into the unity of a notion, is wholly an affair of the understanding ; and, inasmuch as the understanding itself is a unity, we can anticipate a like quality for the principle in demand. Could we but find, indeed, all the functions, all the modi of the understanding, we should then, as we are now warranted to assume, be at no loss for all the subordinate forms of what principle we seek. Our first object, then, is a complete table of the functions of the understanding. Now, what is the understanding — what do I mean when I say I understand a thing ? We have already seen, in reference to the illustrations adduced of the necessity of the addition to the perceptive detail of the unity of a notion, that the understanding was a necessary element or moment even in the everyday THE REPRODUCTION. 63 seeing (perceiving) of objects ; that at all to perceive, it was necessary to understand. Understanding, then, in all the examples alluded to, is seen to consist in the uniting of a perceptive complex into the single whole of a universal notion (and then only it is that we perceive the perceptive complex itself). That is, there was no understanding possible until the class, the universal, was found, of which the perceptive detail was only an example, only a particular. All the objects from abroad, for example, or those that peeped perplexingly from the cupboard, or the unin- telligible quarry of stones — not one of these was understood, not one of these was — literally — perceived, till we found the class of which each was an example, basket, oar, club, brass candlestick, cocked hat, Avail. The process by which these were found was thinking : we did nothing all the time we were longing to per- ceive but gaze and think, though the thinking was but an obscure nisus till, with a light, the thought wanted, the notion, sprang to us. Understanding, then, is so far identical with thinking, and both relate to notions. Thought and understanding, that is, are both discursive and proceed by notions. Again, judgment or judging is a faculty that proceeds by notions, a faculty that compares notions, joining and disjoining them. Judgment or judging, then, is but another name for the understanding, or for thinking. AVas it not, in each case of the adduced perceptive details, an act of judgment that added the notion? Was it not an act of judgment that found out the class, the universal, to which the particular or detail in question belonged as an example or instance ? To understand is to think, then, and to think is to judge. In fact, it will be found on trial that no example of thought or understanding can be taken up that will 64 TEXT-BOOK TO KANT : not demonstrate this. For instance, some one, putting one hand over my eyes, brings with the other my fingers into contact with a certain body. Well, I am puzzled for a moment. It is an unknown variety, an unknown detail, an unknown many ; but, on a sudden, all becomes one, and I shout out, Water. Now, was it not at last by an act of judgment that I was able to identify the variety at the point of my fingers as water? Here, then, in the formation of a single notion, we find judgment necessary, as well as that its act or function consists in the subsumption of a given variety under a certain known universal or class. To take more complicated instances, what is it to understand the theory of heat, of dew, of the heavenly bodies? Is it not to attain by judgment to the reduction of a variety of particulars to the simplicity and unity of a co-ordinated and subor- dinated whole of general or universal notions? Or, once more, what is it to understand the universe? Is it not to discover an ultimate principle (God, the absolute) under which we may subsume the infinite all of things? And what faculty subsumes the lower under the higher but judgment? We do not notice the respective domains of under- standing, reason, etc., that are shadowed out here, but we say again that a complete table of the func- tions of the understanding is evidently our special quest at present, and that by means of such table there is every likelihood of attaining to the recogni- tion of the totality of pure notions. But, further, understanding being judgment, we know that logic treats of judgment. Logic, certainly, at least classifies all formal judgments. Now, are not all possible formal judgments just, in so many words, all possible forms of judgment, and are not all possible forms of THE REPRODUCTION. {jD judgment, just, in so many words again, all possible functions of judgment? But if general logic cata- logues all possible functions of judgment, we have seen already that we may, on its part, safely accept as much, for, admittedly, general logic has nothing to do with the matter, the a posteriori, the empirical element, of thought, but only with the form, the a priori, the pure element, of thought. General logic, too, has existed for 2000 years without suffering either diminution or increase ; and, in its regard, therefore, we may positively rely on the presence of correctness, completeness, and sufficiency. Judg- ments, then, purely regarded, will be found to possess in logic Quantity, Quality, Relation, and Modality. In quantity, judgments are either universal, par- ticular, or singular. In quality, they are either affirmative, negative, or indefinite. In relation, they are either categorical, hypothetic, or disjunctive. And in modality, they are either problematic, assert- oric, or apodictic. It is necessary to admit that this classification is not absolutely identical with any of those that may be met with in the usual treatises. Still we dare assert that an examination and comparison of all that is ordinarily treated of in general logic as concerns judgments will justify us in the assumption of the classification we propose, and will show that, while we have essentially neither added nor subtracted, we have, by greater scientific rigour both of distinction and association, possibly, or probably, in no small degree, improved. Well assured ourselves as regards accuracy and adequacy, we deem it unnecessary to retard our main inquiry by any formal analysis and justification in this place, but simply proceed. Now what is the general function of a judgment E 66 TEXT-BOOK TO KANT: according to logic? When we say, for example, all bodies are divisible, what mental process is indicated by the assertion? I have certainly conjoined two notions : I have asserted the one of the other. But for what reason ? Evidently for this reason, that the one notion, divisibility, is implied in the other, all bodies. Now, that is to say that I analytically found this, for to bring several ideas under one is an analytic act, depending on process of abstraction and generalization according to identity. But this same sort of reason, and similarly constituted, obtains throughout all the other judgments, under whatever name classified. Judgments, in ordinary logic, therefore, are analytic- ally applied, and in regard of notions. Still the action itself of each judgment is a synthetic one; for even the disjunction of negative judgments involves synthesis with an opposite. All forms of judgment, then, are various functions of synthesis, which, logic- ally, are analytically applied, and between notions. But may not these various functions of synthesis be conceived capable of being otherwise applied ? Any object, as first apprehended in consciousness, is but a plural blur of parts, of units of sensation and crude perception. This is their condition as received into imagination. But imagination is productive and reproductive, is capable of movement, is capable of movement among these units. It is capable thus of recognising them, mustering them, comprehending them, and, under the unity of self-consciousness, to a certain extent, performing (in connexion with time and space, which also lie in it) synthesis upon them : it gives them continuity. Such synthesis, however, would still be contingent and subjective. There seems still required something else to bestow objectivity and necessity. Now may not that something else be THE REPRODUCTION. 67 extended to us precisely by these various syntheses of the functions of judgment? The question, then, is this, May not the same functions of judgment that act analytically in logical application to notions be capable of a synthetic action when perceptively applied to the complexes or manifolds of sense? The perceptive units are a disjunct plurality received into the mind, and there are at the same time func- tions of unity in the mind ; but, pluralities received into unities, affections received into functions, why should the latter not grasp and unite the former? In the three classes of judgment, for example, that involve union, connexion, with reference to relation, is it not conceivable that the categorical function, or the hypothetical function, or the disjunctive function, may act in uniting, not mere notions analytically as in the reflection of logic, but the actual facts of ex- perience synthetically as in the perception of sense ? The categorical judgment, as we know, expresses a direct relation between the subject and the predicate ; the two notions are there directly or categorically related. We see, then, that categorical relation is a function of the judgment or the understanding, and is it inconceivable that this function should relate itself, as well synthetically to a perceptive variety already offered to it, as analytically to a notional variety similarly offered? Is it impossible to con- ceive two facts of sense which, operated upon by the function in question, would reduce themselves into a relative or correlative unity? Categorically to attach predicates to subjects is really to affirm qualities of substances. The formal function of thought, then, implied in the categorical judgment is the rela- tion of subsistence and inherence, of substance and accident, This being a formal function of judgment, 68 TEXT -BOOK TO KANT : a rule of synthesis, it is evident that, a complex of sense to suit being introduced into it, such action will follow as shall exhibit, sensuously, perceptively, in actual facts of experience, an example of this function, a case of this rule ; but an example withal, a case withal, which, however empirical, however much a matter of mere sense, shall possess, nevertheless, all the universality and necessity of the intellectual in- sight that lay in the general function, that lay in the general rule. Again, in the hypothetical judgment, the relation is between two propositions, and all that is involved is the truth of the consequence. If there be perfect justice, the hardened sinner will be punished : we see that what is concerned here is a vis consequentice. It is not the truth of either proposi- tion that is considered, but simply that of their rela- tion, simply that of the copula between them. But this is an original function of the intellect, and we may certainly conceive some suitable complexion of facts reduced under it; in which case what would result could, manifestly, be only an example of cause and effect — perceptions, now, not notions, with a vis consequentice between them, which should be universal and necessary. As the ground or reason implies its consequent or result, so the cause implies its effect. Things thus, quite empirical themselves, and, con- sequently, quite contingent themselves, may quite well bring with them in their relation the necessity and universality of an intellectual insight. Of course, it is still evidently a necessity that the sensuous com- plexion should, as said, suit — the intellectual ratio that is ; else subsumption were inconceivable. But how this takes place we know not at all — how objects should present themselves in such synthesis or com- plexion as brings judgment to act upon them, and THE REPRODUCTION. G9 reduce them to its universal rule, we know not at all. Why this tree has its particular shape is quite unknown to us ; still it has to conform to the general laws of space, and present itself in length, breadth, and thickness. So, what may be the nature of the object in itself that underlies any empirical complexion, as of magnet and steel, spark and powder, etc., is absolutely hidden from us ; we only know that it is such as to be necessarily subsumed under the function of judgment that concerns the vis consequently, and comes forward, consequently, in the duplicity of a correlative cause and effect, with the necessity of intellectual insight imparted to it. l A similar train of thinking will, we doubt not, bring the reader to see the legitimacy of all the other members which, as principles of perceptive synthesis, we seek to deduce from, and place parallel to, the various affections, quantitative, qualitative, etc., of the logical judgment. These principles we name categories. Under quantity, there will correspond to the functions or affections of the judgment which we have already seen, the categories of unity, plurality, and totality. Under the other rubrics, quality, rela- tion, and modality, we shall similarly have the cate- gories, respectively, of reality, negation, limitation, — substantiality, causality, reciprocity, — possibility, ac- tuality, necessity. After what has been already said, there will be little difficulty in understanding that these categories are but the various affections of the logical judgment which we usually find in the ordinary text-books — these affections, regarded as functions of unity, and conceived to be synthetically applied in reduction of 1 I shall be found again, in the above, to be attempting to meet, suppositiously in the spirit of Kant, my own objections to Kant. 70 TEXT-BOOK TO KANT : correspondent sensuous complexions (said complex- ions being, as must always be borne in mind, only subjective affections, feelings, or perceptions, of our own within us) : they are notions substituted for the indi- vidual moments of each general logical function — pure notions of the understanding, that would arise from said functions, in their various moments, being ap- plied synthetically to the sensuous or perceptive complexions of experience. Under quantity, the setting of totality and universality, of plurality and particularity, of unity and singularity, as parallel respectively the one to the other, will, presumably, present no difficulty. The analogy between affirma- tion and reality is equally obvious. Negation is alike in both tables. Then the function in an inde- finite proposition is really one of limitation. The soul is not mortal, for example : what I have really accomplished here is only a certain limitation ; the sphere predicable of the soul is limited by the pro- position ; a limit has been set down exclusive of everything that is mortal. The peculiarity of the proposition is, that Ave have veritable affirmation pro- duced by a negative predicate. That the problematic pairs with the possible, the assertoric with the existent or actual, and the apodictic with the necessary, may also be accepted at a glance. The two first moments of relation we have already discussed, and there remains for our consideration only the production of the category of reciprocity by function of the dis- junctive judgment. This, also, is easily made clear. Take the dis- junctive proposition, The world exists either through blind chance, or inner necessity, or an outer cause. It is evident that these three clauses constitute a sphere, a whole sphere, and that, for exhaustion of THE REPEODUCTION. 71 this sphere, completion of this sphere, any one clause is necessary to the others. It is also clear that the acceptance of any one clause is the exclusion of the rest, or that the exclusion of any two clauses is the acceptance of the third that remains. In short, a mutualness or reciprocity of action and reaction is evident among these clauses ; and there can be little difficulty in conceiving that such an intellectual function as is there involved, being applied in corre- lation of phenomena, would educe the category of reciprocity between the active and the passive. In fact, we have but to reach the one general idea concerned in all this, to reach also the central insight not only into the nature of a transcendental logic, but into that as well of our whole inquiry, and, very specially, of our answer to Hume. The same under- standing that, by its system of functions (judgments), analytically conjoins notions, avails to introduce, by the same functions synthetically, necessary objective conjunction and connexion into the perceptive details of sense, as present in consciousness whether generally (space and time) or specially (actual sensation). This is the key-conception of the entire enterprise, and what concerns pure or general sense is but corol- lary and complementary. This is the answer to the Quid juris ? — this is the explanation of all apodictic validity, whether empirical or other. What we have desired to do, then, we hope will be now clear. We have found out the various functions of judgment or the understanding; and we have seen that the operation involved in each is a synthetic (a conjunctive or conjoining) unity. We have decided, too, that the whole business of sense is limited to receptivity, while synthesis, combination of any kind, can come from the spontaneity of the understanding 72 TEXT-BOOK TO KANT : alone. Then we have perceived that the a posteriori elements of our knowledge (and these constitute the great bulk of all and any knowledge) are but con- tingent affections of our sense, whose correlative objects-in-themselves (if any) are wholly denied us ; which affections, then, on being received within, re- quire from within the aid of the regulating function of fixed and necessary principles. But just such capability of reducing a variety, a multiple, a plur- ality, a detail, a complex, a multiplex, a manifold, of sense we discovered to be contained in the functions of judgment; the consequent notions due to these functions, when synthetically applied to any manifold, readily suggesting themselves. There, then, is the want ; and here is what is necessary for the supply of the want: what more reasonable than to bring the one to the other, and transform into the unity, and simplicity, and order of a connected and articu- lated whole of experience, the infinite a posteriori variety, by means of a certain number of original patterns, rules, or standards, under which judgment, with its own necessary insight, should subsume it ? What more reasonable than such conception ? What other source or explanation can we find for that peculiar jas, that apodictic validity, which is cer- tainly present for us, not only in what are called the pure sciences, but just in the ordinary facts of our current, hourly experience? How otherwise can affections that are within be subjected to law, than by functions that are also within? Consider, too, the success of the undertaking. With these categories and the two general-sense forms, we really exhaust the whole field of the apodictic. Any other apodictic principles, namely, will readily subordinate them- selves under those primary ones, taking up the THE REPRODUCTION. 73 position of derivatives ; to which, whether springing from union with one another, on the one hand, or with the forms of sense, on the other, we desire to give the name of predicables. Such are power, action, passion, origin, decease, alteration, etc. As for the primary principles in allusion, we call them categories or predicaments, in view of the rela- tion which they bear to the so-named classes of Aris- totle. These, however, unlike our own, owe their origin to no systematic principle, but seem to have been caught up and set down rhapsodically, as it were. Now, no bare estimate of some such mere mechanical aggregate, gathered, too, and finished, as it might be, at hap-hazard, will, evidently, at all avail in such a case. On the contrary, we must have the systematic guidance of a common principle, and in such manner as to have a voucher or guarantee of the perfect legitimacy and exhaustive completeness of all the members or elements which shall distribute it. Such principle has been found for us in judg- ment. But the predicaments and post-predicaments of Aristotle will be found to constitute a mere indis- criminate medley : some are not pure, as motion, for example, and some relate to sense, as where, when, etc., or, like action and passion, are merely deriva- tive. The great objection of the general reader to our rationale in explanation of apodictic validity of syn- thesis will probably concern the empirical peculiarity in form and connexion of the objects themselves. It will be felt, for example, that the reason is not in me, but in the objects themselves, why one, a stone here, has this shape, and another, a stone or a brick there, has that one. So, in the series of cause and effect, it will be felt that we cannot dictate the terms of it, 74 TEXT-BOOK TO KANT : and that, consequently, it has a necessity of its own. And again, it "will seem very unnecessary to make the notion quantity depend upon an internal principle when we have just to use our eyes to see it before us in the objects themselves. As for quality, the general reader will probably think it absurd and quite super- erogatory to make an a priori provision for a matter that can only be known a posteriori, and that is wholly a posteriori. These are the difficulties of uninitiated empiricism, however. Once for all, were it with things in themselves that we had to do, we could have no a priori knowledge whatever. In that case, to have any particular knowledge, we should simply have to wait for the presentation of every separate object ; and all and any knowledge, con- sequently, would, as after the fact, be simply a pos- teriori, and, as a posteriori, necessarily only empirical and contingent. But as we can perceive only through affections of sense and forms of perception, and think only by notions which relate to objects through these affections, we can neither perceive nor think things in themselves, but only phenomena, only appearances. As also we do possess elements of knowledge apodic- tically necessary and universal, it is evident that we are not limited to an a posteriori knowledge, but possess also such as is manifestly a priori. This a j riori knowledge is found to unite itself to the a posteriori, and to effect there results of the most ex- cellent and, indeed, indispensable nature. The a priori and the a posteriori are found to be mutually correlative and complementary, so that either by itself were inane and futile. It stands to reason, then, that the contingent a posteriori requiring an element of necessity, such element is provided for it in the apodictic a priori. Nor, indeed, is there anything THE REPRODUCTION. 75 easier — the phenomenal nature of the a posteriori and the consequent necessity of an apodictic a priori being borne in mind — than to clear away all the special objections we have mentioned. It is quite conceivable that the particular form of affection set up varies with the object in itself; but, though this form takes up a certain length and a cer- tain breadth and a certain thickness in space, it does not follow that length and breadth and thickness themselves, as such, come from the object in itself at all. Were it so indeed, then, manifestly, elements of apodictic necessity (quantities) would spring from ex- perience, which is quite impossible. So as regards the particular causal form set up — for the particular complexion is but a particular affection — it is quite intelligible how the particularity might depend on the object in itself, at the same time that the general vis consequently should not come, nor be able to come, from the object in itself at all ; for did it so come, a jjosteriori, that is, it were self-contradictor)' to affirm necessit} T and universality of the relation at all. Surely it is conceivable that the peculiarity of the object always induces a peculiarity of modification in our sentiency, and that certain modifications are in such complexion (why, of course, depending on the object in itself, we know not how) that we subordinate them under the relation of antecedent and consequent, or cause and effect. Nay, it is not even necessary that we should always couple (or categorize) rightly ; what we want to do is only to show, when we do couple rightly, the origin of the necessity that then obtains. The particular form or complexion that adjusts itself so and so in time and space must depend upon the object in itself; but it obeys the forms of time and space, as such, and it is reduced to unity by the 76 TEXT-BOOK TO KANT : power of an original pure notion of the understand- ing, that is quite general and universal, be the par- ticular what it may. The object in itself giving rise to the peculiar modification or complexion of our sentiencv that arranges itself so as to invite the action of the category cause, we know not at all ; we only know that this category, so manifesting itself, is a general function of judgment under which the modi- fication or complexion is subsumed into an object, as simply a case of the rule. So it is as regards quan- tity : did we derive the conceptions as well of quantity in general as of unity, severality, totality, etc., from things without us, these conceptions would not have an apodictic, but only a comparative universality. It is not so certain, either, that these notions could be so derived : the sensuous procession can be conceived to pass before us perfectly well without suggesting any thought of causality, or even of quantity. There is nothing supererogatory, then, in making provision in the mind for such principles in articulation of the phenomena. Being, indeed, of an apodictic nature, they cannot be derived from experience at all, but must be a priori. Nor is it differently situated with quality. We possess sense, and every affection of that sense is quality ; and reality, negation, limitation, can be very readily seen to reconcile themselves to what must be the absolutely pure and universal mo- ments of sensation ; for it is evident that every sen- sation has elements that correspond to these moments of quantity ; and there is nothing absurd in providing a priori necessary distinctions for the classifying of sensations themselves. 1 1 As said already, I conceive myself to answer, in the above, my own objections to the scheme of Kant, and in the spirit, presumably, of Kant himself. Elsewhere it will be found that I hold in the end by my own objections, and that I reject, consequently, my own answers. THE REPRODUCTION. 77 The reader by reflection will perceive that our great levers of argument are — 1, the phenomenal nature of objective knowledge, and, 2, the fact that we do possess apodictic synthetic principles. These levers he will do well to use for himself. Still, we hope to accomplish conviction for every reader by the end of the following Book, in which it will be found we subject all our materials to a final examination, and articulate them together into a systematic whole of such perfection and completeness as ought to go far to prove itself. Such, then, is the nature of our deduction — a deduction of the peculiar authority which certain facts apparently in experience seem to challenge (deductio aut declaratio aut explicatio juris) — a transcen- dental deduction, therefore, which can expect success for itself only in investigating the a priori elements even in empirical perception. That of Locke, on the other hand, we may name an empirical deduction of the fact (deductio, declaratio, explicatio facti) ; which, indeed, can present no difficulty, as experience is nothing but an aggregate of examples of the principles sought to be explained. In this, however, it never struck Locke that he was providing a genealogy for all facts of knowledge which, in the case of some of them, would prove quite inadequate to the dignity of their pretensions and the wide sway of the powers they arrogated ; for no mere physiological theory can account for the existence of those principles named apodictic synthetics. Nor, it may be remarked, w r as Reid one whit more successful. Hume pointed out a principle bringing with it a peculiar claim, which claim he demon- strated, on the principles of Locke, to be incompetent to it. Reid rose in wrath and defended the instincts 78 TEXT-BOOK TO KANT: which were implanted by God. But this for answer is simply Hume's own. He emphatically recognised the natural instinct ; but if there was to be a reason for the necessity of insight which we all acknowledged to lie in it, then he challenged the production of it. This, however, would seem to have escaped the notice, not only of Reid and his followers, but of everybody else as yet. Our explanation, then, is such as endeavours to meet Hume's question really as it was meant : it attempts to produce the reason, the insight, by and in which the mind makes always a necessary transition from the effect to the cause. Our explanation, in effect, does more than this ; for it converts the one of Hume into the all of truth ; or, in other words, it univer- salizes Hume's single proposition, and ushers universal mankind, through the doorway to which Hume led up, into a mighty, marvellous, undiscovered region, in which are seen the fixed foundations of the whole huge universe. Book II. — Judgment. [Ending precisely where the translation and the commentary end, this Book would seem, if exhibited in extenso, quite excellently calculated — by completion, to wit, of a threefold statement — exactly to accomplish the purport of the volume as an explanatory text- book. The relative explanations, however, run out into so great a length, and I have already so profusely discussed elsewhere the particular subject principally concerned, that I feel induced, if for nothing but the printer's paper (to say nothing of the reader's patience), to confine myself here only to certain extracts — to such extracts, namely, as shall seem light-giving in themselves, or, especially, as shall bear to defend the main principles of Kant from my own adverse criticism.] The object of our inquiry is, as we have seen, How THE REPRODUCTION 79 are a priori synthetic judgments possible? Or how, as we may otherwise put it, is it possible for us to add to a subject, in independence of experience and absolutely a priori, a predicate nowise already implied in the subject, and not possibly deducible from the subject by any process of analysis whatever? How is it possible to affirm something of something else (B of A, for example), unless we have learned the connexion — learned it either from experience, or from analytic consideration of the state of the case? How is it possible, that is, to affirm B of A without the aid either of experience to try the fact, or of analysis to demonstrate a presupposition ? How is it possible to affirm apodictically and yet synthetically? To the trenchant distinctness implied in these questions we have been gradually conducted by a minute analysis and ultimate generalization of Hume's problem in regard to the relation of cause and effect. No previous investigation of any cause whatever (A) will suggest to the investigator any effect (B) ; and no investigation of any actual nexus whatever as between a cause A and an effect B will ever surest to the investigator the reason of this nexus : l how, 1 These, of course, are the assertions of Hume ; and they have been universally admitted and adopted since Hume, not only in this country, hut I suppose in all others. In this country, however, the first note in controversion (so far) of Hume was sounded in the second edition of my As Regards Protoplasm. The fulcrum of my controversial effort was there referred to a suggestion of Hegel as to identity in instances of finita causality. The suggestion is perfectly distinct in Hegel ; nevertheless, as it does not in the least appear to he Hegel's special theory of causality, but to have distinctly, rather, the character of a partial remark, it has pretty well escaped notice and been allowed to pass by. I was happy to observe, on consultation, that Erdmann makes the identity peculiarly prominent : it is express in Rosenkranz, too. Still, neither by Erdmann nor by Rosenkranz, and quite as little by Hegel himself, has the suggestion, to my mind, been carried home, as it were : it is not by any one of them regarded in connexion with the statements of Hume. I am told that Mr Lewes had observed the reference in my As Regards Protoplasm, and had 80 TEXT-BOOK TO KANT : then, is it that we attribute B to A — or how is it that on the appearance of A we always expect B ? — and this apodictically, or by rigorous necessity and with absolute universality ? To this Hume and Reid gave each his own answer, both being in effect identical. We, for our own part, postpone our answer till we have looked about us over the whole field implied by the question. Is Hume's problem founded on a fact, and, stated quite generally, what does that fact amount to ? The fact is unquestionable, and the ultimate generality it may be raised to is this : It is a fact that we apodictically aver things which are neither inferential from ex- perience nor demonstrable to have been a priori implied. They are not inferential from experience, for they are set forth as apodictic affirmations; and to these experience is wholly incompetent. They are not demonstrable to have been a priori implied, for no analysis can deduce the one from the other. The general fact, then, is evidently of a very unusual and interesting nature ; and it imports much that we should thoroughly understand and come to be fairly at home with it. Were it to be demonstrated by ex- perience, the process of proof would be of the nature of a synthesis ; that is, we should see from experience, then, -without note or notice, bodily carried it across into his Problems. If Mr Lewes did so, it "was probably because he was perplexed between what was to be called specially Hegel's and what possibly mine in the matter. If the thing were current among Hegelians, it would not do to speak of it only in my reference, etc., etc.! I am prepared to show, however, from widely-separate passages, and in widely-separate philo- sophies, that identity as the link of causality was, overtly, expressly, ostensively, the doctrine of the Hindoos. There is no doubt whatever that Hegel made his relative remark in consecpience of what he had seen in them. Mr Lewes, if he had known all this, would have still known that, as concerns this country, as concerns a full explicit con- sciousness, as concerns Hume, there was at least some little suggestion of merit, even at last on my part. THE REPRODUCTION. 81 from actual fact, that B attaches synthetically to A ; or we should just learn from experience that A and B together constitute a synthesis in actual fact. Were it to be demonstrated to have been a priori implied, the process of proof would be of the nature of an analysis ; that is, we should be able to see from intellectual trial that B was analytically implied in A. Now, without either the synthesis of experience, or the analysis of reason, how can we possibly pretend to any such knowledge ? We cannot attain to it by experience, and, consequently, we are excluded from all possible a posteriori process, whether synthetic or analytic. We cannot attain to it by a priori analysis, and therefore we seem wholly cut off from any pos- sible approach to the fact, whether a priori or a pos- teriori. Must w r e, therefore, cover up the mystery from our eyes by simply writing over it "instinct"? — sceptically with Hume, dogmatically with Reid ; in- direct and a posteriori with Hume, direct and a priori with Reid (not but that Hume's "instinct" here is ab- solute enough and need not be seen to rest on his "cus- tom"). Ah, but do we not see that, if, as regards the a posteriori, we are excluded from synthesis as well as analysis, the case is quite different with the a priori, where as yet only analysis is spoken of, and we are still free to put the question, But how of a priori synthesis? May not we ourselves, may not our minds, may not our separate faculties, have power to add — have power synthetically to add predicates to possible subjects, which predicates w r ere neither al- ready implicitly contained in said subjects, nor such as could possibly have been determined by any ex- perience whatever. This, then, is the ultimate theme of investigation : Intellectual a priori synthesis ; its nature, conditions, and limits. 82 TEXT-BOOK TO KANT '. Of the existence of such we are not now permitted to doubt; for causality alone lies there before us, clear, obvious, undeniable, in the great highway. Our inquiry, then, has narrowed itself to the a priori, and, closer than that still, to the synthetic a priori. In fact, if there is to be an a priori at all, there must be a synthetic such ; for the analytic a priori, de- pending on a process of regression, would be speedily exhausted by said process itself, were it not supported at last and definitively on a priori synthetic facts ; for it is evident that the regression of analysis must end in an ultimate fact that is either a posteriori or a priori. Be it a posteriori, it has no relation to our inquiry ; and be it a pjriori, then it is synthetically so, for the analysis is by supposition terminated, and it has no other possible termini than these named. The ground- fact, then, for metaphysical, or even psychological inquiry, is — The possibility of a priori synthesis ; and that amounts to — The possibility of principles of necessity and objectivity towards such subjective sense-experience as this of ours. On the principle of causality we judge that B belongs apodictically to A : this judgment we cannot found on experience, for experience is only contingent ; neither can we found it on any a priori analysis of A, for the notion B is by no means implied in the notion A : it must be founded, therefore, on a priori synthesis — the judg- ment that affirms B apodictically of A founds on a principle of synthesis a priori. To exhaust and com- plete our whole subject, consequently, we have simply exhaustively to demonstrate and tabulate all such principles. But how and where to find these principles can present no difficulty; for they are a priori, they are products of the mind, they are contributions THE REPRODUCTION. 83 of the faculties, and will give themselves to view should our process in consequent analysis be but exact enough. And, indeed, we have already accom- plished this ; for, of the two complementary factors of knowledge, — sense, by which objects are materially given, and understanding, by which objects are for- mally construed, — we have now discovered all the a priori contributions, and by means of processes in regard to which, on the score of infallible accuracy and exhaustive completeness, we are not permitted to doubt. The possibility of a priori synthetic judgments, then, is seen to depend on the existence of certain universal and a priori forms both of sense and under- standing ; those of the former being pure perceptions (space and time), those of the latter pure notions (the categories). It remains for us to see how the a priori synthetic judgments themselves result from the union and combination of said pure notions with said pure perceptions. For it is natural to suppose that, the mind once for all containing pure notions and pure perceptions, these will not remain apart, but, the former subsuming the latter, a system of a priori judgments will result — a universal schematism — to the conditions of which the whole subsequent wealth of the a posteriori— chaos else — w r ill be obliged to submit itself. For it is worth while remarking that the nature of a judgment is by no means accurately specified by affirming that it is the comparison of two ideas, unless the correlativity of these two ideas be also specified. These two ideas, in fact, are always correlatively so situated that the one is higher than the other, and that, by consequence, the latter is subsumed under the former. There is, in truth, no judgment that is 84 TEXT-BOOK TO KANT : not the subsumption of a particular under a general. In this process, however, judgment itself, the faculty, exercises a merely formal function : it is simply the operating power, the agent that, plying between two things (a and A), brings the one to the other, but discriminatively so, — that is, in such wise that the mere example or particular case a is reduced or sub- sumed under the general principle, law, or rule A, and not vice versa. It is as if it were thus : Sense supplies an infinitude of particular cases, understand- ing a finitude of rules, and judgment, in exercise of its discriminating and associating function, subsumes the former under the latter. It is thus easily seen how, as indeed belongs to the proverbial wisdom of all times and of all peoples, judgment is a faculty that is strong or weak just as given us by nature. Instruction, books, schools, etc., may, in a thousand different ways, make us thoroughly acquainted with the rules, but still it depends wholly on our own particular judgment whether we shall subsume under the rules truly or not. Every medical man knows well enough, by name and nature, all the possible genera and species of disease ; but it is not every medical man can rightly subsume the case of each particular patient. It does not follow, then, that, if only cognisant of the rules of the understanding, judgment will forth- with duly subsume. To that there is necessary a certain peculiarity in the faculty itself. No rules can guide what is itself the art of using rules. Or the rules lie there before judgment, which, as concerns choice, action, without assistance from them, is left to its own self and its own powers ; a mighty source of difference between man and man is herein indicated. The general conception so far is that, the pure THE HEPllODUCTIüN. 85 forms of perception in the peculiar varieties of their characteristic details (units side by side, units after one another, etc., etc.), as on the one side, being sub- sumed by judgment under the pure forms of under- standing, as on the other side, there will result a primitive, pure, a priori mental schematism, sub- jacent to, and modificative of, experience ; of which schematism the expression in words will be certain propositions, certain ground judgments, which will re- spectively show as axioms, anticipations, analogies, and postulates. By the word subjacent, we would wish it to be understood that the forms so characterized are not to be conceived to precede, so much as to underlie, experience. Till experience there is no mental life whatever; and all a priori schematism must await the stimulus of experience before it can realize itself in actual operation. Nevertheless, though excited by experience, and wholly calculated for, and directed to, experience, it is quite independent of experience, and takes place in obedience only to its own con- ditions; but forming so an a priori ground-net, a fundamental dirradiation on which, and according to which, the world of experience deposits itself. We thus see that, from this provision of pure forms both of sense and understanding, there is subjacent to experience an entire a priori ground-system, the special peculiarities in the formation of which we have now more particularly to consider. In the preceding Book (Simple Apprehension) we have seen the transcendental forms as well of sense, as of understanding : those forms without which no empirical contribution, no actual sensuous matter, could become for us a perception, or what we call an object of experience. There it has been proved 86 TEXT-BOOK TO KANT : that without the unity of apperception, the synthesis of imagination, and the conjunctive function of the categories, this whole daily life that we name experi- ence would be impossible for us. There also has it been proved that every single empirical particular, introduced by the channels of special sense, must take on the form, or rather adapt itself to the mul- tiple, of pure perception in its two modi of space and time. We know now perfectly, then, the various parts of the a priori machinery, the various members of the transcendental apparatus, which is transcen- dental just for this, that it is at once a priori and a posteriori, i.e., a priori in origin, but a posteriori in use, or as it presents itself in actual fact in the various things we perceive, in the various objects of experience. We have seen all this ; but we have not exactly seen yet the precise operation of the machinery : we have not exactly seen yet how the various parts are fitted into each other, how the various movements are co-ordinated so as to complete the wholeness of the entire fabric and the unity of its function. This, now, is all that we have once for all to see — how the particulars combine and work together into the general. In order that what we name experience may be possible, such and such an appa- ratus must be presupposed ; and we have now only to see how it actually works. We have experience ; but experience involves such and such subjective conditions on our part, else it would be impossible. These, then, are to us the possibility of experience — the possibility of such an experience as this of ours. It becomes us, then, to classify these conditions, or to separate and arrange them towards an explanation of their particular action, and the manner of it. Briefly, we have found that the various parts of experience can be THE REPRODUCTION. 87 grouped under two forms: sense and understanding; for the latter term includes apperception, imagination, etc. The characteristic of sense is plurality, multi- plicity, variety, maniness, — a manifold, a complex, a detail of constituent terms, members, particulars, or parts. That, in a word, is the characteristic of affec- tion as affection, as also of the general forms under which alone it can come into consciousness. The characteristic of understanding, again, is singleness, wholeness, simplicity, unity, oneness. These two agents, then, evidently mutually complement and complete each other. Understanding adds to the many of affection the unity of function ; and without the latter the former would remain a disarticulate, unintelligible blur. Sense is merely receptive and, consequently, passive ; the matter it yields would never present itself as knowledge, perception, were it not re-acted on by the spontaneity of the understand- ing. For it is not things in themselves that ever come before us, but only the states or affections of ourselves set up in us we know not how, by objects in themselves we know not what; and these affec- tions, which we call empirical, as being produced a posteriori, to constitute the connected rational whole, named by us experience, must be combined within, and arranged according to the laws of the understanding within, — of that understanding, namely, into which they are received. What we have now specially to see, then, is how the multiple, the many, of sense is reduced into the simple or unity of the understanding. In fact, there is at once considerable difficulty suggested by the question that presents itself here. How are objects, how is empirical matter, which is wholly an affair of sense, to be brought under the 88 TEXT-BOOK TO KANT : understanding, which, contrariwise, is wholly intel- lectual? There must, surely, be something inter- mediate interposed, if there is unity to be established between two such discrepant extremes. But just such intermediate element do we possess in what we have discussed as pure perception, the forms, namely, of space and time. For these, if on the one side a priori, like the forms of the understanding (the categories), are, on the other side, sensuous, like the matter of special sense : in them and with them docs all such matter present itself. They are at once not less intellectual and a priori on the one hand, than sensuous and, as it were, a posteriori or empirical on the other. We can conceive, then, the possibility of the cate- gories acting directly on the multiples of space and time, and, through these, on those of empirical sense. It is evident, too, that, of these forms, the inmost and most general is time ; for into it all matters of sense, both outward and inward, must be received. If, then, the multiple or multiples of time be conceived as susceptible of the uniting and connecting influence of the categories, it is evident that there is a possibility established of conveying this influence to the whole infinitude of particulars that present themselves in actual empirical sense. We can thus conceive a system of schemata produced and brought about by the operation of the pure forms of the understanding on the pure forms of sense. This operation of the understanding might appropri- ately be named the schematism of the understanding; and to that schematism, to these schemata, it is evident that all other contributions of sense must submit and subject themselves. The schema, we may remark by the way, is dif- THE REPRODUCTION. 80 ferent from what we may name the type or image. The type has a certain definiteness and particularity about it, while a certain indefiniteness and generality attach to the schema. Five points, or five counters, or five pips, or five fingers, are severally a type ol* image of the number five ; but any triangle in general you can construct or think, any horse or dog in general you can imagine, can by no possi- bility represent a type ; it must remain a schema. The general notion triangle is simply a conceived formula whereby you can construct a type ; but it is itself a schema, for it is of no single form, — rather it is of an infinitude of forms. To be an absolutely general notion, namely, the triangle must be neither scalene, nor isosceles, nor equilateral ; neither right- angled, obtuse-angled, nor acute-angled : it must just be a triangle in general ; that is, not a type, but a schema. So with the general notions, dog, horse, man, etc. : these are not types, but schemata. The type is a single image or figure set up by the em- pirical imagination ; whereas the schema is an absolutely general formula for the production of a whole family of types : it is a monogram of pure imagination. Evidently, indeed, the entire operation alluded to here is one of the deeply-hidden arts or processes of the human soul. 1 1 Kant here is seen to make an easy end of our ordinary modern nominalistic quibbling. " An idea," says Berkeley (Prins. Hum. Knowl. Introd., 12), " whicb, considered in itself, is particular, becomes general by being made to represent or stand for all otber particular ideas of the same sort — a line which in itself is a particular line, is nevertheless with regard to its signification general — a line in general." Hume characterizes this proposition of Berkeley, that " all general ideas are nothing but particular ones annexed to a certain term," as " one of the greatest and most valuable discoveries ever of late made" (T., i. i. vii.) ; and, in the Enquiry, section xii., he says further, " Let any man try to conceive a triangle in general, which is neither isosceles nor scaleuuin, 90 TEXT-BOOK TO KANT I We can conceive, then, the possibility of the con- junction of pure understanding and pure sense pro- ducing schemata, and we can further conceive the conjunction of these schemata with empirical multiples (actual sensations) producing the arranged world of experience. Now it is judgment that in both cases will produce the conjunction: it is judgment that will subsume the particular empirical multiples under their respective schemata ; and it is judgment that will subsume the pure multiples under the pure notions to the production of the schemata themselves. Now, the action of judgment in all this is simply formal ; it adds nothing, it merely brings together and subsumes the relatively particular under the relatively general : it decerns the casus legis datce. We cannot lay down a rule, then, with a view to guide judgment; for, under such rule itself, judgment could only again subsume, — standing in need, then, of yet another rule to guide it there! In point of fact, judgment is what we call mother-ivit, and is incapable of being learned. No instruction can supply its want ; and that is the secunda Petri, the pars altera Petri, — stupidity, — that which is the characteristic of the blockhead, dolt, dunce, etc. The statesman, the jurist, the physician, may be very learned men, and possess a thorough knowledge of every general rule ; but, without judgment, they cannot discern the par- ticular cases that apply to the general rules, and so nor has any particular length nor proportion of sides, and he will soon perceive the absurdity of all the scholastic notions with regard to abstrac- tion and general ideas." Dr Thomas Brown (Lect. 46) very clearly shows that Berkeley here only implicitly accepts what he explicitly rejects. As for Hume, again, when he defies a man to make a schema a type, he actually fancies that he is exploding all scholastic absurdities ! Surely it is common sense to see that a general idea involves in imagination only a schema, and that a schema there is not a type, but a general receipt for a whole infinitude of types. THE REPliODUCTION. 01 are unable to subsume. Judgment, however, if it cannot be learned, may be exercised or practised, and accordingly strengthened and improved. For a cer- tain familiarity with examples is necessary to the very best judgment. At the same time it is to be acknowledged that, as no example can come up to the rule, a too great commerce with, and confidence in, examples, are apt to confuse and taint the pre- cision and purity of the rule. And thus examples become, as it were, the go-cart of the understanding, and almost counterbalance by disadvantages the very advantages of them. What we have here, then, in this transcendental loiric where there is matter as well as form, is a doctrine of judgment, a doctrine of the faculty as it plies between the two, as it exercises its function of bringing the one under the other. But what that means is that certain pure multiples of time, being brought under correspondent categories, will give rise to an equal number of similarly correspondent schemata. Taking the categories in hand, then, we have simply to find out what multiples of time re- spectively correspond to them, and the relative schemata will at once show. Now, the succession of time is at once a sensible multiple of what intellectual multiple is implied in the notion of quantity : it is the homogeneous coming together of like with like ; but that is number. As for quality, again, no multiple in time itself can be found to correspond to it ; nevertheless we are quite entitled to take the absolute universal of sensation as sensation, and conceive time filled by it. But so con- ceiving, it suggests itself at once that there is a pro- cess in this filling from any imaginable amount down to zero, or, equally, up from zero to any imaginable 92 TEXT-BOOK TO KANT: amount. That now is degree, and degree will be the schema of quality. By degree we mean the coming together of filled moment with filled moment of time ; as quantity, which is simply the homogeneous addi- tion of pure to pure, is, so to speak, literally sensualized in the mere succession of time as a homogeneous addition of bare moment to bare moment. We may understand, therefore, that, in this latter case, any homogeneous empirical succession is not left in its mere indifferent separation, but, unit being con- nected with unit in the synthesis of imagination, collapses into the unity of apperception through the notion quantity. For it is evident that quantity is the notion under which any such multiple as that which is represented in the succession of time will fall to be subsumed ; which notion, indeed, it is equally evident, is essential to produce unity in any such multiple, the mind otherwise being only passively filled with a perpetually fleeting sequence. Xor is it less manifest that, in quality, a corresponding function is required in order to induce unity on the amount exhibited in the filling of time. Under these two headings (quantity and quality) we have only two schemata ; there is only a schema for each general rubric, and we might expect one as well for each of the six subordinate categories. The reason is that quantity can have in kind only one multiple ; it can differ only in amount (of extension) ; and that quality, similarly, can have also in kind only one multiple ; it can differ only in amount (of inten- sion), and that is in degree. But will this be the case as well with the categories that follow? What of the next category, for example, relation ? Is there only one kind of pure relation, or are there several ? If we glance at the particular categories, I think we THE REPRODUCTION. •'•» shall expect a different schema for each ; and for this reason, that each represents a very different relation. But that necessitates for each different rela- tion a correspondent!)' different multiple. Now, will time be able to supply this? It was no act of usurpation to assume time as (in its succession) a type or exemplar of number and quantity ; and it was certainly perfectly justifiable to regard time when put in relation with its absolute generale of filling: we cannot rightly think usurpation, or assumption, or presupposition, or begging the question, ofthat either. But, really, the multiple of time seems so uniform, simple, and monotonous, that one fears for the possi- bility of extracting from it more in that reference than we have already extracted- one fears, indeed, that it must be altogether impossible and out of the question to extract from it actually no less than three more modifications, and these very special ones too. Sub- stance and accident, cause and effect, action and reac- tion : how find in a mere flux of homogeneous units three special multiples that shall be sensible types of the intellectual multiples implied in these very peculiar — and peculiarly different and distinct — in- telligible relations? We can still, of course, allowably resort to space and the generale of the faculty of ap- prehension for a filling in time. As much as that is certainly to be regarded as a priori and pure, and quite legitimately at our disposal. If we can even wring out of time, and in these references, multiples to suit the three relations in view, it will not be com- petent to any man to except or reject. Let us, then, take said three relations in their order, and correspon- dently examine time, in connexion as well — if neces- sary — with its allowed lemmata or postulates. First, then, can we find so any multiple that will 94 TEXT-BOOK TO KANT: correspond to the relation of substance and accident? Why, yes. Time is a flux and yet it abides. It is as though time, to its own broken and fluent side, turned ever a whole, unbroken, and permanent side. Time is the vast unchanging ocean that gives foundation, place, and room to all its own bubbles. Time as a whole stands in the relation of substance to its own fluent parts, which are as accidents. All passes through times in time, but time passes not : time is as substance, then, and times as accidents. Nay, reality as reality, space itself, can be legitimately assumed as the a priori permanent substrate that, in time, cor- responds to time itself. A void time were unpereeiv- able. Time and times are discernible only on occa- sion of the experience of empirical matter. Time and times, then, are perceived, and they are perceived in this mutual relation. But they were so perceived only through mediation of empirical matter. Said matter itself, therefore, is in the same relation. Or there is a substrate permanent, fixed, invariable, constant, that corresponds to time; and there are modi, unfixed, variable, transitory, which correspond to times. We see. then, that filled time and filled times relatively constitute the sensuous multiple that will correspond to the intellectual multiple which is thought or implied in the notion substance. Empirical realities, to become realities for us, or to enter into and be united with our apperception, will assume this multiple in time, and collapse to unity under the notion. Xow. the principle that turned up under quantity we named an axiom ; for it conditioned the very possibility of objects — no object could be an object for us that did not present itself formed on that principle. Again, the principle that turned up under TUE REPRODUCTION. 9. 3 quality we named an anticipation; and this, too, involved conditions that necessarily entered into and manifested itself as constituent and constitutive of the object. These two principles, then, may be named mathematical ; for they form a large portion of the very structure of every object. The principle in- volved under substance, however, is a relation, and asserts only that empirical realities will be related on analogy with the logical function of judgment implied in substance. This, then, is not a principle that appears as an ingredient entering into the object itself, but a principle that relates or connects object with object, — in other words, a principle that regulates, but does not constitute (or enter into) the pheno- mena — a principle, therefore, that is regulative, not constitutive. Again, and for the same reason, it is not mathematical but dynamical ; it relates to the com- portment of existence to existence mutually ; it does not determine any existence as such. Relatively to any existence as such it is contingent ; it prescribes nothing a priori which will be found necessarily in each and every existence. Relatively to exist- ences mutually, however, it is necessary and pre- scribes their common reference. The difference of its evidence is thus plain : objects do not present themselves according to it as so and so mathematically formed or qualitatively constituted, but as thought in such and such mutual relation, necessarily, on analogy with a certain function of judgment. And yet it is not so much a relation connecting all realities the one with the other, as a condition of such relation. It is simply this, that, in all the vicissitude of accidents, substance abides. In a word, for the analogies and the postulates we may preliminarily sum up thus : Time relatively 96 TEXT-BOOK TO KANT : viewed, but still only in its own self or abstractly, pre- sents three multiples, which, otherwise also, may be regarded as its three inherent relations ; and these are respectively duration, change, and interchange. Again, time relatively viewed, but concretely or with refer- ence to its filling, similarly exhibits another triplet of relations. Said filling, namely, may be possible, or it may be actual, or even necessary. Time, in relation to its own succession, is very conspicuously, as already said, simply a type of duration, substance ; while the succession, in contrast with it, is, as conspicuously, but a type of its accidents. Time, as it fleets from past to present (in which fleeting the present is ever the result of the past, and only reachable through the past), is but type of a reality which being given involves a consequent ; and that is but a schema of causality. Or time, as it fleets, is a type of change ; and change, being the same thing existent and per- sistent under two opposed determinations, implies an antecedent. Time in itself presents no relation of interchange ; but by means of space and the filling of spac% we can see that things may be together. The multiple contemporaneous times, then, is but a type of communion, and communion is the schema of the abstract notion reciprocity. Time, or a time, complete in its conditions, but indefinite, is a type of the agreement of a notional synthesis with the general conditions of time ; and this is a schema of possibility . Time, or a time, complete in its conditions, and de- finite, is a type of existence, and a schema to the notion actuality. Definite existence in all time is a schema of necessity. [Here follows a long discussion in very full detail. As intimated, I only select from it sentences that seem happily to explain or defend.] THE REPRODUCTION. 97 A time is unity of a homogeneous successive multi- tude ; therefore a time is an extensive magnitude ; and in like manner space is an extensive magnitude. But again every object or phenomenon is, though with empirical filling, just a time or a space ; there- fore every object or phenomenon is an extensive magnitude. All empirical multiples, in fact, corre- spond unit by unit with the pure multiples of space and time which contain them ; and as these latter are apprehended by the synthesis of imagination, and conjoined into unity by the category, so must those former. Thus, then, is it that the function of the category is conveyed to empirical matter; and thus, too, is it that such matter contains in it, even as it presents itself to our sensuous consciousness, elements of necessity ; for all such matter, once objectively placed before consciousness, is a compound of ele- ments as well a priori as a posteriori. Or, again, thus is it that all empirical objects contain a priori elements and, through them, an authority apodictic and absolutely universal. It is not things in them- selves we know, but phenomena, into which enter the a priori forms as well of sense as of understand- ing, and what holds of these latter, holds also of the former. Thus it is that we can a priori prescribe laws and conditions to objects which, as a posteriori themselves, have still to be waited for. The necessity of imagination towards the possi- bility of what synthetic processes are involved is obvious ; for through that faculty only can the past items be reproduced for summation with the present : without such reproduction we could have no con- sciousness but of insulated and disconnected units. Imagination, in short, is par excellence the place of ideas; it is the element in which they live; in which G 98 TEXT- BOOK TO KANT : they are received, conserved, and reproduced. But every unit of consciousness, sensuous or intellectual, is an idea. It is in imagination, therefore, that every process whatever of intellection must take place ; or imagination is the universal intellectual menstruum, the universal vehicle of mediation between sense on the one hand and the functions of apperception on the other. Through imagination, acting by means of the pure form of time on the pure form of space — through imagination it is that we have the science of geometry, etc., etc. [Further references to quantity, quality, and sub- stance omitted.] Here, then, we demand, first, what notional mul- tiple is implied in the function itself (of judgment) which is to be the original of the category of cause and effect ? This multiple is evidently a relation be- tween two members such that the one of them, the antecedent, determines the other, the consequent. That is, there is a process implied, with a necessary first and a necessary second. Can time, now, pre- sent a sensuous multiple that will correspond to this notional multiple? Well, the function involves a process; but what is the process of time? It is simply change. The present is, and we feel it ; the past luas, and we feel it no longer : this is change. To be aware of this process of time, we must have been aware of empirical realities in a like process. Therefore there are such, and they assume this pro- cess in time ; for, though we know time on occasion of them, time does not depend on them, but they, on the contrary, depend on it. Still time is not an abso- lute object in which empirical realities present them- selves : were it an absolute object, its contained objects would have each its necessary relative place. It is not, THE REPRODUCTION. 99 then, by the mere perception of time that we perceive the succession of objects. Time is quite apart from these, and they themselves are only related to time by us after we have apprehended them, and according as we have apprehended them. For sense, as it is affected from without, is merely receptive and passive, and all further reference, relation, combination, is product of the various principles of our spontaneous understanding. These affections, in fact, constitute but a succession of sub- jective states, and all that is empirical would appear mere sensuous subjective modification, did not the understanding synthetically interfere. It is this syn- thetic function of the understanding that externalizes mere internal affections, — that throws out our mere inner subjective modification into a very world of outer objective realities. Still, should the question be put to us, we may proceed to admit that, though the synthesis pura constituted by the a priori or transcendental machinery, potentially precedes analysis, an act of analysis must, nevertheless, always be involved in the bringing into operation of the various steps which, materials of sensation being given from without, produce realiza- tion of this synthesis pura into the world of experience. For the synthesis pura is realized only on occasion of experience, and the very first act of experience in- volves analysis. To prove this we have only to con- sider that the synthesis pura consists of various parts. It consists of several ground-multiples of time united, through imagination, by function of several ground- unities or notions (categories), into the one of apper- ception (self-consciousness). The sensuous affections, again (set up in us we know not how, by things in themselves we know not what), will not possibly enter indiscriminately, so to speak, into this system of 100 TEXT-BOOK TO KANT : multiples and unities. They must recognizably pos- sess correspondences respectively adapted to these multiples and unities, and be accordingly distributed and assigned. The a posteriori matter must be as variously modified as the a priori form. Certain affections will be subsumed under this function and its correspondent pure multiple, certain others under that, and so on, till every function is supplied, not only with its own pure multiple (of time), but also with its own empirical multiple (of sensuous affec- tion). That is, judgment will discriminate, in every separate instance, the empirical cases of the pure rule. Discrimination, however, can only take place through analysis. The synthesis and the analysis are simul- taneous, then ; or, indeed, to the production of any empirical object, the analysis, so far as action is con- cerned, precedes the synthesis (emjnrica) ; for the empirical matter cannot enter the non - empirical system of moulds till judgment, acknowledging the right, permits the transit. That indeed seems plain : that, if there is a pure diversity, a correspondent empirical diversity must be admitted and assumed ; for, if not, how could we, without direction or guide, hope correctly to accomplish all these infinite actual subsumptions ? Empirical matter has thus a necessity of its own. We cannot set up our own sensuous affections ; as such, they are independent of us, and as such we cannot alter or reject them. We are subject to their necessity ; we entirely depend on them. But this necessity is for us contingent, in the sense that we do not see into it, that we do not see its rationale and grounds. It comes, and we submit to it, and our submission is absolutely necessary. Still it itself is contingent in this way, that we cannot say, it could THE REPRODUCTION. 101 not have been otherwise. The sense-affeetions I have are entirely contingent. I might have had in their place any others. I only feel that they are as they are, but no more. This — that they are as they are — is, as such, however, entirely independent of me. I have therefore to obey the empirical matter, and group according to its peculiarities. There must be a hint in the empirical succession itself, which determines when quantity, when quality, causality, reciprocity, etc., shall be the proper category to act, and subsume into experience. The empirical succes- sion that is to be grouped under causality, for instance, cannot be the same as that which is to be grouped under reciprocity ; the succession, namely, that calls for a reversible number must surely be cor- respondency different from a succession that calls quite as imperatively for an irreversible number. The empirical variety itself must be furnished with a cue, must, as it were, blow its own prompters whistle, before my judgment can be expected duly to subsume it into the appointed checker. Even when subsumed it will continue to manifest its own individual necessity. The amount of time or space assumed by any empirical object depends on it, the object itself, though its general subjection to time and space depends on me. The particular cpiality or degree, too, of any object depends on it, though the general law depends on me. It is not really an objection, then, the fact that particular shape and particular form are entirely independent of the general, so far as each is particular, though, again, each must assume the laws of the general. The general laws of space are verily in every particular object through me, although the particular form and quantity of this object in space depend entirely on its 102 TEXT-BOOK TO KANT: own transcendental matter. So it is with causality. I can subsume under that notion a sensational or empirical reality only when that reality exhibits its pass, — when it manifests such particular nature as renders it amenable to such particular rule. There is thus, then, a certain pre-established harmony be- tween pure form and empirical matter : the one could never be subsumed under the other, were they wholly disparate, wholly incommensurable. This pre-estab- lished harmony, however, between the empirical and the intellectual does not render the intellectual super- fluous. The intellectual is absolutely necessary in order to raise what is subjective and contingent into what is objective and necessary. In the case of certain categories, however, the fact of even the empirical succession being necessarily credited with the possession of necessity is not to be slurred over, but fairly faced. And here we mean suc- cession as such, succession that remains succession. In quantity and quality, the succession in conscious- ness of the constituent units of the particular detail, complex, or manifold disappears : these units become united into a single object. But in causality and reci- procity it is not with successions of units collapsing into single objects that we are concerned, but with suc- cessions among objects themselves, successions as suc- cessions, successions that are not only at first but even at last successions. Then these successions are different the one from the other, — so different that each must bring with it nothing less than an empirical necessity, an empirical necessity of its own and peculiar to itself. Causality, as we have seen, is distinguished from reciprocity just by this, that its succession cannot be'reversed, while that of reciprocity not only can be reversed, but must. From this it follows that there THE REPRODUCTION. 103 is a certain order in the phenomena themselves, adapt- ing them now to causality and again to reciprocity. That is the difficulty, then. As the order is pre- cisely what constitutes the distinguishing peculiarity of the two cases, why call upon an intellectual order from within for the attribution of necessity, while such necessity is already a manifest possession of the phenomenal order from without? Even phenomen- ally what we term cause must precede what, in the same connexion, we term again effect. Even pheno- menally we cannot transpose the order. And is it not just this incapability of transposition that decides us to call in this category (causality), and not that one (reciprocity) ? Where is the reason for a double necessity, then ? The phenomenal necessity being ad- mitted, what occasion is there for the apparent super- fluity of an intellectual necessity? The intellectual necessity does not even explain the phenomenal necessity ; for the latter must be admitted to be independent of, and to precede, the former. Does not the old difficulty of causality remain, then, and is not the very thing we want explained just this pheno- menal necessity? Before your intellectual category can act, and superinduce its peculiar vis necessitatis, you admit that there must be a fixed phenomenal order. What Hume wants explained, then, — what we all want explained, — is just why is this pheno- menal order fixed? Why is the order always A B, and not sometimes B A ? Or how do we know that it is really so, — how do we know that the order is always A B, and never A C or A D, etc. ? Hume, in examining the whole process itself and every member of it, both before the event and after it, was quite unable to detect the copula that was the vis nexus, the actual necessity in question. He was 1 04 TEXT-BOOK TO KANT : driven, therefore, to look upon the whole business as dependent, naturally, on instinct, or, philosophically, on custom. We simply expected to find conjoined what we had been in the habit of seeing conjoined. And, really, if there can be no objective reason dis- covered, it is good ratiocination to have recourse to a subjective reason. Nor is the subjective reason in question a weak one ; rather it is one that obtains widely, and even deeply, in human affairs. Never- theless, it is quite true, as has been pointed out, that no customary association is adequate to a connexion that is, conspicuously, apodictically necessary and universal ; and Hume's explanation is to be defini- tively rejected. What we have here before us, how- ever, is this, that our own explanation is incom- petent, — that we ourselves seem to fail, indeed, in the very key-stone of our whole system. The presence of apodictic synthesis in contingent matter, — the presence of apodictic synthesis at all, — that we have under- taken to explain, — precisely, too, in connexion with Hume's problem, — and precisely that do we seem to fail to explain. Our answer, after all, — and especially as regards causality, — seems to turn only on a super- fluous and supererogatory pre-established harmony between the facts of sense and the functions of the intellect. The phenomenal necessity being simply assumed, and not explained, an awkward intellectual necessity is then merely mechanically added to it. Nevertheless, we have already said what is neces- sary here. Sensuously we know only the units of a phenomenal succession. That this succession does not remain such, but gets combined into the various objects and connexions of experience, cannot depend upon the receptivity of sense, but must depend upon the spontaneity of the intellect. Let the particulars THE REPRODUCTION. 105 be what they may, the general law and laws which they are found to obey can only issue from within. These laws are still absolutely necessary in order to convert disarticulate units of subjective affection into — just to say it at once — the formed universe. The phenomenal order is really granted to be fixed. Nevertheless, this, in the first instance, is but a matter of subjective judgment, and the necessity it involves is, equally, not more than subjective. Still this is not, exceptionally, so with causality merely. All empirical elements bring with them, in fact, a certain empirical necessity. Leaves are green, rocks are solid, seas are liquid : I must so accept them, I cannot alter them. This is their qualitative necessity, and it is wholly independent of me. In the same way, among these things, and all things, there obtains an empirical order, which is empirically necessary, and which I have simply to accept, without power to move a finger in change of it. There are phenomena in which I cannot return from the second to the first ; and there are equally others in which it is their express distinction that I can take the units of the relative manifolds in any order I please. The former, presenting an analogy to, are subsumed under, the logical process involved in the relation between ante- cedent and consequent, which process synthetically applied to phenomena, is named causality. The latter, presenting an analogy to the logical function that is concerned in a disjunctive proposition, are subsumed under reciprocity. And observe the effects of the subsumption. So long as we have merely the facts of sense before us, we have an unconnected succession ; my imagination, which receives them, finds the one first and the other second, and it is a subjective fact that I always find them so. But this necessity is like 10G TEXT-BOOK TO KANT : that of the greenness of the leaves, etc.; what the things are in themselves that lie under the special quantities, or under the special qualities, or under the special arrangements, — what are the noumena that lie under the phenomena, — I know not all. These things in themselves, these noumena, must dictate : they must prescribe their own quantities, qualities, and arrangements ; but still these quantities, qualities, and arrangements are, as they are in the first instance, or only in sense, simply subjective, merely contingent. As they are I find them, and as they are I accept them ; but still I am quite blind in their regard. What that is that lies under the double we call cause and effect, I know not at all ; and I see no rationale whatever in the duplicity presented until I myself regard it in the light of a logical relation of my own, that of antecedent and consequent. In short, phenomena, while only in sense, must be but subjective and contingent units : they can become objective and necessary only w^hen lifted into the functions of intellect, w r hich functions are alone there beside them, are alone left as any possible agencies to manipulate them further. Thus, then, there are two judgments involved in every act of the perception of experience. There is, first, a subjective judgment that attaches itself only to the phenomena themselves (and, consequently, to the order of them) ; and there is, second, an objective judgment that, through due distribution and assign- ment of the categories, projects these phenomena — our own contingent subjective affections — into the articulate world, into the ruled and regulated con- text of experience. All phenomena are received first into time, and the functions of unity implied in the categories have already potentially converted the THE REPRODUCTION. 107 various multiples of time into schemata — checkers, as it were, for the final completion into experience of our mere subjective sensations. As regards causality, for example, the unity of apperception (self-conscious- ness) possesses a logical function which we name antecedent and consequent (or the hypothetical judg- ment) ; time, again, which is only a form within us of general sense, as it were, of a priori sense, even of pure or non-empirical sense, transcendental sense (a priori in place, but empirical in use) — time, again, possesses, on its side, a multiple or series, — namely, in its potential or dynamical process, — which may be assumed as the type of change ; said pure function of unity and said pure sense-multiple (affection) coalesce into a schema ; this schema receives into itself such multiples of special sense as present an analogous order; and thus it is that certain mere subjective phenomena are perceived as objects bearing to each other the mutual relation of cause and effect. 1 We have probably seen enough of reciprocity in the preceding, and may now pass it. Of modality, the categories possibility, actuality, and necessity are, also, from what has been already said, probably, intelligible enough. They are, so to speak, but the other three categorical classes in act. Fulfil- ment of quantity is to satisfy the requisitions of form, the requisitions of the understanding, and that is what characterizes possibility. Fulfilment of quality, again, 1 An apology is here clue for the length of the discussion, which, how- ever, is much more detailed in the manuscript. As may be known from other works, too, I do not admit at last the validity of the relative de- fence. It itself, however (the defence), will be allowed, perhaps, to be not alien to the spirit of Kant. At all events, it will show that the student at work must have made a very serious business of the various pieces of Kant's machiuery, and must have remained very long, faith- fully, and even trustingly by them. 108 TEXT-BOOK TO KANT: is to satisfy the requisitions of matter, the requisitions of judgment, and that is what characterizes actuality. Fulfilment of relation, lastly, is to satisfy the requisi- tions of connexion, the requisitions of reason, and that is what characterizes necessity. It is not now necessary to repeat either that all these forms are only for the filling of sense, only for experience, and that they are quite idle and useless else. Nay, they are even mischievous should any attempt be made to deduce results from them inde- pendently regarded, or a priori and alone. They themselves are a priori ; but, even so, they are only the possibility of experience, or, for that matter also, only the possibility of experience ; and they are nothing more. We know, too, only the one experience, only the experience we have ; to conceive another, any other, — and it is quite possible to conceive any number of different experiences, — is not to realize it. We are bound, not only to forms of sense on the one hand, but to forms of understanding on the other: we may, indeed, conjecture others ; but conjecture is not discovery. With the same sense-forms, we might possess different categories ; or, the latter remaining, the former might be changed ; or we may conceive both changed ; or we may even conceive an intellect that would not think discursively through categories or notions, and not perceive vicariously through forms of sense, but that, like the divine mind, it may be, would directly perceive, or see into. Thus, then, we have completed all that relates to transcendental judgment. Under transcendental ap- prehension (understanding) we were enabled to perceive the nature and necessity of certain a priori subjective forms, both of sense and intellect, without which this regulated whole which we call experience would be THE ItEPRODUCTION. 109 manifestly impossible ; and now, in the present book, we have seen the manner in which these a priori elements mutually relate themselves into an a priori system antecedent to, and in provisional condition of, experience, as well as that also in which the elements proper of experience — those, namely, which result from special sensation and are termed a posteriori — become arranged and subsumed under said a j/riori system. In correspondence with all the ground- notions of the understanding, time was found to furnish ground-multiples of sense : the subsumption of the latter under the former gave rise to schemata ; and these again readily discovered to us the principles which judgment would adopt in subsuming under them the whole wealth of the empirical all. The nature and evidence of these principles presented such points of relative difference that they were of necessity named differently. The a priori principles that concerned extensive magnitudes were, consist- ently and appropriately, termed axioms. Those that related to degree were similarly denominated anti- cipations. And while, with reference to the intel- lectual relations, the laws of sensuous relations ap- peared as analogies, those of modality could only lay claim to the name of postulates. All of these principles, both sensuous and intel- lectual, were found to have no intention, no validity, no meaning, no reference, unless to the a posteriori empirical matter, from which alone they could receive filling and objective actuality. The world that resulted was seen, however, to be, in every respect, phenomenal ; and this, not only because the empirical filling was required to adapt itself to certain internal forms, but because, also, this very empirical filling itself was only, and could only be, 110 TEXT-BOOK TO KANT: sensuous affection. This phenomenal life we found to be as true for the inward as for the outward world, and that we knew the noumenon subjacent to our phenomenal consciousness just as little as we knew the noumena subjacent to the phenomena we call things without. It by no means follows, however, that this word noumenon is entirely empty and meaningless for us. We cannot assert, indeed, that we affirmatively know any noumenon. Such know- ledge is utterly impossible to beings constituted as we are, whether sensuously or intellectually. Or limita- tion and finitude in both respects have been clearly and convincingly demonstrated ; and a knowledge of noumena is only possible for an intellect which we may name original and primary, while ours, plainly, is but derivative and secondary. Such original and primary intellect we may conceive to belong only to the Deity ; who will not think through finite cate- gories, nor perceive through finite senses, but will intuitively and directly know all things, not as they appear, but as in themselves they are. But if we are so situated as regards affirmative noumena, the case is different with those which we may call negative. These we may assume. Nay, a phenomenal world implies a noumenal; and the assumption of such is absolutely necessary in order duly to subordinate and limit the pretensions of sense. It does not follow, nevertheless, that its phenomenal nature attaches any character of uselessness and meaninglessness to this the world of time which we, in time, inhabit. Here, as evidence from every side assures us, existence is but probationary. There, beyond, is our true and noumenal home awaiting us for eternity, with God, when time and the shows of time shall have worked out their function on us. THE REPRODUCTION. Ill That these are true views, and that our system is co-extensive with the universe of things, will appear more and more evident as we proceed. The world we have yet seen is a world conditioned merely. Under reason (our third book), we shall discover those relations to the necessary unconditioned, that round and complete it (our world) as an object of intellect. Our practical critique, again, will introduce us to the veritable noumenal world ; while our inquiry into judgment will mediate and justify transition from the one world to the other. In this way, the whole of those first principles on which man, intellectually, morally, and aesthetically, founds, will be given to sight; and the course and method of their applica- tion to our whole actual existence will, without diffi- culty, suggest themselves. Before proceeding to our third book (on reason), however, we interpose certain intercalary, introduc- tory, and transitionary matter relative to noumena and phenomena, as well as to what we have named the notions of reflection. These last will be found of considerable interest in themselves, and very strikingly illustrative of the value of our principles as applied judicially to certain views of Leibnitz. Translation. u THE KRITIK OF PURE REASON. (TRANSLATION.) Introduction. I. Of the Difference between Pure and Empirical Cognition. It is beyond a doubt that all our knowledge begins with experience. For by what should our faculties be roused to act, if not by objects that affect our senses, and thus partly of themselves produce impres- sions, partly, again, bring the understanding itself into movement, in order to compare these, to join or disjoin them, and in this manner work up such crude material of the intimations of sense into a cognition or recognition of objects which is named experience. So far as time is concerned, then, no cognition of ours precedes experience, and with experience all our knowledge begins. But, though all our knowledge begins with experi- ence, it does not follow that therefore it all derives from experience. For it is just possible that experience is itself a compound. It is just possible, that is, that there is in experience, besides what is due to the impression of sense, something in addition that comes from our faculties themselves (when merely acting because of impression) ; and in that case, it would take long practice, it may be, to enable us to dis- tinguish the latter, and separate it from the former. 116 TEXT-BOOK TO KANT: It is at least not a question to be summarily dismissed, but one that demands more particular consideration, this, to wit : whether there really be such component part of knowledge as is independent of experience and, indeed, of any impression of sense whatever? Such component part of knowledge, did it exist, were alone to be truly termed a priori; and it would evidently stand in contradistinction to what other component part of knowledge is called empirical : the latter, namely, having its source only a posteriori, or in experience. The expression a priori, at the same time, is not precise enough to designate the entire sense of the preceding question. For of many a mere empirical fact, we say, that we know it a priori, simply because we do not directly derive it from experience, but from a general rule ; and this, even notwithstanding that the rule itself may be so derived. For example, we say of a man that shall have undermined his house, he might have known a priori that it would fall in ; he had no occasion to wait for the experience of the actual event. Nevertheless, he could not have known this absolutely a priori. For that bodies are heavy and, consequently, fall when their supports are withdrawn — this, at least, he must have known by experience beforehand. In what follows, therefore, we shall understand by cognitions a pi iori, not such as are independent of this or that experience, but such as are totally independent of any experience whatsoever. Opposed to these are empirical cognitions, or such as are only possible a posteriori, or from experience. Pure, again, are those a priori cognitions which are quite free from all and every empirical admixture. Thus, for example, the proposi- tion, that all change has its cause, is an a priori proposi- THE KRITIK OF PURE REASON. 117 tion ; but it is not, at the same time, purely such, for change is an idea which can only be derived from experience. II. We do possess certain a priori Cognitions, and even Common Sense is never without such. 1 What is wanted here is a criterion, by means of which we may, with certainty, distinguish what is pure from what is empirical. Now experience informs us that something is so and so, but not that it cannot be otherwise. Firstly, then, should there be a pro- position such that it is thought together with its neces- sity, then it is a judgment a priori ; and, if underived from any other, absolutely a priori. Secondly, experi- ence extends to its judgments never strict or true, but only (through induction) assumptive or com- parative universality ; so that, properly, it can only be said, So far as we are yet aware, there is no excep- tion to this or that rule. Should any judgment, then, be thought in strict universality, or so, that is, that exceptions are impossible, we may be sure that that judgment is no derivative from experience, but directly a priori. Empirical universality, therefore, is only an arbitrary raising of validity from that which obtains in most cases to that which holds good in all, as in the proposition, for example, that all bodies are heavy. Whereas, when strict univer- sality attaches to a judgment, such universality points to a special cognitive source, namely, to a faculty of cognition a priori. Necessity and strict universality, therefore, are sure criteria of a priori cognition, and 1 Rosenkranz has here " der gemeine Stand." I prefer to read " der gemeine Verstand," as, indeed, is supported by the text : examples of ordinary common sense follow examples of science. 118 TEXT-BOOK TO KANT: inseparably found together. In practice, however, as it is easier, now to apply the one and now the other, it will be advisable to avail ourselves, as occasion may suggest, of either criterion separately ; for, even separately, either of them is quite infallible. Now, it is easy to show that there actually are in our knowledge such necessary and, in the strictest sense, universal (consequently pure a priori) judg- ments. Would we have an example from science, we have only to turn to any proposition in mathematics ; while, as for the most ordinary common sense, there is obviously to hand, by way of instance, the pro- position that every change must have a cause, where the very notion cause so manifestly implies necessity (of connexion with an effect) and strict universality (of rule), that it would be altogether lost did we derive it, like Hume, from our conjoining what simply follows with what simply precedes, through the mere habit of the experience, and the consequent simple custom of connecting ideas (where the neces- sity could be only subjective). Besides demonstrating the actual existence in our knowledge of principles a priori by a reference to fact, we might even a priori prove as much. We might demonstrate, that is, the indispensable necessity of such principles to the very possibility of experience. For how should there be any certainty in experience, were all the rules in it only empirical and (consequently) contingent? It were hardly possible, evidently, to allow any such rules the name of first principles. But it may suffice here to have demonstrated the fact of the possession of pure cognition on our part, together with the signs of the latter. Nay, not merely judgments, but even certain ideas, may claim for themselves an a priori origin. Suppose, in the case of our empirical idea of THE KRITIK OF PURE REASON. 119 a body, we successively withdraw all its empirical con- stituents, such as colour, consistency, weight, even impenetrability, we shall still find it impossible to withdraw the space it occupied. This space will still remain when the body itself has disappeared. In like manner, if, in regard to our empirical idea of an object in general, whether corporeal or incorporeal, we Avithdraw all properties known to us from experience, we shall still be unable to withdraw from it those by which we think it as substance, or as attributive to substance (though this notion of substance has more determination in it than that of an object in general). We must, therefore, overborne by the necessity with which said notion forces itself upon us, admit that it has its seat a priori in our faculties of cognition. III. Philosophy stands in need of a Science which shall determine the Possibility, the Principles, and the Limits of all a priori Cog- nition. But, to go still further, it is a fact that there are cognitions which even quit the bounds of all possible experience, and actually, by means of ideas for which, so far as experience goes, no correspondent object can be found, assume to extend the range of our judgments beyond any experience whatever. And just in these latter cognitions, transcending as they do the world of sense, and unaccompanied by experience to guide and correct them, there lie in- terests of reason which we hold to be of far greater con- sequence and loftier aim than anything or all that un- derstanding can teach us in the domain of experience. In these cognitions, indeed, even at the risk of failure, we rather venture everything than, for any reason of 120 TEXT-BOOK TO KANT : doubt, or carelessness and indifference, consent toforego what is of such an import. Such unavoidable problems of pure reason's own are God, Free Will, and Immortal- ity. The science, again, which, as well in the end it con- templates, as in all its complement of means, is alone directed to the solution of these, we name metaphysic — a science that, in its procedure, starts as yet only dogmatically ; that is, having instituted no previous inquiry into sufficiency or insufficiency on the part of reason for so great an enterprise, it yet confidently undertakes completion of it. l Now, it seems no more than natural that, once we have left the solid ground of experience, we should not forthwith proceed to build, without having care- fully assured ourselves, first of all, in regard to a foundation, and that, too, all the more, should we find ourselves provided only with principles which are unauthenticated, and have come to us we know not whence. It seems no more than natural, I say, that, rather than this, we should have long before started the question, How have we got to these principles, and of what extent, import, and value are they ? In effect, nothing is more natural when by the word natural we understand what, rightly and reason- ably, ought to take place ; but, on the other hand, when we mean by natural only what usually takes place, then nothing is more natural than that any such preliminary inquiry should remain long null. For the fact is, that some of the principles in view (as the mathematical ones) possess authentication from of old, and reflect, consequently, a similar presumption on to others which may in reality be altogether different. Besides, when one is beyond experience, one is safe not to be contradicted by ex- 1 Bosenkranz would seem to omit a sentence or two here. TUE KRITIK OF PURE REASON. 121 perience ; and, eager as we are to extend our know- ledge, only when so contradicted is it that we can allow ourselves a halt. But even this may be avoided, should we be but careful with our fictions ; for fictions, in such circumstances, they must be. On the other hand, mathematics afford us a splendid example of success in the cognition in question. Objects and ideas, it is true, are considered there only so far as they are capable of being exhibited in objective re- presentation. But this is easily overlooked, because said representation can itself be a priori given, and is, consequently, scarcely to be distinguished from a pure notion proper. Led away by such a proof of the power of reason, we can see no bounds to the extension we desire. The light dove, in feeling the resistance of the air its free flight cleaves, might very well think to itself that it would have a still better chance in a space that were void. Even so Plato, because of the narrow limits it set the understanding, forsook the world of sense ; and, beyond its bounds, buoyed up on the wings of the ideas, committed him- self to the blank inane of the pure intellect. He did not perceive that, with every effort, his progress was null ; for foothold he had none, against which steadied, he mischt have exerted his strength to bring reason from the spot. It is, however, an ordinary fate of speculative reason, to complete its edifice at the soonest, and only then to examine whether the foun- dations are well laid or not. All manner of excuses, rather, is indulged in to comfort us in regard of their entire sufficiency, or even to prove such late, danger- ous examination wholly inexpedient. What saves us during the work from any fear or suspicion, deceiving us with apparent substantiality indeed, is this. A great part, perhaps the greatest, of the business of 122 TEXT-BOOK TO KANT: our reason consists in the analysis of ideas which we have already formed of objects. This furnishes us with a number of cognitions which, although they are nothing more than elucidations or explanations of what is already (confusedly) implied, are still, at least in form, regarded as new : in matter or contents, not ex- tending our notions, they explicate them. But this process furnishing, as it does, an actual a priori cogni- tion, accompanied, too, by a certain safe gain, our reason interpolates unawares into this false show of extension allegations of quite another nature ; foisting in with given notions other notions quite alien, and that, too, a priori, without our knowing how or whence these latter come, or even without any such question being ever once entertained by us. Accordingly, I shall treat directly now in the beginning of the difference between these two modes of cognition, i IV. Of the Difference between Analytic and Synthetic Judgments. In all judgments in which the relation of a subject to a predicate is thought (affirmatives alone con- sidered — application to negatives being afterwards easy) this relation is possible in two ways. Either the predicate B belongs to the subject A as something that (covertly) is contained in it ; or B lies completely outside of the notion A, though possessing connexion with it. In the first case I call the judgment analytic ; in the second synthetic. Analytic judgments (the affirmative ones) are therefore those in which the connexion of the predicate with the subject is 1 One or two trifling expressions in the above are not found alike in the different editions. THE KRITIK OF PURE REASON. 123 thought through identity ; synthetic, again, those in which this connexion is thought without identity. We might name them also, the former, judgments of explication; the latter, judgments of extension. The former, namely, add, in the predicate, nothing to the notion of the subject, but only separate this notion into its subnotional parts, which parts are already (ob- scurely) thought in the notion. The latter, on the other hand, add to the subject a predicate which was not at all thought in it, and could not by any analysis have been extracted from it. For example, if I say, All bodies are extended, this is an analytic judgment. For, in order that I may find extension as connected with it, I need not leave what notion itself I attach to body. I have only to analyze it, or open my eyes to what complex I think in it, to become aware of this predicate as contained in it. The judgment, therefore, is analytic. On the other hand, if I say, All bodies are heavy, in that case the predicate is something quite different from anything I think in the mere notion of a body as such. The addition of such a predicate produces, therefore, a synthetic judgment. Judgments of experience, as such, are all synthetic. For it were absurd to have recourse to experience for an analytic judgment, seeing that I need not go out of my notion itself to get the judgment, nor require, therefore, any testimony of experience in the case. That a body is extended is a proposition a priori evident, and not a judgment of experience. For, without having recourse to experience, I have already in the notion all the conditions necessary for my judgment. I have only, according to the principle of contradiction, to extract the predicate from the notion. In so acting, I become aware, also, of the neces- sity of the judgment, and necessity is no declaration 124 TEXT-BOOK TO KANT: of experience. On the other hand, although I do not include, in the notion of a body in general, the pre- dicate heavy, still said notion (body) designates an object of experience, by a part of experience, to which part I can add other parts of the same experience, not comprehended in the first. I know the notion body already analytically, say, through the characters ex- tension, impenetrability, figure, etc., which the notion simply implies. But now I extend my knowledge, and in once more consulting experience (from which I had derived this notion of body), I find, always conjoined with the said characters, that also of weight, which, as a predicate, therefore, I add synthetically to the notion in question. It is, therefore, on experi- ence that the possibility is founded of the synthesis of the predicate heavy with the subject body, be- cause, though the one is not implied in the other, still both notions, as parts of a whole (namely ex- perience, which is itself a synthetic conjunction of perceptions), belong to each other, if only contin- gently. But, in the case of a priori synthetic judgments, this expedient (of experience) is altogether inappli- cable. If, in such reference, I am to go beyond the notion A in order to recognise another, B, as connected with it, on what do I support myself, and by what is the synthesis made possible, seeing that I have not the advantage in this case of looking about me for it in the field of experience ? Let us take the proposi- tion, All that happens has a cause. In the notion of something that happens (an effect), I think something come to be, which, therefore, had a certain time before it, etc., and from this something, as it is there before me, it is possible for me to deduce various analytic judgments. But the notion cause lies quite TUE KRITIK OF PURE REASON. 125 out of this notion. Denoting something quite differ- ent from that which happens (the effect), it is not at all implied in it. How do I come, then, to say of any fact in event something quite different from the fact itself, and to recognise the notion cause, though not contained in said fact, nevertheless as belonging to it, and that, too, necessarily ? What is the unknown x on which the understanding supports itself, when it believes itself to discover from the notion A a predi- cate B, alien to it, but which it judges, nevertheless, to be connected with it ? It cannot be experience, because the relative proposition adds the latter to the former, not only with a greater universality than experience can supply, but even with the expression of necessity, and consequently wholly a priori or through mere notions. Well now, the entire end and aim of our speculative cognition a priori concern such synthetic principles, or judgments of extension. For the analytic ones are certainly of the greatest import- ance and necessity, but, here with us, they are avail- able only for the sake of that precision of ideas which is required for an accurate and complete synthesis, as an acquisition veritably new. 1 V. In all the Rcational Theoretic Sciences, Synthetic a priori Judg- ments are present as Principles. 1. Mathematical judgments are all synthetic. This proposition seems hitherto to have escaped the obser- vation of the anatomists of human reason — nay, to be directly opposed to all their suppositions, although it is undeniably certain, and very important in result. 1 Rosenkranz has some slight differences here too, besides some errors of press. 126 TEXT-BOOK TO KANT: For, because it was found that mathematical reason- ings proceed all of them on the principle of contradic- tion (as the nature, indeed, of apodictic certainty re- quires), there ensued the conviction that by means of the same principle also it was that the fundamental propositions themselves were to be seen into. In this they erred. For a synthetic proposition may cer- tainly be understood from the principle of contradic- tion ; still, only in this way, that another synthetic proposition is presupposed from which it may be inferred, — but never independently. First of all, it is to be remarked that mathematical judgments as such are always a priori, and not em- pirical ; for they bring with them necessity, which is not to be got from experience. Should this, however, as a general proposition, appear doubtful, I will con- fine it to pure mathematic, the very notion of which implies that it is not concerned with empirical, but only with pure a priori cognition. We might be apt to think at first that the pro- position 7 + 5 = 12 is merely an analytic proposition, which follows from the notion of a sum of 7 and 5, according to the principle of contradiction. But if we look closer, we shall find that the notion of the sum of 7 and 5 implies nothing but the uniting of the two numbers into one, there being no thought, at the same time, of what this one number itself is which comprehends the two. The notion of 12 is not thought in this, that I think to myself the uniting of 7 and 5 ; and I may analyze my notion of such possible sum as long as I please without finding the 12 in it. TTe must go out of these notions, and take help from perception. We must assist ourselves, that is, by such objective representation as corre- sponds to one of the two numbers (say five points or THE KRITIK OF PURE REASON. 127 the five fingers), and, so assisted, add the units of the number perceived (5), one by one, to the notion of the number thought (7). I take first the number 7; next, for the notion of the 5, I refer to my fingers as perceived ; and then I add the units (which together constitute the number 5), one by one, in guidance of the representation perceived, to the number 7. In this way, for result, I see the number 12 emerge. That 7 should be added to 5, I have indeed thought in the notion of a sum 7 + 5, but not that this sum is equal to the number 12. An arithmetical proposi- tion is, therefore, always synthetic, as we may more distinctlv discern, should we assume somewhat lanrer numbers ; in which case it will clearly appear that, let us turn and twist our notions as we may, we never can, by mere analysis of notions, and unassisted by perception, discover their sum. Just as little is any proposition of pure geometry analytic. That the straight line between any two points is the shortest, is a synthetic proposition. For my notion of straight includes in it nothing of quantity, but only a quality. The notion shortest is wholly something adscititious, something added to it, and cannot by any analysis be derived from the notion straight line. Perception, then, must be here called in to assist, and only by its intervention is the synthesis possible. Some few propositions which are presupposed in geometry are, it is true, really analytic and rest on the principle of contradiction. They serve, however, only as identical propositions, for the chain of the method, and not as principles. For example, it is said a is equal to a, that is, the whole is equal to itself; or a + b is greater than «, which is, the whole is greater than its part. And yet even these, that 128 TEXT-BOOK TO KANT: pass valid on the authority of mere notions, are only allowed place in mathematic because they can be exhibited in perception. What commonly leads us here to suppose that the predicate of such apodictic judgments is already contained in our notion, and that, consequently, the judgment is analytic, is solely the peculiarity of the expression. To a given notion, namely, we must think a certain predicate, and this necessity is already present with the notions. But the question is not what we must think to the given notion, but what we actually, though obscurely, think in it ; and then we see that the predicate belongs to the notion, necessarily indeed, not, however, because of being thought in it, but because of a perception which must be added to it. 2. Natural philosophy possesses synthetic a priori judgments as principles. I will only adduce a couple of propositions in example ; as that in all changes of the corporeal world the quantity of matter remains the same, or that in all communication of motion, action and reaction are always alike. In both, not only the necessity is clear, and by consequence their a priori origin, but also the fact that they are synthetic propositions. For in the notion of matter I do not think its permanence, but only its presence in space as filling it. That is, 1 actually go beyond the notion of matter in order to think a priori to it something that I did not think in it. The proposition, there- fore, is not analytic, but synthetic, and yet a priori ; so it is with the other propositions of the pure part of the science. 3. In metaphysic (though we should only regard it as a science which has been hitherto desiderated, but which, from the very nature of human reason, never- theless, is a science indispensable), synthetic cogni- THE KRITIK OF PURE REASON. 129 tions a priori simply must be. For it is not its business merely to unravel notions which we a priori form of things. On the contrary, the business here is to extend our a priori cognition ; and to that we must avail ourselves of such propositions as add on something beyond the given notion, something not contained in it ; and in this way, by means of syn- thetic a priori judgments alone, advance indeed so far that experience itself is unable to follow us. For example, there is the proposition, among others, that the world must have a beginning. And by this we see that metaphysic, at least in its aim, consists of pure a priori synthetic propositions. VI. General Problem of Pure Reason. It is already not a little won, if we can bring a variety of questions under the formula of a single problem. For in this way, through exact determina- tion of it, we not only lighten to ourselves our own work, but we facilitate for everybody else as well, who will examine it, the judgment whether we have done justice to our own design or not. The problem proper of pure reason, now, is comprised in the ques- tion, How are a priori synthetic judgments possible? That metaphysic, hitherto, has remained in so vacillating a condition of uncertainty and contra- diction, is solely to be ascribed to the fact that we have not sooner attained to the conception of this problem, or even to that of the distinction between analytic and synthetic judgments. On the solution of this problem now, or on a satisfactory proof that the possibility it would wish demonstrated does not i 130 TEXT-BOOK TO KANT : exist, it depends whether metaphysic shall stand or fall. David Hume, who of all philosophers came nearest to this problem, thought it not out, however, by any means determinately enough, or in its gener- ality, but merely took his stand by the synthetic proposition of the connexion of the effect with its cause (jmncipium causalitatis). Accordingly, he as- sumed to make out that such a proposition is, a priori, wholly impossible. His reasonings went to prove, indeed, that all we call metaphysic terminates in a mere delusion of a supposed insight on the part of reason, into what in effect is merely borrowed from experience, and has only taken on, through custom, the semblance of necessity. But such an allegation, subversive as it is of all pure philosophy, would never have occurred to him had he but caught sight of our problem in its universality. For he would have then been conscious that, on his argument, even pure mathematic would be impossible, inasmuch as it is a science built on a priori synthetic propositions — a conclusion, plainly, from which his own good sense would certainly have saved him. In the solution of the above problem there is in- volved, at the same time, the possibility of an applica- tion of pure reason in foundation and completion of all the sciences in which any theoretical a priori cognition of objects is concerned ; that is, an answer to the questions, How is pure mathematic possible? How is pure natural philosophy possible ? Of these sciences, inasmuch as they once for all are, we may certainly with propriety ask, how they are possible ; for that they must be possible is de- monstrated by their actuality. 1 As for metaphysic, 1 This may be doubted as regards a pure natural philosophy. But we bare only to look to the first propositions of physic proper (empirical), as THE KRITIK OF PURE REASON. 1ÖL again, we may reasonably doubt of possibility in its regard, in view, namely, of its unsatisfactory progress hitherto, as well as of the fact that, considering its essential aim, we cannot say it has, in any instance, actually been. And yet, again, knowledge of this kind is really, in a certain sense, to be assumed as given, or metaphysic is actual after all — if not as science, then as natural capability (inetaphysica naturalis). For human reason, not moved by any vanity of mere learning, but im- pelled by necessity of its very nature, strives ever irrepressibly forward towards such questions as can- not possibly be answered by any mere empirical con- sideration, or principles derived thence. So it is that really in all men, so soon as reason has advanced to speculation, a metaphysic of some kind always has been and always will be. And now, from the same source, we have this question also : How is meta- physic as natural capability possible ? That is, how do the questions which pure reason starts for herself, and which, in some way, she must answer — how do these questions originate in the very nature of reason as such? 1 It is the fact, however, that unavoidable contradic- tions have always shown themselves in any attempt yet to answer these natural questions (e.g., whether the world has had a beginning, or whether it exists from all eternity, etc.) ? We cannot, therefore, remain satisfied with a mere natural capability for metaphysic, or with the mere faculty of reason itself, in possession the permanence of matter in quantity, inertia, the equality of action and reaction, etc., to be convinced that they constitute a physicam pur am (or rationalem), which certainly deserves to be separately established, in its whole extent, whether large or limited, as science proper. — K. 1 Rosenkranz omits here a very important " die " (before " Frage "), as, near the end of the preceding section, a not unimportant " und." 132 TEXT-BOOK TO KANT: of which there is always that necessity of a meta- physic of some kind, be it what it may. It must be possible, rather, to bring matters relatively to some certainty as concerns either the knowing or the not knowing of the objects in question, either the ability or the inability of reason to judge in their regard. That is, it must be possible for us either confidently to extend, or else duly limit, reason. This last ques- tion, which flows from the general problem, were rightly put thus : How is metaphysic as a science possible ? A criticism of reason leads, therefore, at last necessarily to science ; while, without criticism, dog- matically to set to work with reason, results only in groundless allegations, to which others equally spe- cious may be opposed, and the end, consequently, is scepticism. Neither will this science be of great and forbidding extent. It is not with the objects of reason, namely, the multiplicity of which is infinite, but with reason's self, that it has to do. The problems it considers take birth in the bosom of reason only : they are not im- posed upon reason by the nature of things, which are different from it, but by its own nature. Accordingly, therefore, if reason has, first of all, come perfectly to know its own powers in regard of objects which may be offered in experience, it must be easy fully and surely to determine the range and extent of its desired application beyond all bounds of experience. We may and must, therefore, regard all these pre- vious attempts dogmatically to bring about a meta- physic as, in effect, null. For, whatever there may be of analytic in the one or the other of them, as regards the mere dissection of the notions which a jiriori attend our reason, such material is not the end THE KRITIK OF PURE REASON. 133 and aim of, but only a preparation for, metaphysic proper. To this science it belongs, namely, to extend our synthetic a priori knowledge, and to that, said analytic material is inapplicable, as it merely shows what is contained in those notions, but not how we a priori attain to them. Accordingly, we are not enabled thereby to determine their due and valid use in regard of the objects of cognition generally. It is no great hardship to abandon such pretensions wholly, see- ing that the undeniable and dogmatically inevitable contradictions of reason have long since cost every previous metaphysic all its credit. It will demand more self-reliance, in view of the difficulty within and the opposition without, to resist, in regard to a science indispensable to human reason (whose root, let us hew off whatever actual stems we may, it is impos- sible to tear up), discouragement from the attempt to further it, once for all at last, into a prosperous and fruitful growth, by means of another, and, to those in the past, wholly opposed method. VII. Idea and Division of a Special Science under the Name of a Critique of Pure Reason. There results from all this, now, the idea of a special science, which may be named critique of pure reason. For reason is the faculty which furnishes the prin- ciples of a priori cognition. Pure reason will there- fore contain the principles towards an absolute a priori cognition. An organon of pure reason would be a whole of those principles, in accordance with which all pure a priori cognitions can be acquired and actually realized. The complete application of such 134 TEXT-BOOK TO KANT: an organ on would have for result a system of pure reason. Inasmuch, however, as this is asking a great deal, and it is yet uncertain whether and in what cases an extension of our knowledge is in this way at all possible, we may conceive rather a science of the mere investigation of pure reason, its sources and limits, as the jjropcedeutic to the system of pure reason. Such a science would necessarily be considered not a doctrine, but only a critique of pure reason ; and its use in speculation would, in reality, be merely negative. Serving not to extend, but to clear reason, it would guard it from errors, which is already much. I call all cognition transcendental, which is occupied not so much with objects, as with the process by which we come to know them, in so far as that process has an a priori element. A system of such elements would be a transcendental philosophy. But, for a beginning, this again is too much. For such a science, necessarily embracing as well analysis as synthesis, would extend beyond our intention at present, seeing that we shall apply the former only so far as is indispensable for a complete survey of the principles of the synthesis a priori ; which is our one general object. It is with this inquest that we now occupy ourselves — an inquest that cannot properly be named a doctrine, but only a transcendental critique, and for this reason, that it has in view, not the exten- sion, but the clearing of our knowledge, and would seek to furnish merely a touchstone of the worth or worthlessness of our cognitions a priori. Such a critique, accordingly, is a preparation, so far as pos- sible, for an organon, or, failing that, a canon, by means of which we shall certainly be able to realize, some time, a completed system of the philosophy of pure reason, as well analytic as synthetic, let it con- THE KRITIK OF PURE REASON. 135 sist, as it may, in extension or in mere limitation of the relative knowledge. For, that this is possible, nay, that such system cannot be of too great extent to allow the hope of completing it, may already beforehand be judged from this, that what constitutes the object here is not the nature of things, which is inexhaustible, but the understanding, which pro- nounces on the nature of things, and this understand- ing itself, too, only in regard of its cognition a priori; its provision of which, moreover, seeing that Ave have not to seek it from without, cannot pos- sibly remain concealed from us, and in all likelihood is small enough to be perfectly taken up, duly esti- mated, and in worth or worthlessness competently appreciated. Still less must there be expected here a critique of the books and systems of pure reason, but only of the faculty itself that is so denominated. Only in such critique as basis have we a sure and certain touchstone whereby to try the philosophical worth of earlier or later works in this department ; otherwise, we have only an unaccredited historian and judge pronouncing on the groundless opinions of others solely through opinions of his own which are equally groundless. The transcendental philosophy is the idea of a science whereto the critique of pure reason shall sketch the entire plan — architectonically, that is from principles, with plenary guarantee of the complete- ness and security of all the pieces which compose the structure. It is the system of all the principles of pure reason. That this critique is not itself already the transcendental philosophy, depends solely on this, that, in order to be a complete system, it ought to comprehend, as well, a completed analysis of the entire human cognition a priori. Our critique, in- 136 TEXT-BOOK TO KANT : deed, must certainly exhibit a complete enumeration of all the primitive notions which constitute the pure cognition in allusion. But from the detailed analysis of these notions, as from the detailed revision of their derivatives, it will rightly refrain ; partly because such analysis were to little purpose, inasmuch as it has not the significance peculiar to the synthesis, which synthesis, properly, is the special motive of the entire critique ; and partly because it would contra- dict the unity of the plan, to undertake the responsi- bility of the completeness of any such analysis and derivation ; from which, indeed, in consideration of one's design, one might reasonably be dispensed. At the same time, it will be easy to supply this com- pleteness of analysis and derivation from the a priori notions themselves, directly they are once for all established as the complete principles of synthesis, with nothing wanting to them in that essential refer- ence. According to this, then, there belongs to the critique of pure reason all that is constitutive of a transcendental philosophy, and of such philosophy it is the entire idea, but it is not that philosophy ; because, on the side of analysis, it goes no further than is required for a complete estimate of the a priori synthesis. As regards the divisions of such a science, the main consideration is, that there must be no admission for notions in anywise empirically tinged ; or that the a priori elements must be perfectly pure. Hence the exclusion from transcendental philosophy of the principles of morality, notwithstanding that the chief moral notions and propositions are really cognitions a priori. And the reason is, that, as regards the notions of pleasure and pain, desire, passion, etc., TUE KRITIK OF PUKE REASON. 137 which are all of empirical origin, said principles do not, indeed, regard these notions as basis of their own moral prescripts ; but yet, in construction of the system of pure morality, they must necessarily admit them as (in duty) obstacles to be overcome or temp- tations to be resisted. The transcendental philosophy, therefore, is a philosophy of the merely speculative pure reason. For all moral practice, so far as it involves motive, refers to feeling, and feeling always is of empirical origin. As concerns division, then, this our science will, on the usual general principles of such, consist of a theory, firstly, of the elements, and, secondly, of the method of pure reason. Each of these parts, again, will have its own sub-parts, the conditions of which, however, we do not discuss here. Only, it may be of advantage, perhaps, to be, introductorily, or pre- fatorily, reminded, that there are two stems of human cognition, sprung, both, it may be, from a common but unknown root, namely, sense and under- standing, by the former of which objects are given to us, and by the latter thought. Even sense, then, if it be found to possess for us intimations a jjriori, which constitute conditions under which alone objects can be perceived by us, will, for that reason, enter as a constituent into a philosophy that is transcendental. And, accordingly, the transcendental sense-elements will necessarily constitute the first part of our theory of elements, inasmuch as the conditions under which objects are given precede those under which they are thought. 138 text-book to kant i Transcendental ^Esthetic. 1 §1- Let cognition refer itself to objects in what manner it may, perception is such reference direct or imme- diate; while the reference of thought, as only through perception, is mediate. Perception, again, takes place only so far as an object is given us, which for us men, further, is only possible in this way, that said object, in some certain way, affects our mind. The capability (receptivity) of receiving intimations through the mode in which we are affected by objects is called Sensibility. By sensibility alone, then, are objects given us, and sense alone affords us Percep- tions ; which being thought by Understanding, there result Notions. Ail thought must, at last, directly or indirectly, refer itself to perception and, consequently, to sense ; for in no other way can an object be given us. The effect of an object on our susceptibility of im- pression, so far as we are affected by it, is Sensation. Any perception that refers itself to an object through sensation is said to be Empirical. The undetermined (afterwards seen to be the uncategorized) object of an empirical perception is what we call an Erscheinung (a sense-appearance, a sense-presentation : it is the mere sensation in crude perception — time and space — but as yet without notion or category). In this object, I call what corresponds to sensation (the mere feeling) its matter; while what again so acts that the units of the impression are, in each 1 Kant's headings are mostly so compound that they give a look of difficulty of themselves. We translate here only what is simple and direct. THE KRITIK OF PURE REASON. 139 other's regard, peculiarly disposed (as so and so beside each other, or so and so after each other), I call its form. As this latter constituent (the order in the units of impression) cannot possibly be itself sensa- tion, we must hold that, while the matter of impres- sion is only a posteriori given, its form, on the con- trary, must a priori lie ready for all impression in the mind, and be capable, therefore, of separate considera- tion, apart from sensation. I call all intimations pure (in transcendental sense), in which there is nothing found that belongs to sensa- tion. The pure form of sensuous perception, conse- quently, will be met with a priori in the mind, wherein all units of impression are perceived in cer- tain relations. This pure form of sense or sensibility, accordingly (as without sensation), may be legiti- mately named pure perception. Thus, when I with- draw from what makes up my consciousness of a body, what elements in it belong to the understanding as substance, force, divisibility, etc., and again what elements in it belong to sensation, as impenetrability, hardness, colour, etc., still, of this empirically per- ceived object, there remains something over, namely, extension and figure. These belong to pure percep- tion which, as a mere form of sensibility, and without any actual object of sense or sensation, exists in the mind a priori. A science of all the a priori principles of sense, I call Transcendental ^Esthetic. 1 There must, there- fore, be such science which, constituting the first part of the transcendental theory of elements, will oppose itself to the second part, which is devoted to the 1 In a note here Kant vindicates this word for his own use of it, and against that of Baumgarten . To this note the second edition adds a clause which is wanting in that of Rosenkranz. The note itself is omitted here as of no importance. 140 TEXT-BOOK TO KANT : principles of pure understanding, and is named Transcendental Logic. In the transcendental aesthetic, we shall isolate sense, first, by withdrawal of all that the understand- ing thinks into it through its notions, and second, by further withdrawal, from the bare empirical sense- presentation that then remains, of all that belongs to sensation. For result we shall have nothing but pure perception and the mere form that adds itself to sense-matter ; and that is all that the sensibility can a priori yield. But, through such investigation, it will be found that, as principles of a priori cognition, there are two pure forms of sensuous perception, namely Space and Time, with the consideration of which we shall now occupy ourselves. Section I. — Of Space. § 2. Metaphysical Exposition of this Notion. By means of external sense, which is a function of our mind, we perceive objects as external to our- selves, and collectively in space. In it their figure, magnitude, and relation the one to the other, are determined or determinable. Inner sense, by means of which the mind contemplates itself, or its internal condition, furnishes, indeed, no perception of the soul itself as an object ; but there is, nevertheless, a single determinate form, in connexion with which the perception of its internal state is alone possible ; or all that belongs to our inner affections presents itself in relations of time. Time can be as little perceived externally, as space can be perceived as THE KRITIK OF PURE REASON. 141 though it were something internal. What, then, are space and time ? Are they actual things on their own independent account ? Are they only affections, or, it may be, relations of things, but such that they would attach to things in their own selves, even if these things were not perceived ? Or are they such (affections or relations) that they only hold of the form of the function of perception, and, consequently, of the subjective conformation of our mind, without which they could not be predicated of anything whatever ? For answers here we will first discuss the notion of space. I understand, however, by discussion or exposition the distinct statement (if not at full) of what belongs to a notion. Such exposition is metaphysical, moreover, when it demonstrates the notion to be given a priori 1 1. Space is not an empirical notion which has been derived from external experience. For, that certain sensations are referred to something out of me (that is, to something in another part of space than that in which I am), and further, that I can perceive them as out of and near each other, conse- quently, then, not merely as different themselves, but as in different places : to that the perception of space must be already presupposed. Accordingly the cognition space cannot be derived from the rela- tions of external impression, through experience ; but, contrariwise, this external experience is itself only possible through said cognition. 2. Space is a necessary perception a priori, which is presupposed by, and underlies, all external percep- tions. We can never realize to ourselves the con- ception of there being no space, though we can 1 This last very important sentence seems omitted by Rosenkranz, as also a single not very important word in the next paragraph. 142 TEXT-BOOK TO KANT : perfectly well think of no objects being found in space. It is taken for granted, therefore, as condi- tion of the possibility of the appearance of objects to external sense, and not as an affection or form dependent upon objects : it is an a priori percep- tion, which is necessarily presupposed as ground (or canvas) for the reception of all external conscious- nesses. 3. Space is not a discursive or, as we say, general notion of the relations of things, but a pure percep- tion. For, firstly, we can conceive only a single space, and when we speak of spaces, we mean only parts of one and the same sole space. These parts cannot precede, either, the one all-comprehending space as though they were the particulars from which it is generalized ; but, on the contrary, they are only thought in it. It is essentially one ; any plurality of parts or units in it (consequently, also, the general notion of spaces) rests solely on limitations of itself. From this it follows that a perception a priori under- lies all notions of it. This is the reason why every geometrical proposition, as, for example, that any two sides of a triangle are together greater than the third side, is never by any possibility to be deduced from mere general notions of triangle, line, etc., but from perception, and a priori, with apodictic cer- tainty. 4. Space is conceived as an infinite magnitude there before us. Now a notion must be conceived, indeed, as common to an infinite number of different possible individuals (it is their common type), which individuals, therefore, it holds under it ; but no notion as such can be so thought as though it con- tained an infinite number of individuals in it. But it is thus that space is thought (for all the parts of space THE KRITIK OF PURE REASON. 143 are at one and the same time together in it ad infinitum). Consequently the original of space is perception a priori, and not notion. § 3. Transcendental Exposition of the Notion of Space. By transcendental exposition I understand the demonstration of any notion as a principle such, that, through it or from it, the possibility of other a priori synthetic cognitions may be understood. The requisites here, then, are : 1, that such cognitions actually do derive from the given notion ; 2, that these cognitions are only possible on presupposition of a certain mode of interpreting or explaining the given notion. Geometry is a science determinative of the pro- perties of space, synthetically, but yet a priori. What must space itself be, then, that such cognition is possible of it ? It must be originally perception ; for no propositions that, as is the case in geometry (Introduction, V.), exceed (contain more than) a notion, can possibly be derived from that notion. The perception, again, must be a priori, or found in us before any special sense-perception ; pure, there- fore, or non-empirical. For geometrical propositions are all apodictic ; that is, they bring with them their own necessity ; as the proposition, for example, that space has only three dimensions. But such pro- positions cannot be empirical judgments (judgments of experience) ; neither can they be inferred from these (Introd., II.) How, now, can there be in the mind an external perception, which yet precedes any perception of objects, and in which (from its nature, namely) the notion of these may be a priori determined? In no 144 TEXT-BOOK TO KANT : other way, plainly, than that this perception has its seat only in the subject, as mere form of general external sense, or as mere formal susceptivity of the subject in assumption of objects when affected by them ; through (and with) which, then, there is obtained immediate cognition, that is, perception, of these objects. Our explanation alone, therefore, makes geometry conceivable as a synthetic cognition a priori. Every mode of explanation which does not effect this, what- ever similarity it may exhibit, can, in the surest way, through this characteristic, be distinguished from it. Inferences from these Ideas. a. Space exhibits no property of things in them- selves, nor yet themselves in their own mutual relations. It neither represents nor conveys any affection or attribute of things, which were theirs in themselves, and which would remain even if abstrac- tion were made from every subjective condition that belongs to perception (as a function). For neither absolute nor relative attributes can a priori be per- ceived, that is, before existence of the things them- selves in which they are found. b. Space is nothing else than merely the form of all presentations in external sense. It is that subjective condition, under which alone external perception is possible for us. Inasmuch, now, as the susceptibility of the subject to be affected by objects necessarily precedes any perception of these objects, we can easily understand how the form of all perceptions may be already present in the mind before all or any actual special perception, and, consequently, a priori. So present in the mind, we can readily understand, THE KRITIK OF PURE REASON. 145 also, how it, this form, as a pure perception, in which all objects, as presenting themselves in it, must submit to determination from it, may possess prin- ciples of the relations of things before any experience. Only, then, from the point of view of a human being is it that we can speak of space, of extended substances, etc. Directly we discount the subjective condition under which alone external perception is possible to us (so far, namely, as we may happen to be affected by objects), the expression space is without meaning. This term is referred to things only in so far as they appear to us, only in so far as they are objects of sense. The invariable form of this recep- tivity, which receptivity we name sense or sensibility, is a necessary condition of all the relations in which objects are perceived as external to us; but these objects being abstracted from, it is only a pure per- ception (a void subjective form) which has got the name space. Inasmuch as we cannot make the peculiar conditions of sense, conditions as well of the very pos- sibility of things, but only of their appearance to sense, it is impossible for us to say that space contains all things as they are in themselves, no matter what subject per- ceives them, and no matter whether they are perceived or unperceived by any subject, but only that it con- tains all things so far as, externally, they sensuously appear, and to us. For, as regards the perceptions of other thinking beings, we cannot at all judge whether they are confined to the same conditions which limit our perception and are universally binding for us. Only when we add the mode to judgments, do they become unconditionally true. The proposition, All things are together in space, holds good under the limitation that these things are understood to be objects of our perception of sense. "When I add the K 146 TEXT -BOOK TO KANT : condition here, and say, All things, as external percep- tions, are together in space, then the rule is valid universally and without restriction. Our exposition asserts, therefore, the reality of space in regard to every- thing that may come externally before us as an object but no less the ideality of this same space in regard to things when these things mean things in themselves as taken up in their truth by reason and without reference to the special nature of our sensibility. We maintain, therefore, the empirical reality of space in regard of all possible external experience, but also its transcendental ideality, in this respect, that it is nothing so soon as we cease to regard it as condition of the pos- sibility of all experience for us, and assume it, rather, to be something that is involved in the very nature of things in themselves. But, besides space, there is no other external cog- nition which, though subjective, can be called ob- jective, and that, too, a priori. For from no other can we derive synthetic a priori propositions, as we derive them from perception in space, § 3. Hence, to speak accurately, ideality attaches to no such others, even though agreeing with space in this, that they belong merely to the subjective nature of the particular form or mode of sense, — of sight, hearing, feeling, for example, through the sensations of colours, sounds, warmth ; which, at the same time as well, being merely sensations and not perceptions, enable us, of themselves, to know no object whatever, and certainly not possibly a priori. I say this only to prevent resort on our part to inadequate exemplification of the ideality concerned, as from colours, taste, etc., which are rightly enough regarded, not as qualities of things, but as changes of our own subject ; which, further, may even be dif- THE KRITIK OF PURE REASON. 147 ferent in different individuals. For, in that case, what is originally only appearance to sense, say a rose, notwithstanding that it may differ in colour to every different eye, is still held valid in the empirical under- standing as a thing in itself. On the other hand, the transcendental understanding of perceptions in space is a critical reminder that nothing at all that is per- ceived in space is a thing in itself, and that space itself is not a form of things which were appertinent to them in themselves ; but that the things in them- selves are not at all known to us, and that what we call external things are nothing else than mere pre- sentations of our own sensibility ; of which presenta- tions the form is space, but of which, again, the true correlates, that is, the things in themselves, neither are nor can be known thereby ; after which things, indeed, there is never in experience even any inquiry. Section II. — Of Time. § 4. Metaphysical Exposition of the Notion of Time. 1. Time is not an empirical notion which has been derived from any experience. For co-existence and succession would not themselves be found in the things perceived, were not time a priori implied. Only on the pre-supposition of time is it conceivable that some things are at one and the same time (to- gether) or that others are in different times (after one another). 2. Time is a necessary cognition which is implied in all perceptions. We cannot suppress time as in regard to things, but we may very well suppress 148 TEXT-BOOK TO KANT: things as in regard to time. Time, therefore, is a datum a priori. Only in it is all actuality of things possible. These may fall away bodily, but it (as the universal condition of their possibility) cannot be dis- pensed with. 3. On this a priori necessity, the possibility of apodictic propositions in regard to relations of time, or axioms in regard to time generally, is established. It has only one dimension : different times are not to- gether, but after one another (just as different spaces are not after one another, but together). These pro- positions cannot be derived from experience, for ex- perience would yield neither strict universality nor apodictic certainty. Were experience the source, we should only be able to say : That is what common observation tells us ; but not : That is what, of neces- sity, must be. These propositions are binding as rules, under which experience, generally, is possible, and advise us before it, not through it. 4. Time is not a discursive or, as we say, general notion, but a pure form of sense-perception. Different times are only parts of precisely the same time. The cognition which can be yielded only by a single object is perception. The proposition, also, that different times are never co-existent cannot be deduced from a general notion. It is a synthetic proposition, and not dependent on mere notions. It is directly implied, therefore, in the simple perception and conception of time. 5. The infinitude of time amounts to no more than that every particular magnitude of time is possible only through limitations of a one universal under- lying time. Hence the original cognition time must be given as unlimited. That object, however, the parts and every magnitude of which can be conceived THE KRITIK OF PURE REASON. 149 as determined only through limitations, cannot, as a totality, be given through notions (for notions only contain subnotions which, as particulars, precede their principals), 1 but must involve a direct perception. § 5. Transcendental Exposition of the Notion of Time. I may refer in this connexion to § 4, No. 3, where, for the sake of brevity, I have introduced into the metaphysical exposition, what, properly, is transcen- dental. I add now that the notion of change (with that of motion as local change) is possible only through and in time : that if time were not perception a priori (internal), no notion whatever could make intelligible the possibility of a change ; that is, of a conjunction in one and the same object of predicates contradic- torily opposed the one to the other (as the being and not being of one and the same thing in one and the same place). Only in time can such predicates be found together in the same thing — i.e., after one another. And so our view of time explains the possibility of 1 I have conveyed here both forms of the parenthesis found in the two editions. Kant's " parts " and " parts of composition " give pause, especially in the number 3 of Space. This very parenthesis (even in its two forms), however, is, perhaps, decisive. Comparison and reflection seem to me undeniably to demonstrate that Kant had no idea but that of contrasting perceptive parts with notional parts, and that even by the words " Bestandtkeile " and "Zusammensetzung" (,§ 2, No. 3) he meant only subnotions and generalization. The words themselves are unhappy, however ; and much is inexact throughout these sections. Consider the confusion, grammatical and other, of the sentence to which this is note. Literally translated, it would run thus : — " Whereof, however, the parts themselves, and every magnitude of an object, can be conceived deter- mined only through limitations, there the whole cognition must not be given through notions (for these contain only part-cognitions), but there must underlie them immediate perception." The " whereof," the "every magnitude of an object," the "whole cognition," the "them," all most obliquely put, refer, however helplessly, to the one subject or object which is alone spoken of. In Rosenkranz all is even worse. 150 TEXT-BOOK TO KANT: as much synthetic cognition a priori as is demonstrated by the general theory of motion, which is not a little fruitful. § 6. Inferences from these Ideas. a. Were abstraction made from all subjective con- ditions of perception, time would not be found to remain, whether as something self-subsistent and on its own account, or as an objective quality inherent in things themselves. For, in the first case, it would be something which, without actual object, were, nevertheless, itself actual. And, in the second case, it would be impossible for it, as a quality or order belonging to things, to precede these things, as their very condition indeed, and be, through synthetic pro- positions, a priori cognised and perceived. This latter circumstance is very intelligibly possible, should time be the subjective condition only under which all per- ceptions in us can take place. For in that case this form of inner perception is in consciousness before the objects, and, consequently, a priori. b. Time is nothing but the form of internal sense, that is of the perception of our own self and of our own inner state. For time results not from any determination of outer objects ; it is not referred to anything that has bodily shape or place, etc. ; on the contrary, it is time that, for all presentations in our inner consciousness, determines their relation. And just because this inner perception offers no shape, we seek to supply its place by analogies. We picture time-succession as a line that proceeds into infinitude, the complex of parts in which, moreover, constitutes a series which is only of one dimension. From the qualities of this line, too, we conclude to all the qualities of time, except this single one that, while in THE KRITIK OF PURE REASON. 151 the line all the parts are at once and together, those in time are always successive or after one another. From this it is evident that time is a perception ; for all its relations are capable of being expressed in external perception. c. Time is the formal condition a priori of all sense-perceptions. Space, as the pure form of all outer perception, is limited, in its function of a priori condition, merely to external objects. On the other hand, because all cognitions, whether due to external things or not so due, do, so far as they themselves are concerned (in that they are affections of mind), belong to our inner state — further, because this inner state must come under the formal condition of inner perception which is time — it follows that time is an a priori condition of all sense-perception, immediately of internal (the soul) and mediately (i.e., through it) of external perception. As, in the external reference, I can say, All external perceptions are in space and a priori determined according to the relations of space ; so, in the internal reference, I can equally say, All perceptions whatever (all objects of the senses) are in time, and fall necessarily under relations of time. If, from our mode of internally perceiving ourselves, and accordingly disposing in consciousness all exter- nal perceptions, we abstract, and, consequently, take objects as they may be supposed to be in themselves, then time is nothing. It is only of objective validity in regard to perceptions, because we recognise these as objects of our senses ; but such validity disappears directly we abstract from what mode of consciousness is peculiar to us (which is that of a perception only through sense) — directly we speak, namely, of things as such. Time, therefore, is solely a subjective con- dition of our (human) perception (which is in every 152 TEXT-BOOK TO KANT ■ case sensuous, objects being conceived to act on us) ; and, in itself, apart from the subject, nothing. In regard of all perceptions, however, consequently of all things which may appear in experience, time is no less necessarily objective. We cannot say, All things are in time ; for such expression bears to con- sider things as they are in themselves, and apart from the mode and conditions of the perception of them ; whereas it is precisely the mode and conditions of perception from which it follows that time adds itself to all objects in consciousness. But subjoin now the mode to the proposition, and say, All things are, as objects of sense-perception, in time; then the judg- ment has its own good objective truth and universality a priori. Our doctrine asserts, then, the empirical reality of time ; that is, its objective validity in regard of all objects which may, on any occasion, be offered to our senses. And as our perception is at all times one of sense, there never can be given us an object in expe- rience which is not submitted to the condition of time. But, again, we deny time all claim to absolute reality, if regarded as intrinsic condition inherent in things themselves, irrespective of the form of our sensuous perception. Such attributes as belong to things in themselves can never be made known to us by the senses. In this, then, consists the transcen- dental ideality of time ; which, abstraction being made from the subjective conditions of sensuous perception, is absolutely nothing; and cannot be attributed to objects in themselves (or apart their relation to our perception), whether as subsist en t or as inherent. But this ideality is just as little as that of space to be put upon a par with the subreptions of sensation ; in whose despite, there is attributed to the subject of THE KRITIK OF PURE REASON. 153 such predicates an objective reality, which is with- out place here (for time or space), unless in so far as such reality were regarded as merely empirical or attributed to the object (time or space) only as a perception of sense : on the distinction in question, however, see § 3, b, two last paragraphs. § 7. Further Explanations. Against this theory, which grants empirical but denies absolute or transcendental reality to time, I have heard an objection so common on the part of intelligent men, that I infer it must occur naturally to every reader, to whom such speculations arc un- usual. It runs thus : Changes are actual, as is demon- strated by the vicissitude of our own mental states, even should we leave out of view all external percep- tions (together with their changes). But changes are only possible in time. Therefore time is something actual. The reply has no difficulty. I grant the entire argument. Time is undoubtedly something actual ; it is the actual form, namely, of internal perception. It has therefore subjective reality in regard of inner experience ; i.e., I have actually the consciousness of time, and of my determinations in time. It is actual, consequently, not as an object, but as the mode of my perception of myself as an object. But if I (or another) could perceive myself without this condition of sense, the same states, which we now call changes, would yield a cognition into which no idea of time, or consequently of change, would at all enter. There remains to it, therefore, its empirical reality as condition of all our experiences. Only absolute reality, in accordance with what has been said, cannot be allowed it. It is nothing but 154 TEXT-BOOK TO KANT: the form of our inner perception. 1 If we withdraw from it the peculiar condition of our sensibility, its very idea disappears ; for time is not inherent in things themselves, but simply in the subject perceiving them. But the reason why this objection is so universal, and on the part of those, too, who have nothing decided to advance against the doctrine of the ideality of space, is this. The absolute reality of space they could not hope apodictically to demonstrate in view of idealism, according to which the actuality of ex- ternal things is incapable of rigorous proof. Whereas the actuality of the object of our internal senses (my own self, my own state) is immediately clear in con- sciousness. The former may, possibly, be a mere show, while the latter is, in their opinion, something undeniably actual. They do not consider that both, without our presuming to deny their actuality in consciousness, are, nevertheless, only appearances to sense, which has always two sides. There is one side, for example, in regard to which the object is viewed as in itself (apart from the mode of its percep- tion, in which respect its nature is always problema- tical). And there is another side where the form of the perception is considered ; which form must not be sought for in the object as in itself, but in the subject to which it appears ; at the same time that said form belongs, nevertheless, actually and necessarily, to the appearance of the object. Time and space, accordingly, are two sources of cognition, from which, a priori, various synthetic 1 I can, indeed, say, my states follow one another ; but that means no more than that we are conscious of them in a sequence of time, i.e., ac- cording to the form of inner sense. Time is not by any means, there- fore, something in itself, nor yet any attribute objectively inherent in things. — K. THE KRITIK OF PURE REASON. 155 propositions may be derived, as is especially exem- plified in pure mathematic with regard to space and the relations of space. Taken together, namely, they are both pure forms of all sense-perception, and thereby render synthetic propositions a priori possible. But these cognitive sources a priori determine their own limits just by this reference to their being conditions (forms) of sense : they concern objects, that is, only so far as objects are considered perceptions of sense, and not things in themselves. Valid only for the former, they at once cease to have objective application directly we go beyond them. Such reality of space and time leaves, for the rest, the certainty of our empirical knowledge unaffected ; for in its regard we have an equal certainty, whether these forms are of things in themselves, or only of our perception. Whereas they who maintain the absolute reality of space and time must, whether they assume subsistence or only inherence, be at variance with the principles of experience itself. For, say they assume the former, as the mathematical inquirers mostly do, then they have before them two eternal, infinite, and self-sub- sistent non-entities (space and time) which, without being themselves anything actual, are there, for all that, for no other purpose than just to embrace all that is actual! Or say they assume the latter (in- herence), as is, in effect, the case with certain meta- physical dogmatists, then, inasmuch as space and time are for them relations of things (the beside one another, the after one another) derived from experience, but necessarily only confusedly so, they (these dogmatists) must impugn the validity, or at least the apodictic certainty, of any mathematical assignments a priori m regard of actual things (e.g., in space). For such certainty is not possibly to be obtained from expe- 156 TEXT-BOOK TO KANT: rience ; and any a priori notions of space and time can, under such suppositions, be no more than creations of imagination. That is, their source must actually be sought for in experience so far as imagination, out of the relations abstracted from experience, has made a certain universal of these, but a universal, neverthe- less, still under the restrictions imposed by nature upon the relations from which it derives. 1 The party for the former opinion have the advantage that, for their mathematical allegations, they leave the field of the experiences of sense free. But then they very much perplex themselves by these very conditions, when the understanding would go beyond this field. The party for the latter opinion, again, have the advantage that space and time are not difficulties to them, when they would judge of objects, not as per- ceptions of sense, but only in relation to the under- standing. But they are unable either to explain the possibility of mathematical cognitions a priori (for any true and objectively valid perception a priori does not exist for them), or to bring the findings of experience into necessary agreement with these cog- nitions. In our theory of the true nature of these two primitive forms of sense, both difficulties are removed. Lastly, that the transcendental esthetic cannot include more than these two elements, is evident from this, that all other notions which hold of sense (even 1 Were all Kant's sentences like the above, De Quincey's ridicule would be very much in place. I have broken it up, and done my best with it, but I fear its import must be still obscure. What is said of the mathematicians seems plain enough ; and as regards the metaphysicians, all that I take to be intimated is, that, all being a posteriori with them, they must find themselves at a non-plus in face of the a priori ; while, further, their universals of time and space, derived only from the action of the imagination on the contributions of sense, must submit themselves to the restrictions of that, their empirical source. THE KRITIK OF PURE REASON. 157 motion, which is a union of both) presuppose something empirical (as subjects or objects of them). Motion, for example, presupposes perception of something that is movable. In space, however, taken by itself, there is not anything that is movable. Therefore what is movable must be something that is only found in space by experience, or that is only an empirical datum. For the same reason, also, the transcendental aesthetic cannot count among its a priori data the notion of change ; for time itself undergoes no change ; only what is in time undergoes change. For that notion there is required, therefore, the observation of some actual existence and of the succession of its states, i.e., of experience. § 8. General Remarks on the Transcendental ^Esthetic. I. First it will be necessary to explain as clearly as possible what, in regard to the nature of sense-cogni- tion, our opinion is, in order to preclude all misunder- standing in that respect. It has been our wish to say, then, that all our perception is nothing but the impression of sense (the state of mind due to sense-presentation) ; that the things we perceive are not in themselves as we per- ceive them ; that this holds good of their relations as well; and that, were our subject abstracted from, or simply the subjective constitution of our senses, all the qualities and all the relations of objects in space and time — nay, space and time themselves — would disappear : for all of these are, as mere appearances to sense, incapable of existing in themselves, but only in us. How it may be situated with the objects in themselves, and apart from our receptivity of sense, remains wholly unknown to us. We know nothing 158 TEXT-BOOK TO KANT : but our manner of perceiving them, which, as peculiar to us, is necessary to us, but not therefore necessary to every other intelligence. But it is with the per- ception peculiar to ourselves that we have alone to do. Time and space are its pure forms, and sensation its matter. Only the former can we cognise a priori, or before actual perception of sense, and for that reason we name them pure perception. The latter, again, as that in our cognition which is only a posteriori, we name empirical perception. The former belong to our sensibility absolutely necessarily, let our sensa- tions be as they may ; and very various they may be. Though we should bring our perception to never so high a degree of keenness, we should not, for all that, be a bit nearer the nature of objects in themselves. For, in every event, we should only be present to our own mode of perception, to our own sensibility — only to this sensibility, moreover, as under the originally - inherent, subjective conditions of space and time. What the objects may be in themselves can never possibly be known to us by even the most luminous cognition of their appearance to sense, and it is that appearance which is alone given us. Wherefore, that our whole complex of sense is nothing but a confused cognition of things, possessed, indeed, of what belongs to them in themselves, but only in the midst of such heaping together of characters and part-perceptions as renders it impos- sible for us consciously to distinguish them — this is such a falsification of the very idea of sense or of object of sense, that it reduces the whole theory of these to vanity and inanity. The difference of an indistinct from a distinct consciousness is simply logical, and does not refer to the contained matter as the contained matter. There is no doubt that the THE KRITIK OF PURE REASON. 159 notion Right (Recht), for example, even as used by the ordinary understanding, involves all that the subtlest speculation can develop from it ; differing from the latter only in this, indeed, that it is without con- sciousness of the many details of the thought. But all this gives us no reason to say that the common notion is only one of sense or appearance to sense ; for Right cannot by any possibility come before sense at all. Right is a notion, its seat is in the under- standing; and it is a (moral) quality of actions which belongs to these in themselves. On the other hand, the idea of a body in perception contains nothing at all that were capable of belonging to an object in itself. We have in it only appearance to sense, or the manner in which we are thereby affected ; and this receptivity of our cognitive faculty is called sense or sensibility, and remains, from a cognition which should concern an object in itself, even if the sense-appearance were seen to the very bottom, nevertheless diametrically different. The Leibnitz-Wolfian philosophy has, therefore, to all investigations into the nature and orioin of our knowledge, assigned quite a wrong point of view. To it, namely, the difference of sense and intellect was wholly logical, whereas, in effect, it is mani- festly transcendental. It is not the form of dis- tinctness or indistinctness that is concerned in this difference, but the origin and nature of our knowledge. From which it results that, through sense we know the nature of things in themselves, not indistinctly only, but absolutely not at all. So soon, indeed, as we leave out of view our own subjective conforma- tion, the qualities of the object (as attached to it by sense) and the perceived object itself are nowhere to be found ; for it is just this subjective confor- 160 TEXT-BOOK TO KANT : mation which determines the form of the object as an appearance to sense. We do usually, it is true, distinguish in objects what is substantial in the perception of them, what in them is valid for every human consciousness, from what, again, is only contingent in them. The latter, namely, unlike the former, is not referred to sentiency as such, but only depends on a special situation or organization of this or the other sense. Or the former is considered such a cognition as perceives the object in itself; the latter, again, only such as perceives the object in its appearance to sense. But even this distinction is, after all, only empirical. If we remain by no more than this (as is commonly the case), and fail to regard (as we ought) such empirical perception as itself again but mere sense- appearance, in such wise, namely, that there is nothing at all to be found in it that concerns any- thing whatever in itself, then our transcendental distinction is all lost. For, so, we believe that we perceive things in themselves ; whereas, nowhere in the world of sense, let us search into its objects as deeply as we may, have we ever anything to do but with sense-appearance. Thus it is that, in the case of a sun-shower, we call the rainbow a mere appearance of sense ; at the same time that we take the rain to be the thing in itself. Nor is this incorrect, in so far as we regard the rain only physically, as what, in experience generally, under whatever position to sense, is, in perception, always thus and not other- wise determined. Should we take, however, the empirical phenomenon all together, and ask, without any reference to the distinction of agreement or not with every human sentiency, whether this pheno- menon as a whole indicates an object in itself (not THE KRITIK OF PURE REASON. 1GL indicates the rain-drops as such object, for they are themselves empirical objects), then the question of the relation of the perception to the object is transcen- dental. For not only these drops are appearances of sense ; but their rounded form, nay, the very space in which they fall, are nothing in themselves unless mere modifications of our senses (the colours, etc.), or groundworks of our sense-perception generally (space, etc.) ; while the transcendental object, for its part, again, remains wholly unknown to us. 1 A second important characteristic of our tran- scendental sssthetic that demands notice is this. It is not something that should gain some favour merely as a plausible hypothesis. On the contrary, it is as certain, and as free from doubt, as can ever be required of any theory that shall act as an organon. In order to make this certainty fully conspicuous, we shall take a case, the evidence of which may prove irresistible as well as throw additional light, perhaps, on what has been said in § 3. Suppose, then, you take space and time to be in themselves objective, or to be conditions of the pos- sibility of things in themselves. In that case this fact courts notice, that a great number of a priori apodictic and synthetic propositions presents itself from both, but more especially from space, which, therefore, we shall preferably refer to in example. Now, I ask you, as the propositions of geometry are known synthetic- ally a priori and with apodictic certainty, whence do you derive these propositions, and on what does under- 1 As we never at all know the object in itself, we may be apt to think that Kant ought to have called it, not transcendental, but transcendent. The Ideas, however, though transcendent as cognitions, are regulatively transcendental in experience. In the same way, the object in itself, as necessary to and in experience, is, though transcendent in cognition, a constitutively transcendental element. L 162 TEXT-BOOK TO KANT: standing support itself in dealing with them ? Source or support there can be none, but either in notions or perceptions ; and from both of these, again, either only as a priori or else as a posteriori. But empirical notions or empirical perceptions are only adequate to what is itself, again, only empirical. They are incap- able of the necessity and absolute universality that are characteristic of the propositions in geometry. Even a priori notions we are called upon to eliminate here ; for it is clear that, from mere notions, there cannot be got any synthetic, but only an analytic cognition. Take the proposition, for instance, Two straight lines cannot inclose a space or construct a figure, and try to deduce it from the notion of straight lines and the number two ; or say even that a figure is possible with three straight lines, and try this with mere notions. All your trying is in vain, and, like geometry itself, you find you are compelled to have recourse to perception. You take an object in percep- tion, then ; but your perception here must, as already shown, be an a priori and not an empirical perception. You must feel, in answer to our question, consequently, that the source and the support required in geometry are a priori perception. Were there not within you a faculty of perception a priori ; were this subjective condition not at the same time, in form, the uni- versal condition a priori under which alone the object of this (external) perception is itself possible ; were the object (triangle) something in itself irre- spective of your subject : how could you say, that what for construction of a triangle lies necessary in your subjective conditions, must necessarily be found also in the triangle itself? It was impossible for you, confined to notions (of three lines), to add to them something new (the figure), which, therefore, must THE KRITIK OF PURE REASON. 163 necessarily be conveyed by the object, inasmuch as this object is a datum before your cognition (of the pro- position in question), and not through that cognition. Were not space, consequently, a mere form of your perception, involving conditions a priori, under which alone things can be for you external objects, as without them they were nothing, you would be quite unable to determine anything synthetically and a priori in respect of external objects. It is therefore indubitably certain, and not merely possible or probable, that space and time, as the necessary conditions of all (outer and inner) experience, are mere subjective conditions of all our perception. In relation to these conditions, conse- quently, all objects are mere sense-appearances and not things on their own account. It is just because of these a priori sense-conditions, too, that much as re- gavds form may be apriori said of sense-objects, though never the smallest word of the things in themselves that may possibly underlie these sense-objects. II. In confirmation of this theory of the ideality as well of external as internal sense, and, consequently, of all objects of sense as mere sense-appearances, it may prove signally serviceable to remark : That what- ever belongs to our external perception, involves nothing but mere relations, as of places in a percep- tion (extension), change of places (motion), and laws determinative of such change (motive forces). But, further, what is in the places, or what apart from the local change acts in the things themselves, is not at all made known thereby. Now, through mere rela- tions, there is not anything in itself given. It is easy to judge, consequently, that, external sense yielding us nothing but intimations of relation, said sense is competent to convey the relation of an object to the subject in perception of it, but not the inner consti- 164 TEXT-BOOK TO KANT: tution that belongs to the object in itself. With internal perception the case is the same. Leaving out of view that even there the contributions of ex- ternal sense constitute the material proper with which we furnish our minds, it is to be said that time, in which we place these contributions, — which itself pre- cedes consciousness of these in experience, and is pre- supposed as underlying formal condition of the man- ner in which we place them in the mind, — that time already prescribes, I say, relations of succession, of co-existence, and of what is implied with succession (a substrate that persists). Now, what, as a cognition, can precede all action to think anything, is perception, and, if it convey nothing but relations, only the form of perception. But, as this form of perception is only operative so far as the mind has an object in it (contents), it can be nothing else than the mode in which the mind, through its own action (the placing of contents), is affected — the mode, then, in which the mind is affected by its own self. This, plainly, amounts to an inner sense, or to an inner sense at least on the side of the form. All that is perceived by sense is always, so far, sense-appearance. An internal sense, therefore, would either require not at all to be admitted, or, if admitted, it would require to be seen that the subject, which is the object of such sense, could be perceived by it only as sense- appearance, and not as it would be judged to be by its own self, were its perception intellectual, or pro- duct of its own spontaneity. All difficulty here con- cerns the question alone of how a subject can in- ternally perceive its own self; but this difficulty is common to every theory. Consciousness of one's self (apperception) is the simple cognition Ego; and were, thereby alone, all complex of elements constitutive of THE KRITIK OF PURE REASON. 165 the subject spontaneously given — in that case the internal perception would be intellectual. In man consciousness requires internal sense-perception of the complex that is given in the subject; and the manner in which, without spontaneous action of the mind, this complex is presented to the mind, must, for the sake of the distinction implied, be called sense. If consciousness is to take up (apprehend) what is in the mind, it must affect it ; and is only able in this way to effect a perception of itself. The form of this perception, however, already in the mind, determines as in time how the complex is collocated in the mind. In a word, consciousness perceives itself, not as it would perceive itself were it immediately self-active in perception, but according to the way in which it is internally affected; consequently, as it sensuously appears to itself, not as it is. III. When I say the object of perception, whether external or internal, is exhibited in space and time only as it affects our senses, or as it appears, I do not mean by that, that said object is a mere deception. For, in sense, the objects, nay, even the qualities we attribute to them, are always regarded as something actually given. Only, the particular subject's par- ticular mode of perception being considered, a dis- tinction is made between the object viewed as appear- ance to sense, and again as a thing in itself. So to speak, the single object of consciousness is, as pheno- menon, or mere appearance before sense, distinguished from its own self as noumenon or thing in itself before reason. When I maintain, therefore, that the quality of space and time, in measure of which quality, as condition of their very being, both external and internal object must set themselves — when I maintain that this, the quality of space and 166 TEXT-BOOK TO KANT: time, lies in me, in my mode of perception, and not in said objects in themselves, I by no means say that the one object (the things without) merely decep- tively seems outside of me, or that the other object (my own soul) merely deceptively seems given in my self-consciousness. It would be my own fault, if I made a deception of what was an object of sense proper — if I made Schein of what was an Erschein- ung. 1 Such a blunder is no result of our principle, however — of the ideality, that is, of all our percep- tions of sense. Rather, were objective reality attri- buted to our said sense-forms, the result then would be the unavoidable transformation of all and every- thing into a mere mock-show. For, were space and time regarded as entities such that, in their very possibility, they were necessarily found in things in themselves, then we should have before us two infinite things which, though not substances, nor even any- 1 In relation to sense, the predicates of sense may be allowably attri- buted to tbe object (Erscheinung), as redness or fragrance to the rose. But illusion, false show (Schein), can never be attributed as predicate to the object. And the reason is that, in the case of illusion, we attribute to the object in itself what belongs to it only in relation to sense, or indeed to a subject generally ; as, e.g., the " two handles" were attributed at first to Saturn. What is not at all to be found in the object in itself, but always in its relation to the subject, and is inseparable from the per- ception of the former, is sense-appearance ; and the predicates of space and time, consecpiently, are rightly attributed to the objects of the senses as such. That is sense-appearance (Erscheinung), and not sense-illusion (Schein). On the other hand, let me attribute the redness to the rose in itself, the "handles" (as existent fact) to Saturn, or extension to all outer objects in themselves, without consideration of the particular relation of the object to the subject in each of these cases, and without accordant limitation of my judgment — then I have involved myself in, or given rise to, illusion. — K. The ring of Saturn, when first seen by Galileo, looked like " two handles." This was in 1612. In 1655, again, Huyghens explained the "handles" by reference to the ring. And in 1715 Cassini discovered that the ring was double. Since 1850 a third ring has been added, and what is now talked of is Saturn's " series " or " system " of rings. THE KRITIK OF PURE REASON. 167 thing actually inherent in substances, were yet some- thing existent, nay, the necessary conditions of the existence of all things ; and which would continue to exist were all other things put an end to. We have only to reflect on the absurdities in which this sup- position would involve us, to find it very pardonable on the part of the good Berkeley that he reduced all things into a mere illusion. Why, even our own existence, were it conceived dependent in such fashion on the self-subsistent reality of a nonentity like time, would, with time itself, be necessarily transformed into a mere show, — an absurdity for which no man as yet has made himself responsible. IV. In natural theology where what is thought is not only for us no object of perception, but never can be even to its own self an object of setisuoas per- ception, we are careful to remove the conditions of time and space from all perception on the part of such object (for cognition in such a case must be perception, and not thought, which always shows limits). But with what right should we do this, if we have first of all assumed both time and space as forms of things in themselves, and such as would continue to be a priori conditions of things, even if these things themselves were once for all annihilated ; for, as conditions of existence as a whole, they must necessarily be conditions of the existence of God ? But if we are not to make them objective forms of all things, then there is nothing left us but to make them subjective forms of our own mode of perception, whether outer or inner — a mode of perception, further, which is to be recognised as sensuous for the reason that it is not original. An original perception, namely, is such that through it the very being of its object is given ; and this is a perception which, so 168 TEXT-BOOK TO KANT: far as we see, can only belong to God. A sense-per- ception, such as ours, on the contrary, is dependent on, and subservient to, the object, and is conse- quently only possible by this, that the perceptivity of the subject is by said object affected. It is not necessary, either, that we should con- fine a perception in space and time to the sensibility of man. It may be that all finite thinking beings must, in that respect, necessarily be identical with us (though we cannot decide as much) ; but it would not follow, from this universality, that such a mode of perception were not still sense. It would still be a derivative perception {intuitus derivativus), and not original (intuitus originarius). That is, it would not be an intellectual perception, such as, for the reason alleged, appears to belong to God only, and never to a being that is dependent as well in its existence as in its perception (which is determina- tive of its existent states in regard to given objects). But this latter remark is only in place here in our aesthetic theory as an illustration, and must not be accounted a ground of proof. Conclusion of the Transcendental ^Esthetic. In resolution of the general problem of our tran- scendental philosophy (How are synthetic propositions a priori possible?), we now possess here one of the required resources. We have now, namely, pure a priori perception, as such resource, the forms of which are space and time. In these, when, in an a priori judgment, we would go beyond a given notion, we have the means of finding what can be a priori discovered (not, indeed, in the notion, but very certainly in the perception correspondent to it), THE KRITIK OF PURE REASON. 169 and may be synthetically united with it (the notion). That, however, amounts to certain judgments ; which judgments can, at the same time, never extend further than to objects of the senses, nor possess validity for any others than those of possible experience. Transcendental Logic. I. Of Logic in General. Our cognition has, on the part of the mind, two sources. Of these the first is the receptivity of im- pressions, and the second the spontaneity of notions. Or the first receives the crude appearances of sense, and the second works them up into the finished per- ception of an object. An object, consequently, is by the first given, but by the second thought — thought, that is, in relation to the sense-impression, the sense- appearance, which, for its part, and solely as such, is merely affection of the sensory. Crude sense-percep- tion and notions, therefore, constitute the elements of all our perfected perception, or perception as ordi- narily understood. Neither notions without sense-ele- ments in some way correspondent to them, nor sense- elements without notions, are capable of furnishing a finished perception. Both, again, are either pure, or else empirical — empirical, when involving special sen- sation (which presupposes the actual presence of an object); and pure, when, in the intimation to con- sciousness, there is no admixture whatever of any element of sensation as such. This element, indeed, sensation as such, may be named the matter of sense- cognition. Pure perception, again (that is, perception 170 TEXT-BOOK TO KANT: as perception properly and strictly so called, and pure as being yet free from either sensation or notion), is tantamount to the form (space and time) under which the perception of anything actual can, in general, take place ; while pure notions, for their part, constitute the form (rather forms) under which an object in general must be thought — in order to be perceived, namely. Pure perceptions or pure notions are alone possible a priori; empirical ones only a posteriori. If we name the susceptivity of mind to receive ele- ments, so far as it is in some certain way affected, sensibility, then the power, on the other hand, to pro- duce elements, or the spontaneity of notions (towards perception, namely), is the understanding. 1 Our con- stitution is such that what we distinctively mean by perception as an element or form, must always be of the nature of sense (though not necessarily a special sensation) ; that is, perception, as perception proper, applies to the manner in which (sensation as sensation conceived apart) Ave are affected by objects. On the other hand, the faculty that, to the sense-perceptive elements in the case of an object, adds the required thought-perceptive elements, is the understanding. Neither of these elements is to be preferred to the other. Without sensibility there were no object per- ceptively given, and without understanding there were no object perceptively thought. Thoughts, 1 Kant's language here will never be understood, if to such words as notion, cognition, spontaneity, understanding, etc., there be given no refer- ence but the usual intellectual one of thought proper as opposed to sense. They must all of them take on, in addition, a direction to perception. Pure perception is time and space. Crude perception is these inspissated, by special sensations, into Erscheinungen, which are objects, but as yet without the foci of the categories. Complete or finished perception, lastly, is, by addition of action from the categories, the ordinary percep- tion of experience proper. THE KRITIK OF PURE REASON. 171 without a content of perception, are void ; perceptions, without the focus of notions, are blind. It is just as necessary, consequently, to add perceptions to one's notions, as to add notions to one's (crude) perceptions. Neither faculty can exchange functions with the other. The understanding does not perceive ; nor the senses think. Only in their union is there what cognition we name finished or perfected objective perception, the perception of experience. But we must not on that account confound their shares in the resultant act; on the contrary, we must carefully separate and distinguish them. Accordingly, we ex- pressly distinguish the Science of the Rules of the Sen- sibility as such (^Esthetic), from the Science of the Rules of the Understanding (Logic). Now logic, again, can be understood in two ways : either as logic general, or as logic special, and in both cases, of course, with reference to the employment of the understanding. The former will contain the absolutely necessary rules of thought, or those rules without which there can simply be no employment of the understanding at all ; it relates to the under- standing without respect of the different objects to which it may be directed. The latter will apply to the rules rightly to think some certain class of objects. We may name the one elemental logic, the other an organon of this or that particular science. The latter is frequently premised in the schools as propaedeutic of the sciences, although it is what is reached latest in the progress of reason — reached, indeed, only when the science itself has long been ended, and requires only the last touch for its due adjustment and final completion. For objects must themselves be under- stood in a pretty high degree, if we are to assign the rules by which a science of them is to be realized. 172 TEXT-BOOK TO KANT: As for general logic, it is either pure or applied. In the former we abstract from all empirical con- ditions under which our understanding is exercised. We abstract in it, for example, from the influence of the senses, the sport of fancy, the laws of memory, the power of habit, inclination, etc. We abstract, consequently, also, in it, from the sources of prejudice, nay, in general, from all causes whence special cogni- tions arise to us or are interpolated, inasmuch as they merely concern the understanding under special cir- cumstances of its application, and to know these experience is required. A general but pure logic has to do, therefore, with mere principles a priarij and is a canon of the understanding and reason, but only as regards the formal element in their use, the matter, on its part, again, being what it may (empirical or tran- scendental). A general logic, on the other hand, is then called applied when it is directed to the rules of the exercise of understanding under the empirical subjective conditions which are taught us by psycho- logy. This logic, therefore, is possessed of empirical principles, although it is in so far general as it con- cerns the exercise of understanding without distinction of objects. It is for this reason also that this logic is neither a canon of the understanding generally, nor an organon of special sciences, but solely a catharticon of the common or ordinary understanding. In general logic, there must be an entire separation between the pure and the applied parts. The former part alone is properly science, though brief and dry, and such as an academical statement requires on the part of an elemental logic. Here, therefore, logicians must have always two rules in their eye. 1. As general logic, it abstracts from all diversity of objects in cognition, and from these themselves ; THE KRITIK OF PURE REASON. 173 it has to do with nothing but the mere form in thinking. 2. As pure, it has no empirical principles, and, consequently, does not (as has been sometimes sup- posed) take anything from psychology, which, in reality, has no influence upon a canon of the under- standing. It is a demonstrated doctrine, and every- thing in it must be quite a priori certain. What (contrary to the common use of the word, which relates to certain exercitia on the rules of pure logic) I name applied logic, is an exposition of the understanding and of the rules of its necessary exer- cise in concreto, namely, under the contingent condi- tions of the subject, which, as such, may either obstruct or promote said exercise, and which collec- tively can only empirically be given. It treats of attention, its obstacles and advantages, the sources of error, the state of doubt, hesitation, persuasion, etc. Pure general logic bears the same relation to it, therefore, that a pure ethic (which contains only the necessary moral laws of free-will as such) bears to the special doctrine of offices, which treats of those laws as under the hindrances of the feelings, desires, and passions, to which mankind are more or less prone. Such doctrine evidently resembles the applied logic, as standing in need, like it, of empirical and psycho- logical principles, and is consequently inadequate to a true and demonstrated science. II. Of Transcendental Logic. General logic abstracts, as we have shown, wholly from the matter of cognition, that is, from any refer- ence of cognition to an object of it; and regards alone the logical form in the relation of the cognitions 1 7± TEXT-BOOK TO KANT : the one to the other, or the form of thought quite generally. Inasmuch, now, as there are (according to the transcendental aesthetic) as well pure as empirical perceptions, it is possible that a like difference may be found between the pure and the empirical thinking of objects. In that case we should have the possibility of a logic in which abstraction from all matter of cognition would not be necessary. For there might be a logic, excluding, indeed, empirical matter, but admitting all that could be a priori cognised (through perceptions or notions) in reference to objects even as experienced in actual fact. Such logic would relate, consequently, to the origin of our actual per- ception and other cognition of objects of experience, so far as that origin did not, or could not, lie in these objects themselves. It is otherwise, of course, with general logic, which, for its part, has nothing to do with any such origin of the actual perception and cognition of objects. On the contrary, it considers only the laws followed by the understanding in its process of thought as concerns objects in their mutual relations generally, without distinction either of these objects themselves or of their origin, whether a priori or empirical. General logic, indeed, treats only of what forms of the understanding ideas must accept, let them originate or be as they may. And here I place a remark which, as influencing all our subsequent proceedings, must be carefully kept in view. The designation transcendental, namely (which means the possibility of such perceptive a priori knowledge, and the rationale of its application in actual fact), is not to be extended to every a priori element, but only to those which enable us to recog- nise the fact that, and the mode how, certain states of consciousness (perceptions or notions) are wholly and THE KRITIK OF TÜRE REASON. 175 solely of a priori possibility and of a priori action. Hence we must neither call space, nor any a priori determination of space, as in forms of geometry, transcendental. What alone is transcendental is, as well the perception that these things (space, etc.) are not of empirical origin, as also the possibility that and how, nevertheless, they may even a priori con- join themselves to actual objects of experience. 1 In like manner, the relation of space to objects generally is transcendental ; but, restricted (with reference to the result) to objects of the senses, it is empirical. The distinction, therefore, between what is transcen- dental and what empirical has place only in the critique of the cognitions, and does not concern the conjunction of these with their objects. In the expectation, then, that there are possibly notions, a priori entrant into objects, not in the manner of perceptions, indeed, whether pure or sensible, but merely as pure thought-functions — notions, conse- quently, which are in origin neither empirical nor aesthetic — we prefigure the idea of a science of pure cognition which, though exclusively holding of under- standing and reason, will enable us to think facts of actual experience even wholly a priori. A science, determinative of the origin, limits, and objective actuality of such cognitions, would necessarily take the name of Transcendental Logic. It would have to do, namely, only with the laws of understanding and reason, and this expressly in an objective application a priori; and not indifferently, like general logic, in reference to interests whether empirical or pure. 1 In the above sentence the word " könne " should evidently be in the plural. Otherwise the only possible nominative to this verb woiüd be " Erkenntniss," which makes tautological nonsense of the sentence, analyze it grammatically as one may. 176 TEXT-BOOK TO KANT: III. Of the Division of General Logic into Analytic and Dialectic. The old and well-known question, with which logicians were supposed to be put to straits and com- pelled either to have recourse to a miserable dialexis * or to admit their ignorance and, consequently, the nullity of their entire business, is this, What is truth? The nominal definition of truth, that it is the agree- ment of cognition with its object, is here admitted and presupposed. But the question we suppose really to be asked, is, What is the universal and certain criterion of the truth of all and every cognition ? It is already no small but an indispensable proof of sagacity and penetration to know what it were rationally proper to ask. For, the question itself being absurd and only calculated to elicit useless answers, it has, besides shaming the questioner, some- times the further disadvantage of misleading the unwary hearer into absurd replies, and suggesting the ludicrous spectacle of one man (as the ancients said) milking the he-goat, while the other holds up a sieve. If truth consists in the agreement of a Cognition with its object, then this object must be thereby distinguished from others ; for a cognition is false, if disagreeing with its object, though possessing some- thing that may well be true of other objects. Now, a universal criterion of truth would be such as holds good of all cognitions, without distinction of their objects. It is plain, however, that, as, in the case of such a criterion, there is abstraction from every matter of cognition (reference to its object), and truth precisely concerns this matter, it is quite im- 1 Rosenkranz has " Dialele " instead of Dialexe. THE KEITIK OF PURE REASON. 177 possible and absurd to ask still after a criterion of the truth of this matter of the cognitions ; and that, therefore, it is impossible also to assign any adequate criterion of truth that shall at the same time be universal. What is to be said here, then, is, that of the truth of cognition as regards matter there is no universal criterion to be required, for any such were a contradiction in itself. But it is equally plain, as regards cognition in mere form (all matter apart), that a logic, confined to the universal and necessary rules of the under- standing, must furnish, just in these rules, criteria of the truth. For whatever contradicts these is false, inasmuch as the understanding would then contra- dict its own universal rules of thought, and conse- quently its own self. These criteria, however, con- cern only the form of truth or of thought generally, and are so far quite correct, but not all-sufficient. For though a cognition were in full agreement with the logical form, and consequently did not contradict itself, it might still, nevertheless, contradict the ob- ject. The merely logical criterion of truth, therefore, agreement of a cognition, namely, with the universal and formal laws of the understanding and reason, is certainly the conditio sine qua non or the negative condition of all truth. Further, however, logic can- not go ; and the error which concerns, not the form, but the matter, is not to be detected by any touch- stone of logic. Now, general logic resolves the whole formal business of understanding and reason into its elements, and exhibits these as the principles of all logical judg- ment in cognition. This part of logic may be called an Analytic, therefore, and is, at least, the negative touchstone of truth ; for by these rules must, first of M 178 TEXT-BOOK TO KANT: all, every cognition, in form, be gauged and tried, before looking to its matter in order to determine whether, with reference to the object, it possesses positive truth. But the mere form of cognition, how- ever much it may agree with logical laws, falls still far short of constituting, as such, material or objective truth. In respect of objects, no one with mere logic can venture to pronounce, or maintain anything; but, having first of all thoroughly inquired into them, logic apart, only afterwards one merely tries the using and connecting of them in a coherent whole on logical laws, or, better still, submits them solely to the test of these. Nevertheless, however poorly off, or quite void, we may be as regards matter, the possession of such plausible art to bestow on all our cognitions the form of the understanding proves so seductive that said general logic, though a simple canon in judging, has, at least for the mere blind show of objective affirmations, been used, or, in effect, misused, as an organon of actual production. Now, general logic, as such supposititious organon, is what we name Dialectic. However variously in meaning the ancients ap- plied this appellation of a science or art, we can always confidently gather from their actual use of it that they intended by it only the logic of {false) shoiv. A sophistical art to give ignorance, nay, intentional trickery, the colour of truth, it imitated the rigour of logic, and applied its topic in concealment of all manner of empty pretexts. Now, we may regard as a safe and serviceable warning the fact that general logic, when used as an organon, is always dialectical, or a logic of show. For, as it tells us nothing of the matter of cognition, but only the formal conditions of agreement with the understanding, which, of course, THE KRITIK OF PURE REASON. 179 in respect of objects, are quite indifferent, we must regard the idea of using it as a means (organ on) of extending and enlarging, at least in pretension, our knowledge — we must regard this as eventuating in nothing but an empty verbiage of affirming, or at will denying, with some show of truth, whatever we please. Such teaching as this is altogether beneath the dignity of philosophy. For this reason dialectic has been included in logic rather as a critique of dialectical show, and it is as such we would have it understood here. IV. Of the Division of Transcendental Logic into the Transcendental Analytic and Dialectic. In a transcendental logic, we isolate the under- standing, as already, in the aesthetic, sense, and make prominent merely the share of thought in our per- ceptive experience, which is alone derived thence. The necessary condition for action of such principles is, that objects be given us in sense-perception, to which then they may be applied. For without such perception, experience, as wanting objects, remains altogether void. That part of transcendental logic, therefore, which propounds the elements of pure understanding in experience, and the principles with- out which no object can anywhere be thought into perception, is the transcendental analytic, and at the same time a logic of truth. For no cognition in ex- perience can contradict it, without losing at the same time all its matter, that is, all its conjunction into an object, and consequently its truth. It is, however, very tempting and misleading to make use of these pure principles by themselves, and even beyond the limits of experience, which can alone furnish the matter or objects whereon to apply them. 180 TEXT-BOOK TO KANT: In this way, consequently, understanding runs risk of making, through mere cobwebs of reason, a material use of its own simply formal principles, and without discrimination judging of objects which are neither given us, nor in any way, perhaps, can be given us. Specially calculated to yield only a canon of judgment in experience, they are merely misused, when, apply- ing them universally and without restriction, we ven- ture, in respect of objects generally, with pure under- standing alone, synthetically to judge, pronounce, and decide. Such use of pure understanding were dia- lectical. The second part of transcendental logic, therefore, must consist of a critique of this dialectical show, and be named Transcendental Dialectic. We are not to expect in it, however, an art dogmatically to produce such show, which, alas ! is a very current art of manifold metaphysical juggleries. Quite on the contrary, it shall be a critique of understanding and reason in their hyperphysical use, in order to detect the false show of their groundless pretensions. Their supposed claims, therefore, to discovery and extension through mere transcendental principles, it will be the business of this critique to reduce to a simple estimate of the pure understanding and the preservation of it from sophistical deceits. Transcendental Analytic. This analytic is the resolution into its elements of what a priori cognition in experience holds of the understanding. And here the following points require to be looked to : — 1. The constituent notions must THE KRITIK OF PURE REASON. 181 be pure, and not empirical. 2. They must belong, not to perception proper and sense, but to thought and understanding. 3. They must be elementary and primitive, not compound or derivative. 4. Their table must be complete, so that they shall cover the entire field of pure understanding. But now this completeness, as of a single science, cannot be expected from any mere rough calculation of some aggregate that owes its existence to a venture. It is only pos- sible through the idea, rather, of a whole of the a priori of experience that belongs to the understanding, and a whole, too, duly distributed into its constituent notions ; or, what is the same thing, it is only possible through the connexion of these notions in a system. Pure understanding, then, is, as such, neither em- pirical nor sensuous. Separated thus, it constitutes a self-subsistent and self-complete unity, that is not to be supplemented or improved by any addition from without. The sum of its elements will constitute a system to be comprehended and determined under a single idea, and so that its completeness and articu- lation shall furnish, at the same time, a touchstone of the purity and truth of every article of cognition that is to be fitted and united into it. This part of tran- scendental logic will consist of two books, the one appropriated to the notions of pure understanding, as the other to its judgments. Book I. — The Analytic of Notions. I understand by analytic of notions not the analysis of these, or the usual resort in philosophical inquiries to the resolution and explication of occurrent ideas, 182 TEXT-BOOK TO KANT: but, what hitherto has been little attempted, the re- solution of the faculty itself, in order to discover the possibility of a priori notions, and in this way, that we look for them in the understanding alone as their place of birth, whose pure function we analyze as such. That is what operation is peculiar to a tran- scendental philosophy ; what is usual else is but the logical discussion of notions in philosophy generally. We shall pursue, therefore, pure notions into their first germs and principles in understanding ; in which germs and principles we are to suppose they lie ready waiting, till, at length developed by occasion of ex- perience, and by the same understanding freed from adherent empirical conditions, they stand forward in their perfection. Chapter I. — Of a Clew to all Puke Notions of the Understanding . When we set a faculty into action, conformably to the various occasions various notions appear, which express the faculty, and may be collected in a more or less completed sum, according as the attendant observation has been longer or shorter, closer or slacker. Where, in such, so to speak, mechanical proceeding, any such inquest is to be regarded as complete, is never with certainty determined. Neither do the notions, thus only casually discovered, unravel themselves in any order or systematic unity, but are at last only sorted according to likeness, and, from the simpler to the more complex, ranged according to contents in series which are nothing less than sys- tematic, though brought about by method of a sort. Transcendental philosophy has the advantage, but the obligation too, to find its notions in conformity THE KRITIK OF PURE REASON. 183 with a principle, for this reason, that they must issue from the understanding in its absolute unity, pure and unadulterated ; and so must cohere among them- selves under a one notion or idea. But such sys- tematic articulation offers a rule, in accordance with which there may be a priori assigned to every pure notion its place and to all together their rounded completeness ; and all this would, any other wise, be a matter of chance, or of one's own arbitrary choice. Section 1. Of Understanding in its Logical Function Generally. The understanding has been already merely nega- tively described as a non-sensuous intellectual faculty. Now, apart from sense, we are insusceptible of any perception proper. The understanding, consequently, is no faculty of perception proper. But, perception apart, there is no cognition but that through notions. Cognition of all, more especially human, understand- ing, is, as through notions, not intuitive, but discur- sive. All perceptions, as of sense, rest on affections ; notions, therefore, on functions. But by function I understand that unity of act whereby the various units in a cognition are ordered into a single common one. Notions found, therefore, on the spontaneity (self-action) of thought ; as sense-perceptions on the receptivity of impressions. Notions, now, can be used by understanding only in so far as it judges by them. But no cognition referring directly to its object unless perception, a notion will be necessarily referred to its object only mediately, that is, through some other intimation of it (whether perception or notion). Judg- ment, therefore, is the mediate cognition of an object, and consequently the cognition of a cognition of it. In every judgment there is a notion which, compre- 184 TEXT-BOOK TO KANT: hending several, is applied to a given one ; and this latter is directly referred then to the object. Thus in the j udgment, All bodies are divisible, the notion of divisibility, as referable to several others, is specially applied to one among these, body ; and that, again, to certain actual objects of sense. These objects, therefore, are only mediately cognised through the notion of divisibility. All judgments are, accord- ingly, functions of unity to the variety in a cognition : in the cognition of an object, namel}', there is em- ployed in judgment, not an immediate element (of consciousness), but a higher one comprehending im- mediate elements under it ; and in this manner several possible units of cognition are combined into a single one. But all acts of understanding may be reduced to judgments, and understanding itself, therefore, may be defined a faculty to judge. For, as above shown, it is a faculty to think. Then to think is to cognise through notions. And notions, again, as predicates of possible judgments, conjoin themselves into the conception or perception of some (till then) indefinite object. Thus the notion body refers to something, which something (a metal, say) can be cognised through said notion. Body, then, is only a notion by this, that other elements of cognition are contained under it, through which it gets referred into actual objects. Or it is a predicate to a possible judgment, as that every metal is a body. The func- tions of the understanding, accordingly, will be cap- able of being exhaustively discovered, if we can but exhaustively enumerate the functions of unity in judgments. But that this is very easy of accomplish- ment, the following section will show. THE KRITIK OF PURE REASON. 185 Section 2 (§ 9). Of the Logical Function of Understanding in Judging. If we abstract from all matter of a judgment, and consider only the precise form of the understanding that is manifested in it, we readily find that the functions of thought, in any such, may be reduced to four titles, with three moments under each. This may, not inaptly, be exhibited in the following table : — 1. Quantity of Judgments: Universal, Particular, Singular. 2. Quality : Affirmative, Negative, Infinite. 3. Relation : Categorical, Hypothetical, Disjunc- tive. 4. Modality : Problematic, Assertoric, Apodictic. Inasmuch, now, as this classification seems, in some, though inessential, particulars, to differ from the usual one in technical logic, the following premoni- tory explanations, as against possible misunderstand- ing, may prove not unnecessary. 1. Logicians say rightly that, in a syllogism, the singular proposition may be regarded as a universal one. For just because singulars have no extension, is it impossible that the predicate in such should be partly affirmed and partly denied of the correspondent subject. The former holds good of the latter, there- fore, without exception, just as though this latter were a universal notion to which, in the entire import of its extension, the predicate applied. On the other hand, again, should we compare a singular with a universal proposition merely as a cognition and in regard of its magnitude, then the former stands to the latter as unity to infinitude, and in itself, conse- quently, essentially differs from it. When I consider, 18G TEXT-BOOK TO KANT: therefore, a singular proposition, not merely in its inner validity, but also, as simply a cognition, in the magnitude which it possesses as compared with others, then certainly it is different from universal proposi- tions, and deserves a place of its own in a complete table of the moments of thought as such (though not, naturally, in a logic that is merely addressed to the functions of judgments in their mutual relations). 2. Just in the same way, infinite propositions must, in a transcendental logic, be distinguished from affir- mative ones, though, in general logic, they are rightly reckoned with these, and constitute no special member of distribution. General logic, namely, abstracts from all matter of the predicate (though merely negative), and considers only whether it is attributed to the subject or opposed to it. Transcendental logic, again, considers the judgment in the value or matter of logical affirmation even through a negative predicate, and what gain such affirmation procures cognition as a whole. Suppose I had said of the soul, it is not mortal, I should, by means of a negative judgment, have at least warded off error. But now, logically, I have here really affirmed, seeing that I have placed the soul in the unrestricted sphere of the non-mortal beings. As now, of the whole sphere of possible beings, the mortal occupy one part, and the immortal the other, there is nothing else said in my proposition than that the soul is one of the infinite number of things which still persist when I suppress the mortal. But here the infinite sphere of relative possibility is limited only in so far as what is mortal is removed from it, and the soul placed in the remaining amount of its extension. This amount remains, however, after this removal, still infinite ; and it is still possible to remove other parts of it, without the notion of the THE KRITIK OF PURE REASON. 187 soul being thereby in the least increased and affirma- tively determined. These infinite judgments in re- gard of the logical extension, therefore, are really merely limitative in regard of the matter (compre- hension) of cognition ; and must, so far, not be neglected in a transcendental table of all moments of thought in judgments, inasmuch as the function of understanding here in play may, possibly, be of im- portance in the field of its pure a priori cognition. 3. All the relations of thought in judgments are these : a, of the predicate to the subject ; b, of the .antecedent to the consequent ; c, of a disjunctive cognition and its members mutually. Of these, there are considered, in the first, two notions, in the second, two judgments, and in the third, several judgments relatively the one to the other. The hypothetical proposition, If perfect justice exists, the hardened criminal will be punished, involves properly the rela- tion of two propositions, namely, that perfect justice exists, and that the hardened criminal gets punished. Whether both of these propositions be in themselves true, remains undetermined. What is thought in such a form of judgment is alone the consequence (between the members of it, not the truth of these). Finally the disjunctive judgment considers also a relation of two or more propositions mutually — not that of the consequence, however, but that, rather, of logical contraposition. That is, it considers such pro- positions so far as the sphere of the one excludes the sphere of the other, and yet so that both, or all together, constitute in common the whole sphere of the special cognition in question. The relation in point, therefore, is one that concerns the parts of the sphere of a cognition, where the sphere of the one part is (towards the whole) complementary of the 188 TEXT-BOOK TO KANT: other or others. We say, for example, The world exists either through blind chance, or internal necessity, or an external cause. Now, here, each of these propositions represents a part, and all together the whole, of the sphere of all possible cognition in reference to the existence of the world. To exclude the truth from any one of these spheres is to place it in one of the others ; while to place it in any one of these latter is to exclude it from all the rest. There is, therefore, in a disjunctive judgment a certain com- munity of the terms of the cognition involved. This community consists in the fact that said terms recipro- cally exclude each other, at the same time that they determine the truth as a whole, inasmuch as collec- tively they constitute the entire import of the single given position. And this is what, merely for the sake of the sequel, I find it necessary to remark here. 4. Modality in judgments is quite a special function of these. What distinguishes it is, that it contributes nothing to the matter of the judgment (for besides quantity, quality, and relation, there are no other constituents of the matter of a judgment), but only concerns the value of the copula in relation to thought as such. Problematic judgments are those where we assume the affirmation or negation as merely possible (we may take either as we please). Assertoric are those where we consider the one or the other alternative as actual (true). Apodictic, lastly, are those where the alternative is regarded as neces- sary. 1 Thus the antecedent and consequent of an hypothetical judgment, as well as the members of a disjunctive one, are all problematic. In the above 1 Just as though, in the first case, thought were a function of the understanding ; in the second, of judgment; and in the third, of reason: a remark which remains to be explained in the sequel. — K. THE KRITIK OF PUKE REASON. 189 example, the proposition, A perfect justice exists, is not assertorically spoken, but only thought as a discretionary judgment, which it is only possible that some one may adopt : only the consequence is asser- toric. Hence such judgments may be manifestly false, and yet constitute, -when problematically taken, conditions of knowledge of the truth. In the same way, in the disjunctive judgment, the proposition, The world exists through blind chance, is only of a problematic value. It is possible, namely, that it should be only temporarily assumed ; and yet, in its place, it serves for discovery of the truth (just like indication of the wrong way among those possible). The problematic proposition, therefore, is such as ex- presses only logical (not objective) possibility ; and this possibility amounts only to a free choice in the admission of such a proposition, or to a merely dis- cretionary assumption of it into the understanding. The assertoric proposition expresses only logical actuality or truth. Thus, for example, in an hypo- thetical syllogism, the antecedent, while problematical in the major, is assertoric in the minor. In such proposition it is seen, however, that it is united to the understanding according to its laws. The apodictic proposition thinks an assertoric one as determined by these very laws of the understanding itself, and as a priori, therefore, in the assertion it makes ; it ex- presses in this way logical necessity. Here, now, inasmuch as all is incorporated into the understand- ing in grades — as of something first judged prob- lematic, then assertorically assumed true, and finally affirmed to be inseparably united with the under- standing, or apodictically necessary — we may evi- dently name these three functions of modality quite as well so many moments of thought as such. 190 TEXT-BOOK TO KANT: Section 3 (§ 10). Of the Pure Notions of the Understanding (the Categories). General logic, as frequently said already, abstracts from all matter of knowledge, and looks for percep- tions to be given to it from elsewhere, in order to convert these into notions ; and this process proceeds analytically. Transcendental logic, on the other hand, already has the matter offered it by the tran- scendental aesthetic (the composites, namely, of time and space in a priori sensibility), as a material for the notions a priori in understanding; and without it, plainly, these would be devoid of all contents and, consequently, altogether blank. 1 Or space and time, as conditions of our receptivity, under which alone objects can be received by us (and which conditions, therefore, necessarily affect the notion of an object), possess in themselves a complex or composite of pure a priori perception (or pure a priori objectivity). But the native energy (spontaneity) of our thought demands that this a priori perceptive or objective matter (laid into imagination) should, first of all, be run over, taken up, and conjoined, in order that a cognition (or, so far, a perception) should be made of it. This process (of imagination), now, I term syn- thesis. By synthesis, in its most general sense, I under- stand the uniting of the various units in a conscious- ness the one to the other, and the combining of their complex into a single cognition (perception). Such synthesis is pure when the materials in it are fur- nished for it, not empirically, but a priori (as those 1 To refer the " sie " (as an it) to " transcendental logic " makes a poor sense. This " sie," then, is better referred to the " notions." The " würde " of the text, consequently, is an error for würden. THE KRITIK OF PURE REASON. 191 that are furnished by time and space). Before all analysis of any of our cognitions these cognitions must, first of all, evidently, be given ; and no notions, so far as matter (not form) is concerned, can analyti- cally originate. It is synthesis (let its matter be empirical or a prion) which first makes a cognition (perception) of that matter. Said cognition or per- ception may, of course, in the first instance, be crude and confused, and require analysis ; but it is the syn- thesis which specially collects the units (towards per- ceptions), and unites them all into a single sub- stantiality or object. Synthesis, therefore, is what first claims our attention, when we would inquire into the origin and nature of our cognition of objects. Synthesis as such (this is made clearer again) is the mere act of imagination, a blind, but indispensable, function of the soul, of which, indeed, we are seldom ever once conscious, but without which we should have no cognition at all. But again, now, to bring this first synthesis of imagination under the action of notions, that is a function of understanding; and thereby, first of all, is there realized for us the cog- nition (perception) of experience, in its proper sig- nification. Pure synthesis, quite generally conceived, is to be further understood as implied in, or exemplified by, each of the pure or a priori notions of the understand- ing. 1 I understand by this (pure) synthesis, a syn- 1 The original runs, " Pure synthesis, generally conceived, gives now the pure notion of understanding." The meaning, even in this way, is not too oblique to be understood, if for " gives " we say constitutes, which really is the force of the German. It is quite possible, however, that the "den" should be der, which would reverse the positions of subject and object, but only place them as, evidently, they naturally should be placed. It is so I have translated the sentence, substituting also the 192 TEXT-BOOK TO KANT: thesis that rests on a ground of synthetic unity a priori. Thus our system of arithmetic (as observ- able more especially in larger numbers) is a synthesis on notions, because it depends on a common ground of unity (e.g., the decade). But in such a case, the unity in the synthesis of the constituent complex is neces- sary. Analytically, a variety of objects are brought under a single common notion ; and this is a business which belongs to general logic. But not to bring objects, rather only the pure or a priori synthesis implied in objects, under the scope of notions — this is a process that is treated by transcendental logic. The first element that must be given for the a priori cognition of objects is the multiple or complex of pure percep- tion (pure objectivity — time and space). The second is the synthesis of this complex on the part of ima- gination ; and so far there is not yet a cognition. The third element towards perception of an object on presentation of itself is constituted by the notions which introduce further unity and unities into this pure synthesis, and which consist, indeed, solely in the consciousness of this synthetic unity, or these synthetic unities : these notions belong to the understanding. The same functions which variously give unity to actual plurals which the generalizing singular is really meant to repre- sent. In short, what is to he understood here is this. A category, as a notion, implies a meaning ; and that is a unity of some certain complex or multiple (of relation, say). A multiple so placed or suspended in such a unity may be called a synthesis. And this synthesis, as held by one of the a priori notions (or categories) in the system of such, may very intelligibly be spoken of as " pwe synthesis, quite generally conceived." Each category is such. Or each category is a unity, but a unity neces- sarily of something. That is, each category, in the system of such (and that system is tantamount to the system of functions which constitute self- consciousness), as a concrete, is the intellectual unity of an intellectual multiple. (It is true, too, that pure synthesis purely cognised just is the pure notion : der for "den" involves reine for "reinen"). THE KRITIK OF PURE REASON. 193 the several terms in judgments, extend a various unity also to the mere syntheses of the different units in perceptions. These latter unities, or sources of unity, are the a priori notions of the understanding (the categories). The same functions of understanding, therefore, which, by means of the analytic unity, brought about the logical form of a judgment in notions, do also, by means of the synthetic unity (which they likewise involve), bring about a transcen- dental objectivity (of union) in the complexions of perception. These functions, in this latter applica- tion, may, consequently, be intelligibly named pure notions of the understanding (categories) : they have, intelligibly also, said a priori action on objects; and that, plainly, is not an affair of general logic. l Now, just in this way we may conceive to arise exactly as many pure notions of understanding (with necessary a priori action on the objects of per- ception) as there are logical functions of all possible judgments in the preceding table. For, through said functions, the understanding as understanding is com- pletely exhausted, and its powers as a faculty duly gauged. We call these notions categories, as follow- ing Aristotle, seeing that our intention with them is originally the same as his, however widely different it will be found in the carrying of it out. Table of the Categories. 1. Quantity: Unity, Plurality, Totality. 2. Quality : Reality, Negation, Limitation. 1 This is one of Kant's very worst paragraphs, and I have been obliged considerably to help it. I shall be found elsewhere to agree with Rosenkranz as to the style of Kant, and to defend it against De Quincey. Nevertheless, it is to be admitted in the end that no author writes more contentedly than Kant what simply comes first to hand. Hence his many confused, over- claused, and cross- claused sentences. K 194 TEXT-BOOK TO KANT: 3. Relation : Inherence and Subsistence (Substance and Accident), Causality and Dependence (Cause and Effect), Communion (Reciprocity of Action and Passion). 4. Modality : Possibility — Impossibility, Existence (Actuality) — Non-existence, Necessity — Con- tingency. This, now, is the catalogue of all the primitive pure notions of synthesis which understanding a priori possesses, and only by reason of which, too, it is a pure understanding, seeing that it is by them alone that it can understand something on occasion of a complex of perception, that is, think an object of per- ception (or, simply, perceive). The classification is systematically constructed in obedience to a common principle, namely, the faculty to judge (which just means the faculty to think). It is no product, there- fore, merely rhapsodical, of a search after pure ideas on chance, the completeness of which then can never be relied on ; for, being realized only by induction, it is impossible to understand in that way how pre- cisely these and not other notions should constitute a pure understanding. To ask after such primitive notions was, on the part of Aristotle, an idea worthy of an acute-minded man. As he had no guiding principle, however, he could only pick them up as they came in his way. In this manner he got to- gether at first ten of them, and these he called cate- gories (predicaments). In the end, however, he be- lieved himself to have discovered other five, which were consequently named post-predicaments. Neverthe- less his table still remained defective and incomplete. Thus some of its articles (quando, ubi, situs, prius, simul) are modi of sense, as another (piotus) is empirical, an d THE KRITIK OF PURE REASON. 195 these ought to have no place in a genealogical tree of pure understanding. Others, again, are mere deriva- tives (actio, passio), while of the primitives themselves there are several wanting. In the last reference, it is to be remarked that the categories, as the true root-notions of pure under- standing, have their equally pure derivative notions which, in a complete system of transcendental philo- sophy, cannot by any means be omitted. At present, however, in a mere critical preliminary inquiry, I may content myself with only mentioning them. Let me beg leave to call these pure but derivative notions the Predicables of pure understanding (as in contrast to the predicaments). Once we have the original and primitive notions, the derivative and subordinate ones may be readily added, with the re- sult of completely depicting the whole tree of the pure understanding. As I have to do here, however, not with the completion of the system, but only with the principles towards it, I reserve this for the business of another work. Still as much as this may be pretty well attained, if, with guidance of the ontological text-books, we range under the category of causality the predicables of force, action, passion ; under that of reciprocity, those of presence, resistance ; under that of modality, those of origin, decease, alteration, etc. The categories being combined with each other, or with the modi of pure sense, furnish a large num- ber of a priori derivative notions. To note these, and, if possible, fully specify them, would be a profit- able and pleasant task, but one that may be dispensed with here. I intentionally omit the definitions of these latter (derivative) categories in the present work, though, possibly, in possession of them ; and I shall not con- 196 TEXT-BOOK TO KANT: sider these notions themselves, in the sequel, any further than may be necessary for my theory of method. In a system of pure reason, they would rightly enough be required from me ; but here they would only cause us to lose sight of the chief interest in question, by suggesting doubts and provoking attacks which, without any loss, we might very well reserve for consideration elsewhere. It is clear enough, just from the little I have said, however, that a complete relative vocabulary, with all needful ex- planations, were not only possible, but even easy to effect. The lines are once for all there ; it is only necessary to fill them up ; and a systematic topic like this does not easily allow us either to mistake loci or to overlook those that are still empty. §n. In regard to this table of the categories some nice remarks suggest themselves, which may not be with- out an important bearing on the scientific form of all general interests of reason. For that, in the theo- retical part of philosophy, this table is uncommonly serviceable, nay, indispensable, in assisting us com- pletely to lay out the plan to the whole of a science, so far as it rests on a priori notions, and mathemati- cally distribute it on fixed principles, is already evident of itself. Said table, namely, contains, in complete- ness, all the elementary notions of the understanding, nay, even the form of a system of such in the human mind, and directs us, consequently, to all the moments of any projected speculative science, not omitting its very order ; and of this I have given an example else- where (Metaph. Prins. of Nat. Phil.) Here, now, are a few of these remarks. THE KRITIK OF PURE REASON. 197 1. The four classes in our table may be thrown into two divisions : one directed to objects of perception (no matter whether pure or empirical), and the other to the existence of these objects (so far as they are referred to the understanding, or the one to the other). I would call the classes in the first division mathe- matical, and those in the second dynamical categories. The latter alone have correlates, the former have none; and this difference must, presumably, have its sufficient reason in the nature of the under- standing. 2. Each of the four classes of categories has under it three subclasses; and this gives to think, the rather, indeed, that all other division a priori through notions is necessarily a dichotomy. Again, under each class, the third category owes its origin to the union of the second with the first. Thus totality is nothing else than plurality regarded as unity; limitation is reality in union with negation; reciprocity is substances exchangeably causal ; and neces- sity, lastly, is actuality given, as it were, by possibility itself. For all that, the third category must not be considered derivative only, and not, in reality, primi- tive. In fact, the union in question for the result in question involves a special act of understanding, which is not the same with that exercised in the case of the first and second. For example, the notion of a number (under the category of totality), is not always possible where there are those of plurality and unity (as in the conception of the infinite). Neither from my uniting the notions of substance and cause is it at once possible to understand in- fluence, or how one substance can be cause of some- thing in another (and v. v.) It is clear that a special act 198 TEXT-BOOK TO KANT : of understanding is required in such cases; and so of the rest. 3. In the instance of a single category, that of reciprocity, is its analogy with the correspondent logical form of the disjunctive judgment, not so strik- ing, perhaps, as, similarly, in that of the others. But for conviction here it is to be observed that, in all disjunctive judgments, the sphere (the amount of what is contained under each) 1 is conceived as a whole divided into parts (the sub-notions). And, further, these parts, as not contained the one under the other, are not thought as subordinated the one to the other, but as co-ordinated the one with the other ; and so that they do not affect one another one-sidedly, as in a series, but reciprocally, as in an aggregate : one term being assumed, all the rest are excluded, and v. v. Now, there is a like connexion thought in a icliole of things, where the one is not subordinated as effect to the other as cause (of its existence), but, on the contrary, co-ordinated as again and reciprocally cause (of affections) in precisely this same other's regard {e.g., in what is called a body, where the parts mutually attract, but also mutually exclude each other). And this is quite a different kind of connexion from what obtains in the mere relation of cause and effect (ground and consequence), where the result does not again reciprocally determine the antecedent, and (like the Creator with the creation) does not, therefore, con- stitute with it a whole. What process of understand- ing refers to the sphere of a distributed notion, that same process we observe in thinking a thing as divis- ible ; and, as the members of distribution mutually exclude each other in the former, and yet together constitute a single sphere, so, in the latter, the parts 1 " Each," for " ihm," here remedies a grammatical oversight. THE KRITIK OF PURE REASON. 199 are conceived as such that existence accrues to each (as a substance) in exclusion of the rest, but yet that all are bound together in a single whole. §12. There are to be found, however, in the transcen- dental philosophy of the ancients, certain pure notions of understanding, which are put forward by them as a priori notions of objects. These do not make part of our categories, and, if to be admitted, would increase their number ; which, on our principles, is manifestly impossible. They occur in that well-known proposi- tion of the scholastics, Quodlibet ens est wium, verum, bonum. The use of this principle, indeed, as issuing in mere tautologies, proved so unsatisfactory that, in modern times, any mention of it in metaphysic is pretty well only honorary. Nevertheless, how empty soever, a thought that has persisted so long merits always some inquiry into its origin, as well as justifies the supposition that it has its source in some rule of the understanding, which, as is often the case, has only been wrongly interpreted. These supposed tran- scendental predicates of things are nothing else, in truth, than logical requirements and criteria of all cognition of things in general. In fact, they only subject it to the categories of quantity- — to unity, plurality, and totality. These, however, which ought, properly, to be only materially taken, as concerned with the possibility of things themselves, the ancients applied only in a formal sense as bearing on logical requirement in every cognition, and yet, at the same time, inconsiderately regarded them, though mere criteria of thought, as characteristics of things in their own selves. In every cognition of an object, namely, 200 TEXT-BOOK TO KANT: there is unity of the notion ; and this unity may be named a qualitative unity, in so far as there is thought under it only the unity of the embraced many of units in the cognition, as the unity of plot in a play, speech, story. The second requirement, truth, is truth in regard to the constitutive relations. The more true relations we have as depending on a given notion, the more signs we have of its objective reality. This we may name the qualitative number of characters in- herent in a notion as their common ground (but not thought in it as quantity). The third requisite of completeness applies thus : the many, namely, are con- versely brought back into the unity of the notion, and with this notion, and no other, they must fully coincide. Now this may be termed qualitative com- pleteness (totality). From all this it is evident that these logical criteria of cognition in general apply here the three categories of quantity (in which, as such, the quantitative unity implied must be con- ceived to be thoroughly homogeneous), to connect as well hetei'ogeneous elements in consciousness, and this through the quality of a cognition as principle. Thus the criterion of the possibility of a notion (not of an object) is the definition, in which the unity of the notion, the truth of its consequences, and the complete- ness of its relations, constitute what for reintegration of the whole notion is the requisite on the part of itself. Or it is thus also that the criterion of an hypothesis is constituted by, first, the intelligibleness of the principle adopted in explanation, or its unity (as without supplementary hypotheses) ; second, the truth (agreement with themselves and with experience) of the consequent relations 5 and, third, the complete- ness of the principle adopted in regard of these rela- tions ; which relations contain no more and no less THE KRITIK OF TÜRE REASON. 201 than was assumed in the hypothesis, and only present again a posteriori and analytically what was previously thought a priori and synthetically, at the same time that they are in entire harmony therewith. The notions, consequently, of unity, truth, and complete- ness, do not at all supplement the transcendental table of the categories, as though it were incomplete ; but the application of these latter is (their reference to objects entirely overlooked) brought under general logical rules of the agreement of a cognition with its own self. Chapter II. — Deduction of the Categokjes. Section 1 (§ 13). Principles of a Transcendental Deduction in General. Writers on jurisprudence, when discussing rights and their violations, distinguish, in an action at law, the question of law (quid juris) from the question of fact (quid facti) ; and, in requiring proof in both respects, they name the former, which is to make good the title (the right), the deduction. We com- monly employ a number of empirical notions, with- out any one thinking to question them, and assume ourselves authorized, even without deduction, to impute to them a certain meaning, because we have always experience to fall back upon in proof of their objective reality. There are also usurped notions, as Fortune, Fate, which pass current with almost uni- versal assent, but are at times called upon for an answer to the quid juris ; and then the deduction of them proves a matter of no small difficulty, for neither from experience nor reason can any clear title be produced for them. But among the many notions which constitute the 202 TEXT-BOOK TO KANT: very mingled web of human cognition, there are some which are destined to serve a purely a priori purpose (entirely independent of all experience) ; and these always require for their title a deduction. For proofs from experience are incompetent in such a case, and yet we would understand how these notions can enter into and refer themselves to objects, for the idea of which objects they (these notions) owe nothing to experience. I call the explanation, then, of how a priori notions can have this application to objects of experience the transcendental deduction ; and distinguish it from the empirical deduction which for the origin of an idea appeals to sensation and reflec- tion, and, in this way, involves not the right of use, but the fact of existence. We have now found two quite diverse elements, which, however, agree in being both a priori consti- tuents of objects of experience ; namely, on the one hand, space and time as forms of sense, and, on the other, the categories as forms of intellect. To require an empirical deduction of these would be a futile want ; for what is distinctive of their nature lies pre- cisely in this, that they connect themselves with objects without owing anything to experience for the idea of these objects. If, then, a deduction at all is required for these, that deduction, plainly, must be always transcendental. At the same time, in the case of these notions, as in that of all cognitions, we can rightly enough inquire, not for the principle of their possibility, but for the occasions of their appearance in experience. It is certainly the impressions of the senses which give the first stir to the production of experience, and the movement of cognition in every reference. Still, experience, or cognition generally, includes in THE KRITIK OF PURE REASON. 203 itself two very dissimilar factors, namely, a matter derived from the senses (sensation), and a certain form (for the ordering and arranging of this matter) which is due to the inner source of understanding and pure perception. Now, it is on occasion of the former element (sensation) that the latter faculties of form are moved to bring forward and introduce their a priori contributions. An inquiry into the earliest struggles of our faculties in order to ascend from particular perceptions to general ideas, has un- doubtedly its own great use ; and we have to thank the illustrious Locke for having first opened the way to this. But then a deduction of what is a priori can never possibly be brought about in that way. What is a priori lies quite in another region ; and must produce, for its license of use in the future (inquiries beyond limit of experience), quite another certificate of birth than that furnished by the senses. The at- tempted physiological derivation, therefore, which concerns only a question of fact, and can never be properly called deduction, I shall denominate the proof of our possession of elements, which elements may still be a priori. But it is clear that of such elements it is a transcendental, and not an empirical, deduction that is required. The latter, indeed, in an a priori reference, must prove always a vain attempt, on which only he will venture who completely mistakes the quite peculiar nature of the interest in hand. Although, however, it be granted that the only possible deduction of what is a priori must be tran- scendental, it does not immediately follow therefrom that such deduction is unavoidably necessary. We have now, by means of a transcendental deduction, traced space and time to their sources, and we have demonstrated and made good their a priori objective 204 TEXT-BOOK TO KANT: validity. Nevertheless, geometry goes its own sure way with mere a priori elements, without any call to exact of philosophy a letter of credit in respect of the pure and legitimate origin of its basal notion, space. But the use of that notion in this science applies plainly to the external world of sense, of the perception of which space is the pure form, and in which, therefore, every geometrical cognition, as founding on a priori perception, has its immediate evidence. The cogni- tion itself here, in fact, assumes its objects as (in form) a priori given in perception. On the other hand, there arises with the categories an absolute necessity to call for their transcendental deduction ; and not for theirs only, but for that of space also. The cate- gories, namely, do not in themselves act upon objects through predicates of perception and sense, but of pure a priori intellect : they refer to objects generally without all conditions of sense. They neither bring their title from experience, nor can they foreshadow in a priori perception objects on which before all expe- rience they might found their synthesis. The conse- quence is that they raise doubts not only as regards the objective validity and limits of their own use, but even make equivocal the notion of space. For space, objectively, is applied by them beyond the powers of sensuous perception. Hence the necessity of a transcendental deduction, as above, for space as well. 1 The reader, then, must convince himself of the indispensable necessity of such transcendental deduction, before he has taken a single step in the 1 The above conveys one of Kant's long and involved sentences. The reasoning is that, though geometry is of an a priori nature and yet calls for no deduction, the case is not the same with the categories. They require to be deduced, and not only they, but the space which they use, and must use, for any possible application on their part to objects at all, pure or empirical. It is the introduction of space into the sentence THE KRITIK OF PU11E REASON. 205 field of pure reason. Otherwise he will proceed only blindly, and will find himself, after many wander- ings, obliged to return to the ignorance from which he had set out. He must, however, make the inevi- table difficulties clear to himself beforehand ; in order that he may not object obscurity when it is the matter is deep, or be too soon disheartened when hindrances obstruct. For it comes to this, either to give up all claim to discoveries of pure reason, and in that her dearest field beyond the bounds of all possible expe- rience, or else to complete the present critical inquiry. We have, with little difficulty, made intelligible above how space and time, though cognitions a priori, join themselves, nevertheless, necessarily to objects, and render, in independence of all experience, a syn- thetic cognition or perception of objects possible. For, inasmuch as only through such pure forms of sense an object can appear to us (that is, before it can be an object of empirical perception), space and time are pure perceptive forms (a priori objects), which are a priori conditions of objects of experience, and synthesis in them is of an objective validity. The categories of understanding, on the other hand, have nothing to do with conditions of perception (in the strict sense), and there certainly may very well be presentations of objects so far as sense is concerned, without there being any necessity to refer them to functions of the understanding at all. Understanding, so far, need not involve, in formation of objects, any a priori influence whatever. In this relation, indeed, particularly, which, as it were, throws all the measures across. I am disposed, also, to see verbal errors here : " redet " should be reden, and I would even altogether expunge the "die" after the "und." Kant's thought is, Categories, not being, directly or properly, perceptive, show a difficulty and a need of deduction (the latter, too, for the space they use) not shown at first hand by geometry, etc. 206 TEXT-BOOK TO KANT: there shows a difficulty which we did not find when employed on sense. How, namely, can subjective conditions of thought conceivably at all exert an objective function — that is, how can they furnish con- ditions of the very possibility of all perception and experience of objects? For, surely, presentations may be given in sense, pure or empirical, without calling in any function of the understanding. I take, for example, the notion of cause, which implies a par- ticular sort of synthesis, where on something, A, there ensues, by necessity of a law, a something else, B, that is quite different from A. Now, it is useless to refer to experience in proof of any such notion, which, as containing necessity, can be proved objectively valid only a priori It is, in the first instance, difficult to understand, then, how sense-presentations should ex- hibit any such virtue ; and we may at first very much doubt whether any such claim, any such idea, be not altogether void, and without correspondent object anywhere among the presentations in sense. For that objects of sensuous perception must obey what formal conditions of sense in general lie a priori in the mind, is clear from this, that, on other terms, they would not be objects for us. It is not so easy to see, however, that they must also obey conditions required by the understanding for synthetic perception on the part of the intellect. For presentations in sense might very well be of such a nature that the understanding would not find them at all accordant to the conditions of its unity, but, on the contrary, all in such confusion, that, for example, in the succession of presentations there were nothing to be found capable of affording a rule of synthesis, and correspondent, therefore, to the notion of cause and effect ; which notion, consequently, were null and void, and without sense. Presentations THE KRITIK OF PURE REASON. 207 in sense would not the less for that furnish us with objects so far perceptive; for perception, so far as strictly sensuous, stands nowise in need of the func- tions of the intellect. Did we think to rid ourselves of the difficulty of such inquiries by saying, Experience affords continual examples of such submission to law in the objects of sense, which examples furnish abundant occasion for abstracting the notion, cause, and thereby ratifying at the same time the objective validity of such a notion, then we forget to observe that the notion, cause, cannot arise in this way, but that it must either be based completely a priori in the understanding, or else utterly abandoned as a mere chimera. For this notion demands absolutely that something, A, be of such a nature that something else, B, ensues from it, necessarily, and by virtue of an unconditionally universal law. Sense certainly, however, gives examples from which we may infer a rule of what usually happens, but never of what necessarily happens. Hence there belongs to the synthesis of cause and effect a dignity which can never be empirically expressed ; namely, that the effect not merely comes after the cause, but is given by it, and ensues from it. The rigorous universality of the rule, too, is not at all a possession of empirical rules which, as through induction, can have no more than comparative universality, that is, a certain extended application. The validity of the categories would be completely changed, then, were we to regard them as merely empirical products. § 14. Transition to the Transcendental Deduction of the Categories. There are only two cases possible, in which syn- thetic perception and its objects can coincide and 208 TEXT-BOOK TO KANT : necessarily refer to one another. Either the object makes the perception, or the perception the object, alone possible. In the first case the circumstances are only empirical, and the perception is not possibly a priori. This case is that of presentation in sense, and, specially, of what belongs to sensation in them. In the second case, again, no mere mental act (for there is no question here of the causality of will) being competent to produce an existent object, a per- ception can only then be a priori operative in regard of an object, when through it alone it is possible for us to perceive something as an object (cognise it as a factor in actual experience). Only under two condi- tions, however, is such cognition of an object possible. There is, first, perception proper, by which the object is given, but, so far, only as intimation to sense. There is, second, notion, by which, in correspondence with the elements of sense, an object is fairly thinkingly perceived in experience. It is clear, however, from what has been said further back, that the first condi- tion, that, namely, under which alone objects can be (taking the word strictly) perceived, must, in effect, be presupposed for, and basally underlie, objects, so far as form is concerned, a priori in the mind. 1 With this condition of sense, therefore, all objects neces- sarily agree ; for only through it is it possible for them to show in sense, or to be empirically given and perceived. Now the question is, whether there are not also precedent notions a priori, as conditions under which alone anything is, though not sensu- ously, yet cognitively, perceived as an object (as such) ; for, in that case, all empirical cognition of objects (formed perception of them as in experience) is necessarily subjected to, or in conformity with, such 1 After " liegen " I add a muss Lere, which seems necessary. TUE KRITIK OF PURE REASON. 209 notions, inasmuch as without presupposition of them nothing whatever is possible as an object of experience. But, now, every perceptive experience involves, be- sides the elements of sense, by which something is given, still further a notion of an object, or a notion uniting the sense-elements into a one object, which one object appears then as given in cognitive or formed perception. On that understanding, conse- quently, there are notions, bearing on objects as objects, a priori presupposed as conditions that basally underlie all cognition or perception of experience. And the objective application of the categories, there- fore, as a priori notions, is based on this, that through them alone is experience (so far as relates to the form of thought — the involved function of intellect) at all possible. For then they have a necessary and a priori bearing on objects of experience, inasmuch as only through them as universal condition can any object whatever of experience be cognitively per- ceived. The transcendental deduction of all a priori ele- ments has, therefore, a principle directive of the whole inquiry, this, namely, that they must be re- cognised to be a priori conditions of the possibility of experience (whether as of sense or of understand- ing). Elements which furnish the objective ground of the possibility of experience are for that very reason necessary. An analysis, however, of the experi- ences in which they occur, would not constitute their deduction, but, as in that way they would still remain contingent, only their illustration. 1 Without 1 In the above three sentences I translate as though a plural Erfah- rungen and a singular Erfahrung had changed places as regards the first and last of them. The plural in the first sentence seems proved wrong by a singular "ihr" which follows and, as I think, in its reference. O 210 TEXT-BOOK TO KANT : this primary reference to possible experience, which holds of all objects in perception, the application of these a priori elements to any object could not be possibly understood. The celebrated Locke, for want of such considera- tion, and because he found in experience pure notions of the understanding, actually derived such notions from experience, and proceeded so inconsequently that, simply in trust of them, he ventured on cogni- tions that far transcended all limits of experience. David Hume saw that, for this to be possible, it was necessary that said notions should be possessed of an a priori origin. As, however, he could not at all explain to himself how the understanding, in the case of elements of experience that were not to it in them- selves connected, did yet feel forced to think them necessarily connected in objective experience ; and as, in view of this inability, it did not occur to him to reflect that perhaps the understanding was itself, and through these very elements, the actual originator of that precise experience in which the objects showed — in such circumstances he believed himself under a necessity to derive them only from experience. He attributed them, namely, to a subjective necessity due to repeated associations in experience (custom), which subjective necessity was, in this way, at last, taken to be objective, but erroneously so. His resultant con- clusion, however, was perfectly consequent : he de- clared it to be impossible to transcend experience with any such principles. The empirical derivation, never- theless, which is all we have in either, cannot possibly be reconciled with the actuality of those scientific a priori cognitions which we have in pure mathematics and pure physics, and is, therefore, refuted by the factum. THE KRITIK OF PURE REASON. 211 The first of these celebrated men opened a wide door to fanaticism ; for reason, if once with any title on its side, is no longer to be kept in bounds by any mere vague exaltations of prudence and moderation. The second, again, quite abandoned himself to scep- ticism, in the conviction that he had once for all dis- covered, on the part of our faculties, a mere deception that was universally held to be reason. We are now on the threshold of the inquiry whether it be not possible to steer human reason between both rocks, assign it its definite limits, and yet keep open for it the entire field of its legitimate action. I begin with the definition of the categories. They are notions of objects generally, by which the sense- elements of these objects are conceived to be deter- mined in respect of one or more of the various logical functions of judgment. Thus the function of the categorical judgment is that of the relation of the subject to the predicate, as, All bodies are divisible. But here, so far as concerns the mere logical use of the understanding, it remains undetermined to which of the two notions the function of subject, and to which that of predicate, accrues. For we might quite as well have said, Certain divisibles are bodies. But by the category of substance, now, it is deter- mined of the notion body, when subjected to it, that its empirical perception in experience must be always regarded only as subject, and never as predicate. And so of all the other categories. 212 TEXT-BOOK TO KANT: Section 2. Transcendental Deduction of the Categories. § 15. Of the Possibility of a Conjunction in General. The constitutive units may be given in a perception which is merely sensuous, or nothing but receptivity, and the form of this perception may lie a priori in our faculty, without being anything else, however, than how the subject is passively affected. But the conjunction of these or any units is not possibly an affair of sense, and can, therefore, not be found as an element or action involved even in the pure form of sense-perception (space, etc.) For it is an actus of the mind's own faculty, and as in contradistinction to sense we must name this faculty understanding, it follows that all conjunction, conscious or unconscious, in perceptions or in notions, in elements pure or in elements empirical, is au act of the understanding to which we would give the general appellation of synthesis} AVe use this term, namely, to signalize the fact, as well that we cannot be aware of any- thing conjoined in the object which we have not previously conjoined in understanding, as of conjunc- tion being, among all perceptions, the only one which cannot be given by objects, but must be effected only by the subject's own self, because it is an actus of self-action or spontaneity. It is easy to be under- stood here that this actus must be originally monome (strictly one), and of force for all conjunction, as also that the resolution {analysis) which seems to be opposed to it, does yet, for all that, always presup- pose it; for where understanding has not already 1 The words " sinnlichen oder nicht sinnlichen Anschauung," if cer- tainly Kant's, must be held to have meant for him, in the one case, special, and, in the other, pure sense. The latter of them is specially mis- leading, as literally contradictory of the most current expressions. THE KRITIK OF PUKE REASON. 213 conjoined, neither can it disjoin, inasmuch as only through it can anything, as conjoined, be offered to our perception. But the notion of conjunction carries with it, besides those of the complex of sense-units and their synthesis, that of their unity as well. Conjunction is synthetic unity of a complex. 1 The cognition of this unity can, therefore, not arise from the conjunction ; rather, by adding itself to the cognition of the com- plex, it first makes the notion itself of conjunction possible. This unity, which precedes a priori all notions of conjunction, is not possibly said category of unity (§ 10) ; for all categories found on logical functions of judgments, in which conjunction is already thought, and, consequently, unity of given notions. The categories, therefore, already (in their own selves) presuppose conjunction. We must, therefore, seek this unity (as qualitative, § 12) still further back ; we must seek it, namely, in what is the ground of that unity in the judgments themselves, or in what, consequently, is the ground of the under- standing itself in its very logical function. § 16. Of the Original or Primary Synthetic Unity of Apperception. The I think must be capable of accompanying all my perceptions ; for otherwise there would be some- thing placed in my consciousness which could not be thought ; and that is as much as to say that the per- ception itself would either be impossible or else 1 Whether the cognitions themselves are identical, and the one, there- fore, can he thought, analytically, through the other, is not in considera- tion here. The consciousness of the one, so far as the complex is in ques- tion, is always to be distinguished from that of the other, and it is only the synthesis of this latter (possible) consciousness that is here concerned. — K. 214 TEXT-BOOK TO KANT: nothing for me. That form of apprehension which precedes thinking is called perception. All the units, therefore, of a perceptive complex is necessarily con- joined with the / think of the subject holding them. This, however, is an act of spontaneity, and cannot be thought as due to sense. I call this the Pure Apper- ception, to distinguish it from the empirical. It may be named also the Original (Primary) Apperception, inasmuch as it is that self-consciousness which, while it produces the all-attendant and ever-identical con- sciousness / think, cannot be accompanied by any further one. I call also the unity in it the transcen- dental unity of self-consciousness, in consideration (or indication) of its being a source of possible cognition a priori. For the units in any perception would not be collectively my perceptive units, did they not collectively belong to a single self-conscious- ness. Or these units as mine (though consciousness may not be specially awake to them as such) must be necessarily submitted to the condition under which alone it is possible for them to stand together in a single self-consciousness, for otherwise they would not, one with the other, belong to me. Now from this original synthesis there follows much. This pervading identity of apperception, namely, throughout the units of a perception, necessarily implies a synthesis of these, and is possible, at the same time, only through consciousness of this synthesis. For what empirical consciousness accom- panies our bare sense-intimations or sense-feelings is naturally loose, and, as regards the identity of the subject, inconsiderate or inadvertent. Synthesis into this identity, indeed, does not straightway result from this, that I accompany each of the units with consciousness : I must, further, actually add the units THE KRITIK OF PURE REASON. 215 the one to the other, and become conscious of this, their resultant synthesis. Only by this, therefore, that I can conjoin the units of given intimations in a single consciousness, is it possible for me to conceive the identity of consciousness in these intimations them- selves. The analytic unity of apperception is only possible under presupposition of a certain synthetic one. 1 The thought, These units given in perception are collectively mine, is, accordingly, as much as to say, I unite them, or at least can unite them, in a single consciousness ; and though this thought is not yet itself the consciousness of the synthesis of the units, it yet presupposes the possibility of this. That is, only by comprehending the complex of units in a single consciousness, do I make them singly and col- lectively mine. Otherwise I should have as many- coloured and diverse a self as I have units in con- sciousness. Synthetic unity of the complex of per- ceptions as given a priori, is the ground, therefore, of that identity of apperception itself which a priori precedes any definite act of thinking on my part. Synthesis, however, is not in the objects, and cannot possibly be borrowed from them, or only first of all taken up into consciousness, through perception : it is 1 The analytic unity of consciousness attaches to all common notions as such. For example, when I think red, I think something which may be found as element in something else, that is, something which is con- joined with others. Only, therefore, through a preconceived (a pre- viously thought) possible synthetic unity can I conceive the analytic one. A unit of consciousness, which is to be thought as common to several, is regarded as belonging to such as, besides it, have something else in them ; it must, consequently, be already thought in synthetic unity with others (though these may be only possible), before I can think by it of the analytic unity of consciousness which makes a conceptus communis of it. And thus the synthetic unity of apperception is the ultimate point on which we must base all intellect, even the whole of logic, and with logic, transcendental philosophy ; nay, apperception is the under- standing, the intellect itself. — K. 216 TEXT-BOOK TO KANT: an act of understanding alone, which itself, indeed, is nothing but the faculty whose single function it is, a 'priori to conjoin, and to bring the complex of given perceptions under the unity of apperception. This principle is the ultimate principle in all human cognition. This proposition, of the necessary unity of apper- ception, is certainly, now, itself, identical, and there- fore analytic. Nevertheless it involves a synthesis of any given perceptive complex as necessary. And without this synthesis, said pervading identity of self-consciousness cannot possibly be thought. For in Ego, as a simple thought, there is nothing of a complex given. In the empirical feeling, which is different from the mere thought, is it alone that the relative complex can be given, and through synthesis in a single consciousness, as it were, perceived. An understanding in which the involved complex were at once given by self-consciousness, would perceive, would be intuitive. Our understanding can think only, and must have recourse to the senses for per- ceptive or intuitive matter. I am conscious, then, of my identical self in regard of the units in any given perception, because I name these units singly and collectively mine ; and collectively they constitute a single perception. That, however, is as much as to say that I am conscious of a necessary synthesis a priori, which is called the original synthetic unity of apperception : that all units of perception given me stand individually under it, but that they must be brought collectively, as well, or through a synthesis, under it. THE KRITIK OF PURE REASON. 217 § 17. The Axiom of the Synthetic Unity of Apperception is the Ultimate Principle of the Understanding. The ultimate principle of the possibility of all perception in relation to sense, was, according to the transcendental a3sthetic, this, That the units of every such complex must stand under the formal conditions of space and time. The ultimate principle of the pos- sibility of all perception, in relation to the understanding, is, That the units of every perceptive complex must stand under conditions of the original-synthetic unity of apperception. 1 All units of perception stand under the former, so far as they are given to us ; and under the latter, so far as they must be capable of being conjoined in a single consciousness. For without such conjunction there would be nothing thinkingly cognised or recognised (as in experience), inasmuch as the units given by sense would not have the actus of apperception, / think, in common, and would not be brought together thereby into a single consciousness. Understanding is, to speak generally, the faculty of perceptive cognitions (or recognitions). These consist in the definite conjunction of given units — their con- junct reference — into an object. Object, again, is that in the notion of which the units of the percep- tive complex are united, are made one. But all such union demands unity of consciousness in its very synthesis. The unity of consciousness, therefore, is 1 Space and time, and all the parts of either, are perceptions, and one, consequently, with the many of their constituent complex (s. Trans. iEsth.) They are not mere notions, then, which, as such, are each a one and the same consciousness in a plurality of individuals. On the contrary, they are, each of them, a plurality of parts in a single consciousness (that, in fact, of them). These parts, then, heing set together in that consciousness, there is plainly involved a unity of consciousness which is synthetic and also original or primary. This singleness of conscious- ness is important in application (s. § 25.) — K. 218 TEXT-BOOK TO KANT: alone what constitutes this conjunction of units into an object, or the objective realization of these, or the fact that they are objectively perceived in experi- ence. It is on the unity of consciousness, conse- quently, that the possibility of the understanding itself rests. The first pure fact of understanding, therefore (towards formed perception), which conditions its whole further action, and which, at the same time, moreover, is apart from any condition of sense- perception, is the principle of the original synthetic unity of apperception. Thus the mere form of external sense-perception, space, is not yet a finished perception : so far, it only supplies the a priori per- ceptive complex towards a possible finished percep- tion. But actually to discern something in space, a line, I must draw it. That is, I must synthetically effect a certain particular conjunction of the space- units (as yet only given), and in such manner that the unity of this act is at the same time the unity of consciousness (in the idea of a line.) Only in this way, plainly, is it first of all possible for an object (a marked off space) to be discerned. The synthetic unity of consciousness, therefore, is an objective con- dition of all formed or finished perception in experi- ence. Not only is it necessary to enable me to perceive an object ; but just to be object every sense- perception must stand under it. In any other way, or without this synthesis, the units of the perceptive complex would not unite themselves together in a single consciousness. This last proposition is, as said, at the same time analytic, even while source of the synthetic unity as condition of all possible intellection. For it says only that all my units of perception in any given case TUE KRITIK OF PURE REASON. 219 must stand under the condition under which alone it is possible for me to reckon them as mine into my identical self, and therefore to comprehend them, through the all-present expression, / think, as syn- thetically conjoined in a single apperception. This principle, however, is not such as to cohere with every possible understanding as such. It is proper only where the I am of pure apperception brings with it no complex, or breadth, as of units in consciousness. An understanding, where self- consciousness brought at once its own constituent discernible many, or an understanding, in whose subjective act its objects at once were, — such under- standing would not stand in any need of a special act of synthesis in a complex in order to attain to a unity of consciousness. Such special act is the necessity of the human understanding which, as an understanding, thinks merely and does not perceive. 1 For the human understanding, however, this act is the indispensable first principle. So much is this the case that our understanding, in fact, is unable to form to itself the least idea of any other possible understanding, whether such as were itself perceptive, or such as, though only perceptive through sense, were yet otherwise perceptive than through forms of space and time. § 18. What Objective Unity of Self-Consciousness is. The transcendental unity of apperception is that unity through which all the complex units given in a perception are united into a notion of the object 1 In the above, there is an awkward use of both cases at once which bedürfen may govern. In these, too, the genitive " deren" really refers to the accusative " Actus," and, consequently, were better dessen. 220 TEXT-BOOK TO KANT: constituted by them. For that reason this unity is called objective, and must be distinguished from the subjective unity of consciousness. This latter is only the inner aifection of sense whereby a perceptive complex is (for such union) empirically given. Whether I shall be empirically conscious of the units in the complex as given together, or as given the one in succession to the other, depends on circumstances, or empirical conditions. Hence the empirical unity of consciousness (through association of the units) is itself a sense-appearance, and quite contingent. On the other hand, the pure form of perception in time, merely as such perception, and involving, conse- quently, a given complex of units, stands under the original unity of consciousness, solely in consequence of the necessary conjunction of the units of percep- tion into the one single / think (or, it is I that am thinking). That is, it so stands, solely in conse- quence of the pure synthesis of understanding, which synthesis (as relating only to an a priori complex), is evidently presupposed a priori to underlie any empirical synthesis. The former unity is alone objectively valid (the same, and in the same way binding, for every one). The empirical unity of apperception (which is not considered here, but which, under given circumstances in concreto, only results from the other), has merely a subjective validity. One man, for example, unites the hearing of a certain word with one thing, and another with another. Thus, evidently, the unity of consciousness, when bearing on what is empirical, is not necessarily and universally valid in regard of anything so given. THE KRITIK OF PURE REASON. 221 § 19. The Logical Form of all Judgments consists in the Objective Unity of the Notions they contain. I have never been able to feel satisfied with the definition which logicians give of a judgment. It is, they say, the statement of a relation between two notions. Now here, without objecting to the defini- tion that it applies only to categorical, and not to hypothetical or disjunctive propositions (where the relation is not between notions but judgments), I remark only (not but that the oversight has been detrimental to logic), that, in the definition, there is no declaration of what the relation consists in. 1 If, however, I more accurately examine the nexus of the given ideas in every judgment, and distinguish it as due to the understanding, from the relation that depends on laws of reproductive imagination (which relation possesses only a subjective validity), I find that a judgment is nothing else than the method of bringing given ideas into the objective unity of apperception. It is the very business of the little word is in them to distinguish the objective unity of given ideas from the subjective one. For this word designates the reduction of these ideas into the original apperception, and their necessary unity, even although the judgment be only an empirical one and, conse- quently, contingent; as, for example, Bodies are heavy. Here, I do not mean to say that these ideas belong in the empirical perception necessarily the one 1 The whole long doctrine of the four syllogistic figures concerns only categoricals, and though it is nothing more than an art, through concealment of immediate inferences (consequentice immediate) under the premises of a pure syllogism, to produce the false show of more kinds of syllogisms than are contained in the first figure, it would not have par- ticularly succeeded in this, had it not contrived to call attention exclu- sively to categoricals as those propositions to which all the others must be reduced ; which, however (§ 9), is not the case. — K. 222 TEXT-BOOK TO KANT : to the other, but that they belong the one to the other by virtue of the necessary unity of apperception in the synthesis of sense-perceptions. That is, they belong the one to the other according to principles of the objective determination of all cognitive ele- ments, so far as they are competent to yield an objective perception ; which principles derive all of them from that of the transcendental unity of apper- ception. So only there is made of said relation a judgment — a relation that is objectively realized, and easily distinguishable from the relation of the very same elements when it is only subjective, as being referred, for example, only to the laws of association. In the latter reference, I should only be able to say, If I lift a body, I feel a sense of weight ; but not, The body is heavy. This last proposition imports that the two ideas or elements are conjoined in the object (without consideration of the mere state of me, the subject), and not simply beside each other in the sense- affection, let it be repeated as often as it may. § 20. All Perceptions of Sense stand under the Categories, as Con- ditions under which alone the Units of their Complex can unite together and coalesce into a single Consciousness. The complex of units given in a perception of sense, falls necessarily under the original synthetic unity of apperception, inasmuch as through this unity alone is the unity of the perception possible (§ 17). But that act of the understanding through which the units of a complex (whether perceptive or notional) becomes reduced into a single apperception, is the logical function of the technical judgments (§ 19). Every complex, therefore, so far as it is given in a single empirical perception, has been determined by one of TUE KRITIK OF PURE REASON. 223 the logical functions of judgment, or by this function of judgment it has been brought into a single con- sciousness. But now the categories are nothing else than precisely these functions to judge, so far as some given complex of perception comes to be determined of them (§ 13). Hence all given perceptive complexions stand necessarily under categories. § 21. Remark. A complex, which is contained in a perception that I call mine, gets reduced into the necessary unity of self-consciousness through synthesis of the under- standing, by the categories. 1 This synthesis implies that the empirical consciousness concerned with a given complex of a one perception, stands as well under a pure a priori self-consciousness, as empirical perception stands under one of pure sense which is likewise a priori. In the above proposition (§ 20), then, is there a commencement made of a deduction of the categories. In this deduction, said categories being independent of sense and wholly in the under- standing, I must still as yet abstract from the manner in which the perceptive complex gets given, in order to direct attention only to the unity which the under- standing by means of its categories produces in the perception. In the sequel (§ 26), it will be shown, from the mode in which the empirical perception is given in sense, that the unity in this perception is no other than what (§ 20) the categories determine in a given perceptive complex. And by this, then, that 1 The proof rests on the assumed unity of perception through which there is an object. Said unity always implies synthesis of the perceptive many, and the reduction of this latter into the unity of apperception. — K. 224 TEXT-BOOK TO KANT : the a priori application of the categories is demon- strated as in regard of all the objects of our senses, will the design of our deduction be at length fully realized. But in the above proof I could not abstract from one element ; from this, namely, that the complex must be given for the perception before, and in inde- pendence of, the synthesis of the understanding ; but in what manner we do not here pronounce. For, did I assume an understanding which (as such) perceived (as a divine understanding which, possibly, produces its objects in its perception, and has them not given only to it), the categories would be without applica- tion in regard of such cognition. They are only rules for an understanding whose whole faculty (as an un- derstanding) consists in thinking; that is, in the process to bring into the unity of apperception, the synthesis of what perceptive complex may be given it from elsewhere. Such an understanding of itself perceives nothing whatever; but only conjoins and orderly disposes the matter, the sense-elements, of perception, which are given it by the object. Of the peculiarity of our understanding, however, only through categories, and only through that kind and number of them, to bring about a priori unity of apperception, we can as little assign a reason as we can explain why we have precisely these and no other functions of judgment, or why time and space are the sole forms of our possible perception. § 22. The Categories have no other Application in Cognition than to Objects of Experience. To think an object, and to perceive an object, are not one and the same thing. There are, namely, in THE KRITIK OF PURE REASON. 225 perception two factors. There is, first, the notion (category) whereby an object is thinkingly perceived ; and there is, second, the sense-elements whereby it is given. For if to the notion no corresponding sense- presentation could be given, the former would be only a formal thought without an object, and, conse- quently, not possibly capable of affording perceptive recognition of anything whatever. There might be, indeed, so far as I could know, not anything, not even possibly anything, whereto my thought might apply. Now, all perception possible for us is sensuous (^Es- thetic); the thinking of an object, therefore, by means of a category, can only become for us a perceptive recognition in so far as this category is brought to bear on objects of the senses. Perception of sense is either pure (space and time), or empirical (what is perceived, through sensation, as directly actual in space and time). Through determination of the former, we obtain (in matheinatic) a priori percep- tions of objects, but only in their form as presenta- tions to sense ; whether there are possibly also actual things which are to be perceived in such form, remains, so far, still undetermined. Consequently no mathematical notion is in itself perception, unless there be presupposed things, also, which are capable of being realized by us only as in accordance with the form of said pure sensuous perception. Tilings in '"> space and time, however, are only realized by us through empirical perception, or so far as they are sense-perceptions, perceptions accompanied by sensa- tion. The categories, consequently, even in applica- tion to a priori perceptions (as in matheinatic), afford perceptive cognition or recognition, strictly, only so far as these a priori perceptions, and consequently also through them the categories, are capable of being p ~2'26 TEXT-BOOK TO KANT : applied to empirical objects. The categories, con- sequently, even with pure perception, yield us no knowledge of things, unless as in reference to their possible application empirically ; that is, they serve only for the possibility of empirical perception and cognition. But that is experience. Consequently the categories have a share in the cognising of things, only in so far as these things actually are, or are taken to be, empirical. §•23. The above proposition is of the greatest importance; for it just as much determines the limits of the share of the categories in objects, as the .Esthetic similarly determined in regard to the pure form of our sense- perception. Space and time function, as conditions of the possibility of how objects can be given us, no further than as regards objects of sense, or no further than as regards experience. Beyond these limits they stand for nothing; for they are only in the senses and have no reality apart from them. The categories are free from this restriction, and apply to objects of perception as such, if only sensuous and not intellectual, let it be like to ours or not like. This extension beyond our sense helps us, however, as on their part, to nothing. For they are then void notions of objects, of which objects, whether they are even possible or impossible, these notions themselves cannot possibly enable us to judge. They are mere thought-forms without objective reality, for we are without a perception on which to apply the syn- thetic unity of apperception involved in them, and so render them themselves available in the relative objective determination. Our sensuous and empirical THE KRITIK OF FÜRE REA- meaning. 1 Let : ; suppose ourselves to assume, for example, an object that is an object of a perception which is Such an object we may determine, of course, by all the predicates which the assumption itself involves — the assumption thai it has nothing of a sense-perception in iL It is not, therefore, extended or in space ; its duration is not a time ; there is no time- succession of modi, no such thing as change, in it, But that is not an objective cognition proper, in rd to which I only name how the perception of the object is not, and remain unable to say anything that it posh: I have not then done anything to indicate the possibility of an object for my cate- r I have not been able to assign a perception which should correspond to it. I have only been able I »y rather, that our perception is, as regards h object, without a bearitj he most important distinction here :hat to any such supposed object, there cannot be applied even any on _ category. How apply that of sub- stance, for instance, or the notion of something, such that it can e: subject, but never as mere predi- cate, and in regard to which I do not at all know I aer anything whatever can possibly exist as cor- r-sj. ?r. 2tl: :: :hr :ljuzi: — :. - empirical perception provide me in an actual case that is actually applicable ? But of this again. Z .. ; ;..:._:._—"... :'..:':._.:..: : -" - ?. --:_>:•;.:_: :.-- :„r. ":Jll-;r. ' '--.-.■- -I'.r. "try -—. :!.:"> -1 ^1 - "t-r ■flak 228 TEXT-BOOK TO KANT : ing, on objects of perception as perception, if only sensuous, no matter ours or another ; but are, for that very reason, mere thought-forms, through which (as such) there is not any actual object cognised. The synthesis or conjunction of the units in their (intellectual) complex held only of the unity of apperception, and was through it the ground of the possibility of perceptive cognition a priori, so far as such cognition depends on the understanding, and is, consequently, not only transcendental, but also merely pure-intellectual. There is basally presupposed in us, however, a certain form of sense-perception a priori which rests on the receptivity or susceptivity of im- pressions (a sensibility as such). Now, understanding, as spontaneity, is to be conceived capable of deter- minatively acting on the units of complex in inner sense, under and in accordance with the synthetic unity of apperception. That is, understanding may be conceived to think synthetic unity of the apper- ception of the complex l of a priori sense -perception, as the condition under which all objects of our (human) perception must necessarily stand. In this wise, then, it is that the categories, though mere thought- forms, get objective reality, or actual presence as factors in objects which may be given us in sense. These objects, too, must be understood as only appearances to, or presentations in, sense while as yet only sense-given ; for only in regard to objects of such a nature are we capable of a perception a priori or of a priori perceptive forms. The sense-synthesis, a priori possible and necessary, may be named the synthesis speciosa or figured one, in 1 It is possible to give a sense, and so reconcile one's self to the phrase "apperception of t the complex." Certainly, however, "apperception" ought to be apprehension. THE KRITIK OF PURE REASON. 229 contradistinction to that which, with a bearing only on perception as such (on any perception whatever) were a matter simply of thought in the category, and accordingly to be named synthesis in understanding, or synthesis intellectualis. Both syntheses — that of a priori sense (time and space) and that of the cate- gories — are transcendental. They are transcendental, too, not simply because they a priori precede other constituents of objective knowledge, but because also they are grounds a priori of the very possibility of that knowledge. But, again, the figural synthesis must, when con- sidered as bearing on the original synthetic unity of apperception, or on the transcendental unity, that is, which functions in the categories, be named, as in contradistinction to the merely intellectual conjunc- tion, the transcendental synthesis of imagination. Ima- gination is the faculty or power to exhibit an object in perception, even ivithout the presence of that object. Inasmuch, now, as all our perception is sensuous, imagination, too, must hold of sense, so far as con- cerns the subjective condition under which alone it can offer to the categories a correspondent percep- tion. So far, again, as its synthesis is an action of spontaneity, which is determinant, and not, like sense, merely determinable — so far, that is, as it can, under and in accordance with the unity of appercep- tion, a priori determine sense in its form — imagination is a faculty which a priori acts upon sense. Accord- ingly, its synthesis of perceptions, under the cate- gories, must be named the transcendental synthesis of imagination. This synthesis is the result of an action of understanding on sense, and is the first application of the former (ground, too, of all its other applications) in the direction of objects of what per- 230 TEXT-BOOK TO KANT: ception is possible to us. As figurate or figural, this synthesis is distinguished from the intellectual syn- thesis that, without imagination, is to be conceived to lie wholly in the understanding. So far as imagina- tion is possessed of spontaneity, I sometimes name it productive imagination, and distinguish it thus from the reproductive. Synthesis with the latter depends solely on empirical laws, those, namely, of associa- tion ; consequently, it cannot contribute anything in explanation of the possibility of a priori principles of objective perception, and, accordingly, being without position in transcendental philosophy, is relegated to psychology. ***** Here now is the place to explain the paradox which, in the exposition of the form of inner sense (§ 6), must have struck every reader : namely, how this sense can exhibit to consciousness even our own selves, not as we are in ourselves, but only as we appear to ourselves ; and for the reason that we only perceive ourselves as we are internally affected, which seems a contradiction, seeing that in that case we should have to relate ourselves to our own selves passively (that is, be acted on by our own selves). Hence we find it usual, in the current systems of psychology, to regard inner sense and the faculty of apperception as identical : we, for our own part, how- ever, desire carefully to distinguish them. What acts upon the inner sense is the native function of understanding to unite a perceptive com- plex, or to bring such under an apperception (as that on which its very possibility depends). As now, in us men, understanding is no faculty of perception proper, and perception itself, grant it even to be actual so far as depends on sensation, cannot, within or of itself, THE KRITIK OF PURE REASON. 231 apprehend, so as to unite and connect, as it were, the manifold of its own self, it follows that the synthesis of understanding, considered apart and by itself, is nothing else than that unity of act (at once, as such and without need of sense, known to understanding) by which understanding is able internally to affect any sense-complex that may be given to it — subject, of course, to the general perceptive form. Said act, therefore, under the designation of a transcendental synthesis of imagination, is exerted by understanding on the passive subject of which it is a faculty ; and so we rightly say that the inner sense is by these means affected. Apperception (with its synthetic unity) is so far from being identical with internal sense, that, rather, as source of all synthesis, it acts (under the name of the categories) on the complex of percep- tions generally (and so on objects generally), in an- tecedence of all sense-perception whatever. AVhile inner sense, again, is the mere form of perception, but without power of conjoining the particulars given in it, and still, consequently, without anything that can be called a certain actual definite perception. That, something that is a perception and distinct, is only possible through the consciousness of the deter- mination of inner sense by the transcendental action of the imagination (the synthetic action of under- standing on inner sense), which I have called the figural synthesis. This we see, too, always in our own selves. We cannot think a line without in thought drawing it, or a circle without so describing it, or the three dimen- sions of space without conceiving three lines perpen- dicularly to meet each other in the same point, or time itself without having regard, in that we similarly draw a straight line (which shall be the external 232 TEXT-BOOK TO KANT : figurate representation of time), merely to the act of synthesis in the complex, whereby we successively determine inner sense, and whereby also we observe the succession of this determination in sense. Motion, as act of the subject (not as determination of an object), 1 consequently the synthesis of the complex in space when we abstract from space itself and regard only the act whereby we affect the inner sense, but still as in accordance with its form, is what gives rise to the very notion of succession. Understanding, therefore, does not just find in inner sense such syn- thesis of the complex implied, but, on the contrary, produces synthesis by affecting sense. But how the ego that thinks itself can, as regards the ego that perceives itself, be different from it (for an under- standing that should at once perceive itself, is con- ceivably possible), and yet, at the same time also, as being the same subject, all one with it — how, there- fore, I can say, I, as intelligence and thinking subject, cognise myself as thought object so far as, in addition to thought, I am given in perception, only, like other phenomena, not as I am before understanding, but as before sense I appear to myself — all this has no more and no less difficulty than how I can be to myself an object at all, an object, that is, of perception and inner sense-consciousness. But that it actually must be so may (space being assumed as merely pure form of the appearances of outer sense) be made clear by considering that, time being no object of external 1 Motion of an object in space is no consideration of any pure science, and, consequently, not even of geometry ; for that something moves can only be known from experience, and not a priori. But motion, as described in space, is a pure act of the successive synthesis of the complex in outer perception, through instrumentality of prodiictive imagination, and belongs, therefore, not only to geometry, but even to transcendental philosophy. — K. THE KRITIK OF PURE REASON. 233 perception, we cannot otherwise represent it to our- selves than by the image of a line, and that, too, only so far as in thought we draw it ; for without this expedient we should be unable to conceive the unity of the dimension of time. Evidence to the like effect is this, that, as regards duration in time, or relative position in time, we must, for realization of concep- tion in such cases, have recourse to the fact of change in external things. Of all the conclusion, therefore, is that we must order the determinations of inner sense as sense-presentations in time, in precisely the same way in which we order those of external sense in space ; and that, consequently, if of the latter we admit that through them we know objects only so far as we are externally affected, we must no less admit of inner sense that through it we perceive our own selves only as we are by our own selves internally affected. That is, we must admit that, so far as con- cerns internal perception, we cognise our own subject only as sense-appearance, and not according to what it is in itself. 1 1 I do not see how there should be so much difficulty in conceiving that inner sense may be affected by our own eelves. Every act of atten- tion is an example of the fact. In it understanding always determines the inner sense (in accordance with the combination that is the subject thought) into that inner perception which corresponds to the complex in tlie synthesis of the understanding. How much the mind is commonly affected in this, every one will be able to perceive in himself. — K. I cannot help remarking here that, in the above (text) Kant has dis- tinguished by stars one of the most signal examples possible of his very worst writing. Not only do the two last sentences occupy fully a whole page with the most spider like sprawl of helplessly-intricate clauses, but what precedes them is even disfigured by actual errors, both verbal and syntactical. One is apt to surmise that Kant never stopped for comple- tion of his thought before he put his pen to paper ; but that he thought as he wrote. So it was that factor after factor and bearing after bearing interposed only by the way ; suggesting thus modification after modifica- tion, condition after condition, and, consequently, clause after clause. To think, too, that what is wanted to be said is so simple that a single sen- 234 TEXT- BOOK TO KANT §25. On the other hand, in the transcendental synthesis of mere mental consciousness as such (that is, in the original synthetic unity of apperception), I am aware of myself, not as I appear to myself, nor yet as I am in myself, but only as that I am. This cognition is a thought, not a perception. But, now, for an objective re- cognition of myself, there is required, besides the act of thought necessary for the reduction of any possible perceptive complex to the unity of apperception, a particular mode of perception, as well, in sujqdy of a convex. Thus, then, my own being is not mere appearance (and still less mere blind show). This, my being, however, can be determined only as in regard of the particular manner in which the com- plex presented for my synthesis is given in inner per- ception, and always, at the same time, as in accord- ance with the native form of inner sense. 1 In this tence — but see Commentary. Kant himself refers to § 6 ; a mucli better reference would be § 8, II. The errors are these. Under the stars, second paragraph, fourth sen- tence, there is an omission, or the punctuation and an " auf " are wrong. Third paragraph, second period, the writer's ear has put a terminal clause, an " Acht haben," which properly refers half way up ; or the references of " ohne," and " indem," and " Acht haben " are all wrong. By-and-by, Rosenkranz has a " Wahrnehmungen " which were better singular. 1 The I think expresses the act to determine my being. The fact of my being is thereby, therefore, already given ; but how I shall determine What it is, or how I shall realize in myself the constituent complex of cognisable units that is peculiar to it — this is not thereby given. To that there belongs self '-perception, and perception basally iinpbes an a priori given form (time), which is sensuous and attaches to the receptivity of my sentiency. Shoiüd I not have, now, further, another self-percep- tion by which what is determinant in me (of whose spontaneity as such I am only as yet supposed to be conscious) is given before the act of de- termining, in the same way as what is determinable (sentient) is given by time — in that case I cannot determine my existent nature (as that of a self-active being) : I only perceive the spontaneity of my thinking, that is, of the determining ; and my constituent existential nature remains THE KKITIK OF PUKE REASON. 235 way, consequently, I have no objective perception of myself as I am, but only as I, through sense, appear to myself. The consciousness of one's self is still, therefore, far from being an objective recognition of one's self, notwithstanding presence of all the cate- gories which compose the thinking of an object (as such) through synthesis of some given perceptive complex in an apperception. For perceptive cogni- tion of an object different from myself, I require, be- sides the general thinking of an object in the cate- gories, a perception as well, whereby to determine, or give filling to, said general notion (in the categories). In the same way, for objective recognition of myself, I require, besides consciousness, or besides the fact that I think myself, as well a perception of the com- plex in me whereby to determine, or give filling to, that bare thought. I exist, therefore, as an intelligence which is only conscious to itself of its bare synthetic function, but, for a complex to act on, is subjected to a limiting condition, 1 named inner sense. So situated, such an intelligence can realize said synthetic function on or in a perception, only in accordance with rela- tions of time, relations, consequently, which lie quite outside of the notions proper of understanding; and can, therefore, objectively recognise itself only as, in regard of a perception which cannot be intellectual and given by the understanding, it merely seems to itself, and not as it would perceive itself, were its per- ception intellectual. always only sensuous, or determinable as that of a sense-appearance. It is the spontaneity, all the same, that empowers me to name myself an Intelligence. — K . 1 " Verbindung" here is a mistake for Bedingung. 236 TEXT-BOOK TO KANT : § 26. Transcendental Deduction of the general possible empirical Action of the Categories. In the metaphysical deduction, the a priori origin of the categories was substantiated by their complete agreement with the general logical functions of thought (§ 10). In the transcendental deduction, again (§§ 20, 21), there was established the possi- bility of these categories as a priori cognitive elements in objects of perception as perception (perception proper, general, not yet special). What is now, then, to be explained, is the possibility of an a priori cog- nition (through categories) of actual objects of special sense, and not as concerns their mere general percep- tive form (the successions of time and space), but in reference to the laws of their synthesis. Such a priori synthesis were, as it were, a prescribing of law to nature, and even a making of it possible ; for, other- wise, it were inexplicable how everything whatever that is an object of our very senses must stand under laws, which have their source only a priori in the understanding itself. I remark, first, that by synthesis of apprehension I understand the setting together of the various ele- ments in an empirical perception, whereby empirical consciousness of such perception (still as sensible appearance) is made possible. A priori we have forms of external as well as in- ternal sensuous perception (in space and time), and the synthesis of apprehension must always conform to these, for it can be realized only in that form. But space and time are conceived to be perceived not merely as forms of sensuous perception, but as them- selves perceptions (objects implying a complex), and, consequently, as already a priori possessed of the de- THE KRITIK OF PURE REASON. 267 termination of unity of complex (see Trans. iEsthet.) 1 There is, therefore, unity of synthesis of the complex (whether as external to us or internal) — that is, a conjunction, to which all that can be perceived as de- termined in space or time must submit — already a priori given (as synthetic condition of all appre- hension), at the same time with (not merely in) these perceptive forms or perceptive objects. This synthetic unity, however, can be no other than that of the union of the manifold of an inherently given percep- tion (the a priori general one), in an original con- sciousness, conform to the categories, and only applied to our sensuous perception. All synthesis, conse- quently, whereby even sense-perception becomes possible stands under the categories ; and as experience is a cognition through connected sense-perceptions, the categories are conditions of the possibility of experience, and have, consequently, an a priori ap- plication to all objects of experience. *J& «V. »Sfc. M. W TT W TV* Suppose, for example, I observe the empirical per- ception of a house (the object, house) by apprehen- sion of its complex, there is presupposed under it 1 Space, as an object (an actual requirement of geometry), involves more than a mere perceptional form ; synthesis of complex, namely, in obedience to the form of sense, and into a perceivable cognition, in such wise that the perceptive form presents merely the complex, while the formal perception, on the other hand, presents the unity. This unity I had accounted in the ^Esthetic merely to sense in order only to signalize that it precedes any notion, not but that it presupposes a synthesis, which does not belong to the senses, but which at the same time conditions the very possibility of notions as in application to space and time. For as through it (understanding so determining sense), space and time are as perceptive objects originally just given, so the unity of perception implied belongs a priori to space and time themselves, and not to any notion (as category) of the understanding (§ 24). — K. This unity, which precedes any individual notion, is, of course, that of the first synthesis. 238 TEXT-BOOK TO KANT : the necessary unity of space and of outer sense gener- ally, and I picture, as it were, its shape, in accordance with this synthetic unity of the complex in space. But, leaving out of view the form of space, precisely the same synthetic unity is to be found in the under- standing as the category of homogeneous synthesis in perception generally (the category of quantity), to which, therefore, said synthesis of apprehension (the thing seen) must completely conform. 1 Or suppose I observe the freezing of water, what I apprehend are two states (liquid, solid) such that they stand towards each other in a relation of time. But I picture necessary synthetic unity of the sense-com- plex in the general underlying element of time, and without this synthetic unity said relation (of ruled place in time) would not be given as a single deter- minate perception. But now, leaving time out of view, this synthetic unity, as a priori condition under which I interconnect generally the complex of per- ception, is the category of cause — a category by means of which, when applied to contributions of sense, I determine all that happens into the form of a ruled relation in time. And therefore apprehension in the case of any such occurrence, consequently this (occur- rence) itself, as regards possible perception, stands under the notion of the relation of effects and causes, and so in all the other cases. 1 It is proved in this way that the synthesis of apprehension, which is empirical, must accord with the synthesis of apperception, which is intellectual and represented by the a priori category. It is one and the same mental spontaneity which, there as imagination, and here as under- standing, effects synthesis in the complex that may be before percep- tion. — K. In the second line of the above paragraph, Rosenkranz has, for appre- hension, the misprint "Apperception," an error which we saw before in § 24. THE KRITIK OF PUKE REASON. 239 Categories are notions which a priori prescribe laws to sense-appearances, and consequently to nature as the totality of such {natura materialiter spectata) ; and the question is, as they cannot be derived from nature nor subject themselves to it as their pattern (for then they would be merely empirical) — the question is, how are we to understand this, that nature must subject itself to them. In a word, How can cate- gories a priori determine the actual syntheses of objects in nature — instead, rather, of being derived therefrom ? Here is the solution of this enigma : It is not at all 1 more surprising how laws of objects in nature should of necessity agree with the under- standing and its a priori forms (its functions, namely, of synthesis in the varieties of the sense-complex), than how these objects themselves should of necessity agree with the a priori form of sense-perception. For laws just as little exist in the objects (mere affections of sense), unless only relatively to the subject (in which these inhere) so far as said subject has an understand- ing, — just as little as objects exist in themselves, and not relatively only to the subject so far as the subject has senses. Rule would necessarily accrue to things in themselves from themselves, and independently of any understanding that might cognise them. But our objects, being only sense-appearances, represent merely such things as, for what they are in them- selves, are wholly unknown. And, as such, plainly, neither can their connexions stand under any law but that which the connecting understanding pre- scribes for them. But what connects any complex in sense-perception is imagination ; and imagination, depending on sense for the complex of apprehension, is equally dependent on understanding for the unity of 1 Instead of um, Rosenkranz lias here the misprint "nun." 240 TEXT-BOOK TO KANT : its own intellectual synthesis. Inasmuch, now, as all possible perception depends on the synthesis of appre- hension, which (empirical) synthesis itself, for its part, depends on the transcendental synthesis, and con- sequently on the categories, so all possible perceptions, consequently all that can ever come to be a constitu- ent of empirical consciousness — and that is all objects of sense in nature so far as their connexion is in regard — must stand under the categories. Nature (merely as such) depends, for its necessary subjection to law and order, on the categories as the original source and ground ofthat law and order, in reference to which latter nature is as natura formaliter spectata. But to more laws than those on which a nature in general (as law and order of sense-appearances in space and time) rests, the competence of even the purest understanding does not extend as regards prescription a priori of laws to such objects through mere categories. Special laws, as concerning only what is empirically determined, cannot be completely derived from the categories, though standing in a body under them. Experience must, in addition, be applied to in regard of such. But of experience as experience, and of what can be known as an object of it, said a priori laws alone (the categories) supply instruction. § 27. Result of this Deduction of the Categories. We cannot think an object without categories ; we cannot cognise any object thought, unless through perceptions which correspond to these notions. Now all our perceptions are in sense, and such cognition, so far as the object of it is given, is empirical. But empirical cognition (or recognition — perception) is THE KRITIK OF PURE REASON. 241 experience. Consequently there is no objective cog- nition a priori possible to us, but one solely of objects of possible experience. 1 But such cognition, though confined merely to objects of experience, is not therefor all borrowed from experience. On the contrary, even such cog- nition has elements which originate a priori within ourselves ; firstly, the pure perceptions (time and space), namely, and, secondly, the pure notions of understanding (the categories). Now, there are only two ways in which we can think a necessary agree- ment of experience with notions of objects in it : either experience makes these notions, or these notions make experience, possible. The one alternative is not true of the categories (pure perception apart) ; for they are a priori notions, and consequently inde- pendent of experience (the assertion of an empirical origin would be a sort of generatio cequivocd). There remains, therefore, only the second alternative (as it were a system of the Epigenesis of pure reason) : that the categories, on the part of the understanding, namely, possess the grounds of the possibility of all our experience. How, however, they make experience possible, and what principles of this possibility they furnish in their application to the intimations of sense, 1 In order to preclude possible premature objection of questionable consequences to be apprehended from this proposition, I will only remark that, in thinking, the categories are not necessarily under restriction of the conditions of our sense-perception, but possess, so far, a quite un- limited field. It is only the perception of what we think, the determina- tion of it as an object, that requires elements of perceptive sense. Even without this latter the mere thought of some certain object may have always its own results of true and authentic application in the reasonings of the subject. Such application, however, as it is not always directed to actual determination of the object, to actual objective cognition, but rather to determination of the subject and its volition, is here not yet in place to be discussed. — K. Q 242 TEXT-BOOK TO KANT: will be further explained in what follows on the tran- scendental function of judgment. Between the two ways named, one might con- ceivably propose a middle-way : namely, that the categories are neither spontaneously-thought first principles a priori of our objective cognition, nor, again, that they are borrowed from experience, but that, implanted in us with our very existence, they are subjective germinal elements, or pre-capacities for thinking, which have been so fashioned by our Maker as in their function accurately to harmonize with the laws of nature which obtain in experience (a sort of preformation- system of pure reason). But, besides that in such an hypothesis it is impossible to see any end to presuppositions of pre-established potentialities of future judgments, this, as in oppo- sition to the said middle-way, would be decisive : namely, that the categories, in such a case, would want that necessity which essentially belongs to their very notion. For the notion cause, for example, which affirms, under a presupposed condition, the necessity of a certain result, would be false, if it were founded only on an arbitrarily implanted subjective necessity in the connecting of certain empirical facts conformably with such rule of relation. I should not be able to say, The effect is united in the object (that is, necessarily) to the cause, but only, I am so made that I cannot otherwise think these facts than as so and so connected. But that is just what the sceptic especially wants. For then all our knowledge, through any supposed objective validity in our judg- ments, would be mere illusion. Neither, in that case, would there be wanting those who would not admit for themselves said subjective necessity (matter of feeling as it must be). There could be no dispute THE KRITIK OF PURE REASON. 243 at least with any one as regards what would merely depend on how he was subjectively organized. A Brief Idea of this Deduction. It is the exposition of the pure notions of the understanding (and with them of all a priori theo- retical objective cognition) as principles of the possi- bility of experience, — of these, again, as determination of sense-appearances in space and time generally, — of these, lastly, from the principle of the original syn- thetic unity of apperception as form of the under- standing in a connecting reference to space and time, and to them, for their parts, as original forms of sense. # # * # # Thus far I have considered it necessary to proceed by paragraphs, inasmuch as we have been hitherto occupied only with the elementary notions. Now, however, that it is their application which is to be made conceivable, our exposition shall, without para- graphs, be in continuous connexion. ] Book II. — The Analytic of Judgments. General Logic is built on a frame which quite accurately coincides with the classified table in distri- bution of the higher cognitive faculties : Understand- ing, Judgment, Reason. Its analytic treats, there- fore, of Notions, Propositions, and Syllogisms, in exact accordance with the order and functions of the intellectual powers named ; of which it is further to 1 Beginning of § 27, "Anschauung" should be Anschauungen; next paragraph, the last parenthetic hook (after " Vernunft ") is wanting ; last sentence, a " da " omitted (in edn. Rosenkranz). 244 TEXT-BOOK TO KAKT : be remarked, indeed, that they are usually collec- tively comprehended, in a wide sense, under the common designation of the understanding simply. This merely formal logic, now, abstracting from all matter of cognition (whether pure or empirical), and occupying itself simply with the form of thinking (the form of discursive cognition), will be manifestly competent to supply, in its analytic, a canon for reason as well. For the form of reason must have its own fixed prescript which, without consideration of the contained matters specially concerned, will be capable of being a priori recognised through merely analyzing the processes of reason into their con- stituent moments. Transcendental Logic, again, involving a certain contained matter (in its pure a priori cognitions), to which matter it is restricted, cannot follow the same rule. Nay, it will be found that the transcendental application of reason is not in any respect objectively valid, and consequently that it does not belong to a logic of truth (an analytic) at all, but that, under the name of transcendental dialectic, it takes the place of a special division in the ordinary formulary of the schools only as a logic of false show. Understanding and judgment, accordingly, have, for their parts, each its own canon of objectively available use in transcendental logic, and are conse- quently comprehended in the analytic. But, on the other hand, reason, for its part, in any attempts it makes to decide something a priori of objects, and so extend our knowledge beyond the limits of possible experience, becomes, it will be found, utterly dia- lectical : its mere mock determinations, therefore, were altogether out of place in a canon, and only a canon has place in an analytic. THE KRITIK OF PURE REASON. 245 The analytic of judgments (or propositions), there- fore, will only provide a canon for the judging faculty. This canon, again, will instruct the faculty how to apply to sense-matter the categories in their quality as possessing the condition (or principle) towards a priori rules. 1 shall avail myself, there- fore, in prosecuting an analysis of the propositions proper of the understanding, of the designation doc- trine of judgment, whereby the business involved will be more exactly indicated. Introduction. — Of Transcendental Judgment Generally. If understanding be considered the faculty of rules, judgment will be the faculty that subsumes under rules, the faculty that distinguishes whether some- thing stand (casus clatce lerjis) under a given rule or not. General logic neither has, nor can have, any prescripts for judgment. For, abstracting from all matter of cognition, there can remain to it no business but the setting out analytically of the mere form of cognition in terms, propositions, and syllogisms, and the production, consequently, of rules in the general use of the understanding that are simply formal. Evidently, then, if logic sought, in such circum- stances, to prescribe how we should subsume, how we should discern whether something stood under its rules or not, it could only do so through yet another rule. But this rule, again, and just because it is a rule, demands anew an instructing of judgment, and hence it appears that, for its part, the understanding is capable of being instructed and qualified by rules, whereas judgment, again, is but a special talent that may indeed be exercised but not taught. Judgment, in point of fact, constitutes what is specific in so- 246 TEXT-BOOK TO KANT : called mother-wit, the want of which cannot be supplied by any schooling ; for let schooling amply offer, or even cram a limited understanding with rules borrowed from another, still the ability rightly to apply them must be the pupil's own. No rule whatever that might be supplied him with this intention would, in default of the natural gift, be safe from misuse. J A physician, therefore, a judge, or a statesman may have in his head many fine patho- logical, jurisprudential, or political rules — may have them in his head to such a degree, indeed, as to be actually a profound teacher of them, and yet he shall easily blunder in the application of them, either because he wants natural judgment (not understand- ing as faculty of rules), and, though he can very well understand the universal in abstracto, he is unable to decide of the particular in concreto as a case in point, or because, it may be also, he has not been sufficiently exercised in examples or inured to actual practice, for the formation of such a judgment. For this, in that reference, is precisely the one sole and great use of examples : they sharpen the judg- ment. As regards precision and exactitude of understanding, they are commonly, rather, preju- dicial to it, inasmuch as it is only rarely that (as casus in terminis) they adequately answer the con- dition of the rule, and because also they frequently weaken the effort of the understanding to see into 1 Want of judgment is -what is properly called dulness or stupidity, and for such an ailment there is no cure. An obtuse or restricted capacity, wanting nothing but the due degree of understanding and the notions proper to it, may very well be educated — educated even up to learnedness ; but the want (that of the Secunda Petri) will still be there. And thus it is that it is not at all unusual to meet very learned men who, even in their very learning, not infrequently betray said irre- mediable failing. — K. THE KRITIK OF PURE REASON. 247 rules according to their sufficiency in general and independently of the particular circumstances of ex- perience — inducing in this way a habit at last of using universal rules rather as formulae than as principles. So it is that examples are the go-cart of the judgment which he who wants the natural talent of it can never let go. But, though general logic has no prescripts for judgment, transcendental logic is quite otherwise. Nay, it would seem the precise business of the latter just, through rules, to guide and safe-guard judgment in its intromissions with the pure understanding. For philosophy appears not at all required for ex- tension of understanding in pure cognition, or as doctrine, but rather misplaced ; for, despite all attempts hitherto, little or no ground has been won with it so ; but as critique or criticism, in order to preclude the errors of judgment {lapsus judicii) in regard of the few pure notions of the understanding possessed by us — in that capacity (though the gain is but nega- tive) philosophy, and with all its sharp-sightedness and power of proof, is specially in request. Now this is the peculiarity of the transcendental philosophy, that, besides the rule (or rather the univer- sal condition to rules) which the category represents, said philosophy can at the same time a priori notify the case on which the rule is to be applied. The reason of this advantage over all the other theoretical sciences (mathematics alone excepted) lies in this, that the notions on which transcendental philosophy is engaged are such as to connect themselves a priori with objects. It is not a posteriori, then, that such notions can have their objective applicability proved ; for they possess a dignity beyond that standard. In their case, rather, there must be given along with themselves (at least in 248 TEXT-BOOK TO KANT : general but adequate characterization) the actual con- ditions under which objects are to be offered them ; or else they would be without matter, consequently bare logical forms, and not pure notions of understanding. This transcendental doctrine of judgment, now, will comprise two chapters : the first treating of the sense- conditions under which the categories can be alone applied (of the schematism, therefore, of pure under- standing) ; and the second of the synthetic proposi- tions (judgments) which a priori result from the cate- gories under these conditions and underlie all other a priori cognitions ; that is, of the ground-proposi- tions of the pure understanding. Chapter T. — The Schematism of the Categories. In every subsumption of an object under a notion the former must be homogeneous with the latter ; that is, the notion must be what is represented in the ob- ject which is to be subsumed under it; for the very expression, an object is subsumed under a notion, means that. Thus the empirical notion of a plate is homogeneous with the pure geometrical notion of a circle, seeing that the roundness which is implied in the former is in the latter visible (objective). But, now, the pure notions in comparison with empirical perceptions (or say, rather, perceptions of sense) are quite heterogeneous, and (the word being used strictly) cannot be possibly perceived. How, now, is the subsumption of these under those — how is application of categories to mere affections of sense possible ? for nobody will say that the categories (e.g., causality) can be sensuously seen or felt and are in- cluded in the sense-affection. This so natural and important question is precisely the reason now, which TUE KRITIK OF PURE REASON. 249 makes a transcendental doctrine of judgment neces- sary — to demonstrate the possibility, namely, of intro- ducing categories of the understanding into the affec- tions of sense. In all the other sciences, because in them the notions through which the object is thought are not so heterogeneous, not so different from those which attend said object as given in concreto, it is unnecessary to resort to any special exposition of the relative application. In this reference, now, it is evident that what is wanted is a iertium quid, which, homogeneous on this side with the category and on the other with sense, will mediate the connexion of the one with the other. This mediating agent, in a word, while wholly pure or non-empirical, must, on one side, be intellectual, and, on the other, sensuous. Such an agent we shall name transcendental schema. The category is a principle of the pure synthetic unity of a complex, no matter what. Time, as formal condition of any complex in internal sense, and con- sequently of all connexions in consciousness, implies, represents, or is an a priori complex of pure or non- empvrieal perception. A transcendental determination of time, then, is, with regard to the category which may be supposed to act in determination of its unity, so far homogeneous: like the latter, namely, it is uni- versal and depends on an a priori law. But, on the other side, again, it is so far homogeneous with sense, as time is an element in every actual empirical complex. Application of category to ingredients of sense, there- fore, will be possible through that transcendental determination of time which, as schema of category, mediates the subsumption of the latter under the former. After what has been shown in the deduction of the 250 TEXT-BOOK TO KANT : categories, is is to be hoped that no one will have any difficulty with the question, whether the categories are of merely empirical or whether they are also of a transcendental value ; that is, whether they are of a priori application to sense solely as conditions of a possible experience, or whether, as conditions of the possibility of things at all, they may have their application extended to objects in themselves (without any restriction to our sensibility). For we saw there that notions are quite impossible and meaningless unless an object be given either to them themselves or at least to the elements of which they consist, and consequently that they do not apply to things in themselves (without respect of whether or how they may be given us). AVe saw, further, too, that the only way in which objects are given us is by modifi- cation of our sensibility. Lastly, we saw also that pure a priori notions must presuppose, besides the function of understanding in the category, formal conditions of sense as well (particularly inner sense), which conditions constitute the universal proviso under which alone it is possible to apply a category to any object. We name this formal and pure con- dition of sense, to which in its action the category is restricted, the schema of this category, and the pro- cedure of understanding with these schemata the schematism of pure understanding. The schema is always in itself only a product of imagination ; but, as the synthesis of the latter then has not in view any single perception, but only the unity of a general process in determination of sense, the schema is not to be confounded with the figure or image. If I set down five points the one after the other thus, , what I have is a picture or repre- sentation (figure, image) of the number five. But if THE KRITIK OF PURE REASON. 251 I think just a number, any number at all, let it be five or let it be a hundred, then this thinking is rather the conception of a method towards the picture of some sum under a certain notion than this picture itself, which picture, in this latter case, it would hardly be possible to realize and compare with the notion. This idea now of a general process of imagination for providing a notion with its correspondent picture or image, I call the schema to this notion. In effect there underlie our pure sense-notions not pictures of the objects, but schemata. There can never be an adequate picture for the notion of a triangle in general. For it would never attain to that generality which enables the notion to hold good of any triangle, right-angled, oblique-angled, etc., but would be limited always to a part of this sphere. The schema of the triangle can never exist anywhere but in thought, and signifies a rule of the synthesis of imagination in regard of certain pure figures in space. But still less does any object of experience, or picture of it, come up to the notion. This last, rather, directly refers to the schema of imagination as a rule for the determination of our sense-perception in agree- ment with a certain general idea. The notion, dog, signifies a rule in accordance with which my ima- gination can figure to itself generally the form of a certain four-footed animal, without being restricted to any single individual shape as offered me in ex- perience, or even to whatever possible image I may construct in concreto. This schematism of our under- standing, in regard of objects of sense and their mere form, is a hidden art in the deeps of the human soul, the veritable trick of which we shall hardly ever come to detect in nature and openly display. We can only say this much : the picture (figure, image) is a pro- 252 TEXT-BOOK TO KANT: duct of the empirical faculty of productive imagina- tion ; while the schema (of sense-notions, as, for ex- ample, of the figures in space) is a product and, as it were, a monogram of pure imagination a priori, by and according to which pictures become first of all possible, which pictures, however, must be conjoined with the notion only through the schema producing them, and with which in themselves they never wholly agree. The schema of a category, again, is something that cannot be brought into any image, but is only the pure synthesis, in agreement with a rule of unity through notions generally (which notions are expressed in the categories), and is a transcen- dental product of imagination, which concerns the determination of inner sense generally according to conditions of its form (time) in regard of all cogni- tions, so far as these, under the unity of appercep- tion, are supposed a priori to cohere in a one notion. Not to stop now for the long and tedious analysis of what is required for transcendental schemata of categories in general, Ave shall rather simply set them down in the order of, and in agreement with, these categories. The pure picture of all magnitudes (quantorum) in outer sense is space ; but that of all objects of sense generally, time. The pure schema of magnitude (quan- titatis), as notion of the understanding, again, is num- ber ; and number is a cognition which represents the successive addition of homogeneous unit to homo- geneous unit. Number, then, is nothing else than unity of synthesis in a complex of homogeneous per- ception in general — by this, namely, that I generate time itself in the apprehension of the perception. Reality in the category is what corresponds to sen- sation ; any sensation, as such : that, then, the notion TUE KRITIK OF PUKE REASON. 253 of which in itself indicates a beingness or fact of some kind or other in time. Negation is that the notion of which represents a non-being in time. The distinc- tion of the one from the other, therefore, lies in the difference of a time filled from the same time void. As time is only form of perception, or of objects as affections of sense, what in these, consequently, corre- sponds to the sensation may be called the transcen- dental matte?' (reality) of all objects conceived as things on their own account. Now every sensation has a degree or magnitude, whereby it fills more or less the same time (that is, inner sense in regard of one and the same perception of an object), till it disappears in nullity (nothing, or negation). There is, therefore, a relation or connexion between reality and negation, or a transition, rather, from the one to the other, which transition exhibits every reality as a quantum. Accordingly, the schema of reality (as quantity of something so far as it fills time) is just this same continuous and uniform generation of filling in time, whether we suppose a certain degree of sen- sation progressively to ascend from nothing in time or regressively to descend to it. The schema of substance is the persistence of the reale in time ; that is, the conception of this reale as a substratum of empirical determination in time taken quite generally, which substratum persists, therefore, while all else changes. (Time itself does not fade away and vanish, but only the existence of the mutable that is in time. To time, therefore, as itself immutable and permanent, there corresponds in the presentation to sense the immutabile of existence, that is, substance : only as referred to it can succession and co-existence of sense-presentation be determined in regard of time.) The schema of cause and the causality of anything 254 TEXT-BOOK TO KANT : generally is the reale on which, whenever it is, some- thing else always ensues. It consists, therefore, in the succession of the elements in the complex, so far as this succession is subjected to a rule. The schema of community (reciprocity), or of the mutual causality of substances in regard of their accidents, is the co-existence of the determinations of the one with those of the other according to a uni- versal rule. The schema of possibility is the agreement of the synthesis of several ideas with the conditions of time generally (as, for example, in the reference that a thing and its reverse, or contrary, cannot both be at one and the same time) : it is the determination of a thing as conceivable at any time. The schema of actuality is existence in a deter- minate time. The schema of necessity is the existence of an object at all times. We see here, then, that the schema of every cate- gory refers to time : as that of quantity to the bringing to pass synthesis of time itself in the succes- sive apprehension of an object ; that of quality to the synthesis of sensation (sense-perception) with the con- ception of time, or to the filling of time ; that of relation to the connexions of the sense-units in each other's regard at any time (that is, as in accordance with a rule of the determination in time) ; and, lastly, those of the three modalities, to time itself, in regard of whether and how an object belongs to it. The schemata, therefore, are nothing but a priori time- determinations on rules : these, in the order of the categories, successively refer to time-range, time- filling, time-order, and time-complexion, as in regard of all possible objects. THE KRITIK OF PURE REASON. 255 From this it is clear that the schematism of the understanding as produced by the transcendental synthesis of the imagination has no other end than the unity of every complex of perception in the inner sense, and, in this way, indirectly, consequently, the unity of apperception as function correspondent to inner sense (which, for its part, is receptivity or affec- tion). The schemata of the categories, therefore, are the true and only conditions for providing these with an application to objects, and, consequently, with meaning ; and the categories are in the end, therefore, of no other use than a possible empirical one : they serve merely for this, namely, To subject, through grounds of an a priori necessary unity (towards the necessary conjunction of every consciousness in an original apperception), presentations of sense to uni- versal rules of synthesis, and thereby fitly exhibit them in the complete interconnexion of an expe- rience. All our objective cognitions, however, are only to be found within the sphere of possible experience ; and it is in the universal application to possible ex- perience that transcendental truth consists — the tran- scendental truth which precedes, and only makes possible, all empirical truth. It is self-evident, however, that, if the schemata of our sensibility first of all realize the categories, they, at the same time, also, restrict them — to conditions, namely, which, as in sense, are outside of the under- standing. So it is, therefore, that the schema is only the phenomenon, or the sensuous notion, of an object, in agreement with the category. (N^umerus is quantitas phenomenon ; sensatio is realitas phenomenon ; constans et perdurabile rerum is substantia phenomenon ; and the schemata of modality are, in the same way, possibilitas, 256 TEXT-BOOK TO KANT : necessitas, etc., phcenomena.) Now, if we omit a re- stricting condition, we certainly appear to amplify what notion was subjected to the restriction ; and the categories, it is true, may, in this way, be applied, without conditions of sense, and in their own pure import alone, to things as they are, whereas the schemata of the categories foreshadow things only as to sense they appear. That is, the categories may have a significance bestowed upon them that is in independence of all schemata and of much wider reach. In effect, the categories certainly do possess, even with removal of every condition of sense, a cer- tain function (only logical, however) of mere unity in ideas, but for which ideas there is no object, and neither, consequently, any application capable of furnishing a notion that has any objective bearing. Thus, for example, substance, with omission of the reference to something sensuously permanent, would not signify anything else than a something that may be thought as subject, and not possibly as at the same time predicate. But of such conception I can make nothing more than that: it does not enable me to know what actual properties that thing shall have which is to be such ultimate subject. Without schemata, therefore, categories only function notions for the understanding, but exhibit no object. That (an object) comes to them from sense which realizes understanding in so restricting it. Chapter II. — System of the Ground-Judgments of Pure Understanding. We have considered, in the preceding chapter, transcendental judgment only as in respect of the general conditions (schemata of sense) under which THE KRITIK OF PURE REASON. 257 alone it is competent for it to apply the categories in production of synthetic propositions. Our business now is to exhibit, in systematic connexion, what judg- ments understanding, under such critical proviso, actually a priori creates ; and hereto, without doubt, our categorical table will supply the natural and sure clew. For it is precisely the action of the categories, as on possible experience, that must effect all pure, or a priori, objective cognition ; and it is precisely their relation to sense that is motive to the complete and systematic exposition of all the transcendental ground-j udgmen ts. Ground -judgments a priori appropriate this name not merely because they are the grounds of other judgments, but because they themselves have no further or more general grounds. This circumstance, for all that, does not exempt them from the obliga- tion of proof. For although, in their case, it were impossible to carry any proof objectively further, seeing that they themselves, rather, are to their own objects the ultimate grounds of cognition, still this is no pre- judice to the possibility of a proof from the subjective pre-conditions of objective cognition generally. Nay, such proof as this latter is even indispensably neces- sary, as without it our position would bring with it the greatest suspicion of a mere surreptitious asser- tion. 1 Secondly, Ave shall restrict ourselves to those ground-judgments alone which immediately spring from the categories. The principles of the transcen- 1 There are grammatical slips in the middle of the above paragraph and the end of the preceding one. I think I have remedied both accord- ing to the author's intention. Rosenkranz makes an aller " alle " in the one case, and inserts a "man" in the other. The "man" does not quite cure the sense, and I do not think it right ; but the " alle " leads to non- sense. R 258 TEXT-BOOK TO KANT : dental aesthetic, bearing as they do on space and time as conditions of the possibility of things : — in that they are affections of sense; the restriction of said ground-judgments themselves, in that they are, namely, not to be applied to things in themselves : — these are considerations already behind us and apart from our immediate special inquiry. Neither do mathematical propositions belong here, for they depend on percep- tion and not on any pure notion of understanding. Still, a consideration of the possibility of these will necessarily have place here, because they constitute withal synthetic judgments a priori — not, indeed, to prove their correctness and apodictic certainty, which is no requisite for them, but only to deduce and make conceivable the possibility of such evidently a priori cognitions. We shall have to speak at the same time of the principle of analytic judgments, and as in contrast to our theme proper, the synthetic ones, because pre- cisely this contrast will free the theory of the latter from all misunderstanding, and distinctly exhibit them in their peculiar nature. Section 1. Of the Ultimate Principle of all Analytic Judgments. Of whatever import our cognition may be, and whatever bearing it may have on its object, the uni- versal, though but negative, condition of all our judgments is this, That they do not contradict them- selves ; or else, in themselves, and without considera- tion of the object, they are null. Still, even should there be no contradiction in our judgment, it is quite possible for ideas to be united in it which are not in the object, or that there should be in support of it no ground given whether a priori or a posteriori, A TUE KRITIK OF PURE REASON. 259 judgment, therefore, that is itself quite free from any inner contradiction, may still be groundless or false. This, then, That there cannot be joined with any- thing a predicate which contradicts it, is known as the proposition of contradiction. It is a universal, but only negative criterion of truth. It has its place, however, only in logic, inasmuch as it applies to cognitions merely as cognitions (their matter apart), and declares only, contradiction altogether destroys and subverts them. A positive use, however, is still possible for it. It is not necessary, that is, to regard it as only negative of (self-contradictor}') falsehood and error ; l it may really be applied as affirmative of truth. For, in the case of a judgment that is analytic, let it be negative or let it be affirmative, its truth is always sufficiently within the determination of the principle of contra- diction. What, namely, in the cognition of the object lies as notion, and is thought there — of that the con- tradictory is at all times legitimately denied; whereas the notion itself must be necessarily affirmed of the object, and just because its contradictory would con- tradict the object. Hence it is that we must recognise the 'proposition of contradiction as the universal and perfectly adequate principle of every analytic cognition. Further, how- ever, it is, as a criterion of truth, neither considerable nor applicable. For that no cognition can neglect it without self-destruction, in that regard it is certainly a conditio sine qua ?ion, but not, nevertheless, a deter- minative ground of truth. In prosecuting, therefore, our business proper, which concerns only the synthetic part of cognition, we shall always be on the watch not to offend against this inviolable principle ; but 1 The " er " here, in Rosenkranz (" so feme er "), were Letter es. 2 GO TEXT-BOOK TO KANT : not to expect from it, at the same time, any light on what cognition concerns us. There is, however, a version of this noted proposi- tion, devoid, as it is, of all material content and merely formal, in which, without consideration and quite unnecessarily, a synthesis has been really mixed up. It runs thus : It is impossible for anything, at one and the same time, to be and not to be. This proposition, besides that the apodictic certainty in it (which should be evident of itself) is, in the word impossible, superfluously appended to it, manifests itself as under a condition of time. Its import is this, namely : A thing = A, which is something = B, cannot, at one and the same time, be non-B. But it can very w r ell be both (B as well as non-B), one after the other. For example, a man that is young cannot at the same time be old, but he can quite easily be young at one time, and not-young, or old, at another. Now, seeing that it is logical only, the proposition of contradiction must not have its decisions submitted to considerations of time. Said version, therefore, is quite opposed to the scope of the proposition itself. The misconstruction is this. First of all we isolate, in the case of something, a predicate of it from the notion of it, and then, this very predicate we immedi- ately conjoin with the opposite of the thing in question. The resulting contradiction, all the same, is not with the subject, but only with the predicate of it (as now synthetically conjoined with its opposite), and only then, moreover, when both predicates are applied at one and the same time. If I say that a man, who is unlearned, äs not learned, I must add the condition of time, for he that is unlearned at one time can very well be learned at another. But if I say again, no unlearned man is learned, the proposition is analytic, TUE KRITIK OF PURE REASON. 261 for the quality (the unlearnedness) constitutes, once for all, so far, the notion of the subject, and the nega- tion enunciated is seen to be valid from the simple proposition of contradiction, without any necessity to add a condition of time. This too is the reason why I have (above) so manipulated the version of the proposition that the nature of an analytical judgment is clearly expressed thereby. Section 2. Of the Ultimate Principle of all Synthetic Judgments. To explain the possibility of synthetic judgments is no business of general logic, to which the very name need not be known. It is, however, what is of most importance in transcendental logic ; nay, it is the sole interest of this latter, when the question is of a priori synthetic judgments, their conditions and extent of validity. For so it is that such logic reaches its object — determination, namely, of the extent and limits of pure understanding. In an analytic judgment, I remain by the notion to reach the result. If affirmative, I only attribute to the notion what is already thought in it, and if negative, I only exclude from the notion what con- tradicts it. In synthetic judgments, again, it is re- quired of me to turn from and leave the given notion in order to find, in relation with it, something quite else than was thought in it. Now, the relation here is not one of identity or contradiction ; not such, therefore, that, in regard of it alone, the judgment itself, simply as it stands, can be seen into as true or false. Assuming the case, then, that, if we would syntheti- cally conjoin a certain notion with some other notion, it is necessary for us not to remain by the former, 262 TEXT-BOOK TO KANT: but to leave it, and look about us elsewhere, it is evi- dent that what we must find is a tertium quid, which shall effect the required synthesis of the two notions. What now, then, can possibly be this tertium quid that shall be medium for all a priori synthetic judg- ments? There is only one common element within which all our ideas are to be found, and that is internal sense, of which time (and a priori) is its form. The process of synthesis in our ideas, again, is the work of the imagination. And lastly, unity in synthesis of the ideas, synthetic unity, without which unity there cannot be a judgment, depends on func- tions of unity in the unity of apperception. The possibility of synthetic judgments, therefore, and of pure synthetic judgments (for all three are sources of a jriori cognition), will require to be sought in these. Nay, such judgments will necessarily issue from these sources, if there is at all to be a cognition of objects which shall solely depend upon a synthesis of mental elements. If a cognition is to have objective reality, that is, if it is to bear on an object, and have sense and mean- ing in it, the object itself must, in one way or other, be capable of being given. Without such object the notions are void ; for, though we have thought through them, we have, in effect, through this thinking cognised or recognised nothing : we have only played with ideas. To give an object, again, when such object is not merely to be mediately supposed, but immediately placed before us in perception, is nothing else than to realize its idea in experience, either as actual or as possible. Even space and time, pure and non-empirical as these cognitions are, and certain as it is that they are set up absolutely a priori in the mind within, would, nevertheless, be devoid of objec- THE KRITIK OF PURE REASON. 263 tive truth, and sense, and meaning, were their appli- cation and necessity not actually demonstrated as in reference to the objects of experience. The very idea in consciousness of them is a mere schema, that refers itself ever to reproductive imagination, whose busi- ness it is to call up the objects of experience, without which objects they themselves (space and time) would have no meaning. And, without distinction, all notions are similarly situated. The possibility of experience, therefore, is what gives objective reality to all our a j.riori cognitions. But experience rests on the synthetic unity of the sense- elements (the impressions, etc.), that is, on a syn- thesis according to notions of an object generally in elements of sense. Without this synthesis notionally towards an object, the contributions of sense would not even be perceptive cognition, but only a rhapsody of sense-impressions which would not cohere in any ruled context of a one thoroughly connected (possible) consciousness, nor, consequently, as assimilated into the transcendental and necessary unity of appercep- tion. Experience, therefore, has principles for its form which underlie it as a priori ground. These, namely, are, for the impressions of sense, universal rules of synthesis, of which, as necessary conditions, the objective reality can always be proved in ex- perience, nay, as in regard of the very possibility of it. Without such application, a priori synthetic judgments were completely impossible; for in that case they would be without a tertium quid — they could have no pure object, on which tried, the syn- thetic unity of their notions might exhibit objective reality. Although, therefore, of space and of possible con- figurations in space on the part of productive ima- 264 TEXT-BOOK TO KANT : gination, we really do a priori know so much in synthetic judgments, and without appeal to actual experience, yet such knowledge would be occupation with what were chimerical only, did we fail to regard space as simply condition to the elements of sense, which elements shall constitute the stuff and material of external experience. Said pure synthetic judg- ments, consequently, do, though only mediately, refer to possible experience, or rather to just this possibility of it, and on it alone they ground the objective truth of their synthesis. Experience, as empirical synthesis, is thus, in its possibility, that single knowledge which extends realization to all other syntheses ; and this latter, again, as a cognition a jiriori, gets truth (agreement with an object), only by this, that it has no interest but what is necessary for the synthetic unity of experience as such. The ultimate principle of all synthetic judgments, therefore, is : Every object stands under the necessary conditions of the synthetic unity of the complex of perception in a possible experience. Synthetic judgments are a priori possible, then, if, to the formal conditions of a priori perception, the synthesis of imagination, and the necessary unity of it in a transcendental apperception — if to this tran- scendental machinery we give the general direction towards a possible empirical cognition, and say : The conditions of the possibility of experience in general are conditions as well of the possibility of the objects of experience, and possess thereby objective reality in a necessary synthetic judgment. THE KRITIK OF PURE REASON. 265 Section 3. Systematic Idea of all Synthetic Ground-Propositions of the Understanding. That there are such propositions at all, is to be solely ascribed to pure understanding. And pure understanding is not only the faculty of the rules that bear on the connexions of experience, but is source as wel/i of those propositions through which it is that whatever is only capable of becoming an object for us stands necessarily under rules ; for without such rules there could never accrue to the sense-impres- rions cognition of an object corresponding to them. /Even the laws of nature, considered as principles of the empirical use of understanding, bring with them an expression of necessity, and, consequently, at least the presumption of being determined by grounds which must be valid a priori and before all experience. But all laws of nature, without ex- ception, stand under higher principles of the under- standing, inasmuch as they only apply these to special cases of experience. These principles alone, there- fore, furnish the notion, which constitutes the con- dition and, as it were, the exponent towards pro- duction of a rule ; while experience, for its part, again, supplies the case which is to come under the rule. There is little danger that empirical principles should be mistaken for those of pure understanding, or vice versa ; for the necessity on notions in the one case, and in the other the want of it (so easily seen, however generally the proposition may hold), readily preclude confusion. There are, however, certain pure a priori propositions which I should not attri- bute to pure understanding. These, namely, are not derived from pure notions, but from pure perceptions 266 TEXT-BOOK TO KANT : (though through understanding), and it is of notions that understanding is the faculty. Such propositions are found in mathematics. As said, they are percep- tive in their nature, but, nevertheless, the deduction of them — what concerns their authority in experi- ence, their objective truth, nay, the possibility of the a priori synthetic cognition they involve — depends still on pure understanding. Under my own principles, then, if I exclude from them the propositions of mathematics, I shall certainly include those on which these rest for their very possibility and a priori objective truth. Of said pro- positions these, therefore, shall constitute the prin- ciples proper — not originating from perception for notions, but from notions for perception. In the application of the categories, again, to possible experience, these, in exercising their syn- thesis, will, as formerly classed, be either mathe- matical or dynamical. The former, that is, will address their synthesis to the perception (the very being), the latter to the relative existence (the simple connexions), of objects of sense. 1 The a priori con- ditions of perception are, in regard of a possible ex- perience, out and out necessary ; while those, again, of the relative existence of the objects of a possible empirical perception are in themselves only con- tingent. Hence the propositions that arise in the mathematical application will be unconditionally 1 I have translated above the single word " Daseyn " by the two words " relative existence ; " and that is exactly what Kant means by it. He has not a thought of bringing into existence in his mind : he only thinks of objects relatively the one to the other in regard of connexion, once their existence has been provided for from elsewhere. It is of their seyn as relatively da he thinks, and not at all of their existence as such. Erscheinung I sometimes translate simply object of sense, when legitimate occasion offers. TUE KRITIK OF PURE REASON. 267 necessary, that is, apodictic ; while those that are of dynamical name will bring with them the character, indeed, of an a priori necessity, but only under the condition of the empirical thinking that shall be found in an experience. These latter, then, will exhibit this character only mediately and indirectly ; and, consequently (without prejudice to the universal certainty introduced by them into experience), they will not possess the same immediate evidence which is proper and peculiar to the others. But this will be better seen in the sequel. The table of the categories guides us quite natur- ally to that of the ground-propositions : these, namely, are nothing but rules of the objective application of those. Accordingly all the ground-propositions of pure understanding are — 1. Axioms of Pure Perception. 2. Anticipations of Sense-Perception. 3. Analogies of Experience. 4. Postulates of Empirical Thinking in General. I have chosen these designations purposely, in order that the differences of evidence and use on the part of those propositions should not pass unob- served. It will be directly perceived, however, that, as regards as well the evidence as the a priori action on objects which is connected with the categories of quantity and quality (in respect of this latter form alone being considered), the relative propositions are, in both references, conspicuously different from the other two : the force or import on both sides being equally that of complete certainty, it is only discur- sive in the latter, while it is intuitive in the former. These, therefore, I shall call the mathematical, and those the dynamical, ground-propositions. 1 It is to 1 All conjunction is either composition or connexion. The former is a 268 TEXT-BOOK TO KANT: be borne in mind, however, that I have in my eye here just as little the propositions of mathematics in the one case, as those of (physical) dynamics in the other, but only those of pure understanding when it is related to inner, a priori sense (without reference, that is, to the elements themselves afterwards given by special sense). Nevertheless, it is on these last pro- positions that the former (the mathematical, etc.) are dependent for their very possibility. The designations I give said ground-propositions, therefore, are due to considerations rather of their application than of their own contents. But I proceed now to the dis- cussion of them, and in the order in which they appear in the table. 1. Axioms of Pure Perception. The principle of these is : All perceptions are ex- tensive magnitudes. Proof. All objects involve in form a perception in space and time ; and this influence of space and time is synthesis of elements, not necessarily belonging the one to the other. For instance, the two triangles into which the diagonal divides a square do not, taken as individuals, necessarily belong the one to the other. Of this nature, now, is the homogeneous synthesis in whatever can be mathe- matically looked at ; and it is capable of being distinguished either into that of aggregation, in extensive, or into that of coalition, in intensive, magnitxides. The latter conjunction (that of connexion, namely) is a synthesis of elements that do necessarily belong the one to the other. Substance and accident, cause and effect, for example, are regarded as constituting a priori (necessary) connexions. The connected elements at the same time are heterogeneous. As of elements relatively existent the one to the other, I call this synthesis dynamical; which on its side, again, is capable of distinction as physical (of objects mutually) or metaphysical (of objects considered on the question of their evidence to the mind). — K. The last " können " here were better kann .' THE KRITIK OF PURE REASON. 2G9 presupposed as a 'priori universal condition that pre- cedes and underlies all objects. These, therefore, cannot be otherwise apprehended (taken up, that is, into empirical consciousness) than through synthesis of the complex of constitutive units, by which syn- thesis there are brought about perceptions of a deter- minate space or a determinate time. This synthesis, then, is a putting together of homogeneous elements, and results in a consciousness of the synthetic unity of just such complex. Now consciousness of any homogeneous complex in perception, so far as it is conceived necessary for rendering possible the idea of an object, is the notion of magnitude {quantum). Consequently even the perception of an object, as phenomenon in our sense, is only possible through the same synthetic unity of the given sensuously perceptive complex, by means of which the unity of homogeneous synthesis is, in the notion of quantity, thought. That is, the phenomena of our sense are all quantities — all extensive magnitudes, indeed — because, as perceptions in space and time, they must come before us in or through precisely the same synthesis as is determinative of space and time themselves. I call that magnitude an extensive magnitude, where the cognition of the parts renders possible the cognition of the whole (and, consequently, necessarily precedes it). I cannot picture to myself a line, how small soever, without in thought drawing it ; and to draw a line is, from a certain point, to generate all the parts of it, one after the other, and so mark out the object itself. And this is equally the case with all the parts, however infinitesimal, of time. I figure to myself in it only the successive progression from moment to moment, and in such manner that at last, through all these parts of time and the synthesis of 270 TEXT-BOOK TO KANT: them, there is generated a certain definite and deter- minate time-magnitude. But the element of percep- tion pure and proper in all objects of our sense, being- cither space or time, every such object must, as per- ception, be an extensive magnitude : it can be recog- nised in apprehension, namely, only through successive synthesis (of part to part). All objects of sense, therefore, are perceived as aggregates (collections of parts previously given) ; and this, be it observed, is not the case in regard of every kind of magnitude, but only of that kind where the magnitudes are apprehended and cognised as, strictly and properly, extensive. It is on this successive synthesis of productive imagination in the generation of figures that the mathematic of extension (geometry) founds. Its axioms express the a priori conditions of sense-per- ception ; and under these conditions only is a schema possible of any pure notion of external perception : as, for example, between any two points only one straight line is possible, two straight lines cannot in- clude a space, etc. These are axioms which apply properly to magnitudes (quanta) as such. But as regards quantity (quantitas), that is, the answer to the question, how much or great something is, there are in that respect not any axioms ; at the same time that in the general reference there are various synthetic and immediately certain proposi- tions (indemonstrabilia) . For that, if equals be added to equals, the wholes are equal, or if equals be taken from equals, the remainders are equal — these are analytic propositions, seeing that I am directly con- scious of the identity of the one amount with the other, whereas axioms are synthetic a priori proposi- tions. And, again, the evident propositions in the THE KRITIK OF PURE REASON. 271 relations of numbers, though synthetic certainly, are not universal like those of geometry, and consequently not axioms. They may be named numerical formula?. The proposition 7 + 5=12 is not analytic. For neither in the 7, nor the 5, nor the conjunction of both, do I think the number 12 (that I do think it in the addition of the two, that is not the question here; for, in an analytic proposition, the question only is, whether I actually think the predicate in the notion of the subject). But though synthetic, it is only a particular proposition. So far as the synthesis of the units simply is considered here, that synthesis can be accomplished only in one certain particular way, at the same time, certainly, that the application of these numbers is afterwards universal. When I say, it is possible to construct a triangle by means of three lines, of which any two are together greater than the third, I am present to the mere function of productive imagination which draws the lines smaller or greater, and allows them to meet in all manner of angles at will. Whereas the number 7 is only possible in one single way, as is the case also with the 12 which results from the synthesis of the former with 5. Such propositions, therefore, Ave may call numerical formulae, but not axioms. We should otherwise have quite an infinitude of the latter. This transcendental ground-proposition of the mathematics of sense greatly enlarges our a priori knowledge. For it, and it alone, renders pure mathe- matic applicable in its complete precision to objects of experience. And this latter fact without it, indeed, is so far from being of itself evident, that it has given rise to much controversy. Perceptions of sense are not things in themselves. Empirical perception is only possible through pure (space and time). What geometry 272 TEXT-BOOK TO KANT: says of the latter, therefore, is necessarily true of the former; and such allegations in resistance as that objects of sense need not be submitted to the laws of construction in space (the infinite divisibility of lines and angles, for example), must sink of themselves. 1 For objective truth were thereby denied to space, and along with it to all mathematics, so that it would be impossible for us any longer to know why and how far the latter were to be held applicable of the objects of sense. The synthesis of spaces and times it is, that, as synthesis of the essential form of all percep- tion, is what renders possible at the same time em- pirical apprehension, and consequently all external experience and all perception of any of its objects ; and what holds of mathematics in application to the former synthesis is necessarily true also of this latter. All objections to this are mere chicanes of an ill- advised reasoning, which would erroneously free the objects of sense from the formal conditions of our sensibility, and represent them as objects in them- selves and addressed so to the understanding — the truth being that they are mere affections of, or ap- pearances to, our senses. In fact, were they such objects and so given to the understanding, then truly there could be synthetically known nothing whatever a priori of them, and consequently nothing whatever also of space through pure notions; nay, the very science which is determinative of space (geometry), would itself be impossible. 1 The "dürfe" and "muss" here ought surely to be dürfen and müssen. THE KRITIK OF PUKE REASON. 273 2. Anticipations of Sense. The principle of these is, In all perceptions of sense, the reale that is matter of sensation has intensive magnitude — that is, degree. Proof. Sense-perception is empirical consciousness, or such that it has at the same time sensation in it. Sense- affections, as objects of sense- perception, are not pure (merely formal) perceptions, like space and time (which, for their parts, can, in themselves, not be perceived of sense). They contain, therefore, over and above the element of pure perception, the material elements towards an object (that element or those elements whereby something is cognised as ex- istent in space or time). These material elements are constituted by the reale of sensation, as mere subjective feeling of which there can only be the consciousness that the subject is so affected, and which is then referred to some object. Now, from empirical to pure consciousness there is a gradual transition pos- sible, in the course of which the reale that is present in it at first may, in the end, completely disappear, and there will remain at last a merely formal con- sciousness (now a priori) of the complex proper to space and time alone. Contrariwise, consequently, there is the possibility of a synthesis in the amount of a sensation, up from its beginning, as nothing in pure perception, until it reaches any conceivable magni- tude of feeling in consciousness. Sensation, now, being in itself not an objective consciousness, and in- volving, as such, neither the perception of space nor of time, is incapable of constituting an extensive magnitude. Still it is a magnitude, and a magnitude s 274 TEXT-BOOK TO KANT: such that, in the apprehension of it, empirical con- sciousness increases, from the nothing of it in a cer- tain time, up to the given actual amount. This, then, is intensive magnitude ; and such magnitude, degree, that is, of influence on sense, must be correspond- ingly attributed to all perceptive objects so far as they involve sensation. Whatever cognition enables us to know and deter- mine a priori, or beforehand, some actual ingredient of empirical perception, that cognition we may name an anticipation ; and beyond a doubt that is the sense in which Epicurus used his term Trp6\ti\Jsi$. Inasmuch, however, as there is something in the perceptions of sense which can never be known a priori, and which, therefore, constitutes the element specially distinctive of empirical cognition as compared with cognition a priori, sensation, namely (as the matter of the per- ception of sense), it follows that it is precisely and specially that which cannot possibly be anticipated. On the other hand, we may very well name any pure determinations in space and time, whether of shape or size, anticipations of the objects of sense, because they a priori present to us a constant constituent of whatever may be actually a posteriori given to us. Suppose, however, that, in mere sensation as sensation, sensation as such (not referring to any particular given sensation), it were possible a priori to know something, such something would, certainly, very specially deserve to be named an anticipation. For it could not but prove surprising to forestall experi- ence even in what was peculiar to it — its matter, namely, which as matter it is to be supposed we could only procure from experience. Now, such is the actual state of the case in point of fact before us. The apprehension of mere sensation occupies only THE KRITIK OF PURE REASON. 275 a moment (of course not referring to succession of different sensations). So far, as there is in appre- hension no successive synthesis (of part after part into a whole), there is no question here of an extensive magnitude : the ceasing of sensation in the moment it occupies would exhibit this moment as void, con- sequently = 0. What in the empirical perception, now, corresponds to the mere sensation is reality (realitas phenomenon) ; as what answers to the want of it is negation, which is = 0. But, again, every sensation is capable of a diminution, in such manner that it may gradually grow faint and disappear. There is between reality and negation in the sense- consciousness, then, a continued series of many pos- sible intervening sensations, differing the one from the other always by a less difference than that be- tween the full given amount and the zero or complete negation. That is, the reale in the sense-presentation has always a magnitude, but a magnitude that does not appear in apprehension as a magnitude, seeing that said apprehension takes place in a moment through the mere sensation, and not through a suc- cessive synthesis of several sensations — not, therefore, of part after part into a whole. This reale of any sense- presentation, consequently, has a magnitude, but not of the kind that is named extensive. A magnitude, now, which is apprehended only as a one, a unit, and in which a many (plurality) is only conceivable as that of an approach to negation = — this I call an intensive magnitude. The reality of any object possesses, then, intensive magnitude, or degree. Were this reality looked at in the point of view of a cause (of the sensation, or of any other ele- ment in some cognition, say of a change, for example), then the degree in it were to be named moment (e.g., 276 TEXT-BOOK TO KANT: moment of gravity), and for the reason that degree betokens a magnitude, the apprehension of which is not successive, but instantaneous. I remark this, however, only in passing; for with causality I have nothing as yet to do. Every sensation, and consequently every reality in the object or perception of sense, let each be as small as it may, have, according to this, a degree — an in- tensive magnitude, which, as such, is always capable of becoming less and less, and so that, between reality and negation in it, there is an unbroken continuity of possible smaller realities, and possible smaller percep- tions. A colour, a red, for example, has a degree which, let it be ever so small, is never the smallest ; and it is situated precisely in the same way with heat, gravity, etc. That property of magnitudes, whereby no one part in them is the smallest possible (no part is simple), we name the continuity of these. Space and time are quanta continua : no part can be taken in them with- out including it between limits (points and moments) ; only in such manner, consequently, that the part itself is again a space or a time. Space, therefore, consists only of spaces, time of times. Points and moments (instants) are only limits, that is, mere loci of limitation in them. Loci of limitation always pre- suppose, however, the objects which they are to limit or determine ; and out of such mere loci as con- stituents which shall be capable of being given before space or time themselves, neither the one nor the other can be made up. Magnitudes of this kind may be termed fluent, seeing that the synthesis (of pro- ductive imagination) in their genesis is a progression in time, the continuity of which we usually charac- terize by the expression ^zw ovflouing. THE KRITIK OF PURE REASON. 277 All objects of sense, accordingly, are continuous magnitudes ; as well in perception proper, where they are extensive, as in sensation proper, where they are intensive, magnitudes. If the synthesis of the sense-complex be interrupted, then we have an aggre- gate of several sense-units, and not properly a single sense-cognition as a quantum ; for a quantum is gener- ated, not by the mere progression of a productive syn- thesis of some kind, but by the repetition of a syn- thesis that is as well perpetually ceasing. If I call 13 dollars a money-quantum, I am right so far as I understand by the expression the amount of a mark in fine silver; such mark being undoubtedly a con- tinuous magnitude in which no one part is the smallest possible, but each part is capable of consti- tuting a bit of coin, with the possibility in it of sup- plying matter for still smaller bits, and so on. If, however, I understand by the expression 13 round dollars, just so many coins (their actual silver amount being what it may), then I name it improperly quantum ; rather, as a number of separate silver pieces, I must call it an aggregate. Still, number, nevertheless, always implying the principle of unity, the conjoint cognition is as unity a quantum, and as a quantum always also a continuum. All objects, now, whether as extensive or intensive, being continuous magnitudes, the proposition that all mutation also (transition of something from one state into another) is continuous, might be very easily proved here, and with mathematical evidence, did not the causality of a mutation at all lie quite without the bounds of a transcendental philosophy, and presup- pose empirical principles. For that there is such a thing as a cause possible — something, that is, capable of channdno; the condition of things, or even determin- 278 TEXT -BOOK TO KANT : ing in them the contrary of some given state — of that there is not a hint given us in understanding a priori. And this, not only because there is no possible understanding of such a thing (for we are similarly situated with other a priori cognitions), but because mutability concerns certain determinations of objects which we can only learn from experience ; not but that what is called cause in a mutation must be sought in principles which are beyond mutation. Here, how- ever, Ave have nothing available before us but the pure primitive notions of all possible experience, which must be free from everything empirical; and it is impossible for us, therefore, prematurely to approach the subject of a physica pura, built, as it is, on certain fundamental facts of experience itself, without injury to the unity of the system. We do not suffer under any scarcity of proofs, how- ever, of the great influence which our principle here possesses in anticipation of perceptions, and even in supplement so far of any want of them, that it shuts the door in the face of all erroneous inferences that mio-ht be drawn thence. o For if all reality of perception has a degree, between which and negation there is the possibility of an in- finite series of ever-lessening degrees, and if, more- over, each sense must possess only a certain degree in its receptivity of sensation, it is evident that no perception or experience can possibly prove, mediately or immediately, or by whatever expedient we may conceive, an entire want of reality before sense. There can be drawn from experience, that is, never any proof of either a void space or a void time. For the total want of reality in the perception of sense can itself, firstly, not be perceived ; and, secondly, from no single cognition of sense and the difference in the THE KKITIK OF PURE REASON. 279 degree of its reality, is it possible for it to be inferred, or ought it even for explanation of it (the reality) to be at any time assumed. For, though the entire perception of a certain definite space or time be out and out real (no part of it, that is, void), still, because every reality has its degree, which, independently of any change in the extensive magnitude of the rela- tive object, may decrease infinitely even to nothing (the void), there must be infinitely different degrees in the filling of space or time ; so that in different objects the intensive magnitudes may be greater or less, at the same time that their extensive magnitudes are equal. We shall exemplify this. Almost all teachers of natural philosophy, inasmuch as they observe, on the part of different kinds of matter, great differences of quantity under equal bulk (partly through the moment of gravity or weight, partly through that of resistance to other bodies in motion), unanimously infer from this that volume (bulk, extensive magni- tude) must, in all bodies, be more or less empty. But who would ever suppose it possible of these mostly mathematical and mechanical inquirers that they rested this their inference solely on — what they pre- tended so carefully to avoid — a metaphysical presup- position ? A presupposition of this kind they make plainly here, however, in that they assume the reale in space (not to call it here impenetrability or weight, which are empirical ideas) to be everywhere alike, and to differ only in extensive magnitude or quantitative amount. To this presupposition, for which they could have no grounds in experience, I oppose a transcen- dental rationale. I do not pretend to explain thereby differences in the filling of space, but only to negate said presupposition in the necessity it assumes that 280 TEXT-BOOK TO KANT : such differences can be explained not otherwise than by the hypothesis of empty spaces (pores). This rationale will have at least the merit to set under- standing free to think differently for itself the differ- ence in question, if an hypothesis at all is to be held necessary here for the explanation of the natural fact. We see from what has been said, namely, that, though two equal spaces may be completely filled with dif- ferent matter, and even in such a manner that in neither of them a point can be found in which matter is not present, still every reale has, with the same quality, its own degree (of resistance or weight) which, without diminution of its extensive magnitude, ma}' become infinitely less and less, before actually passing over into vacancy and disappearing. Thus, an ex- pansible element which occupies a certain space, say heat, and in like manner every other reality (to sense), may, without in the least leaving any smallest part of the space void, infinitely diminish in its de- grees, and nevertheless quite as well fill the space with these smaller degrees, as some other object with greater ones. I do not at all mean to maintain here that, with the diversity of matters, and in the ratio of their specific weights, this is actually the case, but only to demonstrate, from a proposition of pure understanding, that the nature of our sense-percep- tions makes such a mode of explanation possible. But, that being so, it must be erroneously assumed and, through an alleged a priori principle of the under- standing, erroneously maintained, that the reale of objects is always equal in degree, and only different in aggregation and extensive magnitude. Nevertheless this anticipation of perception has for an inquirer who is transcen den tally trained, and ac- customed, therefore, to be on his guard, always some- THE KRITIK OF PURE REASON. 281 thing that surprises and arrests attention. It excites doubt and reflection that the understanding should be capable of anticipating a synthetic proposition such as that about the degree of all reality in objects, and about the possibility, consequently, of an inner difference in the sensation itself, abstracting, that is, from the empirical quality (as so and so) of the objects themselves. It is a question, therefore, well worth answering, How it is that the understanding is able synthetically to pronounce a priori upon objects, and even to anticipate these in what is pro- perly and purely empirical — what, namely, concerns sensation. The quality of the sensation — colour, savour, etc. — is always merely empirical, and cannot be a priori realized. But the reale that corresponds to the sensa- tions taken quite generally, and as in contrast to the negation = 0, represents only something, the notion of which implies being, and means the synthesis in empirical consciousness generally. In inner sense, namely, the empirical consciousness may be raised from up to any possible greater degree ; and thus the very same extensive magnitude (an illuminated surface, say) may excite quite as much sensation as an aggregate of several (less illuminated) others at once. The extensive magnitude of the object, then, may be completely abstracted from; and yet we may perfectly well conceive, in a single moment of mere sensation, a synthesis of uniform successive rise from up to the given empirical consciousness. All sensa- tions, therefore, are, as such, only a posteriori given, but the property in them, that they possess degree, may be recognised a priori. It is remarkable that, of quantity generally, we can a priori know only a single quality (continuity), and of quality generally only a 282 TEXT-BOOK TO KANT : single quantity (intension, degree) : anything further is left for experience. 3. Analogies of Experience. The principle of these is, That experience is only possible through consciousness of a necessary con- nexion in the perceptions (objects) of sense. Proof. Experience is empirical cognition, i.e., cognition that, through perceptions of sense, determines an object. Experience, therefore, is synthesis of said perceptions, a synthesis that is not given by percep- tion, but that rather gives to its implied sense-complex, the synthetic unity of a certain single act of con- sciousness. This synthetic unity constitutes what is essential to a perceptive recognition of objects, i.e., to experience (not, that is, to mere subjective sensa- tion or perception). Perceptions, now, at first hand, come together only contingently ; 1 there neither ap- pears, nor can appear, any necessity of connexion in them so far, any necessity brought by themselves. Apprehension, as yet, is only a taking up, one with the other, of the units of the perception. In these units, so far, consequently, there is not to be found any hint of the necessity which shall effect for them connexion of existence as objects in space and time. 1 Kant adds here "in experience ;" and lie means by the latter word, not the completed process of experience when raised into objectivity and necessity by categories, but only its first stage of subjective affection, the empirical apprehension of mere units of sensation. As, however, the word immediately both precedes and follows in its other and full sense, I omit it here where it is not necessary. Kant using it, almost at the same moment, in this double way, excellently illustrates for us his habitual manner of writing. Perhaps it was " apprehension " he had in mind. TUE KRITIK 01 PURE REASON. 283 Experience, now, is a completed cognition and recog- nition of objects through perceptions of sense. It is on sense-perception becoming experience, therefore, that there is effected a relation of the units of the complex in regard of their existence mutually. The complex is regarded now, that is, not as it merely presents itself at first hand in time, but as at last it is experienced objectively in time. But time, again, is not itself perceived; the ultimate determination of existential objects in time, then, is no product of time itself, but must result from the synthesis in time. But such synthesis, so placed, can only take place through a priori notions of connexion. These notions, now, for their part, lastly, must, as such, or being a priori, bring always necessity along with them. Experience, then, can only possibly result from a recognition of necessary connexion in our various perceptions. 1 The three modi of time are persistence, sequence or succession, and simultaneity. Hence three laws of all relations of objects in time will precede experience, and as conditions, indeed, of its very possibility. These laws will determine for every object its relative existential place in regard of unity (connexion) al- ways or at any time (A being, B will be, etc.) The general principle of all three analogies depends on the necessary unity of apperception as regards every possible empirical consciousness (perception) at any and every time, and, consequently, said unity 1 Apprehension, already possessed of the a priori spectra space and time, receives into these the units of sensation special, hut as yet only confusedly and fortuitously together. Experience does more : it binds, and into a context of objects in space and time. This Unding is, evidently, not possihly due either to unperceivahle time or the sensations them- selves as such. It can only result from the categories, as a priori func- tions of synthesis into the unity of self-consciousness. Then imagination is the one common faculty of apprehension that actuates all. 284 TEXT-BOOK TO KANT : being a priori implied, on the synthetic unity (con- nexion) of all objects in respect of their relation in time. For original apperception refers itself to inner sense (the recipient of all cognitions) — refers itself a priori to the form of inner sense, i.e., to the relation of the units of the complex empirical consciousness, the one to the other, as in the form of time. Said com- plex, again, is, as a whole, to be united, but in accord- ance with its time-relations, into original appercep- tion ; for that is just what the transcendental unity of apperception a priori prescribes, inasmuch as under this unity there must stand everything that is to be a cognition of mine, specially mine — everything, con- sequently, that can be for me an object. This syn- thetic unity in the time-relation of the perceptions, a priori determined, is therefore the law, That every empirical determination in time must stand under rules of general time-determination, and the analogies of experience (which we proceed to treat) must con- stitute these rules. These analogies have this peculiarity, that they do not have in regard the objects or the synthesis of their empirical perception as it is in space, but merely their existence, or rather their relation mutually in regard of their existence. Now how something comes to be apprehended as perception (as construction in space) can be a priori determined in this way, that the rule of its synthesis can at the same time, so far, a priori give the perception (general form of the construction in space), as will be necessarily exemplified in every occurrent empirical case — that, in fact, said rule of synthesis can realize said perception (jwceptio?! being quite generally understood as construction in space). 1 1 In the above sentence " Erscheinung" means only, and probably should be, Anschauung. THE KRITIK OF PURE REASON. 285 But the existence of objects (not their mere percep- tive form as due a priori to that of space) cannot be determined or cannot come to be known a priori; and, though we might in this way (a priori) be able to reason or infer in regard to some certain existence, Ave should be quite unable, nevertheless, literally to cognise or perceive that existence. We should be quite unable, that is, to anticipate that whereby, as an actual empirical object, said existence were dis- tinguishable from others. The two previous ground-propositions, which I called the mathematical ones (as, in eifect, it is they entitle mathematics to be empirically applied), related to objects in their mere possibility as objects (indi- vidually), and instructed us how these objects, as well in their perceptive form (extension) as in the reale of their sense-matter (intension, degree), might, accord- ing to rules of a mathematical synthesis, come to be constructed. Hence, in regard of both of them, numbers may be used, and with these an object as a magnitude determined. Thus, for example, I might make up the degree of the sensations of sunlight by means, say, of 200,000 illuminations of the moon — I might, in this way, a priori, determinately give it, that is, construct it. Hence we may name these two (previous) ground-propositions constitutive. But it must be quite differently situated with those which have to bring under a priori rules the existence of objects. For, that (existence) being incapable of a priori construction, the propositions concerned will only refer to relation of existence, and avail to con- tribute, consequently, only regulative principles. In their case, therefore, there will be no question of either axioms or anticipations. But, one perception of sense, in a certain relation of time to an other (for its part, not 286 TEXT-BOOK TO KANT : necessarily determined), being given ns, they (these propositions) will authorize us a priori to say how, in said modus of time, the latter object or perception is necessarily connected with the former object or per- ception from the point of view of their existence mutually, but not what, from the point of view of extension or intension (quantity and quality) said latter perception actually is. Analogies in philosophy have not the same meaning as in mathematical usao-e. In the latter reference they are formula3 which pro- nounce the equality of two ratios, always constitu- tively, and so that three terms being given, the fourth is thereby also given, or can be constructed. An analogy in philosophy, on the other hand, is not the equality of two quantitative, but of two qualitative relations, where, from three given terms, I can a priori discover and assign, not the fourth term itself, but only a certain relation to this fourth term. Never- theless, I have certainly in this way a rule Avhereby to look for it in experience, and a mark whereby to recognise it there when found. An analogy of expe- rience, therefore, will be no more than a rule or law, by virtue of which the perceptions of sense shall be raised into the unity of experience. But in the pro- duction of these perceptions as empirical perceptions it has no power or part whatever: it is a primary proposition or principle that holds of objects (as mere phenomena of sense) not constitutively, but only regu- latively. Nor shall we be able to say more than this for the postulates of empirical reflection. These pos- tulates consider only the synthesis of pure perception (the form of a presentation to sense) ; the synthesis of sercstf-perception (the matter in the presentation to sense) ; and, lastly, the synthesis of experience (the relation or connexion in the presentations to sense). THE KRITIK OF PURE REASON. 287 They are only regulative principles, then, and to be distinguished from the mathematical ones (which are constitutive), not in the certainty — a quality that stands fast a priori in both — but still in the kind of their evidence. That is, their evidence is not intui- tive (and, consequently, not demonstrative), like that of the mathematical primary propositions. What we remarked, however, in regard to all the synthetic primary principles, and have specially to accentuate here, is this, that, not as principles of the transcendental, but only of the empirical exercise of understanding, is it, that these analogies possess, and can be proved to possess, their entire significance and validity in use; and that objects, consequently (re- garded always as mere phenomena of sense), must not be subsumed directly under the categories, but only under the schemata. For, were the objects, to which these principles are to be applied, things in themselves, it were simply impossible, a 'priori and synthetically, to make anything out in their regard. On the contrary, however, these objects are only phenomena of sense, presentations to sense, appear- ances in sense, and a complete knowledge of them can only come from possible experience. All a priori principles must at last, therefore, refer to that know- ledge. They can have in view, consequently, only the conditions of the unity of empirical cognition in the synthesis of the objects. This synthesis (so con- ditioned) is only thought in the schema of the cate- gory. The category, indeed, is what functions unity to this synthesis as a synthesis, and that without restriction of any condition of sense. We shall be authorized, therefore, by these ground-propositions, to put objects together only as in analogy with the logical and universal unity of the notions (categories) ; 288 TEXT-BOOK TO KANT: and, in the primary proposition itself, we shall avail ourselves, consequently, of the category, but, in the execution (the application to objects), set, in place of the primary proposition, the schema of the category as the key to its (the category's) use, or rather place the schema beside the category, as restricting con- dition under the name of a formula of the primary proposition. A. First Analogy. Primary Proposition of the Permanence of Substance. In all mutation of the objects of sense, substance remains (is permanent), and the quantum of these objects is, in nature, neither increased nor lessened. Proof. All objects of sense are in time, in which, as sub- strate (permanent form of inner sense), simultaneity as well as sequence can alone be conceived or represented. Time, therefore, in which all vicissitude of objects is to be thought, remains and does not itself alter, because it is that in which succession or simultaneity can be conceived or represented only as determina- tions of itself. Time, now, can, per se, not be per- ceived — strictly and properly perceived as though it were an object per se. Consequently, in the elements of sense must lie that substrate which is to relieve (exhibit) time, and by reference to which, through the relation of objects to it, all alternation or all simultaneity can be recognised. But substance, now, is the substrate of all that, as real, constitutes the existence of things, and in such manner that whatever takes place in existence, or comes to exist, can only be thought as a determination of it. That permanent TUE KRITIK OF PURE REASON. 289 element, consequently, in relation to which all time- relations of objects can alone be determined, is the substance in all the shows of sense ; it is that reale of these which, as substrate of all alteration, ever re- mains the same. Inasmuch, therefore, as substance enters not into the alteration of existence, neither can the quantum of it in nature be either lessened or increased. Our apprehension of any sensible complex is always successive, and, consequently, always in alteration. We can never determine in this way alone, then, whether this complex (that is, the units in it), as object of experience, exhibits a case of co-existence or of sequence. For that there must be presupposed to lie under the all of things, something that alwa} 7 s is, something permanent and persistent, in regard of which all alteration and all simultaneity are but so many modes (time-modes) in which it itself — this that is always permanent and persistent — exists. Only in this permanent element, therefore, are time- relations possible (for .simultaneity and succession constitute all the relations in time) ; i.e., this per- manent element is the substratum of the empirical perception of time itself, and only by reference to it is any determination as in time at all possible. As the constant correlate of all states of objects, whether those of alteration or of co-existence, time itself expresses permanence. For alteration does not affect time itself, but only the things in time (and similarly, in effect, co-existence is not a modus of time itself, not any one part of which is at once with another, but each is after the other). Did we wish to conceive time itself as an object such that all its parts were sequent the one to the other, we should, for the realization of this, require to call in another time, in which the T 290 TEXT-BOOK TO KANT: sequence were possible. By reason of a permanent element alone does existence, necessarily in different and only successive parts of time, acquire, never- theless, a magnitude, which we name duration. For in the mere succession existence is always only going and coming, and cannot be said to possess even the smallest magnitude. Without this permanent ele- ment, therefore, there is not any relation of time. Now time cannot in itself be perceived. This per- manent element, consequently, is, for the objects of sense, the substrate of all their determinations in time. This substrate, further, therefore, is the con- dition of the possibility of all synthetic unity in our perceptions, i.e., in experience ; and, by reference to this permanent element, all co-existence or alteration in time can be regarded as mere modus of the exist- ence of that which remains and persists. The per- manent element in all intimations to sense is thus the object itself, i.e., substance (phenomenon) ; while all that alters or can alter holds only of the mode in which this substance or these substances exist, only, consequently, of their mere determinations. I find that, in every age, not only the philosopher, but even men of ordinary understanding, have as- sumed this permanence as substrate of all the changes of things. I presume also that this will be always so ; only, the philosopher will continue to express himself more exactly thus : In all alterations in the world, substance persists, and only the accidents change. I find nowhere, however, even any attempt to prove this synthetic proposition. Nay, it only rarely gets the place that belongs to it, at the head of the pure and completely a priori valid laws of nature. The proposition that substance is permanent is in effect tautological. For it is just because of this per- THE KRITIK OF PURE REASON. 29 L manence that we apply the category of substance to objects, and it ought to have been proved that there is something permanent in objects, in reference to which what changes is but determination of its existence. Being synthetic and a priori, there can be no proof of this proposition dogmatically, or from notions. Neither has it been consequently thought that such propositions are only valid for possible experience, and can be proved, therefore, only by a deduction of such possibility. It is no wonder, then, that, though assumed for all experience (because its necessity for empirical cognition was felt), it has never been proved. A philosopher was asked, How much does smoke weigh ? He answered : Deduct from the weight of the wood that was burned, the weight of the ash that is left, and you have the weight of the smoke. He assumed as undeniable, therefore, that matter (sub- stance) does not perish even in fire, but only under- goes an alteration in form. Just in the same way the proposition, From nothing comes nothing, was another inference from the principle of permanence, or rather of the constant existence of the subject proper in objects. For what we call substance being that in nature that is to be the special subject of all determinations in time, all existence, as well past as future, must be determined wholly and solely in reference to it. Hence we can give an object the name of substance only because we presuppose its existence for all time; and the word durableness does not well express this, the reference it implies looking rather to the future. Nevertheless the inner necessity to endure (continue permanent, persist) is inseparably connected with the necessity to have always been, and our expressions may stand. Gigni 292 TEXT-BOOK TO KANT : de nihilo nihil, in nihilum nil posse reverti, were two propositions which the ancients invariably conjoined, and which we now sometimes mistakenly disjoin, because we suppose that they concern things in them- selves, and the former might seem to controvert the dependence of the world (even in its substance) on a First Cause. The fear is unnecessary, however; for what is spoken of are only objects of sense in the field of experience, and their unity would be im- possible if (in substance) we assumed the origination of new things. Then, namely, that would fail which can alone represent the unity of time — the identity of the substrate, which alone guarantees abiding unity throughout all change. This permanency is, at the same time, nothing but the mode in which (though knowing only affection) we represent to ourselves things. The determinations of a substance which are nothing else than its particular modes to exist, are called accidents. They are in every case real, for they concern the existence of the substance (negations are only determinations expressive of the non-being of something in the substance). If we attribute to this reale in respect of substance a particular kind of existence (e.g., motion as an accident of matter), we call it inherence, in contradistinction to subsistence in the case of substance. But this leads to many mis- takes, and our expression will be more correct and exact if we describe the accident only as the mode in which the existence of a substance is positively deter- mined. At the same time, however, it is unavoidable because of the conditions of the logical exercise of understanding, as it were, to separate what is suscep- tible of mutation in the existence of a substance (the substance itself enduring), and to consider it in rela- THE KRITIK OF PURE REASON. 293 tion to what is specially permanent and radical. Hence this category, then, will come under the head of relations, rather as condition of these, than as itself containing a relation. On this permanency depends now the legitimation of the notion of change. Origin and decease are not changes of what originates or deceases. Change is a mode in which to exist, which mode ensues on another mode to exist on the part of the same object. All that changes, therefore, persists, and only its state alters. Change, then, only concerning such deter- minations as may cease or begin, it is possible for us to use an apparently paradox expression, and say, Only what is permanent (substance) alters; what is changeable suffers no change, but only an exchange, certain determinations ceasing and others beginning. 1 Change is capable of being observed, then, only in regard of substances. Origin or decease absolutely, that is, not beino; mere determination of something permanent, cannot possibly be witnessed. It is pre- cisely this element of permanency, namely, which makes it possible for us to perceive or conceive tran- sition from one state to another, and from non-being to being. These states, therefore, are only to be em- pirically recognised as alternating determinations of what is permanent. Assume something absolutely to begin to be ; you necessarily assume also a point of time in which it was not. To what, now, would you attach this point, if not to what already is ? For an empty time which might precede, is no object of perception. But, were the thing assumed to begin joined on to things which previously were and have 1 It misleads to take Kant's use of "Wechsel" here as of universal or exclusive application. On the contrary, as we have seen a score of times already, Kant usually means by the word only change as such. 294 TEXTBOOK TO KANT : continued till this, then what begins is only a deter- mination of what has continued. And it is not otherwise with decease ; for it necessitates assumption of the empirical perception of a time no longer con- taining anything. Substances (in nature as show of sense) are the substrates of all determinations in time. The origin of some of them, and the decease of others, would even destroy the single indispensable condition of the empirical unity of time, and objects then might be in two sorts of time at once, and in these, two existences would necessarily flow side by side, which is absurd. For there is only a one time, in which the different times are, of necessity, not simultaneous but sequent. Permanency, then, is a necessary condition under which alone affections of sense are determinable as things or objects in a possible experience. But what is the empirical criterion of this necessary perma- nency, and with it of the substantiality of our per- ceptions, — on this the following articles will give us occasion to say what is required. B. Second Analogy. Primary Proposition of Time-Sequence on the Law of Causality. All changes follow from the law of the connexion of cause and effect. Proof. (That all manifestations properly sequential in time are changes, or a successive being and non-being of the determinations of substance, which itself persists, and consequently, that a being following on a non- being, or a non-being on a being, in other words, a THE KRITIK OF TÜRE REASON. 295 coming to be or a ceasing to be, is, on the part of sub- stance, only impossible, — this has been demonstrated under the proposition which immediately precedes. The same proposition might have been expressed thus : All vicissitude (succession) in the perceptions of sense is only change ; for a coming to be or a ceasing to be on the part of substance were not a change of it, inasmuch as the notion of change presupposes the same subject as existing, and consequently as persisting, with two opposed determinations. — This being premised, we proceed to the proof.) I perceive that perceptions of sense follow one an- other, i.e. j that there is a state of things at one time, the opposite of which preceded. I connect, properly, therefore, two perceptions in time. Connexion, now, is no deed of sense or the perception (general) of sense, but is the product of a synthetic act of imagination in that it determines inner-sense in regard of the time-relation. But imagination can connect said two states in two ways, either as that this shall precede that, or that this ; for time cannot itself be perceived, or so, therefore, that, in its reference, as it were em- pirically, what precedes and what follows may, in the object, be determined. I am thus only conscious that my imagination puts the one first and the other second, not that in the object the one precedes and the other follows. In other words, the mere perception of sense leaves the objective relation of the consecutive affections of sense undetermined. In order, now, that this relation should be perceived as determined, the re- lation between the two states must be so thought that it necessarily determines which state shall be neces- sarily set first, and which second ; and not reverse-wise. What notion, however, brings with it a necessity of synthetic unity can only be a category, and a category 296 TEXT-BOOK TO KANT: is no element of the perception of sense as such. That here, now, is the notion of the relation of cause and effect, in which the former determines the latter in time as its consequent, and not as something that in imagination merely might precede (or even, indeed, not at all be). Only by this, therefore, that we sub- ject the sequence of perceptions (and consequently all change) to the law of causality, is experience itself (empirical recognition of these perceptions) possible. These perceptions are themscVes, then, only possible as objects of experience by virtue of this very law. l The apprehension of the sensible complex is always successive. The perceptions of the parts of it follow on one another. Whether they also follow each other as in the object is a second point in the consideration which is not contained in the former. Of course we may name everything — every part-perception as well, so far as it is an item of consciousness — object ; but what this word shall mean in the case of the intima- tions to sense, not so far as they are objects in respect of each being such mere intimation, but so far as they represent an object, that is a matter of deeper con- sideration. So far as they are objects of conscious- ness only in that they are the mere intimations to sense (the part perceptions), there is no distinguishing them from the apprehension of them, that is, from the mere susception of them in the synthesis of imagina- tion. So far, it must be said, then, the sensible com- plex is, as in consciousness, always successively pro- duced. Were objects things in themselves, no man would be able to decide, from the succession of the part-perceptions of their complex, how it was situated 1 In the third sentence of the ahove paragraph, Kosenkranz has an " einerlei " that obviously ought to be a ziveierlei. THE KRITIK OF PURE REASON. 297 with the synthesis of this complex in the object. For always we have only to do with our own units of sense ; and how things may be in themselves (apart from the units of sense whereby they successively affect us), is wholly beyond the sphere of our cog- nition. Now, although objects to us are not things in themselves, and yet all that we can have given us to perceive, it is still necessary for me to demonstrate what that co-existent synthesis is that actually does infuse itself into the sensible complex of our con- sciousness, at the same time that this consciousness of said complex in apprehension is in all cases suc- cessive. Thus, for example, the apprehension of the sensible complex in the case of a house standing there before me is successive. Now, the question is, whether there is a succession of the complex of this house itself and in itself. No one will admit this. But, now, so soon as I consider what an object is to me in its tran- scendental meaning, then the house is not at all a thing in itself, but only an appearance in the affection of sense — a consciousness, therefore, of which the transcendental object is unknown. And the interest is to know, What do I mean by the question as to how it is situated with the synthesis of the complex in the object itself (which, of course, transcendentally, is still not a thing in itself) ? Here that which lies in successive apprehension is considered an affair of mere consciousnesses, while that, again, which appears as the result that is given to me, although it is, in reality, nothing more than a sum of these conscious- nesses, is considered the object of or before these con- sciousnesses, and with which my notion (my notion derived from the contents of apprehension) must agree. It is readily seen, now, that since truth is the agreement of cognition with its object, we can only 298 TEXT-BOOK TO KANT: ask, in such a case, after the formal conditions of empirical truth ; and that loliat appears as the result lohich is given to me, must, as compared with the succes- sive units in apprehension themselves, only then be capable of being distinguished from these as the actual object of them, when the conjunct apprehension stands under a rule which distinguishes it from every other apprehension, and necessitates a specific synthesis of the relative complex. That, now, in the general result as it is before sense, which constitutes the con- dition of this necessary rule of apprehension, is the object. 1 But, now, fairly to take up the problem before us. That something happens, i.e., that something, or some state, comes to be, which previously was not, cannot be empirically perceived, 2 unless there were a some- thing not this something, or a state not this state, preceding either. For an actual something which should ensue on a void time — an origination, that is, with no precedent condition of things — can just as little be apprehended as empty time itself. Said apprehension of an event, then, is an empirical per- 1 I desire always to make Kant's meaning not only clear, but even, perhaps, so far as allowable, acceptable to the reader. Should this reader, then, have had troubles of late — now that Kant is attempting to bring all his principles together practically in use — I beg him not to lay the whole fault at the door of the translation. Kant in the schematism, as I believe, is always now, in effect, heatedly and confusedly, fighting against a difficulty that has, quite unexpectedly, come in at last. His first inten- tion was to confine his " possibility of experience " to " space, time, and the elementary notions of the understanding ;" but, in the end, for his principal categories, a second intention was forced upon him: that of admitting into said " possibility of experience," certain main facts of empirical suggestion. The " condition " above I do not think to mean any such fact, and yet Kant is so various now, that it is just possible it may. — See note, p. 304. (The " condition " meant is just the category.) * Wahrgenommen alone means empirically perceived ; so that Kant's own words here, " empirisch icahrgcnommen" amount to the awkward tautology, empirically empirically perceived. TUE KRITIK OF TÜRE REASON. 299 ception such that it ensues on another. Inasmuch, however, as this, so far, is but a succession, or, with all synthesis of apprehension, only so situated as the complex of the house was, there is no distinction so far of the one thing from the other. But I perceive also that if, in the case of an event, I call the first state empirically perceived A, and the subsequent one B, B can in the apprehension only follow A, while, for its part, A cannot follow, but only precede B. I see, for example, a ship driving down stream. My per- ception of its position down stream follows my per- ception of its position up stream ; and it is impossible that, in the apprehension of these appearances, the ship should be first seen down stream, and after- wards again up. The order in the sequence of per- ceptions in apprehension is here, therefore, fixed, and to this order these perceptions are bound. In the previous example of the house, my perceptions in the apprehension of it could begin with the top and end with the bottom, or, equally well also, begin here and end there. They might, for that matter, quite as well also, apprehend the complex of the empirical object from right to left, or, again, from left to right. In the series of these perceptions, then, there was no fixed order — no order which necessarily prescribed where, in the apprehension, I should make my be- ginning, in order to convert its complex into the due empirical synthesis. Such necessity of rule, however, is always present in any case of an event, and the order of the consecutive perceptions (in the appre- hension of the sensible facts) is thereby rendered necessary. In this case, therefore, it is from the objective suite of the facts that I must infer the subjective suite in apprehension ; for this latter suite (of mere units in 300 TEXT-BOOK TO KANT J sense) is, as such, quite undetermined, and not dis- criminative as yet of object from object. It by itself decides nothing in regard to the synthesis of the com- plex in the object, for its order as yet is quite indifferent. The objective suite, on the other hand, will consist in the order of the perceived complex, according to which order the apprehension of what happens (as in the case of the ship) will follow what precedes by reason of a rule. Only so can I be empowered to say of the object itself, and not merely of my apprehension, that said object implies a consequence, which is as much as to signify that I can dispose my apprehension not otherwise than precisely in such and such order. By reason of the necessity of such a rule, therefore, there must lie in the antecedent of an event the con- dition to a rule such that this event must, in conse- quence thereof, always and necessarily ensue ; but, inversely, I cannot begin from the consequent, and thereby, in apprehension, determine what precedes. For from the subsequent point in time there is no going back of things to the preceding one, though referentially connecting itself, certainly, to some one or other that preceded. From a given time, again, the progression (on the part of things) to the specially following one is necessary. Accordingly, there being something that follows, I must necessarily connect it with something else that precedes, and of such a nature that it follows it in obedience to a rule, i.e., necessarily ; with the general result that the event, as what is conditioned, assuredly points to a condition, and that this condition is what determines the event. Suppose we assume an event to be preceded by nothing which it is necessitated to follow according to a rule ; then all succession of perception were THE KRITIK OF PURE REASON. 301 solely in apprehension, i.e., merely subjective, and it were not at all objectively determined which percep- tion must be properly considered antecedent and which consequent. We should have, so, only a play of intimations in consciousness that had no object in regard. In our perception, that is, no object would, so far as the relation of time is concerned, be distin- guished from another ; the reason being this, that, in our merely apprehending, the succession is always only indifferent, and there is nothing in the appear- ances determinative of them so that a certain conse- cution is thereby rendered objectively necessary. I am able to say, therefore, not that in the object two states follow the one the other, but only that one apprehension follows the other, which, of course, is something merely subjective and indeterminative of any object — such, then, that it cannot pass for the cognition of an object (even as phenomenally under- stood). 1 1 I cannot help offering to come to the reader's assistance here. There is so much repetition in all these words of Kant's that one cannot avoid suspecting that he is merely writing with the hope of gaining such time as will procure light for himself. He is now engaged in answering Hume — in demonstrating the single proposition that is fulcrum to the whole vast enterprise, the whole vast enterprise which seems already to require so little for its triumphant and definitive completion, and, do as he may, try back and back as be likes, turn up his box and shake out his principles with whatever anxiety and minuteness possible, the whole thing seems perpetually to have gone out of sight and to elude the very touch. Objects, namely, being for Kant only states of our own under the synthetic unity of a category, he requires to regard any sensation as a complex of units such that in it the order of these is in the first instance merely indifferent. Even in causality, consequently, he is obliged to assume that the facts are so. In the case of the ship, once the category has acted, we see the necessary order of the positions, and from that objective order we may infer that subjectively the positions, even while mere units of sense, were in precisely the same order. Otherwise, how- ever, Kant would have us to understand that we should have found the order in said positions, while these were mere units of sense, absolutely indifferent. I assume the reader's difficulty to concern this indifference. 302 TEXT-BOOK TO KANT : When we experience, then, something that hap- pens, we always presuppose something to precede from which it follows according to a rule. For without this I should be unable to say of the object that it follows, inasmuch as the mere succession in my apprehension, if undetermined in connexion with something that precedes, through a rule, is no warrant for a consecution in the object. Conse- quently, therefore, it is always by reason of a rule — a rule by which objects are, through a preceding- state, determined in their consecution, i.e., in the manner in which they are a happening — that I make my subjective synthesis (as in mere appre- hension) objective ; and wholly under this presup- position alone is there even the possibility of the experience of something that happens. 1 But, in the first place, how does Kant himself understand it ? It is im- possible to believe that he did not regard the empirical units of impres- sion as coming in their own order. He cannot have supposed the various beds in a garden, or the various events in a battle as witnessed, but while both beds and events were mere units of impression on one's retina, or in one's sense, to have been in their order "beliebig" in this way, that we might set any one bed or any one event, or all the beds and all the events together, into what relative position and positions we pleased. That is all too manifestly absurd. We must assume his indifference ("einerlei") to mean, then — here especially where under causality the question is, and in every stage, of nothing but secpuences — that the succession was merely a following on, until the action of the categoiy converted it into a following from. But even that understand- ing the reader will find himself unable to accept at the hands of Kant. It is all very well for Kant to say with Hume that cases of causality are merely matters of fact, and that all matters of fact are, just as such, necessarily contingent. We tell Kant, for all that, that darkness follows the shutting of tbe shutter and light its opening, and that in no stage whatever was the order in the units of impression an indifferent one : that, in fact, even from tbe first there was only a necessity of order, and that without perception of that necessity we should have had no reason for setting the facts under the rubric of causality at all. 1 So long as what concerns my knowledge of objects is confined to mere units of sensation as mere units of sensation, these are as yet only feelings in my aentiency, and, consequently, only subjective. This is the TUE KRITIK OF TÜRE REASON. 303 It seems, indeed, as though this were in contradic- tion of every observation hitherto made in respect of the process in the exercise of understanding. For, according to such observations, only through percep- tion and comparison of the concordant followings of many events on preceding states is it that we have been led to discover a rule in obedience to which certain events always follow certain states ; and, con- sequently, only thus is it that we have been prompted to form the notion of a cause. But this notion would, on such a footing, be only empirical ; and the rule, to which it gives rise, that everything that happens has a cause, would be just as contingent as experience itself. The universality and necessity of this rule were then only imputed, and would have no true apodictic validity, inasmuch as they would not be a priori established, but only through induc- tion. It is situated here, however, as with other pure a priori elements {e.g., space and time) : we derive them as evident notions from experience only because we ourselves have first of all put them into experience, and only in this way, indeed, brought experience about. Certainly the logical undeniable- ness of such peculiar rule, determinative of the series of events and due to the notion of causality, is only then possible when we have made proof of it in experience ; but a reference to it as condition of the synthetic unity of things in time was still the founcla- first step in perception to Kant and everybody else. The second step to Kant is that, these units being subsumed, through the schema, under the category, experience (objective recognition) is the result. This second step is the application of a " rule," then ; and a rule, Kant held, was no product of mere sense. Afterwards he saw that the very units in sense had their own order, and, accordingly, he was compelled (in the Prolegomena) to postulate a rule subjective which was the cue to the rule objective. 304 TEXT-BOOK TO KANT : tion of experience itself, and a priori, therefore, pre- ceded it. It is important, then, by actual example to demon- strate that, never even in experience, do we attribute (in the case of an event, where something comes to be which previously was not) the sequence to the object, and accordingly distinguish it from the sub- jective sequence of our mere apprehension, unless there be presupposed an underlying rule which com- pels us to observe this order in our perceptions rather than another. Nay, it is properly that compulsion (necessity) which alone makes possible the perception of a succession in the object. 1 We have states of consciousness in us, of which, consequently, we may be aware. But let this our consciousness be as wide, and as minute, and as exact as it may, these states always remain that, and no more than that, i.e., internal determinations of our own mind in this or that relation of time. How now do we come, in addition to that, to assume an object in or as these states of consciousness? Or, over and above the subjective reality of these states as modifications of our own sentiency, how do we 1 The doctrine here clearly is that necessity and, consequently, objec- tivity are due to the a priori elements alone. We must not, then, be misled, when we read such a sentence as this, " That, in the conjunct presentation to sense, which supplies the condition of the necessary rule for apprehension, is the object " (see both text and note, p. 298), to infer that the condition to the rule (as it were the cue to it) lies for Kant in the empirical units. We have just been told that, did we proceed so, any necessity we might arrive at, would be only one of " induction " and " imputed." That point of view, indeed, is what is dominant in Kant : The necessity we ascribe to matters of fact cannot belong a posteriori to them, but must belong a priori to us. He handles such a tangled skein, however, that the "condition" to which he refers seems now to lie in the category, and again only in a generale of empirical suggestion as in the schema. The " subjective rule " (to precede the objective one) was, undoubtedly, his last shift here. THE KRITIK OF PURE REASON. 305 come to put in place of them I know not what objec- tive reality ? An objective value cannot consist in the reference to another idea (ofthat which we would name of the object) ; for then we have just the same question over again : How does this idea, for its part, again, go beyond itself and attain an actual objective value in addition to the subjective value which is proper to it as a determination merely of mind? When we examine what new quality tlie reference to an object extends to our own mere states, and what dignity they attain thereby, we find that this reference does no more than make the conjunction of these states, in a certain way, a necessary one, and this through subjection to a rule ; or, inversely, that only because of a certain order in the time-relation of our own states of mind is it that an objective quality is imparted to them. 1 In the synthesis of perceptions, the units of the complex so constituted always follow one another. So far there is as yet not any consciousness of an object; for through this following, common as it is to all apprehensions, there is not as yet anything dis- tinguished from the rest. So soon, however, as I observe, or by anticipation assume, that there is in this following a reference to the preceding state on which state the present state ensues according to a rule, then I have something before me that so hap- pens, or that is an event, i.e., I perceive an object which I must set in time in a certain definite position — a position such that, in relation to what precedes, 1 The reader's attention ought to be particularly awake to this para- graph. We have here Kant's admission that we can know only ideas within or states of our own, and that, consecpiently, his peculiar problem is, How does that which is manifestly subjective merely become objec- tive ? Kant's answer is his whole system ; and that, also, is his answer to Hume. U 306 TEXT-BOOK TO KANT : it cannot be different. When I observe, therefore, that something happens, then there is implied in this, that, in the first instance, something precedes ; for it is just in reference to such something that what is observed to happen gets its own relative place in time — gets to exist, namely, after a time in which it was not. But its definite time-place in this relation it can only get in this way, that, in the preceding state, there is presupposed something on which it alwaj^s, and in obedience to a rule, ensues. From this it results, firstly, that I cannot invert the terms of the series — set what happens before that on which it en- sues; and, secondly, that, the state which precedes once for all being, the particular event necessarily and inevitably ensues. In this way it is that there is an order in our states, such that in it what state is present (so far as it is a become state) points to some preceding state as its correlate, possibly indeterminate as yet, at the same time that this correlate refers itself determinatingly to the other as its consequent, and thus necessarily connects it with its own self in the time-sequence. As, now, it is a necessary law of our sensibility, and, consequently, a formal condition of all our per- ceptions, that preceding time necessarily determines following time (I can get to the latter only through the former) ; so it is an indispensable laiu of empirical perception in time, that the occurrences of the past determine those of the future, and that these latter take place only so far as they are determined by the former, i.e., follow them according to a rule. For it is only by occasion of the things in time that we are able empirically to recognise this continuity in the connexion of times themselves. 1 1 In the above paragraph Rosenkranz has a "jenes " which ought to be THE KRITIK OF PURE REASON. 307 For all experience and its very possibility, under- standing is necessary, and its first respective action is, not to make the perception of an object clear, but simply possible. It effects this in this way, that it assigns the time-order to things and their existence, even in assigning to each of them, as a consequent, an a priori determinate place in time (it must follow) in regard of what (relatively) precedes. Without such determinate place, it (the effect) would not co- incide with time itself, which imposes on all its parts their places a priori. This determination of place, now, cannot be derived from the relation of things to absolute time (for absolute time is not an object of perception at all) ; but, on the contrary, things them- selves must mutually assign one another their time- places ; they must make the place of each in the time-order necessary, in this way, namely, that what happens must, in obedience to a universal rule, fol- low its preceding state. In this way there comes to be a consequence in things, which, by means of un- derstanding, effects and makes necessary precisely the same order and continuity of connexion in our pos- sible perceptions as exist a priori in the form of in- ternal perception (time), in which all our perceptions must have their places. 1 a jedes. I may remark, also, that nothing in the paragraph really coun- tenances the idea that the element of time, though all must obey its succession, contributes anything whatever to the virtue of causality. 1 This paragraph is a very unsatisfactory one in more ways than one. Grammatically, for example, it is a very glaring instance of Kant's laissez aller in composition. There is a " dieselbe " which must refer very awkwardly back for any antecedent. Then there is " Reihe of per- ceptions " which determines order, etc., in a " Reihe " as something else of a " Reihe." As regards sense, though time is declared to be not an absolute object determinative of things, and these must determine each other ; still causality is spoken of as reducing things into some coinci- dence with time itself. This may mislead us to mix up time itself in 308 TEXT-BOOK TO KANT: That something happens, therefore, is a perception belonging to a possible experience, which experience becomes actual when what happens is regarded as determinately placed in time, and, consequently, as an object which can always be found in the context of perceptions as in accordance with a rule. This rule, now, determinative of something consequentially in time, is, that, in what precedes the condition is to be found, by virtue of which the effect always (i.e., necessarily) follows. And so the proposition of a sufficient reason is the ground of possible experience, namely, of the objective recognition of events as regards their relation, consequentially, in the series of time. The proof of this proposition depends wholly on the following moments. There is required for all empirical recognition, the synthesis of the complex through imagination, and that synthesis is always successive : the impressions in imagination follow, that is, the one the other. Imagination, however, does not at all determine the order (what must pre- cede and what must follow). So far, then, the series of units in the successive impressions may be taken quite as well backwards as forwards. But if this synthesis is a synthesis of the apprehension of the complex of a given event, then the order is objectively determined, or there is an order in the succession determinative of an object, according to which order something must necessarily precede, and, it being, something else must necessarily follow. If, then, my perception is to contain the cognition of an event, of the very virtue of causality. Kant, however, does not mean that ; but only that what is causally in time is peculiarly in time. Eeciprocities are as much in time as, so to speak, causalities ; and Kant does not require to be told that. THE KRITIK OF PURE REASON. 309 something, namely, that actually happens, then this perception must be an empirical judgment in which it is thought that the sequence is determined, i.e., that it presupposes something else in time on which it ensues necessarily or according to a rule. For were it not so, and if I assumed the antecedent, and the consequent did not of necessity follow, I should be forced to regard it as only a subjective sport of my imagination, and did I conceive something objective under it, I should be obliged to name it a mere dream. Wherefore the relation of impressions (as possible perceptions), according to which relation the consequent (what happens) is, through some ante- cedent, determined in its existence in time, necessarily and according to a rule — this relation (that of cause and effect) is the condition of the objective validity of our empirical judgments, in regard of the series of perceptions ; and the condition, consequently, of the empirical truth of these, and therefore of experience. The principle of the causal relation in the succession of things, therefore, is valid for all objects of expe- rience (being under conditions of succession), for this reason, that it itself is the ground of the very possi- bility of such an experience. There comes in here a doubt, however, which must be removed. The proposition of a causal connexion amongst objects is, as now put, limited to successions of them, whereas we "find that, in the actual use of it, it applies to consociations of things, and that the cause and the effect may both be together and at one and the same time. There is a warmth in this room, for example, which is not in the air without. I look round for the cause, and I perceive a heated fire-stove. Now this object as cause is at one and the same time with the warmth as effect. There is here, therefore, 310 TEXT-BOOK TO KANT: as between cause and effect and in point of time, no relation of consecution in a serial form ; the cause and the effect, namely, are here together and at once ; and yet the rule holds good. Most natural causes, indeed, are at the same time as their effects, and any sequence on the part of the latter is only due to this, that the cause does not always realize the entire effect in a single moment. But the instant an effect is, it at once is with the causality of its cause, for if the cause had but a moment previously ceased to be, the effect never would have been at all. What is specially to be attended to here is, that the question is of the order of time, and not of its lapse: the relation remains, let there have been no lapse on the part of time at all. Time, as between the cause and its effect, may vanish, or these may simultaneously be; still the relation of the one to the other is determinable by a reference to time. The cannon-ball that lies on a cushion is, as cause, at once with the dint as its effect. Nevertheless I distinguish between them by means of the time-relation that, dynamically, connects them. For the cushion being smooth, I lay the bullet on it, and the dint follows. But if we invert the facts, and suppose the latter first, it is certain that a lead-bullet does not ensue on a dint in a cushion. Accordingly, subsequence in time is certainly the only empirical criterion of the effect in relation to its cause, which, again, precedes. The glass is the cause of the water rising higher than the level of it, though both facts simultaneously co-exist. For sup- pose from a larger vessel I take water with the glass, a result ensues. There is a change of level, namely. The water, which in the larger vessel was level, has now risen above that level at the sides of the glass. That is, whereas the surface of the water was (in the THE KRITIK OF PURE REASON. 311 larger vessel) previously horizontal, it is now (in the smaller) concave (and this is due to the glass). This causality leads to the notion of an act or action, as that again to the notion of force, on which the notion of substance follows. As I do not wish my critical business (which is occupied solely with the sources of synthetic cognition a priori) to be com- plicated with analyses (which concern, again, the mere elucidation, and not the extension, of our notions), I leave the circumstantial discussion of ana- lytical results for a future system of pure reason. Not but what such analysis is already abundantly to be found in all the current relative text-books. Never- theless, the empirical criterion of a substance, so far as it appears to manifest itself, not by the quality of permanency in objects, but better and more simply by that of action, I cannot help referring to now. Where action is, and consequently efficiency and force, there also must be substance ; and in it alone can we expect to find these conspicuous sources of phe- nomena. That sounds well ; but when we want to explain what we understand by substance, and would at the same time avoid a vicious circle, we do not find it just so easy to do so. How is it that, from the action, we immediately infer the permanency, of the agent, this which is a property so essential and pecu- liar to substance {substantia phoenomenon) ? But, after the preceding, there is no such great difficulty in answering the question, quite insoluble as it may appear in the ordinary proceeding by the analysis of notions. Action already signifies the relation of the subject of causality to the effect of it. Inasmuch, now, as every effect consists in what takes place, and consequently in what is changeable, as represented in time by the fact of succession, the ultimate subject of 312 TEXT-BOOK TO KANT: this change is, as substratum of all that changes, the element of permanency, i.e., substance. For, on the principle of causality, actions are always the first ground of any change in objects : these actions, there- fore, cannot lie in a subject that does itself change, because then there were required other actions and another subject as determinants of the change in it. From this it follows that action, as a competent em- pirical criterion, proves substantiality, without my requiring first of all to discover, through a comparison of perceptions, permanency in this criterion. Such discovery could not, in such manner, be effected, indeed, with that completeness which is required for the full and rigorous universality of the notion. For that the first subject of causality in all that comes or goes cannot itself (in the field of perceptions of sense) come and go, is a sure and certain conclusion point- ing to empirical necessity and permanency in exist- ence, and consequently to the notion of substance as phenomenal fact. When something happens, the mere happening, without reference to what happens, is even in itself an object of consideration. The transition from the non-being of a state to this state itself, even sup- posing this state to be considered apart from any sensible quality, is alone a necessity to be inquired into. A coming to be or into existence of this nature does not (as already shown under section A) affect substance (for substance is not an affair of coming to be), but only its state. It is merely change, then, and not an origination out of nothing. "When such origination is regarded as the effect of some different cause, it is called creation, which as a phenomenal fact is not possibly to be admitted, inasmuch as its very possibility would alone destroy the unity of THE KRITIK OF PURE REASON. 313 experience. At the same time, all things, if 1 regard them not as phenomena, but as things in themselves and as objects of understanding simply, are, notwith- standing that they are substances, capable of being considered dependent for their existence on some cause different from themselves; but this, again, would imply quite another use of words, and be inapplicable in regard of our things of sense as pos- sible objects of experience. How, now, there can be an alteration, simply as such ; how there is the mere possibility that, on one state in one point of time, there can follow an oppo- site state in another — of that, a priori, we have not the least idea. There is required then a knowledge of actual forces which can only empirically be given ; as, e.g., of motive forces, or, what is the same thing, certain successive manifestations (say movements) which indicate such forces. But the form in every change — the condition under which alone it can take place in regard of a previous state (the subject of the change, or the particular state that is changed, being what it may) — consequently the succession of the states themselves (the bare process) is still capable of an a priori consideration in connexion with the law of causality and the conditions of time. 1 When a substance passes from a state a into another state b, the time-point of the latter is distinguished from the time-point of the former, and follows it. The second state, again, is, as reality (in perception), distinguished from the first, as b is from zero. That is, were the difference of b from a only one of mag- 1 It must be carefully observed bere tbat I do not speak of the change of certain relations, but of that of a state. A body in uniform motion, for example, does not at all alter its state (of motion) ; but let the motion increase or decrease, and it certainly does alter its state. — K. 314 TEXT-BOOK TO KANT : nitude, the change were still a production of b— a, a result which previously was not, and in regard of which the previous state a is really =0. The question is, then, How does a thing pass from a state = a into another ==b? Between two moments of time there is always a time, and between two states in these moments there is always a difference, which difference has a certain magnitude (for all the parts of objects are always quantitative). Every transition from one state to another, therefore, takes place in a time which lies between two moments, the first of which is determinative of the state from which a thing passes, and the second of the state into which it passes. Both, then, are limits of the time of a change, and consequently of the interval between the two states. Both, as such, form part, then, of the entire change. Now, every change has a cause which, during the whole time of the change, realizes its causality. This cause, then, brings its change forward not suddenly (at once, or in a moment), but in such manner in a certain time that, as the time increases from its be- ginning in a to its completion in b, so there is also generated, through all the smaller degrees between the first and the last, the magnitude of reality (b—a). All change, consequently, is only possible through a continuous operation of causality, in which, so far as it is uniform, each step is called a moment. The change does not consist of these moments, but is generated by them as the effect of them. That, now, is the law of continuity in all change, and it founds on this, that neither time, nor any per- ception in time, consists of parts which are the smallest possible, and yet that, in the case of a change, the state of a thing passes through all these parts, as elements, into its second state. No difference THE KRITIK OF PURE REASON. 315 of the reale in perception, as no difference in the magnitude of times, is the smallest possible ; and the new state of reality increases from the first step (in which this reality was as yet not), through all the infinite degrees of it (the reality), between which degrees the differences of the one from the other are smaller than the difference between and a. What advantage this may have for the investiga- tion of nature, does not concern us here. But how such a proposition, which appears to enlarge so much our knowledge of nature, should be completely a priori possible, that very expressly demands considera- tion on our part, however much the very first sight of it may seem to prove that it is true in fact, and however much we may seem entitled to believe the question of its possibility superfluous. For there are so many unfounded pretensions to an enlargement of our knowledge by means of simple reason alone, that we must take it to us as a general principle to mis- trust such, and, in the absence of documents thoroughly justificative, to believe and accept nothing of the kind, even on the very clearest dogmatic proof. All increase of empirical cognition, as every ad- vance in perception, is nothing but an enlargement of the determination of inner sense, i.e., a progression in time, let the objects be as they may, perceptions pure or perceptions sensible. This progression in time determines everything, but is itself no further deter- mined by anything, i.e., the constituent parts of it are only given in time and through the synthesis of time, but their synthesis is not before time. Hence every transition in perception to something that follows in time is a determination of time through production of this perception and, as time is always and in all its parts a magnitude, through production 316 TEXT-BOOK TO KANT: of a perception as a magnitude throughout all its in- finitesimal degrees from zero up to the degree actual. In this way we have the possibility of a priori cog- nising a law of changes, so far as concerns their form. We anticipate only our own apprehension, the formal condition of which, inasmuch as it exists in us before entrance of any object, must certainly be capable of becoming a priori known. Time, then, is the a priori sense-condition of the possibility of a continuous progression from what is to what follows. And, in the same way, understand- ing also, by virtue of the unity of apperception, is the condition a priori of the possibility of a continuous de- termination of all positions for objects in time, through the series of causes and effects, where the earlier infal- libly involve the later, and thereby render the empirical cognition of the relations in time universally and objec- tively valid. C. Third Analogy. Primary Proposition of Simultaneity in accordance with the Law of Keciprocity or Community. All substances, so far as they may simultaneously be perceived in space, are in thoroughgoing recipro- city. Proof. Things are simultaneous when, in empirical fact, the perception of the one can follow on the perception of the other, and vice versa (which, as has been just shown under our second primary proposition, cannot possibly take place in the case of the time-consecution of perceptions). 1 Thus, I may first look at the moon 1 Any reader who fancies that he has just seen all things bound to- gether into an iron unity by causality alone, must be startled to be re- THE KRITIK OF PURE REASON. 317 and then at the earth, or, contrariwise, first at the earth and then at the moon, and just because the per- ceptions of these objects may reciprocally follow each other, do I say that they exist simultaneously. Simul- taneity, now, is the existence of the whole of a com- plex at one and the same time. But it is not possible to perceive time itself, in order to infer from the fact of things being in the same time, that the perceptions of these may reciprocally follow one another. The synthesis of imagination in apprehension would bring forward, therefore, each of the perceptions as only of such a nature that it is present in the subject when the other is absent, and so contrariwise ; but not that the objects are simultaneous — not so, that is, that when the one is the other also is, and that such is necessarily the case in order that the perceptions should be able reciprocally to follow each other. There is consequently required a notion of under- standing for the reciprocal series of the determina- tions of things existent there, apart from each other, and yet simultaneously, in order to say that the reci- procal succession of the perceptions is one that takes place in the object, and thereby demonstrate the simultaneity as objective. But now that relation of substances, in which the one is the subject of deter- minations that have their ground in the other, is the relation of influence — a relation that, where this deter- mines that and that this, is known as the relation of community or reciprocity. The simultaneity of sub- minded here, that all objects or perceptions, sensible or pure, are to Kant, as each is a Mannigfaltiges, each also a succession in time, yet that this succession is not always irreversible or causal, but may, as is emphatically declared by Kant himself also, be reversible and reciprocal, etc. Ob- viously, then, Kant has had in view heretofore the necessity of the causal relation only, and only in regard of its own appropriate empirical ante- cedents. 318 TEXT-BOOK TO KANT: stances in space, therefore, is not capable of being otherwise cognised in experience than under presup- position of their reciprocal influence the one on the other, and, consequently, just such reciprocal influence is the condition of the possibility of things themselves as objects of experience. Things are simultaneous so far as they exist at one and the same time. By what do we know, however, that they are in one and the same time? When, in the synthesis of apprehension, the order of such com- plex is indifferent — when it may proceed, that is, from A, through B, C, D, to E, or, reverse-wise, from E, through D, C, B, to A. For were this order an order of simple consecution in time that, beginning in A, concludes in E, it would be impossible to begin the apprehension of perceptions from E and go back again to A, because in that case A would be an affair of past time, and not, consequently, any longer an object of possible apprehension. Let us suppose now, that, in a complex of substances as units of sense, each were absolutely isolated, and not one among them the subject of action and reaction in regard of the others, then I say that the simul- taneity of these would be no object of a possible per- ception, and that the existence of the one could not by any path of empirical synthesis conduct to the ex- istence of the other. For, when it is considered that they would, in effect, be subjects of a separation absolute, it will be understood also that perception, still conceived capable of passing from the one to the other in time, would successively, indeed, determine the existence of each, but be wholly unable to dis- tinguish whether the one were objectively after the other or objectively along with it. There must, therefore, be something besides mere THE KRITIK OF PURE REASON. 319 existence that enables A to determine for B its place in time, and as well, at the same time, B so to deter- mine A ; for only under such a condition is it pos- sible to conceive of substances as empirically co-ex- istent. Now, only that determines for something else its place in time which for this latter is cause, or cause of its modes. Every substance, therefore, must (as it is a consequent only on account of what is determined in it) be the subject at once of the causality of certain determinations in the other, and of the effects of that other's causality in determination of its own self, i.e., they must (directly or indirectly) stand in dynamical unity, if ever the fact of their co-existence is to be possibly perceived in experience. Now, in regard of the objects of experience, every condition is necessary without which experience of these objects themselves would be impossible. It is necessary, then, for all substances in perception, so far as they are simul- taneous, to stand, one with the other, in a thorough- going community of reciprocity. The word community is, in our language, ambigu- ous, and may mean as well commercium as communio. We use it here in the former sense as importing a dynamical community without which even the local one (the communio spatii) would never be capable of being empirically perceived. It is easy to observe from our own experience that only the continuous influences in all parts of space can lead our perception from one object to another. The light, for example, that plays between our eye and the bodies in space effects a mediate community between us and them, and demonstrates thereby the simultaneity of these. Nor, again, can we empirically change our position, or rather perceive such change, without universal matter rendering it possible for us to become aware 320 TEXT-BOOK TO KANT: of our new position, at the same time that, only by means of the reciprocal influence of its elements, it is, that this matter is able to demonstrate the simultan- eity of these, and thereby the co-existence of even the remotest of them. Without community every one perception (of the phenomena in space) would be sundered from the other, and the chain of empirical cognitions, i.e., experience, would, in the case of every new object, have to begin quite afresh, without pos- sibility of any previous one being in the least con- nected with it or standing along with it in the rela- tion of time. I do not mean by this to deny the fact of a vacuum in space ; for that may exist where no perceptions reach, and where, consequently, no em- pirical observation of simultaneity takes place. In such circumstances, however, it is no object whatever for any possible experience of ours. In illustration the following may serve. All per- ceptions must, as belonging to a possible experience, stand in our mind in the community of apperception. So far, also, as the objects are to be perceived con- joined in a simultaneity of existence, they must mutually determine their places in one and the same time, and thereby constitute a whole. If this sub- jective community, now, is to rest on an objective ground, or be referred to objects as substances, the perception of these must, as ground, make the per- ception of those possible, and so vice versa, in order that consecution, which is a necessary character of perceptions as apprehensions, may not be ascribed to the objects, but that these, on the contrary, may be perceived together and at once. This implies, how- ever, a reciprocal influence, i.e., a real commercium of substances, without which the empirical relation of simultaneous existence would be impossible in expe- THE KRITIK OF PURE REASON. 321 rience. Through this commercium, objects, so far as they stand in externality the one to the other, and yet in connexion, constitute a compositum reale, and sucli convposita are possible in many ways. The three dynamical relations, therefore, from which all the rest follow, are that of Inherence, that of Conse- quence, and that of Composition. *JV Jfi. «JE. JL. w