U]3 PS* ^ o CORNELL UNIVERSITY LIBRARY BOUGHT WITH THE INCOME OF THE SAGE ENDOWMENT FUND GIVEN IN 1891 BY HENRY WILLIAMS SAGE Cornell University Library B2798.W33 P5 Philosophy of Kant explairied / bv John W olin 3 1924 029 023 848 The original of tliis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924029023848 THE PHILOSOPHY OF KAJ^T EXPLAINED PUBLISHED BV JAMES MACLEHOSE AND SONS, GLASGOW, ^nbUshers to the ambcrsttg. MACMILLAN AND CO., LTD., LONDON. New York, ' - The MacviiUan Co. Toronto^ - ' • The MacinUlan Co. of Canada. London^ • ■ - SzmJ>^m, Ha^nilton and Co- Cainhridge-, - - Bowes and Bowes. Edinbiirg-h, - • Doug-las and Foulis. Sydney, - - • Angus and Robcrtso?t. THE PHILOSOPHY OF KANT EXPLAINED BY JOHN WATSON, M.A., LL.D. PROFESSOR OF MORAL PHILOSOPHY IN THE UNIVERSITY OF QUEEN'S COLLEGE KINGSTON, CANADA, AUTHOR OF " AN OUTLINE OF PHILOSOPHY*' "hedonistic THEORIES," ETC. GLASGOW JAMES MACLEHOSE AND SONS PUBLISHERS TO THE UNIVERSITY 1908 B PS GLASGOW : PRINTED AT THE UNIVERSITY PRESS BY ROBERT MACLEH08E AKD CO. LTD. PREFATORY NOTE This book is the result of a not unsuccessful experiment in the art of teaching continued over many years, the main object of which was to provide a method by which the tendency of the student to lean upon the authority of his teacher should be counteracted. Nothing can well be more fatal to any real progress in philosophy than the habit of listening to lectures without a corresponding reaction of one's own mind. Various plans have been suggested for the avoidance of this fatal defect. The plan that I was led to adopt with more advanced students a good many years ago was to introduce them to the direct study of the Critical Philosophy through the medium of the translated passages, published under the title of " The Philosophy of Kant in Extracts from his own Writings," which I had made expressly for that purpose. In this way I was able to count upon the co-operation of the class, while the method seemed to me to have the additional advantage of recognising that the mind can only be aroused to powerful reaction when the matter upon which it is exercised is of the first rank. The main disadvantage in this method of slow and elaborate study is the amount of time it consumes, and I have therefore thought it advisable to publish the oral explanations that I have been led to give on the successive paragraphs of my translations from Kant. As these explanations were actually given in class with direct vi PREFATORY NOTE reference to the difficulties found in the text of my trans- lations, it may be hoped that they will be found instructive to others as well. Their value will depend largely upon the use to which they are put. They are not intended as a substitute for Kant's own words, but as a commentary upon them, though they will be found to contain a fairly com- plete account of his philosophy. The first part, dealing with the contents of the Critique of Pure Reason, the only really difficult treatise of Kant, is relatively fuller than the other parts ; and, indeed, I have sometimes condensed rather than expanded the statement of his ethical and aesthetic doctrines. I hope that the additional Extracts from the Critique of Judgment, together with the corresponding commentary, will enable the student to obtain a firm grasp of Kant's aesthetic doctrine. It will be obvious to anyone familiar with Dr. Edward Caird's work on Kant how much I am indebted to that comprehensive and masterly treatise. I also owe a good deal to Professor Vaihinger's invaluable Commentary. The " Historical Retrospect " is not intended to be much more than a summary of the development from Descartes to Kant as I understand it. This is the only part of the work in which I have attempted anything like criticism, that on Kant being purposely limited to exposition. I hope, however, should the present work meet with a fair share of approval, to publish a sequel, containing a discussion of Hegel's criticisms of Kant. The pages in the margin refer to the corresponding pages of the Extracts, except where new passages from Kant are quoted or summarised, when the paging of the second edition of the Critique of Pure Reason, or of Hartenstein's edition, is given. In references to the Extracts the page and paragraph are cited ; in quotations and summaries, the page alone. Thus, 8a on page 48 refers to the first whole paragraph of the Extracts, and 86 on page 49 to the second PREFATORY NOTE vii whole paragraph of the same page ; while ^ 24 on page 342 indicates the page of Hartenstein's edition of the Critique of Practical Season. In the restatements of the Critique of Aesthetic Judgment, a translation of which is given as a Supplement to these Expository Notes, only the number of the original paragraph is cited. The headings in this, as in my former work, are Kant's own. Anyone who desires to check my statements can always find the page of the original by referring to the Extracts. JOHN WATSON. University or Queen's College Kingston, Ontario, 4«A Sept., 1908. CONTENTS Historical Retrospect Critique of Pure Reason- Preface .... Introduction . Transcendental Aesthetic Section I. — Space Section II. — Time General Remarks on the Transcendental Aesthetic Transcendental Logic . Transcendental Analytic Book I. — Conceptions Chapter I. — Guiding- thread for the Discovery of the Categories .... Chapter II. — Deduction of the Categories Book II. — Judgments ..... Chapter I. — Schematism of the Categories Chapter II. — Principles of Pure Understanding 1. Axioms of Perception 2. Anticipations of Observation ,3. Analogies of Experience A. Substance B. Causality C. Community 4. Postulates of Empirical Thought General Remark on the Principles of Judgment Chapter III. — Phenomena and Noumena . PAGE 1 31 46 73' 80 91 100 106 119 119 119 133 168 169 179 180 183 189 195 200 208 213 217 220 X CONTENTS PAGE Transcendental Dialectic . . . . • 227 Book I. — Ideas .... 235 Book II. — Dialectical Conclusions 239 Chapter I. — Paralogisms . 241 Chapter II. —Antinomies . 250 Chapter III.— The Ideal . 284 Ontological Proof 292 Cosmological Proof 297 Physico-theological Proof 303 Metaphysic of JIorality. Section I. — Transition from Ordinary Moral Conceptions to the Philosophical Conception of Morality ..... 308 Section II. — Transition from Popular Moral Philosophy to the Metaphysic of Morality 315 Section III. — Transition from the Metaphysic of Morality to the Critique of Practical Reason 330 Critique of Practical Reason. Book I. — Analytic Chapter I. — Principles Chapter II.— Object . Chapter III. — Motives Book II.— Dialectic . Chapter I. — General Consideration Chapter II. — The Summum Bonum 1. Antinomy 4. Immortality 5. Existence of God 6. Postulates . . . 7. Extension of Practical Reason 8. Faith 339 339 360 364 373 373 375 376 379 380 384 386 388 CONTENTS XI Critique of Judgment. Introduction PAGE 391 Critique of Aesthetic Judgment 408 Section I. — Analytic .... 408 Book I.— The Beautiful . 408 Book II.— The Sublime . 424 A. The Mathematically Sublime 428 B. The Dynamically Sublime . 433 Deduction of Aesthetic Judgment 436 Section II. — Dialectic .... 440 Critique of Teleological Judgment 444 Section I. — Analytic .... 444 Section II. —Dialectic ... 453 Appendix on Method .... 467 Supplementary Extracts from the Critique of Judgment 474 Index 508 EXPOSITORY NOTES ON THE PHILOSOPHY OF KANT. HISTORICAL RETROSPECT. SPEAKraG generally, we may say that every new movement in philosophy is an attempt to solve some problem, which has only come to light as the result of a long process of development. That problem is usually, if not always, one which is forced to the front by the prevalence of apparently irreconcilable views of life. This at least is true of the age which gave birth to the Critical Philosophy ; for the special problem with which Kant had to deal was determined for him by the more or less independent development of two very different ways of conceiving the world, as represented in their final form on the one hand by Leibnitz and his follower Wolff and on the other hand by David Hume. And as Leibnitz is the successor of Descartes and Spinoza, while the doctrine of Hume is the logical outcome of the philosophy of Locke, or at least of one main aspect of that philosophy, it will aid us in our appreciation of the Philosophy of Kant to have before our minds a distinct idea of the method and results of these two divergent streams of thought. All the problems of philosophy may be arranged under the three heads: (1) What is the nature of the object? (2) What is the nature of the subject ? (3) What is the nature of the Unity which comprehends both object and 2 THE PHILOSOPHY OF KANT subject? With these three questions modern thought, beginning with the Eeuaissance and the Eeformation, has never ceased to occupy itself, and in attempting to answer them, it has more and more come to recognise the truth of Kant's remark that an appeal to any authority except that of reason can in the end lead only to scepticism and indifferentism. Eoused from the " dogmatic slumber " of the middle ages, men were impatient and even con- temptuous of all that bore the marks of tradition and authority. It was not that they denied the creed of the past, — for their acceptance of its content was only too facile— but that they protested against the whole medieval attitude of implicit faith in ideas that had never been subjected to free and unbiassed criticism. When therefore the men of the Eenaissance raised the cry, " Back to Greece ! " what they really meant was, " Back to the free spirit of antiquity ! " ; Now this movement of the Eenais- sance, when it expressed itself in a systematic way, followed two main directions : it was either an investigation into the world in which man carries on his life, or an enquiry into that life itself. The whole process of scientific and philosophic thought, from the beginning of the modern world to our own day, is the attempt to find a rational answer to these questions. The Eeformation, again, brought into prominence the problem of the relation between the human subject and that ultimate Unity which was conceived to be implied in both object and subject. Luther, as we know, insisted that in his religious consciousness the individual man must not be the slave of tradition and authority, but must believe only that to which his own spirit bears witness. It is true that the first repre- sentatives of the Eenaissance and the Eeformation did not recognise the full force of their rejection of external authority. Luther, for example, while he rejected the authority of the Church, fell back upon the authority of scripture, and thus violated his own principle, that HISTORICAL RETROSPECT 3 religion must be based upon a man's own direct experience. But we may lay it down as a general rule, that a principle once enunciated will in course of time work itself out to its logical issue. No better formulation, therefore, of the real spirit of the Renaissance and the Reformation can be found than in the remarkable words in which Kant characterised [a his own time. " Our age," he said, " is an age of criticism, a criticism from which nothing need hope to escape. When religion seeks to shelter itself behind its sanctity, and law behind its majesty, they justly awaken suspicion against themselves, and lose all claim to the sincere respect which reason yields only to that which has been able to bear the test of its free and open scrutiny." Just as the physical sciences must come to the study of nature without pre- conceptions, and as the philosophical sciences — including metaphysic, ethics and politics — must study the human subject in the same free and unprejudiced way, so the modern spirit demands that theology, as the attempt to determine the ultimate nature of the Universe, should be absolutely untrammeled in its search for truth. Descartes, as the first representative of the free spirit of the modern world, so far as that spirit has come to a clear consciousness of itself, naturally begins his philosophy with a protest against the uncritical acceptance of traditional or customary ideas. This is the real force of his contention that we must begin the search for truth by doubting everything that can possibly be doubted. Like Luther, however, Descartes is not perfectly true to his own principle ; for he expressly declares that the truths of religion, as formulated by the Church, must be accepted without question, and he makes a similar exception in favour of the laws of the State. It is obvious to us that these reservations are inconsistent with Descartes' own first principle. The free spirit must be entirely free, and can no more accept in unquestioning faith the dogmas of the Church, or the laws of the State, than it can admit the 4 THE PHILOSOPHY OF KANT absoluteness of customary moral or aesthetic ideas. Apart from these indefensible concessions to external authority, Descartes is, at least in intention, a true exponent of the free spirit of the modern world. Accordingly, he begins his search for truth by insisting, as Bacon also does, that the first step in knowledge consists in freeing the mind from all prepossessions, including not only the ideas that are found embedded in formulated systems, but also those natural prejudices which are common to all mankind. We must begin by doubting everything, refusing to accept anything simply because it is customarily held, or because we happen to find it in our own individual consciousness. When we come to examine the positive side of his doctrine, we find that Descartes unwittingly violates the principle which he has himself enunciated as essential to philosophy. He makes an assumption — which, indeed, did not seem to him an assumption — that proved fatal to the satisfactory solution of the problem which he aimed to solve. That assumption is that thought and being, mind and matter, are reciprocally exclusive ; an assumption which is no doubt in harmony with our ordinary ideas, but which is none the less incompatible with a rational account of the principle held to comprehend and explain both. The essence of matter is conceived as extended being, the essence of mind as thinking being, and neither is reducible to the other. Now, when two things are affirmed to differ toto coelo, there is no way in which they can consistently be reduced to the unity of a single principle. Descartes, however, makes an attempt to show that in a sense they are one. Matter and mind are no doubt " substances," and as such each is independent of the other, but they are not " substances " in the absolute sense of the term. Every material being and every thinking being is finite, and as such they are " substances " in the relative sense or as contrasted with each other, while only God can be called " Substance " without reservation. The finite presupposes HISTORICAL RETROSPECT 5 the Infinite, and only the Infijiite is absolutely independent or self-dependent. It is obvious that if only the Infinite is properly called Being or Substance, the assumed being or sub- stantiality of all finite things, whether material or mental, must in consistency be denied. If the only true " Substance " is the Infinite, we must suppose that, from an ultimate point of view, the limits between thinking and extended being disappear. This is the conclusion which Spinoza reaches by pressing home the consequences of Descartes' own statements. If the only being that can properly be said to be " Substance " is God, we must remove from finite things their seeming "substantiality." Spinoza therefore denies the independent reality of matter and mind. There can be, as he argues, only one self- complete and self-dependent Being, and this Being alone is properly entitled to be called " Substance." And as there is but one real Being or Substance, we can no longer speak of thinking beings and extended beings, but only of thought and extension, which now become " attributes " of the one and only Being. Moreover, these " attributes " must cease to be conceived as limiting each other ; for, if they limited each other, they would contradict the unity of the sole real Being. Each " attribute " must therefore be infinite. If it is objected that two infinites are unthinkable, Spinoza answers that, while this is no doubt true if each is conceived as absolutely infinite, it is not true if each is conceived as infinite in its own kind. Infinite extension and infinite thought do not exclude each other, because, as they do not come into contact at all, the one does not limit or interfere with the other. Hence the Infinite Being must be regarded as absolutely complete in each. Spinoza, however, is finally forced to admit that in itself the Infinite Being is not distinguished as thinking and extended ; in other words, he virtually abolishes the distinction between thought and extension — with the result 6 THE PHILOSOPHY OF KANT that the Infinite cannot be characterised as either thinking or extended. Thus by an inevitable logic the Infinite is reduced to pure being. And as Spinoza holds that, strictly speaking, only the Infinite is, the finite disappears and nothing knowable or even thinkable remains. But while this is the logical result of Spinoza's negative conception of the Infinite, we must also recognise that -in two ways he has made a valuable contribution to the perennial problem of philosophy : firstly, in drawing an explicit distinction between thought and being ; and, secondly, in maintaining that the Infinite is at once thinking and'extended. Until matter and mind have been explicitly distinguished and opposed there can be no true com- prehension of the Infinite. Whatever the Infinite may be, it must at least comprehend within itself thought as well as being, and conversely being must be maintained as well as thought. If it is true that being without thought disappears, it is equally true that thought without being disappears ; for thought must have an object, and there can be no object where there is no being. When, (therefore, Spinoza insists that thought and being are ! inseparable in the Infinite, he is rightly protesting against the elimination of either. Descartes had maintained that God is a purely thinking being ; but, if this were true, it is not difficult to show that He thinks nothing. When being is reduced to the extended, thought retains its reality only in so far as it is opposed to extension. If, therefore, extension is denied of the Infinite, we must also deny thought of it. Descartes of course says that thought thinks itself. However true this is, it is not true in the sense that thought thinks itself in independence of its thinking of extended being. It is only in distinction from the extended that thought can think itself as inextended, and if the extended is eliminated, the in- extended also disappears. Now, if extension is not predicable of the Infinite, that must be because extension HISTORICAL RETROSPECT 7 is an illusion. Eliminate this illusion, and we at the same time eliminate the thought of it. Thought in the Infinite can only think the real; hence it cannot think extension, if extension is an illusion ; in other words, it cannot think at all. Descartes is therefore inconsistent in denying extension of God. Spinoza on the other hand has the full courage of his convictions. Since thought and extension are inseparable, they must both be predicated of the infinite. Thought exists only as the thinking of extension, and to deny the one is to deny the other. Therefore the Infinite is held to be infinite both in thought and in extension. But Spinoza, assuming the reciprocal exclusiveness of thought and extension, is forced to deny that the Infinite in itself involves their separation. It is thus obvious that the opposition, or reciprocal exclusiveness, of thought and extension must be broken down, if the unity of God is to be preserved. If thought and being are reciprocally exclusive, both disappear, and we are left with pure being, which is pure nothing. Spinoza no doubt attempts to secure the unity of the Infinite by saying that thought and extension are each a complete expression of the Infinite, but in the end this is merely a device to cover the discrepancy between the asserted unity of the Infinite and the assumed dualism of the attributes. We may therefore say that Spinoza leaves as an unsolved problem the reconciliation of thought and extension, mind and matter. How does Spinoza make the transition from the finite to the Infinite ? He does so by what is really a process of abstraction. Finite things are usually supposed to be real in themselves. INow, in the first place, if we take the finite as external, we must on this view suppose that there is a number of mutually independent things each of which is extended. Extension constitutes the essential nature of all external things, because it is the only pro- perty which is indispensable to their existence, But 8 THE PHILOSOPHY OF KANT extension cannot mark off one thing from another. It is indeed assumed that each thing occupies a limited amount of space, and the distinction of things is supposed to be due to the amount of space occupied by each. But, if the essential property of an external thing is to occupy space, there can be no separation of one thing from another, since space is absolutely continuous. The so-called "parts" of space are in fact merely arbitrary limits, which presuppose the unbroken continuity of space. It is imagination which pictures the " parts " of space as if they were wholes, whereas thought sees that space has no "parts." There can therefore be no separate or independent external things. In the second place, the same line of argument shows that ideas do not form independent things, but that thought is absolutely continuous, or constitutes an unbroken stream of ideas. It is imagination which supposes ideas to be separable and complete in themselves. Both extension and thought are therefore continuous. But with the elimination of iinite things, external and internal, the infinity of extension and thought is established. Both constitute a single unbroken unity, in which finite things are merely limitations, constructed by our imagination, but not representing the real nature of things. The finite is merely the infinite, when the infinite is not viewed in its totality, but is arbitrarily limited. There is in short no real existence but the infinite. I In Spinoza, then, the finite is merely the infinite as it appears to the imagination. The difficulty in this doctrine is, that the infinite as purely continuous being admits of no determination, and therefore it becomes for us an absolute blank. If all determinations of extension are fictions of the imagination, obviously these have no reality whatever from the point of view of the whole. Moreover, fictions though they may be, they are not non-entities ; they have at least the reality of fictions, and therefore need explana- tion. How should fictions arise if reality excludes them ? HISTORICAL RETROSPECT 9 If thought in its real nature is continuous, why should it appear as broken up into separate ideas ? Appearance is ; why should it be, if only reality is ? Spinoza starts from the ordinary separation of finite things and finite minds from one another. This separation, however, as he main- tains, is a fiction of abstraction, which disappears when we see that there is only one continuous extension and one continuous thought. But Spinoza forgets to reconstruct the basis from which he started. If in reality there is only one extension and one thought, the fiction of a number of things and a number of minds is inexplicable. That which has no reality, from an absolute point of view does not exist : why then should it be supposed to exist ? We can understand how, in a world which is in process, there may be fictions due to the limitations of our knowledge ; but in a world where there is no process, and no limitation of any kind, fiction is something quite inexplicable. The finite in other words cannot possibly be explained by a theory which excludes it from the infinite. The only infinite which can be reconciled with the finite is an infinite which comprehends the finite without destroying it. The purely affirmative infinite of Spinoza, as excluding all determination, converts the finite into an inexplicable illusion. i-- Another difficulty which besets the doctrine of Spinoza is that it makes the human mind a mode of thought, and yet endows it with the power of knowing itself as well as the extension which is its opposite. But if the human mind is only a fiction due to the imagination, how can this fiction, this non-entity, have any knowledge whatever ? Spinoza has forgotten that in reality there is no human , mind, but only the divine mind, and that in the divine mind there is no fiction. To suppose that the divine mind is infected by negation — and fiction is negation for Spinoza — is to contradict the infinity which Spinoza ascribes to it. But even if it is admitted that there is a human 10 THE PHILOSOPHY OF KANT mind, how can it possibly comprehend the divine mind ? The human mind at the most is a mode of the one infinite thought, and as such it cannot be co-extensive with infinite thought. How can a mind which is finite, i.e. is confined within a limited circle, get beyond itself so as to embrace the infinite, i.e. to transcend its own limits ? To do so it must itself be infinite. And even if we grant that the human mind can comprehend the infinite mind, how can it comprehend infinite extension ? Extension is assumed to be beyond the mind, and therefore it cannot enter into the mind. If indeed extension and thought were not held to be reciprocally exclusive, one could understand how thought might comprehend extension ; but its com- prehension is inexplicable if they are absolute opposites. Assume that extension lies beyond the mind, and there is no possibility of explaining how it should ever get inside the mind. Nor can we escape from this difi&eulty by saying that from the point of view of God there is no opposition between extension and thought ; for in that case the opposition must be a fiction, like the fiction of finite things and finite minds. If, however, the only reality is God, how should there exist such an inexplicable fiction ? The contradictions to which the philosophy of Spinoza leads show that the radical opposition which it assumes to exist between finite and infinite cannot be admitted. The finite cannot be abolished without the abolition of the infinite. There must, then, be some way of preserving the existence of both. At the same time it has to be admitted that Spinoza has proved the impossibility of any separate reality being ascribed to the finite as such. The finite cannot exist apart from its necessary complement, and that complement is the infinite. From the point of view of knowledge this may be expressed by saying that knowledge of the finite involves knowledge of the infinite. We must, however, observe that the infinite cannot be HISTORICAL RETROSPECT 11 merely the negation of the finite, since an empty negation is for knowledge impossible. Spinoza has also shown that God cannot be defined as pure thought to the exclusion of external being, but must comprehend within himself , both thought and extension. Any true conception of God must therefore include extension as well as thought, but it must include both by transforming them into a unity of distinctions. The philosophy of Spinoza stops logically at the irre- ducible opposition of pure thought and pure extension. To bring these opposites together, it is forced to have recourse to a mediating conception. The unity beyond thought and extension can only be pure being, and pure being is no principle of unity, but the denial of unity. The first step which Spinoza takes in the direction of an inter- mediating principle is through the conception of motion as the universal mode of extension, and ideas or conscious- ness as the universal mode of thought. The " mode " or manner in which change takes place in matter is motion, the " mode " or manner in which change takes place in mind is consciousness. Thus, while extension itself is absolutely unchangeable, the parts of extension are held to be movable relatively to one another. They are, however, movable only relatively to one another, for infinite extension admits of no motion ; hence, whatever movements take place within the one infinite extension, the whole remains the same. Similarly, whatever transition there may be from one idea to another, thought as a whole remains the same, for infinite thought admits of no change. All change of ideas is thus relative. Ideas change relatively to one another, but not to the whole, which is imchanged. Further, the movements within extension correspond to the changes within consciousness, so that there never is the one without the other. There is therefore a perfect harmony between motions and ideas. From this part of Spinoza's doctrine we can see that 12 THE PHILOSOPHY OF KANT extension and thought are coming closer to each other. It is true that they never absolutely unite, their union being made impossible by the assumption of their reciprocal exclusiveness, but, half unconsciously, Spinoza is forced to find some way of mediating between these opposites. The separation between extension and thought is only made plausible by his . assumption that each is complete in itself, and is therefore not limited by the other. But this assumption is untenable, two infinite totalities being a con- tradiction in terms. To get an absolute totality the independence of each must be taken away, and then the problem arises to explain how they can form a single whole. To effect their synthesis by Spinoza's method we must say that the supposed independence of each is a fiction of the imagination. Just as the parts of space are fictions, so also is the whole of space. The whole of extension in short is not a whole, but is only a part of the totality of being, the other part being thought. If we follow out this idea to its logical conclusion, the doctrine of Spinoza will be completely transformed. If the whole of extension is a part, and the whole of thought a part, the true whole must contain all the being of extension and all the being of thought. But this is impossible unless this whole contains extension and thought in a transformed way. No real whole can be obtained by putting parts together externally. The parts in an aggregate are stUl parts, and therefore the whole is no whole. How then- can extension and thought be united ? They cannot be united and yet remain mere extension and mere thought. Extension must involve thought, and thought extension. In other words, extension must be thought in one of its forms, or rather thought must cease to be conceived as inextended and extension as unthinking. The whole must be an organic unity of extension and thought, at once self- externalising and self-unifying. As self-externalising, the whole is extension, or nature ; as self -unifying, it is thought. HISTORICAL RETROSPECT 13 A whole self-externalised and yet self-unified, a whole which is self-unified only as it is self-externalised, is the only conceivable whole. Such a whole is a rational whole, because it is the very nature of the whole at once to externalise itself and to unify itself. As nothing can fall outside of it, it is self-determinitig, and self-determination is free realisation. Thus the opposition between appearance and reality is overcome; what appears is, or is the self- , externaKsation of the one infinite being. ~ ,, ; This, of course, is not the conclusion reached by Spinoza. I The whole to which his philosophy leads does not transform extension and thought, but leaves them in insuperable opposition. Hence, what he calls the whole is not the whole, but merely the abstraction obtained by eliminating the distinction of extension and thought. It is not the iafinite but the indeterminate. No doubt Spinoza says that the whole is the completely determinate; but we . come no nearer to a true whole by the mere addition of new determinations, so long as each of these is conceived to be exclusive of the other. Spinoza tells us that the one 1 Substance has an infinite number of attributes. But an infinite number of attributes is a contradiction in terms. ^ The infinite cannot be reached by adding unit to unit. We ' are really as far from an organic unity when we have added [ a million attributes as when we have predicated only one. . The distinction between an aggregate and an organic unity 'is one of kind, not of degree. The fundamental defect in Spinoza's conception of reality is that his infinite is an aggregate of attributes, not a self-determining unity. We have seen how Spinoza unconsciously brings exten- sion and thought closer to each other, by interposing the conception of motion. There is a process within the _ extended, though it is a process which leaves the whole unaffected. There is no development, but only a move- ment within a whole which remains unchanged, all the changes falling into the parts. Nevertheless, the movement 14 THE PHILOSOPHY OF KANT of the parts requires explanation. Why should an im- movable whole be movable in its parts ? Spinoza has no answer to give except that the movements of the parts are interdependent. Even so, it must be admitted that the parts are acted upon by one another ; otherwise each part would have its own motion, and would therefore be itself a whole. Thus each part must be capable of acting and being acted upon. In other words, we must attribute to the whole of extended being a whole of energy. It is this conception of energy which forms the further mediating idea. Now, this conception is not grasped, or at least not clearly grasped, by Spinoza, but it forms the central principle in his successor Leibnitz. We have therefore to consider how far Leibnitz has avoided the defects of Spinoza. The great value of the philosophy of Leibnitz is that it exhibits the transition to the conception of an organic unity, in which justice is sought to be done at once to the reality of the individual, and the relation of all individuals in a single system ; its weakness lies in its relative failure to grasp the conception of organic unity in a satisfactory way. In the former aspect Leibnitz goes beyond the abstract individualism of Descartes and the abstract universalism of Spinoza ; in the latter aspect he never succeeds in reaching a unity which really transcends the opposition of the individual and the universal. In his conception of all forms of existence as self-activ«< substances, or " monads," Leibnitz is seeking to escape f ron^ '• that dissolution of all finite being in one single Being, whicm is the predominant aspect of the philosophy of SpinozaJ What Spinoza regards as merely " modes " of a single reality, modes which in themselves have no reality, Leibnitz converts into self-centred individuals. This leads him to deny the absolute opposition and correlation of matter and mind, which both Descartes and Spinoza had assumed. That opposition was in them bound up with the assumption HISTORICAL RETROSPECT 15 that we apprehend the true nature of existence only in so far as we set aside all the changeable determinations of things and concentrate attention upon their permanent and unchangeable determinations. These are pure extension and pure thought, which have absolutely nothing in common. Leibnitz, on the other hand, seeks to show that pure extension and pure thought are not the true nature of things ; that the true nature of things lies in their individual energy or force. Matter and mind are therefore not opposite and correlative, but every true individual is at once material and mental, soul and body ; or rather, both matter and mind are identical in being forms of energy. In his endeavour to unite these two opposite aspects of real being, the mental and the material, Leibnitz exhibits two opposite tendencies, or points of view, which he never completely reconciles. On the on e han d his face is turned towards a concrete unity,"in^so laF as he maintains that in every real existence there are two dis- tinguishable aspects, expressive of real individual energy, which, however, is Umited in degree. Thus he seems to preserve at once the independent energy of the individual and to account for the limitation of that energy in different individuals. On the ot her hand, in his endeavour to preserve the independence of the individual, he is led to eliminate as unreal all spacial and temporal relations of things, and to do violence to the qualitative differences which separate from one another the various orders of existence. Thus, for him the individual, when it is grasped by thought as it really is, is not in space or time, and every form of existence is in its essential nature the same in kind with every other. Here, therefore, we have in a new form the same tendency to abstraction as is exhibited in Descartes and Spinoza. Just as these thinkers seek for the essence of things in abstraction from all determina- tions except those of pure extension and pure thought, so Leibnitz seeks for that essence by abstraction from 16 THE PHILOSOPHY OF KANT extension and from all the differences which mark off one order of existence from another. The monad is in its true nature inextended, and every monad is the same in kind with every other, because all alike are self-active. But a force which is conceived as independent of extension could never express itself. Force is actual only as it k exhibited ia the form of motion, and motion is meaningless apart from spacial and temporal relations. To make space and time " confused ideas " that disappear in the clear light of thought, as Leibnitz does, is therefore to make force also a confused idea. If force is real only as expressed in motion, the conversion of motion into a con- fused idea makes the real expression of force a fiction which disappears when the confusion is cleared away by thought. And with the elimination of force the individuahty of the monads is likewise eliminated. We are, in fact, left with a number of empty capacities or pure abstractions. Similarly, the definition of real existence as force cuts off all that is characteristic of the various orders of existence. If the mineral, the plant, the animal, arid the man are all in their essential nature identical, we must reject all that is characteristic of any of them and seek for their true nature in what they possess in common. But what they possess in common is force or energy, and hence the hfe characteristic of plant and animal, the sensation charac- teristic of animal and man, and the self-consciousness characteristic of man, must all be regarded as unessential. We can thus see that the false method of abstraction still rules the thought of Leibnitz, as it ruled the thought of Descartes and Spinoza. He has gone beyond them by advancing to the idea of force or energy, but he falls into the mistake of supposing that to deny the ultimate truth of spacial and temporal relations is to deny as well their relative truth, and that to identify reality with energy is to say that the essential nature of things consists solely in energy. We thus see that, followed out consistently, the HISTORICAL RETROSPECT 17 Leibnitzian conception of the real independence of the monads results in the denial of all reality and in the reduction of the complexity of the world to a dead monotony. Leibnitz, however, seeks to restore the connection of things, maintaining that, separate and independent as each monad really is, it yet mirrors or represents all other things. Hence each individual has an inner life of its own, a life of perception, by which it is ideally connected with all other things. In this aspect of his theory Leibnitz is virtually seeking to restore the complexity of the world, a complexity which he had at first denied. In the conception of all monads as in their essence identical, he had reduced all orders of existence to forces ; now he raises all orders of existence to essential identity by endowing every monad with perception. But here again he is partly under the influence of the method of abstraction ; for, granting that inorganic things have a kind of perception, as he maintains they have, the distinction between the -various kinds of perception is not thereby ehminated. The Chaotic perception of the mineral, the dark perception of the plant, the obscure perception of the animal, and the clear perception of man, are all treated as essentially identical. Thus we have here again an instance of the fundamental weakness in the method of Leibnitz, namely, that qualitative differences are eliminated and reduced to a mere distinction of degree. Leibnitz virtually assumes that to be real each thing must be in essence identical with every other. Hence he denies that there can be any distinctions between things, except those which are quantitative. Now, a really organic unity does not require us to affirm the qualitative identity of all things; it does not demand that because man is perceptive, all other existences must be perceptive. We must indeed hold that all forms of being are so related that they constitute one single system and express the unity of a single principle, but this is very diflerent from saying that each form of B 18 THE PHILOSOPHY OF KANT existence must be aware of that principle. Hence we do not require to say that the mineral has perceptions, but only that in the unity of the whole the mineral is a form of the one single principle which is expressed in all things. The qualitative differences of things, which Leibnitz ostensibly denies, he virtually admits in his doctrine of the graduated scale of existence. There is a regular progress in clearness of perception from the mineral up to man, and this progress, though it is said by Leibnitz to be purely quantitative, turns out, when we examine it closely, to involve qualitative distinctions. There is no sensation in plants, no thought in animals, and, though Leibnitz speaks of these distinctions as merely quantitative, it is only by making them qualitative that he is able to distinguish between the various orders of existence at all. This is especially apparent in the distinction he draws between animals as merely sensitive and man as thinking ; for, on his own showing, it is only to thought that the true nature of the world lies open, while sensible perception so distorts the world that it presents what is really independent as if it were dependent, the discrete as if it were continuous. Obviously, therefore, only thought can properly be said to represent the world at all, while perception merely presents an illusive world that thought transcends. Thus the doctrine of Leibnitz that all beings mirror the universe is contradictory of the absolute distinction which he draws, and is forced to draw, between sense and thought. This contradiction is most clearly seen when Leibnitz comes to deal with man, who is held to be partly a sensitive and partly a thinking being ; for, while he never ceases to affirm that thought is merely clear sense, he practically makes sense and thought yield two distinct kinds of knowledge. There is a truth of fact and a truth of reason, he tells us, the former being contingent and the latter necessary ; the one ruled by the principle of sufficient reason or causality, the other by the principle of identity. Here we have one form HISTORICAL RETROSPECT 19 of that opposition between thought and perception, analysis and synthesis, which plays so great a part in the philosophy of Kant. Truths of reason or thought, according to Leibnitz, are those which are obtained by the bare analysis of concep- tions. When, e.g., we analyse the conception of a triangle, we see that it can have only three angles, that the sum of these angles must be two right angles, etc. All self-evident truths, like the truths of geometry, are therefore held by Leibnitz to be necessary. They are necessary, because they are involved in the individual object of thought which we analyse. The work of thought is therefore merely to make our ideas perfectly distinct, and the judgments which we obtain in this way are necessary judgments. Knowledge in this case must always be analysis, for nothing can ever be brought to distinct consciousness which is not already obscurely contained in it. Thus it would seem that where we cannot by analysis reach such distinctions, we have, properly speaking, no knowledge at all. Leibnitz, however, does not draw this conclusion. Truth of fact he regards as contingent, just because we cannot carry our analysis to the point where distinct conceptions are reached. The reason he gives is, that our representations of the world are complex and confused. When it is affirmed that a triangle has three angles, we have the whole object before us, and we are enabled to see that subject and predicate are absolutely identical ; but, when we are dealing with facts of experience, we have not the whole world before us, and hence we can only analyse our knowledge so far as to see that this given event depends upon another, this second event upon a third, etc. Thus, by the principle of causality, we never reach a whole which we have completely analysed into all its elements. If indeed our intelligence were infinite, all truth would be for us necessary. But such truth is for God alone. Leibnitz is here dealing with a problem which in one form or another has been discussed from the time of Plato. 20 THE PHILOSOPHY OF KANT His " truth of fact " corresponds to Plato's " opinion " (So^a) and his " truth of reason " to Plato's " knowledge " or " science " {eiriai-t'ifj.ri). We cannot say, however, that Leib- nitz has done much more than to state the problem without solving it. If knowledge proper consists in distinct con- sciousness, what are called truths of fact are not really truths at all, and thus all knowledge is reduced to truths of reason. Now these truths, as Leibnitz himself admits, do not reveal what is actual, but only what is possible. If there is a real triangle, it must have the properties ascribed to it by the mathematician, but the analysis of the idea does not prove that a real triangle actually exists. For the knowledge of reality we therefore have to fall back upon experience, and experience, as we have seen, does not give us the real, because our analysis can never be complete. Manifestly, then, reality falls outside both of reason and of experience. It becomes an unknowable realm, the mere idea of a reality that we can never reach. This is why Leibnitz concludes that in our search for causes we fall into an infinite series. We can never reach an absolutely first cause, and have thus to go beyond the series altogether and seel^ for the explanation of the world in the idea of God. This is just another way of saying that we cannot explain the world at all, but have to fall back upon the idea that its true meaning is known only to G-od — a conclusion which leads to the denial of all knowledge on our part. " When a truth is incapable of final analysis," says Leibnitz, " and cannot be demonstrated from its own reasons, but derives its final reason and certitude from the divine mind alone, it is not necessary." From this, it is manifest that all truths of fact are inexplicable. The idea of truth as it is for God is simply the idea of a knowledge that we can never have. The opposition between truths of reason and truths of fact Leibnitz seeks to transcend through his conception of final cause. Finding that we can never reach distinctness HISTORICAL RETROSPECT 21 ia the way of experience, we are forced to refer all objects of experience to God. We have, indeed, distinctness in truths of reason, but these truths do not take us beyond the realm of possibility. We must therefore distinguish, as Leibnitz contends, between metaphysical and moral necessi- ties. The truths of mathematics, e.g., are metaphysically necessary : not even God can make 2 + 2 = 5; but these truths concern only what is possible. The actual existence of things is not determined by metaphysical necessity, for there are many possible forms of reahty, and the necessary in this case involves a choice on the part of God between these possibilities. Here, therefore, we must apply the principle of sufficient reason in a different sense from that in which we apphed it in the case of the particular facts of experience. The principle is that God must choose from the infinity of possible worlds that which is best. This conception of a distinction between the possible and the actual is manifestly a transference to the divine mind of that opposition between truths of reason and truths of fact, which yet Leibnitz affirms to hold only for man. On Leibnitz's own showing, there can be for God no contrast between the thought of the real and the reality which is thought. If we were capable of bringing to distinct con- sciousness all that is obscurely involved in our ideas, there could be no opposition between what we know and what actually is : the possible and the actual would coincide. But this absolutely distinct consciousness is just what characterises the divine as distinguished from the human intellect ; hence, to suppose that there can be in the divine intelligence an opposition between the possible and the actual, is to suppose the divine intelligence to be affected by the same limitations as the human intelligence. There can, therefore, be no opposition for God between the real and the possible ; and if there is not, the very supposition of a choice between an infinite number of possible worlds is a contradiction in terms. 22 THE PHILOSOPHY OF KANT The truth is that we have here simply one form of the contradiction which runs through the whole of the philosophy of Leibnitz, the contradiction between the absolute independ- ence of all monads and their dependence upon God. This contradiction is also found in two opposite conceptions of God, which we find in Leibnitz. I In treating of finite monads he tells us that the individuality of each is bound up with the matter or passive force by which it is determined ; in fact it is by the degree of passive force that the place in the scale of existence of the individual monad ).s fixed. On the other hand, Leibnitz tells us that God is absolutely free from all " matter " or passivity, the activity of God being infinite. How then can individuality be asserted of Him ? God from this point of view becomes simply a name for the unity or harmony which subsists among all the individual monads. // On the other hand, Leibnitz insists that God is not only a monad or individual, but the " monad of monads." This can only mean that all other monads are only individuals relatively to each other, whereas relatively to God they are but modes in which the sole monad, the " monad of monads," manifests itself. From this point of view Leibnitz cannot escape from the absolute substance of Spinoza, in which all finite modes of reality are engulfed./ If " God alone," as he says, " is the primitive unity or simple originative substance of which all created and derivative monads are the productions, born as it were of the continual fulgurations of divinity from moment to moment," how can we any longer speak of the independence or substantiality of finite monads ? Full of suggestion as the philosophy of Leibnitz is, it must be pronounced a splendid failure. The attempt to combine the real individuality of things with their relations in an organic unity, and to unite the self-activity of finite things with the infinite self-activity of God, could not succeed because Leibnitz separated what he afterwards tried to unite. Yet we do injustice to Leibnitz if we do HISTORICAL RETROSPECT 23 not recognise that he had a clear conception of what the problem really is, and thus prepared the way for that deeper solution which was attempted by Kant and his idealistic successors. Meantime, the inadequacy of the various parts of his philosophy was brought into clearer light by the endeavours of his followers, who as usual fixed upon one aspect of his theory to the neglect of its other aspects. The first step in this direction was made by Wolff, a man of little speculative insight but of considerable analytic clearness of thought. The important thing in the phil- osophy of Leibnitz was for Wolff its assertion of the( individuality of every real existence. This individuahty Wolff affirms, dropping the characteristic doctrine of Leibnitz that each individual thing has an inner life in which it ideally represents all other things. Thus disappears the whole of that imaginative synthesis by which Leibnitz sought to preserve the bonds of relation between things, and so to reach a systematic unity. The mineral, in Wolff's theory, is no longer an organism but a piece of dead matter. The plant and the animal, and even the human body, are but aggregates of inert matter, the parts of which have only mechanical relations to one another. The only vestige of the ideal side of Leibnitz which Wolff retaias is the representative character of the human consciousness. But even here Wolff regards soul and body as two iade-, pendent substances, corresponding to each other, but having no real relations; he does not, like Leibnitz, conceive of body as the outer expression of the inner life of the soul. Having thus reduced the world to a mere collection of individual things, Wolff naturally falls back upon the idea of God, whom he conceives to be a purely external being who has put together all things in an orderly way, as the architect disposes the materials with which he works. Thus, just as the ideal side of the monads has disappeared, so the valuable element in Leibnitz's conception of God, the idea that he is a self-determining monad revealing himself 24 THE PHILOSOPHY OF KANT in and to other self-determining monads, has also vanished away. The whole life and unity of the world is thus converted into a lifeless mechanism, systematized by an external artificer. It is quite in accordance with the reduction of the world to a mechanical system that Wolff converts the Leibnitzian theory of teleology into an external teleology, in which final cause or purpose is conceived of, not as in individuals themselves, but only in the form of an adaptation to certain finite ends. All other forms of being are by Wolff virtually denied to have any inner life, and man alone is still held to have anything like self -activity. Other beings, according to Wolff, have been made by God for the sole purpose of ministering to the wants and well- being of man. Teleology is thus conceived of as identical with intelHgent purpose ; it is no longer as in Leibnitz immanent in things. This is the point of view, not only of Wolff, but of all the representatives in the eighteenth century of what the Germans call the Aufldarung. Spinoza regarded external teleology as the source of all bad philosophy, the lesson of philosophy being for him a recognition of the inviolability of the laws of the Eternal and their absolute indifference to the feelings of man. Wolff and his followers, making man, and indeed the individual man, the centre of all things, conceive of the world as expressly constructed for the satisfaction of human wants. Hence for them Spinoza was unintelligible. Men- delssohn could not believe that Spinoza was serious in his rejection of final causes. " If Spinoza was serious in his denial of final causes he has given utterance to the most audacious doctrine which ever proceeded from the mouth of man. Such utterances no son of earth should allow himself to make, for none of us feeds on ambrosia, but must live on bread, eat and sleep. If philosophical speculation leads to such monstrous assertions it seems to me high time that philosophers should make themselves familiar with the views of plain common sense from which they have departed HISTORICAL RETROSPECT 25 so far." In other words, common sense shows that bread has been provided to nourish the human body, and there- fore also all the other means necessary to the production of bread ; hence nature has been adjusted to human wants by a wise and benevolent God. Conceiving the world after the manner of a machine in which all the parts are externally adapted to a certain end, Wolff is naturally led to deny all miraculous interference with the course of events. As all parts of the world have been divinely arranged, each part changes in harmony with all the rest, so that each state of the world arises from its antecedent state. It is therefore morally impossible that God should interfere with the course of events. To do so would be to upset the whole order of things. Such a con- ception in fact contradicts the absolute wisdom of the divine architect and the absolute perfection of his work ; for, if divine interference is admitted, it must be because the world is imperfect, and therefore requires continual readjust- ment. This is to affirm the power of God at the expense of his wisdom. But wisdom, as Wolff maintains, is a greater perfection than power. He who has power can no doubt do what he wills, but he who has wisdom does everything for a good reason. A being of perfect intelli- gence knows everything, a being of perfect will desires nothing but the best. If in the world all proceeds by law, nature is the product of God's wisdom ; but if events occur which have no ground in the nature of things, so that all takes place miraculously, the world is the product of the power, but not of the wisdom of God. From this point of view miracles, inspiration, etc., must in consistency be denied ; hence the philosophy of Wolff is in harmony with the pure deism of the eighteenth century. Leibnitz had sought to reconcile deism with revealed religion by dis- tinguishing between what is contrary to reason and what is beyond reason ; and however inadequate his reconciliation was, it involved the truth that a purely mechanical • 26 THE PHILOSOPHY OF KANT conception of the world is untenable. Wolff leaves out the higher aspect of Leibnitz's doctrine, and thus reduces the world to a dead mechanism. When Kant came to deal with the problems of philosophy, the system of thought which he found in vogue was that of Wolff. It is true that this system had been somewhat modified in a better direction by his master Knutzen, but the main propositions of the Leibnitz- Wolffian philosophy, as it was called, remained unaffected. JSTow, we have seen that Wolff, disregarding the distinction which Leibnitz had drawn between analytic and synthetic judgments, virtually maintained that all truth is derivable from pure conceptions. He also maintained, following Leibnitz, that the distinction between perception and thought is simply one of degree, perception being nothing but confused thought. Moreover, he assumed that there were real objects given in pure conception, which yet transcended our ordinary sensible experience ; for Wolff maintained the existence of God and the immortality of the soul. Lastly, Wolff assumed that the world had been determined in its constitution by God, whom he conceived of as a supreme Architect, adjusting all things for the satisfaction of himian wants. Every one of these propositions Kant was led to question. In regard to the Wolffian doctrine that real judgments can be based upon an analysis of conceptions, he raised the difficulty that such an analysis can never take us beyond the conception itself, and consequently that true judgments in regard to real existence can never in this way be obtained. If thought is the source of all that we can know, the conclusion is that we can know nothing. Kant there- fore asked whether Leibnitz was not right in holding that the judgment of causality is synthetic, not analytic. This problem, however, proved to be one of tremendous difficulty. Kant always maintained with Leibnitz that the mere analysis of a conception could never yield a real judgment. That being so, it seemed as if all HISTORICAL RETROSPECT 27 scientific judgments were impossible. If the principle of causality is not applicable to real events, there can be no physical science. The problem, therefore, for Kant was to show how we can justify a synthetic judgment like that of causality — a judgment which affirms not only a law of thought but a law of things — consistently with the principle that pure thought can never go beyond itself. How he sought to solve this problem we shall learn by the study of his Critique of Pure Reason. In the meantime it may be pointed out that a preparation for its solution was made by Kant in an earlier treatise, The Dissertation on the Form and Principles of the Sensible and Intelligible World, in which he was led to deny the doctrine of Leibnitz and Wolff that perception is simply confused thought. In that treatise Kant maintains that perception is different in kind from thought, inasmuch as it deals with the individual, whereas thought always deals with the universal. This new view of perception obviously changes entirely the problem as it had been conceived by Leibnitz. The question which we have now to deal with takes this form: Granting that perception deals with the individual, how is it possible to base upon it judgments that are universal ? Now perception, as Kant pointed out, deals with the individual because it deals with objects in space and time. Such objects are those with which mathematics deals, and hence if we regard perception as a faculty of apprehending the individual, we can no longer regard mathematical judgments, in the manner of Leibnitz, as deriv- able purely from an analysis of our conceptions. But, on the other hand, if spacial objects and space itself, as well as temporal objects and time itself, are given to us in perception, not in thought, the universality and necessity of mathe- matical judgments seem to be in danger. How from the perception of the individual can we derive a universal judg- ment ? The difficulty of solving this problem was further , intensified for Kant, when he was, in his own words, 28 THE PHILOSOPHY OF KANT "awakened from his dogmatic slumber"' by Hume: for Hume pointed out that if we are limited to perception, or at least to sensible perception, at the very most we can only affirm in om- judgments what holds good of an object when it is actually perceived. Thus, as it seemed to Kant, the whole science of mathematics, which in his view Hume had assumed to be in^^olate, was placed in jeopardy. Then. again, the conceptions of the supersensible — the existence oi Gk>d, the freedom and immortality of man — were in the Leibnitz- Wolffian philosophy based upon pure thought But, objected Kant, if pure thought, as in itself merely analytic, never takes us beyond the conceptions with which we start, and if on the other hand real judgments, as Hume contends, are based on sensible perception, does it not follow that all conceptions of the supersensible are merely fictions of our own min ds ? Here then the choice was forced upon Kant, either to accept the sceptical conclusion of Hume, who denied all knowledge of the supei-sensible, or to find some new method by which the claims of reason to affirm the truth of these rhi-ee great ideas — Gx>d, freedom, and immortality — should lie substantiated. We see then that there are two great movements prior to Kant, with the result of which he is familiar, smd it may be said that his philosophv is an attempt to tret beyond the opposite points of view, indicated ou the one hand by Wolff as a follower of Leibnitz, and ou the other hand bv Hume as a follower of Locke. Formulating the suppositious of. common sense, Locke had held that aU veal knowledge is given in single and momentary acts of consciousness, and hence that the mind is purely receptive in its acquisition of knowledge. There are, according to him, two sources of knowledge, sensation and reflection, or inner and outer sense. The relations introduced by the spontaneous acti\"itT of thought — and thought is for Locke in all cases a facultv of relation — do not constitute but destroy realitv. Xow, if relations of thought are consistently excluded, no assistance HISTORICAL RETROSPECT 29 in the derivation of real knowledge can be obtained from the assumption of an external world or of an internal self. Locke, however, allowed himself to take advantage of both assumptions, and was thus enabled to account plausibly for the knowledge of reality, although at the expense of logical consistency. His illegitimate assumption of the relation of individual feelings to an external world was pointed out by Berkeley, his unproved supposition of their relation to an internal self by Hume. All reality has therefore to be sought in imrelated ideas of sensation and reflection, or, in the language of Hume, in " impressions of sensation " and " impressions of reflection." These indeed do not exhaust the phenomena of consciousness. Impressions are not only originally felt, but they are reproduced, and that in two ways — either in their original or in a new order. These are called by Hume respectively " ideas of memory " and " ideas of imagination." The distinction between an impres- sion and an idea cannot, on this theory, be found in the relation of the impression to an external object or to an internal self ; nor does it consist in any difference in the content of either ; and hence Hiune places it in greater or less vivacity, an impression being a more vivid, an idea a less vivid feeling, as again an idea of memory is more vivid than an idea of imagination. Whatever reality an idea is said to have it must possess in the secondary sense of being a " copy " of an impression, and hence to " impressions of sensation " and " impressions of reflection " all reality must be reducible. The scepticism of Hume thus lies ready to his hand. The only connection in the objects of knowledge which he can admit is the arbitrary order in which feelings succeed one another. There can therefore be no necessary connection either in common experience or in the sphere of mathematical or physical truth. There are no " objects " in the sense of permanent and identical substances, nor consequently can there be any necessary connection of objects or events in the way of causality. All reality is 30 THE PHILOSOPHY OF KANT thus reducible to a series of feelings as they are to the individual, and the supposed identity and causal relation of things must be explained as simply an observed uniformity in the order of succession among feelings. Since no two feelings can be identical with each other — for no feeling can repeat itself — the whole contents of what we call our know- ledge consist of an evanescent series of states of feeling in the individual consciousness. There is therefore strictly speaking no object known, and no permanent self to know it ; hence the belief in the identity and necessary connec- tion of objects is merely a natural delusion, produced by confounding the subjective necessity of custom with the supposed objective necessity of things. THE CRITIQUE OF PURE REASON. PREFACE. The philosophy of Kant is an attempt to show that it is possible to reconcile reason with itself, on the basis of a Critical Idealism, which gives satisfaction at once to our scientific and our spiritual interests.} The necessity of entering upon this " new way of ideas " first impressed' itself on Kant's mind when he discovered that, on the Wolffian assumption that reason by its unaided exercise can determine the nature of real existence, it is possible to prove, with equal logical validity, that the world has and has not a beginning, that there are and are not simple substances, that man is and is not free. Convinced as he was, and always remained, that reason cannot accept a flat contradiction, the way of escape which finally suggested itself to his mind was to challenge the whoje method which had led to such an intolerable conclusion. ', Can reason, he asked, really come in contact with things in themselves through the medium of mere conceptions ? I Are not the contradictions into which it seems to fall due to the attempt to determine the ultimate nature of the world on the basis of our limited experience ? Eeason is no doubt right in maintaining that no conditioned fonn of being can be identical with true reality, but we cannot take this consciousness of a limit in our knowledge as equivalent to a positive comprehension of existence. We must therefore distinguish between the objects of our experience and 32 THE CRITIQUE OF PURE REASON things as they are in themselves. There are certain fixed conditions under which alone we can apprehend or compre- hend objects, and these prevent us from bringing ultimate reality within the sphere of our knowledge. On this view it is intelligible that the attempt to identify the objects of our knowledge with things in themselves must result m contradiction ; a contradiction, however, which cannot be attributed to the weakness of reason, but only to the false assumption that the world as known to us is identical with the world as it absolutely is. The impossibility of basing knowledge upon pure conceptions was confirmed for Kant by Hume's sceptical argument against the necessary con- nection of events in the world of our experience. Reason,, as Kant supposed Hume to argue, cannot determine anything in regard to real objects, because its whole operation is formal. No doubt the conception of cause implies the conception of effect, but it by no means follows that we can therefore affirm the necessary connection of actual events, And as the same argument applies to mathematical con- ceptions, Hume's scepticism leads to the denial of all science, whether mathematical or physical. After long reflection Kant convinced himself that he could establish the legiti- macy of the principle of causality — as well as of other principles of the understanding not specified by Hume — in its application to objects of experience, and that in a way which freed reason in its higher exercise from the burden of self-contradiction under which it had seemed to labour. This double function of justifying the use of the prmciples of understanding employed by the sciences, and proving the illegitimacy of their application to the supersensible, is the main claun made by Kant in favour of his own Critical method, as compared with that of Dogmatism and Scepticism. Li In the Preface to the first edition of the Critique of Pure Reason Kant indicates the relations of Dogmatism, Scepticism, and Criticism. In the realm of metaphysio, the object of which is to determine the ultimate nature of PREFACE 33 existence, human reason finds itself confronted with problems, which it cannot evade, but which it is unable to solve by the method hitherto in vogue. These problems concern the freedom and immortality of man, the origin and limits of the world, and the existence of God. Some solution of them must be possible, for they spring from the very nature of reason, which demands that reahty should be self- complete and self-consistent, while yet no such reality can be brought within the sphere of our knowledge. It is easy to see how reason comes to fall into such perplexity. Naturally it starts with the conviction that reality can be known, and ia deahng with objects of experience it does not hesitate to apply to them those principles without which there can be no connected system and therefore no science. So long as it is dealing with objects of sense all goes smoothly, and its faith in the principles by which i experience is organised seems to be entirely justified. But a great shock is given to this naive confidence of reason in its own powers, when, by the extension of the very same principles to the explanation of the world as a whole, of human life, and of God, it discovers that diametrically opposite conclusions are reached. This transition from unhnuted confidence to sceptical distrust is amply illustrated in the history of philosophy. At first metaphysic is despotic, laying down principles which it regards as absolute laws of existence, and being as yet unconscious of the contradictions into which it will soon be drawn. But this period of Dogmatism is inevitably followed by a period of Scepticism. When the dogmatic assumption that reality may be determiaed by pure ideas of reason is found to result in contradiction, the sceptic, or at least the empirical sceptic, of whom Hume is for Kant the type, concludes that the supersensible is a fiction of the imagination, and falls back upon sensible experience as the sole reality — meaning by sensible experience nothing more than the manner in which impressions of sense usually, but not c 34 THE CRITIQUE OF PURE REASON necessarily, follow one another. Such a Scepticism, as Kant points out, is really a Dogmatism ; for the empirical sceptic is not content simply to deny that the supersensible can be brought within the range of knowledge, but he affirms dogmatically that there is no reality beyond the world of sense. Thus both Dogmatism and Scepticism ultimately derive their force from the same untenable assumption, the assumption that reason has no other than a purely formal function. For, while Dogmatism bases its claim to a knowledge of the supersensible upon the idea that reason comprehends reality through pure conceptions. Scepticism insists upon the contradictions into which reason seems to fall as an evidence of the emptiness of its claim to impose laws upon the world of sense. The opposite defects of Dogmatism and Scepticism can only be overcome by challenging the assumption with which both start, and enquiring whether reason may not under certain conditions have a synthetic power which has been overlooked. To answer this question, the whole faculty of reason must be subjected to the most careful scrutiny, with the object of determining whether the principles which the sciences apply, and legitimately apply, to objects of experience, do not pro- duce a mere illusion of knowledge when they are extended to the supersensible. If the supposed self-contradiction of reason can be traced to a natural but unjustifiable employ- ment of principles which have no proper application except to the world of sense, the claim of Dogmatism to a knowledge of supersensible objects will be proved to be as untenable as the denial by Scepticism that supersensible objects can be shown to exist. Criticism, then, is a systematic attempt to free reason from self-contradiction by an examination of the conditions under which it operates. It differs from dogmatism in the fact that it does not assume the ultimate validity of the method by which in our ordinary experience we introduce system into our ideas, but first enquires whether and how PREFACE 35 far that method enables us to grasp reahty as it is. It differs from scepticism in maintaining that when reason thus examines its own products, it is at least able to determine what the ultimate nature of reality is not. Kant, therefore, although he maintains the limited character of our ordinary sensible experience, does not affirm that there is any absolute limit in human reason itself ; on the contrary, he holds that, as reason is a perfect unity, it must be capable of solving all the questions which it raises. On Kant's theory, in other words, reason in its highest exercise, even when it cannot determine positively the precise character of ultimate reality, can at least deter- mine negatively what reality is not, and is therefore able to rule out all determinations which are inconsistent with the general nature of reahty. There is in his view no fundamental limit in reason itself : the limited character of our knowledge arises from the conditions under which our faculties of perception and thought operate. To the absence of Criticism, as thus under- stood, Kant traces the defects in all previous philosophies. Those systems did not begin, as they ought to have done, by first asking how far reason in its ordinary exercise brings us in contact with actual reality ; or, what is the same thing, they assumed the absolute validity of perception and thought ; with the result that the history of meta- physic is the record of a perpetual rise of new dogmatic systems, which no sooner appear than they are displaced by new forms of scepticism. And yet the human mind can never he satisfied without some solution of the problem with which metaphysic in its highest reach deals. The question therefore is, whether better results may not be obtained by the new method of Criticism. In order to explain the character of the Critical method Kant endeavours to assimilate it to the method of the special sciences. Metaphysic had claimed to be a science of the supersensible, but it had discredited itself, in Kant's 36 THE CRITIQUE OF PURE REASON \ie\v, by its history. The dogmatic method of seeTvUig to liase truth upon ■' mere ideas " liad therefore to his mind pro-\-ed its own iiiadeqiiaey. ^lay we not then, he asks, get some liiut of the true method iu metaphysic if we consider the method aetnally followed by mathematics and physics, which every one admits to be sciences ? Now, what is the method of mathematics ? At tirst, no doubt, as Kant indicates, it was merely a collection of particular observa- tions ; but at a very early period, possibly as early as Thales, by a soit of inspiration its true method was discerned, and from that moment it entered upon " the sure path of science." "What, then, was this method ? It was not, in the first place, based upon immediate sensible perception. The imivei-sal judgments of mathematics cannot he obtained by simply stating in words the character of a given sensible figure, say a triangle. In the second place, it did not consist in the b;ire analysis of such abstract conceptions as triangle. AVherein, then, did it consist i* It consisted in stilting the properties involved in the conception of a triangle, when that conception is employed in the con- struction of a particular tigtire. Thus the true method involves some form of combination of conception and per- ception. Mathematics is not based upon mere sensible apprehension, for then no univei-sal judgment could lie legitimately made; nor is it based upon pure conception, for then it would have no reference to reality, lu short, mathematics, to employ Kanfs technical term, is based upon schemata, a schema being neither an abstiact idea nor a mere sensible perception, but a universal way of deter- mining the sensible. •21) ^^imilarly experimental physics, in Kant's \iew, exhibits a combination of conception and perception. Even in the case of special obser\ation or induction the metliod of physics is not simply to apprehend particular objects by means of sensible perception. No collection of perceptions of sense, as Kant implies, would have enabled Galileo or PREFACE 37 Torrieelii to obtain their results. In the discovery of particular laws of nature, not less than in the establishment of mathematical conclusions, the mind must bring a certain conception with it and, so to speak, reconstruct nature, and only then is it able to interpret the facts of observation. The method of physics is thus at bottom the same as the method of mathematics, allowing for the difference in character of the objects to which it is applied. The sciences of mathematics and physics, then, are not merely the apprehension of objects by means of sensible experience, but imply that the mind contributes something to the determination of the object. We have therefore to ask whether, supplied with this clue, we may not be able to construct a real science of metaphysic. The problem, as Kant here puts it, is to explain how we can have a priori judgments, which yet are not fictions, but judgments expressing the actual nature of things. Now, previous metaphysical specidation had always ignored the element contributed by the mind in the determination of objects, and assumed that knowledge must consist in the apprehen- sion of objects, the nature of which is determined inde- pendently of the mind which apprehends them. Since this method has invariably resulted in disaster, let us now, following the hint of mathematics and physics, ask whether the ^ature of tJh^_objectjiD5s jipt in ^OEae_wa,y_depend upon the mind. of the knowing subject. This new point of view, if it turns out to be true, may be compared with the revolu- tion effected by Copernicus in astronomy. Just as the old Ptolemaic astronomy assumed that the spectator standing on the earth witnessed the movement of the sun, so the old metaphysic assumes that the knowing subject is passive in the presence of the object. And just as Copernicus inverted the old view of astronomy, showing that it is the spectator carried along by the earth who moves, while the sun is at rest ; so this new metaphysical view will maintain that to the character of the knowing subject is due the 38 THE CRITIQUE OF PUKE REASON apparent iiatme of tlie object. Let us tiy this new metJiod first in relation to perception; that is to say. in relation to the direct experience of indi%adual sensible objects, as dis- tingnished from a knowledge of these objects as a conneot«l system. If the mind in perception is. as Locke main- tained, purely piassive, it is quite obvious that no a priori judgments in regard to objects are possible. -( prion judg- ments are concerned, not with what immediately presents itself at a gi\-en moment, but with what nmst always present itself. Now, it is plain that according to Locke's view we are at tlie very most limited to imii\iduad judgments — such judgments. (•.y oxternal things, but a certain activity in our faculty of knowledge, on occasion of the impressions on our senses. This is a view which may be accepted by any one, whatever his explana- tion of the faculty of knowledge may be. The empirical explanation, as we find it in Locke, is that our knowledge of external objects is obtained entirely through "ideas of sensation." This doctrine, as he inalvos clear in the following paragraph, Kant is unable to accept ; but, agreeing with Locke that experience involves impressions of sense, which the understanding subjects to a process of com])arison, distinction and combination, he is able to characterise experienc;e in ttnins which even Locke could have accepted. Such phrases, however, as " aroused to activity " and " the raw material of our sensible im- pressions " convey a subtle suggestion ( if the doctrine which Kant immediately goes (in to indicate, tlrat experi- ence cannot be resolved into mere affections of sense, but involves an element which is contributed by the mind itself. ia The first form of knowledge, then, is that consciousness of a world of sensible objects which we call " experience." On occasion of impressions of sense our faculty of know- ledge is called into activity, with the result that we come somehow to have an actual knowledge of sensible objects and of their relations to one a.iiother. Finding that all knowledge begins with experience, it is natural enough to suppose, with Locke, that there is no other source of know- ledge than experience. But this conclusion is not warranted by the facts. Granting that our knowledge begins with experience, i.e., with the apprehc^nsion of individual objects in space and time and their connection with one another, we must not assume that it contains no ehnnent which has a different origin from that of sensible impressions • for, while it is certainly true that experience is of individual PURE AND EMPIRICAL KNOWLEDGE 49 things and their connection, not only may there be a kind of knowledge which is entirely independent of experience, but even in experience itself there may be an element which is not given to the mind but contributed by the mind. Here, therefore, Kant expressly suggests his own , view, that impressions of sense do not of themselves constitute knowledge, biit that in order to obtain any knowledge whatever the mind must contribute something peciiliar to itself. If it is asked why we ordinarily suppose perception to be a simple operation — in other words, why we assume that it consists in the immediate apprehension of an object already constituted and independent of the mind — the answer is, that the intimate fusion of the two elements, viz., sensible impressions and the element supplied by the mind, is sufficient to explain why it is only after carefiU reflection upon knowledge that we are enabled to see that this complexity actually exists. 8 J We must therefore begin our critical enquiry by asking whether there is not in experience an element which is not derived from experience, and, indeed, whether we are not capable of a kind of knowledge which cannot be obtained through the medium of sensible impressions at all. Know- ledge of this character we shall call a prim-i, to distinguish it from that which, as derived from sensible experience, is called a posteriori. 8 c To prevent confusion, however, we must define more precisely the sense in which a priori knowledge is employed in the critical philosophy. In the popular sense of the term any piece of knowledge, however obtained, which enables us to anticipate what will take place under certain circumstances, is called a priori; as when we say that a man might have known a pric»-i that his house woidd fall, when he undermined its foundations. It is not in this relative sense that we propose to speak of a priori know- 9 a ledge. For us a priori will mean what is absolutely a priori, i.e., it wiU consist of judgments which are D 50 THE CRITIQUE OF PURE REASON independent of all sensible experience, in the sense that special empirical knowledge is not their presupposition; such judgments, e.g., as those of Substantiality, Causality, and Eeciprocity. Setting aside, then, the a priori in the popular sense of that which is relatively a priori, we shall speak of the a priori only in the strict sense of that which is absolutely independent of all experience. This a priori knowledge may be further divided into (1) pure, and (2) mixed. A priori knowledge is pure when there is no particular sensible element whatever involved in the object of which there is a priori knowledge. Thus, the judgment, " A triangle has three angles," is a pure a priori judgment, because a triangle is a determination of space, and space, as Kant afterwards maintains, is a pure form of perception. On the other hand, the judgment, "Every change has a cause," is also a priori, but it is not pure or unmixed, because we have no knowledge of change except in the case of sensible objects, and the knowledge of sensible objects necessarily implies some element of sensible percep- tion. It must be noted, however, that Kant would include states of consciousness, so far as these are in time, within the realm of the sensible. 2. Science and Common Sense Contain a Priori Knowledge. 95 As we have seen, the Critique of Pure Eeason has to consider two main questions : (1) Is there an a priori element in experience ? and (2) Is there a priori knowledge which entirely transcends experience ? The latter question, how- ever, is set aside in the meantime, since an answer to it presupposes the solution of the former question. Now, if there is an a priori element in experience, it is not difficult to see that, unlike the a posteriori element, it must be such that it cannot be other than it is. Hence, firstly, every judgment which can only be thought as necessary A PRIORI KNOWLEDGE 51 is a priori. This is true even of those judgments that are only " relatively a priori^ such as the judgment that " he who undermines his house will cause it to fall," which is necessary under presupposition of the empirical law of gravitation. What Kant has mainly in his mind's eye, however, are " absolutely a priori " judgments — those which are either in themselves necessary, such as the definitions of Euclid, or are derived from those which are in themselves necessary, as, e.g., the proposi- tion that "the angles of a triangle are equal to two right angles." Necessary judgments like these are obviously different in character from empirical judgments. Experi- ence can tell us that a thing is so and so, but not that it must be so ; and, therefore, empirical judgments are always contingent. Secondly, a priori judgments are not only necessary, but they are in the strictest sense i mive rsal, because they are true under all possible circumstances ; whereas empirical judgments, as based upon individual per- ceptions, can never take us beyond the limited number of instances actually observed. A priori judgments, then, are N^lways at once necessary and universal. These two marks, indeed, imply each other ; for, whatever is necessary, as true under all possible circumstances, must be true uni- versally, and whatever is universally true under all possible circumstances, must be true necessarily. a Are there, then, any such a priori judgments, any judg- ments that in the strict sense are necessary and universal ? ^ Kant's answer is that both in our common-sense knowledge and in our scientific knowledge there will be found instances of such judgments. Thus, to take an example from the sciences, all the propositions in pure mathematics are iRant calls these judgments "pure a ^noW judgments" {reine Urtheile a priori), using the term " pure " here in the sense of " that which is independent of anything empirical," not in the sense of "unmixed with anything empirical." Hence the contradiction is only verbal, when, as an instance of a "pure" a priori judgment, he cites the principle of causality, which he had earlier (B 3) characterized as not "pure." 52 THE CRITIQUE OF PURE REASON obviously a priori, since they do not affirm what is true only at the moment or under special conditions, but what 5 is true always and under all conditions. An instance, again, of an a priori judgment of common sense is the proposition that " every change has a cause." Such a judg- ment is manifestly essential to the very existence of ordinary knowledge, and if we deny this principle we at the same time abolish all such knowledge. Thus, if the principle of causality were simply, as Hume supposed, the tendency to feign a connection of events after repeated particular per- ceptions of sequence, the necessity and universality rightly assumed to be characteristic of that principle would dis- appear. Hume's doctrine is, therefore, equivalent to a denial of all knowledge. For, it is obvious, that if the principle of causation is reducible to " custom," we cannot have any " experience " in the sense of a connected system of perceptions : we should in fact have merely an aggregate of particular and therefore contingent judgments. Kant's argument in this paragraph is obviously of a tentative character. He makes as yet no attempt to prove that there are a priori judgments, which determine the nature of objects of experience, but merely starts from the assumption that there is connected experience and science. We may even say that in his first mind it did not seem necessary to him to prove that there are sciences.^ It is a simple matter of fact, he says, that mathematics and physics exist, and since they exist we must be prepared to admit what is implied in their existence. Now, it is implied in their existence that there are a priori judg- ments, which determine the general character of objects of experience. But subsequently, when Kant came to enquire how the special sciences are possible, he found it necessary to show, not only that if there are sciences there are a priori judgments, but that there are sciences containing a priori judgments. One thing that forced this ' See above, p. 36. A PRIORI KNOWLEDGE 53 step upon him was that when the judgments employed by the sciences are assumed to be absolute, that is, to apply to all possible existence, it turns out that reason falls into contradiction with itself. When the principle of causality, e.g., is supposed to determine the nature of absolute reality, it conies mto conflict with the higher interests of nian.^ Since these principles thus show that they are not absolute, a doubt arises whether they can be regarded as in any sense necessary and universal. When Kant became aware of this difficulty, the problem assumed for hiin the new form ; Granting the existence of' experience, in a sense that cannot be denied, viz., as containing at least individual judgments of perception, can it be shown that it involves a priori judgments, judgments which are universal and necessary ? Here, in the Introduction, however, Kapt starts from experience in the higher sense of the term, in which it means a con- nected system of perceptions, and he has no great difficulty in showing that if there is such a system, there must be a priori judgments ; in other words, that a system of experience and a system of a priori judgments in regard to objects of experience practically mean the same thing. Admit that there is a system of experience at all, and it must be admitted that there are a priori principles by which particulars are connected. Were there no such principles, the whole of our knowledge would be con- tingent, uncertain and limited, since we could never reach a conclusion wider than our limited number of observations warranted. The whole basis of the physical sciences would therefore be removed. Thus the denial of a priori principles inevitably leads to Scepticism. Not only is this true in regard to the physical sciences, but even the mathematical sciences must, with that denial, succumb to the attack of empirical scepticism ; a conclu- sion from which, as Kant thinks, even Hume would have 1 See above, p. 31, of. p. 44. 54 THE CRITIQUE OF PURE REASON shrunk, had he clearly seen that his attack on the principle of causality applies with equal force to the principles of mathematics. And obviously common sense cannot fare better than the sciences ; for, with the removal of a priori principles of knowledge, the whole theoretical use of reason is involved in doubt. No stable edifice of know- ledge can be buOt upon the foundation of empirical observation, which can never yield that necessity and universality without which there can be no "cosmos of experience." . 3. A Science is Needed to DeterTnine the Possibility, the Principles, and the Extent of all a priori Knowledge. 11 a Granting the existence of the special sciences, or even of ordinary knowledge as a connected system of experiences, there is no difficulty in showing that a priori judgments in regard to objects of experience have the value of knowledge. But this does not enable us to determine whether we have knowledge of objects which lie beyond experience ; and therefore a special investigation is needed, the object of which is to consider whether we have a priori judgments applicable to ultimate reality, or whether such judgments are impossible. The science which carries on this investigation is metaphysic, or at least the main aim of metaphysic is to determine whether we^can have knowledge of the supersensible, and, if not, whether there is any ground for our affirmation of its reality. The problem of metaphysic is to Kant's mind more im- portant, as he distinctly states (B 7), than the determination of the a priori conditions of experience ; for it concerns the existence of objects, the reality or imreality of which will affect all the higher interests of man. The existence of God, the reality of freedom and immortality, are problems upon the solution of which will depend ultimately our whole view of human life. It is important to observe A PRIORI KNOWLEDGE 55 that it is in this ultimate problem that Kant is mainly interested, because it has been maintained by some writers, such as Huxley, that the whole value of the critical philosophy consists in its limitation of knowledge to ex- perience. This at least is not Kant's own view. For him the enquiry into the nature of experience is merely a step towards the solution of the higher problem, and it is with this higher problem that metaphysic in the narrower sense deals, though in the wider sense it com- prehends the enquiry into the principles and extent of all a priori knowledge. 1 6 Just because that problem is of supreme importance one might have expected that infinite pains would have been taken to answer it correctly by first enquiring into the foundation, real or supposed, upon which it rests. In point of fact, however, it has been blindly assumed that we are able to determine a priori, independently of all experience, the existence and nature of the supersensible. There is one obvious reason why men have been lulled into a false security. Mathematics, though it evidently contains a priori judgments, has so long ago attained to the dignity of an assured science, that it is not unnatural to suppose that we are capable of forming a priori judgments in regard to objects of a totally different nature from those with which mathematics is concerned. As these objects are beyond the range of experience, there is nothing in experience to give us pause, and the instinct of reason to seek for completeness blinds us to the fictitious character of our constructions so long as we are careful to avoid direct logical contradictions. The example of mathematics, however, is quite misleading. What has been overlooked is that the a priori judgments which it employs are not independent of perception ; in other words, that they are not based upon pure conceptions or thoughts, but have a meaning only because their objects are determinations of individual things, though not of individual things as dealt with in the concrete sciences. 56 THE CRITIQUE OF PURE REASON Mathematics certainly advances without the aid of particular impressions of sense. Its judgments are necessary and universal, and therefore do not need to wait upon sensible experience but anticipate it. It was, therefore, not unnatural to overlook the fact, that, though mathematics is not dependent upon particular sensible experiences, it is depend- ent upon the pure forms of space and time, without which there can be no sensible experiences. Misled by the instance of mathematics, it has been supposed that we can build up an a priori science upon the basis of pure conceptions, or, as Plato called them, pure ideas. The real ■*■ truth, however, is that no progress can be made by -the manipulation of mere ideas. One reason why it has been supposed that ideas in themselves are a guarantee of objec- tive reality is that we can undoubtedly derive a priori judgments, provided that these judgments are not synthetic but analytic. We can, in other words, obtain by analysis a priori judgments, which express in an explicit form what is obscurely involved in the conceptions that we already possess. A great deal of our a priori knowledge is of this character, and therefore we are apt to suppose that judgments of quite a different kind — ^judgments which are not due to the mere analysis of our conceptions, but go beyond these to affirm something in regard to the nature of objective existence — are not less certain than those which simply analyze our conceptions. "We have therefore to distinguish clearly between analytic and synthetic judgments. 4. The Distinction between Analytic and Synthetic Judgments. 13 a The problem of the Critique of Pure Eeasou, as pre- sented in the Introduction, is, as we have seen, to explain the conditions under which a priori judgments are employed in determining objects of experience, as a preparation for the ultimate problem, whether such judgments can leo-iti- mately be extended to supersensible objects. Kant finds ANALYTIC AND SYNTHETIC JUDGMENTS 57 it necessary, therefore, to distinguish between judgments which yield new knowledge and those which merely make us more clearly conscious of what we already know. Every' judgment consists in some form of relation between subject and predicate, but it is important to determine the precise character of this relation, because upon it depends the funda- mental distinction in content of analytic and synthetic judgmentsj This distinction, as it must be carefully ob- served, does not correspond to that between a priori and a posteriori ; for, though all a posteriori judgments are synthetic, it does not follow that all a priori judgments must be analytic : on the contrary, the main problem of metaphysic is to consider how there can be a priori\ judgments, which yet are synthetic. ' Stated quite generally, an analytic judgment is one which expresses in the pre- dicate what is already, though perhaps obscurely, implied in the subject ; a synthetic judgment is one which adds some- thing to what is already involved in the conception expressed by the subject. In the affirmative analytic judgment the identity, or rather partial identity, of subject and predicate is asserted, and therefore nothing is added to the conception of the subject, the only change effected being the explicit relation of that conception to the logical species implied in it. To obtain the predicate, it is not necessary to go beyond the conception of the subject, and therefore the judgment is formed independently of sensible experience. By an appli- cation of the logical principle of contradiction, which tells us that nothing can possess a predicate which contradicts it, we directly discern that the predicate is involved in the subject. All analytical judgments, depending as they do upon the exercise of abstract thought, are therefore a priori. More- over, as they only state vdiat is involved in a conception, they tell us nothing as to the nature of reality, The judgment " Pegasus is a winged horse," is true in the logical sense, though to the subject nothing real corresponds. In contrast to analytic judgments stand synthetic judgments, whether 58 THE CRITIQUE OF PURE REASON a priori or empirical, pn these the predicate is not implied in the mere conception of the subject, though it is no doubt connected with it, and therefore thought does not establish the relation of predicate to subject by an immediate application of the principle of contradiction. No judgment can be synthetic unless it is the expression of an actual addition to knowledge, in which the predicate is obtained by going beyond the conception of the subject. From what has been said it is obvious that all analytical judgments are a priori, and therefore are logically necessary and universal. This is true both of judgments which obtain their content from sensible experience and those the content of which is derived from reason itself. Thus, in the judgment, " Gold is a yellow metal," the knowledge of gold and of its properties is obtained from sensible experi- ence, but, when it is obtained, the judgment is a priori, because it simply expresses what is involved in the con- ception " gold." On the other hand, the judgment, " every cause has an effect," is analytic, but its content is a priori, since it is not based upon sensible experience, but upon the nature of thought itself. As an instance of an analytic judgment, Kant here gives the proposition, " Body is extended " ; as an instance of a synthetic judgment, " Body has weight." Both of these judgments are based upon sensible experience. " Body is extended " is said to be an analytic judgment, because the conception " body " involves the conception extension, though no doubt the connection of " body " and " extension " was originally learned from experience. " Body has weight " is called a synthetic judgment, because, to obtain a further determination of the subject, we must go to experience. The distinction in the case of such judgments is evidently not absolute, since the judgment, "Body is extended," as Kant himself says, was originally derived from experience and is therefore synthetic ; and, on the other hand, " Body has weight" must be regarded as an analytic judgment ANALYTIC AND SYNTHETIC JUDGMENTS 59 when once it has become a part of the mind's mental furniture, since the conception " body " then includes the conception " weight " as a logical part of itself. Though Kant illustrates the distinction between these two kinds of judgment by selecting instances from judgments of experi- ence, his main interest is not in such judgments but in those which are at once a priori and synthetic. a (He therefore clears the ground by first pointing out that all judgments of experience are obviously synthetic. How he comes to draw the distinction between analytic and synthetic judgments, and to refer the latter to experience, may be best understood by considering his relation to the Wolffian philosophy on the one hand and to the empirical philosophy of Locke on the other. Leibnitz, as we have seen,i made a distinction between truths of reason and truths of fact. The latter he regarded ; as contingent, because based upon incomplete knowledge ; the former as necessary, because based upon the principle of identity and obtainable by analysis. Truths of fact, again, so far as they imply a relation of things to one another, he referred to the principle of sufficient reason or causality. Now, Wolff sought to reduce all truths to identity, and therefore to abolish the distinction between truths of reason and truths of fact. It soon became apparent, however, that if all truth is reducible to identity, and identity is based upon the abstract operation of thought, it will be impossible to obtain judgments expressing the real nature of things. The analysis of conceptions will not take us beyond thought to reality. This fundamental defect in the philosophy of Wolff was made still clearer to Kant by Hume's development of the philosophy of Locke ; for Hume pointed out that concrete facts of experience must rest upon sensible perception, and cannot possibly be derived from pure thought; while, on the other hand, what is given in perception can never 'Above, p. 18. 60 THE CRITIQUE OF PURE REASON reach the dignity of universal and necessary truth.^ This \ is the point at which Kant takes up the problem. Wliought, as he assumes, is in itself purely analytic, its - principle being that of identity. But this means that thought cannot go beyond itself to determine the concrete facts of knowledge. Thus we seem to be driven back upon sensible experience, and indeed Kant here assumes that we can obtain judgments which go beyond the analysis of thought by the aid of sensible experience. On the other hand, under the influence of Hume, he also came to see that judgments based upon sensible experience must be contingent. Thus, though it seems possible in this way to explain how we can have synthetic judgments, we have not yet explained how we can have a priori synthetic judgments. But, obviously, unless we can show that there are a priori synthetic judgments, we cannot account for science or the system of experience ; and, what in Kant's mind is still more important, the ultimate problems of metaphysic are insoluble unless we can prove the possibiUty of a priori synthetic judgments in regard to the super- sensible. No real proof of the existence of God, e.g., can be obtained by an analysis of the conception, or, as Kant afterwards calls it, the Idea of God. No doubt we can in this way obtain such judgments as, " God is omnipotent," " God is omnipresent " ; but these judgments only express what is involved in our conception of God, and therefore they do not enable us to go beyond our conception and affirm the actual existence of God. No " existential " judgment, as Kant elsewhere says, i.e., no judo-ment affirming the existence of an object corresponding to our thought, can be based upon pure conceptions or ideas. We have, therefore, to ask how the a priori synthetic judgments of experience, which, as Kant in his first mind thinks, we are entitled to assume, can be explained, — an investigation which is a necessary preparation for a dis- iCf> above, pp. 27-8. ANALYTIC AND SYNTHETIC JUDGMENTS 61 cussion of the problem whether we can legitimately maintain a knowledge of the supersensible on the basis [ b of such judgments. Now, to this question we can at once answer that a priori synthetic judgments obviously cannot be derived from sensible experience. But, if we can find no support for such judgments in experience, in what possible way can they be established ? Though we can no doubt obtain a priori analytic judgments from pure thought, this does not enable us to affirm anything except in regard to our own conceptions. Thus we seem to be placed between the horns of a dilemma. Synthetic judgments we may obtain from experience, but they are not a priori ; a priori judgments we can derive from thought, but they are not synthetic. The combination of a priori and synthetic thus seems an impossibility; and yet, unless we can show the legitimacy of judgments which are at once a priori and synthetic, we can have no science. Kant illustrates this dilemma by the case of the principle of causality. "We learn from sensible experience the fact that there are certain events : and having thus obtained the conception of an event, we can derive from it certain a priori judgments, such as that "every event is preceded by another event " ; but these judgments, though they are a priori, are not synthetic. No analysis of the conception of an event will yield the conception of cause ; for cause implies, not merely a series of events, but a series of events necessarily connected with and inseparable from one another. How then are we to pass from the one con- ception to the other ? On what are we to base the truth of our judgment, that " every event must have a cause " ? It is quite useless to fall back upon sensible experience : for, as Hume has shown, sensible experience can never take us beyond the limited knowledge of an actual association of events. The great problem, then, is to find out how we can have a priori synthetic judgments. 62 THE CRITIQUE OF PURE REASON The Princvples of all Theoretical Sciences of Reason are a priori Synthetic Judgments. 15 a (1) Before dealing directly with the question, whether we can have an a priori knowledge of the supersensible, Kant, following the method indicated in the Preface, passes in review the sciences of mathematics and physics. All the judgments of mathematics, as he contends, are synthetic. If it is objected that certain conclusions are reached in mathematics by presupposing the truth of given judgments, and hence that these conclusions are seen to be true analytically, Kant answers that as the judgments pre- supposed are not analytic but synthetic, the judgment in which the conclusion is expressed is really synthetic. Certainly the mathematician assumes, and is entitled to assume, the truth of the logical principle of contradiction, which affirms that A cannot be not-A : but this principle would never of itself enable him to add anything to knowledge. If, therefore, he adds to the sum of knowledge, as Kant assumes that he does, it must be by going beyond the conceptions from which he starts, and adding to them in some way. 16 a And not only are mathematical judgments synthetic, but at least the judgments of pure mathematics are also a priori, since they are universal and necessary. Kant does not here make any attempt to prove that the judgments of mathematics are a priori. He takes it for granted that mathematics is a science and that no one will dispute the proposition that as a science it contains a prion judgments. Indeed Kant elsewhere speaks as if Hume would have been led to revise the whole of his philosophy, if he had seen that the denial of the principle of causality logically led to the denial of the necessity of mathematical judgments. Hume, he says, did not observe that the judgments of mathematics are' not analytic, and therefore cannot be derived, as he supposed they could, from pure A PRIORI SYNTHETIC JUDGMENTS 63 conceptions. So far as Hume is concerned Kant labours under a mistake, for in his earlier treatise Hume reduces mathematical judgments to relative or empirical proposi- tions. What Kant says in regard to Hume is, however, true when applied to the later treatise, the only work of Hume with which Kant was acquainted. The main difficulty, then, to Kant's mind, was not to prove the a priori character of mathematical judgments, but to establish their synthetic character. Hence, though in the course of his investigation he does give a demonstration of the universality and necessity of mathematical judgments, he contents himself in the Introduction with an appeal to the admitted character of mathematical judgments, which, as he says, everybody will grant to be a priori. Arithmetic, one of the branches of mathematics, is based entirely, according to Kant, upon synthetic judgments. He assumes that these judgments are a priori, that is, universal and necessary, and his contention is that they are also synthetic. At first sight it seems indeed as if the product 12 could be obtained by a simple analysis of the conception 7 + 5 ; but Kant contends that the problem to add 5 to 7 can only be solved if we have recourse to pure perception. It is no doubt true that, having the conception 7 and the conception 5, we can obtain by analysis 7 units and 5 units ; but to obtain the product 12 it is not enough to have the separate conceptions 7 and 5, and the problem that they are to be united, but we must actually unite them. Now, this can only be done, as he contends, if there is a pure perception corresponding to our conception. Addition always presupposes perceptible units, i.e., the simplest determinations of space ; and hence we must visualize our conceptions in the form of separate units in order to reach the judgment 7 + 5 = 12. The same thing holds true of geometry, the propositions of which are all synthetic because they go beyond concep- tion to pure perception. The conception straight does not 64 THE CRITIQUE OF PURE REASON yield the judgment, " A straight line is the shortest distance between two points"; but, to obtain this judgment, we must draw the parts of the line one after the other, at the same time combining them into a whole. Thus geometrical judgments, like arithmetical, are at once a priori and synthetic. 17 a (2) Like mathematics physics is a science based upon a priori synthetic judgments. As instances of these, Kant cites the permanency of the quantity of matter in the world, or what is ordinarily known as the indestructibility of matter, and the principle of action and reaction. These propositions, it may be observed, unlike those of pure mathematics, belong to the class of "mixed" a priori judgments. They presuppose a certain matter of sense, whereas pure mathematical judgments are based upon the determination of the pure form of perception. As in the case of mathematics, Kant here assumes that physics is a science, its distinctive characteristic being that it consists of a connected system of perceptions, these perceptions being concerned with material objects, that is, objects in space. Assuming that there is a science of physics — a proposition which he does not as yet seek to establish — Kant points out that the principles of the indestructibility of matter and of action and reaction are obviously a priori in this sense that they are not, like a posteriori judgments, limited to what is immediately perceived, but express what holds necessarily and universally. These judgments are also synthetic, for, as Kant argues, the conception of matter does not in itself imply more than the presence in space of sensible bodies. While, therefore, we can obtain by analysis the judgment, " Matter occupies space," we can only obtain the judgment, " Matter is indestructible," by going beyond the conception of matter with which we start, and adding to it the new attribute of indestructibility or permanence, an attribute not originally found in it. What entitles us to frame such a synthetic a priori judgment Kant does not here attempt A PRIORI SYNTHETIC JUDGMENTS 65 to explain ; he is contented ■ to point out that it cannot possibly be derived analytically from the conception of matter. It has been objected that the principles to which Kant here refers are not taken from the pure science of nature, but from a special branch of it, namely, that which is limited to bodies in space, and which is afterwards dealt with in the Hudiments of Physics. Had Kant, as it is argued, cited the principles of Substantiality and Causality, instead of the Indestructibility of Matter and the Law of Action and Reaction, he would have led up in a natural and logical way to the subsequent development of his philo- sophy ; for the Critique of Pure Reason deals with the former and not with the latter principles. Nor can there be any doubt that as a matter of logical arrangement the criticism is soimd. But it has to be remembered that Kant's attitude in the Introduction is that of one who accepts without question the laws of physical science, and that he could confidently appeal to the general belief that the Indestructibility of Matter and the law of Action and Reaction are necessary and therefore a priori principles. To have entered at this point into a discussion of the distinction between the principles of nature in general and those of physical nature would have unduly complicated the statement of his problem, whereas he could count on the sympathy of his readers when he referred to the dynamical laws assumed from the days of Newton to be universal and necessary. Nor is there more than a formal irregularity in Kant's citation of the principles of Physics ; for, as he afterwards maintains, those principles are based upon the a priori synthetic principles of the understanding, and are therefore, as he here affirms, themselves a priori and synthetic. Th (3) Now, if it is true that both in mathematics and physics there are a priori synthetic judgments, it seems not ; impossible that such judgments may also be found in 66 THE CRITIQUE OF PURE REASON metaphysic, by which is here meant the real or supposed science of the supersensible. No doubt, as Kant has already said, there is a fundamental difference between metaphysic and sciences like mathematics and physics ; for, while these deal with objects of possible experience, the former by its very nature transcends experience. Still, if there is a science of metaphysic, there can be no doubt that it must consist of a priori sjTithetic judgments. It is perfectly obvious that no number of analytic judgments can constitute a science which claims to determine the nature of real existence ; for no analytic judgments, as we have seen, can take us beyond our own conceptions. Metaphysic seeks to enlarge our a priori knowledge by adding to it a whole region, the region of the supersensible, and therefore it must necessarily, if it is a science, consist of real a priori synthetic judgments. Dealing with such questions as the origination of the world — asking whether it began to be at a certain time or has existed from all eternity — metaphysic so obsiously claims to be an a p)-iori science, reaching entirely beyond the limits of possible experience — that is, beyond the limits not only of perception but of imagination — that its object can only be to establish a priori synthetic judgments in regard to supersensible reality. AVhether such judgments are possible is the main aim of our whole investigation. 18 a As the result of our enquiry so far, we may put the problem of pure reason in the form : How are a priori synthetic judgments possible ? for, as we have seen, any a priori knowledge which determines the nature of know- 18 h able existence must necessarily be synthetic. An answer to this question will enable us to see how far reason, that is, theoretical reason, is concerned in the establishment of science. Now, the sciences that deal with experience are the mathematical sciences, which determine the universal form of the sensible world, and the physical sciences which determine the principles underlying the system of connected A PRIORI SYNTHETIC JUDGMENTS 67 objects of sense. We have therefore to deal with these two questions : How is pure mathematics possible ? How is pure physics possible ? At this stage of his enquiry Kant assumes that the mathematical and physical sciences exist, and his problem is to point out what are the conditions in the nature of the human mind which make them possible. We may, therefore, translate the questions, How is pure mathematics possible ? How is pure physics possible ? into the form. Under what conditions of theoretical reason is pure mathematics possible ? Under what conditions of theoretical reason is pure physics possible ? We cannot, however, adopt quite the same attitude in regard to meta- physic ; for the problem here really is, whether metaphysic is a science at all or merely a collection of unproved assumptions. So far as its past history goes, we should rather be led to conclude that a science of metaphysic is impossible ; and, therefore, we cannot, as in the case of the special sciences, assume metaphysic to be a science, and proceed to enquire into the theoretical conditions of it. 8 c But, although metaphysic has not as yet established its claim to the rank of a science, there is no doubt whatever that there is a strong tendency in man to construct a metaphysic. It is true that aU the efforts hitherto made to satisfy this tendency have ended in failure. It has been proved with equal plausibility (1) that the world has had an absolute beginning, and (2) that the world has existed from all eternity. Now, these two propositions are mutually contradictory, and reason can never be satisfied with that which is self-contradictory. We cannot, therefore, conclude from the mere tendency to construct a metaphysic that a science of metaphysic is possible. But this at least we can say : that, inasmuch as the question is whether reason is or is not of such a nature as to be capable of grasping reality as it actually is, we have in our own hands all the data for the solution of the question. We must be able to show that theoretical reason can, or can not, know things in 68 THE CRITIQUE OF PURE REASON themselves ; in other words, we must be able to determine whether real a priori synthetic judgments in regard to the supersensible are possible or not. We may, therefore, even in the case of metaphysic put the problem of pure reason in the form : How is a science of metaphysic possible ? We do not in this case assume metaphysic to be a science, but we ask what are the conditions of there being a science of metaphysic. This is the critical as distinguished from the dogmatic poiat of view. Dogmatism leads to self-contra- diction, because it never asks what are the Umits within which knowledge is possible. As the history of dogmatic metaphysic has abundantly shown, self-contradictory pro- positions are the iaevitable result of the dogmatic procedure of reason.^ But, when contradictory propositions flow from a certain way of treating the problems of metaphysic, the inevitable result is that doubt is cast upon the nature of reason itself. Thus scepticism invariably issues from dogmatism. In order to avoid this result, we must there- fore ask how a science of metaphysic is possible, if it is possible at all. 7. Idea and Division of the Critique of Pure Reason. 19 a The problem of pure reason has been shown to imply two main subordinate problems: (1) How are a priori synthetic judgments of experience possible ? and (2) How are a priori synthetic judgments which transcend experience possible? The answer to these two questions will yield the contents of a science which may be called the Critique of Pure Reason. It is to be observed that this science deals only with the principles of a priori knowledge, not with the special applications of those principles. It is called a science, because it is not a mere collection of propositions, but an organic system, lq which all the parts are related to a single supreme conception. This system is not a doctrine, but a 1 Cf. above, p. 33. ITS IDEA AND DIVISION 69 criticism, of pure reason ; in other words, it does not consist of a totality of judgments, which express the absolute nature of existence. Such a totality of judgments could only be obtained if it were possible for human reason to comprehend or come into direct contact with ultimate reality. But it is the basis of Kant's philosophy that knowledge is limited to "experience," and that experience does not contain the ultimate determinations of things, but only the system expressive of the manner in which we, with our limited faculties of perception and thought, must determine reality. On the other hand, the himian mind is incapable of going beyond experience so as to obtain a knowledge of ultimate reahty, because from pure ideas no direct advance can be made to objective existence. The result of this limitation is that in its theoretical aspect reason does not enable us to determine positively what is the ultimate nature of things. At the same time, while the Critique of Pure Eeason does not yield a theoretical comprehension of ultimate reaUty, it must not be supposed that it is a mere beating of the air ; for, by throwing light upon the nature of our reason and preventing us from confusing the a priori knowledge, which has a meaning in relation to experience, with a supposed knowledge which extends beyond experience, it enables us to keep a space open for the practical reason, by the aid of which a positive determination, though not a knowledge, of ultimate reality may be obtained. It follows from what has been said that " transcendental " knowledge is not to be identified with a knowledge of objects in their ultimate . nature, but only with the way in which a knowledge of objects of experience may be gained, so far as that is possible a priori. The term " transcendental " in Kant always has a reference to the conditions in the nature of our knowing faculty by which certain a priori judgments are made possible. The object, then, of the Critique of Pure Eeason is to determine, by a criticism of pure reason itself, in what sense we have knowledge, and therefore 70 THE CRITIQUE OF PURE REASON to prevent us from falling into error by confusing different kinds of a priori knowledge. 20 a The result of this transcendental criticism wUl be, Kant says, to afford "a complete architectonic plan of philosophy." By an " architectonic plan " he means one which is derived from and connected with a single ultimate principle. It is not enough to point out that there are certain a priori judgments, but we must connect all a priori judgments with » certain principles, and ultimately with a single all-compre- hensive principle. For example, it is not enough to point out that there are certain mathematical judgments which are a priori and synthetic, but we must be able to refer all possible mathematical judgments to a single principle based upon the fundamental nature of our perceptive faculty. Similarly, it is not enough to point out that there are a priori judgments of pure physics, but we must be able to refer all possible judgments as contained in the physical sciences to a single principle expressive of the fundamental nature of our thinking faculty. And these two principles, the principle of perception and the principle of thought, must both be referred to the one supreme principle of self-consciousness, of which they are determinations. 20 6 The Critique of Pure Season, therefore, contains all that is reqidred in the determination of the principles of tran- scendental philosophy, and if we distinguish it from that philosophy, it is only because we do not in it set forth all the specific applications of those principles. For instance, the principle that "action and reaction are equal and opposite " is not one which is established in the Critique of Pure Eeason ; for that Critique only lays down the prin- ciples which are involved in all possible experience, whereas the principle of action and reaction is a specific determina- tion of material bodies. What the Critique of Pure Reason does is to set forth the principle without which this specific physical principle would be impossible. It declares that " all substances in so far as they can be observed to co-exist ITS IDEA AND DIVISION 71 in space are in thoroughgoing reciprocity " ; while the principle of action and reaction goes beyond this universal principle, inasmuch as it introduces the conception of force or energy. Now, what is true in regard to this principle, the " Principle of Community," is true generally ; hence Kant has a special treatise, dealing with the principles of Physics, as supplementary to the Critique of Pure Eeason, and consisting in the apphcation of the universal principles established in the Critique of Pure Eeason to bodies in space. 10 c It is important, in making a division of the various parts of this science, that we should not introduce anything that is of empirical origin, that is, anything which is dependent upon particular sense-perceptions. This of course does not mean that we are to abstract entirely from sensible experi- ence, and deal only with pure ideas ; for, as we know, it is Kant's principle that without sensible experience we can have no knowledge whatever : what is meant is, that we must keep our minds fixed upon the a 'priori element in knowledge, whether that element has a reference to sensible experience or claims to transcend it. Kant therefore separates from this science the principles of morality, because these have a meaning only in relation to particular feelings, the feelings of pleasure and pain. What he means by this limitation we shall afterwards understand more clearly. In the meantime it is enough to say that in his doctrine freedom and duty have a meaning only in relation to a being like man, who is partly under the influence of immediate feelings of desire. Thus the conception of duty commands that our actions should be determined purely by reason, and not at all by the solicitations of desire ; hence the relation of reason to desire is purely negative. Never- theless, duty has no meaning for a purely rational being, but only for one that is at once rational and sensuous. )rf The systematic division of this science is, firstly, into the doctrine of elements, as contrasted with the doctrine of 72 THE CRITIQUE OF PURE REASON method. The doctrine of elements is again subdivided into a priori elements of sensibility and a priori elements of thought. These subdivisions are based upon a distinction, which Kant always insists upon, between perception and thought. This distinction, in fact, is characteristic of the Critical, as contrasted with the Leibnitz-Wolffian, philo- sophy ; for in the latter perception and thought differ only in degree, so that thought is nothing more than clear and distinct perception. Kant does not here deny that per- ception and thought in their ultimate nature may spring from a " common root," but he does deny that under the conditions of our knowledge it is possible to identify perception and thought; and he therefore maintains that for us they remain forever distinct in kind, so that all attempts to assimilate them can only result in confusion. Perception in us is always of the individual as presented in space and time, thought is always of the universal and is independent of space and time. Hence the prin- ciples of perception must be distinct in kind from the principles of thought. Perception involves an element of sense, and as such it contains an element which is not due to the spontaneity of the mind, but is simply received by the mind. It is not with this sensible element that the critical philosophy has to deal, but with the a priori element, as the condition of the determination of this sensible element to objects of perception. The two sub- divisions of the Critique will therefore deal respectively with the a priori element presupposed in perception, and the a priori element presupposed in thought. The in- vestigation of the former is the task of what Kant calls Transcendental Aesthetic, the investigation of the latter falls under the head of Transcendental Logic. 22 TRANSCENDENTAL AESTHETIC. In the Transcendental Aesthetic Kant treats of percep- tion as if it involved the consciousness of an object, independently of any activity of thought. When, however, in the Transcendental Analytic he goes on to consider the conditions iu thought of experience, he clearly indicates that there can be no knowledge of objects proper without the activity of thought. We must, therefore, regard the point of view exhibited in the Aesthetic as in a certain sense provisional. This is partly explained by the fact that Kant had worked out his view of space and time as the a priori forms of perception some time before he had come to see that the understanding has certain constitutive forms of its own which are essential to experience. But we must also regard this anticipatory procedure as a device of method. In setting forth his whole doctrine, which was to accomplish a complete revolution in the ordinary way of conceiving knowledge, Kant found it necessary to start as nearly as possible with the common view of perception as giving a knowledge of iudividual things. What he sets himself to do in the Aesthetic is, therefore, to point out that, assuming perception to be a form of knowledge in which individual sensible things are apprehended, we must, in order to account for the facts, maintain that it contains two elements — a matter and a form — the matter coming from without, and the form belonging to the very constitution 74 TRANSCENDENTAL AESTHETIC of our perceptive faculties. This form is a priori, just because it is inseparable from any perception of objects; and as such it enables us to explain how upon it certain a priori judgments can be based. It is also part of Kant's object to maintain the distinction in kind between perception and conception. In the school of Leibnitz it had been maintained that the judgments of mathematics are a priori ; but this view was held along with another, which Kant regarded as contradictory of it, viz., that per- ception is the same in kind with conception, the difference being merely in less and greater clearness and distinctness. Now, Kant has to show that mathematical judgments are not derived from any analysis of conceptions, for, as he maintains, analysis can never take us beyond our ideas so as to constitute real knowable objects. We have, therefore, to deny that perception is nothing but confused conception, and to maintain that it is different in kind from conception. This view, according to Kant, enables us to understand, not only how we can have a priori mathe- matical judgments, but how these are synthetical. In short, the preliminary view in the Aesthetic, while assuming that knowledge is given in perception without the activity of thought, is directed mainly to two points : (1) to show that the mind has two a priori forms of perception, viz., space and time, and (2) that by taking account of this fact we can explain the a priori synthetic judgments of mathematics. 22 a In the opening paragraphs of the Aesthetic Kant deals with empirical perception, firstly, pointing out its condition; secondly, giving its definition ; and, thirdly, stating what is its object. (1} The condition of empirical perception is sensation, which is (a) a mere modification of the subject, and (b) is called forth through the influence upon the subject of an object. The " object " which is the occasion of the affection must be the thing in itself (Ding an dch). Here, therefore, Kant assumes the existence of objects INTRODUCTION 75 lying beyond the sphere of knowledge, and he speaks as if these objects somehow acted upon the mind and gave rise to impressions. Indeed, it may be said that Kant never entirely got rid of the view that the sensible element in knowledge is in some way due to independently existing things. We shall see, however, in the sequel, that this assumption of independent things tends more and more to sublimate itself, and disappear in a higher doctrine. Sensihility Kant defines as " the capacity of receiving impressions " ; it is in his view the capacity or faculty of having certain " ideas " ( Vorstellungen), taking the term "ideas" in the sense of any mode of theoretical consciousness. Sensation is one of these modes. What is here said, therefore, comes to this, that sensation is a mode of consciousness, which is occasioned by the action upon the subject of an independently existing thing. (2) Empirical perception is contrasted either with (a) intellectual perception or (J) pure perception. Probably in the present case it is the latter which Kant has before his mind. The distinction between sensation and perception is not very clearly ex- pressed. Perception is said to be a mode of consciousness, which refers itself directly to an object. This reference to the object is in us mediated through sensation; in other words, without sensation we could have no perception. Here, therefore, the term " object " must mean the object in consciousness, not the thing in itself. Thus we have already the contrast between the " object " as it exists for conscious- ness, and the " object " as it exists apart from consciousness. We have, in short, already indicated the distinction after- wards drawn specifically between the phenomenal object and the thing in itself. (3) The " undetermined object " seems to mean any object of empirical perception, e.g., house, tree, river, not yet determined by thought. Such an object is a "phenomenon," inasmuch as (a) it is not identical with the thing in itself, and (h) it is an object of consciousness. A phenomenon, we may therefore say, 76 TRANSCENDENTAL AESTHETIC is any sensible object whatever of which we can be conscious. 22 6 On occasion of sensation, then, there is empirical per- ception, and any object of empirical perception is a phenomenon. The phenomenon, however, as Kant goes on to point out, involves two distinguishable elements, which he terms the matter and the form. There is a certain difficulty in understanding what is meant when it is said that the matter " corresponds " to sensation. The most reasonable view seems to be this. Kant has said that a phenomenon is any sensible object whatever, e.g., this house, this tree, this river. Such an object, as perceived, is within consciousness. Suppose, then, that the subject is actually conscious of a certain sensible object, and we ask what is involved in there being for him such an object. The answer is, that the sensations, which in themselves are merely affections of the knowing subject, have been ordered or arranged in a certain way; in other words, we find that the object as perceived is a complex of two elements. When we analyse this complex, we see that the sensations, apart from the manner in which they are ordered, are simply affections of the subject, while in the " object " they are presented as ordered, and, in fact, only as so presented can they be called an " object." Thus a change has been effected in the sensations, from the fact that they have been reduced to order. The sensations are in content the same as before, but this content is now formed. Now, as " matter " and " form " are correlative, we cannot call the sensations hefore they are ordered the " matter " of the object ; what we must say is, that in the object they become " matter." Hence, in the perceived object the " matter " corresponds to what prior to this object was pure sensation. Kant's point is, then, that sensation becomes an element in the perceived object when it receives " form," and that in this new relation it is no longer mere sensation, or the " matter " of INTRODUCTION 77 the phenomenon. No doubt there are passages where Kant says that sensation is the " matter " ; but this is quite a natural mode of speech for one who regards sensation as undergoing no change in perception except that of being ordered in space and time. What Kant must mean is, that sensation becomes an element in the complex whole of perception, when the transition from sensation to perception takes place. It is misleading to introduce passages from the Analytic, where it is pointed out that conception as well as perception is involved in the consciousness of an object. In the Aesthetic Kant is only concerned to show that a perception or perceived object involves more than sensation, viz., the form of per- ception, and he therefore at present keeps in the back- ground the further element which it implies, viz., the form of thought. The sensations become an element in the perceived object when they are ordered in certain relations, i.e., in the relations of space and time. These sensations are then the " matter " to which " form " is given. As particular sensations they differ in content or are a " manifold," and this " manifold " is not sensations, but these viewed as an element ia the perceived object. Kant is not here thinking of the " manifold " as exclusive of simplicity, but rather as iuvolving various given differences. The other element, the " form," is required in order that there may be such a perceived object, and it is due to this element that the manifold is ordered in certain relations. What Kant says, then, is that there can be no perceived object whatever without the ordering of the sensible manifold in the relations of space and time. The distinc- tion of matter and form is in the Aesthetic a distinction of the two elements involved in the perceived object. It is not a distinction of things in themselves, but only of the subject in so far as it is affected by things in themselves. Since sensations are in themselves mere particulars without relations of any kind, we can only explain the perceived 78 TRANSCENDENTAL AESTHETIC object by recognizing that the relations between sensations are due to another element, its form. Kant therefore assumes that in their own nature sen- sations are mere particulars, and from this he infers that the form comes from some other source. And if we grant that sensations are simply particular affections, excited in us by things in themselves, it seems to follow that the order which is found in objects of perception must be due to another source. It is therefore implied that, prior to the ordering of them by the form, sensations are a chaotic mass without order or connection. Accordingly Kant goes on to say that the " matter " of a phenomenon is given to us entirely a posteriori, while the form must lie a priori in the mind. To say that the matter is given a posteriori is merely another way of saying that sensations are affections of the subject as excited by the thing in itself. Again, to- say that the form is not sensation, implies that it is a priori, since only the matter is a posteriori. The dis- tinction of an a posteriori from an a priori element in the perceived object therefore corresponds to, and is involved in, the distinction of matter and form. When Kant says that the form must lie a priori in the mind, he must be taken to mean that while the mind has the capacity of ordering sensations, and so giving form to them, only when this is done is there a perception. 22 c The pure form of sensibility is, however, identified by anticipation with formal perception, i.e., with space and time represented as an object, the object dealt with in pure mathematics. The pure form is the capacity of ordering sensations in certain relations, and this capacity when actually exercised in relation to sensations yields perceptions of sensible objects. Kant here assumes, as he does all through the Aesthetic, that these perceptions are possible without any activity of thought (Verstand). This, however, as has already been said, is a provisional assumption which is afterwards corrected in the Analytic ; for, as Kant there INTRODUCTION 19 shows, there is no formal perception, such as mathematics requires, without the synthesis of imagination as guided by thought. Strictly speaking, therefore, the pure form of sensibility is not identical with pure perception. So Accepting the distinction between the form and the matter of perception, and also by anticipation the distinction between the form and the matter of thought, we shall have two subdivisions of the Critique of Pure Keason, viz., Transcendental Aesthetic, dealing with the a priori prin- ciples of sensibility, and Transcendental Logic, occupied with the a priori priuciples of pure thought. 36 If we look at any concrete object of perception, say a house or a tree, we can distinguish in it the elements contributed respectively by the understanding and by perception. Kant here indicates that in perception there is an element contributed by thought ; in other words, that we cannot even have the consciousness of an object of perception without the activity of the understanding. But, though he thus refers to the part played by the under- standing in perception, he goes on, in the Aesthetic, to speak of perception as if it were possible without the exercise of thought. ISTow, when we set aside the element due to thought, we have before us all that can be referred to perception. But perception, even when thus isolated, has in it the two elements already referred to, viz., the matter of sensation, and the form given to this matter. If now we abstract from the matter, we shall have before our minds the mere form. Looking, then, at the result of this process of abstraction, we find that we can refer everything else in the perception to sensation, except the relations of sensations as determined in space and time. Space and time are therefore the forms of sensible per- ception, and Transcendental Aesthetic has to consider these as principles of knowledge. 80 TRANSCENDENTAL AESTHETIC Section I. — Space. 2. Metaphysical Exposition of Space. 23 e The problem which Kant has to solve in the Aesthetic is m regard to the perception of objects, not in regard to the existence of objects in themselves. Hence when he says that " in external sense we are conscious of objects as outside of ourselves," he means by " objects," not things in themselves, but perceived objects, objects as they exist within our consciousness or for our apprehension. It would of course be an entire misapprehension of Kant's view to suppose that " outside of ourselves " here means beyond our minds or consciousness. What is said to be " outside of ourselves " is simply that which we perceive as extended or in space, and the distinction between " external " and "internal" sense is the distinction between perceived objects which are extended or spacial, and objects which are inextended and non-spacial or exist only in time. It is to be noticed that Kant thinks of " perception " as including inner states or feelings, which are inextended but yet agree with extended objects in being in time. In inner sense, as he explains, we are not conscious of ourselves as an object, i.e., we are not conscious of ourselves except in so far as we are presented to ourselves in inner sense as experiencing certain momentary states. So far as we perceive ourselves we are presented to ourselves as in certain successive states of consciousness, and therefore the " form " of the inner sense is time. Space and time being thus the forms respectively of outer and inner sense, the question arises as to their nature. What are space and time ? Kant refers to the various answers which may be given to this question. In the first place, they may be regarded as " real things," i.e., as existing quite independently of our perception. This was the view taken by Newton, SPACE 81 and indeed it is the natural view of the physical sciences. In the second place, it may be said that space and time are " determinations of real things." Kant is here thinking of the doctrine of Locke, and indeed of common sense, that extension and time are properties of things or of the distinction between one thing and another. Thirdly, space and time may be " relations of real things." This was the answer given by the Leibnitz-Woltfian school, who denied that space and time had any meaning except as the confused apprehension of the distinction between one thing and another. Lastly, space and time, as it may be held, have no real existence except for our perception, being simply the " forms " under which we arrange the manifold of our sensibility. This is Kant's own view, and he immediately proceeds to give the reasons why he holds it to be the true explanation of the facts of our perceptive experience. By a " metaphysical exposition " of space Kant means a direct statement of the character of space as a priori. He probably calls it an " exposition," because he has in his mind the contrast between the direct proof in the Aesthetic that space is a priori and the indirect proof in the Analytic that there are conceptions of thought which are a priori. The latter proof he calls a "deduction," thus distinguishing o it from an " exposition." In his metaphysical exposition of space Kant seeks to show (1) that space is not an empirical idea. It is especially the doctrine held in the Leibnitzian school that he has before his mind, though no doubt he means also to exclude the view of Locke. Thus Wolff tells us that " space as a whole consists in a certain constant relation of elements to one another, so far as this relation is confusedly perceived by us." This is the view indicated in the preceding paragraph, where one of the doctrines mentioned is that space and time are merely " relations of things." On this view of Wolif, we begin with the per- ception of objects in space, or, in other words, we have an F 82 TRANSCENDENTAL AESTHETIC experience of the actual relations of things to one another, the confusion being due to the apparent continuity of objects, which are really discrete substances, or " monads," lying side by side. Space is the abstraction in thought of this relation ; hence, as Kant implies, space is on this view "an empirical conception which has been derived from external experiences," i.e., from experiences of various objects distinguished by their sensible properties, but agreeing ia being all outside of one another. This doctrine he seeks to disprove. He has therefore to show that space is not the abstract conception of the actual relations of things, given in sensible perception, though apprehended in a confused way. Now, the perception of an external object involves (a) the affection of the sensitive subject, (b) the reference of this affection to something which is viewed as out from the subject, and apart from, yet side by side with, other objects. Kant does not here ask (i) whether the sensa- tions which are so referred to something out from the subject remain in content what they were prior to this reference, but rather assumes, or at least does not ques- tion the view, that they afford the consciousness of the sensible properties of things, viz., their colour, hardness, softness, etc. He also assumes (ii) that certain sensations, viz., organic sensations, are directly referred to " something," i.e., to what is perceived as one object. How there can be a consciousness of the sensations as combined into the unity of a single object he does not ask, but leaves this question to be dealt with in the Analytic. Assuming, then, that an "object" is a unity of objectified sensations, he insists that such an object cannot exist for the conscious subject except on condition that it is represented as in space. Thus space is presupposed as the condition of the perception of a sensible object. Without the capacity on the part of the subject of ordering the sensible as out from himself and out from other objects and side by side with them, there can be no perception at all. SPACE 88 We do not perceive objects separated in a real space, and by abstraction obtain the conception of space : but we only have the perception of an object because we have beforehand the capacity of so ordering certain sensa- tions as to present in our consciousness objects as primarily determinations of space. Wolffs view that we perceive real elements or monads, i.e., separate objects lying side by side, is untenable, because every object of perception pre- supposes space as a fixed method of ordering the sensible. In other words, sensations do not become the perception of an object until they are ordered as spacial, and this faculty of ordering them spacially must be presupposed as the condition of their being ordered, i.e., represented as in space. It is not true, then, that perceived objects exist in space prior to the apprehension of them as so existing, but the object as it exists for the conscious subject comes into being in the perception, i.e., the sensations are in the act of perception ordered spacially, and only so become perceived objects. If objects were already related spacially apart from perception, various objects or parts of objects would be apprehended as so related, and by an act of abstraction the conception of space, as a rela- tion of outness found in all, would be formed. In that case space would be an " empirical conception," or would be borrowed from the apprehended relations of objects existing independently of consciousness, and hence would not be a priori. The real explanation is, that the external object as known comes into being with the pre- sentation of it as spacial ; and unless the spacial determina- tion is presupposed, there can be no external object whatever. Kant, therefore, concludes from the logical priority of space that it is a priori. If the determina- tion of objects as spacial is a necessary condition of any perceived object, space cannot be a posteriori, or apprehended in particular sensible experiences, but must 84 TRANSCENDENTAL AESTHETIC be an a priori condition of all sensible experience. This view he sets forth explicitly in (2). 24 6 (2) In the first argument Kant has shown that space is not derived from experience, or is not a posteriori ; and as for him it must be either a posteriori or a priori, he now states positively that it is "a necessary a priori idea," logically preceding every possible presentation of an object as external. Space, Kant argues, is " a necessary a priori idea," because, while we can think away any " sensible object presented as in it, we cannot think space itself away. This does not mean that the non-existence of space is unthinkable, but only that no presentation, no perception or imagination, is possible, if we suppose space to be absent. The idea of space is thus necessary ^- to any presentation whatever. The necessity of space, as the unchangeable condition of any possible presentation, is tacitly contrasted with the contingency and change- ability of sensible objects. What comes from without, or is empirically given, is for the perceptive subject contingent ; and hence, if space were empirically given, it also would be contingent. Kant is here thinking mainly of the Wolffian doctrine, and he argues that, if space were given to the subject in empirical apprehen- sion, we should be able to think it away as we can think away any given object. Since we cannot think it away, it is a priori. Space, therefore, as Kant concludes, is not derived from the apprehension of phenomena, but it is the condition of " the possibility of phenomena, and as such it is an a priori idea making phenomena or perceived - objects possible. The result of the two arguments thus is, that space is the a priori condition of all external phenomena. 25a (3) The next point is that space is a pure perception; the proof being that we can only present to ourselves ~ one single space ; the conclusion from which is that space, as an a priori perception, is presupposed in all SPACE 85 conceptions of space. So far Kant has argued that space is an a priori " idea " : now he specifies the kind of a priori idea it is, affirming that it is an a priori perception (Anschauung), not a conception {Begriff). The distinction between perception and conception is thus stated in Kant's Logic, p. 1 : " All modes of knowledge, i.e., all ideas (Vorstellungen) consciously referred to an object, are either perceptions or conceptions. Perception is an individual idea (repraesentatio singularis), conception a general idea ~ (repraesentatio per notas communes'), or a reflective idea {repraesentatio discursiva). Knowledge through conception is thought {cognitio discursiva). Conception is opposed to perception ; for it is a general idea or an idea of that which is common to several objects, and therefore an idea so far as it can be contained in different objects." So, in the Critique of Pure Eeason (B 377), he says: "Per- ception is related directly to the object, and is individual, whereas conception is related to it indirectly, by means of a mark, which may be common to several things." When, therefore, Kant says that space is a " perception," he means that it is the direct consciousness of an object, as an immediate presentation in an individual image. Space, he says, is " not a discursive or so-called general conception." Kant prefers the term " discursive " to the term " general." In his Logic he says : " It is a mere tautology to speak of general or common conceptions, — a defect which is based upon a wrong division of conceptions into general, particular and individual : it is not the conceptions themselves, but only their use, that can be so divided." The point then, is, that space is not a con- - ception, a product of comparison and reflection. The view which he is opposing is that of Leibnitz as he understood Leibnitz, and hence he says that space is not a " conception of the relations of things." As he says afterwards (B 56): " If we take the view that space and time are abstracted from experience, — though in their separation they are 86 TRANSCENDENTAL AESTHETIC confused ideas of the relations of phenomena as co-existent and successive — we must regard the a priori conceptions o| space and time as merely creations of the imagination, the real source of which must be sought in experience." In other words, Leibnitz regards space as obtained by ab- stracting from the differences in the actual relations of individual things and framing a conception of what is common in them all, a conception which is " confused " because it retains the apparent continuity of individual things. Kant, on the other hand, maintains that space is a perception, not a conception at all. If Leibnitz were right, space would be a mere fiction of abstraction. The contrast here is not between pure perceptions and pure conceptions ; for, though that contrast is referred to later in the Analytic, in the Aesthetic the ordinary opposition of perception and conception is assumed, and space is affirmed to be the former not the latter. The proof of the thesis that space is a pure perception is as follows : (a) Space is something individual : there is only a single space, not several spaces. It must therefore be a perception, not a conception, since a conception is always relative to several objects, which differ in their qualities. It may be objected that as we speak of several spaces, the idea of space must be a conception, containing what is common to these different spaces. Are not these different spaces, it may be asked, several species of space, or various instances of space, just as various species of men or various individual men are subsumed under the conception man ? No, answers Kant : space is an individual, like an individual man,— say, Socrates — and the various spaces are merely parts of this one space. A conception involves a certain number of abstract marks or attributes, which are always found in many individual things. But space is itself a concrete idea, an immediate perception, of which the so-called spaces are parts or determinations. The spaces are in it, not subsumed under it. There are no specifically SPACE 87 different spaces : each particijlar space is space, (b) The "parts" of space do not precede the idea of space as constituents out of which it is made up, but are thought as in the one all-embracing space : the parts of space, i.e., are not first given as parts, and then put together so as to form a whole. Space is a unity, and spaces are limitations of this unity, which arise only by the determination of the one universal space. What before was said of things in space is now affirmed of spaces, viz., that they are possible only through and in it. These "parts" are not actually, but only potentially, in space : they are not independent objects, but dependent parts of an object : they are secondary, not primary. The whole is not possible through the parts, but the parts through the whole. As Kant says in one of his Reflections : " We can only think of spaces by cutting out something from universal space " ; and, again, "what can only be divided by limitation is not possible through composition : therefore space is not so possible." In his Logic, Introd., 5, Kant speaks of the " marks " of a conception as " constituents," that is, logical constituents. His argument here would therefore seem to be : Space is not a conception, which is produced by putting together logical constituents or " marks," but the idea of a whole as logically preceding all its " parts." The two arguments, then, are : (a) Space is a perception, not a conception, because a conception contains individuals under itself, not in itself, as is the case with the idea of space ; (h) Space must be a perception, not a conception, because in a conception the parts precede the whole, while in space the whole precedes the parts, which are limitations of it. The first argument rests upon the logical extension of a conception, the second upon the logical intension or com- prehension of a conception. Both deny space to be a conception. When Kant adds that geometrical pro- positions can only be explained if it is admitted that space is a perception, he anticipates what should 88 TRANSCENDExNTTAL AESTHETIC properly have been reserved for the transcendental ex-, position. (4) In this paragraph Kant states the positive character of space as a perception. He tacitly presupposes that space is infinite, taking here the same view as in the Dissertation, where he expressly says : " What are called many spaces are simply parts of the same unlimited space." The continuity of space is also implied ; and indeed afterwards this is expressly stated (B 211), the continuity of space being connected with the thought of the origin of its parts by limitation. The main points insisted upon are (a) that perception is always of the individual, and (&) that only in perception does the whole precede the parts. Here, therefore, Kant is looking at space as itself an object of perception, not merely as a form of the perception of all external objects ; and his point is, that the possibility of determining it ad infinitum, that is, of continually adding new determinations, shows that it is a perception. Con- ception contains a limited number of determinations, space contains an infinite number. Just as we can go on adding to our knowledge of a sensible object by finding ever new properties in it, so space is of such a character that we never exhaust its possibility of determination. The con- clusion, then, from the whole metaphysical exposition is, that space is an a priori perception. ^ 3. Transcendental Exposition of Space. 26 a In Kant's original statement of the problem of philo- sophy he pointed out that it may be put in the form, How are a prim-i synthetic judgments possible ? Further, the first subordinate question may be put in this way, How is pure mathematics possible ? Now, the transcendental exposition of space so far gives an answer to this question, that it shows how geometry as a branch of pure mathe- matics is possible, or at least states the condition without SPACE 89 which it is impossible. The transcendental exposition of space is, therefore, Kant's answer to the question, How are a priori synthetic judgments of mathematics possible ? when that question is limited to geometry, and abstraction is made from the element contributed by the understand- ing. He calls it a " transcendental " exposition, because it transcends or goes beyond the idea of space and shows how certain specific judgments in regard to determinations of space can be justified. To be successful, a transcendental exposition of space, in the first place, must show that certain actual synthetic propositions are derived from space, and, secondly, it must prove that they can be derived from space only if we admit that it is an a priori perception. 26 6 As to the first point, it is plain enough, as we are told, that geometry actually does determine the properties of space synthetically, and yet a 2'™''"^- Here Kant assumes the existence of geometry as a science, an assumption which, as we have already seen, he makes all through the earlier part of the Critique. Granting the existence of this science, there can be no doubt that it presupposes, and is derived from, the idea of space. Hence, secondly, we have to ask whether space is of such a character that the a priori synthetic judgments of geometry can be derived from it. Now, it is plain from what has been said in the Introduction that the synthetic character of geometrical propositions cannot possibly be explained on the supposition that space is an abstract conception, not a perception. It is further obvious that the a priori character of geometrical propositions can only be explained, if space is itself a priori ; for, all geometrical propositions are universal ; and necessary, and these are the characteristics of a priori ': judgments. It has in fact been shown in the Introduction ; that such judgments cannot possibly be derived from sen- sible experiences, since from these we can never extract 26 true universality or necessity. It is thus obvious that i we can only explain how we can have legitimate a priori 90 TRANSCENDENTAL AESTHETIC synthetic judgments in geometry by holding space as at once a perception and logically prior to objects or a priori. But these two characteristics can only be accounted for if space is a form of the subject, not a determination of the object, since otherwise we could not say a priori that all external objects must always and necessarily be determined 27 a spacially. We therefore conclude that our doctrine, viz., that space is an a priori form of perception, is the only one which satisfactorily accounts for the a priori synthetic judgments of geometry. Inferences. '2,1 h (a) The question which Kant raised in the introductory remark of the metaphysical exposition as to the various theories in regard to the nature of space can now be answered. Negatively, space is not a real thing,^ nor is it a property or relation of real things, if by " real " we mean existing apart from our knowledge or independently; for, as has been shown, the view that space is an a priori form of perception is the only view that explains geometrical propositions, and that view is inconsistent with the doctrine that space is an actual determiaation of things, inde- pendently of the conditions under which our experience takes place. Kant, therefore, rules out the doctrine of Newton that space is a real thing, the doctriae of Locke that it is a property of real things, and the doctriae of Leibnitz that it is a relation of real things. 27 e (b) Positively, again, space is the form of all the pheno- mena of outer sense. It is simply the subjective condition under which we perceive external objects, that is to say, objects that are determined spacially. "When we take this view, there is no difficulty in understanding how space may exist as a form of our perception, and may be determined ' This view is not explicitly mentioned by Kant here, but it is implied, as the corresponding passage on Time, 6 (a), shows. SPACE 91 prior to the actual observation of particular sensible objects. 18 a Space, then, cannot be said to have any independent reality. It exists only for beings who, like us, know under sensible conditions ; for, if we suppose the subjective conditions to be removed, under which we have the percep- tion of external things, it has no longer any meaning. Space, in other words, is a determination given to the sensible matter of our apprehension, or added by the mind itself. This does not mean that it is a mere arbitrary determination of sensible things. Por all human, and probably for all sensitive beings, space is an unchangeable form, and therefore it is the necessary condition of all relations in which objects are perceived ; but this does not entitle us to predicate it of things in themselves. We cannot infer that, because space is a necessary condition of our perception, it is a necessary condition of things in themselves. We cannot even say that all finite beings must order their impressions in the same way as we do. Hence, when we speak of spacial determinations as real, we have to add that they are real only in the sense of being determinations of objects as they present themselves to our consciousness. AVhile we affirm the " empirical reality " of space, that is, while we maintain that spacial determinations are universal and necessary in regard to phenomena, we must also affirm its " transcendental ideaUty " ; in other words, we must deny that space has any meaniag when viewed in relation to things in themselves. Section II. — Time. 4. Metaphysiccd Eocposition of Time. The metaphysical exposition of time is of the same character as that of space, what difference there is being due mainly to the distinctive features of the former. It can therefore be stated very shortly. 92 TRANSCENDENTAL AESTHETIC 29 a (1) As it was argued that space is not an empirical conception derived from external experiences, so it is now maintained that time is not an empirical conception derived from any experience ; the reason of the more general statement being that time, as is afterwards pointed out, is the form of all phenomena, external or internal. Assuming that through perception a knowledge of individual things as co-existent and of their changes as successive may be obtained, it is argued that such knowledge is possible only under presupposition of time. For, unless time is pre- supposed, we should not be conscious of any objects as existing together or of any events as following on one another. Time therefore cannot, as the Leibnitzian School held, be derived from the perception of individual things and events as simultaneous or successive by a process of abstraction. (2) Time, like space, is a necessary a priori idea, with this difference that it is the logical presupposition, not merely of external perceptions, but of all perceptions. For, while we can think away any given object or event, we cannot think away time without destroying the very possibility of perception. Time is, therefore, a priori. 30 a (3) Time is not a conception, but a perception. For there is only a single time, not a variety of times. No doubt we speak of different times, but these are not them- selves individuals, but merely integral parts of the one individual time. As perception is always of the individual, it is obvious that time is a perception, not a con- ception. Kant adds that such synthetic propositions as that " no two moments of time can co-exist," could not be derived from the idea of time were it not a perception; but, like the corresponding remark in regard to space, the statement properly belongs to the transcendental exposition. (4) That the idea of time is not a conception, but a perception, is also evident from its peculiar character. For we are conscious of the parts of time only in so far as we TIME 93 limit the one single time. But it is only a perception which can yield the consciousness of an individual whole. Therefore, a perception must be the foundation of our consciousness of time. The conclusion of the whole argument, then, is, that time is an a priori perception. 5. Transcendental Exposition of Time. 50 (' The transcendental exposition of time, like that of space, must show, firstly, that there are certain a priori synthetic judgments, which actually are based upon the idea of time ; and, secondly, that these judgments can only be explained, if we take a certain view of the nature of time. As to the first point, Kant simply assumes that such a priori judgments as that " time has only one dimension," and that " different times do not co-exist but follow one another," obviously are based upon the idea of time. What he has to prove is that these judgments, which we imme- diately make, and which, as he assumes, are universal and necessary, can only be accounted for on the supposition that time is an a priori perception. Now, firstly, these judg- ments are a priori, and therefore their apodictic or demonstrative character implies that time itself is a priori ; or, in other words, as Kant himself puts it, time is " the necessary a priori condition of all phenomena." Such pro- positions, he argues, as he has done in the Introduction and in the Transcendental Exposition of space, obviously cannot be derived from sensible experience, which can never yield strict universality or demonstrative certainty, because no judgment of experience can take us beyond a limited number of cases. Kant also adds here a remark which has nothing corresponding to it in the Transcendental Exposi- tion of space : he says that principles like those referred to have the force of rules which lay down the conditions of all possible experience ; i.e., he not only argues that such 94 TRANSCENDENTAL AESTHETIC judgments imply the a priori character of time, but he also points out that they are presupposed in all our ordinary experiences of phenomena. In the ease of space he confines himself to showing that the science of geometry requires us to presuppose that space is a priori : here he says, not only that such scientific judgments as " time has but one dimension " presuppose the a priori character of time, but that they are implied in every possible experience of either outer or inner sense. He has here indicated a point which he could only explain fully in the Analytic. 31 a Kant goes on to show that certain apodictic principles are also synthetic, because time is a perception. As he has already argued more than once, no synthetic judgment can be obtained by the mere analysis of a conception; and hence, if time were a conception, it would be impossible to account for the synthetic character of certain apodictic principles. In illustration of this side of his doctrine Kant refers to the familiar idea of change, his object being to show that the consciousness of change is impossible unless we presuppose that time is a pure perception. If we remove the idea of time, change, as he argues, is inex- plicable. This is especially evident if we take motion, which is itself a specific form of change ; for, with the removal of the idea of time, motion will involve the con- tradiction that the same object is at once here and not here. This contradiction is resolved, when we add the qualification that an "object" means that which is in time; for, when this qualification is added, the contradiction disappears, since we can now say : At moment a this body is here, and at moment 6 this body is there. Kant further points out that the doctrine that time is a pure perception must be presupposed as the basis of pure physics; for pure physics consists in a priori synthetic judgments in regard to motion, and such judgments can only be explained on the supposition that time is an a priori perception. TIME 95 6. Inferences. 31 J {a) As in the case of space Kant applies the conclusions reached in regard to time in refutation of the three false theories, held respectively by Newton, Locke and Leibnitz. Time is not itself a real thing ; for, if it were, we must suppose it to exist independently of our perception, whereas it has been shown to be a form of perception, which is relative to the consciousness of perceived objects. Nor is it a property or a relation of real things ; for, as in either ease it would not be a necessary condition of our perception, no a priori synthetic judgments could be based upon it. On the other hand, the doctrine that time is the condition under which alone perception is possible for us makes it intelligible how we can derive from it judgments which are at once a priori and synthetic : a priori, because they are determinations of the pure form of perception ; and synthetic, because they express the character of the universal condition of all perception. 32(1 (S) Time, then, is a form of perception. It must be observed, however, that it is directly only a form of inner sense. By its very nature it in no way characterises ex- ternal phenomena, as having shape, position, etc., but merely expresses the relation to one another of the states of the subject. It is for this reason that time, which in itself is a vanishing series, is figured as a line proceeding to infinity. This of itself indicates that time is not a conception, but a perception. The idea here suggested is further developed in the Refutation of Idealism, where it is argued that we should never become conscious of time, were it not that we can contrast the persistence of objects in space with the evanescence of the states of the subject. 32 6 (c) Time is the formal a priori condition of all phenomena, whereas space is merely the pure form of external pheno- mena. For, while it is true that external phenomena do not fall beyond consciousness, Kant here asserts that they 96 TRANSCENDENTAL AESTHETIC yet must present themselves as modes of consciousness, or appear under the form of tnne. Thus time, unlike space, is a condition of external as well as of internal phenomena. The content of external phenomena, as Kant thinks, implies the form of space ; but as that content must appear in the mind as an idea, it falls under the form of time. The consciousness of external phenomena as in time is therefore not direct, but indirect. This explains why, in the pre- ceding paragraph, it is said that time " cannot be a determination of outer phenomena as such." 33 a As in the case of space, Kant points out that, as time is merely the form of our perception, it cannot be predicated of things in themselves. Since it exists only for beings in whom there is no knowledge except under sensible con- ditions, by its removal knowledge is rendered for them impossible. It by no means follows that time is an arbitrary determination ; it is a imiversal and necessary condition of human kno^vledge, and in that sense is objective. Thus, while it is inadmissible to say without qualification that all things are in time, it is perfectly true that all knowable things are in time. 34 a As a necessary condition of our perception tune is empirically real. Without the capacity of ordering the sensible in time, we could have no experience. But, just because time is the subjective condition of experience, it cannot be predicated as a determination of things in them- selves. Thus, while we affirm the " empirical reality " of time we must also maintain its " transcendental ideality," meaning by this that it is neither a real thing, nor a property or relation of real things, but merely the manner in which objects are presented to us under the necessary conditions of our experience. 7. Explanatory Remarks. 34 h The critical view of time, when it was first put forward by Kant in his earlier work, the Dissertation, was immediately TIME 97 objected to by Lambert, Mendelssohn and Schultz. Kant here refers to the main objection made by these thinkers, and especially to the form in which the objection was stated by Lambert and indeed almost in that writer's own words. The objection is : Whether external changes are denied or not, at any rate it must be admitted that we are directly conscious of changes in our own mental states. To deny these changes, indeed, would be to destroy all consciousness. But we cannot be conscious of the changes in our own mental states without presupposing time. Therefore, the reality of our consciousness involves the idea of time. Time is therefore real. 35 a To this objection Kant answers by pointing out that it rests upon a misimderstanding of the Critical theory. The writer assumes, that to affirm the transcendental ideality of time, i.e., to deny its application to things in themselves, is the same thing as denying that time is real in any sense. But, says Kant, I do not deny the reality of time, in so far as time is a form of our consciousness. It is the real form of inner perception. Time is the mode in which the mind orders its ideas, and this function of ordering ideas is undoubtedly real. But its subjective reality as a function does not establish its objective reality. On the contrary, as the whole argument of the Aesthetic proves, time cannot be a determination of things in themselves, if we are to show it to be by its very nature the condition of a priori synthetic judgments. We have, therefore, to affirm the empirical reality of time, i.e., its reality as the condition of all experience for us, but we have just as emphatically to deny its application to things in themselves. We cannot, therefore, say that things in themselves undergo changes. It does not follow, as Kant afterwards points out, that things are therefore unchangeable. The truth is, that neither the predicate changeable nor the predicate unchange- able has any application to things in themselves, because both changeability and permanence are meaningless except 98 TRANSCENDENTAL AESTHETIC as determinations of things in time. It may be added that, 5 54 in the next paragraph (not translated in the Eoctracts), Kant refers to the source of the objection made by Lambert. Adopting the ordinary view that we are directly conscious only of our own ideas, and indirectly of things in space, which are assumed to lie beyond consciousness, Lambert naturally supposed that the changes in our own states stand on a different footing from changes in the external world. But, as Kant afterwards points out in his Refutation of Idealism, it is not true that we are first conscious of ideas in our own minds, and then infer the existence of objects beyond our minds corresponding to them. On the contrary, we are conscious of our own ideas only in contrast to the consciousness of things in space : for, taken by themselves, our states form a mere series, and the consciousness of that which is permanent is requu-ed in order to give us the consciousness of our states as fleeting. The critical view does away with the opposition between ideas in our minds and objects lying beyond our minds. " External " objects are objects determined spacially, but these exist only for consciousness, being conditioned by the form of space. ; There is, therefore, no fundamental distinction between our ideas as in time and objects as in space. "We are just as directly conscious of external things as of our own states, and the moment we see this, we also see that the whole basis of Lambert's objection disappears. 35 6 Kant goes on to give a statement of the general results of the Aesthetic. In the beginning of section 2 {Extracts, page 23), he enumerated the possible views in regard to space and time that might be held ; now he recapitulates the answer to the problem there raised, pointing out that space and time, as a priori perceptions, are two sources of knowledge, from which a variety of a priori synthetic judgments may be derived. Such judgments are obviously exhibited in pure mathematics, or rather pure geometry, which is the science of space and the relations of space. Kant, however. TIME 99 immediately makes the transition from pure to applied mathematics. Time and space are not only the condition of geometry, but also of the application of mathematics to concrete sensible experience. It is also to be observed that, just because they are the a priori conditions of sensibility, they cannot enable us to establish any a priori synthetic judgments in regard to things in themselves, but are necessarily limited to the sphere of phenomena. This, however, in no way weakens the stability and certainty of pure and applied mathematics, because the whole of the judgments employed by the mathematical sciences are concerned solely with actual or possible objects of experi- ence. For, the limitation to phenomena does not in any way affect the fact that within the sphere of phenomena they are necessary and universal determinations. They are necessary and universal determinations, because they are inseparably bound up with our forms of perception. i6o It may be asked whether space and time are the only principles of a priori sensibility. In one sense no answer can be given to this question ; i.e., we cannot derive space and time from any principle more general than themselves. Just as Kant afterwards argues that we cannot show why the pure conceptions of thought should be just so many and no more, so he holds that no reason can be given why space and time are the only pure forms of perception. On the other hand, we can show from the very nature 'of our experience that there are no other a priori principles of sensibility. At one time Kant was doubtful whether pure motion, the basis . of what he calls Phoronomy, or what has more recently been called Kinematics, should not be included in the a priori principles of sensibility. But though, as he here points out, motion is a synthesis of space and time, it cannot be regarded as a pure form of perception, because it has no meaning except in relation to the sensible observation of moving objects. For, motion presupposes the knowledge of empirical data. It may be objected that pure motion does 100 TRANSCENDENTAL AESTHETIC not involve any sensible element, and that it would therefore seem to be an a priori principle. Kant's answer is that in what is called "pure motion" there is no motion in the sense of motion of the parts of space. Nothing moves, in fact, except body. When the mathematician speaks of the motion of a point, for example, he is thinking of the pro- duction of a line: which is no motion of an object, but merely the ideal motion of the subject. Motion proper is, therefore, always the transition of a sensible object in space, and consequently it is not one of the a priori principles of sensibility. It is even more obvious that change cannot be put among the data of Transcendental Aesthetic. Change necessarily involves sensible experience, whether it takes the form of a mutation of the properties of the object — as, for example, in the case of chemical change — or of a succession in our own states of consciousness ; for in both of these cases empirical elements are implied. The conclusion, therefore, is that as motion and change — which, next to space and time, have the most obvious claims to the rank of a priori principles of sensibility — must be excluded from the Tran- scendental Aesthetic, no other elements but space and time can be regarded as pure forms of sensibility. General ReTnarks on the Transcendental Aesthetic. 36 6 (1) Kant here refers to a distinction which has been made familiar by Locke. According to Locke there is a fundamental distinction between the primary and secondary qualities of things ; the former being actual determinations of external objects, the latter being merely subjective, and depending upon the senses of the percipient. The primary qualities are those which are bound up with extended body or matter: the secondary are sensible qualities, such as colour, taste, smell, etc. Now, the object of Kant in this paragraph is to distinguish between the doctrine of Locke, in which a contrast is drawn between things in themselves GENERAL REMARKS 101 and things as they are for the subject, and his own doctrine, which rests upon the opposition between things in them- selves and phenomena. Kant is willing to accept the distinction between primary and secondary qualities, if it is interpreted in a way consistent with his special theory ; but he cannot accept Locke's view of it. What Locke calls the real object, or in Kant's language the thing in itself, is only a phenomenon; and to that extent its properties stand upon the same level as the secondary qualities, which Locke contrasts with those of the real object or thing in itself. In other words, the distinction between the empirical thing in itself and the appearance of it is a distinction within phenomena, not a distinction between phenomena and things in themselves. Kant indicates the character of what Locke would call a real object, (a) It must exhibit what is essential and necessary to the perception of an object of experience ; (&) it must contain what is valid for every human being ; (c) it must be related to sensibility in general; and (d) it must possess the same character, no matter from what point of view it is perceived. An appearance, on the other hand, is just the opposite, (a) It is not essential, but attaches merely accidentally to the perception of the object ; (/8) it is valid only for particular persons, and especially for particular persons in abnormal states (as for example, in the case of jaundice) ; (7) it is true only in relation to the special senses ; and (S) it is observed only from a particular point of view, or when a sense has a particular organization. As an instance of the distinction Kant refers to the rainbow. From the ordinary point of view the rain, that is, the round material particles, is regarded as the real object, while the rainbow is looked upon as an appearance. This distinction, however, is quite different from that between phenomena and things in them- selves ; for, while it is true that the rainbow is only an empirical appearance, that is to say, is only observable from a certain point of view and under certain peculiar conditions. 102 TRANSCENDENTAL AESTHETIC it is no more an " appearance " {ErscheinuTig) in the critical sense of the term than are the raindrops. The distinction, then, between the primary and secondary qualities is quite different from the distinction between phenomena and things in themselves. The former is a distinction within phenomena ; the latter is a distinction between phenomena and things in themselves. 37 a (2) In confirmation of the Critical view of space and time Kant refers to certain views current in natural theology. What he here says must not be regarded as contributing to the proof that space and time are a priori perceptions. The argument is of the nature of an argumentum ad hominem ; in other words, it is an argument which has force only for those who admit the main positions of natural theology. How, then, is God conceived, both in his exist- ence and in his knowledge, to be related to space and time ? This problem was keenly debated in the controversy between Leibnitz and Clarke, and it is not improbable, from what we know of the development of Kant's thought, that he was partly led to his philosophical doctrine by reflecting upon the difficulties involved in it. There are three points here referred to as maintained in natural theology, (i) God cannot be an object of perception for us ; in other words, our knowledge of God cannot be obtained through sensible perception, inasmuch as he is not a sensible object or phenomenon, (ii) God is not an object of sensible per- ception for himself: in other words, God's perception of himself cannot be limited by the conditions of the inner sense, as our knowledge of ourselves always is, i.e., God cannot know himself in a succession of states, or, what is the same thing, under the form of time, (iii) God's know- ledge of things must be perceptive, not discursive, for we must suppose that his knowledge is absolutely perfect. Now, thought never of itself comprehends reality, for no real knowledge of things in themselves can be derived from any analysis of conceptions, and thought in all cases operates GENERAL REMARKS 103 with conceptions. This limitation to conceptions makes it impossible that thought should have a direct know- ledge of the object. Hence it is always necessary, in order to have any real knowledge, that we should go beyond it. There must be, as Kant has pointed out, something which forms the mediating Irak between thought and the object On the other hand, while God's knowledge cannot be of the nature of thought, neither can it be of the nature of perception, as perception exists for us. His knowledge must, indeed, be perceptive in this sense, that it must be a direct contemplation or comprehen- sion of real existence ; but it caanot be perceptive under the forms of space and time. Now, the natural theologian is careful to point out that the perception of God must be free from the limits of space and time. This he does, because, were it not so, God's knowledge would be limited to what is present at a given time, and what is visible at a certaia point of space. If we accept these deliverances of natural theology, and particularly if we admit that space and time are not limits of reality, what possible theory can be advanced in regard to the nature of space and time except that they are forms of our perception ? If they were actual forms of things in themselves, they must also be limits to the existence of God ; for, as Kant argues, we must in that case regard them as absolutely inseparable from existence as it truly is, and therefore absolutely inseparable both from God's knowledge and from His existence. The difficulty disappears, however, when we admit the Critical view, that space and time are the subjective forms of our perception. Sensible perception, as it exists in us, is dependent upon the existence of the object, and therefore it is possible only if our perceptive consciousness is affected by the presence of the object. It is in fact a characteristic of all finite or dependent beings, that both in their knowledge and in their existence they are dependent upon things distinct from them. The only kind of perception in which this 104 TRANSCENDENTAL AESTHETIC limitation is not present must be one which is original; in other words, it must be a perception which is not dependent upon the existence of an object apart from perception, but one in which the object originates in the perception. This is further explained in the next paragraph. 38 a The question may be raised, whether space and time are forms only of human perception, or are also forms of the perception of all finite beings. Kant refers to this point because it had already been raised by Mendelssohn. His answer is, that, while we have no positive means of determining whether all finite beings perceive under the conditions of space and time, there is nothing to hinder us from supposing that it may be so. In any case all finite beings must be dependent in their perceptions upon an object that is not originated in perception, and if it is the case that there are other forms of perception besides ours, these at least cannot be forms of things in themselves. The perception of every finite being must be limited, inasmuch as the forms under which it exists are forms of sensibility, which is always receptive. "We may express this by saying that sensuous perception is always derivative, in contrast to divine or intellectual perception, which must be original. By this we mean that intellectual perception is not dependent upon or derived from separately existing objects, but is of such a character that the objects perceived are originated or created in the act of perception. Thus, if we say that the form of God's knowledge is that of intellectual perception, we must at the same time say that objects flow from God himself. Kant's point is, that the existence of God is possible only if we distinguish His knowledge in principle from ours. The very idea of God implies that He knows all reality, and in His existence is absolutely self-dependent. The argument, therefore, is that such knowledge and such an existence are only con- ceivable on the hypothesis of a being for whom to know is at the same time to create. The fundamental distinction CONCLUSION 105 between intellectual and sensuous perception is, that in the former that which is perceived is originated by the perceiving subject, whereas in the latter what is perceived exists apart from the perceiving subject and is in some way apprehended by him. This distinction of an intellectual from a sensuous perception is referred to in various sub- sequent passages. Conclusion of the Transcendental Aesthetic. t59 a Kant now refers back to the question of the Introduction, section 6 {Extracts, page 18), where the problem of tran- scendental philosophy is declared to be : How are a priori synthetic judgments possible ? This problem has partly been answered as a result of the discussion in the Transcendental Aesthetic. We have now obtained one of the elements which enable us to solve it, viz., the conditions under which the a priori synthetic judgments of mathematics are possible. The problem was to explain how we can have a priori judgments, in which a predicate is ascribed to the subject which is not analytically contained in the conception of that subject, or, as we may also state it, to find out what is the X or middle term enabling us to connect legitimately the predicate with the subject, that predicate iuvolving a determination tran- scendiug the subject. This X we have found to be the pure forms of perception, space and time. We are now able to defend the a priori synthetic character of all mathematical judgments. It must be observed, however, that such judgments by their very nature have a limited range. As based upon perception, they have no meaning except in relation to objects of our sensible experience ; in other words, we must not suppose that we have justified the application of mathematics to all possible objects, but only to all possible objects of experience. TRANSCENDENTAL LOGIC. 1. General Logic. 40 When Kant passes from Transcendental Aesthetic to Transcendental Logic, he finds it necessary to modify and supplement the conclusions reached in the former. The problem of philosophy, as he states it in the Introduction, section 6, is, to answer the question : How are a priori synthetic judgments possible ? This question, as he further pointed out, divides up into two: (1) How is pure mathematics possible ? and (2) how is pure physics possible ? The Transcendental Aesthetic purports to be an answer to the first question. As a matter of fact it deals with a somewhat wider problem ; for, in maintaining that space and time are a priori forms of perception, Kant is led to point out that as such they are also the conditions of aU sensible perception as well as of all pure perception. When he now passes to the Second part of his task, the problem which has thus been widened is the one with which he is really occupied. The Transcendental Logic, or rather the first part of it, the Transcendental Analytic, ought to be an answer to the question, How is pure physics possible ? that is, according to Kant's first statement of the problem, it ought to be confined to a justification of the a priori synthetic judgments constituting the content of pure physics. Kant, however, finds that the answer to the one problem really involves an answer to the other, and indeed GENERAL LOGIC 107 that they cannot properly be separated. For a priori forms of conception, as he now goes on to argue, are limited in their application to objects of sensible experience, just as are the a priori forms of perception, whUe on the other hand in their application to such sensible objects they have objective reality. The problem, then, with which Kant is really occupied in the Transcendental Logic is to explain the a priori conditions in thought of all possible experience, and thereby to show the limits within which these concep- tions are applicable. 40t( In the Introduction, section 7 (&;a The object of Analytic, as the first part of Transcendental Logic, must be to show that there are pure a priori con- ceptions, and that there are pure a priori judgments based upon them, which determine the conditions under which all sensible experience is possible. Naturally, therefore, the first problem of Transcendental Analytic will be to discover the pure conceptions of understanding, to show that they are pure forms of thought, and to exhibit them as the necessary and universal conditions of certaia a priori synthetic judgments. 5ft Now, if we are to give a true account of the a priori forms of understanding, we must be able to put our hands on all the pure conceptions, and this again implies that we must seek for them on some definite plan or principle. After Kant had convinced himself that space and time are a priori forms of perception, the problem pressed itself upon him to explain such universal and necessary judgments as that " all events must have a cause." This problem he was compelled to face, because Hume had shown that the principle of causality cannot possibly be universal and necessary, on the assumption that it is a generalization 120 TRANSCENDENTAL ANALYTIC from experience. Kant, as he tells us himself, proceeded to ask, whether there are not other principles besides causality, which have the same character of universality and necessity. Thus he was led to generalize the problem of Hume, and in the first instance to enquire how many a priori principles there are. In this difhoulty he remembered that Aristotle had given a list of categories, i.e., of certain universal ways in which the mind judges, and he therefore proceeded to ask whether any light could be thrown on the problem of the discovery of the pure conceptions of the understanding by an examination of the categories enumerated by Aristotle. Now, the categories of Aristotle are substance, quality, quantity, relation, place, time, situation, condition, action, passion. It is, however, evident on the surface, that place, time, and situation are not categories at all, in Kant's sense of the term, i.e., they are not pure forms of thought, but, as the Aesthetic has shown, forms of perception. These three, then, must at once be set aside. Moreover, action and passion are not pure forms of conception, but are obviously special applica- tions of the principle of causality. Condition, again, is not so much a category as a characteristic of all forms of thought. There remain sfiohstance, quantity, quality, and relation. To these must obviously be added causality, the special conception which was brought into prominence by Hume. If these categories are to be arranged in the order of their logical complexity, it is plain that substance must come after quantity, quality, and relation, and must be associated with causality. This gave Kant his starting point, and accordingly he begins with quantity, quality, and relation ; but he places substance and causality under the head of relation, adding to these the category of reciprocity. After further reflection he added a fourth heading, viz., modality. To understand how he was led to add modality, and to arrange his categories in the order in which they appear, we must remember that. ANALYTIC OF CONCEPTIONS 121 after he had rejected the scheme of Aristotle, he turned to formal logic for light, on the ground that it exhibits the principles which underlie the whole activity of thought, in so far as it is analytic or purely formal. It was, therefore, by accepting the conclusions of formal logic that Kant was finally led to give us the table of categories which he sets forth, and this table, as he maintains, is derived from a definite principle, the principle, namely, that all thinking, whether analytic or synthetic, is invariably judgment, i.e., it is some form of introducing unity by the spontaneous activity of thought. Section I. — The Logical Use of Understanding. -' What, then, is the character of thought, ia so far as it is exercised in the formation of analytic judgments ? As we have seen above, imderstanding is, negatively, a non- sensuous faculty of knowledge; that is to say, it is a faculty which does not supply any determinate element of reality. Positively, the understanding as distinguished from the sensibility is a spontaneity. And this spontaneous faculty of understanding, since it is not perceptive, can only consist in operating in some way upon the material supplied by perception. In other words, the understandiug always works by means of conceptions. Now, it is the character of a conception that it presupposes perceptions, and hence understanding is always an indirect or mediating process of knowledge. It operates, not directly with objects, but only with conceptions which are relative to perceptions, and the one function of thought is to connect conceptions with each other, or, it may be, to connect a perception with a conception. The use which the understanding makes of conceptions is to judge by means of them, and we therefore may properly say that all thinking is judgment. And since the materials which judgment uses are conceptions, the judgment is always the indirect knowledge of an object. 122 TRANSCENDENTAL ANALYTIC Sectio-n II. — 9. The Logical Function of Understanding in Judgment. 48 a The essential activity of understanding consists in judg- ment. If, therefore, we fix our attention upon the act of judgment, considered purely by itself, we shall be able to discover what are the various functions of judgment in so far as judgment is analytic. As these functions have already been set forth by formal logic, we can take its results, and, startiug from them, proceed in our task of deriving from them the functions of judgment implied in the synthetic judgments. Kant's principle here is, that understanding, as the faculty of judgment, must have a certain fixed constitution, and that, just as we cannot have any perception irrespective of the forms of space and time, so we can have ' no thought of objects irrespective of the fimdamental forms of the understanding. In the various functions of formal judgment we shall therefore be able to detect the varioxis functions of synthetic judgments. The difference between them will consist, not in the mode of operation of the understanding, but in the kind of material which the under- standing employs. In the case of analytic judgments the materials are abstract conceptions, in the case of synthetic judgments the materials are sensible elements; but the manner in which thought operates with this varying material is necessarily the same, because the understanding cannot divest itself of its fundamental constitution. Here then are the functions of judgment; — (1) Formal logicians divide judgments into universal and particular. Thus, " All men are mortal " is a universal judgment, i.e., the quantity of the subject is taken universally or as a whole, and the predicate " mortal " is asserted to apply to every member of that whole. Again, " Some men are wise " is a particular judgment, because the subject, " Some men " is not co-extensive with the whole class " man." To these two forms of judgment, the universal and particular, Kant ANALYTIC OF CONCEPTIONS 123 adds the singular judgment. Formal logic does not make this distinction, but argues that such a judgment as " Socrates is a man" does not differ in form from the universal judgment; since, though the judgment has reference to a single individual, that individual is taken as a whole or universally. Why, then, does Kant add this form of judg- ment ? He teUs us that his reason for doing so is that, though in formal logic there is no ground for distinguishing the singular from the universal judgment, when we come to look at the function of thought applied in relation to real constitutive judgments, it is necessary to make this distraction, and to set down the singular judgment as a form by itself. What this means is simply, that Kant is not able to derive his categories from the abstract form of the analytic judgment, but has to modify formal logic in anticipation of its requirements for transcendental logic. (2) Quality. A similar remark applies in the case x)f quality. Formal logic distinguishes judgments as to quality into affirmative and negative, and indeed as Kant himself says (Extracts, p. 53), in formal logic all a priori division must be by dichotomy. The infinite judgment he adds, for the same reason as that which led him to add the singular judgment : it is required in anticipation of his subsequent derivation from it of the category of Umitation. The infinite judgment may be illustrated by the proposition, "The soul is not mortal," which may also be put in the affirmative form, " The soul is immortal." Here " the soul " is not only denied to belong to the class of "mortal things," but it is af&rmed to belong to the class of " immortal things." The characteristic, then, of an infinite judgment is that it at once excludes from a given class and thereby includes in the opposite class. (3) Relation. Formal logic distinguishes the categorical, hypothetical and disjunctive judgments ; so that Kant here found ready to his hand all that was required to serve as a basis for the thr6e corresponding categories. (4) Modality. The same remark applies in the 124 TRANSCENDENTAL ANALYTIC case of modality, with its subdivision of problematic, assertoric and apodictic judgments. Section III. — 10. The Pure Conceptions of Understanding or Categories. 49 a Formal logic abstracts from all content, for it deals purely with the forms of analytic thought. Transcendental logic abstracts from all empirical content, but not from the content of pure perception. This content, as has been shown in the Transcendental Aesthetic, consists of the determinations of space and time ; and Kant seems to say here that such determinations are so presented to thought that thought proceeds to bring them to conceptions, for he speaks of Transcendental logic as having " lying before it a complex of a priori sensibility." It has, however, to be added, that thought cannot directly operate with the deter- minations of space and time, until these determinations have been in some way combined and prepared for it. It is true that in the Transcendental Aesthetic Kant has spoken as if nothing more were required to explain the a priori synthetic judgments of mathematics than the consideration that space and time are the a priori forms of all perception, or all sensible experience : that is to say, he has there spoken as if individual objects, as existing in space and time, were revealed purely through perception. But the doctrine which he now wishes to establish is, that the perception of individual objects involves a certain niode of synthesis ; that is, it involves the spontaneous activity of the mind. Hence he tells us here that " the complex content of pure perception must first be surveyed, taken up into thought and combined, before there can be any know- ledge." Now, this can only mean that even the imagination of individual objects, or of specific determinations of space and time, implies the spontaneity of thought. Kant, in fact, is here suggesting what he expressly says in the ANALYTIC OF CONCEPTIONS 125 immediate sequel, that the synthesis of imagination is pre- supposed as the basis upon which explicit thought operates with its pure conceptions or categories. This doctrine is afterwards developed into a complex theory, under the title of "the schematism of the understanding." Synthesis, then, is essential as the condition of any knowledge whatever, no matter whether that knowledge has to do with so-called sensible objects or with pure perceptions. If we are dealing with pure perceptions, the synthesis is called pure, whereas if we are dealing with sensible objects, the synthesis is empirical. Kant further points out that the analytic activity of thought cannot be primary, because no analysis of a conception into its elements can possibly be made unless these elements have previously been combined. If there is a definite content in our consciousness, that content may indeed be present to us in a crude and confused way, and may therefore stand in need of analysis ; but, on the other hand, there would be no content whatever, unless the elements had been previously put together, and this of course means that there must always be synthesis prior to analysis. It is to synthesis, then, that we must direct our attention, if we wish to discover the true secret of knowledge. The first form in which synthesis operates is through the imagination, which Kant afterwards calls the effect of the understanding upon the sensibility. What he refers to is not the reproductive imagination, but the productive. Here thought operates blindly or unconsciously, combining together the elements of the sensibility into a unity, whether those elements are empirical or pure. Only when this preliminary synthesis has been completed, can the understanding with its pure conceptions operate in the constitution of objects of knowledge ; and hence the synthesis of the imagination has, in Kant's words, to be " brought to conceptions," i.e., has to be converted into the 126 TRANSCENDENTAL ANALYTIC express determination of the images of perception as pre- supposing pure conceptions of the understanding, this being the condition of the system of experience. 50 6 If we look at the pure synthesis of thought in its most general aspect, we can see that it rests upon pure con- ceptions of the understanding. Thus, if we are operating with the pure elements of perception, as already determined by the synthesis of imagination, in the manner in which arithmetic operates with them, it is obvious that we must presuppose a certain conception as the form under which the concrete element supplied to the understanding operates. Arithmetic is based upon some conception, — ordinarily upon the conception of the decade — but whatever the conception may be, it is impossible to construct a science of arithmetic without referring the pure elements of number, the pure units, to a conception of some kind. Here, then, we have an instance of the manner in which thought operates with the material supplied to it in the productive imagination. 50 c Transcendental Logic has to show how the pure synthesis of imagination is " brought to conceptions." There are three elements employed, as Kant points out. (1) There is the complex content of pure perception, which we must now regard as an unconnected manifold, or mere multi- phcity, that is not in itself even an object of perception or imagination; (2) the element supplied by imagination, which unites into one view what previously was the unconnected manifold of pure perception ; and (3) the element supplied by the understanding, i.e., the pure conceptions which reduce the blind or unconscious unity of imagination to the explicit unity of the understanding. 51a Now, Kant has to show that the pure conceptions of the understanding may be legitimately derived from an examination of the pure forms of the analytic judgment. What, then, is the character of the analytic judgment? It operates with general conceptions, and brings these into ANALYTIC OF CONCEPTIONS 127 relation with one another, either in a judgment or in a syllogism. Here Kant has in his mind the connection of two general conceptions, or it may be a perception and a conception, as expressed in the analytic judgment. The predicate of such a judgment is obtained by analysis of the subject, for the analytic judgment never adds anything to what is already known, but only states explicitly what is tacitly implied in a given conception. Thus the analytic judgment first separates a certain attribute implied in a conception, and then predicates it of that conception. Here then undoubtedly understandiag, analytic as it is, connects together or combines two distinguishable elements. But understanding must have a certain fixed constitution, ' and, if so, the very same mode of operation of the under- standing will be implied in that prior synthesis, which, as we have already seen, all analysis presupposes. In operating with abstract conceptions, as understanding does in the formation of analytic judgments, the mind necessarily employs certain principles of unity, and those same principles it employs ia the synthesis of perceptual elements by which objects of perception are formed. Thus the very same acts of thought which are employed in the determination of objects of perception are also employed in the constitution of a priori synthetic knowledge. The pure conceptions of understanding, which are the functions of unity, are therefore fitted to determine the universal character of objects a priori, inasmuch as they are the necessary functions of unity which are inseparable from the very constitution of the imderstanding. ii h Following out this principle, Kant finds that there is exactly the same niunber of pure conceptions of the understanding as there are functions of thought in analytic judgments. It must be so, if those functions constitute the essence of the understanding ; for, as we have seen, thought essentially consists in judgment, and, though in the case of analytic judgments we are operating with abstract 128 TRANSCENDENTAL ANALYTIC conceptions, we must nevertheless bring into play those functions without which we cannot judge at all. Assuming that formal logic has correctly specified the various ways in which the understanding is employed in the analytic judgment, we shall be able to derive the whole of the pure conceptions or categories from the table of judgments already given {Extracts, p. 48). Table of Categories. 51 c (1) Now, under the head of qimntity formal logic specifies universal, particular and singular judgments. {a) In such a judgment as " Man is mortal," the quantity is said to be universal, because the predicate "mortal" is affirmed of every member of the class " man." The category corresponding to this subsumption of all iadividuals under one conception must be totality. No doubt the universal is not an abstract idea, but a combination of perceptual elements, nevertheless the function of thought will be fundamentally the same. (&) A particular judgment, such as " Some men are wise," divides up the abstract conception into its logical parts, and the corresponding category will therefore be a number of separate perceptual elements regarded as several or manifold. The category then is plurality, (c) In the singular judgment, such as " Socrates is a man," we are not dealing with a general or abstract conception, but rather with an individual. In the analytic judgment no distinc- tion is made between the singular and the universal judgment, because the predicate is affirmed of the subject without qualification. But, argues Kant, the function of unity presupposed in the singular judgment must be made explicit when we are dealing with synthetic judgments, because here we have to see the object in the making, so to speak. Hence the function of thought implied in this form of judgment is unity. Taking the categories in the reverse TABLE OF CATEGORIES 129 order — probably with a view to his subsequent remark that the third category in each class owes its origin to a union of the second with the first — Kant enumerates the categories of quantity as unity, plurality and totality. (2) In the analytic judgment, quality has three forms, — the affirmative, the negative and the infinite judgment, (a) Now, when thought is synthetic, and is therefore constituting perceptual elements into objects, it seems obvious that the function presupposed in the affirmative judgment is the determination of an object of perception as a reality, (h) Similarly, the negative judgment must yield, when it is interpreted from the synthetic point of view, the category of non-reality or negation. Negation, it must be observed, is not the mere absence of reality, but the negation of a certain given or limited reality, (c) Then, lastly, the infinite judgment, which excludes a conception from one sphere and puts it into another, yields the category of limitation, which is just a synthesis of reality and negation. (3) Relation. — (a) A categorical judgment affirms directly or without limitation. Xow, when we are dealing with actual objects of experience, simple or unconditional predication must consist in attributing properties to a substance : hence our category is inherence and subsistence, (l) The hypothetical judgment takes the form, " If A is B, then C is D " : it does not affirm without qualification, but only under a condition : it does not assert that C is D, but only that if A is B, then C is D. Hence, when we are dealing with actual objects of experience, conditionality, or the dependence of one element upon another, must take the form of real dependence or conditionality, and real dependence or conditionality will be the relation of cause and effect, (c) In the disjimctive judgment, which takes the form " A is either B or C," we have a whole of conception, together with subordinate conceptions, which in their totality constitute the whole. For, in the I 130 TRANSCENDENTAL ANALYTIC disjunctive judgment, " A is either B or C," B and C must exhaust all the possibilities : otherwise the disjunctive judgment would not be valid. When this principle of the reciprocal exclusiveness of two conceptions which together constitute a totality is applied to real objects of experience, we must have the mutual exclusion and yet relation of real substances, and this is the category of community or reciprocal causation. (4) Modality. — (a.) The first form of the modal judg- ment is the problematic, in which it is not asserted that " something is," but only that " something may either be or not be." Thus, for example, we might have the judgment, " The world may or may not have been created." Now, when this principle is applied to possible objects of experience, we get the categories of possibility and impos- sibility, (b) The assertoric judgment, again, asserts a connection of ideas without any limitation. The function of thought in the synthetic judgment will, therefore, be the comprehension of a real object of experience as existing or not existing, i.e., as an actual object of experience, or as one that has no actuality. The category, then, is existence and non-existence, (c) Lastly, the apodictic judgment asserts the absolutely necessary connection of two conceptions ; that is, its principle is that two given conceptions must necessarily be thought as correlative. And when we apply this principle to objects of experience, we get the categories of necessity and contingency, meaning by necessity the necessity of an object of experience, and by contingency the denial of such necessity. 52 a By following the guiding- thread of the foi-ms of judg- ment, as tabulated by general logic, we have thus been enabled to discover the pure conceptions of the under- standing, and to discover all of them. These pure conceptions are the functions of unity constitutive of the very nature of understanding ; without which, therefore, no knowledge of objects of experience can be obtained ; and TABLE OF CATEGORIES 131 we may have perfect confidence in the validity of oxir derivation, because the list of conceptions has not been picked up empirically, but has been derived from a simple principle, viz., the faculty of judgment. 11. ,2 6 (1) When we look at this table of categories, the first thing that suggests itself is that they fall naturally into two groups, which we may distinguish respectively as the mathematical and the dynamical categories. The former are concerned with " objects of perception " ; that is to say, they express the first constitution of an object of experience by the synthesis of elements into individual wholes. Take, for example, the categories of quantity, viz., unity, plurality and totality. By the synthetic activity of thought, elements of perception, which in themselves are a mere manifold, are combined under one of these categories, after they have been prepared for their subsiimption under the category by the synthesis of the imaguiation. The product, however, is the conscious- ness of a single object. Thus, to take an instance from pure perception, the category of unity determines the elements of perception as smgle, the category of plurality as manifold, the category of totality as a whole. Strictly speaking, as Kant immediately goes on to point out, it is only the category of totality which constitutes the object of experience, the remaining categories being rather the two subordinate phases of this complex category. Thus, e.g., before a line can be an object of thought, the perceptual elements or parts of the line must be gi\'en one after the other, represented as a whole by the imagination, and combined by thought into the consciousness of a single line. Tliis will explain what Kant means when he says that the mathematical categories are " concerned with objects of perception." Wliat he means is, that 132 TRANSCENDENTAL ANALYTIC these categories do not deal with the relation or con- nection of objects of experience with one another, but with those objects when they are viewed in themselves or separately. The dynamical categories, on the other hand, deal with the relation or connection of objects not taken singly. The first gi-oup of them — those set down imder the head of relation — are occupied with the inter- dependence of objects of perception on one another, and the main category here employed is that of reciprocal action. The other division of the dynamical categories — those placed under the head of modality — are occupied, not with the actual relations of objects so far as they are interdependent, but with the relation of objects to the subject that knows them. This is what Kant means by saying that the categories in the second group are " con- cerned with the existence of those objects, as related either to one another or to the understanding." The categories of relation are those which deal with objects as related to one another, the categories of modality those which deal with the relation of objects to the understanding. (2) Looking again at our list of categories, it is suggestive that the ni^mber in each class is three. The di^ision of conceptions made by formal logic naturally proceeds by way of dichotomy. AMiy, then, do we find that the categories are divided on a different principle ? The answer must be that the apphcation of the function of the under- standing to actual elements of perception introduces a modification, inasmuch as the understanding is not simply analyzing a conception and expressing it in a judgment, but is combining elements into a new whole. We have also to observe that the categories of each class are not related in the way of mutual exclusiveness. The third category is in all cases a synthesis of the other two. Thus, as Kant points out, totality is just plurality regarded as unity, limitation is reality combined with negation, community is causality in which two TABLE OF CATEGORIES 133 substances mutually determine each other, and necessity is just existence given by mere possibility. This suggestive remark of Kant is not further developed by himself, but it was taken up by Fichte, his immediate successor, and developed ia a more systematic way by Hegel. We have here ui fact the germ of the principle which underlies the whole of the Hegelian Logic, with this difference that Hegel not only finds this principle of triphcity, or the combination of opposite conceptions, manifested in each class of categories, but he makes it the animating principle by which an advance is made from the simplest to the most complex category ; so that Kant's ideal of a complete and systematic account of the functions of thought is attempted to be carried out by Hegel in a more con- sistent and thorough-going way. Chaftee II. — Deduction of the Categories. 13. Principles of a Transcendental Deduction. ia With the first chapter of the Analytic of Conceptions Kant has finished what he afterwards calls (section 26, Extracts, page 78) the "metaphysical" deduction of the categories ; in other words, he has shown, by an examina- tion of the logical functions of thought, that there are certain a priori forms of synthesis, which belong to the very constitution of the understanding. He has now to give a "transcendental" deduction, con-esponding to the tran- scendental exposition of space and time, just as the metaphysical deduction of the categories corresponds to the metaphysical exposition of the forms of space and time. He probably uses the term "deduction," instead of " exposition," because the pure conceptions of the under- standing have to be derived from the logical functions of thought, and because, when derived, their application to experience has to be justified or shown to be legitimate 134 TRANSCENDENTAL ANALYTIC In law it is customary to distinguish between the ques- tion of fact and the question of right. Thus, in point of fact a man may be in possession of a piece of land, but it does not follow that he has a just claim to it. And it may be shown, to apply this distinction to the question of knowledge, that as a matter of fact we do apply certain principles of the understanding in the determination of objects of experience, but we cannot at once pass from this fact to the conclusion that the application of those principles is legitimate. Instances of the actual use of such principles may readily be obtained from the law of causality. We assign a cause for heat and cold, for eclipses of the sun, and for innumerable other objects. The empirical explanation of our right to the use of such principles is simply that the connections are actually exhibited in our experience. This, indeed, was the doctrine of Locke, but, as was clearly shown by Hume, it could never warrant us in the conclusion that the principle so applied was necessary and \miversal. We must, tKerefoife, be able to show that the conceptions of the understand- ing are not only as a matter of fact employed in the determination of our experience, but that we have a right so to employ them. 54 a Kow, there was no difficulty in giving a transcendental exposition of. space and time, because it was easy to show that we can have no perception whatever unless we presuppose space and time as the a -priori conditions of perception. Since only by the application of these forms of sense could we be conscious even of concrete sensible things, we were enabled to prove that space and time are pure perceptions, and from this conclusion there was no difficulty in advancing to the further con- clusion that space and time are the condition of the possibility of our experience of sensible things. Having established this result, we were further able to show how a priori synthetic judgments of perception are possible. DEDUCTION OF THE CATEGORIES 135 j46 But this simple method of establishing the a priori and transcendental character of the forms of perception cannot be resorted to in the case of the pure conceptions of the understanding, for it is quite conceivable that objects might be presented to us even though those objects were not determined by conceptions of the understanding. We cannot say, at least in the first instance, that no sensible experience is possible without the application of the categories of thought. Here, therefore, the difficulty arises to explain, or justify, the contention that there are universal and necessary principles of understanding, based upon the a pnori functions of thought ; for thinking is ia all cases spontaneous activity, and the very fact of the existence of analytic judgments shows that we may think without thereby determining objects. The problem, therefore, is to explain how the spontaneous activity of thought should yet have objective validity. On the one hand, it seems as if we had experience of objects inde- jendently of the acti^^ty of thought, and, on the other hand, it seems equally evident that from such experience we cannot derive universal and necessary principles. The difficulty, therefore, is to show how these apparently contradictory statements may be reconciled. Can it be shown at once that we have experience of objects in some sense prior to the activity of the understanding, and that objects are necessarily conditioned by the activity of the understanding ? Take, e.g., the conception of cause. Such a conception is not based upon mere analysis : if it were, we should find in the conception of the cause the idea of the effect ; but the cause A is different from the effect B, and therefore B cannot be derived from A in an analytic judgment. Now, whenever we have a synthetic judgment, the difficulty arises to explain by what right we go beyond the conception from which we start to add to it what is not originally con- tained in it. Why should our conception of cause be 136 TRANSCENDENTAL ANALYTIC applicable in the determination of objects ? What entitles us to affirm that without that conception experience is impossible ? Why should it not be that the conception of cause is merely an idea in our minds, to which no object of experience corresponds ? 55 a It is no answer to this difficulty to say that the prmciple of causality may be established by an induction from fexperience. It is no doubt perfectly true that we are continually employing such principles as that of causality in the determination of objects of experience ; but, if this is the only basis for such principles, we shall never be able to show that they are universal and necessary, and therefore we shall never be able to show that the forms of thought which they presuppose are the a priori con- ditions of experience. 55 6 We have, then, to give a transcendental deduction of all the a priori conceptions of the understanding, and ob\'iously the principle by which we must be guided in that de- duction is that those conceptions are the a priori conditions of all possible experience. Just as in the Transcendental Aesthetic it was proved that the a priori synthetic judg- ments of mathematics can be explained only because space • and time are the a loriori forms of perception, so we . must be able to show that the pure conceptions of thought , are the a priori conditions of all possible experience. If we can establish this, there will be no difficulty in showing that the principles of the understanding, or the universal judgments which we make in regard to objects of ex- perience, are universal and necessary. But there is no other possible way of justifying them. Section II. — A priori Conditions of Experience. 56 a In the first edition of the Critique of Pure Eeason Kant gave what he calls in the Preface a " subjective deduction," meaning by that a justification of the categories from the DEDUCTION OF THE CATEGORIES 137 point of view of the operations of thought implied in the constitution of objects of experience. This subjective deduction, as he here indicates, might be dispensed with, inasmuch as the main object of a transcendental deduction • of the categories is to show that these are the universal and necessary conditions of objectivity, i.e., of a system of experience. It is, however, of advantage to give also a subjective deduction : (1) because understanding proper comes into play only after certain preparatory forms of synthesis have been in operation, and (2) because there is great difficulty in showing that the understanding can be a condition of the Icnowledge of real objects notwithstand- ing that objects are in a certain sense perceived prior to its operation. )b What has to be explained, then, is the actual systematic connection by the mind of all the objects of experience into a whole. Consciousness must, therefore, be itself a unity. If we supposed it broken up iato a number of distinct and independent states, it is obvious that we should never have knowledge in the sense of a connected system of perceptions. The perception of an object involves the presence to consciousness of various elements, which are viewed as a single whole or totality. This may be called synopsis; but it is evident that, since no distinctive elements can ever be present to the mind unless they have been distinguished and combined by the mind, synopsis must imply some form of synthesis. It is njot enough for the explanation of objects of perception to refer merely to the sensibility, though in the Transcendental Aesthetic Kant seemed to regard this as sufficient : for the sensibility is purely receptive, and as such it can yield only an uncon- nected manifold. With the receptivity must be combined the spontaneity of synthesis, and this synthesis, again, has three forms or objects: the synthesis of apprehension, the synthesis of reproduction, and the synthesis of recogni- tion. 138 TRANSCENDENTAL ANALYTIC 1. Synthesis of Apprehension in Perception. 57 a The whole of the contents of our consciousness come under the form of time, which is the universal form of sensibility. It is to be observed that time is the form of external objects and also of the internal series of feelings. Further, it is the form which is presupposed in particular sensible objects and also in space and time as determinate objects of our consciousness. And as time is the universal form of consciousness, obviously there can be no conscious- ness of objects unless every part of our knowledge or experience is brought into relation with the one time. The importance of this remark lies in the fact tha^ it determines the conditions of the knowledge of phenomena, as distinguished from things in themselves ; for what we have to account for is not the existence and character of things in themselves, but the existence and character of objects as they appear within our consciousness. This general remark has to be kept in mind when we are considering what is involved in the various aspects of synthesis. 67 6 Time, then, as the universal form of our consciousness, is presupposed in the perception of all objects. Whatever other elements may be implied in an object, whatever its sensible differences, there is one thing that is always implied, viz., that it is presented in our consciousness under the form of time. This is important, because it shows us what is the condition of the discrimination of the various elements implied in the perception of an object. Each impression no sooner arises than it disappears, and therefore, if we suppose consciousness reduced to a number of separate or discrete impressions, there will be no con- sciousness of these as separate and discrete. A consciousness limited to the impression of a single moment could not discriminate that impression from the consciousness of the successive moments; therefore perception implies not merely DEDUCTION OF THE CATEGORIES 139 impressions coming one after another, but the consciousness of these impressions as coming one after the other or successively. Further, not only is this consciousness of succession essential to the discrimination of one impression from another, but the various impressions must obviously be combiaed in the unity of a single consciousness of various successions as united into a single object. There- fore the various impressions as they arise must be grasped by the miud, and this is what may properly be called the synthesis of apprehension. It is " apprehension," because there is a direct consciousness of the elements constituting the manifold of an individual object ; and it is " synthesis," because this manifold is united in the one object of con- sciousness. 57 c What has been said in regard to the necessity of a synthesis of the sensible manifold applies equally to pure elements of perception, that is to say, to the manifold of space and time. Here also the consciousness of a succession of elements is presupposed ; i.e., even in the case of the determinations of space, — e.g., a line or triangle or circle — time, as the form of inner sense, is an element in the consciousness of the object, and the manifold of pure perception must be combined in the imity of a single object, — the line, triangle, circle ; so that here also we have a synthesis of apprehension, differing from that in the case of sensible objects only in the fact that it is a pure synthesis. 2. Synthesis of Reprodiiction in Imagination. 58 a We know from ordinary psychology that ideas which have been frequently associated tend to reproduce one another, so that iu the absence of the object from percep- tion the elements or ideas that have been given in previous perception reappear in consciousness. Now, if this associa- tion of ideas is a purely arbitrary or subjective process. 140 TRANSCENDENTAL ANALYTIC it is obvious that it can in no wav help us to determme objeds. The form of our consciousness, no matter what the particular ideas which appear in it may be, is always, as we have seen, a succe^on; but, so far as the mere empirical law of the association of ideas is concerned, we cannot discriminate a real or objective succession from an arbitrary or subjective succession simply on tiie ground that as a matter of fact ideas are associated ; for all ideas are associated, the wildest fancies as well as that connec- tion of ideas which we regard as objective. It is plain, then, that thei-e must be something more in the association of the elements which go to constitute the consciousn^s of a real object than mere association : in other words, the elements must themselves be somehow connected leather in the reproductive imagination in such a way as to enable us to refer them to. or regard them as indicating, real objects of experience. No matter, therefore, what may be the character of the special sensible elements involved in the conscionsn^s of an object, those elements cannot be con- nected in a purely arbitrary way. but must somehow or other be associated on the basis of a rule, TVere it not so, the empirical imagination would never come into play in such a way as to give us the consciousness of objects or facts, and without the consciousness of objects or facts we should not have the consciousness of fictions. 5S6 What, then, is the character of this rule which is pre- supposed in the reproductive imagination ? It must obviously be one which is implied in the very nature of the reproductive imagination itself: in other words, reproduction must be based upon some fixed constitution of the imaginative faculty, and therefore in its operation it must be the condition a priori of a necessary synthetic unity. Let us keep clearly in mind that in the explana- tion of experience we cannot fall back upon the assumption that t hi n g s in themselves have a certain definite nature, and that knowledge consiste in a correspondence subsisting DEDUCTION OF THE CATEGORIES 141 between our consciousness of objects and objects as they really are. For, whatever the things may be in themselves, they are for our knowledge nothiag until they are somehow brought into relation to our minds, and the only manner in which things in themselves can come iato relation to our minds is through the impressions of sense, which are subjective states in us, not determinations of things as they are in themselves. We have, therefore, to explain the consciousness of objects without going beyond con- sciousness, and the only way we can do so, is by showing that, while all our ideas come to us in the form of a succession in time, there is a certain order or rule in the manner in which our impressions are reproduced in the imagination, and that the consciousness of this rule is the necessary condition of the reference of our impressions to objects. What we have to show, then, is that we can have no image of an object, not even of a pure object, i.e., a determination of space or time, unless there is a fixed rule or order, imposed upon us by the character of our imagination. If it is the case that imagination is the bliud or unconscious operation of the understanding in its application to the reproduced elements of perception, then we can explain how we should speak of objects of experience. Kant, ia other words, seeks to account for the consciousness of objectivity by showing that it means, in ultimate analysis, the consciousness of a fixed rule in the reproduction of the elements of perception. Take, e.g., the case of reproduction, where we have a pure object of perception, say a line. The consciousness of a line of course presupposes that there is a direct apprehension of the parts or elements which go to constitute it. But this is not enough : even if we suppose those elements to be grasped by the miad in a unity, we shall not have the consciousness of a line; for a line is not merely this particular image now before me, but it is an image that exhibits the universal manner in which every line must 142 TRANSCENDENTAL ANALYTIC be produced. Therefore, not only must we have the various parts or elements of the line produced as we draw it, but we must be capable of reproducing these elements, and holding them before consciousness. This is the necessary condition of there being for us the consciousness of a line. If the prior elements dropped entirely out of our consciousness as the new elements appeared in imagination, it is plain that we should never have the consciousness of the one single object, the line. Now, the manner in which these elements are reprodiiced must be in accordance with a universal rule; otherwise a line might be produced by any kiud of combination of elements of perception. Thus, though the imagination works blindly or unconsciously, it must operate in conformity with the rule which is after- wards brought to explicit consciousness by the under- standing. 59 a. The synthesis of reproduction, then, is presupposed or implied in the synthesis of apprehension, and we have seen that the synthesis of apprehension, i.e., the con- sciousness of certain elements as distinguished from one another and yet following in succession, and of these elements as combined into a whole, is the necessary condition of the consciousness of objects. Thus we have discovered that there is a transcendental faculty of imagination, just as there is a transcendental faculty of the understanding; in other words, the universal rules, under which the imagination blindly operates, are the necessary condition under which the consciousness of objects is possible and only possible. It is true that the imagination is not the explicit conception of objects as conforming to universal rules, but we must hold that the imagination operates in such a way that when imderstanding explicitly subsumes perceptions under categories, it is simply doing clearly and consciously what the imagination has already done blindly and unconsciously. DEDUCTION OF THE CATEGORIES 143 3. Synthesis of Recognition in Conceptions. )« Kant has argued that for the consciousness of objects it is necessary that there should be, not only apprehension, but reproduction : that is to say, that there should not only be elements of sense grasped by the mind in their separate- ness, but that these should be reproduced, and reproduced in the original order. To the constitution of any image of perception this complex process is therefore required. But something more is needed before we can explain the consciousness of the object proper. The new element referred to is what Kant calls the "synthesis of recognition." It is required not only that the elements of perception should be reproduced in their original order, an order which must conform to a rule of the understanding, but it is further necessary that what is so reproduced should be recognised as identical. Apart from such recognition we could not have the consciousness of objects, that is to say, the consciousness of perceptions as coming under a rule of the understanding. Without the identity of consciousness in recognition there would be no unity in our knowledge, and therefore it would be impossible for us to be conscious of various determinations as constituting the unity of one whole. For example, supposing we are dealing with pure units of perception, as in the case of number : these units must be apprehended, and they must be reproduced according to a rule ; but, unless we can recognise that the units so reproduced are identical with the units originally appre- hended, we shall not be conscious of them as forming a sum. It is only ia so far as there is not merely a synthesis of units, but the consciousness of such synthesis, that we get the conception of a sum. For knowledge, then, the rule which is presupposed in the synthesis of reproduction must be before consciousness; in other words, we must become conscious of the unity of thought in the process of constructing the object. 144 TRANSCENDENTAL ANALYTIC 60 b What, then, is to be understood by an object ? We have seen that experience always involves impressions of sense, and that these, when referred to objects, constitute experience. But the objects to which impressions are referred cannot be objects lying beyond consciousness, for such objects have no meaning for our knowledge. What, then, do we mean when we speak of an object as corresponding to our knowledge ? Here we seem to oppose the object to our consciousness, as if the former existed independently of the latter. But, if knowledge never transcends consciousness, it is plain that such a view is untenable. Our question, then, is, What is the object for consciousness ? What is that which we contrast with our consciousness of an object, and which yet is within consciousness ? 60 c An object within consciousness, as distinguished from a thing in itself, must receive its character, not from anything lying beyond the circle of consciousness, but from something within consciousness itself. Now, we know that the mind does not originate the elements of sense, and therefore we miist seek for the explanation of objectivity in the form applied by the mind to the elements of sense. We have also seen that the consciousness of objects cannot be explained merely from the form of perception, even when it is combined with the elements of sense. The source of objectivity must therefore lie somehow in the form of understanding in its relation to the manifold of perception. Hence Kant says that "the unity which the object demands can be only the formal unity of consciousness in the ■ synthesis of its various determinations." When we say that we know an object, we mean that the understanding has combined various determiuations of perception into unity. An object, in other words, presupposes the subsump- tion of the manifold under a rule, and the consciousness of objects consists in the consciousness of the synthetic unity which is implied in this rule. When we speak of an object, we really imply the universality of a rule, and the DEDUCTION OF THE CATEGORIES 145 universality of the rule implies the synthesis of the under- standing. la Knowledge, then, in the sense of knowledge of an object, is impossible without conception. No doubt the conception employed may be indefinite and obscure, but it must always be present, since without it there can be no consciousness of an object. Thus, if we are conscious of any external thing or body, the conception of " body " is presupposed as a rule to which every individual sensible external thing must conform. And as objectivity involves in some sense the necessary connection of elements of perception in an iudividual object, or the necessary connection of objects with one another in one system of experience, there must be a transcendental condition lying at the basis of our knowledge of objects, or, what is the same thiug, our knowledge of the system of nature. This transcendental system or ground of all objectivity must be absolutely universal, i.e., it must be the precondition of every possible object of experience, and if we can find out such a condition of objectivity, we shall have answered the question: What is meant by the something = X which constitutes the object ? 51 J This absolutely original or primary condition of all objectivity is simply transcendental apperception. The term " apperception " was used by Leibnitz in contrast to perception, the latter being the unreflective consciousness of objects, the former beiug the consciousness of self as distinguished from and related to objects. Kant adopts these terms, and gives them the meaning proper to his own system. Apperception is for him identical with self-con- sciousness. What, then, is implied in self-consciousness ? The identity of self-consciousness is obviously the necessary condition of the consciousness of self as a luiity in its various determinations. But there can be no consciousness of self as a unity unless we presuppose that in its nature the self actually is a unity. This transcendental unity of apperception or self-consciousness cannot be accoimt^d for K 146 TRANSCENDENTAL ANALYTIC merely by saying that in every single mode of our consciousness we are implicitly conscious of self, for no absolutely universal and necessary principle can be established by a simple reference to experience as a matter of fact. We have, therefore, to seek for an explanation of self-consciousness, not by referring to experience as a fact, but by showing that it is the necessary condition of all possible experience. 62 a The transcendental or original unity of self-consciousness is obviously the condition of the unity of our experience; for, unless the various elements of our experience can be connected together in the unity of a single self, there can be no unity in our experience. On the other hand, it is true that there is no consciousness of the unity of self except in relation to objects of experience. But objectivity, as we have seen, implies the synthesis by thought of the elements of perception, and without this synthesis in its various forms there can be no conscious- ness of self as a unity ; in other words, we become conscious of self as a unity in the regress from the con- sciousness of objects. On the one hand, we can explain the unity in our experience only by presupposing the original unity of self ; but, on the other hand, we become conscious of the unity of self only in so far as we actually determine objects through the process of synthetic unity experienced by the understanding. 62 b The transcendental unity of self -consciousness, then, is the supreme condition of all objectivity. Only in and through it can we account for the systematic unity of experience; in other words, all the forms of synthesis which belong to the constitution of the understanding must be applica- tions or modes of this primary synthetic unity. ISTow, the modes of the understanding are just the categories ; hence the categories stand under the original unity of apper- ception. The identity of the self is presupposed in all ■^ knowledge as its absolute condition, and the consciousness DEDUCTION OF THE CATEGORIES 147 of the identity of self is thus shown to be involved in the consciousness of objects. The categories are simply the necessary rules which come to explicit consciousness when the mind not only constitutes objects by reference to the manifold, but becomes conscious of the rules which it has employed in such constitution. When, therefore, the rniud becomes conscious of the identity of self in all the modes by which it determines objects, the blind activity of the imagination is " brought to conceptions." The proof, then, of the objectivity of the categories is, that they are the modes of synthesis by which the under- standing constitutes single objects of experience and the systematic unity of experience under the supreme condition of the synthetic unity of apperception. As perception was explained in the Aesthetic as made possible by the pure forms of perception, viz., space and time, so the Analytic establishes the possibility of the system of experience by showing that the pure conceptions of the understanding are all forms of the one a -priori form of knowledge, namely, self -consciousness. 15. Possibility of any Combination Whatever. 3a The "objective" deduction, to the consideration of which we now proceed, is the form in which Kant states the transcendental deduction of the categories in the second edition of the Critique. He begins the deduction by poiuting out that we cannot explain the consciousness of objects from mere perceptions ; for, not only does consciousness involve more than impressions of sense, but it involves more also than the pure forms of space and time. Objectivity implies the consciousness of the unity of the manifold of perception, and therefore combination (conjundio). Now, combination is a spontaneous act of consciousness, and as such it is characteristic of under- standing or thought as distinguished from sense. An 148 TRANSCENDENTAL ANALYTIC act of understanding is, therefore, presupposed in the consciousness of objects, whether the object is a deter- mination of space and time itself or the determination of a sensible object in space and time. This act of combination is called "synthesis," and as it proceeds entirely from the spontaneity of the subject, it is the single element in the consciousness of objects that cannot be referred to perception. Whatever the mode of the combination of elements into a whole, the character of the act must be fundamentally the same; and, therefore, though the understanding no doubt in certain cases acts by way of analysis of given conceptions, yet this analysis presupposes synthesis ; for unless the understanding had previously combined the elements implied in the concep- tion, there would be nothing to analyse. The synthetic or combining activity of the understanding, then, is absolutely essential to the consciousness of any object whatever. 64 a The combination here spoken of must not be looked upon as the result of the action of thought upon given elements. It is not enough to say that in point of fact the under- standing always combines elements given to it, for this would only show that the understanding has the power of combination. If we are to explain the consciousness of objects, and to show that certain a ■priori synthetic judgments are implied in it, we must further be able to show that the synthetic activity of the understanding is bound up with the very nature of understanding. Syn- thesis, in other words, is not any arbitrary act of combining elements given b^^ perception, but it is presupposed as the necessary condition, without which there could be no distinction of such elements and no combination of them. Further, the unity thus presupposed in the consciousness of ol^jects is not to be confused with the category of unity, which is merely a special mode in which the universal com- bining activity of the understanding operates. The unity to which reference is made here is that fundamental unity. DEDUCTION OF THE CATEGORIES 149 without which there can be no objective synthesis. We have, in fact, to seek for the ultimate ground of the unity of various conceptions in the judgment, as implied in the logical use of the understanding. What, then, is this unity ? 16. The Original Synthetic Unity of Apperception. 5 a The unity is that which is implied in self-consciousness. We are entitled to say, that what is not capable of being combined under the unity of a single self cannot enter into experience. Hence Kant says that we must not only be conscious of various determinations, but we must be capable of accompanying these with the consciousness, "I think these determinations." It is no doubt conceivable that there should be a sensitive consciousness, which is not accompanied by the consciousness "I think" ; but if we call it "consciousness" at all, at least it cannot be called in any sense the consciousness of oljects. All the manifold determinations of perception, then, must necessarily be re- lated to the "I think" in the subject that is conscious of it. The consciousness, "I think," cannot be given to the subject, but must proceed from the spontaneous activity of the subject. It is called p^ire apperception, or pure self- consciousness, because it is the universal form which is necessarily presupposed in aU modes of consciousness whatever. It is, therefore, distinguished from empirical consciousness, inasmuch as the latter involves a particular relation to sense or feeling. It is also called original apper- ception, because it is the primary condition without which • there can be no self -consciousness whatever, and therefore no unity in our experience. And this "I think" is the only idea which occupies the position of being presupposed, ex- plicitly or implicitly, in every form of consciousness. Now, since the "I think" is thus absolutely a priori, it is properly called the transcendental unity of self- consciousness ; for a 150 TRANSCENDENTAL ANALYTIC transcendental unity is that unity without which no a priori synthetic judgment is possible. We have therefore to say that the absolute unity of self-consciousness is the supreme condition of all objectivity. It is no doubt true that the presence of this unity is not in all cases made an explicit object of consciousness ; indeed, in the preliminary operations by which the imagination prepares perception for the express work of the understanding, there is no distinct consciousness of self as a unity ; but, unless this preliminary work imphed the operation of the unity of self- consciousness, it would be impossible to have any unity in our experience, jffl All our ideas, then, must be consistent with the possi- bility of self-consciousness in regard to them. Nothing can come into our minds, and therefore nothing can exist for us • as an object of knowledge, which is not capable of being . brought under the unity of self-consciousness. We can, therefore, lay it down as an a priori condition, that the consciousness of objects is absolutely conditioned by the unity of self-consciousness. What, then, is involved in this unity ? There is no direct consciousness of the unity of self, when the self is taken in separation from the elements of perception and the synthesis by which these are brought under the unity of self-consciousness. The identity of apperception is reached only through the synthetic activity by which the manifold is combined into unity, and, indeed, only when the mind becomes conscious of its own activity as operating in this synthesis. The empirical consciousness is not in itself a unity, but only becomes a unity in so far as the variety of elements of sense is combined by the , understanding; and the consciousness of self, as presupposed in this activity of the understanding, is reached only when the mind becomes conscious of what it has itself done in thinking or combining the elements of perception. In bringing together perceptions, so as to combine them into the unity of a system, we become conscious of a single DEDUCTION OF THE CATEGORIES 151 world of objects. The analytic unity of apperception, as expressed in the judgment, "I am I," presupposes the syn- thetic unity; i.e., while, in Kant's view, we can reach the judgment, "I am I," by an analysis of the idea "I," this judgment could never have been reached at all, had not the understanding exercised its activity in reference to the manifold of perception, and had not the subject thus become conscious of the unity of self as expressed in the synthesis of the imderstanding. The synthetic unity of the various determinations of perception is thus the ground of the consciousness of that unity of self which is the necessary condition of every act of thought; and as combination is the work of the understanding, the unity of apperception which is presupposed in all combination is the condition of all knowledge of objects. 67(1 As we have already seen, the unity of apperception, taken by itself, is an analytic proposition ; i.e., the pure consciousness of self excludes all determinations, and therefore the self is necessarily always identical with itself. But this analytic unity of apperception presupposes the synthetic unity, since it is only in relation to the synthesis of various determinations that self-consciousness arises. This peculiarity is due to the character of our intelligence, which is so constituted that, while it can, and indeed must, combine elements presented to it, it yet is incapable of originating those elements themselves. Kant, therefore, again contrasts our understanding with a perceptive under- standing, the latter of which would not only combine elements given to it, but would originate the elements apprehended, and so in being conscious of self would at the same time be conscious of objects. This, however, is not the character of our inteUigence; for, while the conscious- ness of self as a unity is presupposed in the consciousness of the unity of objects, and ultimately of the unity of the whole world of objects, our understanding cannot operate except in relation to what is given to it in perception. 152 TRANSCENDENTAL ANALYTIC What we can say is, that since the original synthetic unity of apperception is the supreme condition of our experience \of objects, we can have no experience whatever except under the condition of this synthetic unity. 17. The Synthetic Unity of Apperception is the Supreme Principle of Understanding. 68 a In the Transcendental Aesthetic we discovered that there can be no perception of any sensible object which does not presuppose the pure forms of space and time as the condition of such perception ; now we have discovered the supreme condition of that consciousness of objects which is implied in the knowledge of them as entering into the system of experience. Just as all sensible objects must stand under the pure forms of perception, so we can have no knowledge of any system of objects except in so far as the elements supplied by perception are com- bined in the unity of one consciousness. The synthetic unity of apperception is thus the supreme condition of all experience. 68 6 Understanding, as distinguished from perception, may be said to be the faculty of knowledge. By " knowledge " is meant the consciousness of certain determinations as relative to an object : meaning by an " object," not any thing in itself, but the unity which is involved in the combination of perceptual elements into one whole. Now, all such unity of determinations implies the synthesis of the understanding, while the synthesis of the understanding presupposes the synthetic unity of self -consciousness ; so that the synthetic unity of self -consciousness is the primary condition of the consciousness of objects. 69 a By the " original synthetic unity of apperception " Kant means the consciousness of the identity of the thinking self, which is reached only through the synthesis of the understanding and the consciousness of that synthesis as DEDUCTION OF THE CATEGORIES 153 implying the unity of the thinking subject. Without this unity of apperception there could be no synthetic activity of the understanding, and without the synthetic activity of the understanding there could be no consciousness of objects of experience; for, in order to explain the con- sciousness of such objects, it is not enough to point out that there are pure forms of perception which are the condition of objects, since objectivity is possible only in so far as the elements of perception are combined in a synthetic act, and further only in so far as there is a con- sciousness of the identity of the act of combination. Thus the consciousness of the identity of the self, in the various forms of synthesis by which objects are constituted, is at the same time the consciousness of such objects. Apart from the synthetic unity of self-consciousness, then, there would be for us no objects of experience at all, inasmuch as there would be no synthesis of the elements of perception. 59 b But, though the synthetic unity of consciousness is presupposed in all thought, the unity of consciousness, as Kant repeats, is in itself an analytic proposition, which may be expressed in the formula " I = I." No doubt all the elements of perception must be capable of being united in one consciousness or referred to one self, and only in being so united is there the consciousness of self ; but the " I " of which I thus become conscious is in itself a pure imity or identity, containing no differences in itself. It is for this reason that self-consciousness is possible only through the synthetic activity of thought, which is drawn forth by the necessity of combining the elements of per- ception in a unity that the mind can grasp. "We might say that, while the self is a pure unity, the consciousness of self is possible only when differences are supplied to it, - enabling it to exercise its function of synthesis. 70 a While self -consciousness in us is thus conditioned by the • synthetic activity of the understanding, Kant again tells 154 TRANSCENDENTAL ANALYTIC us that we are not entitled to say that every possible under- standing must be of the same character as ours. It is because the pure consciousness of self as a unity does not supply any determinations which can be an object of our consciousness, that we are capable of self-consciousness only through the synthesis of the understanding. The reason why we become conscious of self only through the act by which the understanding combines the elements of perception into unity is that our self-consciousness can ! originate nothing out of itself. If our understanding were of such a character that it originated whatever came before it as an object, we should have a direct self-consciousness, i.e., a self-consciousness not mediated by the synthesis of the understanding. But such an understanding ours is not. Our intelligence, as Kant has previously said, is not perceptive : it always presupposes a given manifold and an act of synthesis, and indeed it is impossible for us to form any idea of an understanding that originates out of itself the determinations which are its object. Nor can we even form any idea of an understanding that is inde- pendent of space and time, inasmuch as the only possible determinations of objects which we ever experience are those which appear under the forms of space and time. 18. Objective Unity of Self -Consciousness. 70 6 The synthetic unity of self-consciousness is the supreme condition of the consciousness of objects. By "objects" is meant, not the variable and changeable elements of perception, or even the unchangeable elements of the forms of perception, but the fixed way in which those elements are combined by our understanding. As Kant has already pointed out (69a), without the synthetic unity of con- sciousness there is no object at all, i.e., no object of experience. Hence we must distinguish the " objective " unity of consciousness from the "subjective" unity. The DEDUCTION OF THE CATEGORIES 155 former has nothing to do with the determination of the elements of perception in their specific character. What these elements shall be, and whether they shall present themselves as simultaneous or as successive, cannot be determined a •priori. The elements of sense, which are referred to an object through the activity of the under- standing, may not only be different for different persons, but they may even vary for the same person at different times. But no matter what may be the variation in these sensible elements, the ways in which thought combines them under rules of the understanding cannot vary : they must be permanent. What can be determined a priori, then, are the universal and necessary conditions under which the elements of perception when supplied to the mind must be combined under the unity of self-conscious- ness. We can, therefore, say a priori, that there can be no consciousness of objects, and therefore no consciousness of one world of objects, unless the elements of perception are so presented that they are capable of being combined in one self-consciousness. 19. The Logical Form of all Judgments consists in the Objective Unity of the Apperception of the Concep- tions they contain. 71 o Kant, in this connection, refers to the ordinary defini- tion of judgment as " the consciousness of a relation between two conceptions." This definition, he says, is not satisfactory ; for, not only is it not true that every judgment is a relation of conceptions — the definition is not appropriate in the case of the hypothetical and disjunctive judgments, where it is not conceptions but judgments that are related — but above all, the defini- tion does not state wherein the relation consists. But when we enquire more closely what sort of relation subsists between subject and predicate in a judgment, we 156 TRANSCENDENTAL ANALYTIC find that judgment is simply the manner in which given ideas are brought to the objective unity of apperception; in other words, judgment is the way in which the elements of perception are combined in one self-consciousness. This is indicated by the copula of the judgment, which claims to express the nature of actual objects, and not merely the subjective association of ideas in the individual mind. A judgment is not merely the bare affirmation that certain ideas are present in the mind, but the affirmation that ideas are connected in such a way as to give us the consciousness of objects. Now, the supreme condition of objectivity, as we have seen, is the necessary unity of self- consciousness in the combination of given elements. This is true, whether the elements combined are pure or empirical, i.e., whether we are dealing with determinations of space and time, as in the mathematical sciences, or with particular sensible objects. Take, e.g., the judgment, "Bodies are heavy." Here the empirical or sensible elements are given to the mind, and given in such a way that they are not necessarily connected with one another. Kant says that, from the point of view of perception, we can only say, " When I lift this body I have a sensation of weight." He speaks here as if in perception there was already a judgment; strictly speaking, however, perception is not yet the con- sciousness even of the association of two ideas in the mind, inasmuch as there is no consciousness of objects without the combining activity of the understanding, and therefore no judgments. What Kant wishes to accentuate is, that, while the empirical elements supplied to the mind may come in any order, yet when these elements are combined under the rules of the understanding the order is absolutely fixed. The judgment, "This body is heavy," therefore presupposes the transformation of empirical elements into the consciousness of an object as having a certain deter- minate character. Every judgment, then, as we must conclude, expresses or implies a synthetic activity of the DEDUCTION OF THE CATEGORIES 157 understanding, and therefore the objective unity of apper- ception. 20. All Sensuous Perceptions stand under the Categories . as conditions under which alone their various Deter- minations come together in one Consciousness. 72 a Kant now sums up the argument of the transcendental deduction so far as he has yet gone. We have seen, firstly, that the elements of sensuous perception must stand under the original synthetic unity of apperception, or there can be no unity of perception, i.e., no consciousness of an object {11). But the act of understanding, by which the elements of perception are brought under the unity of apperception, is judgment (19). It is, then, through the various forms or functions of judgment that the elements of perception are determined in certain universal ways. Now, the categories are just those functions of judgment, when applied to the elements of perception (13). Therefore, the elements of perception must stand under the categories. 22. The Category has no other application in Knowledge than to objects of Experience. 73 a " To think an object is not the same thing as to know it," says Kant. In knowledge there are two elements : firstly, the conception or category ; and, secondly, the elements of perception, which it is the function of the category to unify. If therefore no elements of perception were given to the mind corresponding to the conceptions, i.e., of such a character that they could be combined by thought, we should never get beyond the mere thought of an object, inasmuch as there would be no definite elements to be combined into unity and so to constitute an object. The Aesthetic, however, has shown that perception in us is 158 TRANSCENDENTAL ANALYTIC always sensuous ; hence the thought of an object can only become the knowledge of an object, provided that the elements given to us are sensuous in their character. But sensuous perception is for us either the pure perception of space and time, or particular objects as presented in space and time. In order to obtain knowledge of objects, however, it is not enough that the pure elements of perception — that is, the determinations of space and time — should be combined by the understanding ; for, though in this way we have the consciousness of a single object, it is only the consciousness of a possible object, since space and time are but the forms under which sensible objects are perceived, and these forms have no meaning in them- selves apart from the sensible objects determined under them. It follows from this that the pure conceptions or categories can only yield a knowledge of objects when sensible elements are given to us under the forms of space and time. The categories, therefore, apply only to sensible objects of experience ; in other words, they are merely conditions of the possibility of empirical knowledge. 74 a What has been said above states the conditions under which knowledge of objects is possible, just as the Aesthetic pointed out the conditions of perception. IsTow, if it is the case that the categories apply only to sensible objects as presented in space and time, it of course follows that they have no application beyond the sphere of sensible experience. It is perfectly true that the pure conceptions of the understanding differ from the forms of perception in this way, that they are capable of applying to any possible object of perception, provided only that it is sensuous. This, however, does not enable us to extend our knowledge in the least ; for the pure conceptions of the understanding, as we have seen, have no application DEDUCTION OF THE CATEGORIES 159 except when sensible elements are given to the mind, and in our case no such elements are given except those that are conditioned by space and time. In themselves, then, the categories are mere forms of thought, or functions of synthesis, which have no objective reality, but are simply capacities of combining given sensuous elements in certain fixed ways. 5 a It is of course possible for us to think, or suppose, an object of a non-sensuous character, and to this object we can apply all the predicates that are involved in the negation of sensuous perception. We can say, e.g., that it is not extended, that it has no duration, that it is not subject to change. But such merely negative predicates do not enable us to characterize the conceived object in such a way as to make it an object of knowledge. Such an object can have no positive meaning for us ; nay, such an object cannot even be shown to be capable of existing ; for, in the absence of all concrete elements of perception, it is simply the idea of a possible object that may not be actual. "We must observe, then, that none of the categories which we employ in the constitution of objects of experience, such as substance or cause, has any meaning except within the limits of sensible experience. We can, e.g., think of a supersensible or free subject ; but such a subject we cannot characterize as a substance, nor can we determine it as an object which undergoes changes in the way of cause and effect. In short, when we leave the sphere of sensible experience, we cannot by the sole exercise of the under- standing, or rather of thought in any form, determine anything in regard to the existence or nature of objects. 24. The Application of the Categories to Objects of Sense. 6 a The very fact that understanding is capable of applying its pure conceptions to any object of perception, if only it is sensuous, shows that the categories in themselves are 160 TRANSCENDENTAL ANALYTIC simply forms of thought, in other words, that they in no way determine the specific character of the object. All the concrete elements of an object are therefore given. But the synthesis of these given elements stands under the unity of apperception, and it is only in so far as such unity is possible that we have any a priori knowledge ; in other • words, our knowledge is not a knowledge of things in themselves, but only such a consciousness of a connected . system of sensible objects as is compatible with the imity of self-consciousness. We can, therefore, predicate of ex- perience all that is presupposed in the necessary unity of apperception, though we cannot identify the system of our experience with the ultimate nature of things. The synthesis of the understanding is transcendental, because it is the necessary condition of all objectivity, i.e., of all a priori synthetic judgments which we are capable of making in regard to sensible objects ; but it is also purely intellectual, because, as Kant has repeatedly pointed out, the understanding can originate no concrete element what- ever. The understanding, then, is a spontaneous faculty of determining sensible objects. Now, the universal form of all sensible objects is time, which is the form of inner sense, and all determinations of consciousness, even those that imply spacial extension, must present themselves in the inner sense. The understanding is therefore directly related to the inner sense ; that is to say, it acts upon elements of perception, in so far as these are determined under the condition of time. And since time is the form of all sensible objects, the categories, in combining the elements of sense under the form of time, determine all objects of perception, whether these are external or internal, i.e., whether they are in space or are only events in time. The objects, however, which it thus determines are only pheno- mena, as distinguished from things in themselves. 77 a The direct action of the understanding upon the inner sense is what Kant calls the figural synthesis, or the DEDUCTION OF THE CATEGORIES 161 transcendental synthesis of the imagination, which he distinguishes from the intellectual synthesis, or the synthesis of the understanding proper. What is the difference between these two forms of synthesis ? The former, the figural synthesis, is that form of combination of the various elements of perception which presents the sensible in its relation to time. The figural synthesis, in other words, is the process by which the elements of sense are combined in relation to time. Naturally there are various forms of such synthesis, corresponding to the various functions of the understanding, but the figural synthesis is not an explicit determination of the elements of sense by reference to the pure conceptions of the understanding. It is the intellectual synthesis which " brings perceptions to concep- tions," whereas the synthesis of imagination merely prepares the material for the action of the pure understanding, but prepares it in such a way that it must conform to the understanding, when the understanding comes consciously into play. The figural synthesis, then, is the consciousness of elements of sense as combined into images, whereas the intellectual synthesis is the explicit consciousness of such images as combined by the categories in universal and necessary ways. ?76 The figural synthesis may therefore be called the transcendental synthesis of the imagination. Kant refers it to the imagination, because imagination is the faculty of presenting before the mind a particular object when the senses are not directly affected. Now, the condition of sensibility in us always is that it must conform to the conditions of perception, and this means in the present case that it must conform to the condition of time, which is the universal form of all perception. On the other hand, imagination is not, like sensation, a pure receptivity ; on the contrary, it essentially belongs to the spontaneous activity of the mind ; at least this is the case so far as the imagination is productive or pure. The imagination, L 162 TRANSCENDENTAL ANALYTIC then, while it does not originate the elements that enter into an image, does supply the form to those elements; that is to say, the imagination has certain fixed ways of combining the elements of an image into a whole. It is, therefore, a pure or « priori form, like the pure forms of perception. We must not, however, suppose that the syn- thesis of the imagination is something essentially different from the synthesis of the understanding. On the contrary, the synthesis of the imagination is the unconscious or blind activity of the understanding, in so far as the latter directly acts on the sensible elements given to the mind. The distinction between the two forms of synthesis is, that the productive imagination determines oljjects in the way of images, whereas the intellectual synthesis determines objects by bringing them under the necessary rules of the under- standing. The productive imagination must be carefully distinguished from the reproductive imagination. The latter is simply that arbitrary association of ideas in the individual consciousness which does not necessarily imply the consciousness of objects, although we must observe that even the reproductive imagination must so far con- form to the conditiorus of knowledge that it does not contradict the possibility of such knowledge. 26. Trafisce'fuljintal DedvxfAcm of the Categories as Employed in ExperieTice. IS a The w.etapjhysicol defJ.udion, although not under that name, has already been given in Chapter I. of the Analytic, and especially in Section III. 10. The substance of it is, that the categories are derivable from the universal logical functions of thought, and since they are implied in the very act of the understanding, which these logical functions exhibit, they are necessarily a 'priori conceptions. The transcendental ded^v/itvyn goes on to argue, that from the nature of these a priori conceptions we can understand how DEDUCTION OF THE CATEGORIES 163 upon them should be based a priori synthetic judgments in regard to objects in general. But, though Kant has already incidentally referred to the point, he now goes on to state explicitly, that the categories are not only conditions of a priori synthetic judgments in regard to pure perceptions, i.e., determmations of space and time, but they are also the condition of all a priori synthetic judgments, and therefore even of those which concern objects that are directly related to our particular senses. What has to be explained, then, is how there can be an a priori knowledge of sensible objects, which seem from their nature to be incapable of determination a priori. 79 a The first thing to be noticed is that the knowledge of a sensible object or a sensible event implies a synthesis of apprehension, i.e., the putting together of distinguishable elements of empirical perception, and that without such synthesis or combination of these elements into a unity, there would be no experience of sensible objects whatever. ?9 b Now, in space and time we have a priori forms of outer as well as inner perception, and of course the synthesis of inner perception must necessarily conform to the nature of space and time, because otherwise there could be no appre- hension whatever. But space and time are not merely forms of sensuous perception, i.e., they are not merely conditions of our apprehension of sensible things, but they are themselves determinable, and so are a possible object of definite consciousness. From their character as a priori, all the determinations of space and time are implicitly contained in these forms, and therefore all these determinations are a priori. We have to observe, however, that the synthesis which is presupposed in the knowledge of a determinate object is not given in these perceptions themselves, i.e., is not contained in the separate elements of pure perception, but is possible only through an act of synthesis. It follows from this, that whatever is perceived as in space and time must necessarily submit to synthesis. 164 TRANSCENDENTAL ANALYTIC And this synthetic unity can only be the combination of the elements of perception, which takes place in conformity with the categories, because the categories are just the fundamental forms of the understanding as exercised in relation to the elements of perception. Hence all synthesis, even though the elements combined are given to our senses, must stand under the categories ; in other words, the categories are conditions of the possibility of our knowledge of sensible objects. 80 a Take as an instance the perception of a Tiouse. The various sensible elements, or impressions of colour, weight, etc., involved in the perception of the house must stand under the form of space, and must be in harmony with the form of time. In order therefore to have the perception of the house, the sensible elements must be combined in conformity with the character of space and time ; in other words, the shape of the house must be drawn or presented as an image. Here, therefore, we have a variety of sensible elements, which, when combined by the imagination, give us the image of the house. But the synthesis of the imagina- tion is not in itself arbitrary : it must necessarily conform to the category of quantity; in other words, the sensible elements must be combined, not only in conformity with the nature of space, but in conformity with the category of quantity. And this means that we must have a synthesis of homogeneous units. If we fix our attention upon the pure act of synthesis itself, abstracting from the form of space, we have the pure conception or category of quantity, and it is to this pure conception or category that the synthesis of apprehension necessarily conforms. 80 6 The categories, then, prescribe a priori laws to phenomena, or, in other words, they form the connecting elements in the totality of possible objects of experience. But, since the categories are not derived from experience, and therefore cannot be said to be dependent upon expe- rience, the question arises, how we can possibly say that DEDUCTION OF THE CATEGORIES 165 experience or nature must adapt itself to them. How, in other words, can it be maintained that all sensible objects, which by their very nature are not originated by the understanding, should yet be compelled to conform to the understanding ? How can we show that, instead of finding nature combined in certain ways, nature as a system of known objects only comes into existence in and through the combination of the understanding ? 80 c The answer cannot now be difficult to give. We have seen, in the Transcendental Aesthetic, that all objects of which we are conscious must agree with the forms of perception. Similarly, as we now see, all objects must necessarily conform to the understanding and its a priori forms, because otherwise there could not be for us any system of objects. (Things in themselves, if we could know them, would no doubt be determined in their character entirely independently of our knowledge ; phenomena, however, are not things in themselves, but objects as they are known to our consciousness, and such objects cannot possibly be identical with things in them- selves.) Ajid phenomena must obviously be subject to the conditions under which we can comprehend them. Imagination connects the various elements of sensuous perception, and imagination is itself dependent upon understanding for the unity which it blindly produces by the synthesis of elements of perception, just as it is dependent upon the sensible elements for its concrete matter. As we have seen above, nothing can be an object of experience to us, unless it is taken up and combined through the synthesis of apprehension, while this synthesis itself presupposes the transcendental synthesis of the categories. It follows, that there can be for us no knowledge of a system of nature without the application of the categories to the elements of perception, under the conditions of space and time. (Pure understanding, how- ever, only determines nature in its universal character, i.e., 166 TRANSCENDENTAL ANALYTIC it determines the conditions under which all objects without exception must be brought. ) Special laws of nature, on the other hand, while they must conform to the categories, or, in other words, must be consistent with the system of nature, cannot be explained purely from the categories. These laws, as we shall afterwards see, are derived from the action of judgment under the pure idea of final cause. In the meantime what we have to observe is, that the special laws of nature, such as those of chemistry and physics, while they conform to the universal laws prescribed by the categories, can only be obtained by a special examination of experience. 27. Result of the Deduction of the Categories. 82 a No object can be thought without categories, no object can be known without the manifold of perception to which the categories may be applied. All our knowledge, there- fore, is of objects which in their character are sensible ; in other words, objects of experience. Hence there can be no a priori knowledge except of objects of experience. 82 6 Knowledge, then, applies only to objects of experience. But it does not follow from this, that all knowledge is derived from experience, i.e., from sensible perceptions. Pure perceptions enter into the constitution of knowable objects, and similarly pure conceptions are a condition of that systematic connection of things which we call ex- perience. Both belong to the constitution of the mind and are a priori. There are only two ways in which we can account for the harmony between the sensible data of experience and the conceptions which we form of objects: we must either say (1) that the conceptions conform to the sensible data of experience, or (2) that conceptions are the necessary condition of there being any experience. The former supposition is inconsistent with the nature of the categories, not to speak of the pure forms of perception ; DEDUCTION OF THE CATEGORIES 167 for the categories in their own nature, as the a priori forms of synthesis, are not dependent upon sensible percep- tion, but on the contrary have a much wider range. To derive them from experience would be a sort of generatio equivoca, or spontaneous generation ; in other words, it would contradict the principle, that perceptions without conceptions are blind. We cannot possibly derive a priori synthetic judgments from sensible experience, because the most that we can learn from such experience is the way in which sensible objects usually present themselves. Kant also rejects the preformation view of knowledge, i.e., he denies that sensible perception is of the same fundamental nature as conception, as was held by Leibnitz. It is not the case that perception differs from conception only in its greater obscurity : the difference is one of kind. Hence we must rather adopt what may be called the theory of an epigenesis of pure reason, — the theory, that while knowledge involves the elements of sensible perception, the application to these of the categories is not a mere extension or a mere clarification of perception, but introduces the new element of necessity and universal connection. The categories, then, as proceeding from the understanding, are the necessary condition of all experience. Short Statement of the Deduction. 82 c We may now summarise shortly the whole deduction. The categories, or pure conceptions of the understanding, are the necessary condition of all a priori synthetic judg- ments, and without them no experience is possible. They are the principles by which phenomena in space and time are determined; that is to say, the principles through which the manifold of sense is constituted into that connected system of objects that we call experience. This systematic connection of objects is ultimately dependent upon the original synthetic unity of apperception, which is the 168 TRANSCENDENTAL ANALYTIC principle implied in all the synthetic activity of the under- standing, in so far as that activity is applied to the constitution of objects in space and time. BOOK II.— THE ANAI.YTIC OF JUDGMENTS. Transcendental Judgment. 83 a Following the analogy of formal logic, Kant, having dealt with the conceptions, goes on to deal with the judgments, of the understanding. If we call the understanding the faculty of rules, i.e., the faculty which originates the pure forms of synthesis, then judgment will be the faculty of subsumption under rules; in other words, the faculty of deciding under what rule a particular kind of sensible object must be placed. For instance, the category of causality is the mode or form of synthetic unity by which events are connected in a fixed or necessary way, and transcendental judgment must tell us what is the condition under which events can be brought under the category or rule of causality. Transcendental logic differs from formal logic in this respect; for the latter, since it abstracts from all content, cannot tell us how any specific object should be determined; in other words, formal logic treats all content as precisely on the same level, while transcendental logic must be able to point out the conditions under which we can make objective judgments, and therefore it must be able to in- dicate a priori the case to which each rule or category may be applied. This superiority over all other sciences but mathematics arises from the fact that the categories relate to objects entirely a priori. 84 a The transcendental doctrine of judgment consists of two parts. The first part points out the sensible condition without which the categories cannot be employed. This is called the schematism of the understanding. As we shall immediately see, what Kant means is, that, while the THE ANALYTIC OF JUDGMENTS 169 categories in themselves are pure conceptions, and therefore independent of the particular object to which they are applied, no knowledge can be derived from them taken purely by themselves, since in us all knowledge is limited to objects of sensible perception. Hence the categories must be related both to the sensible matter of perception and to the pure forms of perception, and especially to the pure form of time, which is the universal form of all perception. The sensuous condition, then, of which Kant speaks, is the sensible as relative to time ; in other words, it is the form in which the imagination relates the sensible to the categories. The second part of the transcendental doctrine of judgment deals with the fundamental propositions which arise a priori when the categories are employed under the sensuous conditions specified. Chapter I. — The Schematism of the Categories. 84 6 After his usual method, Kant, in seeking to point out the sensuous condition under which the categories may be applied to objects, begins with the ordinary logical view of conceptions. An empirical conception, it is held, is derived from a number of particular perceptions by a process of abstraction, the conceptions being the grasp by thought of the points which are common to the various perceptions. It is obvious that in conceptions of this kind there is something homogeneous with the conception to be found in the perception ; in other words, since the conception has been derived from perception by a process of abstraction, there is no difficulty in restoring by the reverse process the characteristics that have been abstracted from particular things and referring them to particular things. For example, the conception oi a. plate contains in it the mark of roundness, and since this mark has been obtained by comparison of a number of plates and abstraction from their differences, the judgment, " A plate is round," is easily made, because it is 170 TRANSCENDENTAL ANALYTIC already presupposed in the process of abstraction by which the conception has been formed. 84 c A pure conception or category, however, has absolutely nothing in common with perception. If we take, for example, the category of cause, we find that, looked at purely in itself, it is simply the conception of the relation between a condition and that which is conditioned or dependent upon it. The question, therefore, arises, how it is possible legitimately to bring a sensible object under a pure conception or category, seeing that they have nothing in common. By what right are we justified in applying the category in the determination of an object of sense ? The necessity of answering this question compels us to have a transcendental doctrine of judgment. We must show how pure conceptions of the understanding may be employed in the actual determination of objects of sense, although objects of sense have nothing in common with those conceptions. 85 as Obviously there must be something to mediate between the pure conception and the possible object, something which is homogeneous on the one side with the category and on the other with the object of sense. And this mediating idea must be pure, because what we have to explain is how there can be a priori synthetic judgments in regard to objects of sense. The mediating element must therefore be at once intellectual and sensuous, inasmuch as it is to connect the pure conception, which is intellectual, with the object of perception, which is sensuous. The idea in question is called by Kant the transcendental schema. 85 6 A category is the pure form of producing synthetic unity in any elements of which we can be conscious as different. The question, then, is how the pure conception of under- standing can be brought into play so as to introduce unity into the elements of perception, and to introduce unity in such a way as to entitle us to make universal and necessary judgments. We know that time is the formal condition of all the determinations of the inner sense, and so indirectly SCHEMATISM OF THE CATEGORIES 171 the formal condition of all our ideas, since all our ideas, even those referring to what is spacial, must be presented in our consciousness under the form of time. Time is thus the a priori condition without which we can have no conscious- ness of objects at all. If we find that there is a universal and necessary way of combining the elements of sense in relation to time, we shall have discovered the manner in which the categories can be applied to objects of sense ; for this universal way of determining the sensible in relation to time is so far homogeneous with the category that it is universal, and therefore rests upon an a priori rule. On the other hand, the transcendental determination of time is to this extent homogeneous with the object of sense that without time there can be no consciousness of such an object at all. Thus we can understand how, if there is a universal and necessary way of relating the elements of sense to time, the category may be applied to phenomena, not indeed directly, but in and through the medium of the transcendental schema. 86 a The schema is a product of the imagination, but a peculiar product of it, since the object which it produces is not individual but universal, or rather, it is not the pro- duction of an individual object, but a certain universal method of producing an individual object. If we set down five points one after the other, and combine them, we have before us an image, that is, a singular or individual object. The schema of a number, on the other hand, is the method by which we produce five or any other number ; in other words, the schema is the idea of the successive combination of units into a whole in conformity with a conception. It is obvious that there is a difference between the image and the schema : for, while the schema enables us to realise the conception, it is very difficult in some cases to show that the image is adequate to the conception. If I have the con- ception of 1000, there is no difficulty in schematising it, but it is very difficult to show that the image of 1000 points 172 TRANSCENDENTAL ANALYTIC set down one after the other is identical with the conception. The schema of a conception, then, is the idea of a universal process of imagination by which a conception is presented in an image under the form of time. 86 6 It is schemata, and not images, that lie at the foundation of our pure sensuous conceptions ; in other words, the mathematical sciences work with schemata, not with images. In geometry judgments are laid down in regard to the triangle, which apply to any triangle whatever — right angled, obtuse-angled or scalene. Obviously the uni- versality characteristic of mathematical judgments could not be explained if we supposed the mathematician to be dealing with images. This difficulty, in fact, was pointed out by Berkeley, who denied that we can frame any idea of a triangle which is neither right-angled, obtuse or scalene, and who therefore maintained that the mathematician works with images, which " stand for " any triangle whatever. Berkeley's view, as Kant tacitly maintains, does not explain the universality and necessity of geometrical judgments, for no such judgments can be reached by induction from par- ticulars. Kant, on the other hand, by pointing out that it is the universal process of framing the triangle with which the mathematician deals, believes that he is able to explain how geometrical judgments can be universal and necessary. The schema, though it is a product of the imagination, exists only as a rule according to which the imagination works. This at least is obvious in the case of geometry. Schemata, however, are not limited to pure perceptions, but apply also in the case of the consciousness of concrete sensible things. Here in fact the image with which thought works, when it employs a general conception, even more obviously cannot be adequate to the conception. Every object of sensible perception is specific or individualised, containing as it does something absolutely peculiar to itself, and in this respect it differs from the pure figures of geometry, or the pure elements of number, which are always SCHEMATISM OF THE CATEGORIES 173 precisely the same. Hence the schema is even more necessary in the case of judgments having to do with special objects of sense. Here thought works with an empirical schema, a sort of monogram or outline of a particular kind of individual object, say, a dog, horse or house. This imaginative process of schematising a conception cannot be further explained, but we can at least say, that, while the image is a product of the empirical faculty of productive imagination, the schema is a product of pure a priori imagination, and indeed it may be called the general idea of the possibility of all images of a certain kind. This remark applies to schemata in general, but we have to observe that the schema of a category or pure conception of the understanding is such that it cannot be presented in an image at all. It cannot be so presented, because it is the schema of a pure conception which contains no sensuous element whatever. It is simply the pure synthesis of the imagination which conforms to a rule of unity expressed in the category. This form of schema, then, is a transcen- dental product of the imagination, i.e., it is the a priori condition of a priori synthetic judgments. The schema is a determination of the inner sense; in other words, it brings the sensible into relation with time, which is the form of the inner sense, and indeed the general form of all sense- perception. This determination of the sensible relatively to time is a necessary condition of knowledge, because apart from this process the categories have nothing to which they can be applied.' And inasmuch as the application of the categories to the sensible is the condition in us of self- consciousness, without schemata there would be no unity of apperception. Kant, therefore, argues that the possibility of self-consciousness is only explicable provided we pre- suppose this transcendental process of the imagination, as combining the elements of sense in conformity with the pure conceptions or categories, by relating them in certain universal ways to time. 174 TRANSCENDENTAL ANALYTIC 88a When we are dealing with so-called "external" objects, the pure image must be some determination of space, smce space is the form of outer sense. The pure image, again, of all objects of sense, inner as well as outer, is time ; i.e., the determination of time as containing successive moments i is the image of all objects of sense. Quantity, however, as a conception of the understanding, is the general idea of any unity of the manifold. It is to be observed that Kant, in treating of the schematism, does not in all cases give a schema for each of the categories. In the case of quantity and quality the schemata correspond to the third of the subordinate categories under each head, i.e., to totality in the category of quantity, and to limitation in the category of quality. Here he tells us that number is the schema in the category of quantity. More precisely, it is the schema of the category of totality. Number does not here mean number in the arithmetical sense, but the universal process of adding homogeneous unit to homogeneous unit in a successive synthesis, whether the units so added are numbers in the arithmetical sense or geometrical objects. The schema of number, then, is simply the universal process of combining elements of pure perception, so far as these are homogeneous, into unity, and this process implies the generation of successive moments of time in the act of combining the homogeneous units into a whole. 88 h From the point of view of the pure categories of quality, the category of reality is the conception of purely positive or affirmative being, the category of negation the conception of purely negative being or the conception of the negation of positive being. These two categories are, therefore, as categories mutually exclusive, since the principle of thought in itself is the principle of non-contradiction. The third category of quality, viz., limitation, differs from the other two, inasmuch as it involves their combination. Strictly speaking, as Kant himself virtually admits, the category of limitation cannot be derived from the analytic SCHEMATISM OF THE CATEGORIES 175 judgment. "We must, therefore, regard it as the pure form of the synthesis of being and non-ieing. It is to this category of limitation that the schema properly applies. Kant, indeed, speaks of the category of reality as "that which corresponds to any sensation whatever, and therefore of that, the very idea of which is that it has being in time." But, strictly speaking, the category, as a pure conception, cannot be defined as determined by relation to time, since every pure conception is in itself capable of a wider application than to that which presents itself in time. The schema of limitation is degree, or that in the object of experience which corresponds to sensation ; in other words, there is no possibility of the knowledge of any real object, unless in so far as there is involved the matter supplied by sensation. The schema of degree is, therefore, the deter- mination of the quantity of sensation. For sensation is of this character, that it is not an extensive magnitude; that is to say, each sensation must be conceived as occupying only a single moment of time. The quantity of sensation is, therefore, not extensive quantity, and the only other form of quantity is that of intensive quantity. The kind of reality implied in sensation is thus that of degree or intensive magnitude. Every sensation fills a given moment of time with more or less of itself, or, in other words, it occupies the inner sense with more or less completeness. Hence, the degree or intensive magnitude of a sensation can never be either zero or infinity; that is, the sensation must always occupy a given moment of time in some degree, though it can never occupy it with absolute completeness. The schema, then, in this case is the idea of a moment of time as occupied by a determinate or limited degree of sensation. ;J9a The category or pure conception of svhstance, taken in its absolute generality, is the conception of that which is always subject and never predicate ; and therefore it is the idea of any reality whatever which is not predicable of something 176 TRANSCENDENTAL ANALYTIC else. But since for knowledge every category must be schematised, that is, brought into relation to time in some way, the schema of substance can only be the idea of such a subject and predicate as can be an object of our experience : that is, it must be of a subject and a predicate that exist relatively to one another in time. Now, that which corre- sponds to the pure conception of a subject, when it is determined as an object in time, is substance, that is to say, the permanence or persistence through time of that which nevertheless undergoes changes in time. The schema of substance, then, is the relation of a permanent thing to its changing qualities or activities. 89 6 The category of cause is the conception of ground and consequence, or condition and conditioned ; in other words, the logical dependence of one thing upon another. It may be expressed in the formula, "If A is, so also is B." But the pure conception can only be applied in the determination of objects of knowledge when it is schematised, and, as in other cases, the schema must imply the relation of the sensible to time in some way. The special form which the schema here assumes is that of ordered succession in time; so that cause and effect as known in our experience necessarily imply the regular or ordered succession of phe- nomena in time. In other words, a cause is always an antecedent, an effect is always a consequent ; and the antecedent and consequent are related in this way, that without the former the latter cannot be. 89 c The pure category of community is the conception of the relation of various members of a logical division to one another, these members being so related that they determine in their union the total sphere of the conception or genus. The scheroa of community must therefore involve time, inas- much as knowledge is only possible of that which is capable of presenting itself in time. And we find that the schema in this case is the reciprocal relation of objects that co-exist. SCHEMATISM OF THE CATEGORIES 177 i39rf Passing now to the categories of modality, we have, firstly, the categories of possibility and impossibilitij. The conception or category of possibility is simply of that which does not involve a contradiction, as contrasted with that which does involve a contradiction. Thus, we cannot conceive of an object as being determined by contradictory predicates : it cannot at once be and not be : it cannot have two qualities which are mutually exclusive. Expressed in terms of time, the category becomes the schema of possibility. Here we can have opposites or contradictory qualities in the same object, but we cannot have them at the same time. The schema of possibility, therefore, is the idea of that which is compatible with the general conditions of time, the schema of impossibility that which is incompatible with the general conditions of time. The conception of actiuility is simply the notion of the existence or non-existence of some deter- minate reality. The schema of actuality, on the other hand, is the idea of a determinate thing as existing at a given time, just as the schema of non-existence is the idea of the non-actuality of a determinate thing at a given time. Lastly, the category of necessity is the conception of that which cannot but exist, the category of contingency of that which need not necessarily exist. The schema of necessity, again, is the idea of the existence of an object through all time, the schema of contingency the existence of an object which is fleeting or does not exist through all time. Wa The detailed treatment of the schemata, as corresponding to the categories, has shown that in every case the schema is relative to time. The schema of quantity is the con- sciousness of an object as generated by a successive synthesis in time, a synthesis which at the same time is the genera- tion of time itself; the schema of qxmlity is the synthesis of sensation as filling a given moment of time ; the schema of relation is the relation of different perceptions to one another at all times or in conformity with a rule for the determina- tion of time ; lastly, the schema of modality, in its three M 178 TRANSCENDENTAL ANALYTIC forms, is the idea of the relation of objects to time itself, whether that relation is the relation to time in general, to a specific time, or to all time. The schemata are, therefore, the modes in which the categories are determined in relation to the sensible elements of perception by reference to time. We may say, shortly, that experience exhibits either the series of time, the content of time, the order of time or the comjjrehension of time. 90 6 The schematism of the understanding, then, is just the way in which various determinations of perception are brought to unity in the inner sense through the tran- scendental synthesis of imagination, and thus are fitted to be received into the unity of apperception. It is only through the schemata that the categories come to have significance, that is to say, are employed in the actual determination of objects. The fact that this schematising of the categories is essential to our knowledge shows that in the end the categories, so far as our knowledge is concerned, are limited to objects of possible experience. The categories are the modes in which thought by its pure spontaneity combines the elements of perception under the universal rules of synthesis, and therefore it is only by means of the categories 91 a that we can have a system of experience ; in other words, the system of experience is coterminous with the totality of our knowledge. 91 b Since, however, the categories are applicable to objects of experience only in and through the schemata, they are not employed in their full significance, but are in actual use necessarily restricted to sensible objects. The conditions under which the categories are employed are not conditions of the understanding in its pure operations, but depend upon the relation of the understanding to the sensibility. It is true that the schema must be in harmony with the category, but, on the other hand, it implies a limitation of the cate- gory, the result of which is that the objects we know are not things in themselves but phenomena. If the categories SCHEMATISM OF THE CATEGORIES 179 in themselves have a wider application than to objects of sensible experience, it is naturally asked why we cannot by means of them determine the existence of things in them- selves. It is true no doubt that the categories are forms of synthesis, which are not absolutely restricted to the sensible as given to us : but this possible extension is for us valueless, so far as the extension of knowledge is concerned, since we have no other elements to which we can apply them than those presented to us under the conditions of space and time. Thus, the category of substance, taken absolutely, means that which can be conceived only as subject, never as predicate ; but an object corresponding to this conception we can obtain only in so far as our sensibility is excited, and therefore only in so far as the category of substance is schematised as the permanent in contrast to the changeable. The conclusion, then, is, that the categories are functions of the understanding, which enable us to con- stitute objects only in so far as they are brought into relation with the sensibility, a relation which necessarily restricts them to objects of sensible experience. Chapter II. — Peinciples of Pure Undeestakding. 92 a Having laid down the conditions under which the cate- gories are applicable to objects of experience, Kant goes on to state what are the universal propositions or judgments which are presupposed in the a priori synthetic judgments of experience. What he has to do, therefore, is to give an explicit statement of what is implied in the consciousness of the synthesis of the understanding as working through the schemata. These fundamental judgments, or principles of understanding, he classifies as (1) axioms of perception, (2) anticipations of observation, (3) analogies of experience, and (4) postulates of empirical thought; and subsequently he divides them into two classes, called respectively the mathe- matical and the dynamical categories. Kant might have 180 TRANSCENDENTAL ANALYTIC called his discussion of the mathematical principles a transcendental deduction of mathematics, inasmuch as what he seeks to prove is that mathematics contains a ^priori synthetic judgments, and that these judgments are applicable to all possible objects of experience. We may say in fact that his aim is to give a deduction or justification of pure and applied mathematics. 1. Axioms of Perception. 92 b The principle of these is : All perceptions are extensive magnitudes. 92(3 An extensive magnitude is one in which we become conscious of a whole only through the combination of parts given prior to the whole. We cannot have the idea of a line, however small it may be, without drawing it in thought ; i.e., we must produce it part after part, starting from a certain point, and combine the parts thus successively produced into the unity of a line. So we can only be conscious of a space of time, however small it may be, by the successive generation of one moment after another, and the combination of the parts so generated into a single whole. What Kant is here pointing out is, that we have the consciousness of a pure object of perception, whether it is an external or internal object, only in so far as we bring into play the schema of number. There is no object of perception which does not involve either space or time, or both, and inasmuch as the pure element involved in the object can only be obtained by a successive synthesis of part with part, it is plain that we can have no knowledge of any object whatever, unless that object is an extensive magnitude. This of course does not mean that there are no objects which are not magnitudes, but only that we can have no knowledge of objects except those that are extensive magnitudes. Notice that by extensive magnitudes Kant does PRINCIPLES OF PURE UNDERSTANDING 181 not mean solely external objects, objects presented in space, for a succession of events as occurring only in time is also to be conceived as an extensive magnitude. 93 a The science of geometry, which is a determination of the properties of spacial figures, obviously presupposes the schematising of pure perceptions as number ; in other words, the successive synthesis of elements of space and the consciousness of such synthesis. The universal and necessary judgments of geometry — its axioms, as Kant calls them — rest upon this process of combining the elements of pure perception into a smgle whole. Thus, the definition of a straight line as the shortest distance between two points is a universal and necessary judgment, resting upon the synthesis of the homogeneous elements of the line, a synthesis which is successive. AU the universal and necessary judgments of geometry are, therefore, at the same time synthetic. 936 The case is somewhat different in the mathematical science of arithmetic. Here the propositions, while they are a priori and synthetic, are not universal, i.e., the subject of such judgments is not universal but singular. 7 + 5 = 12 is certainly a synthetic, not an analytic judgment. It differs from such a judgment as, " If equals be added to equals the wholes are equal," for here the very conception of equals — the very conception of the quantity generated — involves the judgment that they are equal. Arithmetical judgments, on the other hand, necessarily imply a synthetic process, inasmuch as each unit must be produced one after the other and combined into the whole. There is, therefore, a distinction between such arithmetical judgments and the universal propositions of geometry. The judgment, e.g., that a triangle contains two right angles is applicable to every possible triangle ; i.e., the subject of the judgment is universal, whereas the subject of the judgment 7 + 5 = 12 is singular. We may say, then, that while both arithmetic and geometry contain a priori synthetic ^<52 TRANSCENDENTAL ANALYTIC judgments, the former differs from the latter in containing singular, not universal judgments. 94 a A very important consequence follows from the proof or deduction of the principle that "all perceptions are extensive magnitudes " ; for, not only is it true that all the judgments of pure mathematics are universal and necessary, but it is also true that the judgments of applied mathematics are of the same character. Whatever is true of space and time as pure conceptions is from the nature of the case equally true of phenomena in space and time. Geometry is applicable to all objects of external perception, that is to say, to all objects that are presented to us as in space. Arithmetic is applicable to all objects of perception whatever, whether these are in space and time, or only in time ; and the justi- fication of the application of mathematics to all passible objects of experience is that those objects are not things in themselves, but phenomena. We can have no experience except of objects that are determined as spacial or temporal; and, hence, whatever mathematics says of pure perception must be applicable to these objects. The idea that objects of sense may possibly be exempt from the mathematical principles rests upon the false notion that those objects are things in themselves. Hence we cannot possibly deny that every object in space and time must be infinitely divisible. Such a theory, e.g., as that there are sensible things which are indivisible atoms, contradicts the very conditions of our experience ; for nothing can be extended which is not infinitely divisible, inasmuch as whatever is extended must have the general character of space. If phenomena were things in themselves, no doubt the judgments of mathematics could not be shown to apply to them; but, on such a supposition, we could say absolutely nothing about objects, inasmuch as we should be entirely dependent for our knowledge of them on their action upon us. It is, then, because the objects which we are capable of knowing are phenomena, that we can justify the a priori PRINCIPLES OF PURE UNDERSTANDING 188 synthetic judgments of mathematics, and at the same time show that they must be applicable to all possible objects of experience. 2. Anticipatioyis of Observation. 96a The principle of these is: In all phenomena the real, which is an object of sensation, has intensive magnitude or degree. 96 6 In the first principle of pure understanding, entitled " Axioms of Perception," Kant has pointed out the conditions under which we have the consciousness of any object, whether it is a determination of space or time, or a deter- mination of a concrete sensible thing. Whatever we perceive must be perceived as an extensive magnitude. He now goes on to show, that, whenever we perceive a sensible object we must determine it as having intensive magnitude or degree : in other words, that we have the knowledge of such an object only in and through the application of the schema of degree. The question here is, whether there is something which we can determine a priori in regard to every possible sensation. It is of course impossible for us to determine a priori what the particular sensation should be. We can never anticipate whether an object will be red or some other colour, hard or soft, etc. Such a determination of a given object we learn only through sensible experience. Nevertheless, Kant contends that, however the object may vary in its sensible content, — whether it is red or blue, hard or soft — we can lay down the proposition a priori that it must have intensive magnitude or degree. We may therefore call the principle here involved an " anticipation " of sense-perception or observation, — not because we can anticipate the sensible content, but because we can antici- pate the element in the object which determines that content as an intensive magnitude. 184 TRANSCENDENTAL ANALYTIC 96 c Now, we have seen, in considering the first principle of understandins, that sensations must be combined in a successive synthesis, in order that we may be conscious of an object. But, besides this successive synthesis of appre- hension, there is implied in the consciousness of a sensible thing the determination of it as not only an extensive magnitude, but as having intensive magnitude or degree. The consciousness of degree is connected by Kant, not with the combination of various sensations, but with the deter- mination of a single sensation, that is to say, with the sensation as occupying a single moment of time. We do not obtain the consciousness of the specific character of the sensible properties of a thing by putting together different sensations in succession ; but a single sensation, as occupying a given moment, is given in itself as a whole or unity, and the determination of such sensation, or its division into ideal parts, does not yield the conscious- ness of extensive magnitude, but of intensive magnitude. We never have the consciousness of any sensible object, unless when there is given to us an element of sensation; and hence, in the absence of sensation, there can be no consciousness of the sensible object at alL We may say, then, that what corresponds to the absence of sensation is the emptiness of a given moment of time. On the other hand, when we are conscious of a given moment of time as occupied by sensation, that moment is never occupied absolutely, that is to say, it is only occupied by sensation somewhere between zero and infinity. So far as the pure conception of reality goes, it must be conceived as that which is absolutely affirmative, or excludes all negation, while the category of negation is that which excludes all affirmation. But we have no actual experience of either reality or negation in this sense. What we experience is the consciousness of an object which has a degree some- where between zero and infinity, that is, between positive affirmation and positive negation; in other words, that which PRINCIPLES OF PURE UNDERSTANDING 185 corresponds to the category of limitation. When, therefore, we have the consciousness of a sensible object, i.e., not of a thing in itself, but of a phenomenon, we are conscious of it as having a certain intensive magnitude or degree. It must not be supposed that sensation in itself reveals to us the quality of the object. The knowledge of the quality of an object implies the process by which sensation is deter- mined through the schema of degree, i.e., is determined as having an intensive magnitude somewhere between zero and infinity. The quality of an object, then, is only known through the synthesis by which we represent to ourselves the quality of the sensible thing as corresponding to a sensation. 91 a The peculiarity of intensive magnitude is that we start from unity, and determine it by breaking the unity up into what might be called internal parts. Inasmuch as we can only refer the sensation to the quality of the object in this process, the schematising of the object as degree is essential to the knowledge of the object as having a certain quality ; and, inasmuch as we can have no knowledge of any sensible object in any other way than by the application of the schema of degree, we can say universally that whatever is real in a phenomenon has intensive magnitude or degree. In the sphere of dynamics, degree takes the form of momentum ; but this is a special application of the idea of degree, which presupposes the category of causality, and therefore cannot properly be dealt with until after the determination of the conditions under which we have the knowledge of the causal connection of things. 976 We see, then, that the consciousness of sensible objects necessarily implies that these are determined, so far as their sensible properties are concerned, as having a certain specific degree. Thus, e.g., we cannot be conscious of a red object without determining it as having a specific degree of red- ness, which is somewhere between zero and infinity ; in other words, the intensive magnitude or degree of the 186 TRANSCENDENTAL ANALYTIC quality of a thing is never the smallest possible, but from the nature of the case is infinitely divisible ; and the same thing applies to all the ways in which we are conscious of degree. 97 c Kant now makes a general remark upon the first two principles of pure understanding. Magnitudes may be looked at from the point of view of their continuity. Space and time, e.g., regarded as objects of perception, are quanta continua, i.e., they are not discrete parts which are separable from one another. As Kant pointed out in the Aesthetic, what we call the parts of space, or the moments of time, are merely limits within space and time ; for these are individual wholes, and therefore the so-called " parts " are not separate things, but limits or determinations within the one space or time. When we speak of positions in space, we do not mean that these are separate things, which can be perceived prior to space or time, and out of which space or time can be constituted. Every magnitude is of such a character, that it is continuous, or is the product of the synthesis of imagination. From this point of view magni- tudes may be called fluent, to indicate that they imply the generation continuously of the so-called " parts " which are combined into a whole. 98 a This determination of magnitudes as continuous applies to all phenomena, whether we look at them as exten- sive or as intensive magnitudes, for to either of these two forms of magnitude continuity is applicable. On the other hand, when a magnitude is determined as discrete, as in the case of numbers, the continuous generation or pro- duction of the magnitude is conceived of as interrupted and then begun again, so that we have the consciousness of the distinguishable units, which are combined in a whole that does not abolish the distinction. 98 6 All phenomena, then, are not only determined as ex- tensive and intensive magnitudes, but also as continuous magnitudes. That being so, it would he easy to show that PRINCIPLES OF PURE UNDERSTANDING 187 all changes in the properties of things must be continuous. We cannot, however, here enter into this proof, because we have a knowledge of changes only in and through special experience. All that we can do, therefore, in determining the principles of pure understanding, is to point out that, whatever form sensible changes may take, these must be continuous. We have to remark that in all cases knowledge contains a specific sensible element as well as an a priori element, and that, while we can lay down the universal conditions of all possible experience, we cannot anticipate the particular or concrete character of the objects of possible experience. It might be thought that we are entitled to accept from pure physics the principles which it establishes, but we have to remember that transcendental philosophy can accept nothing which it has not itself deduced. Hence the principles of pure physics must be derived from the principles of pure understanding, not vice versa. 99 a It is easy, however, to show that the principle of the Anticipations of Observation, while it does not enable us to construct a special science of dynamics, is of great value in pointing out the universal conditions under which the experience of sensible objects is possible. In doing so it enables us to guard against the false inferences that, but for this principle, might be drawn from the absence of sensible perception in certain cases. 99 b No sensible object can be perceived that is not capable of affecting our senses. We cannot, however, infer that when our senses are not affected, there is nothing in the object corresponding to a possible sensation. Our senses are not in all cases fine enough to be directly affected by objects, and therefore we cannot say that, in the absence of such affection, there is no object present. What we are entitled to say is, that no object of knowledge or experience can possibly exist tbit is not capable of affecting sensation, if the senses were fsufficiently refined. It is obvious that we cannot possibly f rove from experience that any part of space 188 TRAXSCEXDENTAL ANALYTIC or time is empty of all degree. Kant, therefore, contends, in opposition to the mathematical phrsicists, that a part of space may be filled with any degree of reality between zero and infinity. The filling or occupying of space, in other words, does noL in his view necessarily mean that each part of space is filled with the same degree of reality : for it is just the character of degree that it occupies a given moment of time more or less, and hence any given part of space may be occupied with an intensive magnitude varying between zero and infinity. "We can perfectly well hold space to be filled with various d^rees of reality ; in short, intensive magnitude does not necessarily correspond to extensive magnitude 100 a The qualitative content of sensation is perpetually vary- ing, and, since it depends upon the receptivity of the subject^ it cannot be known a priori. Xevertheless we can lay down the a priori proposition, that, whatever may be the variation in the quality of sensation, there must be a certain deter- minate degree of that quality. It is not true, however, that intensive magnitude necessaiily corresponds to extensive magnitude. The ordinary view of the seientifie man is that, where there is a greater d^ree of sensation, there is also a greater occupancy of space. Thus we may have apparently the same amoimt of space occupied by different degrees of light, the explanation advanced being that the particles of matter are in the one ease spread over a greater space than in the other, and that, when we make allowance for this fact, we find that the extensive magnitude is precisely cor- respondent to the intensive. This doetriue Kant refuse to admit, maintaining that there is no necessary connection between intensive and extensive magnitude, and that there is nothing to show that the very same intensive magnitude may not occupy a greater amount of space, or, conversely, _that a greater amount of intensive magni.ude may not be present in a smaller extent of space. The main point, however, is, that we cannot be conscious of the d^iee of a PRINCIPLES OF PURE UNDERSTANDING 189 quality except in so far as we have the consciousness of the generation by a synthetic process of the elements of sensa- tion, viewed as occupying a single moment of time. 3. Analogies of £j:pu-iincc. la The principle of these is : Experience is possible only through the consciousness of a necessary connection of perceptions of sense. 16 Knowledge of objects is possible only through the deter- mination of the elements of sense relatively to time, in other words, only through the application of the schemata. Now, the schemata of relation are respectively, permanence in time, fixed order in time, and reciprocal existence in time. It is these schemata to which Kant refers when he speaks of the modi of time as permanence, succession and co-existence. He ought rather to have said, that there are three modi in which objects of perception are capable of being related to time. These three rules express, according to Kant, aU the ways iu which phenomena are connected together in the system of experience, and the whole question here to be considered is how they affect this connection. ; In the deduction of the categories it was proved that the various forms of synthesis, which express the spontaneous activity of thought, must necessarily be such that they enable us to have the consciousness of the identity of self. We are, therefore, entitled to affirm of the objective world all that is necessarily involved in the unity of self-conscious- ness. Without the necessary unity of self-consciousness there can be no connected system of experience, and hence we are entitled to lay down, as conditions of such a system, all that is implied in the unity of apperception. We know that the original apperception is related to the inner sense, inasmuch as time is the universal perceptive 190 TRANSCENDENTAL ANALYTIC condition of all phenomena whatever; and therefore that self-consciousness is possible for us only in so far as the elements of perception are brought into relation to time. It is, then, through the synthesis of the manifold relatively to time that the consciousness of the identity of self arises, and hence we are entitled to say that nothing can enter into our knowledge that does not presuppose the synthetic activity by which the elements of sense-perception are brought under the transcendental unity of apperception. The analogies of experience in fact are just an expression of the three ways in which the subject comes to the conscious- ness of his own identity in and through his consciousness of objects as belonging to a single system of experience. 102 a There is a distinction between these analogies of ex- perience and the two principles already considered, which Kant calls the mathematical principles. The analogies of experience, he says, have to do with the existence of the objects of sense and their relations to one another, while the mathematical principles simply determine the objects as perceptions. The mathematical principles, in other words, express the modes of synthetic unity by which individual perceptions are constituted. The first principle lays down the formal conditions of every possible perception, pointing out that we can have no consciousness of an object at all unless in so far as we determine it by determining the pure form of perception as an extensive magnitude. The second principle points out what are the real components of every possible perception, so far as the sensible qualities of the object are concerned. These principles, therefore, evidently determine the very character of an object, that is, they point out what are the conditions without which we cannot have single objects. But the analogies of experience do not tell us what are the elements which enter into single objects or constitute them as single objects. We can predicate a priori of every sensible object that it will have a certain extensive and a certain intensive magnitude, for PRINCIPLES OF PURE UNDERSTANDING 191 these determinations belong to the very constitution of the object, and without them there would be no object at all. But, when we predicate existence of a sensible object, we are not in the same way predicating a mark or attribute of it. Existence, in fact, as predicated of objects, always refers to the conditions of their existence in the connected system of experience, and therefore the Analogies have to do entirely with the relation or connection of objects. This distinction between the mathematical principles and the analogies Kant expresses by saying that the former are constitutive, the latter regulative. The two first principles are constitutive in this sense, that, when we have three terms given, we can construct the fourth, as when we construct the intensive magnitude of the sun from the intensive magnitude of the moon, multiplying by 200,000. But the analogies of experience do not in this way enable us to supply the fourth term as a definite or concrete object. Thus, the principle of cause and effect is a perfectly general principle, stating the universal condition under which an objective sequence or order of events occurs. This principle, in so far as it involves the schematism of the understanding, means that wherever there is a fixed or objective succession there we must have the connection implied in . the principle of causality. When we have a given phenomenon, say, the fact that a stone grows warm, we cannot immediately pass from it to determine a priori what the particular cause of the heat in the stone is. It is for this reason that Kant speaks of the three principles now under consideration as " analogies." They are not axioms or anticipations, because they do not enable us to determine anything in regard to the concrete character of objects, but only to lay down the condition of their relation or connection. The analogies of experience, then, unlike the mathe- matical principles, do not enable us to determine anything in regard to the special character of objects. The same 192 TRANSCENDENTAL ANALYTIC thing, it is to be observed, applies to the postulates of empirical thought, which constitute the second class of dynamical categories. Like the analogies of experience, these postulates are concerned entirely with the relation or connection of objects. They differ from the analogies of experience in this way, that they do not determine the connection of objects with one another, but only their connection with, or relation to, the knowing subject. Like the analogies of experience they are entirely general in their character ; in other words, they determine nothing in regard to the concrete content of the individual object. It is obvious that, in the transition from the mathematical principles to the analogies of experience, and again from these to the postulates of empirical thought, Kant has been ad\'ancing regularly in his determination of experience as a whole. He begins, in the first principle, with the purely formal determination of the object, pointing out the per- ceptual conditions under which it is determined as possibla In the second principle he goes on to point out that not only must objects be combined so as to present themselves as extensive magnitudes, but that they must further be determined as having intensive magnitude or degree. Having thus stated the conditions of possible indi\'idual objects, he next goes on, in the analogies of experience, to tell us what are the conditions under which objects as such are connected in the system of a single experience. With the analogies of experience Kant holds that the whole character of our experience or knowledge has been determined, so far as objects are concerned ; but in the postulates of empirical thought he proceeds to consider how these objects of experience, now regarded as completely determined, are related to the knowing subject. The main distinction drawn between the mathematical and dynamical principles is, that the former are coiistitiutive, the latter regulative. This use of the term " regulative " must be carefully distinguished from a subsequent use of it, which is PRINCIPLES OF PURE UNDERSTANDING 193 introduced in connection with the application of Ideas to objects. In this latter connection the term " regulative " means that there are Ideas of the unconditioned, which do not give us a knowledge of real objects, but only enable us to bring our knowledge into a subjective unity. The term "regulative," however, in the present connection, refers to the fact that the dynamical categories have to do with the general rules or rcgulac under which objects are brought, in so far as they are connected together in a single system of experience, or are related in universal ways to the knowing subject. 13 6 Kant finds a special difficulty in proving the analogies of experience, because it seems as if there might be an experience of objects independently of the application to them of any category, and especially of any of the categories of relation. But, if we actually have experience without categories, how can it be shown that categories are essential to experience ? The difficulty presses upon us with special force, because philosophy cannot employ the method of proof which is legitimate in mathematics. The judgments of mathematics are based upon direct perception, for mathematics demo/istratcs the truth of its principles by constructing its objects a primn, — a method it is entitled to follow because it deals with the pure perceptions of space and time. Philosophy, on the other hand, cannot argue that the pure conceptions of the understanding may be presented in experience, because no number of particular experiences can establish a universal principle. sha Kant's solution of this problem is, that, while the principles of the understanding cannot be directly based upon particular experiences, there is a sense in which particular experiences are based upon them. No doubt the objective character of a principle cannot be proved by showing that we actually use it in our ordinary experience of objects, but it may be proved by showing that without it we should have no experience of objects at all. Kant has somewhat weakened the force of his doctrine by admitting N 194 TRANSCENDENTAL ANALYTIC that we can have particular experiences independently of the process by which these are judged to be instances of a . universal principle, all that he seems to deny being that' from such experiences, however numerous, no universal and necessary principle can be derived. But, as the present discussion shows, he does not really mean that we can have particular experiences without the application of universal ^ principles, but only that we can have them without the conscious use of such principles. When in our ordinary experience we observe a change to take place, we do not think of the change as involving the principle of causality, but we tacitly assume that principle, and unless we do so, we cannot have the experience of change. The method of proof in philosophy therefore is, not to base certain principles upon an induction from given particulars, but to show that without the express or implied application of those principles in particular cases there would be no experience of objects at all. We start neither from data nor from conceptions assumed to be true, but argue that without certain conceptions there would be no data, and without certain data no fruitful conceptions. 106 a This method of proving a principle by showing that it is the condition without which there could be no experience is not dogmatic, i.e., it does not start from unproved data or from preconceptions, but it is in the strictest sense systematic. It must be so, because, as we have seen, the supreme principle of all possible experience is the unity of self-consciousness, a unity which is possible only because the categories are themselves a system standing under the unity of self-consciousness ; whence it follows that, although experience is only of phenomena, it is necessarily of pheno- mena connected with one another by the universal and necessary principles of the understanding. PRINCIPLES OF PURE UNDERSTANDING 195 A. First Analogy. Principle of the Permanence of Substance. ;106 6 In all the changes of phenomena substance is permanent, and its quantum in nature neither increases nor diminishes. In the discussion of this principle Kant seeks to show that substances, or real objects, in so far as they come within the circle of our experience, are necessarily permanent or unchanging in quantity. He is, therefore, virtually seeking to establish one of the principles which lies at the basis of aU physical science, viz., the indestructibility of matter. It is true that, as he here states it, the principle is not limited to spacial objects ; but in point of fact, as Kant afterwards points out (Extracts, p. 127), the only permanent that we are capable of knowing is the permanent in space, in other words, matter. 106 c In our apprehension of the various elements of perception, as it is argued, there always is a succession of ideas, and therefore a continual transition from one state to another. If this is the form of all our consciousness, it is obvious that we cannot distinguish between things that are co-existent and a real succession of events by a simple reference to the fact that our consciousness always appears under the form of time, for obviously this applies at once to co-existent objects and to a real succession of events. We have to ask, therefore, what it is that entitles us to distinguish between an objective sequence or change and a mere sequence of impressions, and also what entitles us to distinguish between the permanent, the changeable and the co-existent. The answer is, that there must be in our conscious experience something that enables us to dis- tinguish between the changeable and the co-existent, and this again implies that there must be something which is permanent. Kant, therefore, treats the idea of the per- manent as the fundamental presupposition in our experience 196 TRANSCENDENTAL ANALYTIC of the knowledge of what is simultaneous and successive. It is this permanent which constitutes what he calls the substratum of the empirical consciousness of time itself; that is to say, that which endures, notwithstanding the changes which take place in a given object. We cannot possibly explain the consciousness of the permanent by reference to time itself, any more than we can explain the consciousness of co-existence as a mode of time ; for that which is permanent is essentially relative to the changing, and in the mere succession of the moments of time there ie no consciousness of change. It is only in so far as we apprehend the sensible element or matter, and relate it to time, that we come to have the consciousness of change in time. There is, in short, no such thing as a change of time, but only a change of phenomena in time. If any one maintains that, since time is a pure succession, and therefore comes into being part by part, we can be conscious of pure time as a succession, we must answer that there is no such consciousness except in relation and contrast to that which is not successive ; and hence we should have to suppose that this time, which originated part by part, existed in a time which did not originate, but was permanent. This shows clearly that the consciousness of the changeable is essentially relative to the consciousness of the permanent. In a mere succession of states of consciousness, taken as a mere succession, there can be no consciousness of change, because in such a succession there is a perpetual coming to be and ceasing to be. Since time cannot be perceived by itself, it is obvious that we can have no conscious- ness of objects, and therefore no consciousness of the unity of self, unless we have the consciousness of a permanent substrate of all the determinations of phenomena in time. The synthetic unity, then, which, as we have seen, is implied in all consciousness of objects, necessarily implies that objects are themselves permanent in change. It follows that all objects known to us in our experience present PRINCIPLES OF PURE UNDERSTANDING 197 themselves as the permanent which undergoes change, and it is this permanent which constitutes what we mean by the real object. 108 a It is important to observe that the " permanent " which constitutes the reality of an object is not something which has an existence apart from its accidents. The deter- minations of substance are not " accidents " in the sense of something without which the object would still be what it is : they are just the manner in which the substance exists, or they are positive determinations of the essential character of the object, not determinations related to the object negatively. It is true that we sometimes speak as if the modes in which a thing exists were accidents that merely inhere in it and are not essential to its existence. But this mode of speech, though it is natural in certain cases, is not strictly accurate, and is apt to lead to the false notion that the substance can exist, and be what it is, independently of its accidents. Kant finds the explanation of this seeming independence of substance on its accidents in the character of our understanding ; for the understanding, as he always holds, is in itself a pure identity, and therefore, when it predicates something of a subject, it goes on the principle that what is predicated attaches to the subject in the way of pure identity. Hence thought naturally separates sub- stance, or that which is real, from the changeable, and conceives of it as that which is permanent, in the sense of that which is identical or unchangeable. On the other hand, thought, in so far as it is employed in the synthetic activity by which it determines objects, does not proceed on the basis of this principle of abstract identity, but, on the contrary, consists essentially in the combination of differ- ences into a unity. The most that we can say is, that understanding, in operating with the category of substance, is not, strictly speaking, operating with a category of relation, but rather with that which is the condition of relations. The conception of substance, in other words, 198 TRANSCENDENTAL ANALYTIC is the conception of that which, as permanent, is the necessary presupposition of all objective changes and all real co-existences. 108 6 Since, then, the conception of the permanent is the condition of our knowledge of the changeable and the co-existent, it is obvious that we can have no experience of change except in so far as the changing elements of our sensible experience are referred to that which is permanent. Change must not be confused with mere alternation, i.e., with the coming to be and the ceasing to be of separate determinations. To identify alternation with change is to identify a mere evanescent series with the consciousness of such a series, — an identification which is illegitimate, because we can only be conscious of such a series in so far as the mere succession of events is contrasted with the per- manent. We can, therefore, say that " nothing changes except the permanent," since the determinations of the permanent when separated froni it become a mere discrete series. 109 a It follows from what has been said that there can be no experience, or observation, of the absolute beginning or cessation of a real object. Such an experience would mean that the object was absolutely separated from all the other objects of which we are conscious. If we suppose that something absolutely begins to be, we must suppose that there was a point of time in which that something was not. But in an absolutely empty time, which we thus conceive to be prior to a given point of time, there is nothing to enable us to discriminate what begins at a given moment, because there is nothing with which to contrast it. There can be no consciousness of empty time, because all our experience is of that which is relative to time, while time itself is not a possible object of experience. Hence, whenever we experi- ence anything, it must be the experience of some change in that which already exists or is permanent. Similarly, we can have no experience of an absolute cessation of existence, PRINCIPLES OF PURE UNDERSTANDING 199 ■ which would mean that we had the experience of absolutely empty time. 095 The actual objects of our experience are necessarily conceived of as substances, i.e., as things which, in all their changes, yet maintain their identity through time. If we suppose that substances could come into being or cease to be, we destroy the condition under which alone there is any unity in our experience. For, the unity of our experience necessarily implies that, whatever be the changes present to our consciousness, we are capable of connecting them all within the unity of one time, under the supreme condition of the unity of self-consciousness. A substance which came into being for the first time, however, could not be connected with the objects of our experience, and there- fore could not be brought under the unity of a single time, and so could not be brought under the supreme unity of self-consciousness. We should in fact be forced, on this supposition, to think of two separate and distinct series of time, inasmuch as we can have no consciousness of objects at all without relating them to time. But such a disruption in the continuity of our experience is absurd, for there can be only one time. Therefore, whatever we know must be related to this one single time. Wa There is no experience, then, except of objects which are determined as permanent in the process of change. Now, such permanence, in the case of external things, i.e., things in space, necessarily means the permanence of the quantity of such things: and hence this principle of the permanence of substance is one of the fundamental propositions lying at the basis of pure physics, — the proposition, viz., that the substance or quantity of matter neither increases nor diminishes. 200 TRANSCENDENTAL ANALYTIC B. — Second Analogy. Principle of Causal Succession. 110 6 All changes take place in conformity with the law of the connection of cause and effect. 110 c To the principle of causality Kant gives special attention, partly no doubt because it was in connection with this principle that he was first aroused by Hume from his dogmatic slumber. He tells us elsewhere that he gener- alised the problem of Hume, in order to find out if there were not also other judgments, besides that of causality, which were open to a similar objection. Hume partly gave plausibility to his reduction of causality to " custom " by assuming that in immediate perception we have the knowledge of objects that are permanent, or persist even when they are not perceived. In this connection Kant points out here, that, in the principle of Substance, it has already been shown that there is no such thing as absolute change, but only an alteration in the state of a substance that already exists. Hence, when he goes on to discuss the principle of causality, he presupposes that he has already established the necessity of the consciousness of the per- manent as a condition of all knowledge of objects. Further, not only is the principle of substantiality presupposed in the principle of causality, but we must also explain how we come to have the experience of co-existing objects. There are, then, three distinct ways in which the elements of sense-perception are brought into the unity of experience; in other words we have to account for (I) the permanence of things, (2) the changes of things, and (3) the co- existence of things. Hume attempted to account for the changes of things by saying that the repetition of perceptions or ideas in our consciousness comes to be identified with the proposition that events are necessarily connected. The plausibility of this explanation arises from the confusion PRINCIPLES OF PURE UNDERSTANDING 201 between a succession of subjective states and the succession of events. Hume in fact identifies the one with the other. Now, Kant in his discussion of the principle of causality begins by pointing out that, when the problem is univer- salised, the explanation of Hume is seen to be utterly inadequate. We may plausibly account for the objective sequence of events by a reference to the subjective sequence of ideas, so long as we confine our attention purely to objective sequences, but how can we possibly derive, from the very same sequence of ideas, three entirely different determinations of objects, viz., permanence, succession, and co-existence ? The form of our consciousness, he argues, is always a successive apprehension, and therefore, if our consciousness of objects could be derived from it, we ought to have no knowledge except of objects or events as succes- sive. It is plain, then, that some other explanation must be given of causality than that which is advanced by Hume. Hume is right enough in saying that, if there is a principle of causality, there must be a necessary connection of events ; and he is also right in saying that the knowledge of such events must be shown to be somehow involved in the succession of our ideas. How, then, can we derive the con- sciousness of objective succession, without going beyond the circle of consciousness ? It is not enough to direct our attention simply to the sequence of ideas, as Hume does ; for, though we may call any idea an object of consciousness, that does not explain what we mean when we contrast a mere series of fancies which arise in our imagination with a real or objective succession of events. There is no way of distinguishing between an objective sequence and an arbitrary sequence, so long as we confine our attention simply to the succession of states in consciousness. If the objects of our experience were things in themselves, it is plain that we could never predicate objective succession at all ; for, on the supposition that phenomena are things in themselves, we could only have a knowledge of them, if at 202 TRANSCENDENTAL ANALYTIC all, through the affections which arose in our consciousness. But, assuming that we are limited to the immediate states of our consciousness, obviously even the consciousness that there is in our minds a succession of ideas does not prove that there is any succession in the real object. What things in themselves may be we cannot possibly tell, for we ■cannot go beyond our consciousness in order to apprehend things as they , are in themselves. How, then, keeping within consciousness, are we to account for the experience of an objective succession ? How is it that, though the form of our consciousness is always successive, we yet do not in all cases predicate objective succession ? When we perceive a house, there is no doubt that, so far as our apprehension goes, the visible parts of the house present themselves to us in the form of a succession of our ideas ; but no one would think of saying that the parts of the house are successive. How is it, then, that in this case we •do not predicate objective succession, notwithstanding the fact that our ideas are successive ? We have to remember that, from the transcendental point of view, i.e., from the point of view of the possibility of actual experience, the house is not a thing in itself beyond consciousness, but is a phenomenon, which has no existence apart from conscious- ness. Without denying that there is an object corresponding to the house, Kant affirms that what we have to account for is the house as a phenomenon, i.e., what we have to account for is our experience of the house. What then is meant by the connection of various determinations in the phenomenon? We distinguish the phenomenon from the succession of our ideas, and yet the phenomenon is nothing but a com- plex or combination of our ideas. What is the ground of this distinction ? As truth consists in the harmony of knowledge with its object, and as the object here, so far as the sensible element is concerned, is given to us, the only question must be, What is there in the character of our knowledge that converts the given sensible element into the PRINCIPLES OF PURE UNDERSTANDING 203 consciousness of an object ? The explanation must be drawn from the manner in which the understanding combines the given " matter " of sense ; in other words, we have to seek for the explanation of objectivity in the synthetic activity of the understanding, as supplying some rule under which our experience of the various elements of perception must be brought. What we mean, then, by objectivity must simply be the consciousness of this necessary rule of apprehension. As we have already seen, there is no consciousness of any object that comes into existence for the first time ; in other words, we must, in dealing with the question of causality, recognize that it presupposes the permanence of real objects. Every apprehension, then, of an event is the apprehension of something that follows upon a previous apprehension. This, however, does not enable us to dis- tinguish between permanence, objective succession and co-existence. But, when we observe that there is in certain cases a peculiar characteristic in the sequence of our ideas which is not found in other cases, we get the clue to the distinction between an objective sequence of events and an arbitrary sequence of ideas. We never predicate objective sequence except when there is fixed order in our perceptions, an order such that A must precede and B follow. Thus, when we perceive a ship drifting down stream, the order of our perceptions is absolutely fixed, so that we cannot observe the ship first lower down and then higher up the stream, but must observe it in a certain order, making the transition regularly from one point to another. Here then our apprehension is fixed by the character of the events apprehended, while in the case of the house the appre- hension is not fixed but arbitrary. Kant does not mean by this that the house does not involve a rule of the under- standing, for he has already shown {Extracts, p. 80) that the perception of a house implies the category of quantity. What he contends is, that the peculiar form of objectivity 204 TRANSCENDENTAL ANALYTIC which presents itself in our experience as a real succession of events, or an objective sequence, can only be derived from a peculiar mode of relating the elements of perception,— a mode, therefore, which must have its seat in the nature of the understanding, and which must consist in the schema by which the elements of perception are brought under the unity of self- consciousness. 113 a Here, then, the peculiar succession of our ideas can only be explained by the objective succession of the phenomena ; in other words, our ideas arise in a certain order, an order which is invariable, because the object as known implies such invariability. It is only by looking at the matter in this way that we can explain how we should distinguish one kind of phenomenon from another, i.e., in some cases predicate substance, in others causality, and in others reciprocity. What is meant by objective sequence is simply the conformity of our perception to a fixed rule, the rule in this case being that of invariable order of succession. Whenever we have an invariable order in our perception, there we are entitled to predicate objective sequence. 114 a This rule, then, implies that events cannot come in any but one order. The order may be said to correspond to the necessary succession of moments of time themselves, but Kant ought strictly speaking to say, that what determines time to a necessary order of moments is the necessity we are under of combining certain determinations in such an order, i.e., ia an order which is invariable. It is to be observed that, though in all cases of objective sequence the principle of causality is implied, Kant does not mean that in such a sequence itself the cause is explicitly apprehended : what he means is that, whenever there is an objective succession, we necessarily presuppose that the given change has some cause without which it could not be. 115 a The deduction of the principle of causality just given enables us to see the inadequacy of the doctrine of Hume. PRINCIPLES OF PURE UNDERSTANDING 205 On his view the conception of causality is merely a generalisation from a number of observations, i.e., he holds that we have the repeated experience of certain events as following certain antecedents in time, and, comparing these experiences with one another, we frame the general idea of cause. On this explanation we should never attain to pure universality and necessity, but at the most to generality. The real truth, however, is, that the principle of causality is an a priori principle, which cannot be derived from experience, but is absolutely essential to the constitution of experience. Admitting that thought in itself is analytic, it must be observed that the analytic judgments of thought presuppose synthetic judgments. We do not obtain the principle of causality by any mere analysis of given concep- tions, obtained in the first instance by an induction from experience ; but we have the conception or principle of causality to analyse at all, only because in our actual experience the synthetic activity of thought has been exer- cised in the form of the schema which corresponds to the category of causality in the actual determination of changes in the known world. The clear analysis of the conception of cause is, therefore, not the foundation of the principle of causality, but merely the direct or explicit grasp by thought of what that principle is. We first employ the principle in the constitution of phenomena, or objects of experience, and then, reflecting on what is involved in the principle so applied, we state it in the express form of a principle of judgment. But such a principle could never be brought explicitly before the mind, did it not already lie a priori at the basis of our experience. The special form in which the understanding in the present ease exercises its synthetic activity in the constitu- tion of objects is by determining events in time to a fixed or constant order. It is because the understanding thus determines the elements of perception in an irreversible order that we are enabled to fix the temporal relation of 206 TRANSCENDENTAL ANALYTIC events to one another; for time itself is not an object of perception, and therefore we cannot determine the order of events simply by reference to time ; on the contrary, time itself obtains an order only in so far as we are conscious of phenomena as following one another in a fixed order. Thus the experience of an orderly succession of events is presupposed even in the consciousness of one event as preceding and another as following. And as events must conform to the universal condition of time, the series of possible perceptions must present itself in the same order as the series of moments in time. The principle of causality, then, or the principle which expresses the con- formity of all changes in time to the law of causality, is a necessary condition of the connection of phenomena in our experience. 116 a Since the order of time is determined by the order of phenomena, so far as these phenomena imply the synthetic activity of the understanding, we could not be conscious of an event at all, unless we had previously determined the changes of phenomena in a fixed way. No doubt, when we abstract from the particular facts of our experience, and direct our attention to the mere succession of the moments of time, we have before our minds the consciousness of a simple temporal series. But this abstract view of events as mere sequences only states what is implied in the possibility of experience, because in actual experience we must, besides the mere form of time, which applies to all phenomena, have also a determinate sensible element, which is brought into relation with time. It is, then, the rule by which the phenomenon is determined in a fixed way, so far as its changes are concerned, that enables us to have the know- ledge of events or successions in time, and the particular rule is, that in what precedes is found the condition under which an event always or necessarily follows. 116 6 We may now sum up the proof or deduction of the principle of causality. The experience of objects always PRINCIPLES OF PURE UNDERSTANDING 207 implies a synthesis by the imagination of various determina- tions or elements of perception. It is impossible, however, to explain the knowledge of objective or real succession simply by saying that the form of our consciousness is always that of a succession in time ; for, just because it is true that all our ideas follow one another, we are unable by a mere reference to this fact to explain how in some cases we predicate permanence, in others succession, and in still others co-existence. It is not, then, the mere fact that the synthesis of apprehension involves a succession of ideas that accounts for the determination of real events as successive, but it is because certain of our ideas follow in a fixed order, which is determined for us. If therefore we are to have the knowledge of an actual succession of events, there must be in our experience this fixity of order, and without it we should never be conscious of objective succession at all. For, even if we suppose it possible that we should have a mere series of ideas, and that we should further be conscious of these as a series, even then the most that we could affirm would be that there was in our minds a certain series of ideas. There is nothing in a series of ideas, taken by itself, that enables us to distinguish between the mere play of imagination and the knowledge of objective fact, or between a dream and waking reality. There must therefore be added to the mere consciousness of a series of ideas the • actual experience of these as coming in a fixed order in time, — such an order as that, given A, B invariably follows. But this is just the principle of causality in its application to objects of sensible experience. For, though the pure conception of causality is merely the thought of the dependence of one thing upon another, it is only in so far as this pure conception is schematised as a fixed order in time, i.e., only in so far as the elements of perception are determined to a fixed order, that we have actual knowledge of the relation of cause and effect. The principle of causality, then, necessarily applies to all possible objects of 208 TRANSCENDENTAL ANALYTIC experience, because, apart from the experience of events as following in a fixed or invariable order, we could have no experience whatever of real events. / C. Third Analogy. Principle of Community. 118 a All Substances, in so far as they can be ^observed to co-exist in space, are in thorough- going reciprocity. The third analogy differs from the other jtwo in adding the limitation of space, and it is worthy of^'note that this limitation first appears in the second edition^ of the Critique. The limitation is in harmony with a general Eemark added to the principles of judgments {Extracts, p.p. 126-128), where Kant points out that all the principles of judgment involve a relation, not only to perception, but to external perception. In treating of the schemata Kant represented the synthesis of the understanding as determiried always by reference to time. But, in the interval between the publication of the first and second editions of the Critique, it seems to have more and more impressed itself on his mind, that the categories must be schematised, not merely by reference to time, but also by reference to space, so that, strictly speak- ing, we have no experience of substances or co-existences except in relation to objects in space. It is only, however, in the third analogy, that he expressly introduces this limitation. 118 6 Kant begins by pointing out the empirical criterion of co-existence, as distinguished from that of permanence and objective succession. When we say that things co-exist, we mean that they exist at the same time and in the same space. But, as time is the general form of all our con- sciousness, there must be something in the character of the succession of our ideas which accounts for the distinction we draw between the changes occurring in a single sub- stance — in other words, objective succession — and the PRINCIPLES OF PURE UNDERSTANDING 209 changes which we observe in different substances relatively to one another. As a matter of fact we predicate the co-existence of substances when the order of our ideas is reversible, whereas we predicate objective succession when the order is fixed. If the determinations actually fol- lowed in time, the order of our ideas would be fixed, begin- ning with A and going on regularly through B, C, D, E, to F. Hence it is at least plain that the consciousness of the co-existence of objects caimot be derived from the mere fact that our ideas occur in succession. 118 c We cannot, then, explain the co-existence of objects from the more subjective succession of our ideas. We have shown, however, that the experience of objects presupposes permanent things or substances, and also that these sub- stances undergo certain changes which occur in a fixed order. But, even supposing that we could observe separate substances with their changes : supposing, in other words, that experience were determined by the two principles of substance and causality, this would not explain how we come to have the experience of co-existent objects. It is plain that we cannot derive co-existence from the experience of substances in their separation. For suppose we could observe a single substance, and then pass on to the observa- tion of another, it is obvious that since here our experience would take the form of succession we could not say that the two substances successively observed co-existed. The J perception, then, of separate substances does not of itself! involve that those substances are in one space ; in other words, that they co-exist. 19 a There must, therefore, be something more implied in our experience of co-existent objects than the fact that they exist in the same space. The only satisfactory explanation of the experience of co-existence is that which shows that, of two substances, neither can be what it is except in relation to the other ; in other words, there must be some casual connection between the two substances. A 210 TRANSCENDENTAL ANALYTIC cause must either act upon substances themselves or upon their changing states. Inasmuch, however, as substance does not itself begin to be, but on the contrary is perma- nent, it is obvious that causality can only have to do with the states of substances, not with substances themselves. The knowledge, then, of the reciprocal changes in different substances presupposes that these several changes are not independent, but on the contrary imply one another : that is to say, each substance must be a cause in relation to the changes of another substance, and must itself be an effect, not indeed as a substance but in its own states. We have only experience in fact of objects as co-existent, in so far as we determine them as reciprocally dependent in regard to the changes which they undergo relatively to one another. Since we must regard as necessary to experience all that must be presupposed in order to account for its unity, and since we cannot have a unity or system of perceptions except through the synthesis of the understanding, by which objects are determined as reciprocally causal, we are entitled to say that the objects of experience must stand in the thorough-going community of reciprocity. 119 6 When phenomena are said to stand in the community of reciprocity, it is not meant merely that they are co-existent, or in local community, but that they are related by the principle of reciprocal causation, or are in dynamical com- munity. If we look at our experience, it may be easily seen that we attribute co-existence to different things, only in so far as they have causal influence upon one another. By merely passing from the observation of the earth to the moon, and from the moon to the earth, we should not have the knowledge that the moon and the earth are co-existent. We experience their co-existence, because they are con- nected dynamically by the light which plays between us and the moon. Nor could we observe any change in the position of objects, or, what is the same thing, any change in the empirical observation of objects relatively to our- PRINCIPLES OF PURE UNDERSTANDING 211 selves, were it not that all objects of experience are connected together in the way of dynamical community. Without such community, then, there could be nothing but a number of detached observations; and hence this principle is the necessary condition of the systematic unity of our experience. It is obvious that the mere consciousness of the occurrence of similar ideas in succession would never yield the consciousness of the co-existence of objects. Our experience is by its very nature a causal system of phenomena. This does not entitle us to say that space is absolutely filled, but it does entitle us to say that we cannot experience a space which is empty. The principle of community, which has thus been deduced, like the other principles, presupposes the unity of apperception. This unity in the present case takes this form, that we are conscious of objects as co-existent, only in so far as the understanding has combined them through the principle of community, and in constituting them into a single system has related them to the single unity of apperception. What we had to explain was not the independent existence of objects or things in themselves, but our knowledge of objects as presenting themselves co-existently in space ; and, as we have seen, this objective unity of our experience is only possible in so far as there is an objective ground for it ; that is to say, we cannot derive the actual co-existence of phenomena from our own subjective ideas, but, on the contrary, we can only explain the unity of our self-conscious- ness by showing that the synthesis of the understanding in the constitution of the system of experience is the necessary condition of the unity of our experience. We have now established the three dynamical relations which are involved in our experience as a system of phenomena. Upon these relations are based the fundamental principles of pure physics. Thus, as Kant points out in his Budiments of Physics, the permanence of substance is the basis of the principle of the indestructibility of matter, the principle of 212 TRANSCENDENTAL ANALYTIC causality of the persistence of force, and the principle of reciprocity of the law of action and reaction. 121a We may shortly sum up the result of our whole investigation. While the two first principles determine the character of the content of individual perceptions, the Analogies of Experience determine the character of the relation of judgments to one another in time. As the possible relations to time are duration, succession and co-existence, the analogies determine phenomena in so far as they endure or persist through time, exhibit changes which follow a fixed order in time, and are related to one another as existing together in time. The determination of phenomena is, therefore, absolutely dynamical, i.e., there is no experience of objects simply as in time, but only of objects as causally related to one another in time. The unity which these principles introduce is due to the rules, under which the understanding operates, in so far as it is exercised in combining what is given to it in sensible perception into a systematic unity. 121 b We have already distinguished between the two senses of the term "nature" {Extracts, pp. 80-81). By the term " nature " is meant either the sum-total of all phenomena {natura materialiter spedata), or the system of necessary laws by which these phenomena are connected (natura formaliter spectata). Now, the analogies of experience are obviously rules or laws which determine nature in the second of these senses. Without these principles, then, there could be no " nature " at all, i.e., no system of phenomena. This is an important point, because it enables us to determine the fundamental principles which must be presupposed in the special sciences. The empirical laws which these sciences discover — the laws of dynamics, physics, chemistry, etc. — must all stand under the analogies of experience, because these are the primary laws without which )io experience at all is possible. The specific laws of the sciences may thus be regarded as applications or exponents of the analogies of PRINCIPLES OF PURE UNDERSTANDING 213 experience, and these laws also must necessarily be laws of phenomena, for no knowledge at all is possible except in so far as the analogies of experience are presupposed as a necessary condition of the unity of apperception, and there- fore of the unity of experience. Taking the three analogies together, we can say, that all phenomena necessarily belong to a single system of nature, since apart from the a priori unity of apperception, which again presupposes these principles, there could be no unity of experience, and there- fore no determination of objects in experience. 4. Postulates of all Empirical Thouglit. ,122 a With the analogies of experience Kant has finally determined all the universal conditions of our experience of objects. He now goes on to consider the conditions under which the objective world is apprehended by the knowing subject. These conditions are expressed in the Postulates of Empirical Thought, which do not determine the constitution of objects, but simply point out the conditions under which we have experience of objects We may say, generally, that the three Postulates af&rm that we can have no knowledge of an object at all except in so far as it is brought under the schema of possibility, no know- ledge of a given object unless it is brought under the schema of actuality, and no knowledge of a necessary object unless it is brought under the schema of necessity. These principles may therefore be said to sum up, and express in concise form, the conditions of our knowledge. 122 6 The categories of modality do not determine the content of objects of experience, as is done by the categories of quantity and quality, nor do they determine the relation of such objects to one another, as is done by the categories of relation, but they state the relation of objects as already constituted and connected in the system of experience to the knowing subject. We may, therefore, assume that we 214 TRANSCENDENTAL ANALYTIC know all the conditions under which objects of experience are possible, and we may yet ask what are the conditions under which we predicate about such objects possibility, actuality and necessity. The postulates of empirical thought are the answer to this problem. 123 a (1) The first postulate demands that the object which is to be known must conform to the formal conditions of aU experience. But the formal conditions of experience are the forms of perception, viz., space and time, and the forms of thought, i.e., the categories. No object, then, as we are entitled to say, can be known which does not conform to I the conditions of space and time, and to all the principles of understanding previously laid down, including the analogies of experience. It is natural to suppose that the possibility of an object may be shown from its mere conceivability, but the whole course of our investigation has proved that from a mere conception nothing can be determined in regard to the possibility of knowledge. We can think, e.g., of a figure that is enclosed by two straight lines, for the conception of two straight lines and the conception of two such lines meeting does not involve the negation of figure, i.e., we can think any elements that are not in themselves absolutely contradictory ; but we cannot from this conclude that the object so thought is a possible object of experience, for nothing can be an object of experience that does not conform to the necessary conditions of experience. 123 6 It seems at first sight certain that whatever is con- ceivable is at least possible. This, indeed, was the doctrine assumed by Descartes, Spinoza and Leibnitz, who took it for granted that whatever can be conceived may exist. Leibnitz, indeed, went further, and maintained that what is ' conceivable, provided always it does not conflict with what is otherwise conceivable, is actual. Thus, the conception of an infinite being is a possible conception, and as there is nothing to contradict it, Leibnitz maintained that such a PRINCIPLES OF PURE UNDERSTANDING 215 , being existed. The ontological argument for the existence of God, as put forward by Descartes and Leibnitz, rests, in fact, upon the principle, that what is conceivable is possible, and that where there is no other conceivability, that which is conceivable is actual. ma If we take a particular conception, such as that of a triangle, it seems as if we could affirm that an object corresponding to it is possible. And no doubt it is true that we can construct an object a priori corresponding to the conception ; but we must observe that we do not prove the possibility of a triangular object merely by showing that we can construct a triangle a priori. A triangular object is possible only if it conforms to the necessary conditions of perception and thought; in other words, the proof of the possibility of the triangular object implies that there should be sensible experience, and that the elements contained in this sensible experience should be combined by thought in the unity of an object under the conditions of perception. It is certainly true that we can characterise the possibility of things prior to particular experiences, but we can do so only because we can lay down a priori the conditions under which any experience at all is possible. By possibility, . then, we are to understand, not merely conceivability, but the possibility of the experience of real objects. 24 6 (2) The first postulate merely states the negative con- ditions of an object, the conditions without which we cannot have any experience at all. Provided only the conceived object does not contradict the formal conditions of our experience, we can affirm the object to be possible. But more than this is required in the actual experience of an object. The object must not only conform to the formal conditions of experience, but it must be an actual object ofi sensible perception. This relation to the sensible is what! distinguishes the postulate of actuality from the postulate of possibility. For, actual perception is always relative to a definite moment of time, and therefore we have no know- 216 TRANSCENDENTAL ANALYTIC ledge of anything as actual except that which is presented before us as at a given moment of time, or, more generally, that which is presented when it is experienced as at a given moment of time. It may be objected that if we limit our knowledge to what is actually perceived, the range of our experience will be very narrow, and that the physical sciences in many cases aiJEirm the actual existence of objects I which are not perceived. Thus, the doctrine of an ether as pervading all space seems to be incompatible with the principle that nothing can be affirmed to be actual except when it is related to a perception. In answer to this objection Kant explains that the postulate does not demand the immediate perception of the object affirmed to be actual : all that it demands is that that object should either f be directly perceived by us or be legitimately inferred from our direct perceptions. We can say that no object is actual which is not capable of being perceived by sense, but we must interpret this to mean, not that we can only affirm the existence of what we actually perceive, but that we can only affirm the existence of what can be proved to be capable of being perceived, provided our senses were fine enough. The postulate, then, entitles us to say, that whenever we have direct perception of objects as present in and forming part of the content of our experience, or whenever we are entitled to affirm the existence of such objects by a legitimate inference from the perceptions we actually have, there we have experience or knowledge of actual objects. 125 a (3) The third postulate, like the other two, shows the relation of objects of experience to the knowing subject. Kant points out that the necessity of an object cannot be established by pure thought. Logical necessity is the conception of the conditioned as implying a condition. But from this pure or logical conception of necessity nothing can be determined in regard to the necessity of known objects. This postulate, therefore, has to do with PRINCIPLES OF PURE UNDERSTANDING 217 what Kant calls material necessity ; that is to pay, it has i to do with necessity in relation to concrete sensible things or phenomena. Now, the existence of an objept of sense, in so far as it contains a special sensible eleBaent, cannot be determined a prion ; for that which is a ^prim'i always has to do with the forms of perception or thought under which variable sensible objects can be known. In what sense, then, can we speak of this postulate as an a priori principle of the understanding ? Only ia the sense that, when definite sensible objects are given, or u:ather when the sensible matter presupposed in such objects is given, we can lay down the principle a priori, that in order to be known they must enter into the single context of our experience ; in other words, they must be related in the way of necessary connection. And, obviously, we cannot assert that phenomenal objects or substances are known to be necessary in themselves, becau&e, as has been said, the only substances that we can know are those that imply a variable sensible matter. Necessity, then, while as a postulate it must deal with phenomena, cannot deal with I these as substances ; in other words, necessity here means \ the necessary connection of the changing states of phenomena ; that is to say, it means necessary connection in the way of causality. General Ecmarh on the Principles of Judgmervt 16 a Kant makes a general remark on the principles of judgment, which, as has been mentioned above, was intro- duced into the second edition of the Critique, along with another addition, called the Refutation of Idealism, which he brought forward in connection with the two first postulates. In the interval between the publication of the two editions of the Critique, Kant was charged with making the reality of experience doubtful or impossible, inasmuch as his theory was held to reduce objects to 218 1 TRANSCENDENTAL ANALYTIC mere states of the individual subject. It was in order to meet this objection, and perhaps also because the objection had made it more clear to himself that the principles of the understanding apply only to external phenomena, that these two additions were made. In the Refutation of Idealism (B. 274-8) he distinguishes two forms of so-called "Idealism," viz., the problematical and the dogmatic: The former is the Idealism of Descartes, who, as Kaut represents him, held that we are directly conscious onjy of ideas in our own mind, and from these infer the existence of external objects. Dogmatic IdeaUsm, again, is the doctrine advanced by Berkeley, which agrees with that of Descartes in maintaining that we are directly conscious only of our own ideas or perceptions, but differs from it in denying that there are any objects beyond consciousness. Kant maintains that his own doctrine differs fundamentalltj from both. Descartes and Berkeley alike assume that we are directly conscious only of our own states, i.e., of a succession of ideas. But, as Kant argues, the reduction of knowledge to a mere succession of ideas makes all knowledge impossible. Therefore, neither pro- blematical nor dogmatic Idealism can be defended. What he contends is that, if we can be said to know either class of objects prior to the other, our first consciousness is of external objects, not of internal states. Now, external objects are not, as Descartes and Berkeley assume, objects existing beyond knowledge, but simply phenomena as determined in space, in contradistinction to the mere succession of our ideas as only in time. We have, then, to maintain that the objects of our experience — the objects, that is, which are determinable by the principles of judgment — are not a mere succession of our ideas, but are spacial phenomena, or what we call ordinarily material things. 126 6 This limitation of experience or knowledge to external or spacial phenomena may be best seen in connection with the PRINCIPLES OF PURE UNDERSTANDING 219 analogies of experience. Take first the analogy of sub- stance. We have seen that we can know real objects only so far as we determine them as permanent, and contrast them with the changing states of our consciousness as these arise in time. We have now to observe that it is only in relation to external phenomena that we have the experience of anything permanent ; for, taken by themselves, the states of our consciousness, in so far as these are simply in time, do not present to us anything permanent, but, on the con- trary, are in perpetual flux. S econd ly, we can have a knowledge of objective change only in so far as the change takes place in an external object or material thing ; for we have a direct knowledge of change, and indirectly a know- ledge of the succession of our own ideas, only through our experience of motion or change of place. \Df objects through the categories is not a know- ledge of itself as an object to which categories may be applied. For we must remember that objects exist for the subject only as the product of its synthetic activity in the determination of the manifold of perception. The subject as the ground of the idea of time cannot determine its own existence by the idea of time. Therefore it has no manifold to which the category can be applied. > The source of the dialectical illusion of Rational Psychology is now perfectly obvious. The Idea of Reason, the Idea of a pure intelligence, is confused with the per- • feetly indeterminate conception of a thinking being, and it ' is assumed that what is true of the former is true of the latter. In order to make clear to myself the possibility of an ultimate experience, I abstract from all possible experience and concentrate attention upon the self. Then I falsely assume that I can be conscious of my own exist- ence apart from experience and its conditions. Thus I confuse the possible abstraction of my empirically condi- tioned existence, the mere possibility of self-consciousness, with the imaginary existence of my thinking self. Hence I come to believe that I have a knowledge of the substance in me as the transcendental subject, while in fact I have in thought only the unity of consciousness which is presupposed as the mere form of my experience. But, while there can be no knowledge of a pure intelli- gence, the idea of such an intelligence is not without value even for theoretical Reason, though it has no other than a 250 TRANSCENDENTAL DIALECTIC regulative value. It supplies us vcith the ideal by reference to which we may seek to determine the phenomenal self. The substantiality, simplicity, self-identity, and independence of the soul are, says Kant, to be regarded merely as a schema for this regulative principle, not as the real ground of the properties of the soul. These may rest on quite other grounds, which are not known to us, nor could we in any proper sense know the soul by means of these sup- posed predicates, even if they were admitted to apply to it, since they constitute the mere idea and cannot be presented in concreto. Nothing but advantage can come from such a psychological Idea, if we are careful to observe that it is only an Idea, i.e., that its sole value is to reduce the phenomena of our soul to system by the exercise of Eeason. The Idea further serves to indicate that the self as the object of inner sense is merely a phenomenon. There is therefore nothing absurd in the supposition that the self in its real nature is independent and self-determined, and thus the way is left open for that rational faith in freedom and immortality, which Kant afterwards seeks to base upon the moral consciousness. Chapter II. — The Antinomy op Puke Eeason. 155 a The dialectical arguments in the case of Rational Psychology take the form of paralogisms, because the un- conditioned is here the Idea of a self-complete subject. In the case of Rational Cosmology the illusion assumes a different form, which Kant calls Antinomy. Here the unconditioned is the Idea of the world as a complete totality of phenomena, i.e., of objects in one time and one space. The world of experience, as we have seen, is one fin which no single object is known by itself, but is 'apprehended only in relation to other phenomena. The ^[fitfirmination of phenomena, however, in experience^ is never complete, but constitutes an unending series. But THE ANTINOMY OF PURE REASON 251 Reason by its very nature cannot be satisfied with any- thing short of completeness, and therefore its demands are obviously incompatible with the conditions of experience. Now, Antinomy arises from this conflict between Under-7 standing . and Reason ; for, while we can obtain a know- ledge of phenomena only by determining the elements of perception through the synthesis of the Understanding, Reason is not satisfied with this relative determination, and hence arises a conflict. This conflict is not due to a mere logical oversight : it results from the inevitable character of our intelligence, as contrasted with the conditions of experience under which it operates. The peculiarity in the) case of the Idea of the world is that the problem of Reason) assumes the form of antithesis or dilemma. Now, when the mind has before it two mutually exclusive alternatives, by the law of excluded middle it cannot accept the one without rejecting the other. On the other hand, in accordance with that law, either of the alternatives may be established, if we assume the principle that the opposite of a false principle must be true. Here, however, we find that each of thel opposites can be proved with equal cogency by the principle of reductio ad absurdum. What we have to enquire into is the source^ of this peculiarity in the idea of the world 165 6 as a totality. Our object is the transcendental Ideas which relate to the synthesis of phenomena. The conception of the world as a whole presupposes the Idea of the uncondi- tioned, and the unconditioned inits- relation to phenomena implies the Idea of a completed synthesis of phenomena. Hence the Ideas of Reason in the present case may be called cosmical conceptions. Just as we have already con- sidered the claims of Rational Psychology, so we have now to consider the claims of Rational Cosmology. 252 TRANSCENDENTAL DIALECTIC Section I. — System of Cosmological Ideas. 156 a Two things are to be observed in connection with the Cosmological Ideas. In the first place, these Ideas do not enable us to extend our knowledge, inasmuch as, unlike the categories of the understanding, they are not constitutive relatively to sensible experience ; in other words, the Ideas of Eeason are simply certain conceptions of understanding when these are freed from their limitation to possible ex- perience and are regarded as absolute. Eeason, therefore, (abstracts altogether from the sensible conditions of experi- Jence, seeking to give completeness to the empirical synthesis |by carrying it up to the unconditioned. In doing so its principle is, that if the conditioned is admitted, we must presuppose a totality of conditions in order to account for the conditioned, and hence that the conditioned presupposes the unconditioned. The second thing to be observed is, that it is not every category of the Understanding, but only those which in their application to the elements of per- ception give rise to a series of conditions subordinated to one another, that are here in place. These Ideas, therefore, in every case start from the conditioned, and seek for completeness by a regress to the totality of conditions. 157 a When those categories are set aside which do not imply a series of conditions subordinated to one another, we find that four cosmological ideas remain, which exhibit the demand for the conditioned as regards quantity, quality, relation and modality. All these ideas imply a series in the synthesis of phenomena. (1) In the first place, since phenomena are always extensive magnitudes, whether they exist in space and time, or only in time, we naturally have an Idea of absolute completeness in regard to the com- position of phenomena ; for all phenomena, as in time and space, constitute for knowledge a series. It is obviously impossible to know without having a consciousness of one phenomenon as related to another phenomenon. THE ANTINOMY OF PURE REASON 253 Eeason demands totality in the synthesis of phenomena in time. Similarly, if we look at objects in space we find that one phenomenon is determined by relation to another, and that in our knowledge this process means that objects in space are determined serially. Thus the first cosmological Idea demands an unconditioned synthesis of phenomena in time and space. (2) The second Idea is concerned with matter or the object of external perception, which, as we saw in the first Principle of the Understand- ing, must always have intensive quality or degree ; in other words, no object of experience is found which is absolutely simple. As every phenomenon consists of parts, which are the condition of its existence as a whole, we cannot complete our knowledge of it unless we divide it into its ultimate parts and enumerate the whole of them. Thus we have the Idea of absolute completeness in the division of a given whole of phenomena. (3) The third Idea, which is connected with the general title of Relation in the table of the categories, deals with the totality of synthesis in the causal relations of phenomena to one another. All phe- nomena, as objects in time, are determined as effects of causes, and as these causes are again themselves effects presupposing other causes, Eeason demands an absolute totality of causes. Thus we have the Idea of absolute completeness as regards the origination of a phenomenon. (4) Phenomena in themselves are contingent, or only hypo- thetically necessary, i.e., necessary under condition of the existence of something else upon which they depend. Eeason, which always demands absolute completeness, demands in this case absolute completeness in the way of the depen- dence of phenomena. Section II.- — Antithetic of Piore Beason. I We may use the term Antithetic to express the conflict between two propositions, either of which seems to be true. 254 TRANSCENDENTAL DIALECTIC but both of which can put forward equally valid claims to acceptance. As thus understood, Antithetic is not the dogmatic assertion that there is an unsolved contradiction between two propositions : the source of this conflict lies in the appHcation of Eeason to objects beyond experience, instead of its use purely within experience. When our conclusions are based upon pure Ideas, the propositions which thus result are such that we can neither establish nor overthrow them. Taken separately, they are not only logically valid, but their conflict with each other arises from the very nature of reason, which, as we have seen, cannot be satisfied with anything short of the unconditioned. 158 a The Antinomies follow in the order of the transcendental Ideas as given above. The Antinomy of Pure Eeason. First Gonfiict of tlm Transcendental Ideas. THESIS. ANTITHESIS. The world has a beginning The world has no begin- in time, and is enclosed with- ning in time, and no limits in limits of space. in space, but is infinite as regards both time and space. Proof. Proof. Assume the opposite, viz.. Assume the opposite, viz., that the world has no be- that the world has a begin- ginning in time. Then, at ning in time. Then, there any given time an infinite must have been a time when series of states of things must the world was not, i.e., an have passed away and come absolutely empty time. But, to an end. But an infinite in an absolutely empty time series, from the nature of the there is nothing to account case, can never come to an for anything coming into end. Hence only a finite being. Hence, there cannot series can have passed away; have been an absolutely THE ANTINOMY OF PURE REASON 255 which is the same as saying that the world has a begin- ning in time. As to the second point, again assume the opposite, viz., that the world has no limits in space. Then, the world must be an infinite whole of co-existent things. But an extensive magnitude can only be presented by the successive synthesis of its parts, and in the case of an iafinite magnitude this syn- thesis must occupy an infinite time, or, in other words, an infinite series of moments must have passed away and come to an end. But this is impossible, as we have already seen. Hence the world cannot be infinitely extended in space, and the opposite proposition must be true, viz., that it is enclosed within limits of space. empty time, i.e., the world must always have existed. Thus it is proved that the world had no beginning in time. As to the second point, 159 as assume the opposite, viz., that the world is limited in space, i.e., that empty space extends beyond the world. Then, the world must be related to this empty space. But, such a relation of the world to empty space is im- possible, because it would be the relation of the world to no object. We must there- fore deny that there is any empty space beyond the world, i.e., we must affirm that the world is infinite in its extension. Second Conflict of the Transcendental Ideas. THESIS. ANTITHESIS. Every composite substance in the world is made up of simple parts, and nothing whatever exists but the simple or that which is com- posed out of the simple. No composite thing in the world is made up of simple parts, nor does anything simple exist anywhere in the world. 256 TRANSCENDENTAL DIALECTIC Proof. Assume the opposite, viz., that composite or continuous substances are not made up of simple parts. Now, com- position by the very nature of the case is an accidental relation, and can therefore be conceived to be absent without the destruction of the substance. But, when all composition is thought away, substance must either be simple substance or no- thing at all. It has been assumed, however, that there is no simple substance, and hence there must be nothing at all. Since the hypothesis is that substances do exist, we must therefore deny the proposition that substances are not made up of simple parts, i.e.., we must affirm the thesis that every substance in the world is made up of simple parts. Proof. Assume the opposite, viz., that a composite thing or substance is made up of simple parts. Now, it may be argued that no object in space can possibly be made up of simple parts, because as existing in space an ob- ject must have the same characteristics as space itself Space, however, is not made up of simple parts, but con- sists of spaces. Since, there- fore, we cannot get rid of composition in space, we must hold that every real thing which occupies space is composite. It may be said perhaps l that internal phenomena may be made up of simple parts, because they are not in space but only in time. This ob- jection, however, may easily be answered, because no ob- ject of perception, whether external or internal, can be presented to us which does ;not contain a manifold. Hence there is nothing ex- isting anywhere in the world which is absolutely simple. THE ANTINOMY OF PURE REASON 257 Third Conflict of the Traiiscendental Ideas. THESIS. ANTITHESIS. Causality in conformity There is no freedom, but with laws of nature is not aU that comes to be in the the only causality from which world takes place entirely all the phenomena of the in accordance with laws of world can be derived. To nature, explain those phenomena it is necessary to suppose that there is also a free causality. ProoJ. Assume the opposite, viz., that the only causality is that in conformity with laws of nature. Then, whatever comes to be implies an ante- cedent state, which is its condition ; for otherwise the effect would have existed always. And since this cause is itself a change, it requires a prior change to account for it, and so on ad infinitum. jNow, if all changes must con- /form to the law of nature, \ there is never a complete (cause, but only a conditional (cause. Assuming the truth of the law, therefore, that every effect must have a cause, we are compelled to suppose that there is another kind of cause besides that Proof. Assume the opposite, viz., that there is an absolutely spontaneous or free cause. Then, not only must this cause originate the series of causes and effects, but it must determine itself to originate that series ; that is, its act must take place without any antecedent de- termining it to act in ac- cordance with fixed law. But an act which begins presupposes a state in which the cause has not yet begun to act. And this state can have no causal connection with the preceding state of that cause. Hence, tran- scendental freedom contra- dicts the law of causality, which is essential to the 258 TRANSCENDENTAL DIALECTIC according to laws of nature, i.e., that there must be an absolutely spontaneous or free cause. unity of experience ; it is a mere Idea, which cannot be verified, and must therefore be denied. That is, there is no absolutely spontaneous or free cause, but everything takes place entirely in ac- cordance with the laws of nature. 165 a Fourth Conflict of the Transcendental Ideas. THESIS. There exists an absolutely necessary being, which be- longs to the world either as a part or as the cause of it. Proof. As the world of experience is a world in time, it contains a series of changes, each of which is necessarily depen- dent upon a condition prior to it. Now, the conditioned presupposes for its existence a complete series of conditions, ending in the completely unconditioned or absolutely necessary. Hence, change demands an absolutely neces- sary being. Moreover, this necessary being must be within the world of sense ; for the beginning of a series ANTITHESIS. There nowhere exists an absolutely necessary being, either in the world, or out- side of the world as its cause. Proof Assume the opposite, viz., that the world itself is a necessary being, or that a necessary being exists in it. Then, there must be either an absolutely necessary be- ginning in the series of its changes, i.e., an uncaused beginning, or the infinite series of changes must as a whole be absolutely neces- sary, while the parts are contingent. But, the former supposition contradicts the phenomenal law of all deter- mination in time, and the THE ANTINOMY OF PURE REASON 259 of changes must be deter- mined by what is antecedent in time, or has existed in a time prior to the series. In other words, the contingent implies a necessary being existing within experience, that is, a necessary being contained in the world. latter supposition contradicts itself, since the whole series cannot be necessary if no single member of it neces- sarily exists. Hence, there nowhere exists an absolutely necessary being in the world. Assume again, a necessary being beyond the world as the cause of it. Then, this being, as the highest member in the series of phenomena, initiates the series, and its causality must therefore fall into time. But, if its act falls into time, it comes within the sphere of experi- ence. Hence, there does not exist a necessary being out- side of the world. Putting the results of these two arguments together, we reach the general conclusion, that neither in the world, nor as a cause outside of the world, does there exist an absolutely necessary being. 166 Section IV. — Necessity of a Solution of the Transcendental ProUems of Pure Reason. i6 a The problems raised by the fact of antinomy must admit of solution. For, it cannot be admitted that any question which Pure Eeason presents to itself is incapable of being solved by Pure Eeason. When we are dealing with 260 TRANSCENDENTAL DIALECTIC phenomena, which necessarily imply an element not con- tributed by the mind but given to the mind, problems may well be suggested which cannot be solved on account of insufficient data. But in Transcendental Philosophy, as Kant argues, " the very conception which enables us to ask the question, must also give us the means of answering it, because the object to which it refers has no existence except in the conception." The only ground we have for saying that an object corresponding to the Idea exists is that Eeason gives us the Idea, and therefore we must be able to tell by an examination of the Idea what is involved in it. We cannot in this case appeal to the limits of our intel- ligence as a reason for saying that the problem is unintelligible, for its solution does not require us to go beyond the Idea itself 167 a It is, however, only in connection with the Cosmological Ideas that this absolute demand for solution applies. In the case of Eational Psychology the question was whether we are entitled to infer the existence of an independent soul or substance from the consciousness of the pure self; and in the case of Eational Theology the problem, as we shall see, is whether we are entitled to conclude to the existence of a Being who is the absolutely necessary cause of all things. In both of these cases, therefore, we have no knowledge of the object, since knowledge is necessarily I confined within the limits of experience. But, in the case of the cosmological ideas there is no question of the existence of a world beyond knowledge ; the whole question is whether the world as it falls within experience corre- sponds to the Idea of the unconditioned which Eeason demands. The Ideas in this case are relative, not to a thing in itself, but only to the objects of experience ; and, as these are known, the only question is how far the empirical synthesis corresponds to the Idea. Here, there- fore, Eeason must be able to solve its own problem. THE ANTINOMY OF PURE REASON 261 Section VII. — Critical Solution of the Cosmological Froblem. U The antinomy of Pure Reason rests upon this dialectical argument : " If the conditioned is given, the whole series of conditions is given. But objects of sense are given as conditioned. Therefore the whole series of conditions of objects of sense is given." As Kant immediately explains, this argument is dialectical, because the term " conditioned " is taken in two different senses. In the major premise it' means the conception of the conditioned, but in the minor ( premise the term conditioned is applied only to that which is given in experience and which therefore involves an^ , element of perception. Now, the conditioned in this latter sense does not directly imply the unconditioned. That is, we cannot reason directly from the experience of the con- ditioned to the experience of the unconditioned. The conclusion is therefore invalid, because it assumes that_the u ncondition ejL can be ex Berienced. Thus a purely logical principle, or a connection of ideas in thought, is changed into a material principle. 86 The sophistical character of the argument may be readily shown. In the first place, it is obvious that if the con- ditioned is presented in experience, reason demands a regress in the series of its conditions, and therefore it demands the unconditioned. For the very conception of the conditioned implies a condition, and ultimately the unconditioned. It must therefore be admitted that, starting from the conditioned as presented in experience, we are entitled to seek for a totality of conditions. But it by no means follows that we can obtain an actual experience of the unconditioned. J a In the second place, the antinomy rests upon a confusion between phenomena and things in themselves. It is no doubt true that, if the conditioned as well as its conditions are things in themselves, we are entitled to say that if the conditioned is given the unconditioned must be given. For, 262 TRANSCENDENTAL DIALECTIC whatever is absolutely real must be a whole, whether that whole is finite or infinite. As a matter of fact, however, we have no knowledge of things in themselves, but only of phenomena ; and hence all that we are entitled to say is, that the experience of the conditioned is impossible apart from a prior empirical condition : we cannot pass from the I necessity of a given condition to the experience of a totality '■ of conditions. Phenomena always presuppose an empirical synthesis in space and time, and therefore the synthesis exists only in the regress and in no sense apart from it. Eeason certainly demands that the regress should go on ad infinitum, because it demands completeness. But, on the other hand, it is obvious that this demand of reason for com- pleteness can never be satisfied by an empirical synthesis, which from the nature of the case is unending. 169 6 We thus see that both the thesis and the antithesis of the antinomies rest upon an illusion. In the major premise the conditioned is taken in the sense of a pure conception or idea, while in the minor premise it is taken in the sense of an object of experience. Logically, therefore, we have here the fallacy of sophisma figurae dictionis (the fallacy of an ambiguous term). We must not suppose that the fallacy is a mere logical trick ; for the illusion in this case is inevi- table. Prior to the distinction of phenomena and things in themselves we naturally suppose that we are dealing with things in themselves, and, as we have seen, on that presupposition wherever anything is given as conditioned, the unconditioned is presupposed. Eeason here acts upon its fundamental postulate, that thought cannot be satisfied until it has obtained an ultimate conception. Moreover, it is naturally supposed that the conditioned is connected with its condition independently of any succession in time ■ and it is just as natural, in the minor premise, to regard phenomena as things in themselves, as it is to take the conditioned, in the major premise, in the sense of a pure conception. In doing so, however, we fall into the confusion THE ANTINOMY OF PURE REASON 263 already pointed out between phenomena and things in themselves. For, the synthesis of the conditioned with a totality of conditions, which is expressed in the major premise, being a purely formal or logical principle, is inde- pendent of time, whereas the conditioned referred to in the minor premise concerns phenomena in time. Obviously we cannot legitimately pass from a purely logical synthesis to an empirical synthesis. In order to solve the antinomy, therefore, we have only to show that it rests upon a con- fusion between these two very different forms of synthesis. 1 a Take, for example, the first antinomy, where the thesis argues that the world is finite in extension, and the anti- thesis that the world is infinite in extension. If we assume that the world, or the whole series of phenomena, is a thing in itself, it must be admitted that one or the other of these propositions must be true, for the world must then be complete either as finite or as infinite. On that supposition, therefore, the antinomy cannot be solved, and Reason falls into contradiction with itself. The solution is, that the world is not a thing in itself, and therefore has no existence apart from the regressive series of ideas in which it is known. We do not therefore require to say either that the world is finite in extension or that it is infinite in extension, for a third alternative remains, viz., that it is neither finite nor infinite in extension, but is infinitely extensible ; in other words, that there is no limit to the possible deter- mination of the conditions for the conditioned of experience. If we take any given point in the regress, we find that it is finite ; but, on the other hand, there is no point at which the regress comes to an absolute end. 1 h What has been said of the first antinomy applies with equal truth to the others, so far as they deal with a series of conditions which exists only in the regressive series itself. In the second antinomy, where it is argued on the one hand that nothing in the world is made up of simple parts, and on the other hand that every thing in the world must be made 264 TRANSCENDENTAL DIALECTIC up of simple parts, it is assumed that we are dealing with things in themselves. But, when we see that the substances here spoken of are objects of experience, it becomes obvious that the regress implied in the synthesis cannot be said to be either finite or infinite, because it can go on ad infinitum,. In other words, the world is not infinitely divided or finitely divided, but infinitely divisible. The same thing holds of the series of causes and effects, for here, while Eeason demands a regress from the conditioned, the actual process of experience can never lead to absolute completeness either by summing up a finite series or an infinite series. Similarly, the synthesis which proceeds from conditioned existence in search of an unconditioned necessary existence can never be completed, because of the character of our experience as involving a successive synthesis. 172 a The Antinomy of Pure Eeason in its cosmological ideas thus disappears, when we see that it involves a confusion between the idea of absolute totality as applied to things in themselves and the necessity of seeking for completeness in the series of phenomena. At the same time it is of indirect value as a confirmation of the conclusion reached in the transcendental aesthetic, viz., that space and time are tran- scen dentally ideal and only empirically real. If there were any doubt of the truth of this conclusion, it would be set at rest by the proof based upon the antinomy of pure reason. This new proof would run thus : " If the world is a self- existent whole, it is either finite or infinite. But it is neither finite nor infinite (as shown in the antithesis and thesis respectively). Therefore the world (the sum-total of all phenomena) is not a self-existent whole." Here the illusion is clearly brought out, because the major premise obviously has the meaning of " the world as it is in itself," and the minor premise shows that on this supposition we must accept either the one alternative or the other. Since neither is true, the conclusion inevitably follows that the world is not a thing in itself, but simply the sum-total of THE ANTINOMY OF PURE REASON 265 all phenomena. Now, phenomena have no existence except in our experience, and this is what we mean by transcen- dental ideality. Section VIII. — Regulative Principle of Pure Beason in the Cosmological Ideas. 173 a The cosmological principle of totality demands that we ' should seek for completeness in the series of conditions as regards the objects of sense. Such completeness from the nature of the case cannot be found. Nevertheless, the prin- ciple is true in its own sense, for we can never be contented with anything short of an absolute totality of conditions, as is demanded by Eeason. Sensible objects, from the very fact that they are in space and time, can never present to us the unconditioned. The principle of reason, therefore, is merely a rule which demands a regress, and the series of conditions of given phenomena will not permit us to assume that we have reached anything absolutely unconditioned. Unlike the principles of judgment, which state the con- ^ ditions under which experience is possible, this rule does not tell us what are the conditions of possible experience. Nor again is it a constitutive principle, which can be employed beyond experience ; for in that case it would determine things in themselves. It is therefore a regulative principle of Eeason, meaning by this that it is constitutive neither of objects of experience nor of things in themselves, but is simply a principle which prevents us from assuming completeness in the series of conditions. By calling it " regulative " we draw attention to the confusion that arises from attributing objective reality to an idea which , serves merely as a rule. 266 TRANSCENDENTAL DIALECTIC Section IX.— Empirical Use of the Regulative Principle oj Reason. 174 a We have seen that no transcendental use can be made of pure conceptions, whether these belong to understanding or to reason. In other words, no conception, taken by itself, is constitutive. For the conceptions of the understanding become constitutive only in relation to sensible objects, and the conceptions or Ideas of Eeason are in no sense consti- tutive, whether we seek to apply them to phenomena or to things in themselves. Since from the nature of the case absolute completeness in the series of conditions never can be found, there is no meaning in asking whether the series is finite or infinite. It is obviously neither, since reason demands that we should seek for a totality of con- ditions, whereas the nature of experience is such that a totality of conditions is not a possible object of experience. The principle of reason, then, has been shown to be merely a rule for the extension and continuation of possible experience. If we keep this steadily before us, the conflict disappears ; for the critical solution reveals the illusion from which the contradiction arises, and it also enables us to see that reason is not in contradiction but in harmony with itself. 1. Solution of the First Antinomy, 175 « The solution of the first cosmological problem is readily discerned, when we see that the sole question is whether there is a regress to infinity, or merely a regress which is capable of being continued indefinitely. 176 a The idea of a totality in the series of conditions is merely the idea of a possible regress which has not yet been realized. The world as a whole does not exist as an object of experience : it is a pure Idea of Eeason. Hence THE ANTINOMY OF PURE REASON 267 we cannot conclude from the quantity of the world to the quantity of the regress, i.e., to the quantity of phenomena ; in fact the conception of the quantity of the world can only be discovered by determining what is the quantity of phenomena. From the very character of experience, as dealing with things in space and time, there is no limit to the regress in the way of quantity, and hence we cannot say that the regress proceeds to infinity. To say that it proceeds to infinity would be to say that we had already completed an infinite series, whereas it is just the character of quantity as applied to numbers that there is no ultimate limit to the series. The first and negative answer to the first antinomy therefore is, that the world has no first beginning in time and no extreme limit in space. 76 6 The affirmative answer directly follows, viz., that the regress in the series of phenomena proceeds in indeflnitum. The world of sense has no absolute quantity, i.e., its quantity is neither finite nor infinite. Every beginning is in time, and therefore presupposes a prior time, and every limit of ^ that which is extended or in space presupposes a space beyond that which is given. But space and time, as we know, are not determinations of things in themselves, but only of phenomena. Hence, while things in the world are conditionally limited, we cannot say that the world itself is either conditionally or unconditionally limited ; that is, we cannot say that its quantity is either finite or infinite. And as on the one hand the world cannot be presented to us in experience as complete, while on the other hand a series of conditions of that which is given as conditioned cannot be given as complete, the conception of quantity is simply a conception of the process which is involved in the continu- ous determination of phenomena. The regress does not proceed to infinity, but is merely capable of proceeding indefinitely. And since the quantity of experience does not exist outside of the regress itself, there is no meaning in saying that the world is either finite or infinite in quantity. 268 TRANSCENDENTAL DIALECTIC 2. Sohotion of the Second Antinomy. 117 a The solution of the second antinomy is of a similar character. If we start from a whole of perception, we can proceed to determine the conditions under which it exists, i.e., we can divide it into parts. Whatever is extended, or in space, is of this character. The division into parts thus involves a regress in the series of conditions ; for, we must proceed in our division regressively, i.e., by first taking one half, then subdividing that, and so on. If the parts involved in the process of division are themselves divisible, the process of division must go on in infinitum. The regress in this case, Kant contends, cannot be merely a regress in indefinitum, because the whole is already given, and therefore necessarily contains all the parts in it. At the same time we have to observe that in our knowledge the parts exist only in the regress ; and as there is no limit to this regress, we are not entitled to say, because the whole is divisible to infinity, that it contains an infinite number of parts. The regress in this, as in the former case, is such that it can never be completed, and therefore there is no possibility of our reaching either a finite number of parts or an infinite number of parts. 178 a The application of this principle to space is easily made. Every space perceived within its limits is a whole, and as from the nature of the case division can go on to infinity, each so-called "part" of space is itself again a space. Hence space is infinitely divisible. 178 6 The application of the principle to external objects or bodies directly follows. For, since every body is determined to be external from the fact that it is spacial, the same principle which applies to space also applies to body. Every body is therefore infinitely divisible, but without being made up of an infinite number of parts any more than a finite number. THE ANTINOMY OF PURE REASON 269 Transition from the Mathematical to the Dynamical Antinomies. Ida In the solution of the two first antinomies we have assumed that the conditions and the conditioned are both in space and time. This assumption, which is always made by common sense, inevitably gives rise to antinomies ; for, it implies that when any event is given in time as conditioned, its condition must be another event in time ; and that when any object is given in space, its condition must also be in space. Thus the conditioned and the condition are assumed to be contained in the same series, or are homogeneous, i.e., the same in kind. As we have seen, it is impossible to complete the series of conditions, on the assumption that these are in space and time; and if we assume that the conditions as in space and time are completed, or, in other words, that we have reached the unconditioned, we fall into the mistake of supposing that it is possible to reach an absolutely first member of the series. The result is that the antinomies that arise in connection with objects in space and time can only be solved by recognizing that it is impossible for the Ideas of Reason to coincide with the products of the Understanding. For, if we say that there is a finite series of conditions in space or time, we find that this series is too short for the understanding, which demands that every event in time and every object in space should be explained in reference to an antecedent event or a new space. In this case, therefore, Reason makes the series too short for the understanding. On the other hand, if Reason maintains an infinite series, then, since it is impossible for the Understanding to reach such a series, Reason makes the series too long for the Understanding. Thus in neither case can Understanding coincide with Reason. 796 When we go on to consider the third and fourth anti- nomies, we find that we can no longer assume that the 270 TRANSCENDENTAL DIALECTIC totality of conditions must necessarily belong to the same sphere as the conditioned. The two first antinomies may be called mathematical, in accordance with the distinction already drawn in the case of the categories, while the two second employ a dynamical relation of phenomena. In other words, the two first antinomies deal entirely with the magnitude of the series, while the two second deal with the dependence of the series on that which produces it, or the dependence of conditioned existence upon unconditioned or necessary existence. Now, when we reflect on the distinc- tion between the mathematical and dynamical categories, the possibility is suggested that there may be a different mode of solution in the two cases. In the former both thesis and antithesis are false, for neither a finite nor an infinite series can be shown to be a possible object of experience. But, in the dynamical antinomies Eeason itself suggests a distinction between phenomena and things in themselves, and therefore it may be that both thesis and antithesis are true, since the one may be true when we are dealing with phenomena, and the other when we are dealing with things in themselves. Thus both may perhaps be shown to have valid claims when they are properly delimited, a thing that was impossible in the case of the mathematical antinomies. 180 a No doubt even in the dynamical Ideas the same principle applies, so far as we are dealing simply with the question of the magnitude of the series. That is, the third and fourth antinomies presuppose, like the two first, a regressive series, and of this series we must say that it is either too long or too short for the understanding. But, on the other hand, the conceptions of the understanding which are employed in the case of the third and fourth antinomies need not necessarily be limited to this regressive series. That is, the unconditioned, whether in the way of an unconditioned cause or an unconditioned being, may not be homogeneous with the conditioned. For, there may be a THE ANTINOMY OF PURE REASON 271 non-phenomenal cause, and a non-phenomenal existence, to which the totality of phenomena is related in the way of dependence. This new mode of solution is manifestly possible. BO 6 From the point of view of the mere magnitude of the series of phenomena we cannot allow any condition to be supposed which is not sensuous ; but in the dynamical series of conditions, there is nothing to hinder us from presup- posing a heterogeneous condition, i.e., one which lies beyond the series of phenomena. Such a condition, as non- phenomenal, will be purely intellectual. What Reason demands is that there should be no contradiction in our judgments. But, there is no contradiction in maintaining that within the sphere of phenomena no unconditioned member of the series can be found, while yet such a member is possible beyond phenomena. On this supposition both Understanding and Reason would receive satisfaction ; for the Understanding would not find the series of conditions abruptly terminated, contrary to its demand that there should be a condition within the same series for whatever is given as conditioned, while Reason would receive satisfac- tion in the admission of the possibility of an unconditioned, existing outside of the sphere of phenomena. 81a Granting that there may be a non-sensuous condition of phenomena, a condition lying entirely beyond the series of sensuous conditions, we reach a different conclusion from that to which we were brought in the case of the mathe- inatical antinomies. In these there was no means of escape from the conclusion that both thesis and antithesis were false. But, while no doubt the dynamical series, from the point of view of phenomena, must be solved in the same way, our new view shows the possibility of a non-phenomenal condition which is empirically unconditioned. Thus satis- faction will be given both to Understanding and Reason. No more shall we attempt to find the unconditioned by the impossible method of summing up a series of phenomena. 272 TRANSCENDENTAL DIALECTIC We now see that both propositions may be true when they are properly understood. This method of solution was impossible in the former case, because there the whole question was whether we could find an unconditioned member in the series of phenomena, and of course it would have been contradictory to suppose that this member was non-phenomenal. 3. Solution of the Third Antinomy. 182 o There are only two kinds of causation conceivable, viz. natural or free causation. By natural causality is meant such a connexion of states with one another that the one necessarily presupposes the other, or follows in accordance with an inviolable rule. Causality of this kind is relative to phenomena, i.e., to events which occur in time ; and as the state on which the other depends must itself have come into existence, since otherwise its effect would not for the first time arise, it is obvious that every cause presupposes another cause, and hence that in this way an absolute or 182 6 unconditioned cause is impossible. Freedom, again, so far as it is a form of causality conceived as related to pheno- mena, is the power of bringing a state into existence spontaneously. This state must of course be in time : but the causality, if it is free, does not exist in time, and there- fore does not stand under another cause as its condition. So understood Freedom is obviously a transcendental idea, i.e., it is absolutely a priori. For, in the first place, it contains no element derived from experience ; and, in the second place, it cannot be presented as an object, in the sense in which we have experience of phenomena as objects. " Whatever comes to be must have a cause " : this is the principle of all possible experience, and no extension of experience can ever bring us to any cause which is not itself an effect, so long as we remain within experience. Eeason, however, discerning the impossibility of satisfying its own THE ANTINOMY OF PURE REASON 273 , demand for completeness in the way of causality, sets up the idea of a non-phenomenal or spontaneous cause, a cause which is not determined to act by anything but itself S3 a Now, there would be no possible solution of the antinomy, if the ordinary assumption that phenomena are things in themselves were true; for in that case no cause could possibly be found, which did not belong to the same series as its effect. In this as in the other antinomies we should have to say, that no unconditioned cause could possibly exist that was either finite or infinite ; in other words, we should have to regard the opposition between what reason demands and what understanding supplies as incapable of solution. But the peculiarity of this antinomy is that it is a question, not in regard to the possibility of summing up a series, but in regard to the question of the dynamical infiuence of a cause in relation to its effect. Here, there- fore, we may abstract from the finitude or infinitude of the series, and direct our attention solely to the dynamical relation of conditioned and condition. The only question is whether we can admit unreservedly that every event follows upon another event in accordance with an in- violable rule, and at the same time maintain that all events proceed from a free or spontaneous cause. If it is possible to show that both natural and free causation can be legitimately maintained, though of course in different senses, then we shall have a solution of the problem, not by the denial of both thesis and antithesis, but by the accept- ance of both as true in their own sense. No attempt must be made to set aside the law of natural causation, so far as phenomena are concerned; for, to do so would be to destroy what it has taken us in the Analytic so much trouble to . establish. The question is : Granting in the fullest degree the principle of natural causation, can we yet defend the reality of freedom ? How important it is to get rid of the false assumption that phenomena are things in themselves is obvious. If that were true, the problem of freedom would s 274 TRANSCENDENTAL DIALECTIC be absolutely insoluble; for, no absolutely free cause can possibly be found within the realm of phenomena. But, if phenomena are clearly seen to be merely phenomena, and not things in themselves : if, in other words, they are simply objects of our consciousness, determined in accordance with the ordering principles of our understanding : then we can see how these phenomena as a whole may have their ultimate source in that which is not itself phenomenal. This intelligible cause, being entirely free from the restric- tions of phenomena, is compatible with the idea of a free or spontaneous causality, while yet its effects may present themselves within the world of phenomena. By this method of solution, then, we can understand how the whole series of phenomena may proceed from an absolutely free cause, while yet as a phenomenon every effect presents itself as following upon a phenomenal cause according to the neces- sary law of nature. Possibility of Causality through Freedmn. 184 a The problem is : Admitting that no event can be found which is not subject to the inviolable law of causation, can we yet hold that every event proceeds from a free cause ? or must we maintain that natural causation and freedom are mutually exclusive ? 185 a As we have seen, there is no free cause in the realm of phenomena, i.e., nothing can be found in the series of events that is not dependent upon a prior condition in the way of an event. For, every cause is itself an event, in so far as it is an element in the phenomenal world ; and therefore it is useless to attempt to defend freedom by maintaining that certain events are exempted from the law of natural causation ; in other words, we must deny that within the sphere of phenomena there is any self-determined cause. 185 6 Granting, however, that the cause of every phenomenal effect is a phenomenon, are we bound to say that the THE ANTINOMY OF PURE REASON 275 causality of every cause is empirical ? May it not be that, while every phenomenal effect must be determined as a phenomenon by the laws of empirical causality, yet this empirical causality is itself the effect of an intelligible cause ? Is it not possible, in other words, that the empiri- cal causality proceeds from a cause which originates phenomena, and which as originating phenomena is not phenomenal but intelligible, while at the same time this causality presents itseK to us simply as a link in the chain of natural causation ? 185 c Is there anything in the nature of experience to confirm this . solution of the problem ? In the case of man we certainly have a being who on one side of his nature belongs to the world of sense ; and, so far as he is regarded from this side, he is just as much subject to the law of natural causation as other objects of experience. Like them he must have an empirical character, and what this is we learn from the effects which follow from the exercise of his powers. In the case of inorganic things, and even of the animals, we do not find anything that compels us to suppose that, besides being subject to the law of natural causation, they have self-activity or freedom. But in the case of man we have to recognize a marked distinction. No doubt he obtains a knowledge of other objects, and even of himself as an object, only by the application of the forms of thought to the perceptual element received through the sensibility ; but he is also conscious of himself as contrasted with other objects, and even with himself as an object, in the analytic judgment of self-consciousness, which, though it arises only in the return from the synthetic judgment, cannot be identified with it. Even the understanding, ( though it can make no real use of its conceptions except in relation to the matter of sense, is distinct in its nature from the sensibility ; and reason is obviously and emphati- ; cally a faculty which operates in entire independence of ' sensible conditions. Now, the fundamental character of the 276 TRANSCENDENTAL DIALECTIC objects of experience makes it impossible that the demand of reason for a unity corresponding to that which is involved in the analytic judgment of apperception should be satisfied in the way of knowledge. Thus, out of the con- sciousness of the essentially limited character of knowledge there arises by way of contrast the idea of a possible development of the rational life which is free from the restrictions of experience. In this way, as Kant thinks, the very consciousness of the limitation of reason in its theoretical use points beyond itself to a use in which it is free from that limitation. But the only other use of reason is the practical. Here the object is not something opposed to the self, or even the self as a phenomenon, but the self as a subject, which is set up as an ideal, the known world being regarded as the means by which this ideal is to be realised. 186 a Is there anything to show that this ideal of a completely realised self is more than a fiction ? Kant's answer is that we at least suppose Keason to have causality, because we impose upon ourselves imperatives, which imply that we believe ourselves to have certain powers of self-determination. There is obviously no meaning in speaking of obligation in the cases of lifeless things or animals, neither of which have will. The understanding lays down the rules which govern phenomena, and inasmuch as all phenomena are in time, it has to do only with the necessary connexion of events, present, past or future. There is therefore no meaning in asking what nature ought to do. The only question in regard to nature is what is true as a matter of fact, just as in the case of the triangle the only question is what are its 187 a actual properties. The ought is entirely independent of sensuous conditions, and expresses a possible activity based upon a pure conception or Idea. It is no doubt true that an act which ought to be done must be possible under conditions of nature. But what ought to be done is not in any way determined by these conditions. The will, in other THE ANTINOMY OF PURE REASON 277 words, which wills the ought, must will it independently of all sensuous desires, though it is true that what is so willed expresses itself as an effect in the phenomenal world. No j number or intensity of natural desires can ever give rise to the ought, because the sequence in the case of natural desires is purely in accordance with natural law. Hence any volition that is determined by such desires is con- ditioned, or presents exactly the same features as any other natural sequence. A volition proceeding from such desires is therefore never necessary in the sense of the ought : on the contrary Eeason refuses to be influenced by such desires, and demands conformity to its own idea, the idea of moral law. Eeason can either forbid or authorise such a volition, — forbidding it when natural desire is contrary to reason, authorizing it when natural desire is in conformity with Eeason, — but in either case it demands that volition should not be determined by desire but by the moral law. Thus Eeason presents to the mind an idea of things, which is contrasted with the idea of the phenomenal world ; and i though this new order must be compatible with the natural order, Eeason is guided by the ideal order constructed by 1 itself; and thus it refuses to admit that actions are neces-i sary in the sense of natural causation, maintaining that on I the contrary they ought to conform to its own idea. 1 Obviously, therefore, Eeason assumes that it has in itself the power of originating action independently of all natural 1 desire ; for, under no other supposition could it expect to 1 influence experience. ! 87 6 Let us assume, then, that it is possible for Eeason to be a cause of change in the phenomenal world. Even under ' this supposition Eeason must manifest itself in the pheno- . menal world ; in other words, the activity of Eeason must be conformable to the inviolable law of natural causation. Every cause expresses a fixed or invariable order in phenomena, and inasmuch as we must apply the conception of cause to the connexion of our own desires and volitions, 278 TRANSCENDENTAL DIALECTIC so far as these are phenomena, it is obvious that here also there is an inviolable connexion of cause and effect. This fixity or uniformity in the sequence of our desires and volitions is what is meant by the empirical character. No matter what may be the changes in the particular desires and volitions which are connected in the way of cause and effect, the principle of inviolable law is always maintained ; so that the empirical character is unchangeable, although the effects appear in changeable forms, according as there is a change in the empirical conditions. 188 a The will of every man, then, has an empirical character, which, on the supposition that we have made, is the exter- nal manifestation of free causality. When we fix our attention upon the empirical character, we must view the sequence of desires and volitions precisely in the same way as we view other phenomena. The sole question in this case will be what desires or motives can be inferred from these volitions. The only way in which this question can be answered is by tracing out the connexion of the actions with certain desires. We must apply the universal rule of natural causation, and this is the only possible way in which we can determine the actions of man so far as these are viewed simply as events in time. If therefore we had a complete knowledge of all the conditions in the way of desire, we should be able to tell in every case how a man would act, and thus to show that his actions followed necessarily under the given conditions. If man is thus as much a being of necessity as other things, so far as he is regarded from the phenomenal point of view, it is plain that we cannot establish freedom from any con- sideration of his empirical character. As we cannot possibly find any break in the chain of natural causation, there is no reason for affirming that man is free, so long as we keep to the phenomenal point of view, or regard 189 a him as a purely natural being. But, it may be possible to show that the inviolability implied in the empirical character THE ANTINOMY OF PURE REASON 279 of man is not incompatible with freedom, if there is ground for supposing that the very same action which comes under the natural law of causation, may yet proceed from reason, — not of course speculative reason, or, what is the same thing, ^ understanding, but reason in so far as it originates the idea of the ought and prescribes the law of action. From the point of view of moial obligation, what has taken place in conformity with the law of nature may yet be affirmed to be contrary to the law of reason. It is no doubt true that under the given conditions a certain effect in the way of action must follow, but this is not necessarily incompatible with the origination of these conditions by pure reason. There are cases at least in which we find, or believe we find, that the ideas of reason have determined the actions of man ; and if this is the case, the empirical law will merely be the outer expression of the manner of action of the real cause, in so far as its effects are presented to us under the conditions of our knowledge and experience. 1896 Supposing that reason actually is a cause in regard to phenomena, is it possible to maintain at the same time that as phenomena actions are determined in accordance with the law of natural causation ? in other words, is it possible that the empirical character is the outward expression of the intelligible character ? We have no direct knowledge of the intelligible character, because our knowledge is necessarily conditioned by time, and the intelligible character we have supposed to be out of time. The intelligible character we must therefore symbolise or) indicate by representing the dependence of phenomena on, it precisely as if the relation between it and phenomena' were that of antecedent and consequent. The expression of this intelligible character will present itself to us in the form of a modification of our consciousness in time, and as such the effect will of course appear to us as if it were determined by an antecedent event also in time. But the causality of reason in the intelligible character, as we have 280 TRANSCENDENTAL DIALECTIC to rerpember, is not strictly speaking in time ; and therefore it is not subject to the law of natural causation, or, in other words, it is free. We must then represent the matter to ourselves in this way : just as in the sphere of phenomena the cause is the antecedent, which is the condition of an effect, so reason is a cause which is not an antecedent, because it is not in time, but is that which is the condition of phenomena. For, since the condition that is implied in the exercise of reason is not sensuous, it cannot be said to begin to be. In this way, then, we can understand how there may be a free cause, while yet within the realm of phenomena the law of natural causation is inviolable. 190 a-; It must be observed that so far no attempt has been made to prove freedom as an actual fact, and indeed such a proof is impossible on the basis of theoretical reason ; for, as we have seen, there can be no knowledge of a free cause, since the conditions of our knowledge are such that only those objects which present themselves in the context of experi- ence, i.e., only phenomena, can be known. Theoretical Reason has simply shown us the limits of our knowledge, and has set up the idea that completed knowledge would necessarily be of the absolute or unconditioned ; and the antinomy into which reason falls in the present case arises from the conflict between the idea of an unconditioned cause and the conditioned cause which alone appears within knowledge. But, while we have not proved the existence of a free cause, we have shown that there is nothing con- tradictory in the idea of such a cause, if our doctrine is admitted that knowledge in us is not of things in them- selves but only of phenomena. 4. Solution of the Fourth Antinomy. 191a In the third antinomy we are dealing solely with the question as to the specific connexion of given events with their antecedent conditions. In the fourth antinomy, on. THE ANTINOMY OF PURE REASON 281 J the other hand, it is not a question of the causal dependence of one specific event upon another, but the more general question as to whether all events as contingent do not presuppose a necessary being. The problem, therefore, is not vyhether there is an unconditioned causality, but whether there is an unconditioned existence. J916 It is obvious that we cannot find any necessary being ■ within the world of phenomena. Every phenomenal object, 1 from the very nature of the case, is changeable, and ' change, as we know, necessarily presupposes an antecedent I condition without which the change could not take place, since otherwise something would come to be without a ! cause. As this applies universally within the sphere of i phenomena, no phenomenal object can be found which is in f itself necessary. Necessity in the case of phenomena only ' means the necessary existence of a given change, provided i that a certain condition precedes it. Absolutely necessary i- existence, on the other hand, means existence that is il necessary in itself, quite irrespective of its dependence i, upon anything else. If there were no existences except 1 phenomena, it is obvious that we could not maintain « even the possibility of an absolutely necessary being; for 5 phenomena as conditioned always imply conditions which i themselves are phenomena, and which therefore must be 1 192 a contingent, not necessary, existences. But we have to ( remember what has already been pointed out in regard to the distinction between the mathematical and the dynamical i antinomies. In the mathematical antinomies we are con- cerned solely with the completion of a series by composition and by division, and therefore the conditions must always be, like the conditioned, phenomenal. In the dynamical antinomies, on the other hand, where it is not a question of '' the completion of a series by composition or division, the condition does not necessarily belong to the same sphere as the conditioned. Just as in the third antinomy it is possible that there is an unconditioned and non-phenomenal cause, 282 TRANSCENDENTAL DIALECTIC so we may also hold that an unconditioned or non- 1926 phenomenal existence is possible. Here, as in the third antinomy, a way of escape is provided by a distinction being drawn between phenomena as contingent and a non- phenomenal being as necessary. Both the thesis and the antithesis of the fourth antinomy may therefore be true when understood in different senses. All phenomena may quite well be contingent, and therefore have only an empirically conditioned existence, while yet there may be a condition of the whole series which is not conditioned, i.e., an absolutely necessary being. Such a being must be outside of the whole empirical series, and therefore it cannot be said to exist in any sense in the phenomenal world. And as it must be related equally to every member of the series, we cannot regard any one member as empirically uncon- ditioned, nor can we grant that there may be any interference with the necessary dependence of one phenomenon on another. There is a distinction between the manner in which an unconditioned existence is conceived, as compared with the manner in which we represent an unconditioned free cause. In the case of the free cause there is a specific relation to a certain event which is the empirical condition of another event, and all that is maintained is the possibility that the causality of this free cause is not empirical but intelligible. In the present case, on the other hand, a necessary being, if it exists at all, must exist in entire independence of every member of the series, and must therefore be purely intelligible. This is necessary when we are speaking of existence, because otherwise the being would not be free from all contingency and dependence. 193 a The regulative principle of reason in the present case may be thus stated. So long as we are dealing with objects of experience, we must seek for an explanation of their pro- perties by reference to that which is their condition. Hence we can never come to a point where we can say that no further advance can be made in the discovery of the condi- THE ANTINOMY OF PURE REASON 283 I tions of phenomena. Moreover, we can in no case assume that there is a break in the necessary series of conditions ; i.e., we cannot in any case refer a particular mode of experience to a non-phenomenal condition, or regard a particular mode of existence as self-dependent. But, while this is true, there is nothing impossible in the supposition ' that the whole series of phenomena is dependent upon a necessary being, which, as free from all empirical conditions, is itself the condition of the possibility of all phenomena. This necessary being will be purely intelligible, i.e., it can never be an object of experience, but is only demanded by theoretical reason as a possible explanation of the contra- diction which arises when we suppose phenomena to be things in themselves. Concluding Remark on the whole Antinomy of Pure Reason. 1936 Our ideas are cosmological so long as they are related simply to possible objects of experience, which we seek to , complete in accordance with the demand of Eeason for the unconditioned. Since these ideas are related to objects of experience, they are transcendental; but in the course of our investigation we have learned that the unconditioned which Eeason demands cannot be found within the sphere of experience. Thus the cosmological Ideas turn out to be . merely regulative principles, i.e., principles which set before us an ideal of completeness in our experience, an ideal that can never be realised. We are thus forced to seek for the unconditioned by going beyond experience altogether. The idea of the unconditioned now separates itself entirely from experience, and sets up the idea of a supersensible J object, in support of which nothing in experience can be ' found. Here, therefore, we have ideas which claim to be absolutely self-supporting, or to present to us objects which by their very nature cannot present themselves within the realm of phenomena. It is especially the cosmological 284 TRANSCENDENTAL DIALECTIC Idea connected with the fourth antinomy that forces us to take this step. We can find within experience no object which is absolutely necessary, and as reason cannot be satisfied with an infinite series of dependent or conditioned existences, we naturally attempt to find the necessary exist- ence which reason demands in a purely intelligible object. Having thus gone entirely beyond the realm of experience, we must now attempt, on the basis of Ideas or pure conceptions, to determine the nature of the absolutely necessary being. The problem, therefore, to which atten- tion must be directed is to ascertain what is involved in the conception of an abolutely necessary being, and to determine how far the existence of such a being can be established on the basis of pure thought. 195 BOOK II.— THE IDEAL OF PUEE EEASON. Section I. — The Ideal in General. 195 a It has been shown in the Transcendental Analytic that no object can be known by means of a pure conception of Understanding, but that knowledge or experience necessarily implies an element of sense ; or, in other words, that the pure conceptions of the Understanding must be schematised. On the other hand, when elements of sense are presented, the Understanding is able to bring them under pure concep- tions or categories, and so to present them in concreto, i.e., in certain determinate forms; in fact a conception of experience is simply a conception of the Understanding as determined through its relation to the sensible. In the Transcendental Dialectic, again, we have seen that Ideas, unlike the categories of the Understanding, cannot find anything in the world of sense corresponding to them; and therefore the Ideas of Reason cannot be presented in concreto at all. These THE IDEAL OF PURE REASON 285 Ideas are simply regulative principles, setting up the idea of a systematic unity, an idea which can never be com- pletely realised in experience, but which nevertheless is of great value, because it prevents the mind from resting in anything short of a complete unity. When Reason in its inevitable course has carried us beyond experience altogether, it is obvious that we are still further removed from objective reality than in the case of the cosmological Ideas ; for here Reason sets up what may be called an Ideal, i.e., an Idea which does not consist in the completion of objects of experience, but which sets before the mind an object that transcends all the bounds of ex- perience, because it is absolutely complete in itself. The Ideal of Reason is, therefore, that of an individual reality determined purely by the Idea itself. What reason now demands is not merely a complete system of experience, but a totality of Reality comprehended within a single individual reality. This Ideal of an absolutely complete individual, which is to be determined by purely a priori rules, is one which, as we shall immediately find, cannot possibly be more than an Ideal. Since it is absolutely separated from all the con- ditions under which knowledge is for us possible, it cannot be known, or, in other words, it cannot be established by the theoretical reason. Thus it is really a transcendent conception. Section II. — The Transcendental Ideal. If we are to have knowledge of an object corresponding to this Ideal of Reason, we must have a complete knowledge of all that exists, and we must be able to determine it either affirmatively or negatively. We have on the one side an Ideal of a completely determined object; and if an object corresponding to this Ideal is to be known, we must be able to bring it completely within the sphere of our knowledge. 286 TRANSCENDENTAL DIALECTIC so as to see that it coincides with our Ideal. Now, we know from the nature of knowledge that experience is never complete, and therefore that we never can have the know- ledge of an object corresponding to our Ideal. We are thus left simply with an Ideal which can have no other than a regulative value, the Ideal of absolute completeness of knowledge, by which the understanding is guided, though we have no right to convert it into a constitutive principle. 196 6 The Idea of the absolute individual — an individual which contains within itself all possible existence — excludes certain predicates which are incompatible with it. It excludes all those predicates which have a meaning only in relation to other predicates, and it also excludes all the pre- dicates that are contradictory of positive existence. Kant's conception of thought is, that by its very nature whatever is positive — in other words, whatever is real — must be real or complete in itself. From this point of view it is obvious that everything conceived to be real must be independent of all relation to anything else. If thought can only admit that which is self-complete to be real, and excludes from this reality all contradictions, clearly reason will demand an individual which contains within itself all positive predicates, to the exclusion of all relations and all negations. This 196 c is what is meant by an Ideal of Pure Eeason. We may therefore view this Ideal as that which contains the material or transcendental content by reference to which all 197 a real determinations of things must be found. Like Spinoza Kant regards thought as of this character, that the predi- cates by which existence is determined must be purely af&rmative. Negative predicates, in other words, when applied to existence, merely indicate non-existence or want of reality, and therefore they express limitations in the unlimited reality which Eeason presupposes. 197 6 It is only by supposing a thing in itself to contain the totality of Eeality that we can think of it as completely determined. For, if all negations are simply the absence THE IDEAL OF PURE REASON 287 of reality, nothing can be absolutely real except that which contains no negation. But a being that is completely determined positively is at the same time an individual being, since we must assign to it every possible real predi- cate. Here, then, we have a transcendental Idea, which compels us to think of an individual reality that is com- pletely determined, and by reference to which all objects must be judged, so far as the question of reality is concerned. And this is the only genuine Ideal which Reason can possess ; for no other conception is that of an individual existence which is completely determined in itself or apart from all other existence. 197 c It is manifest that the mere possession of this Idea does not at once entitle us to assert the existence of a real individual corresponding to it. All that Reason requires ' for the determination of the degree of reality of any. object is that we should be in possession of this Ideal ; it is not required that we should have actual knowledge of an object corresponding to it. Reason, therefore, employs this Ideal as the pattern or prototype, of which all finite things are imperfect copies or ectypes. By reference to this Ideal things are found to contain more or less of reality, but all finite things must ever remain at an infinite distance from it. iSSa We think, then, of all finite beings as deriving their possibility from that Being which contains all reality within itself. All the predicates by which we characterise finite things as such are negative, since they are thought of as limitations of the supreme Reality, from which their content is conceived to be derived. The various determinations by which such things are characterised, in so far as they are distinguished from the unconditioned Being, express the different ways in which they are limited, just as any finite space is a negation or a limitation of infinite space. The object which Reason sets up is conceived, in the first place, as the Original Being, i.e., as the Being from which all other 288 TRANSCENDENTAL DIALECTIC being proceeds, but which does not itself proceed from any other. It is also conceived as the Supreme Being, since it contains within itself all reality, whereas other beings contain only a limited degree of reality. And, lastly, it is conceived as the Being of Beings, to indicate- that all other beings are conditioned and subject to it. But, while this Ideal is presupposed in the determination of finite things, it by no means follows that an actual Being corresponding to the Ideal exists. 198 6 This Original Being cannot be conceived as made up of a number of derivative beings, for obviously no number of such beings can yield the idea of a Being containing in 199 a itself all reality. The Original Being must therefore be conceived as one or simple. The derivation of all other possibility from the Original Being cannot be a limitation of that being; for, if that were true, the Original Being would be an aggregate of derivative beings, which we have just seen to be impossible. The Original Being is, therefore, not an aggregate or sum of all finite beings, but the pre- supposition of the possible existence of all finite beings. The distinction between one finite being and another is not a distinction within the original being, but all finite things must be regarded as a product of that being ; and consequently all the modes of reality that come within our experience, including our own sensibility, must be related to the Original Being as effect to cause, not as part to whole. 199 6 If we assume that this Ideal guarantees the existence of a real object corresponding to it, we shall be able to establish the existence of this object by setting forth the predicates that are involved in our conception of it, such as unity, simplicity, all-sufficiency, eternity. This supposed object is exactly what is meant by the con- ception of God in the transcendental sense, and thus the Ideal of reason is the object of a transcendental theology. THE IDEAL OF PURE REASON 289 By the substantiation or hypostatising of this trans- cendental idea, theology oversteps the limits prescribed by the very nature of the idea. It is no doubt true that Eeason cannot be satisfied with anything short of the conception of a totality of Eeality, but it is not necessary that this conception should have an actual object corresponding to it. In the actual extension of our knowledge, indeed, we cannot be satisfied without setting up the idea of the complete unity of the world and the self, and such a unity is conceivable by us only by reference to the Ideal of reason ; nevertheless, we have no right to assume that an existence corresponding to it is even possible : nor can we infer its existence from the fact that it brings to light the limitations of the finite object. X)a Why, then, does reason assume that the possibility of everything implies the existence of an Original Being containing within itself all reality ? This question we ." can easily answer by recalling the result of the Transcen- dental Analytic. It was there found that there is no system of experience, no knowledge proper, except in so far as thought combines elements of sense into unity. The sensible element must be given to us, and without it we can have no objects of experience at all. It is ' impossible to determine an object of sense except in so far as we apply positive and negative predicates to it. The real element of a phenomenon, however, must present itself to us within a single whole of experience. No sensible object can be known except that which appears within the context of one experience. Hence we can say that '' experience is necessarily for us a totality of empirical reality. What could be more natural than that we should ,' confuse this totality of empirical reality with a totality ; of real existence ? For, only by extreme care in the ' analysis of knowledge do we discover that knowledge consists only in the apprehension of phenomena. Hence, T 290 TRANSCENDENTAL DIALECTIC if the distinction between phenomena and things in them- selves is overlooked, we naturally take the unity of our experience as entitling us to affirm the unity of ultimate reality. 201 OS The unity which we find within experience is a ^distributive unity ; i.e., it is a special form of unification, implied in the process of the understanding in so far as the understanding employs categories under the guidance of the idea of completeness. This distributive unity is only an ideal, because completeness can never be found within experience. We suppose, however, not only that we actually reach unity in this way, but that we reach the unity implied in the totality of experience ; and this col- lective unity we conceive of as an individual thing containing all reality within itself. Having reached this point we take still another step, and identify the supposed totality of experience with ultimate reality, maintaining that there actually exists a Being corresponding to the idea of unity which reason has set up. The indi- vidual thing, in other words, which we suppose to have been established as the totality of all empirical reality, we confuse with an individual whole, which is not merely a whole of experience, but a whole of actual existence ; and as the totality of empirical reality is supposed to include all particular reality, this supposed individual existence is conceived of as the source of all possible finite reality, and therefore as accounting for the whole of the character of each finite thing. Section III. — Arguments of Speculative Reason for the Existence of a Supreme Being. 202 a Seeking to prove the existence of a Supreme Being, Eeason takes the following course. (1) Since conditioned existence always implies something which is its condition Eeason cannot be satisfied with anything short of an uncon- THE IDEAL OF PURE REASON 291 ditioned or necessary Being. (2) Having concluded that some necessary Being exists, whatever its character, the next question is what kind of Being that is, which is necessary ; and the answer is that t he only necessary Being is that which is the condition of all other reality, but which as itself unconditioned contains all reality in itself. In other words, the second step is to afSrm an ens realissimum, a Being which contains in itself all reality. Then (3) reason concludes that that which contains within itself all reality must be a Supreme Being, and must be the source or cause of all other beings. la Even if the premises of this argument were true, the conclusion deduced from them is fallacious. For, suppose it to be true, in the first place, that we can legitimately pass from conditioned or contingent existence to uncon- ditioned or necessary existence ; and suppose it to be true, secondly, that an ens realissimum, or a being containing all reality, is consistent with the idea of a necessary being ; it does not follow that the only necessary being is that which contains all reality within itself. For, there is nothing to prevent us from supposing the existence of a necessary being which is at the same time finite. That being so, we cannot reason from the existence of a necessary being to the existence of an infinite being. In short, granting that we have legitimately reached the conclusion that there is a necessary being, we cannot take a step further. 3 a But while the argument really leads to no conclusion, it is so simple and natural that it always commends itself to the human mind the moment its meaning is understood. It is an undoubted fact that things arise and perish, and hence we cannot be satisfied without asking what is the cause of the changes which they undergo ; and as we never find within experience an ultimate cause, we come to the conclusion that there must exist a first cause, which, as the supreme cause, is the sufiBcient explanation of every possible effect, and at the same time contains 292 TRANSCENDENTAL DIALECTIC within itself all positive reality. And because we cannot be satisfied with anything short of this absolutely unconditioned existence, we naturally suppose that such an existence is not only demanded by our reason, but is an objective reality. We think a supreme cause to be absolutely necessary, because it is absolutely necessary for us to think it. Hence, even in nations which have not advanced beyond the stage of polytheism, there is always a tendency towards monotheism. This idea has not been reached by any explicit process of reason or speculation, but simply from the natural operation of reason in its unreflective form. 203 6 There are only three possible arguments for the existence of God : (1) the physico-theological, or the argument from design, (2) the cosmological, or the argument from a first cause, (3) the ontological, or the argument from the Idea of God. 204 a The actual process which the mind has followed is to begin with the first, then to go on to the second, and to end with the third. But, while this is the case, what really impelled the mind to follow this process was the implicit idea that the existence of an infinite being is implied in the very idea of that being. It is thus the Idea of Eeason which gives rise to the process by which the mind, beginning with the specific forms of things as given in experience, passes to the idea of a necessary Being, and from that to the idea of a single supreme Being. It will, therefore, be advisable to examine these arguments in their logical order; and hence we shall begin with the ontological proof, the proof which rests upon a pure transcendental conception. Section IV. — The Ontological Proof. 204 i It is obvious that the conception of an absolutely necessary being is a pure idea, the objective reality of which cannot be proved by the mere fact that we have it, THE IDEAL OF PURE REASON 293 or that reason cannot be satislied with anything else. For, though Reason demands completeness, we find that, as completeness is unattainable as an object of knowledge, the idea of it merely serves to indicate that understanding operates legitimately only within the sphere of experience. °204c While it has been assumed that the idea of an absolutely necessary being is possible, no attempt has been made to prove the existence of such a being, or even to show that it is positively conceivable. JSTo doubt it is easy to define a necessary being as one the non-existence of which is impossible. But this does not tell us whether that, the non-existence of which is impossible, is really conceivable. Now, what we wish to know is the conditions which compel us to affirm that the non-existence of the absolutely neces- sary being is impossible. The unconditioned is no doubt thought of as a negation of the conditions by means of which the understanding is able to regard anything as necessary ; but this does not give us a positive conception of the unconditioned, and therefore it may be that it is merely a product of abstract thought, to which nothing whatever corresponds. 'U05a An attempt has been made to establish the existence of a necessary being by appealing to geometrical judgments, such as that a triangle necessarily has three angles, and it is supposed that such instances entitle us to reason from I !05 h the conception to the existence of a necessary being. All such examples, however, fail to establish the required conclusion. To show that the elements in a judgment are inseparable does not prove anything in regard to existence. A necessary judgment is merely one in which the predicate is inseparable from the subject; so that if the subject is admitted, the predicate must also be admitted. It is a necessary judgment that a triangle has three angles, because if we grant the existence of the triangle, we must also admit that it has three angles. But the judgment that the predicate belongs to the subject does not establish the 294 TRANSCENDENTAL DIALECTIC existence of the triangle as a real object. This confusion of logical necessity with necessary existence has been a fruitful source of illusion. It has been supposed that because we can form the conception of a Being that includes existence in its content, we can therefore go on to assert the existence of an object corresponding to it. The contention is that, because existence is a predicate inseparable from the subject as conceived, therefore the subject must have existence. But such an argument rests upon a confusion between logical and real necessity, i.e., between the necessary inseparability of a predicate from a given subject and the necessary existence of the subject itself. 206 o In an identical or analytic judgment the denial of the predicate involves at the same time the denial of the subject, for the predicate belongs to the very conception of the subject. Hence I cannot deny the predicate, and at the same time aifirm the existence of the subject ; since in that case I should be afiirming that the subject contradicts itself I cannot deny that a triangle has three angles, and yet affirm the existence of a triangle. But there is no contradiction in saying that there is no triangle, and there- fore no three angles. And it is precisely the same with the conception of a necessary Supreme Being. I cannot deny the existence of that Being without at the same time denying all the predicates involved in that conception. But there is no contradiction in denying that a necessary Being exists. If I deny that "God is almighty," I fall into a contradiction, because the conception God tacitly involves the conception almighty ; but there is no contradiction in denying the existence of God, because I at the same time deny all the predicates involved in the conception of God, and therefore the predicate almighty. 207 a It is argued, however, by Descartes, that there is one conception, and only one, the object of which cannot be denied to exist without contradiction, viz., this very con- THE IDEAL OF PURE REASON 295 ception of an absolutely necessary Being. For, it is said, the only possible conception of an absolutely necessary Being is that of a Being who possesses in himself all reality, and that which contains all reality must necessarily exist. Here, then, as it is maintained, is a case where the concep- tion of that which is necessary implies the existence of that which is conceived. If the existence of the object is denied, there is an internal contradiction, because the very possibility of the thing implies the predicate of existence. ,1076 The whole question plainly depends upon whether the judgment that a necessary Being exists is analytic or synthetic. If it is an analytic judgment, then, since the conception involves the predicate existence as part of its content, it is obvious that that thought cannot take us beyond the conception. To say that the conception con- tains existence as one of its predicates is not to say that an object lying beyond the conception actually exists. Either, therefore, there is nothing but the conception, or the existence of the object in separation from the conception has been assumed. It does not matter whether the predicate is called reality or existence, for it still remains true that in an analytic judgment nothing can be predi- cated except what is already involved in the subject. On the other hand, if it is admitted that the judgment is synthetic, as indeed it obviously is, we can no longer argue that there is any contradiction involved in denying the predicate existence. The only case in which the denial of a predicate is self-contradictory is in that of an analytic judgment. 08 a The illusion here is due to the confusion between a logical and a real predicate, an illusion into which there is a perpetual tendency to fall. Logic abstracts from all real content, and therefore anything at all may be taken as a logical predicate ; but the moment we seek to go beyond a conception and determine it further, the predicate must be established in some other way than by an analysis of the 296 TRANSCENDENTAL DIALECTIC conception. All real determinations, therefore, — or, what is the same thing, all existential judgments,— imply something 208 h more than the conception with which we begin. From the point of view of logic leing is not a real predicate, but simply the copula of the judgment. In the judgment, "God is omnipotent," the word is does not mean that God exists, but merely expresses the relation of the predicate to the subject in the judgment. The judgment " God is," or " There is a God," is a totally different kind of judgment, if it means that an object corresponding to the conception of God exists beyond the conception. On the other hand, if the judgment only means that I have the conception of God, all that is implied is that, having that conception, I also have all the predicates tacitly contained in it. So far as the content of a conception and the content of an actual object corresponding to it are concerned, it is no doubt true that the one is identical with the other. The conception of a hundred possible dollars contains precisely one hundred dollars, and the actual one hundred dollars also contains precisely one hundred dollars. We cannot therefore argue, from the mere existence of a conception, that an object ■ corresponding to it exists. It is no doubt true that to possess a hundred dollars is not the same thing as merely to have the conception of them; but this does not invalidate the fact, that the actual one hundred dollars has exactly the same content as the conception. The fundamental distinction is that the existence of the one hundred dollars implies something more than the mere conception of them: something therefore which must be obtained through an empirical synthesis. 209 a Even if we could think a thing in the absolute com- pleteness of its determinations, our conception would not -be enlarged by saying that it is. For the whole question is not whether certain determinations exist in our thought, but whether they exist independently of our thought. Hence, complete as the conception of a Being of the highest reality THE IDEAL OF PURE REASON 297 may be, I cannot affirm the objective existence of such a Being without going beyond my conception. And here we I see the source of the whole difficulty. In the case of objects of sensible experience it is at once obvious that from the mere conception of the thing we cannot pass to its * actual existence. The conception merely states the condi- i tions under which an object of experience is possible, and : we can only convert this possibility into actuality by means of perception, which enables us to obtain a knowledge of the object as contained within the context of our experience. Thus it is through perception that we are able to pass from ' the conception of an object to the knowledge of it. But, in I the case of the Idea of an ens realissimum there is no ! possibility of enlarging our conception by means of percep- tion, and hence it is not surprising that we cannot find in our Idea anything to distinguish its content from mere possibility. Valuable, therefore, as the conception of a Supreme Being is, as a mere Idea it cannot in any way ^ enlarge our knowledge ; nay, we cannot even say with Leibnitz that an actual Being corresponding to our Idea 210 a possibly exists. The famous Ontological or Cartesian proof of the existence of a Supreme Being is therefore invalid. We can no more extend our knowledge by the addition of predicates to our conception than the merchant can better his position by adding a few noughts to his cash account. Section VI. — 27ie Cosmological Proof. Like the ontological proof, the cosmological affirms that ilO 6 there is a connexion between a necessary being and a being that contains all reality. But, instead of beginning with the highest reality and arguing that it must be identified with an absolutely necessary being, the demonstration proceeds in the reverse direction, and reasons from an absolutely necessary being to the highest reality. The steps are as follows. Some being or other is given in 298 TRANSCENDENTAL DIALECTIC our experience; this being presupposes the existence of an absolutely necessary being; and this absolutely neces- sary being can only be a being which contains all reality within itself. This seems a perfectly natural 211 a and reasonable method of proof. Stated in syllogistic form it runs thus : " If anything exists, an absolutely necessary Being exists. Now, at least I myself exist. 211 h Therefore an absolutely necessary Being exists." The minor premise states an undeniable fact, for, whatever else may be denied, it will not be denied that I know my own existence. The major premise contains the inference from this fact to the existence of an absolutely necessary being. The argument, then, begins with a fact of experience, and thus in form it is quite different from the ontological argument, which proceeds entirely a priori. It is also different from the physico-theological proof; for, though like that argu- ment it rests upon experience, it differs in this way, that it takes no account of the peculiar characteristics of the objects of our experience, but reasons entirely from the general 212 a character of finite reality. The final step in the argument is to reason from the existence of the absolutely necessary Being to the existence of a Being containing all reality within itself. The only absolutely necessary Being, it is argued, is the one which contains in itself all possible predicates — a Being, in other words, that is completely determined. N"ow, the only being that corresponds to this demand is the Being which contains all reality within itself, an ens realissimxim. We must, therefore, identify the absolutely necessary being with the Supreme Being. In other words, since an absolutely necessary being exists, a Supreme Being necessarily exists. 212 6 This argument contains a whole nest of dialectical assumptions. It is really the ontological argument in a disguised form; for, though an appeal is made to experience, the whole weight of the proof consists in reasoning from an Idea of reason. In order to have a firm foundation for THE IDEAL OF PURE REASON 299 itself, the argument, unlike the ontological proof, appeals to a fact of experience. But the only use it makes of this fact is to pass from it to the conception of a necessary Being. For obviously experience can tell us nothing as to the nature of this necessary being, and hence the argument from this point onward is made to depend entirely upon an idea of reason. The question is asked, What kind of being is a necessary being? and the only valid answer, as it is held, is that a necessary being is one which is absolutely real, or contains within itself all reality. In this argument, then, i the necessary being is declared to be identical with the Supreme Being. The only difference between this argu- ment and the ontological is that the latter starts with an absolutely real being and identifies it with an absolutely necessary being, while the former starts with an absolutely necessary being and identifies it with the absolutely real being. There is therefore no distinction in principle between the two arguments; in other words, the cosmological argu- ment is just the ontological argument, with superfluous additions which in no way add to its value. Nothing is gained by the appeal to experience, for the whole force of the argument depends upon the inference to absolutely necessary reality. 14 a The cosmological argument has the additional defect of an ignoratio elenchi. It pretends to rest its case upon experi- ence, while in reality it appeals to pure reason. The new way which it offers us is immediately abandoned, and we 14 & enter upon the old path. There is no difificulty in exhibiting the dialectical assumptions involved in the argument, and it will be quite enough to point them out without further elaboration. iie In the first place, the argument reasons from effect to cause. Now, the principle of causality has no relation to actual existence except within the sphere of experience. When a certain event is given in perception, we can legitimately reason back to its cause, but this cause is itself 300 TRANSCENDENTAL DIALECTIC a member of the series of empirical causes. The cosmo- logical argument, however, reasons from a fact of experience to a cause entirely beyond experience ; i.e., it proceeds by 2]4(^a method which is obviously illegitimate. Secondly, the impossibility of an infinite series of causes as an object of experience is made the ground for the inference to an absolutely first cause. But, as we have seen, such an inference is entirely illegitimate within experience, and much more so when it is extended beyond experience 215 a altogether. Thirdly, the argument assumes that, by abstrac- ting from all the conditions of the series of causes and eifects, we obtain absolute completeness of the series. But what this means is simply that we assume completeness of conception because there is no object whatever before the 215 h mind. And fourthly, it is taken for granted that the conception of all possible reality entitles us to affirm the possibility of a Being containing all reality. The assump- tion, however, is inadmissible. In order to pass from a possible conception to a possible reality, we should require something enabling us to make the transition from the conception to objective existence ; in other words, it would be necessary for us to have an experience corresponding to the object of our conception, and such an experience is from the nature of the case impossible. Source of the Dialectical Illusion in all Transcendental Proofs. 215 c Both the ontological and the cosmological proofs are transcendental, i.e., they attempt to show that we can obtain a knowledge of the existence of God from pure conceptions or Ideas. What is the source of the Dialectical Illusion which induces us to pass from the conception of necessity to the conception of the highest reality, and then to realise and hypostatise this Idea ? 215 c^ It is no doubt true that the knowledge of contingent THE IDEAL OF PURE REASON 301 existence compels us to think that there must be something which is not contingent or necessary. On the other hand, we have a knowledge of what is necessary only in the sense that we know the necessary condition of that which is given as conditioned. It therefore follows that we can predicate neither necessity nor contingency of things in themselves. It is obvious that ultimate reality cannot be both con- tingent and necessary, and therefore a contradiction would arise if we were compelled to affirm both necessity and contingency. Neither of these judgments, therefore, can be constitutive ; i.e., we cannot say that things in themselves are either necessary or contingent. But, while this is true, there is nothing to prevent us from regarding them both as subjective principles. The experience of the contingent compels us to seek for something necessary as its condition, and we cannot be satisfied until we have found something absolutely necessary, i.e., something not itself contingent. On the other hand, the impossibility of finding anything necessary within the sphere of experience warns us that we must not convert this search for necessary being into the dogmatic assertion that we have found it. These two principles, taken in this sense, are principles of discovery, serving to keep before us the impossibility of our ever being satisfied with anything short of the unconditioned ; in other words, they are regulative principles, and as such not inconsistent with each other. The one principle tells us that in all our speculations on nature we should proceed as if there were a first cause. This principle is a valuable regulative principle, because it enables us to systematise or unify all our knowledge, a knowledge which can be obtained only if in all our enquiries into the causes of phenomena we keep before our minds the Idea of a supreme cause as the goal of all our efforts. The other principle prevents us from supposing that we have actually reached a first cause in the process of extending our knowledge, reminding us that any cause we assign must from the nature of the case 302 TRANSCENDENTAL DIALECTIC be conditioned, and therefore that our knowledge is a 216 process that can never come to an absolute end. As it is impossible to find in the whole system of phenomena any- thing absolutely necessary or unconditioned, that which is supposed to be the unconditioned must be conceived as lying outside of experience. It is the Idea of a supreme cause, which can never fall within the sphere of our knowledge, but is simply an ideal guiding us in the con- tinuous extension of our knowledge. 217 a The Ideal of a Supreme Being is, therefore, a purely regulative principle of Keason, an Ideal which we must keep before ourselves in all our efforts to discover the causal connexion of phenomena. It is a rule by reference to which that systematic unity which is necessary for the explanation of the world by universal laws is made possible for us, but it does not entitle us to assert the existence of a necessary Being. It is, however, natural and indeed inevitable that this regulative principle should be supposed to be constitutive. Just as space, which is merely a form of our sensibility, is supposed to be an actual existence, making possible all the figures contained in it, so the idea of an absolutely real Being, as the supreme cause of all phenomena, is naturally converted into the objective exis- tence of such a Being. Because this idea is demanded by Eeason as a condition for the systematising of our experience, it is naturally enough objectified, and it is supposed that an actual object corresponds to it. Thus a regulative principle is converted into a constitutive principle. We can see at once that this substitution has been made, if we consider that, while there is a meaning in speaking of the Supreme Being as the necessary condition of all the changes in the world, there is no meaning in speaking of this Being as itself necessary. The idea of necessity means for us necessary connexion, and therefore we cannot regard it as involving a necessary existence beyond experience. THE IDEAL OF PURE REASON 303 Section VI. — The Physico-theological Proof. 18 a The ontological proof is an argument from the Idea of Reality as a whole ; the cosmological an argument from ! the general character of existence as known to us in :' experience. As neither of these arguments from the general : character of things has been successful, the only other way in which we can hope to establish the existence of a Supreme Being is by a consideration of the peculiar or i specific character of objects as known to us. Things may ; be so constituted and related to one another that we may ,:' be able to base upon them an argument for the existence of i a Supreme Being. If such a proof can be given, we may properly call it a physico-theological proof. Should this also fail, we must submit to the conclusion that there is no possible way in which speculative reason can estabhsh the ■1 existence of a Being corresponding to the transcendental Idea. ;186 Now, even without entering into a special examination j' of the new course suggested, we can see at once that it ( can hardly be successful. The argument is to be based i upon the character of objects of experience. But we know i already that it is impossible to find within experience , anything corresponding to the idea of a necessary and all- I sufficient Original Being, since this Idea takes us entirely 1' beyond the sphere of the empirical and conditioned. jl8c The physico-theological argument, which is the oldest and 1 simplest of all, must always have considerable weight with the popular mind. To the study of nature it gives life and movement, because it enables us to view things as if they were produced by intelligent design, and indeed the argument itself arose from the natural tendency to look at things in 9 a this way when we are studying nature. But, though the usefulness of this way of conceiving things is undeniable, it cannot be shown to have any claim to demonstrative certainty. The truth is that the physico-theological argu- ment depends on the cosmological, and as the cosmological 304 TRANSCENDENTAL DIALECTIC rests upon the ontological, ultimately there is only one argument, the ontological. 219 6 The main steps in the physico-theological argument are these: — (1) We find in the world distinct evidences of purpose, or the adaptation of means to ends, and this purpose or adaptation exhibits great wisdom, being found in the minutest objects and extending as far as our experience goes. (2) This adaptation does not belong to things themselves ; i.e., it cannot be accounted for on purely me- chanical principles, for these will not explain how different things are so adapted to one another as to conspire to a single end. We must, therefore suppose that they have been specially adapted by a rational principle, acting from the idea of an end to be secured. (3) There must, then, be one or more causes of this adaptation, and we must further suppose the cause or causes to be intelligent or free. (4) That there is only one intelligent cause we are entitled to infer from the fact that the world as a whole must be regarded after the type of a skilfully constructed edifice; and, though it is true that our observation is limited in extent, we are entitled to infer by analogy that we should find the same unity and harmony of things if it were unlimited. 220 a The physico-theological argument, based as it is upon the adaptation and harmony of the different forms of nature, therefore presupposes a matter or substance which is to be adapted to an end. In order to prove what is required, viz., a Creator of the World, we should have to suppose that this matter is not in itself of such a character as to adapt itself to an end, and therefore that the substance of the world is the product of supreme wisdom. Now, it is impossible to prove this by an argument which appeals to the analogy of human art, for that analogy rests upon the very idea that the matter is not in itself adapted to an end, but is only adapted by an intelligent being distinct from it, who acts upon it. The argument from THE IDEAL OF PURE REASON 305 design therefore cannot prove more than an architect of the world, who is capable of shaping a given material, but is at the same time limited by the character of the material with which he works : it cannot possibly prove a Creator of the world, who is the source of the matter as well as the adaptor of it. If we are to establish the contingency of matter itself, i.e., to show that it is dependent upon an intelligent principle, we can only do so by a transcendental argument, — the very thing which the argument from design was intended to avoid. 20 J The physico-theological argument, then, presupposes the existence of matter as incapable of adapting itself to an end, and therefore reasons that the adaptation must be contingent, or due to a cause distinct from matter itself And as this cause must be adequate to the production of the effect, it is argued that there must exist a Being who possesses all wisdom, power, etc. ; in a word, a Being who is absolutely perfect or self-sufficient. 20 e Now, however true it may be that in our observation or experience we find everywhere marks of adaptation or pur- pose, it is impossible for us, just because our experience is limited, to say that this adaptation is due to a Being possessing infinite power. Nor can we infer from the order of the world that absolutely perfect wisdom is required for its production, or that the unity of the world establishes the absolute unity of its Author. In all these respects the argument reasons from the limited to the unlimited. It is therefore impossible by this method to establish the existence of a supreme cause of the world, or to obtain from it a principle of theology which is to serve as the basis of religion. 21a This argument, therefore, at the most cannot take us beyond the very great power, wisdom, etc., of the Author of the world. Hence we are compelled to abandon experience altogether, and to fall back upon the contingency which we had already inferred from the order and design of the V 306 TRANSCENDENTAL DIALECTIC world. The argument from contingency to a necessary Being, which is the first step in the new process, is simply the cosmological argument over again ; and, as we already know, the cosmological argument derives its support entirely from the ontological argument, which identifies the abso- lutely necessary Being with the Being that comprehends all reality. The physico-theological argument is unable to establish the existence of a Supreme Being in its own \ manner, viz., by an appeal to the specific character of objects of experience, and therefore it suddenly falls back upon the cosmological proof; and as the cosmological proof is simply the ontological proof in disguise, the argument from design really rests upon Pure Reason, though it began by professing to make use of nothing except that which is proved by experience. 221 b When we thus see that the physico-theological proof derives all its support from the ontological, we are forced to conclude that Pure Reason cannot establish the existence of a Supreme Being ; for, besides the three arguments which have already been considered, there is no other argument, based upon purely theoretical Reason, which has even apparent validity. Section VII. — Criticism of all Speculative Theology. 222 a Although Reason in its speculative use cannot prove the existence of a Supreme Being, it prepares the way for a valid proof, provided that can be given from some other source. The conception of a Supreme Being, as demanded by pure speculative reason, is the only conception which is consistent with the nature of such a Being, if it exists at all. Hence pure speculative reason brings the conception of the Supreme Being into harmony with itself and with the aims of our intelligence, removing from it all that is incompatible with the conception of an original Bein" and 222 h purifying it of all empirical limitations. Thus, though the THE IDEAL OF PURE REASON 307 Supreme Being is for purely speculative Eeason a mere ideal, it is an ideal without a flaw, expressing the demand which Eeason necessarily makes as essential to the com- pletion of the whole of human knowledge. If we can establish by an argument from the moral consciousness the existence of a Supreme Being, as Kant maintains, then our ideal will be of great importance, as it will enable us to state what must be the character of the Supreme Being. For, the enquiry into the nature of the speculative reason ' has liberated us from the tendency to identify phenomenal with ultimate reality, and thus it prevents us from predi- cating of a Supreme Being characteristics which apply only to objects of experience. And not only does speculative Eeason free us from this conclusion, but it enables us to construct an ideal of the Supreme Being that is in all respects consistent with itself The conception which is thus formed is that of a Being that is absolutely infinite, one, existing apart from the world, eternal, i.e., not subject to temporal conditions, unaffected by spacial limitations, etc. All these are purely transcendental predicates, pre- supposing liberation from the confusion of the phenomenal with the real, and thus providing us with a purified conception of the Supreme Being which can be obtained in no other way. Theoretical reason, therefore, is the ally of practical reason, in so far as it supplies a perfectly self-consistent ideal which is the only possible basis of Theology. 225 THE METAPHYSIC OF MORALITY. Section I. — Transition from Ordinary Moral Conceptions to the Philosophical Conception of Morality. i In entering upon the enquiry into the conditions of the ifloral consciousness, Kant is seeking to solve the problems, which he originally announced as those in which he was most interested, viz,, the existence of God, Freedom and Immortality.!) His investigation into the conditions of knowledge, aa-' carried on in the Critique of Pure Keason, is preparatory to the discussion upon which he now enters. The former problem was forced upon him by the apparent impossibility of reconciling the reign of law in nature with the freedom and immortality of man and the existence of God. For, if man, like other beings, is subject absolutely to the law of natural causation, there seems to be no place for freedom, and therefore no place for immortality ; and if, as the school of Locke held, knowledge is limited by what falls within sensible experi- ence, it is impossible to prove the existence of God. Kant, therefore, found it necessary to make a critical investigation into the power of human reason to solve these high pro- blems. ^ The result of his investigation is, as he contends, to show that we have no knowledge of the objects of what he calls " Ideas " ; in other words, that the supersensible in any form lies beyond the reach of the theoretical reason. At the same time we can think or believe in the super- sensible. Thus, although the Critique of Pure Eeason THE CONCEPTION OF MORALITY 309 compels us to deny that we have a knowledge of noumena, it by no means excludes the possibility of the existence of noumena.} In fact the very character of our knowledge is such that we cannot fail to see its essential limits. Knowledge for us is not a complete whole, and as the mind cannot be satisfied with less than a complete whole, we are compelled to ask whether we cannot establish the existence of objects corresponding to the Ideas of reason by an investigation into the moral consciousness. Morality is obviously bound up with the reality of a free subject; for, if man is not free, his actions cannot be attributed to himself, and therefore there will be no distinction between him and the lifeless things of nature. Our investigation,) however, into the nature and limits of knowledge has shown j the possibility, and even the probability, of the existence of a noumenal self, which is not subject to the conditions of sensible objects, but is free or self-determined. It is to the consciousness of such a self, a self not subject to the limitations of the objects of experience, that we must look for the defence of freedom and the explanation of morality. The limitation of knowledge to phenomena gives us a point of view from which we can see the possibility of this free self-determining subject. The principles of experience point beyond experience, and the main object of the critical consideration carried on by Kant in his ethical works, and especially in the Critique of Practical Reason, is to deterA mine whether the nature of the moral consciousness does not enable us to establish the reality of freedom, immor-i tality and God. In the Metaphysic of Morality, Kant therefore takes his stand on the ordinary moral con- sciousness, and tries to find his way through it to the essential ideas of morals. Just as the Critique of Pure Eeason started from experience and asked for its conditions in knowledge, so this ethical treatise starts from the moral law, which Kant calls a quasi-factum, and goes on to ask what are its conditions. His first 310 THE METAPHYSIC OF MORALITY object, therefore, is to set forth the nature of the moral law in its purity. 225 a / The common moral consciousness is quite clear that there ""is nothing absolutely good except a good will ; and Kant suggests that this is affirmed in the most unqualified sense, because it is believed that no being, human or superhuman, can be called good that is not distinguished by a good will.y The ordinary mind has no hesitation in setting aside, as irrelevant to morality, all those inward or outward advantages which distinguish one man from another. It will not admit that a man is good because by nature he is endowed with superior intelligence, nor does it allow, that external advantages, such as great wealth, can be regarded as contributing in any way to the essential character of a man. On the contrary, when a man whose will is not good is prosperous and happy, the moral consciousness feels profound dissatisfaction. There can, therefore, be no doubt that the common opinion of men is, that no one is worthy to be happy who has not a good will. 225 h \ What, then, is a good will ? The will is good, not because it brings good consequences, or because the end aimed at is realised, but solely because the good is willed. When Kant denies that the moral character of an action is determined by its consequences, he no doubt means conse- quences in the way of happiness to the individual or others. His view is, therefore, the direct opposite of that taken by John Stuart Mill, who maintains that the character of an action is determined entirely by its felicific consequences. ; An act is good, according to Kant, if the will is good, or, what is the same thing, if the motive of the man is solely and entirely good. Kant's second point is that the goodness of a man's will is not to be judged by his success in obtaining the end at which he aims. . Thus, to take Mill's illustration, the moral act of a man wbo attempts to save another from drowning purely from the motive of duty is absolutely good, even if he does not succeed in saving the man. And THE CONCEPTION OF MORALITY 311 so in all other cases. It must be observed, however, as Kant cautions us, that a good will is not the same thing as a mere good wish. The good will is a persistent and resolute endeavour to secure the good end, and nothing 226 a short of that can be called good. We must, therefore, ask 2266 what is the ground of this belief in the good will. That the true end of life is not happiness seems evident fromj the fact that, when reason is devoted to the pursuit o^ happiness, the end is never obtained. This partly explains why those who begin by assuming that happiness is the end of life, are very apt to fall into misology, or hatred of reason, when they find that the attempt to secure it by the exercise of reason only results in greater unhappiness. The"\ true lesson from this failure is that the real function of] reason is to enable man to secure the true end of life, which; !27ais not happiness but goodness. The aim of reason in its practical as distinguished from its theoretical use, is to produce an absolutely good will, not a will which is good only as a means to happiness. This will, it is true, is not the complete good, but it is the supreme good, and the condition of there being a complete good. In other words, we can only justify the natural desire for happiness on the ground that happiness should in some sense be conditioned by goodness. Since goodness or virtue is the object of the wisdom of nature, there is nothing inconsistent with that wisdom in the fact that the attempt to secure the complete good by aiming solely at happiness should result in failure. S7 6 To get a clear idea of what is meant by an absolutely good will, we must first analyse the conception of duty, a conception which is not indeed identical with that of an absolutely good will, but only with a will which is good as ,1 manifested under the limitations of our sensuous nature and 1* notwithstanding the obstruction of external circumstances. 227 c What, then, are we to understand by the term duty ? In J the first place duty excludes, not only all direct violations / ! 312 THE METAPHYSIC OF MORALITY M morality, but all acts the motive of which is inclination, (.even when these are not in themselves opposed to duty. Direct violations of duty, such as murder or theft, may be at once set aside, for no one would say that a man does his duty, who acts in conscious defiance of it. We may also set aside actions which, though they are neither contrary to duty nor proceed directly from natural inclina- tion, are yet done in order to gratify some other inclination, as when a shopkeeper deals fairly with all his customers because he believes that honesty is the best policy. There is much greater difficulty when duty and inclination happen to coincide. Thus, self-preservation is at once a duty and in normal cases a natural inclination. Now, there is no moral value in self-preservation which is the outcome of natural inclination, but only in that which springs from duty, though natural inclination would lead a man to do away with his life. So benevolence is a duty, but it has no moral value when it merely proceeds from a sympathetic disposition. It is only when a man is bene- volent purely because he regards benevolence as a duty that his action is moral. I An action in short has moral value in so far as its motive as well as its content is duty.\ 228 a But, secondly, the moral value of an action is determined purely sensitive being never rises above immediate inclina- tion; and such a being is good when his will is good, irrespective altogether of the consequences of his act. pa What, then, is the character of a law, the idea of which I is to determine the will independently of all desire and therefore of all consequences in the way of happiness, a law which we are entitled to call good absolutely ? Since the will must in no way be moved by desire for happiness, it ' is obvious that the principle of the will must be absolutely * universal, i.e., it must be a law, not for persons under I particular conditions, but one which is applicable to all men under all circumstances. The law affirms that we must ' act in such a way that the maxim of our action may have the force of a universal law. Nothing less than this ' is consistent with moral law, and even in their ordinary . judgments men recognise the validity of this principle. When they judge the actions of themselves or others, they tacitly or expressly ask whether those actions are binding 314 THE METAPHYSIC OF MORALITY 230,6 upon all men. "As a matter of faqt- no moral principle is based upon feeling, as is usually supposed, but is real^ly an obscurely thought metaphysic, which is present in the mind of every man as a part of his rational faculty. This is soon discovered by any teacher who subjects his pupil to a little Socratic interrogation in regard to the imperative of duty and its application to the moral estimate of conduct " (H. 7. 178). The conscience of men gives forth an unmis- takable sound, the moment it is freed from the deceptive influence of passing natural inclination. No one, for example, has any difficulty in rejecting lying as inconsistent with the very idea of law ; for he sees at once that if he may make false promises, every one may do so with equal justification. Thus his maxim is self-contradictory; for, if every one lied there would be no lying, since no one would believe another. Obviously, therefore, we cannot without self- contradiction will that the maxim to make false promises 231 a should become a universal law. Duty, then, as we may conclude, is the obligation to act from pure reverence for the moral law, i.e., in entire independence of sensuous desire. This is the only motive of a good will, a will which is good without limitation. 231 b Man, however, not only possesses reason, but he has by nature certain wants and desires, in the complete satis- faction of which he places his happiness. Hence arises a ''conflict in his mind,- — reason prescribing absolute con- formity to duty, and desire inciting him to seek for , happiness. From this conflict issues a natural dialectic, i.e., a kind of practical illusion, proceeding from the disposition to relax the binding obligation of duty in order to allow of 231 c the satisfaction of the natural desires. This practical conflict of motives forces men to enquire into the ^relation of desire^ t.a_reason_; in other words, to enter upon the problem of moral philosophy, a problem which cannot be solved in the practical any more than in the theoretical sphere, without a thorough criticism of human reason. THE CONCEPTION OF MORALITY 315 Obviously we cannot be permanently satisfied with two 1 mutually contradictory principles, and the only way of determining which is to be accepted is by a special investigation. 232 Section II. — Transition from Popular Moral Philosophy to the Metaphysic of Morality. { Kant's object in this section is to show the necessity of going beyond the ethics which is based simply upon conscience or the ordinary moral judgments of men, and enquiring into the ultimate basis of all morality, an enquiry which leads us beyond experience to the basis of morality in pure reason, f 12 a The conception of duty, as we have seen, is employed in the ordinary judgments of the moral consciousness. I± would be a mistake, however, to suppose that it is based upon experience. So far from this being true, we cannot I point to a single instance in which we can affirm with absolute certainty that an action has been done purely from duty, though no doubt it is easy enough to find examples ' of actions which are not opposed to duty, as in the case of self-preservation previously referred to. Any attempt in', fact to base duty upon our experience of what men have! done must end in failure, and inevitably lead to doubt of the whole conception. We must therefore recognise that \ the sole basis of duty is reason, which tells us what ought ' to be done without paying the least regard to what has j been done. Absolutely disinterested friendship is a duty, although there may never have been an actual instance ' of it. 13 a Nor is the moral law binding only upon man : it is ' equally obligatory upon all_rational_beings, and upon these * absolutely and necessarily under all conditions and without reservation. If this is not admitted, morality cannot be objective, but can only be a statement of what should be 316 THE METAPHYSIC OF MORALITY done under special conditions. Obviously a law of this apodictic character, a law which prescribes what should be done universally and necessarily, cannot be derived from experience ; for no generalisation from experience can take us beyond generality ; it may tell us what has been done in many cases, but it can never tell us what ought to be done in all possible cases. H. 256 " It would in fact be very unfortunate if morality could be derived from examples. If a moral example is set before me, I must first bring it to the test of principles of morality before I can determine whether it is worthy to serve as an original example, i.e., as a pattern, and it can in no sense furnish me with an authoritative conception of morality. Even the Holy One of the Gospels must first be compared with our ideal of moral perfection before we can recognise Him as a pattern ; and so He says of Himself, ' Why call ye Me (whom ye see) good ; there is none good H. 257 (the archetype of the good) but God only (whom ye do not see).' Whence, then, do we derive the conception of God .as the supreme good? 1^ Entirely from the idea of moral perfection, which reason frames a priori, and connects I inseparably with the conception of a free will. In moral I action imitation has no place : examples merely serve to stimulate our moral effort by showing beyond doubt that what the law commands can be realised ; they supply us with concrete instances of what the practical rule expresses in a more general form, but they can never justify us in setting aside the true original, which lies in reason, and guiding ourselves by examples." 233 6 The principles of morality must therefore be entirely independent of experience, and must derive their authority solely from reason. If there is any real philosophy of morality at all, we may assume that moral conceptions, and the principles based upon them, are absolutely a priori. 233 1 This philosophy or metaphysic of morality must studiously I exclude all empirical elements, and cannot be based even MORALITY IN PURE REASON 317 upon theological conceptions. And not only is it the indispensable basis for a sound theory^ of duty, but it is also of great value in the actual conduct of life ; for the idea of duty, when it is grasped by reason in its purity, has so powerful an influence on the heart of man, that it is able to master the strongest natural desire. A a From what has been said we learn three things : firstly , Ij that all moral concgpti ons proceed from reason entirely i a priori; secondly, that it is of supreme importance . to set 3 the conception s and laws of~inoraIity~before the mind in I, their purity, and thatlbot merely in the interest of a true i* theory of morality, but also as an a id in practical life; j[ and, thirdly 7 that "The" principles of pure practical reason, unlTKe" those of "theoreticaneason, do_not_de£end in any 5 seiise"upon the peculiar character of man, but are derived I from tEe^ery conception of a rational being, and therefore ^ applyTolill'Tattonar'Bemgs. ,■46 ^"Popular philosophy makes an appeal to the common li judgments of men in support of its contention that certain J acts are just, honest, etc. Since, however, no philosophy Ij can rest simply upon an appeal to experience, it is necessaryl j\ to seek for the ultimate basis of all moral judgments.! , To show that a metaphysic of morality is necessary, we , must start at the point where practical reason prescribes general rules of action, and follow the steps by which it reaches the conception of duty. |4c There is a fundamental distinction between a being, the I i operations of which simply conform to law, and a being \ I that consciously conforms to law. The former is deter- [ i mined externally, or in the way of mechanical causation,; ' while the latter may be determined by reason. Tor only i a rational being, a being that has the consciousness of law, I can act from principles ; or, what is the same thing, only a ^ rational being has a will. As reason is essential to the I deduction~oracts^ from kws, will is the same thing as practical reason: it_is reason expressed inaction. Now, 318 THE METAPHYSIC OF MORALITY will or practical reason may exist either in a being whose will is infallibly determined by reason, i.e., a being who desires only that which reason prescribes as absolutely good, or in a being whose desires are not invariably in harmony with reason. As a matter of fact man is of the latter character. In his case acts which he recognises as objectively necessary are subjectively accidental ; i.e., though his reason recognises that he ought to do them, he yet may act, not from reason, but from sensuous desire? Hence the I determination of his will by objective laws presents itself ( to him as an obligation to obey them, not as the spon- 235 *, taneous expression of his will. Eeason, therefore, commands 'obedience to its objective principle, and the formulation of 235 6 its command is an imperative. The term " ought," as expressive of an imperative, indicates that the will is not necessarily in conformity with the objective law of reason. Now, that is practically good, which determines the will by objective principles, i.e., principles which apply to every rational being. There is a fundamental distinction between ggod and pleasure ; for pleasure as a motive is due to subjective causes, which vary with the sensibility of the individual, while a principle of reason is valid for all. 236 a A perfectly good will agrees with the rational will of man in conforming to objective laws, or laws of the good, but it differs in not being under an obligation to conform to them. An imperative has no meaning as applied to the divine will or any other holy will, such a will being by its very nature in harmony with the law of ^ reason. Imperatives are therefore limited to beings whose )' will is imperfect, such as the will of man, expressing as they do the relation of objective law to an imperfect will. 236 6 All imperatives are either hypothetical or categorical. In the case of a hypothetical imperative, such as that which arises in the pursuit of happiness or wealth, if we will the end we must also will the means by which the end can be obtained ; while the categorical imperative commands us MORALITY IN PURE REASON 319 to will an act for itself, and not merely as a means to an end. Every practical law affirms that a certain action is good, and therefore ought to be done by a rational being. If the act is good merely as a means to something else the imperative is hypothetical. If the act is good in itself : the imperative is categorical. Thus an imperative supplies the practical rule for a will that may not will the good, either from ignorance or through the influence of maxims opposed to the objective principles of a practical reason ; in other words, maxims which are based upon sensuous desire. Hypothetical imperatives may be further distinguished as problematic and assertoric ; the former telUng us what actions are good relatively to a certain possible end, such as wealth, the latter what is good relatively to an actual end, viz., happiness. In contrast to these a categorical imperative, which commands an act withoiit limitation, is an apodictic or abso lutely neces sary^ practical principle. Problematic principles are infinite in number, since they include every act that is capable of being willed by a rational being with a view to some particular end. There is, however, one hypothetical principle, namely the asser- toric, the principle of happiness, which all rational finite beings as sensitive do as a matter of fact seek to realise, since nature makes them desire pleasure and dislike pain. This is a hypothetical imperative, for the means is distinct from the end and is not willed for itself It may be called the maxim of prudence, because it involves nothing more than skill in the choice of means to an end ; for, though it is always willed by us as sensitive beings, it is not imposed upon us by reason as a necessary law of action. Setting aside hypothetical imperatives there remains, as the only categorical imperative, the direct command to do an act as an end in itself, not as a means to an end. This imperative must spring directly out of reason, and must directly connect the act with the con- ception of the will of the rational being as such. It has 320 THE METAPHYSIC OF MORALITY nothing to do with the special kind of act, or with the consequences expected to result from it in the way of pleasure, but solely with the principle or motive of the agent, and may properly be called the imperative of •morality. 238 6 There is an analogy between the problem of knowledge and the problem of morality. The question in which the former was summed up was, as we have seen : How are a priori synthetic judgments of knowledge possible ? Simi- larly, the main problem of ethics is : How are a priori synthetic judgments of morality possible ? Now, when we consider the three kinds of imperatives, which have just been distinguished, we find that there is no difficulty in giving an answer to the question : How are problematic imperatives possible ? for in this case the imperative merely tells us that if a particular end is sought, we must employ the means for the attainment of that end. This imperative of skill, as it may also be called, is therefore an analytic] judgment. All that the imperative implies is, that in the idea of willing the end there is directly implied the willing of the acts necessary to the attainment of the end. No doubt certain synthetic judgments of knowledge are pre- supposed in the willing of the means to the end, but these are not synthetic judgments of morality, having nothing to do with the principle of the will. 239 a It is not quite the same in the case of the assertoric imperative. Every one desires happiness, but no one can give a precise statement of what he means by it. If we could tell precisely wherein happiness consists, the imperative of prudence would obviously be analytic; and we may therefore say, notwithstanding its difference from, the problematic imperative, that like it the imperative ol prudence commands that he who wills the end should will the means. No deduction or justification of either of these imperatives is therefore required, both being analytic. 240 a Tjie real dif&culty js^to justify^ the categoric al imperative MORALITY IN PURE REASON 321 ■ We cannot say that here the end is presupposed, and that the only question is with regard to the means. Nor can ^; we base such an imperative upon experience, for no number of instances can ever establish a universal and necessary proposition. Experience may tell us what has been done, but it can never tell us what ought to be done. We musti therefore be able to show that the categorical imperative ■ cannot be reduced to hypothetical imperatives. ' -Ob It is quite evident, to begin with, that only a categorical imperative can have the dignity of a practical law, that is, of a principle which is binding upon all men in all possible ;■ circumstances. The two other imperatives presuppose a certain end which is not absolutely necessary ; and there- ^ fore they must be regarded as contingent, since the end ' may be denied ; but, in the case of the categorical i imperative, the end cannot possibly be denied without - denial of the absolute obligation of the law. ! Oc But, how are we to justify a categorical imperative, or unconditioned law of morality ? Here we have what claims ^ to be an a priori synthetic judgment of morality, a judgment therefore which cannot be derived from experience ; and we F can hardly expect that the deduction, or justification, of such 1= a judgment will be less difficult than the corresponding deduction of the a priori synthetic judgments of knowledge. 5 c? Let us begin with an analysis of the conception of a t categorical imperative, in order to see if we cannot extract - from it the formula of morality. This enquiry will correspond to what Kant in the Critique of Pure Reason 6 called the metaphysical deduction of the categories. W^ '■ assume in the meantime that there is a categorical impera-y '■■ tive, and we simply ask what must be its character. This is a comparatively simple question, the only difficult ques- '- tion being how we can establish the objective validity of ■■ the asserted categorical imperative. la From the conception of a hypothetical imperative :• nothing can be derived until we know the conditions X 322 THE METAPHYSIC OF MORALITY under which it applies. It is different in the case (rf a categorical imperative ; for, froni the mere conception -of such an imperative we can tell at once what it must imply. This imperative affirms that the subjective principle by t which the will of the agent is to be determined must conform to the ob|ective^j)rinciple of the law, i.e., that the motive and the content of the law must coincide. The law applies universally, or independently of all special ■^ conditions, and hence the maxim on which the agent acts must be solely and entirely this, that he ought to obey the 241 h law under all possible conditions. There is, therefore, but one categorical imperative, which may be thus stated : "Act in conformity with that maxim, and that maxim only, which you can at the same time will to be a universal law." 241c If all imperatives of duty proceed from this single imperative, we can indicate what we mean by the cate- gorical imperative, although we may not be able to show \ 241 d that from it specific duties can be derived. When we speak of a law of nature, we mean a law which applies universally to all events. In the present connexion the law is that of mechanical causation. If there are laws of duty, they must agree with laws of nature in being universal ; in other words, though they proceed from the will of a rational being, they will yet outwardly resemble fixed laws of nature. The universal imperative of duty may therefore be stated in this form : " Act as if the maxim from which you act were to become through your will a universal law of 242 a nature." If we examine our consciousness in cases where we act in violation of duty, we -shall find that we do not really will that the maxim from which we act should become a universal law : what we really will is that the law should remain in force, but that an exception should be made in our favour, or under what we try to think are very exceptional circumstances. Hence, if we looked at every- thing from the same point of view, the point of view of MORALITY IN PURE REASON 323 reason, we should see that there was really a contra- diction in our will, the contradiction that a principle is recognised to be objectively necessary, and yet to be subjectively contingent or to admit of exceptions. Instead of doing so, we admit the validity of the categorical imperative, but we try to avoid the contradiction by making / our own case an exception to the universal law. >ii~_ The reasoning by which Kant reaches his first formula of morality is substantially as follows. Man as a rational] being is not subject to a law which determines his action! in a purely mechanical way ; on the contrary, his action! presupposes the consciousness of a law which he may or) may not obey. This law presents itself to him as ad imperative, because he finds in himself certain naturaA desires, the claims of which to determine his action reason \ refuses to acknowledge. Thus the pure consciousness of j seK is bound up with the willing of a universal law. j Hence the formula : " Act as if the maxim from which you ' act were to become through your will a universal law of nature." This does not mean that action is to be determined by a law of nature, but only that a law of nature is employed as the type or analogue of a free act ; in other words, the law must have the universality and)* inviolability characteristic of a law of nature, but not its/ external necessity. ' By the application of this formula, as Kant contends, we can at once determine the moral character of any proposed course of action. Consider (1) duties of perfect obligation to oneself, and take as an illustration the question whether suicide is in any case justifiable. Suppose that the natural desire to maintain one's life has ceased to operate in a man, is he justified in making away with himself ? In order to test the maxim, the individual is asked to consider what will be the result if not merely he but every one acts in the same way. The result must evidently be a contradiction, for " it is impossible to conceive a natural system in which the 324 THE METAPHYSIC OF MORALITY same feeling, the tendency of which is to impel men to the preservation or furtherance of life, should by a universal law of nature lead them to self-destruction." (2) As an instance of a duty of perfect obligation to others, consider the question of the right to borrow without the possibility of repaying the debt. A promise to repay a debt implies belief on the part of the person to whom the promise is made in the good faith of the person who makes the promise. Suppose, then, that we universalise lying promises, and it is evident that there could be no such promises, since no one would believe another. Hence the very conception of a promise is contradictory of the idea that it is not universally binding. (3) As an example of the breach of a duty of imperfect obligation, Kant cites the case of a man who refuses to cultivate a natural talent with which he has been endowed by nature. Here, as he admits, there is nothing unthinkable in the supposition of a whole community in which every one fails to cultivate his peculiar gifts, but he argues that the universalisation of the maxim of idleness contradicts the rational will. Every man in virtue of his reason must admit that what exists for a certain purpose should be applied to that purpose. Natural talents exist for the purpose of serving the com- munity, and therefore a community of rational beings all violating the plain dictates of reason is inconsistent with the formula of morality. (4) Lastly, the maxim of selfish- ness cannot be universalised without contradicting the rational will. No doubt the human race might subsist although every one refused to help his neighbour, but it is not possible to will that such a principle should become a universal law of nature ; for if every one were equally selfish, no one would get help from others, and thus he would defeat his own end. It has been objected to this reasoning of Kant that the validity or invalidity of a maxim cannot be established by asking whether it can or cannot be universalised. There is MORALITY IN PURE REASON 325 nothing self-contradictory in the conception of universal suicide, or universal lying, or indeed of any conception whatever in its abstraction. What underlies Kant's argu- ment is the tacit^ assumption that life and society are good ; whence it no doubt follows that the destruction of life and society is contrary to a rational will. This assumption becomes more explicit in Kant, when he goes on to defend duties of imperfect obligation on the ground, not that the opposite is inconceivable, but that it is incompatible with a rational or impartial will. ":3 a As the result of our investigation so far, then, we have learned, firstly, that the conception of duty must be expressed in the form of a categorical imperative, and we have even discovered, secondly, the formula in which it must be expressed if it is possible at all. What we havej now to ask is whether we can prove a priori that there] actually is an imperative, which is binding upon all rational! beings, and therefore upon ourselves as rational. !'36 It is of the greatest importance to observe that the existence of such a principle cannot be derived from the peculiar constitution of human nature. In the investigation into the conditions of knowledge, Kant was led to maintain that the forms of perception and of thought are peculiar to man, or at least to all finite beings as such : they are not determinations of ultimate reality, but are only forms of our experience. The principle of morality, as he contends, must have absolutely universal validity, and therefore it is ' not enough to show that this principle applies to all men : we must be able to prove that it applies to al l rationa,! beings, and that in the case of beings who possess a sensuous nature it must take the form of an imperative. 3 c Must all rational beings, then, determine the value of their actions by asking whether the maxims of their will : can be universalised ? If there is such a law, it must be ; derivable a priori from the very idea of a rational will, i without any reference to the peculiar constitution of human 326 THE METAPHYSIC OF MORALITY nature. Thus we are forced to go entirely beyond the realm of experience, and to enter the realm of metaphysic or pure Ideas, for the whole question is in regard to the will in itself, so far as it is determined solely by reason. The will must therefore be separated altogether from every empirical element, and determined entirely by reason or a priori. The truth is that there is no will in the proper sense of the term, except for a being capable of determining itself to act from the consciousness of certain laws ; in other words, will belongs only to a rational being. The object or eud which reason originates must be binding upon all beings possessed of reason. We must therefore dis- tinguish between subjective ends, as based upon natural inclination, and objective ends, which spring from motives that hold for all rational beings. Principles of action which jare independent of all subjective ends are formal, i.e., they fare based purely upon the universality of the law. On the I other hand, principles that depend upon subjective ends are 'material, being determined by the special susceptibility of the individual under certain given conditions. From their very nature the latter cannot be universal and necessary 245 a principles, but are merely hypothetical imperatives. But, if there is an absolute end, an end in itself, in this will lie the only possible categorical imperative or practical law. 245;6 / A rational being by its very nature is an end, which 'can never be regarded as a means to some other end. But no object of natural desire can have more than a con- ditional value, for its value depends entirely upon its being desired by this or that individual. Nor can any natural desire itself, or the want that springs from it, have an absolute value ; on the contrary, every rational being must wish to be free from the influence of natural desire. All non-rational beings have only the relative value of means, and are therefore called things, while rational beings are called persons, because they are ends in themselves, and ought never to be treated as means. It is only in persons, h MORALITY IN PURE REASON 327 then, that we find objective ends, or beings whose existence is an end in itself; and indeed apart from them we could not point to anything that had an absolute value ; there- fore, it would be impossible without them to show that there is any supreme practical principle whatever. 46 a Granting the existence of such a principle, it must evidently be an objective principle of the will, or a univer- sal practical law. This principle must be based upon the existence of that which is an end in itself, and therefore that which is necessarily an end for every rational being. Every man, in virtue of his rational nature, presents his own existence to himself as an absolute end. Hence, there is in man (1) a subjective end, consisting in his own existence as an end in itself, and (2) an objective end, consisting in the rational nature common to all men. Thus we get the practical imperative : " Act so as to use humanity, whether in your own person or in the person of another, always as an end, never merely as a means." This second formula may be illustrated by the same examples as were employed in the illustration of the first formula. (1) Suicide is inconsistent with the demand to treat oneself as an end, since it regards one's existence merely as a means for the attainment of an endurable state of feeling. (2) In making deceitful promises, I am using another as a means to my ends, not treating him as an end in himself. (3) To neglect the development of one's natural powers is not to treat oneself as an end, for as a man I am called upon to further whatever belongs to me as man. (4) If I am to treat humanity in the person of others as an end, I am called upon to further their ends as if they were my own. 466 Combining the two first formulas, we obtain, as Kant contends, a third formula. For, firstly, the categorical imperative applies universally to all rational beings, and not merely within the limits of human experience ; and, secondly, it is not only a subjective end, an end which 828 THE METAPHYSIC OF MORALITY as a matter of fact men desire, but an objective end, to which all subjective ends must be brought into subjection, and therefore it must proceed entirely from reason. The objective ground of all laws of action is the rule and form of universality, the subjective ground is the end desired. By our second formula every rational being is an end in himself, or is the subject of all ends. Hence follows a third formula : " Act in conformity with the idea of the will of every rational being as a will which lays down universal laws of action." 247 a A will of this kind cannot obey the law which it originates from interest in it ; i.e., it is not real obedience !to the law to obey it because of the consequences in the way of pleasure that may accrue from obedience. It is a violation of the law, for example, to act honestly from fear of consequences in the way of punishment either through the medium of the state or in a future life. No doubt man has an interest in his own happiness, and this he conceives, and properly conceives, to be connected with obedience to the law, but this interest in the law must not be allowed to determine his will. 247 h We now see why no one hitherto has discovered the true principle of morality. It was clearly enough perceived that man is under obligation to obey the law of duty, but it I was not observed that, as this is a law which he imposes upon ^himself, in obeying it he is but realising his own rational (will, a will which by its very nature lays down universal jflaws. So long as it was supposed that man must simply submit to a law not originated by himself, his action was inevitably referred to interest of some kind as acting externally upon his will. Without discarding this assump- tion it is not possible to show that there is a supreme principle of duty. But, as that principle is really self- imposed, it may very properly be called the principle of the autonomy of the will, and distinguished from all other principles, which are rightly called principles of heteronomy. MORALITY IN PURE REASON 829 The idea of every rational will as prescribing universal laws for itself leads to the cognate idea of a Jdngdom of e nds. By " kingdom " is meant a system or community of rational beings united by common laws. The law of a rational being is to treat himself and other rational beings never merely as a means but always at the same time as an end. This is the basis of that system or community of? rational beings, united under common objective laws, which] we have called a kingdom of ends. This kingdom of ends is no doubt merely an ideal, for it cannot be shown that there actually exists a society in which every one is deter- mined in his actions purely by regard for the moral law ; ^, nevertheless, it is the basis of all morality. A moral action,^ then, is one which -agrees with the system of laws implied' in the kingdom of ends. The principle of the will is that the maxim of every act must be consistent with a universal law. Hence the third formula : " Act so that the will may regard itself as in its maxims laying down universal laws." The reason why this principle presents itself as a duty is that the maxims from which as a matter of fact we act are not necessarily in harmony with it. For man is at once a subject and a sovereign in the kingdom of ends ; a subject, because he must submit to the universal laws binding upon all ; a sovereign, because these laws are imposed upon him by his own reason. As a member in the kingdom of ends, he is, like all the other members of it, under obligation to obey its laws, but the foundation of aU moral value in man, and indeed in every rational being, is the autonomy of the will. We now see that an absolutely good will, as determined by the categorical imperative, cannot be influenced by the desire for objects, but must be determined purely by the form of volition, as based upon the autonomy of the will. The law which every rational being imposes upon himself is to act entirely independently of all subjective interest, or from maxims capable of serving as universal laws. 330 THE METAPHYSIC OF MORALITY 249 c What has been said must not be taken as a proof that such a synthetic a priori practical judgment is possible. It is not the province of a metaphysic of morality to give this proof So far from holding that the problem has been solved, we have not even ventured to affirm that there actually is such a principle. To prove that morality is real, and that practical reason is synthetic, we must enter upon a criticism of the practical reason itself. Meantime we may point out the general character of a Critique of Practical Eeason, leaving the systematic treatment of the problem to be dealt with in a special treatise. 250 Section III. — Transition from the Metaphysic of Morality to the Critique of Practical Reason. The Idea of Freedom as the Key to the Autonomy of the Will. ^Will is that cause which is possible only in living beings who are rational. Freedom is that form of causality in which the subject is not determined by any cause other than himself. Natural necessity is that form of causaUty which is found in non-rational beings, and which consists in determination by causes external to themselves.?. 250 6 The definition of freedom, as a cause that is not deter- mined to activity by any other cause, merely tells us what freedom is not, but it prepares the way for a positive idea of it. Every cause must be subject to law (Gesetz), and as an effect it must be posited (gesetzt) by a cause. As a form of causality, therefore, freedom must not be regarded as inde- ' pendent of all law (gesefzlos), but only as indep,gndent of natural la^ /|A free cause conformsTo^awTEhough it is to a law of its own, and indeed a free will which acts without law is inconceivablcjy Natural law, however, expresses a sort of causality which is heteronomous, for the cause is determined THE CRITIQUE OF PRACTICAL REASON 331 to activity by something other than itself. Freedom, there- fore, must consist in autonomy, i.e., in a will which is a law to itself. But this is the same as saying that the will is to be determined by no other maxim than that the object of which is a universal law. And this is the formula of the categorical imperative and the principle of morality. Hence_a_free will_is a will which confor ms to moral law. If we begin by assuming the existence of freedom, we can pass directly by analysis to the principle of morality ; in other words, a free will we must conceive of as a self^ determined_will, or ajwilLwhichjwills the universaHaw, i.e., as a moral jwilL But we have no right to assume freedom without proof The principle of morality is synthetic, for we cannot conclude, from the bare conception of an absolutely good will, that the maxim by which a good will is determined is objective or universally binding. The transition from the negative idea of freedom, as a kind of causality which is independent of everything else, to the idea of morality can ■ only be made through the positive conception of freedom. In other words, we must prove that morality involves freedom. To make this clear, how- ever, demands some preliminary investigation. Freedom is a Property of all Rational Beings. To show that the will of man is free it is necessary to show that he is free because he is a rational being. The very nature of morality is such that it has no meaning except for rational beings, while, on the other hand, it is an expression of the will of all rational beings. Now, it is obvious that no being can be moral that is not free, and hence we must be able to show that freedom belongs to the will of aU rational beings. It is, however, impossible to establish the existence of freedom on the basis of know- ledge; for the conditions of our knowledge are such that nothing falls within the sphere of knowledge except 332 THE METAPHYSIC OF MORALITY phenomena, and all phenomena are subject to the law of natural causation. But, while there is no possible know- Hedge of a free being, it is obvious that a being who has the f consciousness of freedom, and who can only act under that i consciousness, must be regarded as free so far as his actions )are concerned. Such a ' being, in short, must be capable 'of acting purely from regard for the universal law of morality. A rational being we must conceive as of such a character that he can act purely from reason. It is impossible to conceive of reason as in any way influenced from without, for the subject would in that case not act under the idea of freedom, but under the influence of natural impulse. We must therefore hold that practical reason, as the will of a rational being, must be regarded by itself as free, and hence the idea of freedom in the practical sphere must be ascribed to all rational beings. The Interest connected with Moral Ideas. 253 a It has been shown that the conception of morality is inseparable from the conception of freedom. This does not prove that man actually is free, but only that, so far as man is a rational being, he must be free. We have also shown that every being endowed with reason must act 253 h under the consciousness of freedom. We have further seen that if man is free he must act from the consciousness of a universal moral law. But it may be asked : Why should I as an individual being act from this consciousness when I am not impelled to do so by any interest in the law? What, in other words, constitutes the difference as regards feeling between action that is determined by natural impulse and action that proceeds from regard for the moral law ? The difficulty that Kant feels in this case is that all action which proceeds from a motive seems to him to be action in which the will is determined by something other than itself. This he expresses by saying that the subject has THE CRITIQUE OF PRACTICAL REASON 333 an interest in the object which he is said to will. The moral law, as he has already pointed out, precludes the operation upon the will of anything but the moral law itself; and thus we seem to be left in this position, that we must will the moral law without a motive, or without personal interest in it. But surely there must he some 4 sense in which, though we do not act from interest, we yet 8 take an interest in the moral law. As we shall afterwards see, Kant seeks to solve the problem here raised by drawing a distinction in kind between all sensuous desires as motives and the single motive of rev erenc e for the moral law. The latter, as he contends, is a purely rational feeling or emotion ; it is not so much that which acts upon the will as rather the feeling that arises in the subject when he acts from reverence for the moral law. 53 c It has been shown that only a being who acts from the consciousness of moral law can be free. But it may be objected that no proof has been given that the moral law iSrfis universally obligatory. Nor can it be denied that there is here a sort of circle, from which it is impossible to escape. We cannot be subject to moral law, unless we are free beings in the kingdom of ends, capable of determining ourselves by universal principles ; and we cannot conceive of ourselves as free unless we are under moral law. Freedom and morality therefore imply each other, but for that very reason neither can be reduced to the other. The difficulty which Kant here raises, as we shall afterwards see, is sought to be solved by saying that freedom is essential to the existence of moral law, while it is from our knowledge of moral law that we prove the reality of freedom (see Ex- tracts, p. 268, where Kant says that freedom is the ratio essendi of the moral law, while the moral law is the ratio cognoscendi of freedom).f No one but a free being can originate moral law, and only a moral being can have the consciousness of freedom, j 334 THE METAPHYSIC OF MORALITY Hovj is a categorical imperative possible ? 254 a It is as a member of the intelligible world that a rational being regards himself as a free cause, or as having a will. Nevertheless, he is conscious of himself as belonging to the world of sense, and therefore he regards his actions as having also a phenomenal side, though they proceed from his will. And as the conditions of knowledge preclude him from going beyond phenomena to the free cause lying behind his actions, he is compelled to regard his actions on their phenomenal side as, like other phenomena, determined by phenomenal antecedents or causes ; in other words, by natural desires or inclinations. Were man only a member of the intelligible world, all his actions would harmonise with the autonomy of the will ; were he a purely natural being, they would be determined solely by inclination or desire. In the one case he would act purely from regard for the moral law, in the other case solely from the desire of happiness. It must be observed, however, that, in Kant's own words, " the intelligible world is the condition of the world of sense, and, therefore, of the laws of that world." The will belongs entirely to the intelligible world, and therefore its laws are laws of the intelligible world. Hence, though man is a being with sensuous desires, as an intelligence he is subject only to the laws of the intelligible world. These laws, therefore, come to him as an imperative, commanding that he should act in accordance with universal laws of morality, notwithstanding the obstructions of desire ; and the actions implied by them therefore present themselves as duties or obligations. 255 a y To the question, then: How is a categorical imperative /possible 1 the answer is, that it is made possible through I the idea of freedom, which shows that man is a member l^of the intelligible world. Were man a member only of this intelligible world, his actions would necessarily con- form to the autonomy of the will ; but, as he is also a THE CRITIQUE OF PRACTICAL REASON 335 member of the world of sense, all that we can say is, that his actions ought to conform to the autonomy of the will. The categorical oioght is therefore an a priori synthetic proposition. In other words, the idea of duty cannot be derived from any consideration of the actual operation of the sensuous desires, but we have to go beyond the con- ception of man as the subject of desire, and add to it the conception of man as a free or rational being who imposes upon himself a universal law. The will as rational is thus seen to be the supreme condition, to which the will as sensuous must be absolutely subject. The method of proof here employed is similar to that by which the categories were deduced in the sphere of knowledge. There it was argued that, as experience cannot be explained purely from sensation, it is necessary, in order to account for experience, to pay heed to the categories, or a priori conceptions, which are essential to the constitution of knowable objects. In a similar way we argue here, that the idea of the moral law cannot be derived from the operation of sensuous desire, and therefore that we must presuppose the will as rational in order to account for the objectivity of the moral law. Limits of Practical Philosophy. 3 a From the point of view of knowledge freedom is only an idea of reason, because, though experience is obviously not the whole of reality, we cannot bring a free being within knowledge, since that would mean that we could transcend the necessary limits of knowledge. Thus, though speculative reason sets up the idea of freedom as possible, it cannot establish its reality. Hence there arises a^ dialectic of practical reason, the dialectic that, while reasoni demands freedom, knowledge assures us only of thej necessity of nature. Nevertheless, speculative reason must so far come to the aid of practical reason as to show 336 THE METAPHYSIC OF MORALITY the possibility of freedom ; in other words, to show that it is not a self-contradictory idea ; and it does so by revealing the limits of experience, and thus making it possible for us to maintain that, though we present our- selves to ourselves in the sphere of knowledge as under the law of natural causation, we yet as noumenal subjects may be free. Speculative philosophy in this way prepares the way for practical philosophy. 256 6 The thought of the intelligible world, which is a product of reason, does not imply that we are seeking to transcend the proper limits of knowledge ; for, as we have already seen, reason necessarily sets up the idea of a noumenal or intelligible world, just because the world of experience shows itself to be limited. In so thinking itself, reason in the first instance conceives of itself negatively — i.e., as not belonging to the world of sensible experience — without regarding itself as prescribing laws in determination of the will. But the idea of freedom is not merely negative, for it must be conceived as a kind of causality which, as self- determining, is called will. By will, therefore, is meant the principle of action which is in conformity with the nature of a rational motive ; in other words, a principle which is in harmony with universal law. If reason sought to derive a definite object of the will, or a motive, from the idea of the intelligible world, it would transcend its proper limits, because such an object would necessarily be an object of knowledge, and as we have seen there is no knowledge of the actual operation of a free will, but only of its effects in the world of sensible experience. The idea of an intelligible world is therefore merely a point of view, set up by reason beyond the world of sense, upon which reason feels itself compelled to take its stand, because otherwise it could not be conceived as of itself the source of action. If man were purely a beint' of ■ sensuous desire, he would never have such a conception at all. It is only because he is capable of setting up a THE CRITIQUE OF PRACTICAL REASON 337 purely intelligible world, as contrasted with the actual world of experience, that he can conceive of himself as an intelligence. This conception of an intelligible world in- volves the idea of a complete system of law, which man imposes upon himself as the law that he ought to obey, quite irrespective of what the actions of himself and of other men may have been. This intelligible world, or kingdom of ends, is thus an ideal world, in which all rational beings are conceived as ends in themselves, and therefore as imposing absolute laws upon themselves and ' upon all others. But this ideal does not enable us to enter into the intelligible world by means of theoretical ■ reason ; in other words, we have to think of it from the point of view of a system of legislation, in which all the maxims of the will are in harmony with universal laws. 7 a Keason, then, would entirely transcend its proper limits, if it made any attempt to explain how pure reason can be practical, or, what is the same thing, how freedom is 7 b possible. All explanation involves the application of natural laws of causality, and therefore nothing can be explained except that which is subject to those laws. Freedom can- not be brought under the laws that apply to facts of experience : it is a pure idea, the objective reality of which can never be verified by being exhibited in con- crete sensible form. Its necessity is the necessity ; which compels reason to presuppose its own freedom as practical, i.e., to presuppose a will that is entirely independent of practical desire. But, though we cannot explain freedom, we can defend it from the attack of those who deny its possibility, by pointing out that in that : denial they are assuming a knowledge of the ultimate ; nature of things of which the human mind is incapable. ; It would no doubt be a contradiction to maintain that there is a free subject, whose will has no connexion in ; its effects with the sensible world, and which is there- : fore absolutely opposed to natural law ; but, there is no Y . 338 THE METAPHYSIC OF MORALITY contradiction in maintaining that a free cause is not determined by laws of the sensible world, while yet its effects present themselves in the sensible world. If it is granted that there is nothing contradictory in the conception of such a cause, we can understand how the laws which determine the action of things in themselves are different 258 a from, and yet harmonious with, the laws of phenomena. I While, therefore, we cannot comprehend the unconditioned necessity of the moral imperative, we can comprehend how it comes about that it is incomprehensible. It is incom- prehensible, because the limits of our knowledge prevent us from directly knowing a free cause ; while, on the other hand, we are compelled to presuppose such a cause as the fundamental condition of morality.! { THE CRITIQUE OF PRACTICAL REASON. 261 In the last section of the Metaphysic of Morality Kant has given in short outline the contents of the Critique of Practical Eeason. What he now does is, therefore, not to add absolutely new matter, but to give a complete systematic statement of the Practical Eeason, as the basis of the existence of freedom, immortality and God. BOOK I.— ANALYTIC OF PUEE PEACTIOAL EEASON. Chapter I. — The Principles of Pure Practical Eeason. In the first chapter of the Metaphysic of Morality it is argued that moral action consists in reason acting from a motive supplied by itself, as distinguished from action the motive for which is derived from desire. This doctrine is re-stated in the three " theorems " with which the Critiqioe of Practical Season begins, the only difference in the two statements, apart from the form of expression, being that in the latter all determination by objects is identified with determination by pleasure. 1 . Definition. i61 a Practical principles Kant defines as " propositions which contain a general determination of the will, a determination that has under it various practical rules." These principles again are either maxims or laws ; maxims, when they are 340 CRITIQUE OF PRACTICAL REASON subjective, or hold only under special conditions peculiar to this or that individual ; laws, when they are objective, or apply to every rational being. Remark. 261 6 As an instance of a subjective principle Kant gives the maxim always to avenge an injury, a maxim which no one would hold to be a universal law. The obligation never to make a deceitful promise, on the other hand, is an example of an objective principle or law, since it is obligatory whether it agrees or does not agree with the individual's private ends. As morally necessary, it is a categorical imperative, and therefore a law that is binding upon the will quite irrespective of the consequences anticipated by the agent or actually following from it in the sensible world. 2. Theorem 1. 262 a Practical principles which presuppose an object or matter of desire as a motive of the will are empirical and cannot yield practical laws. By the matter of desire is meant an object the con- ception of which excites in the subject a desire for pleasure. Now, if the desire for this pleasure is the basis upon which the practical rule rests, then, in the first place, it is obvious that the principle of the will must be empirical. It must be empirical, because nothing can be shown to be a priori except that which is a condition of every possible instance that falls under it, whereas the motive of the will in the case supposed depends upon the manner in which the conceived object acts upon the sensibility of the subject. It is obviously impossible to say a priori how a given object will affect the subject, just as it is impossible to anticipate a priori the particular sensations which shall arise on the presentation of a given object of perception. PURE PRACTICAL REASON 341 The motive, then, must in this case be altogether empirical, and as a consequence the principle based upon it. In the second place, a principle of this kind, one which depends upon the peculiar susceptibility of the subject to pleasure or pain, must from the nature of the case be a maxim, not a law. The second Theorem, therefore, directly follows. 3. Theorem 2. All material practical principles are of the same kind, being reducible to self-love or individual happiness. The pleasure that an agent takes in the idea of the existence of a thing depends upon the peculiar sensitive constitution of the agent as related to the object desired. One person is susceptible to the influence of one object, another to the influence of a different object, and such sus- ceptibility is entirely independent of the will of each. Now, pleasure belongs to sense or feeling, not to understanding. The understanding always works through conceptions or ideas. In the sphere of knowledge it constitutes objects of experience through its categories or a priori modes of synthesis, and in the sphere of practice it gives rise to ideas of practical laws. The understanding is thus a faculty which is identical in all men, and indeed in the sphere of practice, it is, in the form of reason, identical in all rational beings. Pleasure, on the other hand, as peculiar to this or that sensitive subject, can have an influence on action, or be practical, only in so far as it acts on the imagination of the subject, leading him to anticipate a certain satisfaction in the realisation of an object. If we generalise the desire for various forms of pleasure, we get the conception of happiness as the unbroken experience of agreeable feeling continuing through the whole of life. The desire for happiness, therefore, proceeds from the principle of self-love. Hence, all material principles which appeal to 342 CRITIQUE OF PRACTICAL REASON the motive of the pleasure or pain anticipated from the realisation of some object are of the same kind, since they are all simply different expressions of the principle of self- love or individual happiness. 263 e Corollary. All material practical rules or maxims assume that the lower faculty of desire determines the will, and unless it can be shown that the will may be determined purely by the form of law, there can be no higher faculty of desire at all. Remark 1 . 264 a The distinction commonly made between lower and higher desires on the ground of a reference to their respective sources cannot be justified. All pleasures are the same in kind, whether they are pleasures of sense or pleasures of intellect. Whatever may be the source from which they spring, or however they may differ in content from one another, all pleasures, in so far as they are motives, agree in kind, and can only be known empirically. Even if the pleasures proceed from the understanding or the reason, as pleasures they differ only in degree from pleasures of sense. Thus, if a man acts from the idea of a moral law, but acts from it only because of the pleasure he expects to receive from so acting or the pain that he thereby wishes to avoid, his act is none the less contrary to the idea of duty. It is, for instance, not a moral act to be just from fear of consequences in the way of punishment, here or hereafter, for the motive in such a case depends upon the particular pleasure, or avoidance of pain, expected H. 24 to accrue from the performance of the act. " Just as it is a matter of indifference to a man who uses gold to pay his debts, whether the gold was dug up in the mountains or washed out of the sand, so long as it has the same value in exchange, so no man who estimates life solely by the PURE PRACTICAL REASON 343 li pleasure it brings asks whether the agreeable feeling is due 1(| to ideas of understanding or to ideas of sense, but only j!' how much pleasure they produce and how long it will last. Only those who would fain strip reason of the power to determine the will without any aid from feeling allow themselves to fall into the obvious inconsistency of first referring all action to one and the same principle of it pleasure, and then speaking as if pleasures were different in 't kind. Thus, it is a matter of experience that we take i pleasure in the mere exercise of our powers, in the con- st sciousness of strength of soul in overcoming the obstacles which oppose our purposes, in the cultivation of our talents, etc. ; and we rightly call these pleasures and enjoyments E more refined, because, being more under our control than ; others they do not pall upon us, but strengthen our capacity J for greater enjoyment, and minister at once to our delight (I and to our culture. But it is absurd to maintain that I they determine the will in a different way from pleasures ; of sense, when in reality they would not be felt as pleasant were there not in us a natural disposition to be pleased in these ways, a disposition which is the primary ,; condition of the satisfaction that we feel. The idea that the more refined enjoyments are, as pleasures, essentially different from coarser gratifications, is on a par with the metaphysic of those untrained speculators who think of matter as reduced to the utmost fineness, and suppose that they are in this way able to conceive of an extended H. 25 substance as a thinking substance. If we accept the doctrine of Epicurus that virtue determines the will merely by the pleasure which it promises, we have no right to blame him for holding that this pleasure is the same in kind with the coarsest gratifications, since there is nothing to show that he regarded the ideas by which this feeling is excited in us as due entirely to the bodily senses. So far as we can learn, Epicurus found the source of many pleasures in the exercise of the higher faculty of knowledge 344 CRITIQUE OF PRACTICAL REASON though this did not prevent him, and could not prevent him, from maintaining that the pleasure which comes from these intellectual ideas, and which alone on his theory can in this case determine the will, is of exactly the same kind as pleasures of sense." Of course understanding and reason may be employed in promoting individual happiness, but this in no way shows that the principle of the will contains any other motive than that of the lower faculty of desire. The conclusion, then, is, that there is either no higher faculty of desire at all, or that pure reason may of itself determine action independently of all feeling of pleasure or pain. H. 26 " Every rational finite being must desire to be happy, and happiness is therefore an inevitable ground of determina- tion of his faculty of desire. For satisfaction with his whole existence is not an original endowment — a state of tranquillity due to the consciousness of independent self- sufficiency — but a problem forced upon him by his finite nature. As a finite being he is the subject of wants, and these are connected with the matter of his desires, that is, with something which is relative to a subjective feeling of pleasure or pain, as indicating what is needed for his complete satisfaction. ''But just because he can discover this material ground only by experience, it is impossible to regard this problem as a law ; for a law, as objective, must contain the same ground of determination of the will in all cases and for all rational beings. It is true, indeed, that the idea of happiness furnishes a kind of unity, under which all the different objects of desire may be brought ; but it is in this point of view a mere general title for all subjective motives of will, and does not yield any principle of determination which could give us the specific direction that we require from a practical principle. It is the particular feelings of pleasure or pain which he experiences that determine each man's idea of happiness, and even if we confine ourselves to the same subject we find that as PURE PRACTICAL REASON 345 his feelings change so also do his wants. A law which, as a law of nature, is subjectively necessary, is therefore objectively a very contingent practical principle, which may or rather must be very different in different subjects, and hence it can never have the force of a practical law. In the desire for happiness it is not the form of law, but solely its matter, which determines the will, that is, whether in following the law I may expect to secure pleasure, and how much. Principles of self-love may no doubt contain universal rules of skill, telling us how to adapt means to ends, but they are in that case purely theoretical principles. On the other hand, no practical precept based upon them can ever be universal, since desire is determined by the feeling of pleasure or pain, which cannot be universally directed to the same objects." Bemark 2. g4j Even, therefore, if we supposed that there was an absolute agreement between all finite rational beings in regard to the pursuit of pleasure and the avoidance of pain, it would still be true that no universal law could jl be based upon such a ground. 'A consensus of opinion in , regard to the objects fitted to bring pleasure is not the same thing as an objective law ; for, under the supposition made, the principle governing the will would still be self- love or the desire for individual happiness. The motive would therefore be subjective and empirical, as distinguished from an objective law based upon a priori grounds. From these considerations follows the truth of the third Theorem. 4. Theorem 3. 65 a If a rational being is to conceive of his maxims as universal laws of action, he must conceive of them as determining the will, not by their matter, but simply by their form. 346 CRITIQUE OF PRACTICAL REASON 265 6 The matter or object of the will either determines the will or it does not. In the former case, the will is evidently subjected to empirical conditions ; i.e., it is deter- mined by a feeling of pleasure or pain. If that is the case, there can be no practical law, but only a maxim. Now, if all matter is removed from the will, nothing is left but the pure form of a universal system of law. Either therefore a rational being cannot conceive of his maxims or subjective principles as laws binding upon all rational beings, or his will must be determined purely by the form of law itself. Bemark. 265 G Men in their ordinary judgments virtually admit that it is the form of a maxim, and not its matter, which alone can serve as a universal law. Suppose, e.g., my maxim is to become as wealthy as possible, and there has been left in my hands property in trust for others, without any document to show that I have received the money ; can I act on the maxim that any one who receives money in trust may use it for his own purposes ? If we suppose every one acting from this maxim, it is obvious that we fall into self-contradiction ; for, in a community in which it was recognised that money left in trust was not to be applied to its proper object, no one would leave money in trust. This instance shows that a practical law must be applicable universally or without any exception, and indeed this is an identical proposition. If my will is to stand under a practical law it must be free from the influence of all natural inclination. Problem I. 266 a Assume that the mere form of universal law is the only form of a maxim that is sufficient to determine the will ; then the problem is to discover the character of a will so determined. PURE PRACTICAL REASON 347 ;66fc The supposition is that the will is determined purely by i the form of the law, in distinction from all objects of ii sensuous desire. Now, this form cannot possibly belong to K the world of sense, and therefore a being who is capable of J being determined by the mere form of law must in idea be i] beyond the world of sense. The idea which is to determine \ the will is essentially different from the principles that ( determine the relation of phenomena to one another in the ^i sphere of nature, for these are all subject to the law of mechanical causation or reciprocal influence. Events in the world of nature, in other words, are all determined by other events, i.e., the connexion is purely of one phenomenon with another. But, if the will is determined by the form of law, which can be apprehended only by reason, it must be independent of the law of mechanical causation. ''Now, the independence of natural law is freedom in the sense of independence of the whole sphere of phenomena. There- fore, a will which is determined purely by the form of the maxim must be a free will. Problem II. ;i67« Assume that there is a free wUl, then the problem is to discover the law which alone is fitted to determine it necessarily. jw h The matter or object of a maxim can be learned only from experience, but a free will must be entirely independent of * experience, i.e., of all sensuous conditions, and yet, as it is a "' will, it must be capable of being determined to action. f; Since this will is free from all empirical or sensuous condi- I tions, the principle by which it is determined can only be the law itself, taken in abstraction from the matter of the ( law. Now, apart from the matter there is nothing but the , form of law in general. Therefore, the form of law in general, as expressed in a maxim, is the only thing capable of determining a free will. 348 CRITIQUE OF PRACTICAL REASON Remark. 267 c Moral law and freedom, as these two problems show, are intimately connected. Passing over the question at present whether they are ultimately identical, let us ask how know- ledge of them' is possible. Do we start from the knowledge of freedom and infer morality, or do we start from morality and infer freedom ? It is evident that we cannot have an original or primary knowledge of freedom, for our first conception of freedom is negative, i.e., it arises in contrast to the consciousness of phenomena. Freedom, therefore, cannot be derived from experience, for experience deals only with phenomena, and with the natural law of causality as their condition. Nature is just the opposite of freedom, and it is in the contrast to the necessity of nature that we first have the idea of freedom. '' The answer to our question, then, is that we are not primarily or originally conscious of freedom but of moral law. j/This law we apprehend in thinking of our maxims purely in their form. Reason prescribes moral law as a principle of action, a principle which is entirely independent of all sensuous conditions, n Thus the consciousness of moral law compels us to presuppose freedom. We may properly enough be said to have a knowledge of moral law, since it is a direct object of reason, whereas we cannot have a knowledge of freedom, because nothing is a possible object of knowledge which falls beyond the sphere of experience. But, while we infer freedom from moral law, it must be observed that it is freedom which is the condition of the moral law ; for were there no freedom, there could be no moral law. We may therefore say that freedom is the ratio essertdi of the moral law, i.e., it is the condition of the actual willing of a universal law, while the moral law is the ratio cognoscendi of freedom, i.e., the basis upon which we justly infer the reality of freedom. No doubt there is nothing self-contradictory in the idea of freedom, but unless we first PURE PRACTICAL REASON 34» had a consciousness of moral law, we should not be justified in affirming its actual existence. Supreme Law of Pure Practical Reason. Ba Act so that the maxims of your will may be in perfect harmony with a universal system of laws. tei The consciousness of this law may be said to be a fact of reason, because it cannot by reason be resolved into ' anything higher than itself Thus, it cannot be derived from the consciousness of freedom, because from the nature of the case we cannot establish the existence of a free subject on the basis of knowledge. It is quite true that if we had such knowledge, we could then derive from it this universal law, because, as has been shown above in Problem II., on the supposition that there is a free will, this will must be determined by the pure form of law. Since, however, we cannot establish the actual existence of a free subject on the basis of knowledge, the only way in which it can be shown to exist is by passing from the conscious- ness of moral law to the belief in freedom. The reason why Kant denies that we can have any knowledge of a free 1- subject is that for him knowledge is identical with experience. The conceptions of understanding or reason taken by themselves are empty, for it is only in relation to sensible experience that knowledge is possible for us at all. If indeed we possessed an intellectual perception, then we can understand how we might have actual knowledge of a • free subject. But our understanding is never perceptive. We have no knowledge, then, of freedom. But the con- sciousness of moral law is in a sense given to us. When we say this, however, it must be observed that it is not " given " in the sense that it is something related to our perception. It is not a " matter of sense," to use the - technical term ; it is " given " only in so far as it is /'originated purely by reason, and therefore in this case, and 350 CRITIQUE OF PRACTICAL REASON in this case alone, reason declares itself to be the source of law (sie volo, sic Jubeo). Corollary. 269 a Pure reason is practical purely of itself, and gives to man a universal law, which is called the moral law. Bemark. 269 b The principle of morality prescribes a universal law, i.e., a law which is binding notwithstanding the solicitations of desire, and which therefore is the supreme ground of the determination of the will. Now, a law which is thus absolutely universal applies by its very nature to all rational beings that have a will. Hence, it is not to be conceived simply as a law for man, but as also a law for all other possible finite beings possessed of reason, if there are such,> and even of an infinite being or supreme intelligence. But, since in the case of finite beings natural desire or inclination is opposed to reason, the law in their case takes the form of an imperative. ISTo finite being can possess a holy will, i.e., a will which by its very nature always wills the good. Such a will can exist only in a supreme intelligence, an intelligence which by its very nature expresses itself in willing the good. A being of this kind does not act contrary to law, i.e., the action is not arbitrary, but the law is itself the expression of the absolute nature of the being. Although holiness is unattainable by any finite being, since that would mean the entire transcendence of natural desire, it yet serves as an ideal which man can set up as the goal of all his efforts, and to which he can indefinitely approximate. Virtue in fact just consists in the assurance that, while our will is not absolutely conformed to this ideal, we are nevertheless making con- tinuous progress towards it: No man can possibly be PURE PRACTICAL REASON 351 I certain that in all his acts he is conforming to the moral law, but he may well have a conviction that he is gradually- approximating to the ideal. 5; 8. Theorem 4. )o Autonomy of will, or determination of will purely by reason, is the sole principle of all moral laws and of the duties that are in conformity with them. Heteronomy of will cannot be the basis of moral obligation, for no • I universally binding law can be derived from perpetually K fluctuating desires. And not only so, but heteronomy of ill will is contradictory of the principle of obligation and of ■- the morality of the will, since the only principle of ,1 obligation is that prescribed by reason. The sole principle ;: of morality, then, is the determination of the will through III the mere universal form of law. '* This independence of all f desire is freedom in the negative sense, while the self- b; legislation of pure practical reason is freedom in the r:: positive sense. Hence the moral law simply expresses the - autonomy of pure practical reason. If desire enters in ft the slightest degree into the will so as to influence it, c; the will is no longer determined purely by practical r; law, but becomes heteronomous. Kant therefore denies •: that morality can be any combination of desire and '•y. reason ; for, as he argues, if the will is influenced by ,- natural desire, it must be dependent upon a law of nature, ,ji and it is the very character of a law of nature to be J opposed to freedom. On the supposition that the will is jji influenced by natural desire, the function of reason will not 51; be to supply the motive to the will, but only to discover, ■f. the means by which the end indicated by desire may be attained. Even if an action coincides with the moral law, ,,; but is not done from the principle of the moral law, it is u( not moral, any more than an action which is directly ,|. contradictory of it. 352 CRITIQUE OF PRACTICAL REASON Bemark. SVl a There is no doubt that, as a matter of fact, every finite being does seek happiness, but we cannot base upon this fact a law declaring that every one ought to seek his own happiness, much less that every one ought to seek the happiness of all men including himself. Kant, in other words, denies the utilitarian principle, that the basis of moral law is to be found in an induction from the facts of experience. John Stuart Mill, e.g., argues that as every one desires his own happiness, we are entitled to say that the happiness of all is the true end of life. Kant objects to this reasoning, firstly, that granting it to be a fact — and he admits that it is a fact — that every one does desire ^ his own happiness, we cannot pass from this proposition to the conclusion that every one ought to desire his own happiness. We must, in other words, distinguish between that which is desired and that which is desirable. And, in 5^ the second place, admitting that every one desires his own happiness, we cannot derive from this the law that every one ought to desire the happiness of all. There is no possible way, in short, of establishing the law, that every man ought to seek to promote the happiness of all, except on the ground that a maxim must be universalised before it can be the basis of moral action. In other words, a maxim of self-love cannot attain to the dignity of an objective law: unless a maxim has the universality of a law, the idea of moral obligation is absolutely inexplicable. 1. Deduction of the Principles of Pure Practical Reason. 272 a The result of the enquiry presented in the Analytic of Practical Eeason is to show, firstly, that pure reason does supply a principle by which the will may be determined. And this principle is directly known, i.e., we have a direct PURE PRACTICAL REASON 353 consciousness of moral law through pure reason. In the second place, as we have seen, the fact of this moral law , being presented to us compels us to infer the existence of i, freedom; for a rational being is conscious that in his will I he does not belong to the sphere of nature, but is raised in j idea above it, although he also knows that in so far as , he belongs to the world of sense his will must come under the natural law of causality, which applies to all i phenomena. ,72 5 There is a fundamental difference between the Analytic . ,of Practical Reason and the Analytic of Speculative , Reason. In the latter we do not start from fundamental principles, and proceed to derive from these certain laws, but we start with the data of sensible perception, as ordered by the forms of space and time, and proceed to show that under these conditions certain laws are applicable to all objects of experience. The consequence is that we are , compelled to deny a knowledge of all objects lying beyond , the boundaries of experience. It is true that speculative I reason cannot be contented with the objects presented in , experience as ultimate ; for it shows that the mind cannot . be satisfied with less than a complete or unconditioned .[ unity, and such a unity it is impossible to attain, just ,: because knowledge is conditioned by the forms of perception and understanding. What speculative reason is able to show, however, is that there is nothing inconsistent with the nature of our knowledge in the supposition that there is a free cause, i.e., a cause which is independent of sensible conditions. But, though it thus leaves the way open for the practical reason to establish the existence of objects corresponding to its ideas, it in no way extends our know- ledge beyond the limits of experience i13c Nor does even practical reason enable us to come directly in contact with a free cause, for that would involve the extension of knowledge beyond the sphere of experience. What it does is to supply us with an indubitable fact, viz.. 354 CRITIQUE OF PRACTICAL REASON the fact that we are conscious of an absolute moral law, which presents itself to us as binding upon ail rational beings. From this fact we cannot fail to see that the idea of moral law is unintelligible unless we presuppose freedom. Now, this fact indicates that beyond the world of experience there is an intelligible world, and that the law of this world must be that which is expressed in the moral law. ''For 273 i> the moral law applies to the world of sense, or rather to beings that are rational and yet have a sensuous side of their nature, and supplies the form of an intelligible world, without in any way interfering with the fixed conditions of the sensible world. Nature, in the most general sense, means the existence of things under laws. So far, therefore, as there exist beings with a sensuous nature, these, in that aspect of their existence, are subject to the laws of nature. But, as contrasted with the pure legislation of reason, which remains in absolute conformity to itself, a being determined by sensuous desires must be described as subject to a heteronomy of the will ; i.e., the will is not self- determined, but is determined through the influence of conditions independent of it. In contrast to this stands the autonomy of pure reason. Now, laws which are presupposed as the condition of something coming into being are practical laws. The supersensible nature of rational beings is therefore just their nature as under tlie autonomy of pure practical reason. But the law of this autonomy is the moral law, which is therefore the fundamental law of the intelligible world, or of man as a purely rational being. This intelligible world, though it exists only in idea, is yet the conception of a world that ought to be realised, and ought to be realised in the world of sense without interfering with the laws of that world. 'The intelligible world might, therefore, be called the archetypal world, since it supplies the ideal or pattern after which all rational beings should order their conduct ; while the world of sense, in so far as it is the result of the PURE PRACTICAL REASON 355 action of free beings who conform to the law of the "^ intelligible world, may be called the ectypal world. The ^ moral law commands absolutely, and therefore claims that * it ought to be realised. It is true that it is not always '^" possible for a finite being to overcome the obstacles in ' nature which prevent its realisation, but a rational being must always act from the idea of the moral law, and in so '' far as he does so, his act is in conformity with the intelligible world, even if the consequences of his action are *^ not such as he expected would follow from it. ^■74(1 In point of fact we do not find that man is by nature " determined by maxims, which could be taken as a system '■ of universal laws, or which are even in harmony with such ■^ a system. On the contrary, he naturally seeks the '''■ satisfaction of particular or private inclinations, which are '- contrary to the laws of reason. While the maxims ■^ based upon desire are subject to law, the law to which P they are subject is not moral but physical or pathological ; * for so far as the natural desires are concerned, man must " be viewed, like other beings, as standing under the ^ conditions of natural or mechanical causation. But, on the i^i other hand, reason gives us the consciousness of a law to "^ which all our maxims ought to conform, precisely as if - these maxims had been imposed upon us by nature. ''Thus ^ we get the idea of a system of nature different from that ^ which we have learned to know through experience, a J' system in which aU men at all times act purely from the ' motive of the moral law. We may therefore distinguish 1 175 a between the laws of a nature to which the will is subject, ''■ and the laws of a nature which is subject to the will. In f" the former case the will is determined to act by the I influence upon it of the natural desire for an object that ' is conceived to be fitted to bring pleasure. In the latter ' case, the object is originated by the will, or, what is the same thing, the will is determined purely by an ' idea of reason ; and when this is the case, reason 356 CRITIQUE OF PRACTICAL REASON actually is practical, i.e., it determines the character of the action. 275 6 What has been said constitutes the metaphysical exposition of the supreme principle of practical reason, i.e., it sets forth the character which must belong to such a principle, on the supposition that it is absolute. As it is an a priori synthetic proposition, the deduction or justification of this principle — in other words, the proof that it actually is objective or universally valid — we must expect to find even more difficult than the deduction of the principles of pure theoretical understanding. The deduction of the latter, Kant means, is comparatively easy, because without the principles of understanding we can have no system of experience whatever. It is different, as he implies, in the case of the supreme principle of practical reason ; for here we cannot appeal to experience at all, inasmuch as the principle, sup- posing it to be valid, applies not merely within the limits of experience, but with absolute universality, or to all possible rational beings. How can we establish the existence of a law, which, supposing it to exist, must proceed purely from an idea of reason ? Is there not here even a greater difficulty than in the case of theoretical reason, viz., that while we can think an absolute moral law and a free subject, we cannot take this capacity of thinking as a warrant for the existence of an object corresponding to our thought ? 275 c It is not possible to establish the objective reality of a moral law by any appeal to theoretical reason, nor can it be based upon experience. But, while a deduction of the principle of morality is impossible in either of these ways, we find that it can be deduced by a method that is peculiar to pure practical reason. The moral law forms the basis upon which we must rest the reality of a free subject. It does not itself require any deduction, but inasmuch as we see that it could not PURE PRACTICAL REASON 357 exist unless it were originated by a free subject, we can reason back from the fact of the moral law to the existence ' of such a subject. This, then, is the basis of a super- sensible system of nature. Just as the law of natural '' causation is a condition of there being a sensible system of' nature, so the moral law is the condition of there being a ' supersensible system of nature. Thus practical reason is ' able to do what speculative philosophy failed to do, viz., ' '■ to give objective reality to the conception of a free cause. ■■76 a The moral law is established even to the satisfaction of - speculative reason in this way, that, being itself an idea " without which moral action is impossible, it presupposes a 'i free cause, which is not only unaffected by sensuous '' desire, but is directly determined by reason. " In this way ' the moral law is proved to be objective, although it does ■ not fall within the sphere of theoretical reason ; for, while ^ nothing can be objectively known except phenomena and '= the laws of phenomena, it is easily shown that the moral law is not a law of phenomena but of a noumenal or ■ free cause. Since the Critique of Practical Reason proves ■■ that man may be a free cause, it follows that, although on ' the phenomenal side his actions belong to the realm of experience and are under the law of natural causation, yet on the other side they may have an influence upon the phenomenal world, and thus have an immanent use. !?66 tcin the world of phenomena there is no cause which is not itself an effect, and therefore we cannot find in experience any unconditioned cause. At the same time as reason can never be satisfied with anything short of the unconditioned, it sets up the idea of a causality that is self-determined. Now, the possibility of a self-determined or free cause is guaranteed by the character of theoretical reason. Just because there is no completeness of causation to be found in the world of phenomena, reason becomes aware of the limited and conditional character of experience. The possibility 368 CRITIQUE OF PRACTICAL REASON of freedom does not require us to have recourse to a postulate of practical reason, i.e., we do not require to show that there may be a free cause, for the possibility of such a cause is sufficiently guaranteed by speculative reason. But, since we cannot find in experience any instance of an action which is the effect of an unconditioned or free cause, the only way in which speculative reason can defend the idea of a free cause from attack is by showing that a being who on the one side belongs to the world of sense may yet on the other side be a free or noumenal cause. Having thus shown that freedom is not self-contradictory, i.e., that reason may quite consistently maintain both free and natural causation in regard to the same act, the way is prepared for the next step, which can only be taken by practical reason.// So far as theoretical reason is concerned, the idea of freedom is not constitutive, but only regulative ; i.e., it is an idea which enables us to think of our actions as if they were freely determined, even though we have no actual knowledge of them as free. But it is only by the practical reason that the actual existence of a free cause can be established. Thus the empty place which was left open by speculative reason is now filled up by a law applicable in an intelligible world to a free subject. Speculative reason does not in this way gain any additional knowledge, but it acquires certainty in regard to the existence of a free cause. We cannot even say that, when we have established the objective reality of a free cause, we have extended the conception of causality itself beyond the world of experience ; for it still remains true that that conception has no meaning except in its application to phenomena. If it could be shown that the logical con- ception of reason and consequent is determinable, not only in relation to events in time, but also in relation to 3 noumenal or free cause, then we should actually have an extension of the law of causality. But without having a perceptive reason such an extension is impossible and PURE PRACTICAL REASON ,359 therefore practical reason does not extend the conception of cause. Nor indeed has it any motive for trying to extend it ; for all that we require in order to establish the objective validity of the moral law is the certainty that we are capable of determining our actions by it, in other words, the certainty that we are free causes. 2. Extension of Practical as compared with Speculative Reason. a Besides its relation to sensible objects the intelligence has also a relation to desire, and in this connection we speak, not of intellect or understanding, but of will. And as reason in relation to desire may issue in action purely through the idea of law, it is rightly called in this case pure wUl. The objective reality of a pure or moral will, or what is the same thing of a pure practical reason, may fairly be said to be given to us as a fact of our moral consciousness : meaning by this, not that it is a fact of sensible experience, but that it is a fact actually present in our higher consciousness. The con- ception of a will involves the idea of the will as a cause, and therefore the conception of a pure will involves the idea of a free cause, i.e., of a cause that is not subject to the laws of nature, and therefore cannot be shown to exist by any reference to sensible experience. There is no possible way of justifying it except a priori, or through a pure practical law. Now, the idea of a being who has free-will is that of a noumenal or non-phenomenal cause. That there is nothing self -contradictory in the idea of such a cause is evident if we only observe that the conception of cause proceeds from the pure understanding, and indeed that we are able to justify its objective validity only on the ground that it is independent of all sensuous conditions. In its origin there is nothing in the idea of cause itself to limit it to phenomena ; in other words, there 360 CRITIQUE OF PRACTICAL REASON is nothing to prevent us from supposing that the pure conception of cause is realised in a being that does not appear within the circle of experience. But, because we have no object to which we can apply the idea of cause except that which presents itself within experience, it is not possible for us to claim knowledge of a free cause, and therefore for theoretical reason it remains an empty idea. We do not, however, require to hold that our knowledge can be extended beyond phenomena, in order to ensure the existence of a pure will or free cause. Since the conception of cause, as we have seen, does not arise from experience, there is nothing to hinder us from maintaining that there actually exists a free cause, especially as we see that except on that 280 a supposition there could be no moral law. And as the pure conception of causality has a meaning and reality within the sphere of the supersensible, all the other categories subordinate to it receive from it objective validity, though no doubt only in so far as they are related to the moral law by which the will is determined. 280 6 Chapter II. — The Object of Puke Practical Eeason. The question in regard to the object of pure practical reason is quite different from the question in regard to an object as presented to us within the sphere of knowledge. In the former case we are not asking whether a certain object within the sphere of nature can exist in accordance with the law of natural causation, but whether a certain act ought to be willed, assuming that we have the power to bring into existence the object to which the act is directed. It is this question of moral possibility that we have here to consider, since it is the law of the will, or of 280 6 a free subject, with which we have to deal. Excluding, then, all objects in the sense of phenomena, and dealing purely with objects as they present themselves to the PURE PRACTICAL REASON 361 n practical reason, we find that the onlj objects are good 11 and evil, the former beiQg the necessary object of desire, the tt latter of aversion, and both resting upon a principle of reason, s30c Now, pleasure and pain cannot be originated by pure i! reason, because they depend upon the relation of an imagined fi object to the subject as susceptible of them. Hence any ij attempt to base moral judgments upon feeling must be' i] unsuccessful. All hedonistic theories identify the good H with that which is a means to the pleasant, and evil with 9 that which results in pain to the agent. If we accept this- 1, conception of good and evil, we must deny that there is. !i anything which is good or evil in itself, that which we ti call good being so called merely because it is a means tO' il something else, viz., the experience of pleasure, and evil It being merely that which issues in pain. ii i a The terms " weal and woe " may be employed to express s respectively the feelings of pleasure and pain as excited by ij certain objects. Good and evil, on the other hand, have si strictly speaking no meaning except in relation to the will as determined by reason. " It follows that the will which wills the good is never determined by the idea of a given object as fitted to bring pleasure, but solely by universal law./' Good and evil in this sense have nothing to do with il the state or feeling of the agent, but are affirmed only of his j action. If therefore there is anything absolutely good or evil, it cannot be the object of the action, but is due solely J to the character of the maxim by which the agent is { determined to act. I'lJ Admitting that there is a principle capable in itself of , determining the will, that principle wiU be an a priori law i of action, and pure reason will supply from itself the motive j for determining the act. An action so determined is good in itself, and not merely as a means to the experience of anticipated pleasure. Hence a will, the maxim of which is in harmony with law, is absolutely or in every respect good, and the supreme condition of all good. On the other 362 CRITIQUE OF PRACTICAL REASON hand, when the maxim presupposes an object fitted to bring pleasure and pain, the motive is not in harmony with absohtte law : in other words, it is not a moral motive. If in such a case we speak of actions as good, we do not mean that thev are good absolutely, but only that they are good as means to an end. Such maxims, therefore, can never be laws, but are merely practical precepts. :2S2a Good and e\Tl, in the strict sense, then, have a meaning only in so far as the agent determines himself by an a jiriori law ; in other words, in so far as he is a cause acting through pure reason. The agent's acts are therefore in all cases A'iewed as good or evil solely in so far as they proceed from himself as a free cause. Thus reason shows itself to be practical. y But. although an action receives its moral character from its relation to reason as setting up a law of freedom, and therefore belongs to the intelligible world, vet, on the other hand, it must express itself as an event in the world of sense, and as such it must be subject to the law of natural causation, which appHes to all phenomena. Practical reason can, therefore, determine itself only in relation to phenomena, though not by phenomena ; in other words, the acts of the agent, in so far as he is a rational being, must be in harmony with the laws of nature as constituted by the categories of the ttnderstanding, and especially the category of natural causation. Eeason, therefore, cannot here be employed theoretically : and as the laws of knowledge are determined purely by the under- standing, it can only wiU the miiversal law and thereby subject the various desires to the imity of self-consciousness, as implied in practical reason or piu'e will. TTu Type of Pure Practical Judgment. ■2S2 h Good and evil as objects have no existence except in relation to a rational will. The question however arises, how we are to determine whether a certain action pro- PURE PRACTICAL REASON 363 ceeds from a practical rule of reason, or is determiaed by the influence of an imagined object upon desire. It is the function of practical judgment to answer this question ; i.e., practical judgment must subsume a given action under the universal law, or it must apply to concrete eases of action the form of law prescribed by reason. In the sphere of theoretical reason there is no difficulty in judgment finding the case to which a principle of the understanding applies, for there we are dealing with phenomena. Thus the principle or law of causality, when applied to objects of sensible perception, gives rise to the schema of causality, viz., the invariable suc- cession of one event upon another. In the case of practical reason, however, no schema is available, because practical reason does not operate with phenomena, but with pure ideas. '/The moral law, therefore, cannot be applied by the ' judgment through a schema, and nothing remains but that it should operate through the pure understanding. / Now, understanding works, not with schemata but with laws, and as in the present case the law is a pure a priori law, the juc^ment does not here employ a schema but a type ; in other words, while in the sphere of morality we cannot admit that the subject as free is subjected to an raviol- able law, we are justified m employing the form of an inviolable law of nature as an analogue of the manner in which the free subject determines itself. A moral being; is not subjected to an inviolable law, but he uses it; as a pattern to guide him in determining whether his' 3 a action is moral or not. The rule which judgment, looking to inviolable law as a type, thus obtains is this : " Ask yourself whether a proposed act can be willed by you as if it took place from an inviolable law of nature " ; and, in point of fact, this is the rule by which every one judges '73 whether his action is good or evil. " This comparison of the maxims of our actions with a universal law of nature , is not the motive which is to determine our will to perform 74 them. The law of nature serves as a type for our CRITIQUE OF PRACTICAL REASON judgment upon the maxim according to moral principles. If the maxim is not of such a character that it can stand the test to which it is subjected in giving it the form of a law 283 b of nature, it is morally impossible." We may, therefore, take natural law as a type of the intelligible nature, so long as we are careful to observe that reason must govern itself simply by the form of law and is not subjected to the influence of natural inclination or desire ; for there is this in common with all laws, whether those of nature or of a free agent, that they must be universal. 284 Chapter III. — The Motives of Puke Practical Reason. 284 a The will must be directly determined by the moral law, if an action is to have any moral value. Even when the act is not contrary to the moral law, it may yet be immoral. The introduction of feeling of any kind into the will destroys the moral purity of the act, because in that case the will is not determined solely by the law, and the act, even when it harmonises with the law, is only a legal not a moral act. It is because man may be determined by desire, that we speak of motives in relation to the will. Hence, in the first place, there is no meaning in speaking of the motive by which the divine will is determined, because such a will by its very nature wills the moral law; and, in the second place, the only motive by which the human will, or that of any finite rational being, ought to be determined is the moral law. The objective ground of the will is thus at the same time the only moral motive. Hence an action may agree with the letter of the law and yet violate its spirit. 284 6 From the conditions of our knowledge it is not possible for us to understand how the will can be directly deter- mined by a law that is not of the same nature as the law of phenomena. A will that has a motive seems to be one PURE PRACTICAL REASON 365 that is acted upon by something other than itself ; but in the present case the agent must act purely from himself, and therefore it seems as if he could not legitimately act from any motive. It is impossible for us to explain how we may act purely from the motive of moral law, and yet act freely. But it is not necessary to show how a free agent determines himself by the moral law. The existence of the idea of moral law compels us to postulate freedom, and , therefore it is enough that we see the actual influence upon the mind of the moral law, in so far as that law supplies the motive of action. What Kant here refers to, as he immediately goes on to explain, is the feeling of reverence for the moral law, which he claims to be, not ordinary pathological feeling, but the feeling which arises in a sensitive being in so far as his sensuous nature is subjected ■ to practical reason. ,ic The moral motive, as we have seen, must always be the ; moral law, and nothing but the moral law. The agent acts I freely only in so far as this is his motive, his act being , determiaed purely by moral law without even the co- I operation of natural desire. The will which is free must therefore restrain all natural inclinations that run counter to the moral law, and where they are contradictory of the moral law, it must not only restrain, but entirely suppress them. The influence of the moral law is therefore, in relation to desire, of a negative character, and the motive by which it is determined must present itself to us a priori, i.e., as a law that is contrary to all particular , motives. When the will is determined by natural inclination or sensuous desire, the motive takes the form of a feeling, the possession of which is conceived as fitted to bring pleasure. The moral law, since it refuses to allow the influence of feeling upon the wUl, must itself exercise a certain influence upon the desiring subject. When the , natural inclinations are thwarted, there results a certain feeling, which may be called a feeling of pain, though we 366 CRITIQUE OF PRACTICAL REASON must observe that it is the kind of pain that only a rational being who possesses a sensitive nature can experi- ence. This is the only instance in which we have been able to show a 'priori how reason can have an influence upon feeling. There are two sources of natural inclination, viz., self-love and self-esteem, both of which may be regarded as specific forms of self-regard. Self-love is natural to all rational beings who possess a sensitive nature, for it is natural and reasonable in such beings to desire their own happiness. Pure practical reason does not affirm that self- love is necessarily in conflict with the moral law ; i.e., it does not assert that it is impossible to desire one's own happiness and at the same time to be virtuous ; what it afiirms is that happiness must not be willed, though it may be the consequence of willing the moral law. Pure practical reason, therefoio, merely brings self-love into harmony with the moral law, and it is then called rational self-love. Self-esteem, on the other hand, it completely destroys ; for no man can claim merit for obeying a law which he ought to obey. The moral law, however, though its influence is thus negative, is in itself positive : it is in fact the form of a being who lives in an intelligible world, or is capable of acting freely. Now, so far as the moral law brings self-love into harmony with itself, it produces in the rational subject a feeling of reverence ; and so far as it uproots all self-esteem, it produces a feeling of the highest reverence. This feeling, which is not of empirical origin but is the result of the action of reason upon the desiring agent, is known a priori. Reverence for the moral law has therefore an intellectual origin, and indeed it is the only feeling which can be known absolutely a priori and perceived to be necessary. 286 a The moral law, then, is, in the first place, the form by which pure practical reason determines action ; in the second place, it is the objective condition under which those acts which we call good and evil are determined ; and, in PURE PRACTICAL REASON 367 I the third place, it is the subjective condition or motive of all moral action, inasmuch as it acts upon the subject and ) produces in him a feeling which conduces to the influence of , the law upon his will. It is no doubt true that the feeling of reverence would not arise in us were we not beings capable i of being influenced by sensuous desire, but it by no means i follows that the feeling of reverence is nothing but a sensuous desire. In its origiu it is not pathological but practical, I, being the result of the influence of reason upon desire. (, All sensuous motives are external to the will and act ii externally upon it. The feeliug of reverence, on the ,, other hand, is not an external motive ; it is the moral j; law itself, in so far as it is made the motive of the j will. By setting aside all the claims of self-love that jj are iu conflict with itself, pure practical reason secures ;; absolute authority for the moral law. As reverence for K this law is the result of the action of reason upon a (1, beiag who on one side of his nature is sensuous, we cannot J, say that a Supreme Beiag feels reverence for law, and on i,.j the other hand a purely rational being, even if he were , finite, provided he was not influenced by sensuous desire, iij would not have a feeling of reverence ; in short, this feehng jj is only found where there is a conflict between reason and i,82 desire. " Eeverence is so far from being a feeling of pleasure J that we surrender ourselves to its influence with great J, reluctance. We even try to avoid yielding reverence to the ^ stern majesty of the moral law itself. On the other hand, ^ reverence is so far from being a feeling of pain, that when ., we lay aside our self-conceit and allow it to exercise its In practical influence, we can never be satiated with gazing J upon its solemn splendour; the soul believes itself to be exalted just in the measure in which it recognises the Ik elevation of the holy law above itself and the frailty of its I own nature." J, 83 "Reverence for the moral law is indubitably a moral ;, motive, and the only moral motive ; it is a form of feeling S68 CRITIQUE OF PRACTICAL REASON which is never directed to an object on any other ground. In the judgment of reason the moral law first determines the wUl objectively and directly ; freedom, the causality of which is determinable entirely through the law, just consists in limiting all inclinations to the condition of obedience to the pure law, and upon this obedience all personal esteem is based. This limitation has an effect on the sensibility, and gives rise to a feeling of pain, which can be derived a priori from the moral law. So far its effect is negative : arising as it does from the influence of a pure practical reason, its main office is to prevent the subject from being determined by inclination, and to make him feel that apart from the moral law he has no personal worth whatever. Thus the effect of the law on our feeling is to humiliate us, as we can perceive a priori, though in this feeling we cannot discover the power of the pure practical law as a motive, but only its resistance to sensuous motives. But as this same law is in the idea of pure reason a direct and objective ground of determination of the will, and as this feeling of humility is merely relative to the purity of the law, that which on the sensuous side lowers our claim to moral esteem or humiliates us, on the intellectual side -elevates our moral or practical esteem of the law itself, and intensifies our reverence for it. Hence, when we consider its intellectual origin, this feeling is known a priori as positive. For that which lessens the obstacle to an activity furthers this activity itself. Now, the recognition of the moral law is the consciousness of an activity of practical reason on objective grounds, an activity which is only prevented from manifesting its influence on our actions by subjective or pathological causes. Hence reverence for the moral law, so far as it weakens the opposing influence of the inclinations by humbling our self-esteem, must be regarded as also a positive, though indirect, effect of that law upon feeling, and therefore as a subjective ground of activity or motive for obedience, as well PURE PRACTICAL REASON 369 as the foundation for maxims of life in conformity with it." E81 " Duty and obligation are the only names which properly express our relation to the moral law. No doubt we are legislative members of a realm of morality, which is possible through freedom and is presented to us by our practical reason as an object of reverence, but we are at the same ' time subjects, not sovereigns, in that realm. To overlook our position as creatures, and to exhibit a proud disregard of the authority of the holy law, is to revolt from its spirit, even if we conform to the letter." ' " This view is entirely in harmony with the possibility of such a command as. Love God above all and thy neighbour as ' thyself. For as a command what is required is reverence " for a law which commands love, and does not leave it to ' our arbitrary choice whether we shall make this our principle. ^ Love to God as inclination (pathological love) is impossible, ■" for God is no object of sense; and love to man, though ' possible, cannot be imperative ; for it is impossible to love another merely at command. It is, therefore, practical love -?88 that is meant in that kernel of all laws. To love God is - gladly to obey his commands ; to love our neighbour is gladly '-■• to do all our duties to him. But the law that makes this our '' rule of action cannot be a command to have this temper of " mind in acting, but only to strive after it. A command to ^ do something gladly would be a contradiction, for if we ■; already know what we are bound to do, and are also ' conscious of pleasure in doing it, no command is necessary ; ' and if, on the other hand, we do it without pleasure ' from reverence for the law, a command that makes this '■ reverence the motive of our maxim would act in direct '■ opposition to the disposition commanded. That law of all '* laws, therefore, like all the moral precepts of the Gospel, is ' to be regarded as setting the true moral habit of mind before '■ us as an ideal of perfection which can be attained by no *• created being ; though it is the antitype to which we should i 2a 370 CRITIQUE OF PRACTICAL REASON endeavour to assimilate ourselves iu an uninterrupted but endless progress." 5"89 " The characteristic grade of moral life at which man (and so far as we can see, every rational creature) stands is that of reverence for the moral law. The temper of mind that ought to bind him to obey it is a sense of duty, and not a spontaneous impulse such as might lead one to undertake a task to which he had no call of obligation. The highest moral state in which he can maintain himself is virtue, that is, a goodness which continually maintains itself in effort and conflict ; and not holiiiess, which would involve the attain- ment of perfect purity of mind and will. It is nothing but moral fanaticism and an exaltation of vanity that we are likely to produce, when we urge men to do certaiu acts because they are ' noble,' ' lofty,' and ' magnanimous.' For by such exhortations we set aside the plain motive of duty, that is, of reverence for the law, whose yoke (though in a sense easy, as it is laid on us by reason itself) is one to which we are not merely permitted, but obliged, however unwillingly, to submit ourselves, and in submitting to which, therefore, we have to humble ourselves and give up all claims of merit. The harm of acting on such principles is that it does not satisfy the spirit of the law, which demands an inward temper of obedience, and not a mere outward conformity of action ; and that it substitutes the pathological motive of sympathy or self-love for the moral motive which lies in the law itself. More than this, it gives rise to a windy, extravagant and fantastic habit of mind, in which we Hatter ourselves as though we were in possession of a spontaneous goodness which needs neither spur nor rein, and forget our duty m the vam idea of our merit. It may, indeed, be allowable to speak of actions of others H 90 which have cost great personal sacrifice as noble and grand, though we must so speak only if we have good reason to beheve that such actions have been done entirely from regard to duty and not from mere impulses of the heart. PURE PRACTICAL REASON 371 ! But, if we set up such actions as models for imitation, we must be careful to lay the whole weight on the motive K of reverence for the law, which alone constitutes genuine H moral feeling. For so only will the holy and earnest pre- s scription of duty keep us from deluding ourselves with • 1 pathological impulses, which are at best analogous to '■■: moral principles, and from pluming ourselves on our own s merits." ! " If fanaticism, in the most general sense of the word, is c: the endeavour, made on express priaciple, to transcend the i limits of human reason, moral fanaticism may be defined as E the effort to transcend the limits which pure practical If' reason sets to humanity, when it commands that the I! subjective motives of moral action should be found nowhere i but in the law itself, and that the habit of mind shown in our maxims should be one of pure reverence for the i law." ;!866 Critical JExamination of the Analytic of Pure Reason. The apparent contradiction between the mechanism of nature and freedom of action can only be solved if we bear in muid the results of the Critique of Pure Reason. The law of natural causation has a meaning only in reference to that which presents itself as an event in time, and therefore it stands under the conditions of time. Now, the actions of man have a phenomenal character, that is, they present themselves to us as events in time; consequently no act can be done which is not capable of being traced back to an antecedent. It is quite true that the acts of the agent, in so far as they belong to the past, are no longer within his power, and therefore he is not free in relation to the past; but, in so far as he is conscious of himself as a free being, he sees that all his past actions could have been freely determined. There is nothing in his existence as a free being that we can speak of as acting externally upon 372 CRITIQUE OF PRACTICAL REASON his \?ill, so far as his existence as an intelligence is con- cerned. The changes that take place in the inner sense he ascribes to his own free causality. From this point of ^^ew a rational being can sav with tmth that every wrong act which he has done he coidd have left undone, and this is in no way in conflict with the fact that from the pheno- menal point of \'iew the act takes its place lq the chain of natural causality. The whole series of acts belongs to the phenomenal character which he has made for himself, and which has been determined by bis free action. Hence from the poiat of view of his freedom, or independence of all external influence, he regards himself as responsible for the whole chain of actions which are the outer manifestations of his inner being. A difficulty still remains in regard to the reconciliation of the mechanism of nature and freedom in the case of a being that belongs to the world of sense. How is it possible, it may be said, that a being can be free, if at the same time it is admitted that all reality must proceed from God ? If God is the originator of all things, he must also be the cause of the existence of men in their noumenal as well as their phenomenal character. The actions of man must, therefore, idtimately be referred, not to himself, but to God. The diflicidty may easily be solved, if we remember what is impUed in the distinction between phenomena and noumena, A noumenal being is conditioned by the principle of natural causality, because, though his actions have an influence on events in time and therefore present themselves to him as events, this mode of representation in no way affects his existence as a free cause. The creation of thinking beings must be regarded as the creation of beings who are really free, though they or rather their actions are presented to themselves tinder the form of events in time. "VVe cannot properly speak of the ci-eation of the sensuous form in which objects of experience present PURE PRACTICAL REASON 373 (, themselves. The beings that are created must be beings as 15 they are in themselves, and we have seen reason to hold (I that man in his true nature is free. Since phenomena are 1^ not realities, it would be a contradiction if we maintained Iji that God created phenomena, and at the same time created If, beings who in their real nature are non-phenomenal. Let jj it be admitted, then, that existence in time has a meaning }f only in relation to phenomena. Now, we have seen that ,j freedom is not incompatible with the natural mechanism of n actions regarded as phenomena ; i.e., we have already proved j that man, from the very fact that he has the idea of a t moral law, is free ; and therefore the freedom of man tj cannot be destroyed by the admission that in his existence he is dependent upon a Creator, for it is man as a free 5 being whom we must suppose to be created. There would ; be a contradiction, no doubt, if we held that beings in the , world of sense are real beings ; for in that case the creation ; of man would mean that he was created as subject to the 5 law of natural causation ; but, there is no contradiction in maintaining that his existence proceeds from God, so long as we hold that in his real nature he is non-phenomenal, and therefore free from the bondage of natural mechanism. 289 BOOK II.— DIALECTIC OF PUEE PRACTICAL , REASON. Chaptee I. — General Consideration. }Qa Pure reason is dialectical, or subject to an inevitable illusion, whether it is employed theoretically in the acquisi- tion of knowledge or practically in relation to action. Reason can in no case be satisfied with anything short of the unconditioned; i.e., it demands an absolute totality of conditions as the necessary presupposition of anything that 374 CRITIQUE OF PRACTICAL REASON is presented as conditioned. But, as this totality of con- ditions cannot be found in the sphere of phenomena, reason proceeds to seek for it in the sphere of things in themselves. When reason operates purely in itself, however, it is im- possible in its theoretical use to penetrate to things in themselves, because the conditions of knowledge are such that without perceptions our conceptions or ideas have no objective application. We cannot, therefore, by the use of reason theoretically, find the unconditioned; we can only pass from condition to condition in search of an uncon- ditioned, which from the nature of the case can never be found as an object of knowledge. Eeason, however, is so convinced of the necessity of the unconditioned, that it inevitably falls into the illusion that the unconditioned can be found in the sphere of phenomena; and it proceeds under this natural illusion until it is confronted with an absolute contradiction, and only then does it begin to siispect that the unconditioned cannot be an object of knowledge. This truth is brought to light by the criticism of the whole faculty of reason in its theoretical use. At first sight reason seems to be entirely baffled in its quest for the unconditioned, but it turns out on closer considera- tion that the contradiction, or antinomy, into which it falls, is just the means of bringing to light the higher nature of things. When we discover that we have been treating phenomena as absolute realities, we are led to see that reality as it is in itself must be of a higher character than the objects of sensible experience. And when we have once seen the possibility of the existence in our own case of freedom, as forced upon us by the fact of our consciousness of moral law, we are able to solve the con- tradiction, which for theoretical reason was insoluble. 291 la The Critique of Pure Reason has shown how the natural dialectic of speculative reason may be explained, and how we are thus enabled to guard against the false inferences drawn from the natural illusion to which we are subject. DIALECTIC OF PURE PRACTICAL REASON 375 oj;. But reason is dialectical in its practical as well as in its ij$ theoretical use, and therefore the Critique of Practical ilij Eeason must seek for an explanation of the illusion which j^ arises in this case also. Here reason demands the uncon- .); ditioned for the practically conditioned. Just as in the i,; sphere of knowledge it sought for the unconditioned in is: reference to sensible perceptions, so in the sphere of prac- ;g tical'feason it starts from natural inclinations or desires,; (|i; and demands an unconditioned for them, although the; ^ unconditioned is not in this case to be conceived as, I,, dstefmining the will, but only as the unconditioned totality! ij, of the object of pure practical reason. This object is the jjf, highest good. ,,!906 "Before we proceed to the dialectic of pure practical j^i reason, it is important to observe that the idea of the J, highest good presupposes that the will is determined purely J I by the_form of a universal law, not by any matter. The , ■ Eghest good, as we shall immediately see, has two in- ij gredients : it involves | the conception of the realisation of J, perfect morality or ^^^tue, and} the realisation of complete g' happiness. This is an object which reason demands, but ,p It is not the motive by which the will is to be determined. ;i; The only pure motive is the moral law itself, for if the wOl were determined by an object called the good, it would not :,. be determined purely by the moral law. As we have seen I in the Analytic, it is necessary to morality that determina- "J, tion by the pure form of law should be the only motive. Chaptek II. — The Summtjm Boxum. 291a The Summum bonum, or highest good, contains two ele- ments, which must be carefully distinguished from each other, viz., the supreme (suprcmum) and the complete (con- summaium). The supreme good is that which is absolutely unconditioned, and presupposes nothing higher than itself. It is therefore not subordinate to anything else (originarium). 376 CRITIQUE OF PRACTICAL REASON The complete good is a whole which is not a part of any other larger whole of the same kind {perfectissimum). Now, the condition of virtue, or worthiness to be happy, is the supreme condition of all that we can regard as desirable, and therefore it is the supreme condition of complete happiness. Virtue is thus the supreme good, but it is not the whole or complete good, which finite beings not only seek to obtain, but which impartial reason declares to be a legitimate object of desire. On the supposition that there is a rational being of infinite power, we must suppose that He desires that His creatures should not only be virtuous but happy, provided always that happiness is the result of virtue. The highest good of a possible world must therefore consist in the \mion or harmony of virtue and happiness in the same person, i.e., it must consist in happi- ness in exact proportion to morality By the summum honum or highest good, therefore, is meant the whole or complete good. What has to be especially observed, however, is that virtue, or the supreme good, is the necessary condition of the complete good, because no one has a right to expect happiness unless he is virtuous. Happiness is thus not a good in itself, but only a good under the condition that conduct is in conformity with the moral law. 1. The Antinomy of Practical Beasori. 292 a The summum honum or highest good, then, demands the union of virtue and happiness. Now, the conception of virtue does not necessarily imply the conception of happi- ness, nor does the conception of happiness necessarily imply the conception of virtue ; i.e., we cannot pass from the one to the other by a purely analytical process ; on the contrary, we can perfectly well conceive that virtue may not bring happiness, and, as we have seen, the desire for happiaess, if made a principle of action, is contradictory of virtue. The only way in which virtue and happiness can be THE SUMMUM BONUM 377 '! combined is by a synthetic principle, and indeed a " syntTietic principle which connects the one with the other •'■ through the conception of cause and effect. The whole '■ question is in regard to the good of action, a good that " is possible only through the wiU. Hence we must say, ' either that the desire for happiness supplies the motive for ^:, the maxims of virtue, or that the maxims of virtue are the '* efficient cause of happiness. But the former is absolutely * impossible, for anyone who makes happiness his motive ^' thereby destroys the morality of his action. And the latter ■ is also impossible in another way, for though a man may i will the moral law, it does not foUow that the result of his E action will be to secure happiness. Conformity to the i: moral law may exist without happiness, since happiness F is^dependent upon the whole connexion of things in the must be sought for. "When the law or principle is of the V former character, Kant calls it the determinant judgment, ,, because it starts from the universal and specifies or deter- mines it by reference to particulars of perception ; in the latter case the law has to be discovered, and it is therefore called the reflective judgment. J In the reflective judgment, ■ however, some principle must be employed as the guiding- thread by reference to which judgment operates. This ',: guiding-thread Kant finds in the idea of purpose or end. ,, The determinant judgment, then, is simply the judgment as , employing the principles of understanding already set forth „ in the Critique of Pure Reason, namely, the axioms of per- I ception, the anticipations of observation, the analogies of 5 experience, and the postulates of empirical thought. The , reflective judgment, on the other hand, is judgment as r employed by the special sciences in the discovery of par- j ticular laws of nature. 1 12 a The determinant judgment brings particulars under the universal transcendental laws supplied by the understanding ; in other words, its function is to determine or specify such laws as are expressed in the principles of the understanding. In this case, therefore, the law is not supplied by judgment ' but by understanding. Although in this way we can I explain the universal conditions under which any experience at all is possible, we are stUl, after the application by judgment of the principles of understanding, very far from having a complete knowledge of nature. For, while it is 398 THE CRITIQUE OF JUDGMENT true, e.g., that nature is a system the parts of which are connected by the principle of causation, the particular forms displayed in the connection of definite objects and events are infinitely various. Hence it is necessary to go beyond the general system of nature, conceived as determined by the principles of the understanding, and to seek for the laws of those specific phenomena. These laws must be regarded as contingent, so far as our intel- ligence is concerned ; i.e., it cannot be shown that without them we could have no system of experience whatever. On the other hand, if we are to discover specific laws of nature, there must be some principle which regulates our procedure — a principle which cannot be derived from the observation of particular facts, because its object is to 312 h combine those facts under laws. The principle of a I reflective judgment must therefore be conceived, not as a j unity imposed upon nature by an intelligence distinct from I it, but as a unity which enables us to treat nature as if it- were an organic whole, all the parts of which are deter- 1 mined by an intelligence distinct from ours. We cannot, however, affirm that there actually is an intelligence of this kind ; all that we can say is, that judgment, in seeking for specific laws of nature, must proceed on the principle that nature is an organic whole, the parts of which have been combined by an intelligence different from ours with the 313 a object of making it intelligible to us. Now, if we are to treat nature as having an end, it is obvious that the different parts of nature must be conceived as related in such a way as to be purposive. This, then, is the concep- tion under which the reflective judgment always operates. f It proceeds on the principle that the specific laws of nature ' can be viewed as if they were the expression of an intelli- gence acting purposively. FORM OF NATURE IMPLIES PURPOSE 399 5. The Principle that the Form of Nature implies Purpose is a Transcendental Principle of Judgnunt. \h The transcendental principle of judgment is " one which enables us to think a priori the universal condition without . which things could not be objects of our knowledge at all." ; This is an application of the general idea attached by Kant to the term " transcendental." The transcendental exposi- . tion of space, e.g., set forth the universal conditions under , which mathematical judgments are possible, and the principles of the understanding are an expression of the , universal laws under which experience in general is possible. If, then, there is a transcendental principle of judgment, it will be one which is the necessary a priori condition of the knowledge of objects. From such a principle Kant dis- tinguishes what he here calls a " metaphysical " principle, i.e., a principle which depends for its application on something more than the universal nature of objects of ' experience. It is in fact a principle which is applicable to objects determined in a specific way. The principle, e.g., that all changes in the world of experience must have a cause, is a transcendental principle, because it applies to objects in general ; but the principle that certain changes must have an external cause, is metaphysical, as when we say that a body movable in space must be acted upon by another body external to it. Now, the principle that nature is purposive is not a metaphysical but a transcen- dental principle, because it is not limited to certain specific objects of experience, but applies to every possible object. As such it is a condition of experience in general. The ■■ idea of purpose, then, in so far as it is applied to objects of; , experience, or to nature, is a transcendental principle. On! the other hand, the idea of purpose may be applied in the determination of a free will, and in this case the principle is metaphysical, because the idea of purpose here gets its specific meaning from the sphere in which it is applied. 400 THE CRITIQUE OF JUDGMENT But, while this is true, we must observe that the principle of purpose, no matter how it is employed, is a priori not empirical, since it is a principle which precedes and condi- tions knowledge, not one which is gathered from the observation of particular facts. Kant means, then, that the principle of purpose, as employed in the natural sciences, is brought into operation only in the endeavour to determine the specific objects of nature by reference to universal laws ; while the principle in accordance with which judgment itself operates is not one derived from the particular facts, but one which precedes the discovery of special laws. It is a trans- cendental, but at the same time only a regulative, principle. 314 a That the conception of nature as purposive is a transcen- dental principle, is confirmed by the fact that certain maxims, upon which scientific investigators habitually act, plainly imply that nature is viewed as if it were purposive. Thus it is said that " nature takes the shortest way " {lex parsimoniae), that it " makes no leaps " {lex continui in natura), and that it "has many laws, but few principles" (principia praeter necessitatem non sunt multiplicaiida). These sayings all imply that nature may be viewed as working towards a definite end, and employing the means 314 6 best adapted for securing that end. Such propositions cannot be explained by simply saying that, as a matter of fact, we have found that we do employ them ; for in this way we could only reach at the most the conclusion that they hold good in a limited number of cases, whereas what is characteristic of them is that they apply necessarily and universally. And as the idea of which these are particular applications is the idea of purpose, we must hold that the conception of nature as purposive is a transcendental principle ; i.e., it involves an a priori synthetic judgment, and as such requires a transcendental deduction. 314 c The universal laws or principles of the understanding, as we have seen, are essential to the existence of any experience whatever; for, without them there could be no FORM OF NATURE IMPLIES PURPOSE 401 connected system of phenomena, i.e., no system of nature, These laws rest on the categories, as applied to particulars of sense, through the medium of the a priori conditions of .15a perception. In relation to them judgment is determinant, since the understanding supplies the laws, while judgment merely brings the particulars of sense under them. Thus, e.g., the understanding lays down the law, that every change has , a cause. This law is a universal principle of nature, because without it there would be no system of phenomena, i.e., no nature ia the sense in which we speak of nature from the point of view of its form, — that which Kant elsewhere calls natura formaliter spectata. Transcendental judgment, on the . other hand, receiving from the understanding the universal law or principle, merely presents the a priori condition by means of which the universal law obtains specification. In the case of causality, this a priori condition is the idea of fixed order in the succession of events ; in other words, it is the schema correspondiag to the category of causality. The law of causality is thus known to be an absolutely necessary condition of nature as an object of possible experi- ence. But the objects of our knowledge are determined in many other ways besides this general determination of them; at any rate we may say a priori that they are capable of being determined in many other ways. The extension of knowledge thus involves, not merely the apphcation of the laws of the understanduig to particulars of sense, but the discovery of the specific ways in which objects are connected ia nature, these specific ways consti- tuting the special laws which form the subject matter of the special sciences. As the laws thus discovered must each have its own determinate character, it follows that they cannot be derived from the universal laws of the understanding, and are therefore for us contingent ; in other words, what these laws may be we caimot tell a priori, but must proceed to subject nature to special observation, with the object of finding out how its specific forms are 2c 402 THE CRITIQUE OF JUDGMENT connected. So far, then, as the empirical laws of nature are concerned, those laws that are obtained only from special observation and inference must be regarded as related to our faculty of knowledge in a contingent or accidental way. Nevertheless, we must presuppose that nature may be viewed as if it were a unity, because otherwise our experi- ence would fall into fragments, and we should have no connected system of knowledge. It is of course true that the universal laws of nature enable us to affirm that nature is a system; but as this system is of the most general character, if we are to have a knowledge of nature in its multifarious forms, it is necessary that judgment should bring to its consideration a principle of its own, by the application of which it will be prevented from treating nature as if it were not a connected whole. Now, the presupposition advanced by judgment, that nature may be treated as a unity, though it is for us contingent, is identical with the idea that nature is of such a character that it is not in disharmony with our faculty of knowledge ; in other words, that between nature and our intelligence there is an adaptation or purpose. Hence judgment, in so far as it seeks for the laws under which particular facts of observation must be brought, is reflective, not determinant, and is forced to think of nature as purposive in relation to our knowledge. This is what is expressed in the familiar axioms mentioned above as practically assumed by the scientific discoverer. This conception of purpose is neither a conception of nature Qor a conception of freedom. It is not the former, because if it were, it would determine or constitute the system of nature ; and it is not the latter, because it is not a law or principle of the will, but is concerned only with the extension of knowledge. The idea of purpose, therefore, is simply a maxim, or subjective principle of judgment, which serves as a guide in the extension of knowledge, but which is not constitutive. THE FEELING OF PLEASURE 403 6. The Feeling of Pleasure connected with the Conception that Nature is Purposive. 11a Employing this idea of purpose as a guide, the scientific man discovers that he can in certain cases bring a number of particulars under a law, or a number of particxilar laws under a higher law. The effect of such a discovery is to ezeite in him a feeling of pleasure, a feeling which evidently arises from the conviction that in some sense nature is adapted to our faculty of knowledge. This pleasure is of a very marked character, and persists even after long familiarity with the law thus discovered. 7. 2Tie Aesthetic Consciousness of Purpose in Nature. 317 6 We must carefully distinguish the aesthetic character of an idea from its logical character. The form which is presupposed as the condition of the consciousness of external objects is space. Now, although space is a form of our perceptive faculty, it is also true that it is essential to the constitution of perceptual objects, for objects of external perception must present themselves as extended in space. Sensation, again, whUe it is a subjective element in the perception of external objects, is also essential to the consciousness of such objects, since, when it is determined by the forms of perception and thought, it enters into our consciousness of them, and without it such consciousness would be impossible. But, although sensation is a form of feeling it is not the only form of feeling; for, in connexion with our perception of objects there sometimes arises a peculiar feeling of pleasure or pain, which does not enter into our knowledge of the object, and is therefore not an element or constituent in that knowledge. The question therefore is, how we are to explain this peculiar feeling. Kant's answer is, that it arises from the harmony of the various faculties of knowledge of the subject. "We present 404 THE CRITIQUE OF JUDGMENT before ourselves an object in space by what may be called generally the faculty of imagination, and when this object , is in harmony with our understanding, — when, in other words, we have the presentation of a sensuous whole — then there arises in us a peculiar feehng of pleasure. The object of imagination is then said to be heautiful, the faculty which judges it to be beautiful being called Taste. 318 a There is another form of the aesthetic consciousness, viz., the sublime. The pleasure which arises from the con- templation of an object that is viewed as beautiful, whether it is an object of nature or of art, is due to an adaptation between the form of the object and the reflective judgment. The sublime, on the other hand, arises from a certain adaptation of the subject as a free or moral being to the form, or even formlessness, of the object. In this latter case the subject, though as a sensuous being he is affected by a feeling arising from the consciousness of his own finitude, yet as a free being reacts against that consciousness and asserts his superiority to anything in nature, however vast. Thus arispsthe emotion of the sublime, and hence the Critique of Judgment has two maia divisions, the first dealing with the beautiful, the second with the sublime. 8. Logical Idea of Purpose in Nature. 318 6 So far we have been dealing with the aesthetic judgment, which rests upon a certaia adaptation of the object to the subject, or of the subject to the object. What is characteristic of this form of judgment is that in it there is no explicit conception of the object, inasmuch as the feeling of the beautiful or the sublime arises in the direct contemplation of the object without the intermediation of any conception. There is another form of adaptation, however, viz., that in which the object is regarded as purposive, not simply in its relation to the subject, but in itself; and in this case there must be a conception of the form of the PURPOSE IN NATURE 405 thing as by its very nature purposive. What Kant is here referring to is the adaptation or purpose implied in organic beings, which, as he maintains, are for us inexplicable without the introduction of the idea of purpose as realised in their form. Aesthetic purpose rests upon the pleasure which immediately arises in the contemplation of the form of an object, whereas the logical idea of purpose in nature is possible only through a distinct conception of the character of the object. This latter form of purpose, therefore, is independent of any feeling of pleasure which may arise in the contemplation of the object, and pre- supposes a judgment of understanding; in other words, while it is true that an organised being may give rise to the feeling of beauty, this feeling is quite independent of the consciousness of the object as organised, and therefore con- ceived as purposive. Now, the conception of an object may either precede the actual presentation of the object, as in the case of art, or the object may be presented to us and we may then proceed to judge of it. In this latter case we are compelled to employ the idea of purpose, in order to explain how this peculiar kind of object can exist at all. It does not foUow that the purpose which we thus attribute to the object actually exists in the object, but the only way in which we can make the character of the object intelligible to ourselves is by conceiving of it as if it were purposely formed by nature. When we are dealing with natural objects from the point of view of the aesthetic consciousness, we may think of natural beauty as the presentation of the conception of a formal or subjective purpose, and the natural end which we attribute to nature we may regard as the presentation of the conception of a real or objective purpose. In the one case we think of^ nature as adjusting objects to our faculty of knowledge, in ] - - the other case as constructing objects themselves with a ' definite purpose. The former is the object of aesthetic judgment or taste, the latter is the object of certain logical 406 THE CRITIQUE OF JUDGMENT judgments of the understanding or the reason. The Critique of Judgment has thus two maia divisions, dealing as it does respectively with the Aesthetic Judgment and the Teleological Judgment. 9. Connection of Understanding and Reason through Judgment. 320 a We have seen that understanding prescribes the a priori laws which make a theoretical knowledge of nature as an object of sense possible. Eeason, again, as practical pre- scribes the a priori laws of freedom, and thus gives rise to an unconditionally practical knowledge of moral laws as the laws of a free subject. The realm of nature and the realm of freedom are not in any way connected with one another in their own character, for the conception of nature does not prevent the free realisation of moral law, nor does the idea of the free subject as acting under such law in any way extend the theoretical knowledge of nature. But, while there is thus an absolute separation between the two realms of nature and freedom, so that the sensible can have no possible influence upon the supersensible, there is a certain sense in which we may say that the supersensible must have an influence upon the sensible. As the moral law is unconditionally binding upon man, it must be capable of realisation. It is that which ought to be realised in the sensible world. No doubt we cannot speak of a free subject as a cause, in the sense in which we apply the term cause within the sphere of phenomena; but we are entitled, and indeed compelled, to conceive of a free subject as the ground or ultimate condition of certain effects in the world of nature. That a free cause should express itself by manifesting its influence in the world of phenomena we have seen to be possible, even when we limit ourselves to the theoretical reason ; and when we think of man as a being who is capable of realising the moral law, we must UNDERSTANDING AND REASON 407 conceive of him as a free cause and yet as sensitive. Now, it is judgment which mediates between the two realms of the supersensible and the sensible, supplying us with the conception of purpose in nature, and thus making possible the transition from nature as a system of laws to freedom as capable of realising itself as an ultimate end in nature. SI a In prescribing a priori laws to nature, understanding is dealing only with the phenomenal; in other words, the whole conception of nature as a system under laws of the understanding presupposes a supersensible substrate of nature. Reason as practical, on the other hand, determines this supersensible substrate by the willing of the moral law. Judgment, again, by means of its a priori principle of purpose, enables us to see how understanding on the one hand, and reason on the other, may be connected, although it does not give us any further insight into the ultimate nature of things. (21 b The three higher faculties of the mind are understanding, ' judgment, and reason. Understanding supplies the con- stitutive principles of knowledge; judgment gives us the principles which are relative to the feeling of pleasure or pain; reason has to do with the principles relative to desire. The conception of purpose in nature is merely a regulative principle of knowledge. The aesthetic judgment is con- stitutive in so far as it supplies the constitutive principle relative to the feeling of pleasure or pain. It has also a certain relation to the moral consciousness in this way, that it intensifies the susceptibility of the mind for the moral feeling. Thus the conception of purpose in nature naturally enables us to effect the transition from the conception of nature to the idea of freedom as manifested in its effects. THE CEITIQUB OF AESTHETIC JUDGMENT. BOOK I.— ANALYTIC OF THE BEAUTIFUL. I. The Judgment of Taste as Eegaeds its Quality. 1. The Judgment of Taste is Aesthetic. Asa object is pronounced beautiful or ugly, not because it is comprehended by the understandiug as an actual object of knowledge, but only because it is related to the subject, in whom it excites a feeling of pleasure or pain. In this case, as we are told, the special faculty that operates is the imagination, — the faculty which presents images before the mind, — or perhaps rather the imagination work- ing in union with the understanding. Here in fact Kant suggests that in a certain sense we may hold that there is in us a perceptive intelligence ; for the immediate union of imagination and understanding must mean that the intelli- gence works perceptively or intuitively, as distinguished from the operation of the understanding in so far as it employs universals or conceptions. A judgment of taste is therefore not a judgment of knowledge, i.e., it does not make any affirmation in regard to the character of the object known. The judgment is aesthetic, meaning that the pre- dicate beautiful or tigly expresses only how the subject feels when affected by the idea of the object called beautiful or ugly. This is the only case, as Kant claims, in which a judgment is subjective ; in all other cases judgments rest ANALYTIC OF THE BEAUTIFUL 409 upon something characteristic of the known object. Even sensations, subjective as in a certain sense they are, are yet essential to the constitution of phenomena. It is therefore only in this single case of the aesthetic judgment that we find a relation of ideas that is entirely subjective. Take, e.g., the instance of a building which is pronounced beautiful. So far as the building is an object which is judged to be well adapted for the purpose of human habitation it is looked at from the point of view of the understanding, and therefore the judgment is iu this case logical. On the other hand, when the building is judged to be beautiful, the predicate rests, not upon the idea of the adaptation of the parts to a certain end, but simply and solely upon the feeling of pleasure or satisfaction that arises directly in the contemplation of it. The feeling which thus arises belongs to the subject and indeed to the subject as a sensitive or living being. Now, the faculty which here pronounces the object to be beautiful is quite distinct, for the judgments it makes do not in any way contribute to our knowledge of the object, but rest upon the peculiar character of the relation of our ideas to one another. There are of course cases in which the ideas implied in a judgment rest upon sensible experience, and these ideas we may in a sense call aesthetic, because they belong to the faculty of percep- tion as distinguished from understanding ; but the judgment which is formed by means of them is logical, in so far as judgment is related to the object of knowledge. On the other hand, if the ideas are connected in our mind purely ia a rational way, i.e., if they are free from all empirical' elements and yet are harmonious, then we have an aesthetic judgment. 2. The Satisfaction which is expressed in the Judgment of Taste is disinterested. The satisfaction which arises in all cases of action is quite different from that which is connected with the beautiful. 410 CRITIQUE OF AESTHETIC JUDGMENT The former necessarily implies that the agent is interested in the object which he sets up before his mind as something desirable. When the action springs from desire, then the feeling of pleasure which precedes the act and is set up by the imagination as something desirable is one form of this interest. On the other hand, when the action is determined purely by reason, — when, in other words, it is willed because it is in conformity with the moral law, — the agent still has an interest, though it is different in kind from the interest connected with the satisfaction of immediate desire. But, when we ask whether a thing is beautiful or not, we separate it entirely from any relation to our desire, whether in the form of pleasure anticipated from the satisfaction of natural inclination, or of the peculiar pleasure which springs from reverence for the moral law. Here the pleasure has nothing to do with the satisfaction of the subject, and the object is judged to be beautiful in simple perception or the reflection on it without any regard to his interest. 3. Satisfaction in the pleasant is interested. Pleasure is in all cases relative to the sensitive subject, but it is very important to point out that there is a marked distinction between different forms of sensation, and indeed it only leads to confusion when we speak of all forms of satisfaction as a sensation of pleasure. The fundamental weakness of this mode of speech is that it does not distinguish between the pleasure which arises from the consciousness of an object agreeable to our desires, the pleasure which results from the operation of practical reason as willing the moral law, and the pleasure connected with the contemplation of the beautiful. If we do use the term sensation to designate the feeling of pleasure or pain, we must take care to distinguish these three senses ; not to do so is to confuse the satisfaction which is connected with the knowledge of an object with that form of satisfaction ANALYTIC OF THE BEAUTIFUL 411 which in no way contributes to knowledge, not even to the knowledge which the subject has of himself. It is better, however, to limit the term sensation {Empfindung) to that subjective element which enters into the constitution of objects of perception, reserving the term feeling (G-efuhl) for the expression of what is purely subjective, or does not enter at all as an ingredient into the consciousness of an object. The green colour of a meadow, e.g., must properly be regarded, in so far as it is apprehended by the subject, as sensation, because it enters into the object which is perceived ; but in so far as the perception of the meadow gives rise to a form of sensation which is properly called feeling {Gefiihl), the state of mind is purely subjective. When an object is said to be pleasant, we make the affirm- ation always by reference to our interest in it. This is plain from the fact that the sensation excited in us produces, or may produce, a certain desire for the object. This desire, then, implies that the satisfaction found in the object pre- supposes not only a judgment in regard to it, but also a certain satisfaction in the subject perceiving it. 4. Satisfaction in the Good is interested. Good is in all cases that which is regarded as bringing some kind of satisfaction to a rational being. A thing may be good either because it is a means to something else, i.e., is useful, or because it is good in itself. In the former case the satisfaction arises from the fact that what is called good is a means to pleasure ; in the latter case it is said to be good because it brings satisfaction in itself. In both of these cases the conception of an end is implied, and there- fore the relation of the good to the will ; the good, in other words, always implies some kind of interest. Nothing is regarded as good except that of which we have a definite conception. The beautiful, on the other hand, does not imply such a conception. A flower, a free drawing, ornamental 412 CRITIQUE OF AESTHETIC JUDGMENT foliage, have no meaning in the sense of implying an act of explicit thought, and yet the contemplation of them gives rise to aesthetic pleasure. The satisfaction in the beautiful thus results from the contemplation of an object in the absence of any definite conception, and in this respect it differs both from the good and the pleasant. The pleasant and the good, though they are not identical, agree in this, that they always imply an interest in their object. This is true not only of the pleasant, and of that which is good as a means to something else, but also of that which is good in itself, that is, moral good. The good is an object of will, and therefore it implies a faculty of desire as determined by reason ; but to will anything and to take an interest in it are the same thing. 5. Comparison of the pleasant, the good, and the beautiful. Both the pleasant and the good are related to the faculty of desire, but the former consists simply in the satisfaction produced in the sensitive subject through an external stimulus, while the latter arises from the relation of the subject to an object which is to be brought into existence through the will. In these two cases the existence of the object is essential to the satisfaction, i.e., there is a relation between the subject and the object. The judgment of taste, on the other hand, is purely contemplative. The object is not something to be brought into existence by the subject, but something which the subject sets before his mind, and in which he finds direct satisfaction. The contemplation of the object, again, is not for the purpose of obtaining a know- ledge of it ; for the judgment of taste is not based upon a conception, nor has it a conception as its end ; in other words, it neither operates through the understanding nor through the wUl. It is thus evident that Taste is a judg- ment based upon the satisfaction or dissatisfaction arising from the mere contemplation of an object, quite irrespective ANALYTIC OF THE BEAUTIFUL 413 of any interest in it possessed by the subject. The object of this satisfaction is called beautiful. II. The Judgment of Taste as Eegards its Quantity. 6. The Beautiful is that which is viewed as the object of universal satisfaction apart from conception. v. That the beautiful should be viewed as an object of universal satisfaction without the interposition of concep- tions follows directly from the fact that it is an object of pure or disinterested satisfaction. When we are conscious that the satisfaction which we experience is not at all due to any inclination of our own, but is entirely of an impersonal character, we cannot but judge that the object will give satisfaction to every one. The satisfaction which we experience is in no way determined by what is peculiar to ourselves as individuals ; and therefore, finding that it is of this character, we naturally assume that every one will have the same experience in the presence of the same object. Hence we are apt to speak of beauty as if it were in the object, and to suppose the judgment of taste to be logical. These, however, are natural illusions ; for the object in itself is not beautiful, though we experience a feeling of satisfaction when it excites our faculties to harmonious action; and the judgment of beauty, while it no doubt agrees with the logical judgment in being valid tor all, differs fundamentally in this respect, that it does not rest upon conceptions, but upon the direct perception of the object. 8. The Universality of the satisfaction is in the judgment of taste viewed as merely siibjective. The quantity of a judgment of taste is individual. This arises from the fact that as this form of judgment rests upon a feeling of pleasure or pain, which arises without 414 CRITIQUE OF AESTHETIC JUDGMENT the interposition of a conception, it cannot be objectively universal. The aesthetic judgment, on the other hand, always predicates something of the individual object directly before us either in perception or imagination, and hence in quantity it is always individual. If it is objected that we may make such judgments as " all roses are beautiful," where the quantity is universal, Kant answers that, though this is no doubt a judgment universal in quantity, it is not an aesthetic judgment, but a genuine logical judgment, resting upon a number of aesthetic judgments each of which is individual in quantity. 9. Does the feeling of pleasure precede or follow the judgment of taste ? There is a contradiction in the supposition that pleasure in a given object precedes the judgment that it is beautiful. For, as the pleasure would on this supposition simply be an affection of the sensibility, and would therefore depend upon the pecuhar sensitive constitution of the individual, we should be unable to explain the conviction that it is com- municable to every one. In a judgment of taste the state of feeling that is experienced in the consciousness of the beautiful must be capable of being experienced by every one, and therefore the pleasure in the object follows from the judgment. But nothing is capable of being communi- cated to all men except knowledge in some form : in no case is there any objective consciousness except when there is knowledge, and therefore knowledge is presupposed as the condition of universality. If the ground of the judgment is purely subjective, i.e., does not imply the definite con- ception of an object, it cannot be anything but that state of mind which accompanies the relation of our faculties to one another, when our consciousness of an object is referred to knowledge in general, as distinguished from specific knowledge resting upon a definite conception. There is ANALYTIC OF THE BEAUTIFUL 415 here a free play of our faculties of knowledge. When knowledge is based upon a definite conception, such as that of causality, the understanding is exercised in the constitu- tion of the object ; but in the free play of our faculties of knowledge there is no such definite conception prescribing and hmiting their exercise. Now, speaking generally, the faculties of knowledge are imagination, which brings together the elements of perception, and understanding, by which those elements are combined in a unity. Hence, when the idea of a given object results in the free play of the faculties of knowledge, the state of feeling which thus arises must be universally communicable, since the faculties of knowledge, and therefore the feeling which accompanies their exercise, are the same in all men. The aesthetic judgment, then, goes before our pleasure in the object, and is the ground of it ; but the universal communi- cability of the pleasure is due to the universality of the subjective conditions of the judgment, i.e., to the identity of the operation of the faculties of knowledge in all men. It must not be supposed that we have in this case an intellectual consciousness of the purposive activity of our faculties : all that we have is a feehng of the excitation of the imagination and the understanding to indefinite and yet harmonious activity. The beautiful, then, is that which gives pleasure imiversally, without the intermediation of a definite conception. III. The Judgment of Taste as Regards the Eelation OF Ends. 10. Purpose in general. By an end, in the transcendental sense, is meant an object the existence of which is only possible by means of a conception. Such an end obviously cannot be explained by a reference to the particular facts of experience. An object, for instance, may produce a feeling of pleasure 416 CRITIQUE OF AESTHETIC JUDGMENT in the subject, and this pleasure may be called an end : but since it depends upon the peculiar susceptibility of the subject, it is not an end in the transcendental sense, for we cannot state a priori the conditions under which it will arise. An end, in the transcendental sense, is such that the idea of the effect or result precedes and is the ground of the cause. The kind of causation we here have is, therefore, that of the final cause {forma finalis), which is distinguished from natural causation by the mark just men- tioned, viz., that the effect can only come into existence by means of an antecedent conception of it. The consciousness of the idea as maintaining the subject in a certain state may be called pleasure, whereas pain consists in preventing or removing pleasure. Now, will is the faculty of desire, so far as it can be brought into activity through the conception of an end. Here we have an instance of purpose in the strict sense of the term : the subject has the definite con- ception of the end or object to be attained, and determines himself by reference to it. But an object or state of mind or even an action may be said to be purposive even when there is no explicit conception of an end, if we cannot account for its existence without introducing the idea of a cause acting purposively. In this case, as we may say, we have an instance of " purposiveness without purpose " ; for, though we do not affirm that the object has actually been produced by a will, we have no other way of making its existence intelligible to ourselves than by conceiving it in that manner. 11. The Basis of a Judgment of Taste is the purposive Form of an Object. When the idea of an end is the ground of satisfaction, it creates an interest in the subject, and upon this the judg- ment is based. The judgment of Taste, on the other hand, cannot arise in this way, nor can it be based upon the ANALYTIC OF THE BEAUTIFUL 417 conception of an object as purposive in itself or as in itself good. As aesthetic it has to do only with the relation of our faculties of knowledge to each other, and there- fore it must rest upon a purely formal or subjective adaptation. 12. TJhe Basis of the Jxidgment of Taste is a priori. We can only learn from experience what effect in the way of pleasure or pain a sensation or conception will pro- duce. No doubt it has been shown a priori, in the Critique of Practical Reason, that a feeling of reverence may be derived from universal moral conceptions ; for there we were able to go beyond experience, and connect that feeling with a free or supersensible subject, though we were not able to derive the feeling itself from the idea of morality, but only the determination of the will. The truth is that any determination of the will implies an effect on the sensibility, and is therefore not an effect of it. Something similar takes place in the case of the aesthetic judgment. Just as practical reason acts upon the sensibility, giving rise to the feeling of reverence for the moral law, so the harmonious exercise of the higher faculties of knowledge results in the pleasure called aesthetic. The difference between the two cases is, that in the former there is an interest in the object, whereas in the latter the pleasure is purely contemplative and therefore disinterested. Hence the consciousness of the harmony of our faculties of know- ledge in the contemplation of the beautiful object is itself aesthetic satisfaction, and stimulates them to harmonious activity. The pleasure experienced is only causal in the sense of maintaining the state of consciousness without leading on to definite knowledge : the mind dwells upon the beautiful object, and the contemplation of it strei^thens and reproduces itself. 2d 418 CRITIQUE OF AESTHETIC JUDGMENT 13. The Pure Judgment of Taste is ivdependent of charm arvd emotion. Personal interest in an object destroys the impartiality essential to a pure judgment of taste, especially if it arises from natural inclination and is based upon the pleasure anticipated from the realisation of a given object. It is a barbarous taste, which is not satisfied with contempla- tion unless the object at the same time has sensuous charm or produces sensuous excitation, and especially barbarous if these accidents are made the foundation of the judgment that the object is beautiful. What Kant here refers to as " sensuous charm " is an immediate sensible quality, such as the green of a meadow, or the visible brilliancy of an object under the influence, say, of sunlight. 15. The Jvdgment of Taste is entirely independent of the Conception of Perfection. An object can be known to be purposive only in so far as there is a relation of its various parts or elements to a definite end, and therefore only when there is a definite conception of that end. It is thus at once evident that the judgment that an object is beautiful does not depend upon any conception of an end, but is connected in some way with the form of knowledge. Purposiveness in an object is either external or internal : the former, when the object is regarded as adjusted to something other than itself ; the latter, when the object is viewed as internally purposive, i.e., as displaying a certain perfection in the combination of its parts. It is plain that aesthetic satisfaction cannot arise from the idea of the utility of the object ; for, in that case it would depend upon the comparison of the object with that to which it is adapted, and such a comparison is an act of the under- standing; whereas the judgment of the beautifid is direct ANALYTIC OF THE BEAUTIFUL 419 and independent of any explicit activity of the under- standing. The perfection, or internal purposiveness of an object, seems to be much nearer to the predicate of beauty; and hence thinkers Like Baumgarten have identified the perfection of an object with its beauty, adding only that in the case of the aesthetic judgment the conception ia confused, whereas in the case of the judgment of perfec- tion it is clear and distinct. It is of great importance to determine whether this identification is tenable. If an object is pronounced to be in itself purposive, we must be able to show that it is inconceivable without the introduction of the conception of an end, and indeed with- out the introduction of the conception of internal purposive- ness. In this case, therefore, we must first have the con- ception of the kind of thing that the object ought to be ; in other words, we must have an idea of its qualitative perfection, as consisting in the harmony of its parts or elements. Qualitative perfection must be distinguished from quantitative perfection, the latter of which consists in the object coming up to the standard of the class to which it belongs and containing all the elements characteristic of the class. The mere harmony of the various elements- of a thing gives no knowledge of the object as purposive, unless there is first a definite conception of the kind of unity in which the harmony should consist. But, when abstraction is made from this unity as an end, and we do not ask what the thing ought to be in order to be perfect, nothing remains but the subjective purposiveness implied in the harmonious operation of the faculties of the mind in the individual who contemplates the object. No doubt this implies a certain facility in the subject of presenting the form of the object in imagination, but it is quite independent of the perfection of the object, since there is no conception of the object as purposive. Now, the judgment of taste is aesthetic ; in other words, it rests entirely upon subjective grounds, and as it does not imply 420 CRITIQUE OF AESTHETIC JUDGMENT any conception of purpose in the object, it cannot depend upon the perfection of the object. In judging an object to be beautiful, there is no consciousness of its perfection as an object; and it is therefore a mistake to suppose that the idea of perfection and the idea of beauty are identical. The distinction is not one of degree but of kind. The judg- ment of taste does not differ from a logical judgment in being less distinct : as entirely independent of any know- ledge of an object, it is generically different. IV. The Judgment of Taste as eegaeds Modality. 18. The Modality of the Judgment of Taste. The beautiful is that which is regarded as implying a necessary relation to our satisfaction. But the necessity is of a peculiar kind : it is not a theoretical or objective neces- sity, depending upon the fundamental constitution of the understanding in its relation to the elements of perception; in other words, it is not a necessary or invariable conn&xion of objects of experience through the law of causality. In pronouncing an object beautiful, we cannot say a priori that as a matter of fact every one will feel the same satisfac- tion, as would be the case were it due to a law of nature. Nor again is the necessity to be identified with moral obli- gation, as arising from a law or rule applicable to aU rational beings. The necessity in this case must be called exemplary; i.e., it simply means that every one on the presentation of this object should agree in pronouncing it to be beautiful. 19. The Subjective Necessity which we ascribe to the Judgment of Taste is conditional. The judgment of taste is universal and necessary, but only in the sense that we are entitled to hold that every one ought to agree in the judgment that we ourselves make. The oiight in this case is different from the ought of moral ANALYTIC OF THE BEAUTIFUL 421 obligation; for the latter is true unconditionally, since it applies to all possible intelligences and to man under all possible conditions, while the ought of the former is conditional, because it assumes a correct subsumption of the individual case under the rule of assent. 20. The Necessity asserted in a JudgmerU of Taste rests upon the Idea of a Common Sense. The judgment of taste is not unconditionally necessary ; unlike the judgment of knowledge, it is not the application of a definite objective principle ; for we do not in making it bring before our minds a principle, such as causality, and subsume the particular case under it. On the other hand, the judgment of taste is not independent of all principles ; for in that case, like particular judgments of perception, it would have no necessity whatever. This form of judgment must, therefore, be based upon a principle, but upon one which is subjective, i.e., which rests upon the manner in which the sensibility is affected by the mere contemplation of a beautiful object. Now, a principle which yet is based upon feehng may well be said to be based upon a common sense, if only we carefully observe that this common sense is quite distinct from that common understanding to which the name is often given ; for the latter is never based upon feeling, but only upon conceptions, though no doubt these conceptions are not made an explicit object of reflection, but are present in an immediate or unrefiective way. 21. Are we jiistifled in presupposing a Common Sense? All judgments of knowledge, together with the belief that accompanies them, are universally communicable, i.e., they are capable of being made by all who realise their meaning. If this were not true, every one would make judgments peculiar to himself, expressing merely the subjective play of his own consciousness — a view which logically results in 422 CRITIQUE OF AESTHETIC JUDGMENT complete scepticism. Now, if knowledge can be shared in common by all, there must also be a possible agreement in the state of feeling which accompanies knowlec^e ; i.e., in the feeling of the harmonious exercise of the faculties with a view to knowledge in general. In point of fact, knowledge actually arises whenever the senses bring the imagination into play in the synthesis of the different elements of perception, and the imagination brings the understanding into play to carry up the imaginative syn- thesis so produced into the unity of the understanding. The harmony of the powers exercised in knowledge differs according to the character of the object, but it must be such that it is best fitted to bring the imagination and the understanding into the proportion most suitable for knowledge. It is therefore determined, not by conception, but by feeling. Since, then, this harmony of the faculties of knowledge must be capable of being shared by aU, so also must be the feeling which is its sign or index, and the universal communicabihty of feeling presupposes that there is a general agreement in feeling between all who ex- perience it ; in other words, it presupposes a common sense. Thus, from the mere consideration of the harmonious exercise of our faculties of knowledge, in the proportion which results in agreeable feeling, we have sufficient evi- dence for supposing that there is a common sense. We may, therefore, say that we can prove the existence of this common sense a priori, without falling back upon the uncertain evidence of psychological observation. From all these considerations we conclude that beauty is known, without the intermediation of any definite conception, as the object of a necessary satisfaction. General Bemarks. The result of the whole discussion is as foUows. Taste is the faculty by which an object is judged to be beautiful, not ANALYTIC OF THE BEAUTIFUL 423 because it can be brought under a definite rule, but simply because it implies the free conformity to a rule or law of the imagination. The judgment of taste, however, must not be regarded as based upon the reproductive imagina- tion ; for this form of imagination consists simply in the arbitrary play of ideas, as suggested by association. The form of imagination upon which the judgment of taste is based is productive or self-active. No doubt in presenting before itself any given object, imagination is limited by the definite form of this object, and to this extent it is not free ; but there is nothing in this fact to prevent us from holding that the form thus supphed by the object contains such a combination of parts or elements as the imagination would itself create, if it were left to its own absolutely free exercise; a combination which must therefore be in harmony with the law or rule of the understanding. We cannot, of course, say that the imagination is in itself at once free and subject to law, for that would be a direct contradiction; but there is nothing to hinder us from saying, that the imagination freely operates in conformity with a law prescribed by the imderstanding. When imagina- tion is forced to conform to a definite law, the form assumed by its product is determined by the conception of what ought to be. But the satisfaction which arises in this case is not in the beautiful but in the good, and therefore the judgment is not a judgment of taste. We thus see that imagination, as employed in the aesthetic realm, is confor- mity to law without law; in other words, the imagination does actually freely conform to the law of understanding, but without doing so under the condition of the express and definite conception of an object. 424 CRITIQUE OF AESTHETIC JUDGMENT BOOK II.— ANALYTIC OF THE SUBLIME. 23. Transition from the Beautiful to the Sublime. The beautiful and the sublime agree in being pleasing in themselves. They also agree in being based upon a judg- ment of reflection, not upon a judgment of sense or a logically determinant judgment ; i.e., the judgment in then- case is not the subsumption of a particular object under a rule that is given or presupposed, nor is it a judgment based upon the sensuous characteristics of particular objects ; it is a judgment in which the object is directly presented, but in which the rule under which the object is to be sub- sumed does not precede the judgment. The satisfaction in the case of the sublime and the beautiful is therefore independent of sensation, and also of all definite con- ceptions. At the same time it bears a certain relation to conceptions, though these are indefinite ; and hence aesthetic pleasure arises in the mere presentation of an object, or is related to the faculty of presentation. Here, then, the faculty of imagination in the case of a given presentation is regarded as in harmony with the under- standing or the reason. Both the judgment of the beautiful and the judgment of the sublime are also individual, while yet they claim to be universally valid for every subject, but without basing that claim upon any definite knowledge of the character of the object. Although the beautiful and the sublime agree in being pleasing in themselves, and in presupposing a judgment of reflection, they yet exhibit striking differences. The beauty of nature is connected with precisely limited objects, while the sublime arises in connexion with an object which has no definite limits and may even be perfectly formless. Thus, while the beautiful implies definite limits, the sublime involves the effort towards a complete whole, an effort that is never entirely successful. The beautiful, there- ANALYTIC OF THE SUBLIME 42^ fore, may be regarded as implying an implicit exercise of the understanding, inasmuch as the understanding is the faculty which determines objects by specific conceptions ; while the sublime may be viewed as an implicit exercise of the reason, for it is the peculiarity of the reason to seek for an unconditioned totality. In the beautiful the satis- faction is, therefore, connected with the consciousness of the quality of the object ; in the sublime it is connected with the idea of quantity. The consciousness in the two cases is also different in kind ; in the case of the beautiful there is a feeling of pleasure, arising from the free outflow of the vital activity, and hence it may be associated with sensuous charm and the play of the fancy. The pleasure which arises in connexion with the sublime, on the other hand, is not direct but indirect. The first effect of the sublime object is to check the outflow of vital forces, and it is only secondarily that this is followed by their stronger outflow. Hence, the sublime excludes immediate sensuous charm, and involves a severe and solemn exercise of the imagination. Since the feeling of pleasure arising from the sublime is indirect or secondary, it is not so much a positive as a negative pleasure, analogous to the feeling of reverence which accompanies the consciousness of the moral law. In considering more particularly the distinction between the sublime and the beautiful we may limit ourselves to the sublimity of nature, for the sublime in art is simply in some way a reproduction of the sublimity of nature. Natural beauty consists in the harmony of the elements implied in the form of the object, and this harmony directly excites in us a feeling of pleasure, upon which the aesthetic judgment is based; so that in the case of the beautiful there is a direct judgment, which seems to rest upon the direct adaptation of the object to our faculties of knowledge. In the case of the sublime, on the other hand, the form or formlessness of the object is rather the negation of any appearance of purpose, so that the judgment is not direct. 426 CRITIQUE OF AESTHETIC JUDGMENT but indirect. The imagination struggles to represent what is beyond its power to represent. The object, being of such a character that it cannot be compressed within limits, escapes from all attempts to represent it, and therefore the judgment, that the object is sublime, is indirect, and indeed it is to be noticed that it is viewed as all the more sublime, the more it baffles aU the powers of the imagination to represent it. Here, therefore, there seems to be an opposition between the object and our faculties of knowledge, instead of a direct adaptation of the one to the other. We cannot, strictly speaking, call an object of nature itself sublime, although we may properly enough speak of objects of nature as beautiful. For, in the case of the sublime, there is no direct adaptation between the form of the object and the faculties of the subject, such as is implied in the case of the beautiful. We cannot, therefore, have an immediate feeling of satisfaction in the object of nature which we call sublime, because it is rather inharmonious than harmonious with the direct exercise of our faculties. What we must say is, not that any object of nature is sublime, but that it is fitted to produce a feeling of sublimity by the reaction of the mind against the object. The truly sublime cannot be compressed within any sensible form. It implies an immediate exercise of the reason, which, as ever seeking for an absolute whole or uncon- ditioned, revolts from all objects which are incapable of realising it. It is the very inadequacy of the object sensuously presented to realise the idea of reason, that calls up in the mind the feeling of sublimity. The ocean roused to fury by a tempest is not in itself sublime, but .simply terrible, and only when the mind is fiUed with ideas of reason is there a reaction against the natural shrinking of the sensitive nature from that which threatens its destruction, so that the mind reacts against the external tobjeot, being stimulated to rise above all that is sensible, ANALYTIC OF THE SUBLIME 427 «i and to occupy itself purely with ideas. Thus the pur- 's posiveness in the case of the sublime is in the subject, and H indeed in the rational subject, not in the object. fe The beauty of nature is of such a character that we are * naturally impelled to figure it after the type of a conscious *i,, agent realising a purpose or combining different elements 'I'j with the object of securing an end. The principle of H purposiveness, which we thus introduce, suggests that ilf phenomena should be judged to belong, not merely to a "j mechanical system of nature, but to something analogous to art. It is true that the beautiful does not actually te extend our knowledge of nature ; for, as we have learned, p knowledge is limited to the determination of objects by S'(L the principles of the understanding, and the highest of sij! these principles is that of reciprocal causation, which does i- not take us beyond the conception of nature as a mechanical i£ system. But, while this is true, it is also true that the ft beautiful forces us to introduce the idea of purposiveness, li- and therefore to think of nature as if it were an artist. Is On the other hand, there is nothing in the sublime V which induces \is to apply the idea of purpose directly to I'J' nature itself; for nature, in its wildest and most unregulated jl state, is calctJated to excite the feeling of the sublime, all ■^ that is required being that it should display tremendous a! magnitude or power. The idea of the sublime, therefore, e does not lead to such rich applications as that of the )4 beautiful. It suggests no purpose in nature itself, but f ' only in the use that we make of it in producing the con- is | Bciousness of elevation above the sensible. The beautiful we tf, are forced to explain by something in the character of the te| object; the sublime we must refer to the peculiar state !S) of mind that arises in us from the contemplation of the vastness of the power of nature. Thus the idea of purposiveness, which in the case of the beautiful is referred to the object, arises in the case of the sublime only from the consideration of the subject; and therefore the theory 428 CRITIQUE OF AESTHETIC JUDGMENT of the sublime is an adjunct to the aesthetic judgment of purposiveness in nature. 24. Forms of the SMime. The analysis of the sublime compels us to make a division between the mathematically sublime and the dynamically sublime, a di^dsion which was unnecessary ia the case of the beautiful. The former is the sublimity of magnitude ; the latter the sublimity of force or power. The distinction between the feeling of the sublime and the feeling of the beautiful is that in the former there is a movement of the mind itself, while in the latter the mind rests upon the object. This movement must be regarded as implying a certain purposiveness in the subject, since the consciousness of the sublime is accompanied by a feeling of pleasure. As related through the imagination either to knowledge or to desire, the feeling of the sublime implies an adaptation, which must be attributed to the harmony of our faculties ; for the feeling of pleasure does not arise from any idea of the satisfaction to be secured through the object, but is entirely disinterested. In the one case the adapta- tion is attributed to the object as a magnitude ; in the other case to the object as a force. There is therefore either a mathematical or a dynamical determination of imagination. A. The Mathematically Sublime. 25. Uxplanation of the term "Sublime." That which is sublime is said to be absolutely great. We must therefore distinguish it from magnitudes, or definite quantities {quanta). The absolutely great is that which exceeds every definite quantity. To say that a thing is absolutely gi-eat is not the same thing as saying ANALYTIC OF THE SUBLIME 429 absolutely that it is great ; for in the former case we merely affirm a thing to he great taken hy itself, while in the latter case we declare it to he great heyond all comparison. Now, when we say that anything is great or small or of moderate size, we are not employing a pure conception of the understanding, for a conception is something fixed and unchangeahle. Nor are we making use of a perception of sense, since nothing is said to be great or small or of moderate size except in comparison with something else. Nor again is it a conception of reason that we employ, since it does not involve a principle of knowledge. It is therefore an idea of reason that is here in question. If we take a magnitude simply by itself, we know it to be a quantum, or definite quantity, because it implies the combination of homogeneous units within itself; but we can only tell how great it is by taking some other magnitude as a standard, and comparing it with this standard. In judging magnitudes, we therefore consider not merely the units combined into a whole, but the magnitude of the units, i.e., we estimate the quantum of a thing by the number of units which are taken, as the standard of measurement. Hence the determination of the magnitude of phenomena, being always relative, cannot possibly yield an absolute conception of magnitude. But when a thing is said to be, not merely great, but absolutely great, or sublime, it is obvious that it is useless to seek for a standard of comparison in anything but the thing itself. It follows at once that the sublime is not to be found in nature, but only in our ideas. This may be expressed by saying that the subHme is that in comparison with which all else is smaU. It is plain that, since nature is simply the sum of sensible phenomena, nothing in it can be judged to be infinitely great or infinitely small. These predicates indicate merely a relative point of view, affirming only that the thing is great in comparison with that which is smaller, or small in comparison with that which is 430 CRITIQUE OF AESTHETIC JUDGMENT greater. Hence no object of sense can be sublime. Nevertheless, the imagination perpetually presses forward in the endeavour to represent something which is infinite or absolutely great, and the very failure of imagination indicates that reason demands absolute totality ; for, we could not be conscious of the failure of imaghiation to represent the infinite, if we had no idea of the infinite. We may, therefore, express the character of the sublime also in this way : that is sublime the mere ability to think which proves that we possess a faculty which transcends every standard of sense. 26. WTiat is requisite for the Idea of the Svhlime in estimating the magnitude of natural things. The estimation of magnitudes by means of numerical conceptions is mathematical, their estimation in perception by measuring them with the eye or by means of touch is aesthetic. Now, we can only have definite conceptions of the size of an object by making use of numbers, i.e., by using the conception of a unity, and therefore all estimation of magnitudes by an act of thought is mathematical. At the same time, this logical process cannot be carried on except upon the basis of perception ; for the unit which is to serve as the measure of magnitude must be assumed before the logical process of numerical estimation can go on. And since any unit, judged logically, is relative to another measure, ultimately the unit which is to serve as a standard must be presented in perception, i.e., the unit may be said to be aesthetic. In the logical estimation of magnitudes there is no maximum, for we can go on adding unit to unit ad infinitum, but there is undoubtedly a maximum when we endeavour to estimate magnitudes by perception, or, what is the same thing, in imagination. There is thus a maximum in the aesthetic estimation of ANALYTIC OF THE SUBLIME 431 i magnitudes. Here, if it is asserted that there is an li absolute measure than which no greater can be found, e it is obvious that we are setting up the idea of that which 4 transcends every possible presentation. It is in this way If tiiat the idea of the sublime arises, producing in us a I pecuhar emotion, which is entirely absent from the mathematical estimation of magnitude by means of numbers. is. The infinite is not merely comparative but absolutely ill great. Hence all other magnitudes of the same kind, as jil compared with it, are small. The important thing, however, in the present connexion is the capacity of thinking of it as a whole, which shows that there is in us a faculty of mind that transcends every standard of sense. To represent ' 1 the infinite sensibly would require a power of compre- hension in which a standard was employed that had a lis definite relation to the infinite supposed to be expressible ifc in numbers. This is obviously impossible, but the bare ili power of thinking this infinite without contradiction impUes itfj in the human mind a faculty which transcends perception. |i It is only by means of this faculty, as giving rise to the sj idea of a noumenon, that the infinity of the sensible world ]! can be completely grasped in thought, though it can never ,j; be comprehended by the mathematical method of employing jj numerical conceptions. Bji The measure of definite quantities is not absolute but ,, relative. The only absolute measure of nature is given in , ji the conception of absolute infinity. Now, when this conception is applied to nature as a phenomenon, it is found to be self- [^ contradictory, because it is impossible to present an absolute totaHty, or the infinite, inasmuch as the only method of ji presenting it to the imagination is by adding unit to unit in ^ an endless progress. The consequence is, that imagination j^i is foiled in its attempt to present that which is adequate ta / the conception of infinity, and therefore the mind is carried j back to a supersensible substrate of nature, as presupposed both in it and in our faculty of thought. As supersensible 432 CRITIQUE OF AESTHETIC JUDGMENT this substrate obviously transcends every measure of sense, and therefore strictly speaking it cannot be called an object, but rather consists in an idea in our minds which we apply in criticism of the object presented by imagination. It is thus properly speaking our own state of mind which must be regarded as sublime. 27. The quality of the satisfaction involved in our Judgment of the Sublime. The feeling which arises when the absolute totality involved in an idea is unattainable is the feeling at reverence, because an absolute law is imposed on us by our reason, to which in our utmost efforts as sensuous beings we cannot completely attain. Now, the idea of a complete synthesis of phenomena is demanded by a law of reason, for reason is always of this character that it cannot be satisfied with anything short of the unconditioned. In the present case it refuses to recognise any absolute measure except that of the complete whole ; and as imagination even by its utmost effort never reaches such a whole, and there- fore never succeeds in presenting an object adequate to the idea of reason, it betrays its own limits and inadequacy, while at the same time it shows that the ideal which is guiding it is the idea or law of reason. The feeling of the sublime, which we ascribe to nature, is therefore properly reverence for our own character as rational; for reason pays no respect to the limitations of sense, but demands absolute conformity with its own law. It is therefore by a kind of subreption that we transfer to nature what strictly speaking is pertinent only to humanity in our own person, i.e., to humanity conceived of as imposing a rational law upon itself. We have here a new evidence of the superiority of the rational determination of our faculties of knowledge over even the greatest effort of our perceptive or sensible nature. ANALYTIC OF THE SUBLIME 433 B. The Dynamically Sublime of Nature. Nature is regarded as dynamically sublime, when it manifests a power which exceeds definite computation, but at the same time does not overmaster us. Nature is never regarded as dynamically sublime except when ia the first instance it is fitted to excite fear, — though of course we fear many things that are not sublime. In an aesthetic judgment, which, from the nature of the ease, cannot be based upon a definite conception of the under- standing, the consciousness of superiority to an opposing force must be estimated by the magnitude of the resistance to it. AVhatever we seek to resist is from the point of view of our sensitive nature an evil, and when it is so great that it is beyond all our powers of resistance, it is an object of fear. In the case of the aesthetic judg- ment, nature can therefore be regarded as a power, or as dynamically sublime, only when it is an object of fear, i.e., when it excites the natural shrinking from physical evU. But an object may be viewed as fearful, i.e., as calculated to bring physical evil upon us, whUe yet we may not be afraid of it. This arises when we simply think of a case in which we might resist, recognising at the same time that resistance would be in vaiu. The virtuous man may be said to fear God, but not to be afraid of him ; i.e., he thinks of the possibility of resisting the commands of God, but he puts this from him as having no personal interest for himself, inasmuch as he has no desire to disobey those commands. In the case, however, where resistance to what is recognised as a divine command is thought of as possible, God is recognised as fearful. He who is afraid of nature cannot judge it to be sublime ; for the sublime arises only in so far as the mind transcends merely physical terror, just as the consciousness of an object as beautiful arises only when it excites a purely disinterested 2e 434 CRITIQUE OF AESTHETIC JUDGMENT feeling, i.e., a feeling which is not connected with the satisfaction of appetite and inclination. " Bold, overhanging, and as it were threatening cliffs, masses of cloud piled up in the heavens and alive with lightning and peals of thunder, volcanoes in all their destructive force, hurricanes bearing desolation in their path, the boundless ocean in the fury of the tempest, the lofty waterfall of a mighty river ; these by their tremendous force dwarf our power of resistance into insignificance." But we are impressed all the more by such natural objects, the more fearful they are, provided that we are in a state of mind, which lifts us above physical terror. Under such circumstances we pronounce nature sublime, just be- cause it calls out an unusual exercise of our own strength of the mind, and reveals in us a power to resist the utmost efforts of mere external force, thus giving us courage to measure ourselves against the omnipotence of nature. The immensity of nature, combined with the impossibility of finding a standard proportionate to the aesthetic estimate of its magnitude, brings home to us the limitations of our physical being, but it also reveals to us the fact that reason has a standard higher than anything that can be presented to us in phenomena. The unlimited power of nature is after all not adequate to the idea of absolute unity, in contrast to which everything in nature is small. We thus have in our own minds an idea which lifts us above nature in all its immensity. So the irresistible power or force excited by nature, while it makes us conscious of our physical impotence and limitations, at the same time brings to light the faculty of judging which is independent of nature. As an individual we feel weak and powerless in the presence of the irresistible forces of nature, but reason, putting us at the point of view of humanity, remains unhumiliated. Thus nature comes to be regarded, not as fearful, but as sublime. Strictly speaking, in this as in the mathematically sublime, it is not nature that is sublime, but man himself, in so far ANALYTIC OF THE SUBLIME 435 as in virtue of his reason he is conscious of that in himself which lifts him entirely above nature. The sublimity, then, which we attribute to nature is really in our own mind, in so far as, rising above our own natural being, we at the same time rise above all that is external to us. 29. Modality of the Judgment in regard to the Sublime in Nature. In the case of beautiful objects of nature we can in most cases count on the agreement of others with our aesthetic judgments, for these are directly excited by the immediate character of the object. We cannot, however, have the same assurance that others will agree with us in regard to the sublime ; for va. this case we must presuppose a con- siderable amount of culture, not only of the aesthetic judgment, but of our faculties of knowledge. Openness to the sublime implies in the subject a capacity for Ideas, and involves the exercise of reason, as originating the absolute or unconditioned; for the sublime essentially consists in a contrast between the limitation of nature, as presented before the imagination, and the Idea of the infinite, which is incapable of being presented by imagination even in its utmost efforts. Since therefore the sublime implies a capacity for the Idea of the infinite, and since moral ideas rest upon the conception of an absolute moral law, it is only those who are to a certain degree cultivated, who are capable of rising above the first aspect of nature, in which it is presented as terrible, and of judging it to be sublime. But, though culture is required for the consciousness of the sublime in nature, much more than in the case of the beautiful, it does not follow that we can explain the sublime merely as a product of culture and convention ; on the contrary, it has its foundation in human nature, and indeed in that which we may fairly demand of every one, 436 CRITIQUE OF AESTHETIC JUDGMENT viz., the tendency to the recognition of moral ideas. It is on this tendency that we may base the demand that others should agree with us in our judgments about the sublime. Just as we say that a man is wanting in taste who is indifferent to the beautiful in nature, so we say he is wanting in feeling when he is unaffected by an object which we judge to be sublime. Both are required of every man, and both may be presupposed, where there is any culture at all. l^evertheless, we make a distinction between the beautiful and the sublime. The former we expect every one to experience, because here there is a direct relation of the imagination to the understanding, and the understanding is a faculty without which there can be no experience what- ever. In the case of the sublime, on the other hand, where there is a relation of the imagination not to the under- standing but to reason, we presuppose moral feeling as its subjective condition ; for moral feeling arises in the con- demnation of immediate desire, in so far as it is in opposition to the idea of the moral law. Hence, though we cannot attribute this moral feeling to every one, we yet hold that every one should experience it, and therefore we regard the judgment of the subhme as also necessary. 30. Deduction of Pure Aesthetic Judgments. A deduction or justification of an aesthetic judgment is required only when the judgment implies satisfaction or dis- satisfaction in the form of the object. By a " deduction " Kant means a proof that the judgments made by the subject are necessarily in harmony in some way with the object. Thus the deduction of the categories consisted in showing that there are certain principles of judgment, which are presupposed in the object of experience, because without them there would be no such object. In the present case he argues that a deduction is required only when it has to be shown that the judgments made by the subject imply a certain form in the object. Now, it is judgments of taste ANALYTIC OF THE SUBLIME 437 which are of this character. It is true that such judgments do not rest upon conceptions of the understanding by means of which objects of knowledge are constituted, but they do rest on the presupposition that the form of the object is Buch that it is so adapted to our faculties of knowledge as to excite the feeling of satisfaction or dissatisfaction. The sublime in nature, on the other hand, does not involve a relation between the form of the object and our faculties of knowledge ; for the object may be regarded as without form or figure, and yet it may give rise to a feeling of satisfaction, and thus imply purposiveness, not in relation between the form of the object and the faculties of knowledge of the subject, but between the faculties of knowledge of the subject himself. No deduction therefore is necessary in the case of the sublime, for strictly speaking we do not require to go beyond the subject, there being no sublimity in nature itself; so that, though by a subreption we speak of it as sublime, the basis of sublimity really lies in human nature. When therefore in the exposition of the sublime it was shown that it springs from the character of the subject, as capable in virtue of reason of rising above all phenomena, what was virtually a deduction was already given. The only aesthetic judgments, then, which require deduction are judgments in regard to the beauty of nature. 31. Method of dedv^tion of Judgments of Taste. No deduction is necessary except in cases where judgments lay claim to necessity. But in this class we must also rank judgments which claim to be binding upon every one, although they rest, not upon knowledge, but upon a feeling of pleasure or pain. The judgment of taste is not a judg- ment of knowledge, for it is neither theoretical nor practical ; i.e., it is based, neither upon the conception of nature, nor upon the idea of freedom. Here therefore we have to justify a priori a kind of judgment which does not express either what belongs to the nature of the objective world, 438 CRITIQUE OF AESTHETIC JUDGMENT or that Tiniversal obligation which is implied ia the moral law. What has to be shown is the universal vaUdity of an iadividual judgment as expressing a certain harmony between the form of an object and the faculties of the subject. This viniversal validity cannot be proved by an appeal to experience on the ground that all men agree with us in our sensations, for a universal and necessary judgment cannot in this way be established. What we have to show is that every one must agree in basing his aesthetic judgment upon a certain feeling of pleasirre, i.e., upon his individual taste, while yet this agreement does not rest upon conceptions of the undei-standing. Xow, the judgment of taste has two distinctive characteristics : in the first place, it is universally vaHd a priori, while yet it is an individual judgment ; and in the second place, it is necessary, or rests upon a prioi-i grounds, although its necessity cannot he proved a priori by any logical process. 36. The problem of a deduction of the Judgment of Taste. The difficvdty connected with a deduction of the judgment of taste arises from the fact that what has to be proved is that there is a valid a priori judgment, while yet it cannot rest upon conception, but only upon feehng. The problem, therefore, is, how we can base a universal judgment upon our own feeling of pleasure in an object, and claim that every one should make the same judgment, while yet this claim cannot be based upon any empirical generalisa- tion. This problem may be brought under the general problem of transcendental philosophy, viz., how are a priori synthetic judgments possible ? 37. WTiat is properly averted a priori of an object in a Judgment of Tasted All judgments of taste are by the nature of the case individual, because there is no explicit conception under ANALYTIC OF THE SUBLIME 439 «3 which the individual case is brought, the judgment being k. based upon the idea of the individual instance now present. i: But, though the judgment is individual, it yet must have It! universal validity, and hence what is predicated in the i judgment of taste is the universal validity of this pleasure. ?: The judgment of taste, therefore, affirms a priori, that the e pleasure felt by the subject in this individual case is a ;;i universal rule vahd for every one. ■'- 38. Deduction of the Judgment of Taste. ''' The judgment of taste is connected purely with the form of the object, not at all with its sensible matter. Nor again can it be based upon a conception of the under- standing ; but, inasmuch as the form of the object presents ^" itself to all in the same way, being dependent upon the ■' character of our perceptive faculty, and since the feeling of the beautiful arises from the direct perception of this form, we are entitled to conclude that there must be an adapta- tion between the form of the object and the faculties of knowledge of the subject. For, the subjective conditions ,; under which the faculties of knowledge operate in all men in relation to the same object must be identical. We can ; therefore assume a priori, that there is here a haiinony of an Idea with the conditions of judgment. Note. The deduction of the judgment of taste is a very simple '•i matter, because it is not necessary to justify the objective f vaUdity of a conception, as was the case in the deduction of ' the categories. "Were that necessary, we should be under I the difficulty of explaining how conceptions can determine objective reality. As beauty does not rest upon a concep- tion, the judgment of taste is not a judgment of knowledge; all that it asserts is that we are justified in assuming the same conditions of judgment in all men, and that the object 440 CRITIQUE OF AESTHETIC JUDGMENT in question has been correctly subsumed under those condi- tions. No doubt it is always possible to make a mistake in this subsumption, whereas m judgments of experience such a mistake is virtually impossible. The conceptions of the understanding are the condition without which no experi- ence of an object is possible at all, whereas in the aesthetic judgment we are dealing with a relation of imagination and understanding which presents itself in consciousness only in the form of feeling. But, though a mistake is possible in our subsumption, it is nevertheless true that the judg- ment of taste has a right to claim universal agreement, since it only asserts that every one who correctly subsumes the object under the subjective conditions of knowledge will make the same judgment of taste as we ourselves do. 56. The Antinomy of Taste. The antinomy which arises in connexion with the principle of taste may be thus stated : (1) Thesis. The judgment of taste is not based upon conceptions ; for if it were, it would lead to controversy, and would admit of proof. (2) Antithesis. The judgment of taste is based upon conceptions ; for if it were not, notwithstanding its diversity there would be no controversy, as there is when we demand that others must necessarily agree with us in our judgment. 57. Solution of the Antinomy of Taste. As explained in the Analytic, the conflict between these two principles, which are simply the two characteristics of the judgment of taste, can only be brought to an end by distinguishing the different senses in which " conception " is spoken of in the thesis and antithesis respectively. This double sense of the term " conception " is not to be dis- missed as a mere confusion of ideas ; it is due to the whole ANALYTIC OF THE SUBLIME 441 character of the transcendental faculty of judgment, and therefore the illusion which arises in this as in other oases is inevitable. The judgment of taste must refer to conceptions in some sense, i.e., the individual instances must be brought under a imiversal rule, because otherwise its claim to Tm.iversal validity could not possibly be justified. But it does not follow that this claim can be justified from a conception. A conception is sometimes capable of determination, as is the case with the logical judgment, where, as we have seen, the rule of the understanding is given, and the judgment, when the sensible matter is presented to it, consists in deter- mining the conception by reference to that matter. A conception, on the other hand, may be undetermined, and indeed incapable of determination. This is the case with the Idea of the supersensible, which is presupposed in aU our apprehension of the sensible, but which cannot be further determined by theoretical reason. For the only way in which such determination is possible for us is by the application of conceptions of the understanding, and these are inevitably limited, since they apply only to objects of sensible perception. Bearing this in mind the solution of the antinomy of Taste is not difficult. The judgment of taste has an application to objects of sense, but it does not through this application determine and thereby limit a conception of the understanding. It is not a judgment of knowledge ; being based upon a feeling of pleasure, which arises from the bare contemplation of an individual object, it is a judgment of the individual subject. What is affirmed is that this object is for me an object of satisfaction no matter what it may be for others. Never- theless, in the judgment of taste a wider relation is implied than can be expressed by simply saying that every one has his own taste. There is implied a certain universal relation of the object to the subject, and it is because of this relation that we can claim universal vahdity for our individual 442 CRITIQUE OF AESTHETIC JUDGMENT judgments. Now, universal validity can never exist unless a conception of some kind is implied, though it is not necessarily brought to explicit consciousness. There must, therefore, be presupposed in this case a conception, but it is one which cannot be determined or limited by perception. Hence, no object can be known through this conception, nor can a proof of the judgment of taste be based upon it. Now, a conception which is indeterminate by reference to perception is a pure Idea of the supersensible, and it is conception in this sense that underlies the phenomenal object as well as the judging subject. The judgment of taste, then, is based upon the conception or Idea of the supersensible. The apparent contradiction formulated in the antinomy of taste disappears when we see that a conception is implied, though not a conception of the understanding. The conception of a general ground for the subjective adaptation of nature to our faculty of judgment does not enable us to determine anything in regard to the object that is presented, because it is itself incapable of deter- mination. The judgment, however, does not for that reason cease to have validity for every one, since it is based upon the universal character of man, in so far as he is regarded in his noumenal or supersensible aspect. Thus the judgment is valid for all, though of course only as an individual judgment which directly accompanies perception. The solution of the antinomy of taste, then, is based on the distinction between the different senses in which the term " conception " is used in the thesis and antithesis respec- tively. The thesis, which affirms that the judgment of taste is not based upon conceptions, is true in the sense that it is not based upon definite conceptions ; but the antithesis, which affirms that the judgment of taste is based upon conceptions, is equally true, when by the term " conception " is understood an indefinite conception, viz., the conception or Idea of the supersensible substrate or noumenon. When ANALYTIC OF THE SUBLIME 443 * this distinction is made, the apparent contradiction dis- " appears. The two apparently contradictory principles are * therefore reconcilable with each other. No doubt we cannot '• base the reconciliation upon positive knowledge, for the f limitation of our faculties of knowledge prevents us from I* bringing the supersensible within the sphere of experience. '« But the distinction just made shows that both thesis and ■', antithesis may be true, and nothing more is needed for a '''- solution of the antinomy. If an attempt is made to base ■'' the judgment of taste upon pleasure, on the ground that it '''- implies the idea of an individual, or if we take the view - of writers like Baumgarten that it is based upon the principle of perfection, in either case no solution of the -; antinomy is possible, because both of these views assume ! that phenomena are things ia themselves, and on that E supposition the contradiction remains, since both thesis If and antithesis, as referring in the same sense to the same - object, will be false, being not merely contradictory but £ contrary. The antinomy which arises in connexion with ' the judgment of taste is similar to the antinomy with which ' we had to deal in the Critique of Practical Eeason, and the '■ solution is of the same character ; for in both cases we are 3 forced to go beyond the realm of the sensible, and to seek : for the reconciliation of our faculties of knowledge in the ■ idea of the supersensible ; and indeed there is no other •■ possible way in which reason may be saved from self- contradiction. 323 THE CRITIQUE OF TELEOLOGICAL JUDGMENT. Section I. — Analytic of Teleological Judgment. 62. Formal Objective Purpose. 323 a It often happens that geometrical figures, which are not constructed deliberately with that end in view, are found to be of such a character that they enable us to solve several problems by a single method, or one problem in a number of different ways. The adaptation is here obviously in the geometrical figure as an object, and it is intellectual, i.e., we have not here an instance of adaptation, such as has been discussed in the case of the aesthetic judgment, where the adaptation is subjective. But these figures, though they are found to be adapted to a special purpose, are not constructed solely for that purpose. Hence, though the adaptation to an end is intellectual and objective, it is not an adaptation in the object, but is merely formal. We have no instance of teleology proper, except when the adaptation is at once objective and directly purposive ; i.e., when it rests upon the conception of end or purpose, and when that purpose is attributed to the object. We may, therefore, set aside the case of geometrical adaptation in our consideration 323 b of the teleological judgment. When, on the other hand, things or objects are presented as external, and yet are arranged in a certain definite way, there is real adaptation, as distinguished from formal adaptation. The arrangement of trees and walks in a garden exists because it has been fl: ANALYTIC OF TELEOLOGICAL JUDGMENT 445 definitely and purposively produced, and hence there is no possibihty of knowing that such an arrangement exists unless we have actual experience of the object. Such an instance of real purpose is, therefore, different from the geometrical adaptation of figures, which are found to be useful in the solution of problems. 63. Relative as Contrasted with Internal Purpose. \a Experience compels us to apply the idea of purpose to natural objects only when in no other way can we account for the facts. This may take place either when the effect is regarded as itself the result of purpose, or when it is f, merely the material or means for the act of other possible it natural beings ; in other words, we employ the idea of ji purpose either in the sense of an end, or in the sense of a If means to an end. In the latter case purpose is called I utility in relation to man ; i.e., man himself consciously n employs the object as a means for the reahsation of his g own end ; and it is called advantage when we are speaking ., of other creatures, where the means are not consciously J employed by those creatures themselves with the purpose ; of securing certain ends. Purpose, again, when the effect ; is conceived as itself a product of art, is an internal or immanent purpose, which is viewed as necessary to the ,146 very conception of the existence of the object. Take, as an , instance of advantage, the case of the relation between a sandy soil and pine trees. There is no doubt that pine trees flourish in sandy soil, and as a matter of fact the withdrawal of the sea from the shores of the Baltic laid ' bare large tracts of sand, which served to nourish pine ^ trees, and were therefore of advantage to them. The question may, therefore, be put, whether the withdrawal of the sea is to be regarded as taking place with a view to the I nourishment of pine trees. If this is so, then the sand will be regarded as a relative end, the withdrawal of the 446 CRITIQUE OF TELEOLOGICAL JUDGMENT sea being a means ; just as the pine trees are an end, for which the sand is a means. Whatever may be said of the legitimacy of this application of the idea of purpose, it is at least evident that we have here a purely relative or contin- gent purpose ; for the purpose is not attributed to the pine trees themselves, or the sand, or the sea, but only to the external relation between these different phenomena. It is thus obvious that we can only justify the idea of external natural ends, if we assume that there is some absolute end which nature intended ; otherwise, we should never reach beyond a means, and we should in fact have a progressus ad 325 a infinitum. But, in order to establish the existence of such an absolute end, we must go beyond nature altogether, since nature materially considered is simply the sum-total of particular phenomena. Eelative purpose, therefore, although it suggests the hypothesis that nature is pur- posive, can never justify an absolute teleological judgment ; in order to obtain such a judgment, we must be able to show that there are actual things which must be conceived of as themselves purposive. 64. The Properties of Things which are Natural Ends. 325 & The attribution of purpose to an object of nature must be based upon the impossibility of explaining the com- bination of parts exhibited in the object by ordinary mechanical laws. The object must therefore be of such a character that we cannot explain its existence at all, except on the presupposition of conceptions of reason ; in other words, the form or combination of parts must be perceived by reason to require as its presupposition the idea of pur- pose. The fact that the object in the present case exhibits a form which is not necessary, so far as the ordinary laws of nature are concerned, is a sufficient reason for our regarding the form as possible only through reason. Now, reason, in so far as it is practical, is will, or the faculty of acting with I ANALYTIC OF TELEOLOGICAL JUDGMENT 447 a view to an end, and hence in this case the object must : be regarded as if it had been produced by reason with 5 fi reference to a conceived end. It is not enough, however, that the object should be conceived of as produced by reason, for : this is true also of artificial products. A thing is a natural end only when it is its own cause and its own effects, i.e., when it is self-caused, and when this is true, at once of the ; species, the individual, and the parts of the individual. In I the first place, to take an example, a tree produces another f tree in accordance with a well-known law of nature. The tree produced is of the same species as that which produces L it, so that we may correctly say that the tree is self -produced, ? or is its own cause and its own effect. In the second place, the individual tree is self-produced. This is usually called ; growth ; but growth is not merely a mere increase in size, explicable by mechanical laws, but it consists in the assimilation of the material which the individual tree takes up as a means of developing itself. It is of course true that the tree receives the material for its growth or self-produc- tion from without; but it exercises upon this material a separating, combining and shaping activity, which converts ' it into the means of its own maintenance. Lastly, every part of the tree is self-produced ; so that, unless each part exhibits the same power of self-perpetuation as the whole, the tree itself will die. Thus, a bud which is inoculated on the twig of another tree does not display the characteristics of the other tree, but produces a plant of its own kind. We may therefore properly say, that each tvng or leaf of the same tree is in a sense engrafted on it, and forms of itself an independent tree, being only externally attached to the other tree and nourished by it. But it is also true that while the twigs and leaves are nourished by the tree, they also contribute to the nourishment of the tree, so that a tree repeatedly stripped of its leaves will die. 448 CRITIQUE OF TELEOLOGICAL JUDGMENT 65. Things which are Natural Ends are Organised Beings. 327 a Causal connexion, as thought by the understanding, always constitutes a regressive series of causes and effects ; for here the cause precedes the effect as its condition ; and inasmuch as the cause is itself again an effect, it is preceded by another cause, which is itself an effect, and so on ad infinitum. This kind of connexion we call that of efficient causes. But there is a causal connexion, which does not in this same sense imply a regressive series, viz., the conception of ends. Here the series cannot be strictly called a series, except from the point of view of our apprehension of it, since it may be taken either backwards or forwards ; in other words, the elements causally connected co-exist and are mutually dependent. The cause may, therefore, just as properly be called the effect, and the effect the cause. This form of causal connexion is that of final cause (nexus finalis). 327 6 A thing is a natural end, in the first place, when its parts are possible only in relation to the whole ; in other words, when the combination of the parts presupposes a conception or idea which determines the particular form in which the parts are disposed. This, however, is not suffi- cient to express all that is implied in a natural end ; for what has been said is equally true of artificial products, where the cause is an intelligent being, distinct from the parts, which are combined in a certain way in order to realise which the subject has in view. Hence, in the second place, a natural product must not only presuppose the con- ception of an end, but it must by its very nature be such as to realise that end independently of any intelligent cause external to it ; in other words, the purpose is not external but immanent. And this can be true only when the parts of the natural product are reciprocally cause and effect of each other's form. Under no other condition can the idea of the whole determine the form and combination of all the ' ANALYTIC OF TELEOLOGICAL JUDGMENT 449 parts. Here we cannot speak of a cause which brings ^ together parts in a certain way ; what we must say is, j that in no other manner can we conceive of the particular i: combination of parts exhibited in the object except liaas under presupposition of immanent purpose. A body is, i therefore, a natural end only when its parts are reciprocally ' dependent, both in their form and in their combination ; ', while, conversely, the idea of the whole may be regarded as ■ the cause of the body in accordance with a principle. In 1 this case, therefore, there is a combination of ef&cient and i final causes; in other words, the efficient cause is con- i ceived as the means by which the final cause is realised. «286ln a natural product each part not only exists by means of the other, but is conceived to exist for the sake of i the other and of the whole, so that each part is an ' instrument or organ ; and not oidy so, but these organs reciprocally produce one another — a fact which distinctly marks off organised beings from artificial products. The dnly natural end, then, is found in self-organising beings. 128 fl Organised beings are the only things in nature, which, taken by themselves, must be conceived as existing only as ends ; for in no other case is purpose attributed to the object. For this reason the conception of an end of nature, as distinguished from a practical end, i.e., an end secured artificially by an intelligent subject distinct from the object, first obtains objective reality from the consideration of organised beings ; and if it had not been found that there are natural objects that can be explained in no other way, there would have been no justification for conceiving nature teleologically. 66. The Principle hy lohich Organised Beings are judged to he internally purposive. 129 rt The principle which is applied when an object is judged to be internally purposive may be thus stated : 2f 450 CRITIQUE OF 'V[A.V.O\A)iUV\L .IIIIHJMKNT / An ()r}j;iuiiH(Hl |)n)(lii(;l; of iiiil.um in one. in whicli a,ll l,ho parts are ro(;i|in)c,;iily oml luui irit«um. Tliiw iiii)iiioH tliat every jmrl. of mi iirtj;miiH(Ml bciiij^ is iKMnwHury to l,li'>">-ori laws of the understanding, or it may rest upon laws which are of a ;; specific character and are capable of indefinite extension. In M applying the laws of the understanding to nature, judgment : does not require any special principle of reflection ; its sole : task being to subsume sensible particulars under the laws ; prescribed for them by the understanding. But the specific laws of nature are so numerous and diverse that reason is compelled to go beyond the mere system of nature as constituted by the determinant judgment. Here, there- fore, a special principle for the regulation of judgment is i necessary, and that principle must be supplied by judgment itself, if an investigation into the phenomena of nature is to be conducted in an orderly way. Unless judgment has such a guiding conception it must obviously proceed in a perfectly unsystematic manner. But, since there are two distinct ways of judging of nature, viz., that pursued by the determinant judgment and that which is charac- teristic of the reflective judgment, an antinomy may arise ; for, when it is assumed that both forms of judgment apply ' to objects of experience, and apply to them in the same sense, a dialectical contradiction inevitably arises, the principle of reflection being assumed to be constitutive, like the principle of the determinant judgment. i2a The first maxim of judgment is the position: All pro- duction of material things and the forms of material 454 CRITIQUE OF TELEOLOGICAL JUDGMENT things must be judged to be possible according to purely mechanical laws. 332 h The second maxim is the counterposition : Some products of material nature cannot be judged to be possible accord- ing to purely mechanical laws, but require quite a different law of causality, namely, that of final cause. 332 « So stated, these are regulative principles, which can be employed in enquiring into the specific laws of nature. If they are converted into constitutive principles, deter- mining the conditions under which objects of nature are possible, they will run thus : 332 rf Position: All production of material things is possible according to purely mechanical laws. 332 e Counterposition : Some production of material things is not possible according to purely mechanical laws. 332/ Now, if we take the last pair of propositions as con- stitutive, or as determining the character of objects of nature, the one is obviously contradictory of the other. Here, therefore, we have an antinomy, which arises from the character of judgment. Eeason cannot prove either pro- position, for reason deals purely with the supersensible, and therefore it supplies no a priori principle constitutive of sensible things. 333 a The first two propositions, on the other hand, simply express maxims of retiective judgment, and are therefore not really contradictory. When it is said that all events in the material world, and therefore all the products of nature, must be judged to be possible by purely mechanical laws, there is nothing in the statement which implies that there is no other possible way of judging in regard to them. What is affirmed is that, in seeking to obtain a knowledge of the specific nature or laws of sensible objects — in other words, in advancing our knowledge of the special sciences — we must in all cases regard objects as if they were purely mechanical products of nature. For, since nature, so far as our experience goes, is a mechanical system, in which DIALECTIC OF TELEOLOGICAL JUDGMENT 455 every object is determined by its relation to another object outside of itself, all our judgments in regard to nature as such must be governed by the principle of mechanical causation. But, though this is true, there is nothing to prevent us, if we meet with objects which are obviously not completely explained when they are determined as parts of the mechanical system of nature, from follow- ing the guiding-thread of the conception of final cause, and employing it as a subjective or regulative principle, by which we advance our knowledge of the specific forms of nature. It is not denied that determination by the principle of mechanism is the only determination of objects ; what is said is only, that we miist not assume that there is no other possible way of determining them. When we set up the principle of final cause, and guide ourselves in all our attempts to discover the specific laws of nature by it, we are not affirming that objects may not be explicable purely by the principle of natural mechanism; all that we are affirmiug is, that our reason, constituted as it is, can' never positively discover the hidden ground of nature theoretically, and therefore that we are not in a position either to affirm or to deny that the forms of nature are ultimately explicable in a purely mechanical way. The question is, therefore, rightly left undetermined, whether, if we could penetrate to the inner ground or substrate of nature, thus abolishing the limits of our know- ledge, we should not find the two methods of judging natural objects — viz., by mechanism and by final cause — ultimately reducible to a single principle. "We must admit, however, that our reason ia its theoretical use is by its very character incapable of having a direct or perceptive knowledge of this supersensible ground of nature, and therefore that we cannot take the principle of reflective judgment, viz., final cause, as determining certain forms of nature by means of its own peculiar principle. 456 CRITIQUE OF TELEOLOGICAL JUDGMENT 76. Remark. 334 a Theoretical reason cannot make objective or synthetical judgments. Setting up the idea of the unconditioned, ia its three forms of the world, the self and God, it yet cannot pass legitimately from these Ideas to the assertion of objects corresponding to them. Eeason, in fact, supplies us with knowledge only in so far as the conceptions of reason are determined or limited by their relation to sensible objects in space and time ; in other words, all positive knowledge through reason takes the form of the under- standing, which enables us to construct a system of phenomena, but does not justify the claim to a knowledge of ultimate reality. Theoretical reason, therefore, just because it deals with the supersensible, while our know- ledge is limited to the sensible, does not contain any constitutive principle, but merely regulative principles. Now, the very nature of our theoretical reason or intelligence forces us to distinguish between the possible and the actual. The conception of the unconditioned is undoubtedly an actual idea in our minds, but it does not entitle us to affirm that there is any corresponding reality. We can think of that which may or may not be, as distinguished from that which actually is. This distinction would never arise for us were it not that our intelligence, by its imalterable con- stitution, is dependent for concrete particulars upon sens- ible perception, while sensible perception is of such a character that it prevents the possibility of our knowing things as they are in themselves. Were our intelligence free from the limits of sensible experience ; in other words, were it a perceptive intelligence ; then the object of knowledge would always be actual, because whatever we should per- ceive would be at the same time real. The distinction, then, between the possible and the actual is one that cannot apply to ultimate reality, or things as they are in them- DIALECTIC OF TELEOLOGICAL JUDGMENT 457 J selves. It is a distinction which arises simply from the , peculiar character of our faculties of knowledge, and which has a meaning only because we can think that which we cannot know. It is certainly true that the possible may not be actual, and hence that we cannot derive actuality from possibility. At the same time, it must be clearly observed that this distinction is true only of human reason, or at least only of a reason that like ours is not perceptive but discursive. That it does not apply to things in themselves is evident even from this, that if there were no idea of the unconditioned set up by reason, we should never come to see that the objects of our experience are but phenomena. Eeason, therefore, exhibits an ; irrepressible tendency to presuppose an unconditionally j necessary condition, or an original ground of things, in which the distinction of possible and actual no longer holds 15 a good. This conception of an absolutely necessary being, — that is, a being, in which whatever is possible is actual — is an indispensable Idea of reason, but at the same time . it is an Idea of that which can never be positively estab- ' lished. The contrast of the possible and the actual / is due entirely to the character of our faculties of know- 'f ledge, and therefore it can have no objective application. We cannot say that every intelligence must make such a distinction, for we have no right to assume that all intelli- I gences operate under the limitations which are applicable to our intelligence. There is nothing to prevent us from supposing an intelligence in which thought is perceptive, a I perceptive intelligence, and therefore there is nothing to hinder us from supposing that there are beings for whom there ' is no opposition between the possible and the actual. For i- a perceptive intelligence whatever is possible is actual, and ; therefore such an intelligence can never draw a distinction between the contingent and the necessary; whatever is would for it be necessary, or what is the same thing, the possible and the actual would be identical. 458 CRITIQUE OF TELEOLOGICAL JUDGMENT 335 6 Just as theoretical reason presupposes the idea of the unconditioned necessity of the original ground of nature, so practical reason presupposes its own unconditioned causality or freedom, as inferred from the absolute commands which it imposes upon the subject. There is here a contrast between necessity, in the sense of an act that is universally binding upon all rational beings who possess a sensitive nature, and necessity, in the sense of the inviolable con- nexion in experience of events through the law of natural causation. An act that is morally necessary is yet regarded as physically contingent, since moral obligation does not inevitably carry with it the translation of the free act into the form of physical sequence. It is due to the character of our practical faculty that moral laws are presented to us in the form of commands, and the acts conforming to them as duties ; for this arises from the fact that, as we are at once rational and sensitive, our actions do not always proceed purely from reason. Hence reason expresses neces- sity, not in the form that this act inevitably takes place, but in the form that it ought to take place. Were reason in us a cause acting altogether independently of sensuous desire, every act that we now declare to be one that ought to take place actually would take place, and there would therefore be no distinction between what is and what ought to be ; in other words, the distinction would disappear if man were purely a denizen of the intelligible world, and therefore a, being all of whose acts were completely in accordance with the moral law. If that were the case, it would not be necessary to distinguish between being and doing, between a practical law, expressing what we are capable of willing, and a theoretical law of that which is actual through us. In a purely intelligible world, in short, whatever is possible would be at the same time actual, because nothing would be possible except that which was imposed by reason, i.e., that which is absolutely good. While the consciousness of moral law entitles us to infer the existence in us of freedom I DIALECTIC OF TELEOLOGICAL JUDGMENT 459 1' as the formal condition of an intelligible world, it is for us 1 a transcendent conception, and therefore it cannot serve as a i- constitutive principle determining our acts, since these acts : are for us theoretically part of the natural or phenomenal s world. Though our nature is partly sensuous, being S affected by natural desire, it is also true that freedom, as = implying the idea of conformity to reason, is for us and for ; all other rational beings that are possessed of a sensuous s nature, a universal regulative principle. This principle does ; not objectively determine the nature of freedom, but it 5 commands every one to act in conformity with the idea of J freedom, and that as absolutely as if it were a constitutive J principle. s6a From these considerations we may learn the nature of » the principle of teleology as employed by judgment. The i reason why we are compelled to distinguish between the t mechanical laws of nature and the teleological connexion of i nature is that our intelligence, not being perceptive, is i compelled by its constitution to proceed from the universal ; to the particular. As we can have no knowledge of the i adaptation of the particular to an end, judgment cannot be determinant unless it has a universal law, under which it may subsume the particular. We cannot, as Kant argues, advance simply by an accumulation of particulars ; nor can we, on the other hand, have knowledge merely through universal conceptions of thought. In aU cases there must be some way of bringing together the universal and the particular in one act. Now, the particulars which are presented to us in sensible observation are not neces- sary or immediately conformed to the universal, but exhibit a certain contingency. Yet reason cannot be satisfied with anything short of the reduction of the particulars to law. The very character of reason, in fact, is that in all cases it seeks for unity. And as the particulars and the universal are not inevitably harmonious with each other, it is not possible directly to subsume the former under the latter. 460 CRITIQUE OF TELEOLOGICAL JUDGMENT The matter with which we are here dealing is contingent, and wherever there is such contingency of the particulars by reference to the universal, we can only proceed as if there were conformity between them. Hence the con- ception that natural products are purposive, i.e., that there is an adaptation between them and the idea of law, though it is necessary for our judgment, cannot be at- tributed to objects themselves. If it were so attributable, we should have to maintain that there was no distinction between matter and form; in other words, that the par- ticulars were such that they must conform to an inviol- able law. But this again would mean the reduction of nature to a mechanism. The idea of purpose, then, is simply a subjective or regulative principle of reason, although for our judgment it has the same validity as if it were an objective or constitutive principle. 77. The Conce2Mon of Natural End as due to the Peculiar Character of our Intelligence. .337 a There are certain peculiarities of even our higher faculties of knowledge, which we naturally apply as objective pre- dicates to things. They are not really so applicable, because they are only presented in Ideas, and, as we know, it is not possible to find in experience objects corresponding to Ideas. This holds good even of the conception of a natural end, which cannot be predicated of objects, but exists only as an Idea. Nevertheless, the effect which corresponds to this Idea, the natural product, actually exists in nature, and therefore it seems as if we had a right to employ the Idea of natural end as a constitutive principle. There is thus a marked distinction between the Idea of a natural end and all other Ideas of reason. Kant means that in the case of all other Ideas no object at all can be found in nature to which the Idea may be applied. Thus the Idea of the soul is so entirely separated from the particular states of the individual M DIALECTIC OF TELEOLOGICAL JUDGMENT 461 "t subject as in time, that there is nothing in these states to t compel us to apply the Idea of natural ends. Similarly, in i the Idea of God we are lifted entirely above the whole sum f;aof sensible things. The distinction between the Idea of t natural end and all other Ideas is due to the fact that, lii while it is no doubt a principle of reason, it is not one f: which can be employed by the understanding. If tit it were, there would be a direct subsumption of the k particulars of sense under a universal law or rule. The ft Idea of natural end is entirely relative to the judgment, ifi: and is therefore merely the application of our intelligence t : in general to possible objects of experience. The judgment I: is in this case not determinant but merely reflective, and E hence, though the object with which it deals is presented in ii: experience, the Idea under which the object is brought is merely a principle of judgment, which cannot deter- mine the object itself, but is only employed reflectively to bring it under the subjective or regulative idea of ! 6 purpose. It is therefore a peculiarity of our human inteUigence that we must judge of natural things by refer- b enee to the idea of purpose. Xow, the very limitation of !!• our intelligence, as exhil^ited in this case, suggests to us the j; ' Idea of an intelligence, not so limited ; in other words, the r: Idea of a perceptive intelligence, in which the object would ,i be directly known, instead of being indirectly brought under :; a subjective conception with a view to the organisation 2 of experience. Just as, in the " Critique of Pure Eeason," a by setting up the idea of a perception different from ours, vff we were led to see that our perception is limited, so the jji character of our teleological judgment is clearly perceived is; by us in contrast to a higher intelligence not so limited. if When we see that from the point of view of a possible higher is intelligence natural products would be known as they are, I while we must consider them purely by reference to the con- .1;; ditions of our knowledge, we also see that it is a subjective i principle that natural products should be considered by us 462 CRITIQUE OF TELEOLOGICAL JUDGMENT as if they could not exist at all unless they were produced by an intelligence operating by reference to a conceived end. But, though we cannot avoid this mode of conception, we are not justified in taking it as entitling us to affirm that an intelligent cause actually exists, which produces certain natural objects by reference to the idea of an end. For aught we know a higher intelligence would find that objects are completely explicable by the conception of mechanism, without recourse being had to the idea of final 338 c cause. We must therefore expect to find that there is a certain contingency in the relation between our intelligence and its faculty of judgment, and the determination of the character of this contingency will bring out the distinction between our intelligence and other possible intelligences. 339 a The contingency obviously lies in the fact that our intel- ligence does not originate particulars, but has to depend upon sensible perception for them. Hence judgment can only consist in bringing the particulars so supplied under the universals of the understanding. It is due to the very character of our understanding that, while it gives us the universal, i.e., consists in a universal and necessary faculty of combination, there are many ways in which the parti- culars of perception may affect our sensibility, even when they agree in some common characteristic. Knowledge with us in all cases implies perception as well as conception. If we possessed a perceptive faculty, which spontaneously originated particulars, our knowledge would not be limited in the manner described; but such a faculty would not be a form of sensibility, since the very character of sensibility is that it is receptive. In contrast to our intelligence, however, we are able to conceive of a perceptive intelligence ; though, having no experience of it, we can only describe it negatively by saying that it is not discursive or indirect, but perceptive or direct. This intelligence we think of as differing from ours in this way, that it does not proceed from the universal through the particular, in order DIALECTIC OF TELEOLOGICAL JUDGMENT 463 to constitute the individual ; for, in such an intelligence there would be a direct connexion between particular laws of nature and the understanding, and therefore there would be no contingency. It is through the conception of au intelligence of this perceptive character that we are able to think the possibility of the adaptation of natural things to our faculty of judgment. The intelligence so conceived is not one that like ours sets up in idea a certain end, and then proceeds to realise it, but in it the object must be conceived of as directly presented, so that there is no possible distinction between the actual and the possible. We can only represent the relation between natural laws and our faculty of judgment by conceiving natural laws to be so adapted to our faculty of judgment that we are able to systematise our experience by means of the regulative con- ception of purpose; in other words, by employing the idea of final cause in the explanation of the relation between them. ! Our understanding necessarily proceeds from analytic universals to particulars. Operating as it does with ab- stract conceptions, it is in itself incapable of going beyond the abstract universal from which it starts ; and therefore it is compelled to depend upon empirical perception for particulars. These particulars, on the other hand, when they are viewed in separation from the universal, are not brought into unity, and the function of judgment is to effect their combination by bringing them under the universal. This does not exclude the idea of an intelligence different in kind from ours, an intelligence which is per- ceptive and not discursive, and which therefore proceeds from synthetic universals to particulars, comprehending the whole directly, and so does not need to obtain particulars for the universal by going beyond the universals with which it starts. Such an intelligence, supposing it to exist, would not view the whole as separated from the parts, or the parts as separated from the whole ; therefore there would be for it no contingency in the connexion of the parts. 464 CRITIQUE OF TELEOLOGICAL JUDGMENT The very nature of our intelligence, however, compels us to view every real whole in nature as the result of the combined motive forces of the parts, i.e., as a mechanical system. By this process we pass from the parts to the whole, but in contrast to this manner of procedure we can conceive of an archetypal intelligence, which comprehends the parts in the whole, both in their specific nature and in their connexion. Now, in the case of a discursive intelli- gence like ours, since the connexion of the parts does not necessarily presuppose the whole, obviously it must be the idea of the whole by which the form and connexion of the parts is explained. But such a whole is viewed as an effect or product, the idea being conceived as the cause of the product. Here, in other words, we regard the product as an end, which is secured by previously setting up the idea of the product and viewing the parts as means to its realisation. It is therefore due to the character of our intelligence that we look upon certain natural products as due to a different kind of causality from that of the laws of nature, viz., a final causality. But, as the application of this idea of final cause arises from the peculiar character of our intelligence, we cannot attribute it to things in them- selves. Even phenomenal objects we cannot claim to know, since they are viewed as means and ends only because our intelligence is compelled so to determine them. Thus, while we must attribute immanent purpose to living beings, as the only way in which we can explain their peculiar character as phenomena, it does not follow that this predicate of immanent causality actually determines or constitutes their nature. It is the feeling that the idea of final cause is only subjective that accounts for the dissatis- faction which we experience when any attempt is made to explain natural products by that idea. The reason for this dissatisfaction is the tacit conviction, that we are not entitled to attribute purpose to objects ; in other words, that tlie idea of purpose is merely the method by which our DIALECTIC OF TELEOLOGICAL JUDGMENT 465 ; reflective judgment enables us to explain what otherwise would be for us inexplicable. It must be observed that, in thus setting up the idea of an intelleetus archetypus, we are not affirming the actual existence of a perceptive intelligence. Whether such an intelligence exists or not must be deter- mined by other considerations ; all that we have affirmed is ■ that we discover the relative or limited character of our own intelligence by contrast to a perceptive or archetypal intelligence. la The conception of a material whole as in its form pro- * duced by the parts acting and reacting on one another, '■ gives us the idea of a mechanism. This idea, however, does ' not imply the conception of a whole as end, and therefore it is inadequate to explain the peculiar character of organised beiugs, in which there is presented to us an object, the possibility of which we cannot explain apart from the idea of the whole as determining the character of the • form and combination of the parts. This does not mean : that organised beiugs may not after all be the product of '. mechanical forces ; for that would be equivalent to saying that no intelligence could possibly think the parts as com- bined in a vmity unless the idea of a unity was the cause of ' the whole, i.e., unless the product was regarded as purposive. A unity of this kind is based upon the character of objects i as in space ; and space is not a real ground of products, but i^ simply their formal condition. It is true that space is of ' such a character that the parts cannot be determined except ! in relation to the whole ; but, on the other hand, as merely a •! formal condition, it in no way determines the character of the parts or their mode of combination. Now, the material world is for us phenomenal, and as such it implies the ', possibility of a substrate corresponding to the idea of an intellectual perception. We thus obtain the idea of a I supersensible and real ground of the world of nature, although from the conditions of our knowledge that ground cannot be brought within the system of experience. So 2g 466 CRITIQUE OF TELEOLOGICAL JUDGMENT far as we are dealing with phenomena we must apply mechanical laws, because these are the only laws which are applicable to objects that present themselves under the forms of space and time; but harmony and unity of the particular laws and forms of nature is not thereby excluded, since the mechanism of nature applies only to phenomena. The supersensible ground of nature is purely an object of reason, and when we confront the idea of the supersensible ground of phenomena with the different character of phenomena themselves, we can only harmonise the two by means of the idea of final cause. Thus, nature has to be judged on two distinct principles, the mechanical and teleologioal, which in no way conflict with each other, because the former applies to objects viewed purely as phenomena, while the latter is relative to the possible harmony of the supersensible and the sensible. 242 a It is thus obvious that there is no contradiction between the principles of a mechanical derivation of natural products and the teleological principle. When phenomena are of such a character that we conceive of them as natural ends, or organised beings, it is impossible to explain them by the mechanical laws known to us, or even to imagine how any extension of those laws should account for a single organised being. Thus the principle of final cause is absolutely indispensable to us in the extension of our knowledge. No human intelligence, and indeed no finite intelligence, will ever account for the production of even the simplest organised being by mechanical causes ; hence the principle of final cause is indispensable to us in our judg- ments in regard to such objects. It is true that there is nothing in external phenomena which compels us to affirm that they can only exist as the product of an a priori intelligence. The reason for coming to this conclusion must be sought in the indispensable substrate of pheno- mena. Since, however, we cannot have any knowledge of that substrate, our knowledge being necessarily DIALECTIC OF TELEOLOGICAL JUDGMENT 467 limited to objects of nature, we must seek in the con- stitution of our intellectual faculty for the ultimate ground of teleological connexion; in other words, we must set up an original intelligence as the cause of the world, in order to explain to ourselves the existence of organised beings and through them of nature as a whole. Appendix on Method. 87. The Moral Proof of the Existence of God. There is sufficient justification for theoretical reason to maintain, on the ground of a physical teleology, that there is an intelligent cause of the world. From another point of view, however, we must infer a moral teleology. In our moral consciousness we have the certitude of a law which admits of no exception, and this law, as we have seen in the " Critique of Practical Eeason," compels us to postulate free causality in ourselves. But as the moral - ends or purposes which we pursue, and the laws which express the universal principles of action, are determined a priori, we cannot from the mere consciousness of the moral law infer the existence of an intelligent cause outside of ourselves. The moral law is in itself absolute, being imposed upon us purely by our own reason, and iadeed the purity of moral action would be destroyed, did we seek to base it upon the will of a supreme intelligence. At the same time moral teleology, though from the point of view of the motive of action it is entirely independent of the world of nature, must yet be connected with • that world in this sense, that the actions by which the moral law is carried into effect must necessarily produce an influence upon the phenomena of nature. Man, in other words, is not merely a moral, but also a natural being; and therefore the translation of his moral purposes into action involves an influence directly 46 S CRITIQUE OF TELEOLOGICAL JUDGMENT upon his own sensirive nature, and so indirectlT on other external obiecis. Xow, in the conception of ourselves as under moral law we find the ultimate standard, br reference to which we judg-e other things to be ends, or to be Sub- ordinate to om-seives as ultimate ends. The reaiisarion of morality, in Kant's ^-le^v. is the one ultimate end which we are entitled to affirm as that towards which all things are snivin^. for nothing in the world can be rec;vrded as Ir^.- imare except the jood will. Moral teleology, then, has to do with the relation of oiu-sehes as causes to the moral ends which we se: before ourselves, and indeed to the ultimate end of the reahsation of an ahsolutelv moral commmiity. in wliich each makes himself and others ends in themselves. The realisiirion of this tiltimate end pre- supposes that the external world is not in absolute opposition to it. bur permits of its realisation. The quesrioa aiises. therefore, whether we ai'e not compelled by reason to seek for a principle, outside of the world and independent of it. which shall ac-count for the possible union of uarm-e and the ultimate end of morality. Thus there arises a moral teleology, which has to do, on the one hand, with the self-legish^.tion of a moral or free subject, and. on the other hand, with nature as the sphere in which such a subject must realise himself. 344 a Granting the contingency of the things of nature, which is the ordinary point of ■view of our sensible experience, each object must be regarded as dependent upon something else as its cause, or as not self-caused. On the other hand, this cannot be an ultimate point of Wew. suiee reason demands a supreme or ultimate cause, and therefore we may seek for this supreme cause, or unconditioned gioimd of the con- ditioned, either in the physical or the teleological order : in other wonis, we may ask, what is the supreme cause wliich produces tilings ? or what is their supreme or uncouditioueti end, the end implied in the production of a certain class of tiiingrs, viz., organised beings, or even of all things ? ZS^ow, i APPENDIX ON METHOD 469 £ the cause in the latter case we conceive as an intelligence, which sets before itself an end that it seeks to realise, or at ; least, as Kant guards himself by saying, we must conceive of that cause as if it were an intelligence, and therefore as if it acted in accordance with the laws of an inteUigence. ;.5a From the teleological point of ^iew it is admitted by all : that the only ultimate end which reason demands a priori is that of man as under moral laws. A world in which . there were no moral beings, i.e., no beings that were ends m. \ themselves, and which consisted only of inanimate things or : even of mere animals, would have no meaning or value, ■ because in such a world there would exist no rational or intelligent beings, and therefore the whole of it would •r necessarily be of a purely mechanical character. And even :• supposing that there were intelligent beings in the world, but that these beings were iucapable of self- ;. determination or freedom and could only estimate things from the point of view of the pleasure they were fitted to bring, though there would no doubt be relative ends, there would be no absolute or idtimate end ; for no ultimate end can be derived from the relation of objects as fitted to bring pleasure to the sensitive subject. But man as a moral being prescribes for himself ends that are absolutely obligatory, and therefore, as a self-legislative and self- determining being, his morality satisfies the conception of . an ultimate end. The only ultimate end, then, in the world is that of rational beings as living under absolute j moral laws which they prescribe for themselves. fbh Now, the moral law, as the formal condition of freedom, imposes itself on us purely by its own authority, refusing to recognise that the natural desires as such have any claim upon us. At the same time it determines for us, and indeed a priori, an ultimate end as the goal towards which aU our j efforts ought to be directed; and that end is the highest I good possible in the world through freedom, in other words, [6 a the complete realisation of the moral law. On the other 470 CRITIQUE OF TELEOLOGICAL JUDGMENT hand, the subjective condition that we are entitled to regard as reasonable is happiness. This end must be sub- ordinate to the ultimate end of virtue or goodness. The highest physical good possible in the world is happiness, and this end we must seek to promote, although it must never be the motive of our action, and to promote always under the objective condition that the only person worthy to be happy is he who makes the moral law his motive. 346 6 Now, it is not possible in consistency with all the faculties of our intelligence to maintain that these two ends, virtue and happiness, will be secured simply by the operation of natural causes; hence, if the only cause in the world is nature, it follows that there is no guarantee of the realisa- tion in the world of the demand for the moral law. In fact there is nothing in the character of natural necessity which entitles us to say that the practical necessity of the moral law can be realised at all ; in other words, we may will the moral law, and yet the character of the world may .346 c be such that happiness is an impossibility. If therefore we are to satisfy all the demands of reason, and the summum honum, in both of its senses, is to hold good, we must suppose a cause different from nature, which brings the two ends of morality and happiness into harmony with each other, and this cause must be conceived as moral, for otherwise we should be affirming simply a natural or mechanical cause. The conception, therefore, of an absolute end, i.e., of the possibility of the free realisation of absolute moral laws by beings who are ends in themselves, pre- supposes the existence of a moral cause or author of the world ; in other words, it presupposes the existence of God. 80. Limitation of the Moral Proof. Zil a The ultimate end, as merely a conception of our practical reason or moral consciousness, cannot be derived from the facts of experience, and therefore it does not in any way *^ APPENDIX ON METHOD 471 li extend our knowledge. Its only possible use is in the r. determination of action by a free being, who sets before f himself an ideal law and acts purely by reference to it. ;5 The ultimate end of creation must be conceived as that :; constitution of the world which harmonises with the only f end that reason recognises as absolutely binding upon us. :: The moral law, because it imposes upon us an absolute :': obligation, refusing to recognise that there are any exceptions £ to it, entitles us to presuppose that the realisation of morality r. is possible, and therefore that nature cannot be of such a character that its realisation is impossible. It is on the i: ground of the moral consciousness, and only on the ground ± of the moral consciousness, that we can maintain the 1 existence of an ultimate end of creation. :U1b The conclusion just reached is that there is a moral teleology ; in other words, that the world must be fitted for i the realisation of that which reason demands as an ultimate ; end, viz., the existence of moral beings. We must, however, in order to justify the demand of reason that this ultimate end should be reahsable, take a step beyond moral teleology to theology ; for we cannot conceive of the possibility of the harmony with nature of the ultimate end demanded by reason except by presupposing that this harmony is produced by a Being who is both intelligent and moral, in other words, by God. But, whUe this conclusion is demanded by reason, it must be observed that it holds good only for the judgment which enables us to represent to ourselves how the demands of reason can be reahsed. The judgment in this case is made, not by the determinant, but by the reflective judgment. No doubt reason, in laying down absolute moral laws, is independent of reason as presenting the conditions in nature under which those laws may be realised. The latter in fact belong, strictly speaking, not to practical, but to theoretical reason. But we cannot assume that in the supreme cause of the world, which we must con- ceive of as an intelligence, there is the same contrast between 472 CRITIQUE OF TELEOLOGICAL JUDGMENT reason as practical and reason as theoretical, and that a kind of causality is required for the ultimate end which is different from that required for natural ends. We are not entitled, therefore, to infer that the supreme cause of the world is a moral being, in the sense of a being who sets up the idea of certain ends and proceeds to realise them. Nor can we conceive of this being as one who purposely adapts nature so that it may harmonise with moral law. T^Tiat we can say, however, is that by the constitution of our reason it is impossible for us to conceive, in any other way than by the adaptation of nature to the moral law, how the opposi- tion of relative ends and the ultimate end can be harmonised with each other. We must, therefore, conceive of the supreme cause of the world as not only an author or ruler of the world, but also as a moral law-giver. 348 a The consideration of nature thus proves for theoretical reflective judgment an intelligent cause of the world ; i.e., nature presents to us certain objects which cannot be con- ceived by us to be possible at all except under presupposi- tion of an intelligent cause ; and indirectly it proves that all objects of nature must be conceived from this teleological point of view. Moral teleology, on the other hand, establishes the existence of such a cause for the practical judgment, by compelling us to recognise that there is an ultimate end of nature, when we bring it into relation with the idea of moral law. It is quite true that we cannot prove the existence of God as the moral author of the world simply from a consideration of nature as implying purpose. But it is the very character of reason that it cannot be satisfied with anything short of an absolute unity of principles, and therefore the knowledge of physical ends, when it is brought into relation with the knowledge of the moral end, is the means by which we are enabled to connect the practical reality of the idea of God with its theoretical reality as already existing for judgment. 348 b Two things must be kept in mind in regard to the moral APPENDIX ON METHOD 473 proof of the existence of God. In the first place, since we have no positive knowledge of the existence of a Supreme Being, on account of the necessary limitations of our know- ledge to objects of sensible experience, we can only think or conceive the attributes of this Being by analogy. It is not possible for us actually to know God, because there is nothing within our experience which reveals to us the nature of a being who transcends all experience. Secondly, it follows from this that, though we are entitled to say that the Supreme Being must by His nature correspond to what we mean by intelligence and morality, this does not enable us to know Him as He is, nor can we predicate these attributes positively of Him. The only manner in which we can realise to ourselves the nature of the Supreme Being M. through the application of the idea of final cause or \ purpose, and that idea, as we have seen, is only a regulative tiot a constitutive principle. Eeason must take the form of the determinant judgment before we could absolutely determine the nature of God, and this is contrary to its fundamental character. The final result of our whole en- quiry is, however, to place the belief in God, freedom and immortality upon a thoroughly rational basis, a result which at first sight seemed to be excluded by the necessary limitation of knowledge to the world of sense. SUPPLEMENTARY EXTRACTS FROM THE CRITIQUE OF JUDGMENT} 207 THE CRITIQUE OF AESTHETIC JUDGMENT. BOOK I.— ANALYTIC OF THE BEAUTIFUL. I. — The Judgment of Taste as Eegaeds its Quality. 1. The Judgment of Taste is Aesthetic. To decide whether a thing is beautiful or not, we do not bring our idea of it into relation to the object by means of the understanding for the sake of knowledge ; we bring our idea of it into relation to the subject and to the feehng of pleasure or pain experienced by the &ubject, and that by means of the imagination, or perhaps the imagination in union with the understanding. The judgment of taste is therefore not a judgment of knowledge, and consequently not a logical judgment ; it is aesthetic, meaning by this that its sole basis lies in the subject. In all other cases the relation of ideas may be objective. This holds good even of the relation of sensations. But it is not so in the case of ' The passages here translated, if embodied in The Philosophy of Kant in Extracts would naturally follow p. 322 of that work. They contain a fairly complete statement of Kant's theory of Aesthetics in his own words. The numerals in the margin indicate the pages in Hartenstein's edition, vol. V. ANALYTIC OF THE BEAUTIFUL 476 the feeling of pleasure and pain, which tells us nothing of the object, but only how the subject himself feels, when he is affected by the idea. ' To know more or less clearly that a building is regular and fitted for its purpose, is quite a different thing from having this idea together with the feeling of satisfaction. In the latter case the idea is related solely to the subject, and indeed bo his feeling of life, which we call the feeling of pleasure or pain. The faculty which decides and judges here is quite peculiar ; it contributes nothing to knowledge, but merely contrasts the given idea in the subject with the whole faculty of ideas, of which the mind is conscious in the feeling of its own state. Certain ideas implied in a judgment may be empirical, and therefore aesthetic ; but the judgment which is formed by means of them is logical, when they are related in the judgment only to the object. Conversely, if the given ideas are rational, but are referred in the judgment solely to the subject, or rather to the subject as feeling, the judgment is always so far aesthetic. 2. The Satisfaction which is expressed in the Judgment j of Taste is disinterested. By interest we mean that species of satisfaction which we combine with the idea of the existence of an object. Such an object always involves a relation to desire, either as the ' incentive to it, or as necessarily connected with the in- centive. Now, when we ask whether a thing is beautiful, we do not desire to know whether anything is dependent upon the existence of the thing for oneself or anybody else ; the only question is how we judge of it when we simply , contemplate it (in perception or reflection). 209 3. Satisfaction in the Pleasant is interested. '210 The pleasant is that which is agreeable to the senses in sensation. This is the proper place to draw attention to a 476 CRITIQUE OF AESTHETIC JUDGMENT common but reprehensible confusion between two very different meanings of the term sensation {Empfindung). All satisfaction, as it is supposed, is a sensation of pleasure, and therefore whatever is agreeable is called pleasant. On this view no distinction is made between impressions of sense which determine inclinations and principles of reason which determine the will, and merely reflective forms of perception which determine the judgment, so far as their influence upon the feeling of pleasure is concerned. If sensation is a mode of pleasure or pain, it is something quite different from the apprehension of a thing by sense, which is a receptivity belonging to the faculty of knowledge. In the latter the relation is to the object, in the former the relation is solely to the subject, and does not contribute to knowledge, not even to the knowledge which the subject has of himself. As employed in the above explanation of the pleasant, sensation {Umpfindung) is an objective presentation of sense, and to avoid misunderstanding it will be better to use the word feeling {Gefuhl) to express what is always purely subjective and does not enter at all into the consciousness of an object. The green colour of a meadow belongs to objective sensation, being the apprehension of an object of 211 sense, while its pleasantness belongs to subjective sensation, which is not the consciousness of an object, or concerns that feeling of satisfaction which affords no knowledge of the object. Now, that a judgment in which a thing is declared to be pleasant implies an interest in it, is at once evident from the fact that through sensation it excites a desire for it ; hence the satisfaction which is experienced not only pre- supposes a judgment in regard to it, but also the relation of its existence to my state, as affected by it. ANALYTIC OF THE BEAUTIFUL 477 4. Satisfaction in the Good is interested. Good is that the very idea of which satisfies us as rational beings. A thing is called good either when it is a means to something else, i.e., is useful, or when it gives satisfaction of itself or is in itself good. In both cases . there is implied the conception of au end, and therefore the relation of reason at least to a possible act of wUl, and consequently satisfaction in the existence of an object or act ; in other words, the good implies interest of a certain Idnd. In order to find a thing good, I must know what sort i of thing it should be, i.e., I must have a conception of it. With the beautiful this is not necessary. Flowers, free drawings, outlines interwoven with one another without purpose into what is called foliage, have no meaning, not being brought under any definite conception, and yet they give aesthetic pleasure. Satisfaction in the beautiful must proceed from reflection upon an object, which leads to a perfectly indefinite conception. It is thus distinct not only from the good but from the pleasant, which rests entirely upon sensation. b Although the pleasant and the good differ in various ways from each other, they yet agree in always implying an interest in their object. This applies not only to the pleasant and to that which is useful or good indirectly or as a means to what is itself pleasant, but also to that which is absolutely and in every respect good, viz., moral good, ; which brings with it the highest interest. For the good is the object of will, i.e., of a faculty of desire determined by reason ; but to will anything, and to have satisfaction in its , existence, or to take an interest in it, are the same thing. IZ 5. Comparison of the pleasant, the good and the beautiful. The pleasant and the good are both related to the faculty of desire, the former involving a pathological 478 CRITIQUE OF AESTHETIC JUDGMENT satisfaction due to an external stimulus, and the latter a pure practical satisfaction, which arises not merely from the consciousness of an object, but from the conscious 214 connexion of the subject with the existence of the object. It is not only the object, but the existence of the object, which gives pleasure. The judgment of taste is purely contemplative ; it is a judgment which is ia- different to the existence of an object, and merely compares its nature with the feeling of pleasure and paia. Nor is this contemplation itself directed to concep- tions ; for the judgment of taste does not lead to knowledge, theoretical and practical, and therefore it is neither based upon conceptions nor has these as its end. 215 From what has been said it follows that Taste is the faculty of judging an object by means of satisfaction or dissatisfaction apart from all interest in it. The object of such satisfaction is called beautiful. II. — The Judgment of Taste as Eegaeds its Quantity. 6. The Beautiful is that which is vievjed as the object of a universal satisfaction apart from conceptions. This view of the beautiful follows from the ex- planation given above, in which it was characterised as an object of entirely disinterested satisfaction. For when a man is conscious that his satisfaction in a thing is entirely disinterested, he cannot refrain from judging that it must contain the ground of satisfaction for every one. Since it is not based upon his own inclination nor upon any reflective interest, but on the contrary the subject in judging is perfectly free as regards the satisfaction with 216 which he contemplates the object, he can discover no private inclination in himself as the ground of his satisfaction, and therefore he must regard it as having its source in that which he is entitled to presuppose also in others ; hence he ANALYTIC OF THE BEAUTIFUL 479 cannot but believe that he may ascribe a similar satisfaction to every one. It is therefore natural for him to speak of the beautiful as if it were in the object, and of his judg- ment as logical, though it is purely aesthetic and implies only a relation between the subject and his consciousness of the object. It is true that the judgment of beauty agrees with the logical judgment in claiming universal validity. But the universality of the former cannot rest upon conception. For there is no transition from concep- tions to the feelings of pleasure and pain, except in the case of pure practical laws, which as interested differ from the pure judgment of taste. It follows that the judgment of ' taste, as entirely disinterested, claims universal validity, while yet this universal validity does not depend upon the object, but bases its title upon a subjective universality. 18 8. The universality of the satisfaction is in the jvdgment of taste viewed as merely subjective. J9 AU judgments of taste are singular in quantity. Since I must directly refer the object to my feeling of pleasure and I pain without the interposition of conceptions, these judg- SOments cannot be objectively universal in quantity, though no doubt a logical universal judgment can be derived from them. The rose which I see is pronounced beautiful in a , judgment of taste ; on the other hand, when, by a compari- son of many singular judgments of this kind, I obtain the judgment " all roses are beautiful," my judgment is no longer merely aesthetic, but is a logical judgment based upon an aesthetic judgment. Now, the judgment "the rose is pleasant" is no doubt aesthetic and singular; it is ' not however a judgment of taste but a judgment of sense. The universality of the judgment of taste is an aesthetic quantity, being valid for every one, whereas the judgment that the rose is pleasant is not. It is only judgments about the good which at once produce satisfaction and yet have 480 CRITIQUE OF AESTHETIC JUDGMENT logical and not merely aesthetic universality ; for these imply knowledge of the object and for that reason are of universal validity. 221 9. Does the feeling of pleasure 'precede or follow the jvdyment of taste ? If pleasure in a given object preceded the judgment that it is beautiful, and that judgment merely affirmed that this pleasure may be communicated to every one who is con- scious of the object, there would be a contradiction in the process. For the pleasure would be merely a pleasant sensation, and would therefore by its nature hold only for a particular individual, as directly conscious of a given object. In a judgment of taste it is therefore implied that the state of mind that I experience in the consciousness of the beautiful can be coramunicated to every one, and the possibility of such commimication is the subjective con- dition of the judgment, while the pleasure in the object 222 follows from it. But nothing can be communicated to all men except knowledge and the form of consciousness which constitutes knowledge. For, there is objective consciousness only so far as there is knowledge, and only through know- ledge is there a universal point of reference with which consciousness must harmonise. But, if the ground of the judgment as to the universal communicability of this form of consciousness is merely subjective, in the sense that it does not involve a conception of the object, it can be nothing but the state of mind which goes along with the relation of our faculties to one another, when we refer our conscious- ness of an object to knowledge in general. By bringing into play the faculties of knowledge this consciousness confines them to a definite rule of know- ledge. Hence the state of mind in this case is that of a feeling of the free play of the faculties with a view to knowledge in general. JSTow, the faculties implied in the ANALYTIC OF THE BEAUTIFUL 481 consciousness of an object from which knowledge in general may arise are, on the one hand, imagination to bring together the elements of perception, and, on the other hand, understanding to supply the conception which combines those elements into unity. The state of mind in which there is a free play of the faculties involved in knowledge of a given object must be universally communicable, because knowledge, as a determination of the object in which given ideas must harmonise, is the only mode of consciousness that is identical in all men. What is communicable to all, then, in a judgment of taste, which must take place without presupposition of a definite conception, is that state of mind in which there is a free play of the imagination and the understanding, so far as these must be in harmony with each other as the condi- tion of knowledge in general. We are conscious that this subjective relation of our faculties as fitted for knowledge in ' general must be the same for every one, and must therefore be universally communicable, just as if it were a definite knowledge, resting always upon that relation as its subjective condition. !23 This merely subjective aesthetic judgment precedes the i pleasure in the object and is the ground of the pleasure felt r in the harmony of the faculties of knowledge, but the universality of the pleasure rests upon the imiversality of the subjective conditions of the judgment. : The stimulation of the imagination and the understanding to indefinite and yet harmonious activity — that kind of ji24 activity which leads to knowledge — is a feeliag, postulated by the judgment of taste as universally communicable. . . . An idea that is individual and independent of all compari- son with other ideas, and yet is in agreement with the I conditions of universality implied in the exercise of the i understanding, brings the faculties of knowledge into that If proportionate harmony which is essential to all knowledge, in and is therefore valid for every one whose nature it is to 2h 482 CRITIQUE OF AESTHETIC JUDGMENT judge by the combination of understanding and sense, i.e., for all men. "We conclude, then, that the beautiful is that which pleases universally without the interposition of a conception. III. — The Judgment of Taste as eegaeds the relation OF ENDS. 10. Purpose in General. If we desire to state what is meant by an end, con- sidered in its transcendental determinations, apart from anything empirical, such as the feeling of pleasure, we must say that the end is the object of a conception, so far as the conception is viewed as the cause of the object, or the real ground of its possibility. Purposiveness {forma finalis) is thus the causality of a conception in respect of its object. Where therefore not merely the knowledge of an object, but the object itself, is thought of as possible only through the conception of its form or existence, there we say that there is an end. The idea of the effect is thus the ground of the cause and precedes it. The consciousness of the causality of an idea to maintain the subject in a certain state may here be taken to denote what we call pleasure ; whereas pain is that mode of consciousness which contains the ground by which the opposite arises, i.e., it implies the prevention or removal of pleasure. The faculty of desire, so far as it can be determined to activity only through conceptions, or in accordance with the idea of an end, is will. But an object or a state of mind or even an action may be called purposive, even if its possibility does not necessarily presuppose the idea of an end, provided that we cannot explain its possibility with- out employing the idea of a cause acting from a purpose, i.e., a will which has ordered it in accordance with the idea of a rule. There can therefore be purposiveness without ANALYTIC OF THE BEAUTIFUL 483 purpose,! jj^ ao far as we do not attribute the cause of this form to a wUl, while yet we can explain the possibility to ourselves only by deriving it from a will. 11. The basis of a judgment of taste is the purposive form of an object. Every end that is regarded as a ground of satisfaction brings an interest with it, which determines the judgment in regard to the object of pleasure. Hence the basis of a judgment of taste cannot be a subjective end. Nor again ;!26 can it be any idea of an objective end, as implying the possi- bility of the object itself according to principles of pur- posive combination, and therefore it cannot be a conception of the good. For the judgment is aesthetic, and therefore does not rest upon a conception of the internal or external possibility of the object through this or that cause, but concerns only the relation of the faculties of knowledge to one another, so far as they are determined by an idea. The ground of an aesthetic judgment can only be a subjective adaptation in the idea of an object without any purpose, either objective or subjective ; i.e., it can only be the mere form of purpose, in the idea through which the object is given, which constitutes the satisfaction judged to be universally communicable without the interposition of a conception. 12. The basis of the judgment of taste is a priori. It is absolutely impossible to determine a priori what effect a sensation or conception will have in the way of a feeltag of pleasure or pain ; for what effect a cause shall have can be learned only from experience. It is true that in the Critique of Practical Eeason we have derived a priori the feeling of reverence from \miversal moral conceptions. 1 Reading with Rosenkrantz : Die Zweokwiissigkeit kann also ohne Zweck sein. 484 CRITIQUE OF AESTHETIC JUDGMENT But there we could go beyond the limits of experience and call in a causality which rested upon a supersensible 227 constitution of the subject, viz., freedom. And even there, strictly speaking, we did not derive this feeling from the idea of morality as its cause, but we only derived from that idea the determination of the will. But that state of mind which accompanies any determination of the will is itself already a feeling of pleasure and identical with it, and therefore it is not an effect of that determination. A feel- ing of pleasure must be presupposed only if the conception of the moral as good precedes the determination of the will through the law ; for otherwise the pleasure connected with the conception could not be derived from the concep- tion as mere knowledge. Something similar is true of the pleasure connected with an aesthetic judgment. The difference is, that here the pleasure is contemplative ; for we are not personally interested in producing the object, as is the case in the moral judgment. The consciousness of the harmony or adaptation of the faculties of knowledge in the idea of a beautiful object is itself the pleasure, because it is the ground of the activity of the subject in the stimulation of those faculties. The cause is internal and purposive as regards knowledge in general, but it is not limited to a determinate act of knowledge, and therefore it implies a mere form of the subjective purposiveness of an idea in an aesthetic judgment. This pleasure is in no way practical, being due neither to a desire for pleasure nor to a concep- tion of what is morally good. Yet it is causal in the sense of maintaining the state of consciousness itself, and keeping the faculties of knowledge in operation without any further object. We linger over the contemplation of the beautiful, because this contemplation strengthens and reproduces itself. This is analogous to that lingering over some physical charm, which repeatedly awakens attention, except that in this latter case the mind is passive. ANALYTIC OF THE BEAUTIFUL 485 1 3. The pure judgment of taste is indepeTident of charm and emotion. All personal interest corrupts the judgment of taste and destroys its impartiality, especially if the purposiveness 228 does not, as with the interest of reason, precede the feel- ing of pleasure, but is based upon it. . . . Taste is always in a barbarous stage when it can only be satisfied by an intermixture of sensuous charm or emotional excitement, and all the more so if it makes these the standard of approbation. 231 15. The judcffnent of taste is entirely independent of the concejjtion of perfection. We can only know an oiject to be purposive by means of a relation of the manifold to a determinate end, and there- fore only through a conception. From this it is at once evident that the beautiful, which is judged as beautiful on the ground of a merely formal purposiveness, i.e., a pur- 232 posiveness without purpose, is entirely different from the idea of the good, because the good presupposes an objective purpose, i.e., a relation of the object to a definite end. Objective purposiveness is either external or internal ; the former implying the utility, the latter the perfection, of ; the object. Obviously the satisfaction in an object which leads us to call it beautiful cannot depend upon our idea of its utility, for in that case our satisfaction in the object would not be direct, as is required by the judgment that it is beautiful. Perfection or internal purposiveness comes nearer to the predicate of beauty, and hence it has even been thought by eminent philosophers to be identical with beauty, though it is explained that beauty is the confused thought of perfection. In a Critique of Taste it is of the greatest importance to determine whether beauty may actually be resolved into the conception of perfection. 486 CRITIQUE OF AESTHETIC JUDGMENT To determine that an object is in itself purposive, we must be able to show that it is not possible apart from the conception of an end, and indeed of an internal end, which is the ground of the internal possibility of the object. As every end implies that there is a conception of it which makes the object possible, so, if we have the idea of an object as purposive, we must first have the conception of what sort of thing it ought to be ; and the harmony of the various elements in the object with this conception is the qualitative perfection of a thing. Quite different from this is quantitative perfection, which consists in the completeness of each thing as one of a class, or the mere conception of its magnitude or totality ; what sort of thing it ought to he is here taken for granted, and the only question is whether it contains all that is requisite for such a thing. That which is formal in the idea of a thing, i.e., the harmony of the various elements in a unity, so long as it is left undetermined what 233 this unity ought to be, affords no knowledge whatever of any purposiveness in the object; for, since abstraction is made from the unity as an end (what the thing ought to be), there remains merely the subjective purposiveness of the ideas present in the mind of the individual who contemplates the object. No doubt this implies a certain purposiveness in the conscious state of the subject and a facility in taking up a certain form into the imagination, but it does not imply any perfection of an object, since there is no conception of the object as purposive. Now, the judgment of taste is an aesthetic judgment, i.e., one which rests upon subjective grounds ; and as it does not imply any conception, it cannot be dependent upon the conception of a definite purpose. In the thought of Beauty, as a formal subjective purposiveness, there is no thought of the perfection of an object, as supposed to imply formal and yet objective purpose. It is vain to imagine that the only distinction between the conceptions of the beautiful and the good is in their logical form, the former ANALYTIC OF THE BEAUTIFUL 487 being a confused, and the latter a distinct, conception of perfection, while in content and origin they are the same. If that were true, they would not be different in kind, but a judgment of taste would be a judgment of knowledge, like a judgment in which a thing is affirmed to be good. As has already been pointed out, an aesthetic judgment is sui generis, and gives no knowledge whatever of an object, not even a confused knowledge, since it is not a logical judgment. 242 IV. — The Judgment of Taste as kegaeds Modality. 18. What the modality of a judgment of tade is. We think of the heaviifvX as that which has a necessary relation to satisfaction. But the necessity is of a peculiar kind. It is not a theoretical objective necessity, in which I can tell a 'priori that everyone will fed the satisfaction which I experience in the object I call beautiful. Nor, again, is it a practical necessity, in which, by means of conceptions of a pure or rational will, which serves as a rule for beings who act freely, a satisfaction arises as the 243 necessary consequence of an objective law ; for such a necessity merely means that we are under absolute obliga- tion to act in a certain way. The necessity in question can only be called exemplary, i.e., it is a necessity that all should agree in a judgment which is regarded as an example of a imiversal rule that cannot be presented. 19. The subjective necessity which we ascribe to the judgment of taste is conditional. The judgment of taste implies the agreement of every one ; and whoever affirms something to be beautiful claims that every one ought to agree with him in affirming it to be beautiful. The ought of the aesthetic judgment is therefore 488 CRITIQUE OF AESTHETIC JUDGMENT expressed in accordance with the data required for the judgment, and yet it is expressed only conditionally. We demand the agreement of every one, because we have a basis for our judgment which is common to all; and we could even count upon such agreement, were we always certain that the particular instance was correctly subsumed under that ground as the rule of assent. 20. The necessity asserted in a judgment of taste rests upon the Idea of a common sense. If judgments of taste, like judgments of knowledge, had 244 a definite objective principle, we could claim unconditional necessity for them ; if they had no principle, like those based upon the senses, no one would think of claiming necessity for them. They must therefore have a subjective principle, which determines what pleases or displeases purely through feeling and not through conceptions, and yet is of universal validity. Such a principle may be regarded as a common sense ; which, however, must be carefully distinguished from that common understanding, to which the name of common sense (sensus communis) is sometimes given, and which never bases its judgments upon feeling, but only upon conceptions, though these are usually merely principles obscurely understood. 21. Are we justified in presupposing a common sense? Judgments of knowledge, together with the belief which accompanies them, are universally communicable : otherwise they would not harmonise with the object, but would all be a mere subjective play of consciousness, as the sceptic affirms. But, if knowledge can be communicated to all, so also the state of mind, i.e., the harmony of our powers with a view to knowledge in general, and indeed that due pro- portion in the exercise of those powers which results in ANALYTIC OF THE BEAUTIFUL 489' knowledge, must also be universally communicable; for, unless every one could have this harmony, as the subjective condi- tion of knowledge, there could be no knowledge as a result. Now, knowledge actually arises in every case in which a given object by means of the senses brings the imagination into play to combine the manifold, and the imagination brings the understanding into play to reduce the manifold 146 so combined to conceptions. But there is a different propor- tion in the harmony of the powers exercised in knowledge according to the difference of the objects. Yet it must be such that it is in general best fitted to bring the imagina- tion and the understanding into the proportion most suitable for knowledge, and this harmony must be determined by feeling, not by conceptions. As this harmony of the faculties must be communicable to all, so also must the feeling which is its index. But the universal communica- bility of a feeling presupposes a common sense; and therefore we have good ground for assuming the existence of a common sense, and that without falling back upon psychological observation, but simply as the necessary condition of the universal communicability of our know- ledge, which is presupposed in every Logic, and in every principle of knowledge that is not sceptical. !46 We conclude, then, that beauty is known without con- ception as the object of a necessary satisfaction. General Remark. To sum up the results of the above analysis : Taste is. the faculty of judging an object by reference to the free conformity to law of the imagination. Now, if in the judgment of taste the imagination must be considered as free, it is not reproductive and subject to the laws of association, but productive and self -active. It is true that, imagination, in the apprehension of a given object of sense, 24V is tied down to a definite form of this object, and so far has. 490 CRITIQUE OF AESTHETIC JUDGMENT no free play (as it has in poetry), but there is nothing to hinder us from supposing that the object may furnish it with a form which contains such an assemblage of elements as it would itself project, if left to itself, in harmony with that conformity to law which is characteristic of the understanding. We cannot, however, suppose without contradiction that the imagination is autonomous. It is only the understanding that prescribes a law. If the imagination is forced to proceed in accordance with a definite law, the form assumed by its product will be determined by conceptions of what ought to be ; but in that case the satisfaction, as shown above, will not be in the beautiful but in the good, and the judgment will not be a judgment by taste. Imagination is therefore a con- formity to law without law ; and only a subjective harmony of the imagination with the understanding, without such an objective harmony as is implied in the reference of an idea to the conception of an object, is compatible with the free conformity to law of the understanding and with the characteristic of a judgment of taste. 251 BOOK II.— ANALYTIC OF THE SUBLIME. 23. Transition from, the Beautiful to the, Svhlime. The beautiful and the sublime agree in being pleasing in themselves. Moreover, both presuppose a judgment of reflection, but not a judgment of sense, or a logically determinant judgment. Hence the satisfaction does not depend on a sensation, as in the case of the pleasant, nor on a definite conception, as in the case of satisfaction in the good. At the same time it is related to conceptions, though these are indefinite, and hence it is connected with the mere presentation of an object or with the faculty of presentation. Accordingly, the faculty of imagination is in the case of a given perception regarded as in harmony ANALYTIC OF THE SUBLIME 491 ?i with the faculty of conceptions of the understanding or the k reason, and as leading to their formation. Both sorts of a judgment are also singular, and yet they announce them- t- selves as universally valid for every subject, although they 1 lay claim merely to the feeling of pleasure, not to any \ knowledge of the object. ! But there are striking differences between them. The t beauty of nature concerns the form of the object as ; having definite limits; the sublime, on the other hand, is to be found even in a formless object, in so far as we i! picture the unlimited in it, or on occasion of it, and at ! the same time think of it as a totality. Thus the beautiful aeems to be viewed as the presentation of an indeterminate ;.M eonception of the understanding, the sublime as the presenta- tion of an indeterminate conception of reason. Hence in : the former the satisfaction is connected with the idea of • quahty, in the latter with the idea of quantity. The t satisfaction is also different in kind : the beautiful brings I with it a direct feeling of the expansion of life, and hence it may be associated with sensuous charm and the play of imagination ; the feeling of the sublime is a pleasure, which ■ arises only indirectly, being produced by the feeling of a I momentary checking of the vital forces followed by a stronger outflow of them, and as involving emotional excitement it does not appear as the play, but as the serious exercise, of the imagination. Accordingly, it cannot be united with sensuous charm ; and as the mind is alternately attracted and repelled by the object, the satis- faction in the sublime implies not so much positive pleasure as wonder or reverential awe, and may be called a negative pleasure. But the most important and intimate distinction between the sublime and the beautiful may be thus explained. We may confine ourselves here in the first instance to the sublime in objects of nature, for the sublime of art is always limited to conditions of harmony with nature. 492 CRITIQUE OF AESTHETIC JUDGMENT Natural or independent beauty involves purposiveness in its form, so that the object seems to be as it were adapted to our judgment, and thus constitutes in itself an object of satisfaction. But that which in our mere apprehension excites in us without ratiocination the feeling of the sublime, may appear by its form to thwart the purposes of our judgment, to baffle our faculty of representation and as it were to do violence to the imagination, while yet it is judged to be all the more sublime on that account. It is thus at once apparent that we express ourselves incorrectly, when we call an object of nature sublime, although we may quite correctly call many objects of nature beautiful. For, how can that be marked by an expression of approval, which in itself is apprehended as 253 being a violation of purpose ? We cannot say more than that the object is fitted for the presentation of a sublimity which can be found only in the mind ; for the truly sublime can be contained in no sensible form, but indicates only ideas of reason, which, although no presentation adequate to them is possible, are by their very inadequacy (which can be presented sensuously) aroused and called up in the mind. Thus the wide ocean roused to fury by a storm cannot be called sublime. Its aspect is horrible ; and one must have his mind filled with many Ideas, before it is determined by such a perception to a feeling that is itself sublime, and is stimulated to forsake the sensible and occupy itself with Ideas, which imply a higher purposiveness. The free beauty of nature reveals a technic of nature, which enables us to represent it as a system in accordance with laws, the principle of which we do not find in our whole faculty of understanding. This principle is that of purposiveness as regards the use of our judgment about phenomena, so that phenomena must be judged not merely as belonging to nature in its non-purposive mechanism, but also as belonging to something analogous to art. The beautiful does not actually extend our knowledge of objects ANALYTIC OF THE SUBLIME 493 ; of nature, but it extends our conception of nature as a mere ! mechanism to the conception of it as a kind of art. This 1 invites us to profound enquiries as to the possibility of such i. a form. On the other hand, in what we usually call the 1^ sublime in nature, there is nothing whatever which leads : to particular objective principles and forms of nature i corresponding to them. Nature in its chaotic state, or in its wildest and most unregulated disorder and desolation, j excites mainly ideas of the sublime, if only greatness and i power are displayed. Hence the conception of the sublime of nature is much less important and rich in consequences ; than that of the beautiful ; it indicates nothing purposive in nature itself, but only in the possible use of our ( perceptions of it, by which we feel in ourselves a purposive- ! ness quite independent of nature. For the beautiful of I nature we must seek a ground outside of ourselves, but in the ease of the sublime we must seek it merely in ourselves and in the attitude of mind which introduces sublimity into 264 the idea of nature. This important distinction entirely separates the idea of the sublime from the idea of a purposiveness of nature, and makes the theory of the sublime a mere adjunct to the aesthetic judgment of purposiveness in nature. 24. Forms of the Sublime. The analysis of the sublime involves a division which was not necessary in the case of the beautiful, the division into the mathematically/ sublime and the dynamically sublime. The feeling of the sublime is characterised by a movement of the mind, which is connected with our judgment of the object, while in the beautiful Taste presupposes that the mind is and remains in restful contemplation. This movement we must judge to be subjectively purposive, since the sublime gives pleasure ; it is related through the imagination either to the faculty of knowledge or the faculty 494 CRITIQUE OF AESTHETIC JUDGMENT 255 of desire, but in either case the adaptation must be judged to belong to our faculties, and to exclude purpose or interest. In the former case the adaptation is ascribed to the object as a mathematical determination of imagination, in the latter as a dynamical determination. A. The liATHEJIATICALLT SUBLIME. 25. Ua2'>h'inat>o)i of the term " SvMivie." We call that sublivie which is ahsolufelif great. Magnitude (magnitudo) and quantity (qiiantitas) are entirely different conceptions. In like manner to say absolutely (sivipliciter) that something is great is not the same as saving that it is absolutely great (absolute, nan comparative magnum). The latter is that which is great beyond all comparison. What, then, is meant by saying that something is great, or small, or of moderate size ? It is not a pure conception of the understanding which is here meant, still less a perception of sense, and just as little a conception of reason, because it does not involve a principle of knowledge. It must therefore be a conception of judgment, or be derived from such a conception, and a subjective adaptation of the idea in relation to judgment must be its foundation. We know that something is a definite magnitude (quantum) without comparing the thing with anything else ; its unity is constituted by the combination of the manifold as homo- geneous. But to know how great it is always requires some other magnitude as a standard. Now, in judging magnitudes we attend not merely to multiplicity or number, but also to the magnitude of the unit (which is the standard of measure), and the magnitude of the latter again requires something else as a standard, with which it may be compared. We thus see, that the determination of the magnitude of phenomena cannot possibly yield an absolute conception of magnitude, but only a relative conception. ANALYTIC OF THE SUBLIME 495 ;j67 But if we call something not only great, but absolutely and in every respect or beyond all comparison great, i.e., sublime, we see at once that it is not permissible to seek for an adequate standard beyond the thing itself, but we must seek for it in the thing. It is a magnitude which is only equal to itself. It thus follows that the sublime must not be sought in the things of nature, but only in our Ideas. What has been said may be thus expressed : The Sublime , is that in comparison ivith which all else is small. It is readily seen that there is nothing in nature, however great it may be judged to be, which may not be reduced to the infinitely small if it is regarded from another point of view, and nothing so small, that it may not be expanded by our imagination to the dimensions of a world when compared with still smaller standards. Hence no possible object of sense is sublime. But, just because our imagination strives onward into the infinite, while our reason demands absolute totality as a real Idea, the very inadequacy of our imagina- tion to estimate the magnitude of things of sense relatively to this Idea awakens in us the feeling of a supersensible faculty. i68 We may therefore add to the above formula this other : That is sublime, the mere ability to think which proves that we possess a faculty which transcends every standard of sense. 26. What is requisite for the Idea of the sublime in estimating the magnitude of natural things. The estimation of magnitudes by means of numerical conceptions (or their signs in algebra) is mathematical, but their estimation in mere perception by measuring them with the eye is aesthetic. We can only have a definite con- ception of how great a thing is by means of number, the measure of which is the unit, and so far all logical estimation of magnitudes is mathematical. But, as the magnitude of the measure must be assumed as known, and 496 CRITIQUE OF AESTHETIC JUDGMENT this again can be estimated mathematically and by means of numbers, the unit of which must be another measure, we can never have a first or fundamental measure, and there- fore we cannot have a definite conception of a given magnitude. The estimation of the magnitude of all objects of nature is ultimately aesthetic, i.e., subjective and not objective. Now, there is no maximum in the mathematical estima- tion of magnitudes, for the power of numbers extends to infinity; but there is undoubtedly a maximum in the aesthetic estimation of magnitudes, of which we may say that, if it is judged to be the absolute measure beyond which no greater is subjectively possible, it involves the Idea of the sublime, and gives rise to an emotion which no 259 mathematical estimation of magnitude by means of numbers can produce. 262 The infinite is not merely comparatively but absolutely great. Compared with it everything else of the same kind of magnitude is small. But what is most important is, that even to be able to think it as a whole points to a faculty of the mind which transcends every standard of sense. For, to represent it sensibly would require a comprehension which should supply a standard or unit having a definite relation to the infinite expressible in numbers, and this is impossible. Yet even to be able to think of this infinite without contra- diction requires in the human mind a faculty which is itself supersensuous. For, it is only through this faculty and its Idea of a noumenon, which cannot itself be presented per- ceptibly, but which is presupposed as the substrate of our perception of the world as a mere phenomenon, that the infinite of the world of sense is completely embraced in one conception in the purely intellectual estimate of magnitude, though it can never be completely grasped in the mathe- matical estimate by means of numerical conceptions. 263 The true unchangeable measure of nature is its absolute totality, or the comprehension of infinity in it as a ANALYTIC OF THE SUBLIME 497 I* phenomenon. But as this fundamental measure is a self- ^* contradictory conception, because of the impossibility of fej absolutely completing an endless progress, that magnitude F; of a natural object, on which the imagination spends in vain te its whole faculty of comprehension, must carry the concep- i! tion of nature to a supersensible substrate, which is pre- supposed both in it and in our faculty of thought. This fc substrate transcends every measure of sense, and therefore i-"; it is not so much the object, as our own state of mind in i estimating it, that we must regard as sublime. "264 27. The quality of the satisfaction involved in our judgments of the sublime. ^'- The feeling of our impotence to attain to an Idea which is a law for us is reverence. Now, the Idea of the synthesis ' of every phenomenon that may be presented to us into ^ the perception of a whole, is one which is imposed upon us 1 by a law of reason. For, reason does not recognise any definite, universal and unchangeable measure except the 265 absolute whole. But our imagination, even in its greatest j effort to combine a given object into a perceptible whole, and so to present the Idea of reason, reveals its limits and I inadequacy, and yet shows that its ideal is to be adequate ! to this Idea or law of reason. Hence the feeling of the J sublime in nature is reverence for our own character, which ' by a certain subreption we transfer to an object of nature j instead of referring it to the Idea of humanity in our own person. This gives a sort of perceptive evidence of the superiority of the rational determination of our faculties of knowledge over the greatest faculty of our sensibility. 2i 498 CRITIQUE OF AESTHETIC JUDGMENT B. The Dynamically Sublime of Nature. 28. Nature as Power. Nature, considered as a Power which does not overmaster us, is dynamically sublime. If Nature is to be judged by us as dynamically sublime, it must be viewed as an object of fear (though we cannot say, conversely, that every object of fear is held to be sublime by our aesthetic judgment). For, in an aesthetic judgment, which is not based on conception, superiority to an opposing force can be judged only by the magnitude of the resistance. Now, that which we strive to resist is an evil, and if we find our power of resistance inadequate, it is an object of fear. Hence for the aesthetic judgment nature can be regarded as a Power, and therefore as dynamically sublime, only in so far as it is an object of fear. "We can, however, view an object as fearful without being afraid of it. This takes place when we merely think of a case in which we would resist, while recognising that all resistance would be utterly in vain. Thus the virtuous man fears God without being afraid of Him, because he thinks of resistance to God and His commands as what in no way concerns himself. But in every case where such resistance is thought as not impossible, man recognises God as fearful. He who is afraid is unable to judge Nature to be sublime, just as the man who is under the influence of inclination and appetite cannot regard an object as beautiful. He flees from the sight of an object which inspires him with terror, and there can be no satisfaction in what is felt as truly terrible. Hence the joy which arises when his perturbation ceases. 269 Bold, overhanging and as it were threatening cliffs, masses of cloud piled up in the heavens and alive with lightning and peals of thunder, volcanoes in all their destructive force. ANALYTIC OF THE SUBLIME 499 hurricanes bearing destruction in their path, the boundless ocean in the fury of a tempest, the lofty waterfall of a I mighty river; these by their tremendous force dwarf our power of resistance into insignificance. But we are all the more attracted by their aspect the more fearful they are, when we are in a state of security ; and we at once pronounce them sublime, because they call out unwonted strength of soul and reveal in us a power of resistance of an entirely different kind, which gives us courage to measure ourselves against the apparent omnipotence of nature. In the immensity of nature and the inadequacy of our faculty to find a standard proportionate to the aesthetic estimate of the magnitude of its realm, we discover our physical limitation, but at the same time we find in our rational faculties another standard, which as non-sensuous brings that infinity itself under it as a unity, in contrast to which everything in nature is small. Thus we become conscious that in our own mind we are superior to nature in all its immensity. So the irresistible power of nature, which shows us our physical impotence as natural beings, at the same time reveals in us a faculty of judgment which is independent of nature and superior to it. . . . Thus humanity in our own person remains unhumiliated, although the individual man must submit to the power of nature. In this way nature is in our aesthetic judgments regarded, 370 not as fearful, but as sublime. . . . We therefore call nature sublime, merely because it elevates the imagination and leads to the presentation of those eases in which the mind feels its own real sublimity and its superiority even to Nature. 272 There is therefore no sublimity in any natural object, but only in our own mind, in so far as we can become conscious that we are lifted above the nature within us, and therefore also above the nature without us, so far as it has an influence upon us. 2 1 2 500 CRITIQUE OF AESTHETIC JUDGMENT 29. Modality of the judgment in regard to the Sublime in Nature. There are numberless beautiful things in nature, in regard to which we can anticipate, without much liability to mistake, that every one will agree with us in our judg- ments ; but we cannot have the same assurance that aU will agree with us in our judgments as to the sublime. For here much greater culture is required, not only of the aesthetic judgment, but of the faculties of knowledge. 273 In order that the mind may be open to the feeling of the sublime, it must have a capacity for Ideas ; for, it is just the inadequacy of Nature to Ideas, and therefore only under presupposition of them, that imagination is put on the strain to use nature as a schema for Ideas. ... In fact, without development of moral Ideas that which culture prepares us to regard as sublime appears to the rude and uncultivated man as merely terrible. But, though the judgment as to the sublime in nature requires culture, much more than the beautiful, we must not suppose that it is entirely due to culture and the conventions of society ; it has its foundation in human nature, and indeed in that which we are entitled to expect from every one as well as a sound understanding ; in other words, it has its source in the tendency to the feeling for practical Ideas, i.e., in the tendency to morality. On this tendency is based the necessity that others should agree with us in our judgment about the sublime. Just as we charge a man with want of taste, who is indifferent to what we regard as beautiful in nature, so we 274 say that one who is unmoved in the presence of an object which we judge as sublime has no feeling. We demand both of every man, and presuppose that he has both, if he has any culture at all ; still we make a distinction between them ; we expect the former directly of every one, because here the imagination is related to the understanding as a ANALYTIC OF THE SUBLIME 501 faculty of conceptions; but in the case of the latter the imagination is related to reason. As the faculty of Ideas, we presuppose moral feeling as its subjective condition, but we believe ourselves entitled to attribute it to every one. Hence it is that we regard this aesthetic judgment also as necessary. Deduction of Puke Aesthetic Judgments. 30. The deduction of aesthetic jtidgments in regard to objects of nature is not required for the sublime, but only for the beautiful. The claim of an aesthetic judgment to universal validity for every subject, as a judgment which is based upon some a priori principle, stands in need of a deduction or justifica- tion of its pretensions ; and such a deduction must be added to the exposition, when judgment implies a satisfaction or dissatisfaction in the fo7-m of the object. Of this character are judgments of taste in regard to the beautiful in nature. For, the purposiveness has then its ground in the object and its figure, though it does not indicate the relation of the i object to other objects through conci'ptions in judgments of knowledge, but has to do only with the apprehension of this form, so far as it shows itself to be in harmony at once with the faculty of conceptions and the faculty of presenting these in the mind. Many questions may therefore be raised in regard to the beautiful in nature. But the sublime in nature, when we pass upon it a purely aesthetic judgment, which is kept free from conceptions of perfection or objective purposiveness — since this would convert it into a teleological judgment — may bo regarded as entirely without form or figure, and yet it may be the object of a pure satis- faction and exhibit a sulijective purpose. It may therefore be asked, whether in an aesthetic judgment of this kind, besides the exposition of what is implied in the thought of 502 CRITIQUE OF AESTHETIC JUDGMENT it, there is also required a deduction of its claim to be regarded as an a priori (subjective) principle. To this question the answer is, that to speak of the sublime in nature is not strictly correct, and that it is properly attributed only to a state of mind, or rather to the foundation of this state of mind in human nature. . . . Hence our exposition of judgments in regard to the sublime in nature was at the same time their deduction. 289 We shall therefore have to seek for a deduction only of judgments of taste, i.e., of judgments in regard to the beauty of natural things. 31. Method of deduction of judgments of taste. We are under obligation to give a deduction or justifica- tion of a class of judgments only when the judgment lays claim to necessity. This, however, applies also when it demands subjective universality or the agreement of every one, though it is not a judgment of knowledge but only of pleasure or pain in a given object. Now, a judgment of taste is not a judgment of knowledge, being neither theoretical as resting upon the conception of nature in general, nor practical as based upon the Idea of freedom given a priori by reason. We have therefore here neither to justify a priori the validity of a judgment expressive of what a thing is, nor one which prescribes what we ought to do in order to produce it ; all that is required of us is to exhibit the universal validity of an individual judgment, which expresses the adaptation to the subject of an empirical consciousness of the form of an object. This universal validity cannot be established by finding that others agree with us in our sensations, but must rest upon what may be called an autonomy of the judging subject as regards the feeling of pleasure in the given idea, i.e., upon his own taste ; while, on the other hand, it must not be derived from conceptions. A judgment of this sort, ANALYTIC OF THE SUBLIME 503 — such as the judgment of taste in fact is — has two peculiar 290 logical characteristics. In the first place, it is universally valid a priori, and yet its universality is not logical or based on conceptions, but is the universality of an individual judgment. And, secondly, it is necessary, and must there- fore rest upon a 'priori grounds, while yet its necessity cannot be proved a priori on grounds which every one who assents to the judgment of taste is forced to acknowledge. 296 36. The problem of a deduction of the judgment of taste. 297 This problem may be thus put : How is a judgment possible, by which, merely from our own feeling of pleasure in an object, independently of conception, we judge a priori that this pleasure attaches to the consciousness of the same object in every other subject, and that without waiting for any such agreement ? This problem of the Critique of Judgment comes under the general problem of Transcendental Philosophy : How are a priori synthetic judgments possible ? 298 37. What is properly asserted a priori of an object in a judgment of taste ? All judgments of taste are singular, because they do not connect their predicate of satisfaction with a conception, but only with a given individual empirical idea. Hence it is not the pleasure, but the universal validity of this pleasure which is perceived to be mentally boxmd up with the mere judgment of an object. It is this universal vahdity which in a judgment of taste is affirmed a priori to be a universal rule and valid for every one. 38. Deduction of the Judgment of taste. If it is admitted, that in a pure judgment of taste satis- faction in the object is connected with the mere judgment 504 CRITIQUE OF AESTHETIC JUDGMENT of its form, it is merely its adaptation for the judgment of the subject that we feel to be mentally combined with the consciousness of the object. Now, the faculty of judg- ment in its formal rules, being devoid of all matter either 299 in the way of sensation or of conception, can be directed only to the subjective conditions of the exercise of judgment in general, which are applied neither to a particular mode of sense nor to a particular conception of the understanding. Hence it is based upon subjective conditions which we have a right to presuppose in all men as essential to the possibility of knowledge. We can therefore assume the harmony of an idea with these conditions of judgment to be valid a priori for every one. In other words, we are justified in ascribing to every one the pleasure or subjective adaptation of the idea to the relation between the faculties of knowledge in the judgment of a sensible object in general. Note. This deduction is so easy because it has no need to justify the objective validity of a conception ; for beauty is not a conception of the object, nor is the judgment of taste a judgment of knowledge. It merely asserts, that we are justified in presupposing that the same subjective conditions of judgment are found in others as we find in ourselves, and, further, that we have correctly subsumed the given object under those conditions. No doubt this subsumption has inevitable difficulties, which do not attach to the logical judgment; for, in the latter we work with con- ceptions, whereas in the aesthetic judgment we have to do with a relation of imagination and understanding which can 300 only be felt. But, though we may easily make a mistake in our subsumption, this does not destroy the rightful claim of the judgment to count upon universal agreement, a claim which merely asserts, that it is correct in principle to judge as valid for every one what is based upon subjective grounds. DIALECTIC OF AESTHETIC JUDGMENT 505 Dialectic of Aesthetic Judgment. 56. Antinomy of Taste. The following antinomy arises in connexion with the principle of taste : (1) Thesis. The judgment of taste is not based upon conceptions ; for if it were, it would lead to controversy and would admit of proof. (2) Antithesis. The judgment of taste rests upon concep- tions; for if not, notwithstanding its diversity, there would be no controversy, as there is when we demand that others must necessarily agree with us in our judgment. 57. Solution of the Antinomy of Taste. The only possible way in which the conflict between the principles which underlie every judgment of taste — those principles being nothing but the two characteristics of the judgment of taste explained in the Analytic — can be brought to an end, is by showing that the conception to which the object is referred in this class of judgments is taken in different senses. This double sense or point of view is necessary to our transcendental faculty of judgment, but so also is the natural illusion which inevitably arises from the confusion of the one with the other. The judgment of taste must refer to some sort of concep- tion, for otherwise it could make no claim whatever to necessary validity for every one. But it does not follow that it can be proved from a conception ; for a conception may be either determinable, or in itself at once undetermined and undeterminable. Of the former kind are the conceptions of the understanding, which are determinable by predicates of sensible perception corresponding to them ; but the tran- scendental conception of reason, the Idea of the supersensible, which lies at the basis of all sensible perception, is of the 506 CRITIQUE OF AESTHETIC JUDGMENT latter kind and does not admit of further theoretical deter- mination. Now, the judgment of taste is applied to objects of sense, but not with a view to the determination of a conception of them for the understanding; for it is not a judgment of 351 knowledge. Being therefore an individual idea, which is related perceptively to the feeling of pleasure, it is merely a private judgment. So far its authority is limited to the judging individual ; the object is for me an object of satis- faction, no matter what it may be for others : every one has his own taste. Nevertheless it is undeniable that there is implied in a judgment of taste a wider relation of the idea of the object, as well as of the subject, and upon this relation is based the necessary extension of this class of judgments to every one. Such judgments necessarily presuppose some sort of conception, but it is a conception that cannot be determined by perception. Hence no object can be known through it, nor can any proof of the judgment of taste be based upon it. Such a conception is the pure Idea of the supersensible, which underlies the phenomenal object of sense as well as the judging subject. Thus all contradiction disappears from the statement that the judgment of taste is based upon a conception, — the con- ception of a general ground for the subjective adaptation of nature to our faculty of judgment — from which nothing can be known or proved in regard to the object, because it is in itself undeterminable and useless for knowledge. At the same time the judgment has validity for every one, though of course only as a singular judgment which directly accompanies perception ; for its determining ground may well lie in the conception of that which is regarded as the supersensible substrate of humanity. 352 The solution of the Antinomy consists in seeing that while, in the two contradictory judgments, the conception upon which the universal validity of a judgment is based is DIALECTIC OF AESTHETIC JUDGMENT 507 taken in the same sense, two opposite predicates are applied to it. The thesis properly means that the judgment of taste is not based upon definite conceptions, the antithesis that the judgment of taste is based upon an indefinite conception, viz., the conception of the supersensible substrate of pheno- mena. There is therefore no real contradiction between the two. Thus the two apparently contradictory principles are reconciled with each other : hoth may he true, and nothing more is needed. If on the other hand we adopt the view that, as it presupposes an individual idea, the judgment of taste must be based upon pleasure, or if we hold that as claiming universal validity it is based upon the principle of perfection, and frame our definition of taste accordingly, there arises an antinomy which is absolutely insoluble. For both propositions will then be false, being not merely contradictory but contrary, and hence the conception on which they are based must be self -contradictory. . . . Here, therefore, as in the Critique of Practical Eeason, the antinomies force us against our will to look beyond the sensible, and to seek in the supersensible for the point of union for all our a priori faculties. There is in fact no other expedient by which the harmony of reason with itself may be secured. INDEX Aesthetic, Transcendental, 73, Aesthetic distinguished from Logic, 79. Aesthetic deals only with space and time, 99. Aesthetic implies contrast of phenomenon and thing in itself, 100. Aesthetic, Summary of, 105. Affirmation and Negation, 286. Analogies of Experience, 189. Analogies of Experience : Sub- stantiality, 195. Analogies of Experience: Causality, 200. Analogies of Experience : Com- munity, 208. ' Analogies of Exxjerience, General character of, 212. Analysis perhaps main work of reason, 56. Analysis presupposes synthesis, 125, 148. Analytic judgments contrasted with synthetic, 56. Antinomy of pure reason, 2.>0. Antinomy of pure reason : First Antinomy, 254. Antinomy of pure reason : Second Antinomy, 2.j5. Antinomy of pure reason ; Third Antinomy, 257. Antinomy of pure reason : Fourth Antinomy, 258, Antinomies of pure reason, .Solution of First and Second, 259, 261, 266, 268. Antinomies of pure reason, Transi- \ tion from mathemati'^al to \ dynamical, 269. Antinomy of pure reason, Solution of Third and Fourth, 272, 280. Antinomy of pure reason. Contrast of, and Ideal, 283. Antinomy of practical reason, 376. Antinomy of Taste, 440, 505. Antinomy of teleological judgment, 453. Apperception, Transcendental unity of, 145, 149. A priori and a posteriori, 49, 78. Archetypal and ectypal intelli- gence, 465. Asbooiation does not explain Causality, .52. Association presupposes reproduc- tion, 139. Autonomy of will, 328, 351. Axioms of perception, 180. Bacon rejects preconceptions, 4. Beautiful an oVjject of aesthetic judgment, 408, 474. Beautiful comjjared with the pleasant and the good, 412, 477. Beautiful regarded as the object of unis'ersal satisfaction, 413, 478. Berkeley and I/Ocke, 29. Categories, Guiding-thread to, 119. Categories imply synthesis, 124. Categories correspond to logical functions, 12(J. Categories, Table of, 128. Categories distinguished as mathe- matical and dynamical, 131. Categories, Third of each class of, the union of the other two, 132. Categories, Deduction of, 133, 136, 147. INDEX 509 CategorieB, I'l'icoptiona stand under, 157. Ctttegories apply only to experience, 167, 169, 162. Categories, Result of Deduction of, 166. Categories, Short Deduction of, 167. Ctttogorios of Modality, 213. Categories i ctliiotoil to external perception, 217. Caviaality in Hume and Kant, 40. Ciiiisttlity a pure a priori judgment, f.l. tlausality synthetic, 61. t'liusality. Category of, 129. Causality, Sfhoina of, 17(). Causality, I'root of, 2(X). Causality implies motion, 218. Causality, Aniinomy of, 2r)7, 272. Cause, I'irst and .Supriino, 291. Cause, Final, in physioo-tliuotogioal argument, 304. Cause, Final, Rational lioiiig us, :t20. Cause, Final, a principle of judg- ment, 399. Cause, Final, Aesthetic consciousness of, 40.f. Cause, Final, Lngiual idea of, 404. Cause, Final, as formally objective, 444. Causo, Relative and internal, 44."i. Cause, Final, in nature, 44(i. Chiuige conditioned by time, 94. Change consi.stent with ideality of time, 97. (Change empirical, lOH. ('I\ange continuous, 1S(1. Change relative to the peinianeiit, 197. Change and the causal, 200. Community, Oatei;ory of, 129. Community, .Sehema of, 176. Coumiunity, I'roof of, 20S. Cmumunity dynamical and spatial, 210. Community presupposes spaee, 2)9. Coueeptiou an element in all know- ledge, 107. Conception a function of under- stfindinj.', 121. Coueo|itieu limitative, 2'J3. See Calfijorif.i and Idi'n.i. Conslitutive prineiples, 191. Constitutive luinoiples impossible for reason, 2li5. Continuity of magnitudes, 186. Copernicus, the type of the critical philosopher, 37. Critical philosophy in relation to previous thought, 1. Criticalphilosophy demands scrutiny of all Doliefs, 3. Critical philosophy. Problem of, 31. (Critical philosophy in relation to llnme, 32, 40. Critical philosophy contrasted with Dogmatism and Scepticism, 32. Critieal philosophy and the special sciences, 3.^). Critieal philosophy, Method of, 38. Critieal philosophy. Object of, 54. Critical philosophy. Idea aud Divi- sion of, 68. Critical philosophy architectonic, 70. Critical philosophy, Proof of, 193. Deduction of the Categories, 133. Deduction of principles of practical reason, 352. See Citte(jories aud Priiiciplex. Degree the Schema of Quality, 174. Demonstration limited to mathe- matics, 193. Descartes, Kant's relation to, 1. Descartes, Critieal estimate of, 3. I1esinn, Argument from, 303. Desire, Natural, not an object of I reveienee, 312. Deslie contrary to duty, 314. I )esire not a moral motive, 364. Diatectie, Transeeiuleutal, 117, 2.'!0. Dialectic of practical reason, 118. liialecticof aesthetic judgment, 440, ."i0.'>. Dialectic of telcologieal judgment, 4.'>:i. l^i.ssirlalioii. Contents of Kaiit'.s, 27. Piiu'inatisni in relation to Oiiticism, 32. Duty a pure Ideo, 276. Duty, Analysis of, 311. Duty, Idea of, not derived from I'xperienee, lil.'t. Duty implies a will not absolutely good, 31.S, 350. Duty expressed in a categorical imperative. 319. Duty opposeil to self-interest, 364. Dynnniieal categories, ).'!!. filO INDKX DyiiiLinlciLl jn'incipltwi, 19*2. I >y iiainii'Hl luiil iiiiiMii'inat ii^al I'Ogl'i'xM, •2S1. Kiii|iiii<'Jil ri'.'ilily of H|)ni'i>, Itl. ICnipii'liiiil ri'iilll V "I liiiio, llli. li'.iil|iirii'alilc'iliii',l ,uiHil'(,lliM'iili'u;i)ili'n, Knipiiiiial aiiiior.'i'pliiin, Ifld, \M. l<',iiipiiii'M,l cnaiMili'i' >>f mail, '27^. IOxImI^i'iiiM', iSolu'iiia of, 177. Kxpcu'ii'il()in )iiiii( adaplal inirolil()in of lOIIMOll, rt4. (!oil iiol I'onilidoiiiMl liy ripiuw ami finii', 10','. (>oil iM>niM>i\oil to ('oii(.fiin all t'<*ality, 2S(1. (loil tlio oliji'iit of trannO('ml«nl;al (lu'oloHy. 'Jus. V (loil, .Spi.,.n|alivn pi'oofn of I'xintnnoi! of, •2\M. (lo.l, Oiil.olopji.al iiroof of, '2II>2. (loil, ('oriinoionii'lll [ll'oof of, '2l)7. (loil, 1'liyHH^o llioolo^jcal pi'oof of, :i(i;i. i an. I Will. llappinoMM tho I'oniplott' Matiiifaetion of .j.^niro, ;ll I. irappinnKri an annoi'loriii liypotlni tuial liiipi'iallvii, MIO, HappincHFi iinlolliialtlo, .'I'.'O. llaptiiiiMiM oinnproliondn all material prlin'iplon, ^(1 1 . llappininn lin])lliiil in Htimttinm, hiiiiiiiii, 37f>. H»il,('roninriv of will, 't5'2. 1 1 iiino aiiil Kanf, I . lliinio, (IritiiMil lu'i'oniif of l-lio]iliilo Hopliy of, '211. llninii, Caiinality in, 10. Moan, 'rraiiru'i'nilon(. Imperatives, Categroricai. 320, S34. Induction not the b:V5i5 of universal jcdgmenis, 50, 21.4. Ia;eUectual perception proper to Original Being. 104. Intellectaal perception. Appercep- tion not the principle of, 151. Intellectual perception indefinable, 159, 224. Intellectnal perception implies an archetypal intelligence, 463. Judgments, A pf-icri, 49. Judgments anidrtio or syniherie. 56. ■Jcdimsent the essence of tioniihr, 12~1. Judgment, Transcendenta'.. lf>>. Jndsment, Type of pure practical, 362. Judgment connects theoretical and practical philosov by, 396. 'Judtrment dereru^inant or redective, 397. Judgment, Aesthetic, 4C'S. 474. Judgment, Aesthetic. Deduction of, 4^,501. Judgment, Aesthetic, Dialectic of, 440, 505. Jr.d:nnent, Te'eolcdcal, 4-it4. Jtidcment, Te!eolocical. Dialectic of, 453. See Tmit, I>-a'.t:i/'.iI and S biioie. Knowledge begins with experience, 46. KLnowledge a priori or n posteriori, 49. Knowledge pure or muted, 50, 51 n. Knowledge, Criteria of, 50. Knowledge. Transcendental, 6S. Knutzen and Kant, 26. Leibnitz, Kant's relation to, 1. Leibnitz, Critical account of philo- sophy of, 14. Leibnitz. Conception of reason in, 42. Locke. Kant's relation to, 1. Locke, Criticjdacoountof philosophy of. 2>. Logic and Aesthetic, 79. Logic, Divisions of. 117. Luther and the Reformation, 2. Magnitade, Images of. 174. Magnitude, Perception an extensive, iki. Magnitude, Continnity of, 1S6. iiagnitude, Keiation of extensive and intensive, IS^. Man, Intelligible and empirical character of, 277. Materi.^lism unprovable, 24S. Mathematical categories. 131. Matheriarical principles, 190. Mathemaiical antinomies. 251. Mathematics, Method of, 36, 193. Mathematics. Judgments of, syn- thetic, 62. Mathematics, Pure, how possible ? 66. Matinematics applicable to all phenomena, 1S2. Mathematics, Figures of, display formal c"L';ec:ive purpose, 4-14. Matter, ladestraciibibty of, 64. Matter of sense. 76. Matter of desire, 340. Mechanism compatible with free- dom, 4o. Mechanism not inconsistent with te.eolojy. 463. Metaphysie, False method in, 32. Metaphysic and the supersensible, 41. Met.iphysic a search for a priori synthetic judgn-.ents, 65. Me-.iphysic, Natural disposition to, o7. Metaphysic, Science of, possible. 512 INDEX Metaphysic of Morality, 308. Misology, 311. Modality of judgments, 123. Modality, Categories of, 130, 213. Momentum implies degree, 185. Morality compatible with Mechan- ism, 43. Morality, Principles of, not part of Transcendental Philosophy, 71. Morality, Metaphysic of, 308, 325. Morality rests upon idea of duty, 311. Morality limited to rational beings, 313. Morality distinguished from prud- ence, 320. Morality a law for all rational beings, 325 Morality implies kingdom of ends, 329. Morality, Laws of, determine action by their form, 329. Motion made conceivable by time, 94. Motion an empirical conception, 99. Motives of pure practical reason, 364. Natural theology assumes subjec- tivity of space and time, 102. Necessary Being as condition of phenomena, 281. Necessity a criterion of a priori knowledge, 50. Necessity, Category of, 130. Necessity, Schema of, 177. Necessity a postulate of empirical thought, 216. Negation, Category of, 129. Negation involves limitation, 286. Non-existence, Category of, 129. Non-existence, Schema of, 177. Noumena contrasted with pheno- mena, 220. Noumena the idea of a limit, 223. Object, Meaning of, 143. Ontology not a science, 222. Organism implies internal purpose, 448. Paralogism of rational psychology, 241. Perception either pure or empirical, 74. Perception an element in know- ledge, 107. Perception as object an intensive magnitude, 180. See Intellectual Perception. Phenomenon defined, 75. Phenomenon implies time, 96. Phenomenon in a sense real, 100. Phenomenon determinable by mathe- matics, 182. Phenomenon possesses degree, 183. Phenomenon a continous magnitude, 186. Phenomenon contrasted with nou- menon, 220, 372. Phenomenon, Man viewed as, 275. Philosophy, The three problems of, 1. Philosophy divided into Theoretical and Practical, 391. See Gritical Philosophy. Physico-theologioal argument, 303. Physics, Method of, 36. Physics contains a priori synthetic judgments, 64. Physics, Kant's Rudiments of, 65. Physics, Pure, how possible 1 66. Plato's contrast of Opinion and Knowledge, 19. Plato's Ideas, 56. Pleasant, Satisfaction in, interested 410, 475. Pleasant compared with the good and the beautiful, 412, 477. Pleasure relative to desire, 340. Pleasure, No qualitative differences in, 342. Pleasure contrasted with the good, 361. Plurality, Category of, 128. Possibility, Category of, 130. Possibility, Schema of, 177. Possibility a postulate of empirical thought, 214. Possibility has no meaning for per- ceptive intelligence, 456. Postulates of empirical thought, 213. Postulates of pure practical reason, 384. Principles of understanding, 179. Principles, Mathematical and Dyn- amical, 190. Principles, Immanent and Tran- scendent, 228, 234. Principles of reason, 230. INDEX 513 Principles, Regulative, 265. Principles, Practical, 339. Principles, Material, 341. Prudence, Precepts of, hypothetical, 319. Psychology distinct from Logic, 111. Psychology, Rational, Paralogisms of, 240. Psychology, Rational, a discipline, 248. Purpose. See Teleology. Quality of judgments, 123. Quality, Categories of, 129, 131. Quality, Schemata of, 174. Quantity of judgments, 122. Quantity, Categories of, 128. 1 Quantity, Schema of, 174. Quantityof Substance unchangeable, I 199. Quantity implies external percep- ' tion, 220. [ 1, Reality, Empirical, of space, 91. Reality, Empirical, of time, 96. ■ Eeality, Category of, 129. Reality, Schema of, 174. Reality, Sum-total of, 286. Reason an organic unity, 35, 259. Reason, Problem of, 66. [ Reason the faculty of principles, I 230. Reason, Principles of, derived from form of Syllogism, 232. Reason, Dialectical conclusions of, 239. Reason, Paralogisms of, 241. Reason, Antinomies of, 250. Reason, Antinomies of. Critical solution of, 261. Reason, Speculative and Practical, 335. Reason, Practical, Critique of, 339. Reason, Practical, source of moral law, 350. Reason, Practical, Object of, 360. Reason, Practical, Motives of, 364. Reason, Practical, Extension of, 359, 386. Reciprocity, Category of, 129. Reciprocity, Schema of, 176. Reciprocity, Proof of, 208. Recognition, Synthesis of, 143. Reformation, Problem of the, 2. Refutation of Idealism, Kant's, 95, 98, 217. Regulative principles of under- standing, 191. Regulative principles of reason, 265. Regulative principles in regard to Necessary Being, 282. Regulative principles in regard to Supreme Being, 302. Relation, Function of, in judgments, 123. Relation, Categories of, 129. Relation, Schemata of, 175. Renaissance, Problem of the, 2. Reproduction, Synthesis of, 139. Reverence for moral law, 312. Satisfaction in judgment of Taste disinterested, 409, 475. Satisfaction in pleasant interested, 410, 475. Satisfaction in good interested, 411, 477. Satisfaction, universal. Beauty the object of, 413, 478. Schemata of the Categories, 169. Schemata contrasted with images, 171. Schemata the object of mathe- matics, 172. Schemata, Classification of, 174. Schemata restrict the categories, 178. Schemata distinct from type, 362. Scepticism of Hume, .32. Scepticism, Dogmatism and Criti- cism, 32, 67. Sciences, Experimental, Method of, 36. Self-consciousness. See Appercep- tion and Paralogism. Self-love, Material practical prin- ciples reducible to, 341. Sensation, definition of. Ambiguity in, 75. Sensation, inner, Quantum of, 220. Sensibility and tjnderstanding, 71. Sensibility, External and internal, 80. Sensibility receptive, 110. Sensibility, Degree of, 183. Soul regarded as the object of Rational Psychology, 241. Space an object of Trauscendental Aesthetic, 78. 514 INDEX Space, Metaphysical Exposition of, 80. Space, Transcendental Exposition of, 88. Space not a thing, property or relation, 90. Space the form of outer sense, 90. Space empirically real, transcen- dentally ideal, 91. Space the condition of a priori synthetic judgments, 97. Space not a form of divine percep- tion, 102. Space the image of external magni- tudes, 174:. Space a quantum continuum, 186. Space, Empty, not knowable, 187. Space, Transcendental Ideality of, contirmed by critical solution of antinomies, 264. Spinoza, Kant's relation to, 1. Spinoza, Critical account of philo- sophy of, 5. Subject, Thinking, not Substance, 244. Sublime an object of aesthetic judgment, 404. Sublime contrasted with Beautiful, 424, 490. Sublime, Mathematically, 428, 494. Sublime, Mathematically, Quality of satisfaction in, 432, 497. Sublime, Dynamically, 433, 498. Sublime, Dynamically, Capacity for Ideas presupposed in, 435, 500. Substance, Category of, 129. Substance, Schema of, 175. Substance, Proof of, 195. Substance implies dynamical com- munity, 208. Substance presupposes apprehension of matter, 218. Substance contrasted with thinking subject, 244. Summum bonum defined, 375. Synopsis implies synthesis, 137. Synthesis an operation of imagina- tion and thought, 125. Synthesis of apprehension, 138, 163. Synthesis of reproduction, 139. Synthesis of recognition, 143. Synthesis essential to knowledge, 147. Synthesis, Figura! and intellectual, 160. Synthetic judgments distinguished *,| from Analytic, 56. '*• Synthetic judgments. Some, rest on *,; experience, 59. -j Synthetic judgments coutained in ""j mathematics, 62. '"■ Synthetic judgments contained in * physics, 64. ;' Synthetic judgments the object of •' search in metaphysics, 65, 234. Jj Synthetic judgments, Examples of, ' from arithmetic and geometry, 63. ' Synthetic judgments. Conditions of, - 66. -; Synthetic judgments based on space -^ and time, 98. -' Synthetic judgments implied in "• principles of reason, 234. Taste, Judgment of, aesthetic, 403, 408. - Taste, Judgment of, disinterested, ' 409, 475. I Taste, Judgment of, singular, 413, '■■ 479. - Taste, Judgment of, based on ■ purposive form of object, 416, 483. ' Taste, Judgment of, a priori, 417, 483. ' Taste, Judgment of, independent of charm and emotion, and of the ' conception of perfection, 418, 485. Taste, Judgment of, conditionally necessary, 420. Taste, Judgment of, rests on Com- mon Sense, 421, 488. Taste, Judgment of, General remarks on, 422, 489. Teleology of physico-theological argument, 303. Teleology in nature, 399. Teleology a maxim of judgment, 400. Teleology implied iu organised beings, 448. Teleology as a system, 450. Teleology, Relation of, to mechanism of nature, 453. Thales perhaps the discoverer of method in mathematics, 36. Theology, Natural, Conception of God iu, 102. Theology, Rational, Object of, 238, 241. Theolo£-, Rational, Criticism of, 306. INDEX 515 Theology, Moral, 467. Thing in itself, 90, 95. See NouTnenon. Thought one of the two stems of knowledge, 71. Thought a spontaneity, 110. See Onderstajuling and Reason. Time an object of Transcendental Aesthetic, 79. Time, Metaphysical Exposition of, 91. Time, Transcendental Exposition of, 93. Time not a thing or determination of a thing, 95. Time directly the form of Inner sense, indirectly of all pheno- mena, 95. Time empirically real, transcen- dentally ideal, 96. Time : objection of Lambert an- swered, 96. Time a source of a priori synthetic judgments, 98. Time not a form of divine percep- tion, 102. Time homogeneous both with cate- gory and object of sense, 17'J. Time the image of all magnitudes, 174. Time an element In all schemata, 177. Time a quantum continuum, 186. Time, Empty, unknowable, 187. Time, Modi of, 189. Time, Transcendental ideality of, confirmed by critical solution of Antinomies, 264. Torricelli a discoverer of scientific method, .36. Totality, Category of, 1'2S. Transcendental knowledge, 68. Transcendental criticism, 70. Transcendental Aesthetic, 72. Transcendental Logic, 106. Transcendental Analytic, 118. Transcendental Deduction of the Categories, 133. Transcendental Judgment, 168. Transcendental Schematism, 169. Transcendental Principles, 179. Transcendental Ideas, 235. Type of pure practical judgment 362. Unconditioned not a self-contra dictory idea, 41. Unconditioned the object of reason 233. Understanding, Logical use of, 121 Understanding, Functions of, 122. Understanding essential to know ledge, 205. Understanding and Reason, 23 1 . See Thought. Unity, Category of, 128. Unity implied in all the categories, 14S. the only absolute only for Will, A good, good, 310. Will, A good, possible rational beings, 313. Will, Divine, not subject to im- peratives, 318. Will, Autonomy of, 351. Will as desire, 392. Wolff, Kant's relation to, 1. Wolff, Critical account of philosophy of, 22. GLASGOW : PKINTED AT THE UNIVERSITY PRESS BY BOBEP.T MACLEHOSE ANIJ CO. LTD. WORKS BY PROFESSOR JOHN WATSON THE PHILOSOPHY OF KANT. As contained in Extracts from his own Writings. Selected and Trans- lated by John Watson, LL.D. Sixth Edition, Crown 8vo. 7s. 6d. THE PHILOSOPHICAL BASIS OF RELIGION. Post 8vo. 8s. 6d. nett. AN OUTLINE OF PHILOSOPHY. Fourth Edition, revised and greatly enlarged. Crown 8vo. 7s. 6d. nett. HEDONISTIC THEORIES, FROM ARISTIPPUS TO SPENCER. Crown Bvo. 6s. nett. GLASGOW : JAMES MACLEHOSE AND SONS Publishers to the University London and New York : MACMILLAN & CO. LTD. '43flp|pi