UNIVERSITY OF CALIFORNIA. SAN DIEGO 3 1822 00850 7352 Y l^ ' --^ - ^ s :iili S^ 1105'a central University Library University ol Calilornia, San Diego Note: This item is subject to recall after two weeks. Date Due C/ 39 (1/91) UCSDUb. irjIVI MM I ■ I il uAl II DIUIIA ',AIi nil CO 3 1822 00850 7352 LECTUEES ON THE PHILOSOPHY OF KANT AND OTHER PHILOSOPHICAL LECTURES & ESSAYS LECTUEES ON THE PHILOSOPHY OF KANT AND OTHER PHILOSOPHICAL LECTUEES & ESSAYS BY THE LATE HENRY SIDGWICK KNIOHTBRIDOE PROFESSOR OF MORAL PlrfLOSOPHY IN THE UNIVERSITY OF CAMBRIDGE ILantian MACMILLAN AND CO., Limited NEW YORK: THE MACMILLAN COMPANY 1905 KRAUS REPRINT CO. New York 1968 L.C. Catalog Card Number 5-35670. KRAUS REPRINT CO. A U.S. Division of Kraus-Thomson Organization Limited Printed in U.S.A. EDITORIAL NOTE The first portion of this volume consists of lectures given by Professor Sidgwick as part of a long course on Metaphysics, which he delivered for the last time in the academic year 1899-1900. It was his intention eventually to work up these lectures into a book on Kant and Kantism in England. The gap between the lectures on Kant and those on Green and Spencer was to have been filled up with a sketch of the influ- ence of post-Kantian philosophy on English thought. But the two fragments, placed one as appendix to the last lecture on Kant, and the other as ' introductory ' to the lectures on Spencer, are all that seemed now available of the material prepared for this sketch. The lectures on Kant, the author felt, were left " toler- ably complete," but " the study of Green " he knew was " not in the form required for a book." Appended to it is the chief part of a lecture — the last he ever gave — on Green's philosophy, which the author thought " might be somehow combined with the lectures " as here printed. And no doubt it may be, but the editor is of opinion that most readers will ▼I riiK riiii.usorHv of kant prolVr to ilo the i'(»inl»iiiiiii; tlu'iusolvcs. This decisioii to mi'diUe as little as possible Nvilli what the author hixs left us lias also entailed the retention of sundry repetitions which he would doubtless have removed (ef. v.(j. pp. 235, 244). The second portion of the volume consists of articles, all but the first of which have — with the editor's permission — been reprinted from Mind. The first, on the Sophists, from the Journal of Philology,^ has been inserted, though incomplete, on the advice of Dr. Henry Jackson, who has kindly undertaken its revision for the press. A small portion of the last article, that on " Criteria of Truth and Error," occurs also in the lectures on Spencer (cf. pp. 318, 456) ; and as already stated in the editorial note to the author's Philosophy, its Scope and Relations, a few passages from the same article are reproduced there. This article too was left unfinished, but there is now appended to it portions of two lectures which show the lines on which the author intended to com- plete it. These lectures were themselves an amplifi- cation of a paper read to the Metaphysical Society and afterwards published in the Contemporary Review (July 1871). Passages and references in square brackets, other than those occurring in quotations, are editorial additions. ^ Published by Messrs. Macmillan and Co. EDITOEIAL NOTE vii The Index has been kindly prepared by Miss E. E. C. Jones, the Mistress of Girton College, and the proof sheets have been carefully revised by Mrs. Sidgvvick. JAMES WAKD. Trinity College, Cambridge, April 23, 1905. ERRATA Page 4, line 3 from foot, delete (1). 64, ,, 6, /iw principle rearf principles. 67, ,, 15, /or context r«a«? content. 121, ,, 15, for scheme read schema. 188, ,, 3, add a comma after derivative. 202, ,, 4, delete comma after as God. 231, ,, 9, /or relation 7-(;a«J relations. 286, lines 9 and 10 from foot,/(W "obvious to me' read ' obvious ' to me. 331, Hue 5, add to al end of line. 372, ,, 3 of title, /or vol. iii. read vol. vii. PAOE CONTENTS LECTURES The Metaphysics of Kant — 1. The Critical Standpoint .... 1 2. The Transcendental Esthetic . . .21 3. Kant's ' Expositions ' of Space and Time . .38 4. The Transcendental Analytic . . .58 5. The Mathematical Categories and Principles . .75 6. Substance . . . . . .98 7. Causality, Community, Modality . . .106 8. The Transcendental Dialectic . . .128 9. Rational Psychology . . . .143 10. The Mathematical Antinomies . . .152 11. The Dynamical Antinomies . . . .162 12. Rational Theology . . . . .179 Appendix to Lecture 12 : The Unconditioned . 196 ix X THK ruiLosornv of kant Thk Mkiaiiivsh s ov T. 11. (Iukkn — I-AIIK 1. Siuniujiry Atc-ount ..... 'Jl)9 •J. Tlu' Sj.iritu;il I'riiicipK' in Knowlod^c and in Nature 222 3. The l\elation o{ Man to the Spiritual Principle in Niituro .238 Appendix to these Lectures .... 257 The Philosophy of Mr. Herbert Spencer — Introductory : Agnosticism and Relativism 267 1. Metaphysical Doctrines . . . .275 2. Metaphysical and Epistemological Doctrines . 302 ESSAYS 1. The Sophists. (Two papers from The Journal of Philohgy, 1872 and 1873) . . .323 2. Incoherence of Empirical Philosophy (from Mind, vol. vii. O.S., October 1882) .372 3. A Dialogue on Time and Common Sense (from Mind, vol. iii. N.S., October 1894) . . .392 4. The Philosophy of Common Sense (from Mind, vol iv. N.S., April 1895) 5. Criteria of Truth and Error (from Mind, vol. ix N.S., January 1900) Appendix to the preceding Essay INDEX ..... 406 430 461 469 THE METAPHYSICS OF KANT LECTURE I THE CRITICAL STANDPOINT Kant is selected by me as a philosopher to study, not merely on account of his historical importance — that is a consideration for another department of study, undertaken by another teacher ^ — but because it is partly at least to Kant that we trace the origin of the systems of metaphysical thought which have most vogue at the present day — the Agnosticism of Spencer (though here the influence is indirect, through Hamilton and Mansel), and more directly the Idealism or Spiritualism of which I take Green as a representative.^ ^ And, I may add, if that were the sole reason, it would be an instance of the irony of fate that Kant should be studied on that ground. Cf. Pro- legomena, Mahaffy's Trans, pp. 1 f. [References throughout to this edition.] 2 However, I may support my selection by a reference to the space given to Kant in current histories. You will observe that Falckeuberg gives Kant much the largest space that he gives to any one thinker in the whole history of modern thought ; and, if you suggest that this is due to German patriotism, I point out that Falckenberg allots to Kant nearly three times the space that he allots to any other German philosopher. And I point out that in other cases Falckenberg's preference for Germans is kept within bounds : since he gives Locke a somewhat larger space than either Leibniz, Fichte, Schelling, or Hegel. 1 B 2 rilK MKrATUVSlCS OF KANT lkct. But wiuil tioatise of Kant's shall we study V His groat troatiso.the one to which his influence is mainly ilue, the Critiipw of Pure Jx'casou (1781), or the Frolegoynciia to any Future Mctaphysic, written two years later ? The aim of the latter book (as he explains, p. 10) is to remove a " certain obscurity arising partly from the extent of the plan " of the earlier work, which rendered it difficult " to gather into one view the principal points of the investigation." This difficulty is no doubt diminished by Professor Watson's abridgment.' But if we want to learn what a philosopher is driving at, no one can tell us quite as well as the philosopher himself; and I often think that if every eminent thinker who has wTitten an epoch-making work had also written a supplementary one to explain what he aimed at doing, and what he believes himself to have done, in the first, — there would be fewer unsettled questions in the history of philosophy than is actually the case. I cannot, however, take the Prolegoinena (intelligently trans- lated — though not with perfect accuracy — by the versatile Professor Mahaffy) as the primary text- book of this course, because it presupposes the earlier work too much ; but I shall endeavour, so far as I can, to make the lectures suitable both to those who have read the CHtique of Pure Reason (either as Kant wrote it or as abridged by Watson) and to those who have read the Prolegomena. I shall have ^ The Philosophy of Kant, as contained in Extracts from his ovm Writings (1888). I THE CEITICAL STANDPOINT 3 to refer to this for certain important parts of the argument, and shall point out the passages that should be especially read along with the Critique. What, then, is briefly Kant's aim ? It is clearly stated in the Prolegomena (but not quite rightly translated by Professor Mahaff'y) : " My aim is to convince all who find it worth while to busy them- selves with metaphysics, that it is indispensably necessary for them to suspend their business for the present, and start with the question ' Whether such a thing as Metaphysic is at all possible ? '" ^ What, then, is the answer to the question, and are the metaphysicians allowed to resume their business ? Well, this answer properly and logically comes at the end of the book. But as there are some who seem to me slightly to misunderstand Kant's attitude to Metaphysics, I will presently give you my view of his verdict before we examine the arguments in detail. But, first, there is a prior question on which we may profitably spend a few minutes. Why suspend metaphysicians in particular from their business, among all the groups of persons engaged in the pursuit of truth ? The human mind has a moral preference for equality of treatment. Why not suspend Mathematicians and Scientists also, and have a general closing of intellectual workshops, until this prior question as to the possibility of producing the commodity offered has been tried with regard to all branches of what is currently taught as * Cf. p. 37, where — the question having become more definite — "all meta- physicians are solemnly and legitimately suspended from their occupations " till they have answered it. 4 THK MKTArHVSU'S OF KANT lect. knowledixc / 'io this question Kant's answer is simple, and I think ck\ir. (I) Metaphysics has not tlie eharaeteristics by which a (Science is known. It has not been able to obtain "universal and permanent approval." (•_') " Every other science is continually advancing, while in this, notwithstanding its high pretensions, we perpetually revolve round the same point without gaining a step." ^ On the other hand, as regards ' pure mathematic and pure physical science ' ^ " we can say with certainty " that these parts of pro- fessed knowledge ** are actual and given." What then is meant as ' given ' ? Kant answers that both contain propositions which obtain thoroughgoing recognition as apodictically certain : (a) partly by mere reason, (6) partly "by general consent arising from experience and yet as independent of experience."^ Mathematics and Physics, then, stand in no need of criticism ; and the only reason for this, as it seems to me, is that they have the consensus and steady progress which Metaphysics lacks. This is not, indeed, the only reason that Kant gives. In fact, in another passage (§ 40, p. 114) he seems to give only other reasons: viz. (l) that Mathematics "rests on its own evidence " and (2) Physical Science on the confirmation of experience. But neither of these reasons is really available. For (1) Metaphysics, in the view of the dogmatic metaphysicians whom Kant criticises, rested on its own evidence ; and it is ' Prolegomena, pp. 2 f. Observe that ' Science ' is used for any Systematic Knowledge, not as I used it in Philosophy : its Scope and Helations, pp. 2 f. ' By ' pure ' Kant means what is a priori in these sciences. * Prolegomena, § 4, p. 32. I THE CRITICAL STANDPOINT 5 only to a mathematician that Mathematics rests on this. We cannot therefore make this characteristic a difference between the two that necessitates a critical inquiry in the latter case which does not exist in the former. The real difference is the con- sensus in the former case, the uncontested condition of the evidence in contrast with the absence of "universal and permanent recognition" in the latter. The case is different with Physics. Here the basis is said to be " experience and its thoroughgoing confirmation," and certainly the Metaphysics that Kant has in view cannot claim any such basis. But then can this basis be adequate even for Physics ? Certainly not for Pure Physics as conceived by Kant. For the distinctive characteristic of this — what is meant by its ' purity ' — is that it *' propounds a priori, and as necessary, laws to which nature is subject" (§ 15, p. 64); and there is no point on which Kant is more emphatic than he is on the impossibility of establishing such laws by induction from particular experiences. But if the universals of Pure Physics cannot be thus established, it would seem clear that they cannot receive from such experiences adequate confirmation. We are left, therefore, with the lack of consensus and steady progress as the only valid reasons for suspending metaphysicians from their work, until a preliminary critical inquiry into the possibility of accomplishing that work has been completed.^ ^ But now observe the ' presuppositions ' : — Consensus implies plurality of minds; Progress implies Time. Cf. below, p. Zb fin. 6 TIIK MKTArilVSIUS OF KANT lkct. It inav be said that tlieso provisional criteria are not essential, but that in every ease it is important, before attempting to gain knowledge on any subject, that we should satisfy ourselves of the possibility of gaining it. 1 answer that this must also apply to the knowledge of the possibility, etc. Indeed the reasons Kant gives for suspending Metapliysicians from their business must be admitted to apply now to Critieists or Critical Kpistemologists.^ However, we will grant the need of inquiry, and only demand consistency in the assumptions and conclusions of Criticism. One point we may note in the view of knowledge from which Kant starts, because it throws important light on the movement of the modern mind in respect to the relation of Metaphysics to Physical Science. According to Kant, as we have seen, Physical Science has no occasion for a critical inquiry to remove doubts as to the validity of its fundamental principles : it does not require this " for its own safety and certainty." It is, indeed, important in the systematic study of human knowledge to show — as Kant holds that he has shown — that Physics has an a priori element, contains certain universal and necessary principles, " sprung from pure sources of the under- standing." But though this is important for the study of human knowledge as a whole — what we now call philosophy — it is not required for the secure establishment and steady progress of Physical Science * Cf. my article, " A Criticism of the Critical Philosophy," Mind, 1883, vol. viii. pp. 73 f. I THE CKITICAL STANDPOINT 7 itself. This Kant emphatically declares ; and, so declaring, he was no doubt in harmony with the instructed common sense of his time. But turn back something less than a century and a half, to the system which begins distinctively modern thought, and you find a very different view, Descartes, in his treatise on Method, when describing his state of mind at the outset of his independent study, speaks of the Philosophy offered to his youthful mind very much as Kant speaks of the prevalent dogmatic meta- physics : "Of Philosophy ^ I will say nothing, except that when I saw that it had been cultivated for many ages by the most distinguished men, and that yet there is not a single matter within its sphere which is not still in dispute, and nothing therefore which is above doubt," etc. (Discourse on Method, Veitch's edn. p. 9). But unlike Kant, Descartes holds that this defect of Philosophy extends to the Sciences, " Inasmuch as these borrow their principles from Philosophy," he continues, " I judged that no solid superstructures could be reared on foundations so infirm," Between 1637 and 1783 the Sciences and Natural Philosophy seemed to have managed to struggle out of the mire of controversy in which Metaphysics is still up to the neck. They have got their feet on firm ground and are making steady progress, to which the critic points as a contrast that ^ Philosophy as here used included more than Metaphysics, i.e. it included Natural Philosophy, which became effectively independent in Newton, and has since — like other subjects who have achieved independence — shown a disposition to turn and trample on its former lord. But Philosophy was throughout conceived by Descartes as a system of which Metaphysics formed the fundamental part (cf. Preface to the Principles, Veitch's edn. p. 185). 8 THK MKlArHYSlC^ OF KANT lect. puts tus. Practically, then, the Metaphysics into tlie possibility of which he is iiKjuiring may be taken to be modern Metaphysics, not going back further than the seventeenth century. But we may limit the inquiry still further to Continental Metaphysics from Descartes onward. For with the English line of metapliysical thought, developed side by side with the Continental, Kant has again only imperfect acquaintance.* He does not seriously argue with either Locke or Berkeley. He treats the former as the author of a celebrated but unsuccessful attempt to derive the pure concepts of the understanding from experience, and an obviously inconsistent attempt to use the notions so derived for obtaining knowledge beyond the limits of experience. He finds, indeed, in Locke's fourth book, a hint of the distinction between analytical and synthetical judg- ments ; but Locke's undeniable want of definite, systematic coherence seems to have prevented Kant from finding in him the instruction which — I denying unconditioned or free causality and an unconditioned or absolutely necessary Being. Having conapared the advantages and drawbacks of the two lines of thought in an impartial manner, Kant says that " this opposition all along the line of Empiricism to Dogmatism constitutes the opposition of Epicureanism to Platonisra." But this remark strikingly shows the imperfec- tion of his historical knowledge ; for two cardinal points in Epicureanism — which by the way was primarily opposed to Stoicism rather than Platonism — are its a3snmi)tion of material atoms and its maintenance of the freedom of the Will in antithesis to Stoic Determinism. " One might as well swallow the fables about the gods as bow to the yoke of Destiny " is an Epicurean dictum.* Similarly, his references to Plato show only a general popular knowledge of Platonic Idealism. ^ The mentalislic Empiricism which leads in its three stages to the very diverse conclusions of Locke, Berkeley, and Hume is not to be confounded witli the cosmological Empiricism to which I just referred. • Cf. Diogenes Laertius [x. 134. R.D.H.]. I THE CEITICAL STANDPOINT 11 venture to think — he might have found, on this topic. Again, he shows no sign of having understood Berkeley, whom he treats as a mere visionary idealist not requiring serious refutation. Of Hume he speaks with emphatic admiration, and acknowledges that Hume's discussion of causality first " woke him from his dogmatic slumber " ; but he only knows Hume's doctrine in the later and more guarded form in which it appears in the Inquiry concerning Human Under- standing — of the frank, comprehensive, and uncom- promising scepticism of Hume's Treatise on Human Nature he seems to have known nothing. It is, then, on the metaphysical doctrines of the line of Continental thinkers which begins with Descartes and ends with Wolff, that Kant's attention is almost entirely concentrated when he thinks and speaks of Metaphysics and metaphysical dogmatism. And here again we may make a yet further reduction : we may omit Spinoza. There is, I think, no direct reference to Spinoza in either of the books we are to study, certainly no evidence that Kant had ever seriously considered his position and arguments. Apart from Hume — whose metaphysical view, as I said, Kant only knows in re- spect of the concept of Cause — the only leading- thinkers whose metaphysical doctrines Kant knows sufficiently well to criticise with real grasp and penetration are Descartes, Leibniz, and Wolff. Speaking broadly, Wolff's philosophy is that of Leibniz, with the paradoxical element pared down so far as to make the doctrine acceptable to Common Sense. Kant refers to both together as ' Leibniz- 12 THK MKTA I'll VSR'S OF KANT lkct. Woltliiui.' Torliaps, on the wliole, it is Wolff's system tliiit ho h:is most before liis mind : partly, I think, be- cause Leibniz, though a more ori<]^inal and penetrating thinker than Wolff, was less of a system-maker, and Kant himself had a decided turn for system-making. But it is more important to note that the philosophy of Wolff, with minor modifications introduced by disciples, was the prevalent philosophy — the system that held the field, though by no means una.ssailed — not only when Kant was a learner, but for some time aft^r he began to teach : though we gather from the Preface to the first edition of the Cmtique that in 1770-80 its influence had rather given way before the stream of general culture and enlightenment flowing from France ; and that " Indifferentism, mother of chaos and night," was tending to take its place. I propose therefore, when we come to study Kant's criticism of Metaphysics, to state briefly under each head the chief doctrines of Metaphysics as conceived by W^olff, with such references to Descartes and Leibniz as seem to be required. And it is all the more important for us to try to get an idea of the scope and method of pre-Kantian Metaphysics, be- cause it is not easy to get it from Kant himself. For the new view of the problems of philosophy which Kant is introducing requires new lines of distinction which he does not always draw, or does not draw clearly and consistently. To show this, it will be convenient to give by anticipation Kant's answer to the question ' Whether Metaphysics is possible ? ' The answer is ' Yes ' and I THE CKITICAL STANDPOINT 13 * No ' according as the term is used ; and Kant seems to me to say ' no ' or ' yes ' according as he has the old method or doctrine of Metaphysics — what he some- times calls 'dogmatical ' Metaphysics, sometimes " the common Metaphysic of the schools " — in view, or the new method to which the Critique has shown the way. He means the former when he says that " all vain wisdom lasts its time but finally destroys itself," and that ''this time has come for Metaphysics." He means the latter when he says that one who has grasped the principles of the Critique will " look forward to Metaphysics, which is now indeed within his power, with a certain delight." He means the former vain wisdom when he explains the genesis of Metaphysic, how " before men began to question nature methodically, they questioned isolated reason, which is ever present . . . ," and " so Metaphysic floated to the surface like foam — like it also in this, that when what had been gathered was dissolved there immediately appeared a new supply on the surface."^ On the other hand, it is not this 'vain wisdom ' but true knowledge that he means when he say in the concluding section of the Prolegomena that " Metaphysics alone of all possible sciences can be brought" — at once seemingly — "to such com- pletion and fixity as to be incapable of further change or any augmentation by new discoveries." It is largely this doubleness of view which gave Kantism its vogue both in the age of its appearance and in times nearer our own. It appealed both to ^ {Prolegomena, § 4, p. 27.] 14 THK MKrAl'llVSlCS OF KANT lect. tlie foes and llie friends of iMeta])hysics. Were you inolinoil to despise Metaphysics as antiquated rubbish, eternal sterile word-debates, speculative spinning of unsubstantial thought-cobwebs — here was a professor of philosophy who used the same language, and justified your vague contempt by laborious demonstra- tions, conducted according to all the rules of the scholastic game. Were you, on the other hand, disposed to think that these many centuries of efforts of great minds must have some deep meaning, some true end and goal, must spring from an intellectual need for which satisfaction was to be found somewhere in the nature of things — the same professor undertook to explain to you the meaning, show you the goal close at hand, satisfy the philosophic need by a symmetrical, well-articulated, coherent system of far- reaching truths. Whether you ran with the hare or with the hounds, Kant ran with you : you might not quite understand him, but you knew that he was on your side. Let us look closer at the two kinds of Meta- physics : the good and the bad, the sham wisdom and the true. In the first place, it is true of both kinds of metaphysical propositions, the good and the bad alike, that they are synthetic a prio7'i pro- positions, and that neither they nor the concepts used in them can be derived from experience. That they must be a priori is implied in the very concep- tion of them ' metaphysical ' knowledge has always been understood to mean knowledge lying ' on the other side ' {jenseits) of the physical knowledge of I THE CRITICAL STANDPOINT 15 which external experience is the source. Again, they must be ' synthetical ' : that is, the truths which it is the end and aim of metaphysical inquiry to ascertain must be expressible in judgments or propositions in which the predicate is not implicitly thought in thinking the subject. Analytical judgments no doubt belong to Metaphysics, and are of course independent of experience. But as such judgments merely state in the predicate what is implicitly thought in the subject -notion, we can get no ex- tension of knowledge by making them : e.g. we can reflect on the metaphysical notion of substance, and make it more distinct by the purely analytical judg- ment " substance is that which only exists as the subject of predicates " ; but this merely tells us what is meant by substance, and does not extend our knowledge of substances. And such analytical judg- ments are in no way distinctive of Metaphysics, as we can equally well analyse merely empirical concepts as ' body ' and get from them equally certain judgments with regard to it — as that * body is extended ' — which equally add nothing to our knowledge of bodies. To make the definition of metaphysical proposi- tions complete, we require both characteristics, they must be at once ' synthetical ' and ' a priori ' : neither alone will do. Such propositions extend our know- ledge, and at the same time are not empirical : the latter point is otherwise clear from the fact that they are universal and necessary. For merely empirical judgments cannot have true and strict 16 THK MKTArHVSU'S OF KANT i.ect. universality, ami therefore not necessity; "experience can only tell us that, so far as our observation has gone, there is no exception to this or that rule." But, as I have said, these characteristics belong equally to the sliam knowledge and the true, the Metaphysics that we are to adopt and the Meta- physics we are to eschew. The question then is : What is the distinction between the two ? Perhaps the best way of expressing this distinction is to take Kant's phrase that " metaphysical know- ledge," as its very term implies, must be knowledge " on the other or further side (jenseits) of experience " ; and to show that the term * on the further side ' may have two different meanings, which we might express briefly as ' beyond ' and * behind ' experience. Metaphysical study before Kant had tried to go beyond experience : that is to say, it had tried to get, and professed to have succeeded in getting, real knowledge — synthetical judgments — with regard to realities that never were nor could be objects of experience. Whereas the Metaphysics that Kant offers aims mainly at going behind experience : by analysing the object and conditions of experience it seeks to separate and exhibit systematically that element in our thought about experience and its objects which is not obtained from without, but from the nature and constitution of the knowing mind — regarded first as perceiving through its senses, outer and inner, secondly as conceiving and judg- ing, thirdly as reasoning, passing from step to step of inference, and tending to unify its knowledge I THE CKITICAL STANDPOINT 17 into a systematic whole. It is this latter kind of knowledge that Kant sometimes calls Critical Philosophy : what he gives under this name is not, he tells us, a complete metaphysical system of the right kind — Transcendental Philosophy, as he some- times prefers to call it ; for it does not profess to contain a complete detailed analysis of all the pure non-empirical concepts that the human mind possesses, the derivative as well as the primary. But — diH Prolegomena, p. 177, shows — Kant does not think it a difficult matter to work out such a system, final and complete ; and the fundamental principles and plan of such a system he thought he had com- pletely given. Is this, then, all ? it may be said. Is this the end of all the high aspirations and pretensions of the Metaphysics — that seemed so fair, Such splendid purpose in her eyes ? Is she to confine herself to the task of making clear and systematic the a priori elements in our knowledge of the empirical world — which seems quite able to get on without her — and to tell us nothing of the great realities that she once sought to know : of God, and the human soul, and the relation between the two ? Are we doomed to know nothing of God by the exercise of our reason, and nothing of the soul except what empirical psychology can tell us ? No : that is not exactly Kant's meaning. The ultimate aim of the whole of his philosophy is to establish the beliefs in ' Immortality, Freedom, and God.' It is true 18 TllK MKTArilVSlUS OF KANT lfxt. that ho establishes them ])rimarily as postuhites of the practieal reasou, resting ultimately on our certaiu, irrefragable conviction of duty, together with our equally strong conviction that, in order that morality may be more than an idle dream, reason must assume a supersensible world in which happiness depends on the performance of duty. But though this is the basis of the certitude of our faith in God, Freedom, and Immortality, speculative reason has nevertheless a function with regard to these postulates : although, as I understand Kant, it is of very different import- ance in the three cases. In the case of Immortality, speculative reason — the non- empirical study of the soul, when duly critical — appears to do nothing but guard against materialistic explanations of mental phenomena. Rational psychology, with its idea of an absolute subject, " is merely a discipline which prevents us . . . from throwing ourselves into the arms of a soulless materialism," ^ and serves as a regulative principle totally to destroy all materialistic explanations of the internal phenomena of the soul — for these can never account for self-consciousness, — but it gives no ground for inferring the permanence of the soul beyond the period of mundane life. I may observe that as regards the practical postulate of Immortality, Kant's ideas appear to have undergone a development between the Critique of Pure Reason (1781) and the Critique of Practical Reason (1788). In the former, he does not distinguish between the belief in ^ Watson's Selections, p. 153. I THE CKITICAL STANDPOINT 19 immortality and the belief in ' a future life ' or ' future world ' in which the connexion which reason demands between morality and happiness may be realised. But by the time he came to compose the Critique of Pi^actical Reason, it seems to have occurred to him that the postulate of a future life, adequate to the rewarding of desert with happiness, does not necessarily involve endlessness of life. Here, accordingly, he rests the argument for immortality on the necessity for the realisation of the highest good by man, of ' perfect harmony ' between this disposition and the moral law. " Such a harmony," he says, " must be possible, as it is implied in the command to promote the highest good " — a form in which the command to do duty may be conceived ; on the other hand, ' a finite rational being ' cannot attain moral perfection, it is only " capable of infinite progress towards it." Hence, as we must postulate that our " existence should continue long enough to permit of the complete realisation of the moral law," we must postulate that it will continue for ever. I shall have occasion to refer to this argument later. It always seems to me to illustrate well both the ingenuity of Kant and what I may perhaps be allowed to call his naivete. I turn to the second practical postulate. Freedom of Will. Here, again, our positive certain conviction of Free Will is based entirely on the conviction of duty. Still, speculative reason has a not unimportant function with regard to this belief, though only in the way of showing that it is not excluded by the 20 THK METAl'llYSiC^ OJb' KANT lkut. i no less necessary assumption of physical science. We may say that a discussion of the possibility of exphiiniug natural effects by natural causality only, shows us a gap in our system of empirical knowledge which mail ^^ tilled b}- the ' free causality ' of the human individual as a transcendental reality, though we cannot positively say that it is so filled. But in the case of Theology somewhat more is done.^ * Cf. [in the Critiqiu of Piire Reason the section entitled "Criticism of all Speculative Theology "] Watson's Selections, p. 222. [In the Critique of Praclical Ikason that entitled " Possibility of an extension of Pure Practical Reason without a corresponding extension of Pure Speculative Reason," Watson's Selections, pp. 300-302. See also the Appendix at the end of these Lectures.] LECTURE II THE TRANSCENDENTAL AESTHETIC Having given this bird's-eye view of its conclusions, I pass now to examine in detail the principles and method of the True Metaphysics. It must be remembered that, according to Kant, we are not to expect from him a complete metaphysical system, according to his definition of Metaphysics, i.e. mainly a complete systematic statement of the a priori concepts and synthetic judgments — of the knowledge attainable by the human mind, apart from particular experiences. Such a system may be worked out hereafter : the Critique only gives the principles and method of constructing such a system. The exposition of the Critical or Transcendental Philosophy is divided into three parts, in accordance with the traditional threefold division of the cognitive faculties of the human mind into Sense, Under- standing, and Reason. It is to be observed, however, that Reason seems to be also used in the title in a wider sense, to denote the source of the a priori elements in cognition as a whole. This is due to another antithesis, which Kant finds in the thought 21 22 THE MKrArilVSiCS OF KANT lkct. IuuuUhI down to him, between ' rational ' and ' em[)iiieal ' knowledge. For elements of a priori knowledge real or supposed are found not only in tlie ideas and conclusions of Reason in the narrower senate : they are found also in the forms of Sense- perception, Space, and Time, and in the forms of synthesis by which the understanding constitutes empirical objects, and connects them into coherent elements of an empirical world, conceived as extended through space and perduring through time. But the treatment of the three sources has to be fundamentally different. For the a priori element derived from the forms of Sense, and the forms of Understanding, has been in the main rightly conceived by the thinkers who have employed it in the systematic sciences of Mathematics and Physics. The recognised appeal to intuition in the case of Mathematics, and the control of experience in the case of Physics, have kept the human mind — on the whole — from serious vagaries in these departments. In fact, as we have seen, these sciences are now enjoying uncontested acceptance and steady progress, and Transcendentalism assumes them as given. The case is otherwise with the a priori ideas peculiar to Reason — which are, in fact, various forms of the idea of unconditioned being or existence — the temptation to use these in answers to questions that carry us beyond the limits of possible experience has been too strong, and has produced the long stream of bad dogmatic Metaphysics, which Kant hopes effectually to dam up. Here, therefore, in this third part of II THE TKANSCENDENTAL ESTHETIC 23 Transcendental Philosophy, we have first to expose the vain semblance of knowledge by which the human mind has so strong a natural tendency to be deluded : and then, after destroying the vain semblance of knowledge, a sound criticism of these a priori ideas will show their use — (l) in systemat- ising as far as possible the additions to real know- ledge which we are continually obtaining through experience ; (2) in so making clear, when we stand at the limits of empirical knowledge, what may be reasonably thought of its relation to the un- known realities that lie beyond these limits; and thus (3) clearing the ground for the erection not of a structure of speculative knowledge, but still of well-grounded, rational, positive conviction on the great questions of the Existence of God, the Freedom of the Will, the Immortality of the Soul, and generally the Moral Order of the World. As we said, for these great convictions — always fundamentally important to Kant — the Practical Reason, in his view, affords the only adequate rational basis. In this last part of Kant's work — as will appear from what I have said — the true use of the ideas of Reason, the right direction of man's natural, ineradicable impulse to penetrate beyond the con- ditioned to the unconditioned, can only be understood when we have fully seen with his eyes through the illusions of the old Metaphysics. We have therefore to begin by examining Sensi- bility and Understanding, as sources of a 'priori 24 THE MKTArHYSU'8 OF KANT lkct. knowleilt^e. Tlio ii prion (.-Dgiiitions of wliich Seusibility is the primary source have been elaborated into !i gro:it, colteront, progressive system of know- lodge, which, from the outset of modern philosophy, has presented itself to the philosophii* mind as a model of certainty in its premises, method, and conclusions, and as at the same time entirely independent of empirical basis. This we call Pure Mathematics. Here Transcendental Philosophy, Kant holds, has no work to do in distinguishing and separating the pure or non-empirical element of the object of knowledge from the empirical element : it finds the separation completely made and universally recognised. It has only to make clear the source of this non- empirical knowledge, in the universal forms — Space and Time — in which the human mind receives and arranges the particular data of Sensibility. The case is different with Physical Science — including the application of Mathematics to that world of empirical objects with undetermined limits of extension in Space and duration in Time, concerning which Physical Science seeks systematic general knowledge. These objects and all their parts and their relations and changes in Time and Space are all measurable and numerable, and so far objects of the a priori mathematical knowledge just mentioned. But there is another non -empirical element, besides the mathematical, in the knowledge we commonly conceive ourselves to possess of the II THE TKANSCENDENTAL ESTHETIC 25 general laws of our common world of empirical objects ; and this element is much more difficult to exhibit in clear separation from the empirical element that is blended with it in the view of ordinary physical science. Here, in fact, lies the most difficult task for Transcendental Philosophy, so far as its work is constructive rather than destructive. This will occupy us in detail hereafter : the fundamental question is, How, from the subjective data of sense — the various impressions on each individual's sensibility which we distinguish as sights, sounds, touches, pressures, muscular feelings, etc. — is it possible to pass to universally valid knowledge of the laws of an objective world, common to all human minds ? The uncontested establishment and progress of Physical Science shows that we commonly conceive this transition to be legitimate, and that experience confirms the assumption of its legitimacy ; but how is it legitimate ? There is a great gap between the data of sense-perception, as reflective analysis shows them, and the general truths of science which we all accept — e.g. the laws of motion. How is the gap filled up ? The presence of a non-empirical element is manifest, according to Kant, in the conclusions of science if there are — as physical science holds — any ascertained universal laws of the physical world. For a universal conclusion cannot be validly attained by any number of mere particular experiences. But to show what this non-empirical element is, and how it is related to the empirical element, requires elaborate analysis. In the Critique this is given in 26 THK iMKTArHYSICS OF KANT lkct. the 'rransi'oiulentiil Analytic, aiul again in tho 'second piirt of tlie General Tmnsccudeutal Problem ' in the Proh'ijonwna. Tlie argumeuta in the first })art of the Transcend- ental Philosophy, the Transcendental ^Esthetic, are comparatively easy of apprehension ; and they seem to have been found convincing by thinkers who have been able only very partially to assimilate the elaborate and difficult system of the forms of understanding expounded in the second part (the Analytic) or the anti-spiritualistic conclusions — negativing speculative knowledge of Self and God — of the third part (the Dialectic). The arguments of the Esthetic may be read in Watson's Selections, pp. 22-39. What is called the * metaphysical exposition ' gives the context and characteristics of the notions of Space and Time : in the ' transcendental exposition ' they are regarded as sources of synthetic a priori judgments. The con- clusion is simple and striking. Space and Time are unalterable forms of sensibility, and therefore necessary conditions of the apprehension of phenomena by the human mind, but not attributes, elements, or conditions of the existence of things apart from their relation to the percipient human mind, nor even of human minds themselves, regarded simply as existing. Even Time is only a form of the appearance of a human mind to itself, not an attribute of its real existence. " If," says Kant, " I could be perceived II THE TEANSCENDENTAL ^ESTHETIC 27 by myself or by any other being without the con- dition of sensibility, the very same determinations, which now appear as changes, would not be known as in Time, and therefore would not be known as changes." ^ Distinguishing the two forms, Space is the necessary form of external perception — perception of things outside me — Time the necessary form primarily of the perception of ourselves and our mental state ; but, as external perceptions are states — or elements of states — of the perceiving mind, Time is a formal a priori condition of all phenomena without exception. This brought on Kant the charge of Idealism, vehemently repudiated by him in the Prolegomena and also in the second edition of the Critique.'^ It is, then, undeniable that Kant's metaphysical view as here given is not to be classed as Idealism or Mentalism, on account of its strong assertion of the existence of things other than percipient human minds, "unknown to us as to what they are in themselves," but yet * known ' — in a sense — as operating on us and causing impressions on our senses. It is rather to be called Phenomenalism — so far as the existence of a material world is concerned — since it holds that all the attributes of what we commonly call body, Locke's primary qualities as well as his secondary, are mere phenomena. But it is remarkable how little proof Kant ever ^ [Watson's Selections, p. 35.] ^ Prolegomeiia. § 13, Remark ii. pp. 54 f. Critique, second edition, "Refutation of Idealism." 28 TllK METAPHYSICS OF KANT lect. otFors oi' {\\c anii-iiu'iitalistic element iu his doctrine. In the passage in the Prolegomena we have simple assertion and not proof. In tlu' ' Refutation of Idealism ' in the second edition of the Critique, Kant is apparently demonstrating the existence not of things independent of human perception but of phenomenal things in space, which are ultimately only impressions on our minds, received in the forms of sensibility and combined into connected objects of experience by the judgments of the understanding. As regards, indeed, the reality underlying the phenomenal subject which Common Sense conceives as a soul or spirit, Kant (in the Critique of the Practical Reason) finds evidence of its existence in the freedom which our moral consciousness leads us to attribute to the ' noumenal ' self. But as regards ' body,' no such evidence is of course avail- able, and yet Kant does not anywhere offer any other. The explanation may be partly found in the fact that Kant's thought is not consistent on this funda- mental point, though of course this fundamental inconsistency, in a thinker so acute and so laboriously systematic, itself needs explanation. He never, indeed, denies the existence of an unknown thing- in-itself which, acting on our minds, produces the manifold sensations that, when bound together by the understanding, we call a ' body ' ; but in the concluding chapter of the Transcendental Analytic he certainly treats its existence as problematical. The most definite passage is the following : " The II THE TRANSCENDENTAL ESTHETIC 29 understanding limits the sensibility without enlarging its own scope ; and, warning the latter not to presume to deal with things-in-themselves, but only with phenomena, it forms the thought of an object in itself; but only as a transcendental object that is the cause of the phenomenon (and hence not itself phenomenon), and that cannot be thought of either as magnitude or as reality or as substance, because these concepts always require sensuous forms in order to be applicable to an object. We cannot say, there- fore, of this transcendental object, whether it is in us or also outside us ; or whether, if sensibility were taken away, it would disappear along with it or would still remain. " ^ It is impossible to reconcile this passage — especially the last sentence — with that in the Prolegomena, where Kant says : — " I grant by all means that there are bodies without us, that is things which though quite unknown to us as to what they are in them- selves . . . are not therefore less real." ^ In both cases, indeed, not only the ' secondary ' qualities of Locke, which Common Sense, but not physical science, attributes to bodies as they exist unper- ceived — colour, odour, flavour, heat, — but also Locke's ' primary ' qualities — extension, place, figure, impenetrability, — are regarded as merely phenomenal, merely mental, results of the understanding com- bining the data of sense. But in the Prolegomena, ^ Critique, Max Miiller's trans., p. 250. [Italics Prof. Sidgwick's.] This passage is not given by Watson, but the whole chapter is in this sense. See Watson, pp. 129-134. '^ Prolegomena, p. 54. 30 THE METATIIYSK'S OF KANT lect. empirical body is tlio appoanince of a thing ' iiifluenc- inaragiaj>lis, tlien, it seems tt« u\c clear liiat Tiint' and Space, as objects of oitlinaiy anil o{ scientific thought — jipart from any question of tlicir apriority or mentality — are not relations or 'abstracts' of relations, but entities of relational quality. We no doubt conceive the manifold tliinj^s of the material world as arranjifed in Space, and connected through their spatial rela- tions in a kind of order difl'erent from the order which they occupy in a scientific classification that systematises their relations of resemblance ; the essential characteristics of the spatial relations of real things, as Common Sense thinks, is that they are relations of position in one space. ^ And as there is one apparently real Space for all things, so there is one Time in which all events are temporally related. This remains true of Space and Time as ordinarily conceived, whether we regard them as belonging only to percipient and conscious human minds as such, or also to a real world existing independently of such perceptions. But are we to regard them as belonging only to the percipient mind ? To Kant's arguments in sup- port of this momentous conclusion I now turn. First, however, let us consider for a moment how momentous it is. I ask you to realise this, because I am not sure that Kant always realises it. For he seems to suppose that, even after being convinced by the argu- ments of the Transcendental Esthetic, when we come ' I say ' real things ' because, as Sigwart points out, we may and do con- struct scenes and geometrical figures in imaginary Space, having no definite relation to real Space. II THE TKANSCENDENTAL ESTHETIC 35 to the third part of the treatise, we shall still take a serious interest in the great questions of Rational Cosmology : — whether the physical world has bounds in Space, and had a beginning in Time, whether its parts are ultimately simple or infinitely divisible, etc. etc. But surely, for a mind of the least intelligence, all these questions are altogether cut off and precluded by the acceptance of the conclusions of the ^Esthetic : we can no more ask them than we can ask how many angels can stand on the point of a pin (a question which is said to have interested the mediaeval mind). For the real physical world, as we must then hold, not being in Space, can have no bounds ; and not being extended, the question of ultimate divisibility cannot be raised with regard to it. Again, not being in Time, neither beginning, nor duration, nor succes- sion of events can be predicated of it ; and, neither changing nor enduring, it can have no causality, in the sense of necessary connexion of antecedents and consequents. These conclusions, indeed, are what Kant himself draws ; but there are others, that concern us more intimately, which he has not expressly drawn, and which indeed I hardly see how he could have drawn without something like inconsistency. For these latter negations are true of the spiritual no less than the material world : since all temporal determina- tions must be held to belong to appearance, not to real existence, in the case of spirits no less than in the case of bodies. As Kant says in a passage before quoted : " If I could be perceived by myself or by any other being without the condition of sensibility, the 36 rilK MKTArilVSlCS OF KANT i.ECT. vory sanio (lotormiiiatiuns which now a})pcar as chansjes wduKI not he known as in time, and there- t\)iv WDuhl not 1h> known as chances."^ The notion of spiritual j)rogres8 is therefore merely plienomenal and unreal : and henee it would seem that the objection to Metaphysics, put forward as the stiirting-point of the transcendental inquiry, that it does not progress like other sciences, but goes on turning round and round without advancing, is deprived of its force — since the progress is in any case merely apparent. And this, of course, applies to moral as well as to intellectual progress. Hence the conception of moral progress, on which the practical postulate of immortality — as we saw — is based, is a conception that represents no real fact of any soul's existence, but merely an appearance due to the imper- fection of its faculty of cognition. But if moral progress is thus reduced to mere appearance, what becomes of the belief in the immortality of the soul which Kant (in the Critique of the Practical Reason) bases on it ? Indeed, in any case, if Time is merely a form of human sensibility, — due to an imperfection of man's nature which prevents him from knowing things as they are, — the postulate of immortality seems to become a postulate for the endless con- tinuance of an imperfection. It does not seem that this can afford an inspiring hope for a truth-loving mind. I do not find that Kant has fully contemplated these consequences of his doctrine of Time : though I ought to say that in his practical Philosophy he ' Cf. above, p. 27. II THE TKANSCENDENTAL ESTHETIC 37 certainly throws over Time— if I may so express myself — wlien he finds it convenient. Since, indeed, his defence of the notion of Freedom is expressly based on the assumption that the momentous choice between good and evil which every human soul makes is in reality not subject to the condition of time, so that any change that may appear in a man's character is illusory : his character as manifested in his conduct is made by himself though a timeless act of will in which there is no before and after/ Well, the consequences, we see, are tremendous : in the next lecture we shall have therefore to consider carefully the proof of the doctrine from which they flow. ' [Cf. Methods of Ethics, 6tli edn. Appendix.] LECTURE 111 rant's 'exposition' of space and time I.KT us, for simplicity and definiteness, concentrate attention on the notion of Space : and take first the 'Metaphysical Exposition.' Here Kant's points are two : (l) The notion of Space cannot be derived from external experience ; because, in order that I may apprehend things as out of me and out of each other, I must have the notion of Space already in my mind ; and (2) that the notion of Space is a necessary, a priori one ; for I cannot imagine Space annihilated, though I can very well think it emptied of objects/ Now it appears to me that in discussing these arguments — and all that Kant says on the subject — we are liable to two confusions of thought : one relating to the notion of ' externality,' and the other to the notion of * a-priority ' (if I may be allowed the word) : and that when these confusions are cleared away, Kant's arguments are clearly inadequate to prove their conclusion. First as regards externality. What is meant ' Cf. Watson's Selections, p. 24. 38 LECT. Ill 'EXPOSITION' OF SPACE AND TIME 39 here by 'external,' 'outside of? There are two distinct meanings possible : (l) 'Spatial externality.' This seems clearly meant in speaking of the apparent perception of things ' outside of and beside ' one another : the word ' beside ' definitely determines ' outside ' to this meaning. But ' outside ' {ausser) is sometimes used by Kant, definitely in the sense of (2) ' otherness of existence ' — ' distinct and inde- pendent existence.' ^ Now if we get these two meanings quite distinct, and then turn to the argu- ment that I have just summarised, we shall find, I think, that any force it may seem to have is derived from a more or less unconscious fusion of the two : and that if we apply either separately, it loses all force or contains a manifestly unwarranted assumption. First, take externality in the sense of spatial externality. Then ' outside of me ' must mean ' outside my body,' as Kant does not conceive my mind as occupying space. This being so, the state- ment that I cannot apprehend things as being outside my body and outside each other, without apprehend- ing them as occupying difi'erent parts of Space, is undeniable but insignificant ; since material outside- ness is a spatial notion, involved in and involving the notion of 'location in different parts of Space.' But the statement has no tendency to prove that the whole notion of Space and spatial externality is not empirical. I might as well argue that the notion of colour is not derived from visual perception, but ^ Cf. Prolegomena, p. 54, where 'without us' must mean 'having an existence distinct from and independent of our existence, an existence made known by some action on our senses.' 40 I'HK MKTAI'HYSICS OK KANT lkct. ' prosuppDSi'il in it,' beciuiso I oaiiUDt visually perceive tilings to be there at all without perceiving them to be coloureil. It is of course true — and I think this partly accounts for Kant's view — that so far as, in any fresh apprehension of things around me, 1 definitely apply spatial notions, — perceiving and judging that they are in front or to the right, of such and such size, at such and such distance from me or each other — I seem to bring these notions with me to the fresh experience and not to derive them from it. But this applies equally to my perceptions and judgments of colour, or any other admittedly empirical conception. I can only definitely apprehend any fresh experience by applying to it the system of notions that my mind has derived from past experience : though so far as the fresh experience contains novel elements, it will tend to modify and enlarge my previously formed system of notions — sometimes perceptibly, but more often imperceptibly. Observable progress in our experience of objects almost always takes place, not by sudden definite acquisitions of entirely new notions, but partly by new combinations of old notions, partly by the gradual consolidation into definiteness of vague apprehensions of new differences and resemblances. 1 see no reason why we should not suppose a similar gradual emergence into definiteness of our spatial notions, along with other notions admittedly empirical. Here perhaps it may be suggested that when Ill 'EXPOSITION' OF SPACE AND TIME 41 Kant says that the notion of Space is already pre- supposed in external perception, he only means ' logically presupposed ' : and similarly that * a- priority ' in the second argument ^ only refers to logical not chronological priority. Now the dis- tinction between these two meanings of priority has often been drawn — in the form of a distinction between what is ' naturally prior ' in knowledge and what is ' prior for us ' it is as old as Aristotle — and it may be said to be now current and familiar. But it is not easy to get it quite clear : that is, to get the conception of logical priority purged of all chronological suggestion : but when this purgation is effected, it seems to me that a merely logical presupposition of the notion of Space in external perception is quite irrelevant to Kant's argument. For what is meant by priority in a purely logical sense ? Merely that the concept (or judgment) said to be logically prior to another requires to be made explicit before and in order that the concept to which it is prior may be perfectly clear and distinct (or that the judgment may be arrived at by a perfectly cogent process of inference). In this sense the notion of a straight line is logically prior to the notion of a triangle as a figure bounded by straight lines : and Euclid's axiom relating to parallels is logically prior to his 29th proposition — it is a more elementary pro- position, without which the other cannot be cogently established. In this sense the notion of pure Space may no doubt be said to be logically prior to the ^ Watson's Selections, p. 2'^ fin. 42 rilK MKrArUVSK'S of KANT i.ect. notion (»l" a mati'iial iIuul:,-. Hiil wlicii this iiu'aiiing is mad*' i-loar, it is, 1 think, t'vidiMit that ' K)L:;ical jiiiorit V ' is ijiiiti' irrch'vant to tlic (jiu'st ion wliethcr Space really liolongs to the object perceived, in- (.lepeiulently ; or is t)nly a form under which the human mind is hy its constitution compelled to perceive it. SeeoniUy, Kant argues that the notion of Space is necessary, as is shown by the psychological experi- ment of trying to get rid of it. " By no eflbrt can we think Space away, though we can quite easily think Space empty of objects." This argument has been regarded as weighty by writers deserving of respect : but I confess that it seems to me to have all the w^orst defects that an argument can have : (1) it is not strictly true; (2) the distinction drawn in it between Space and Matter is inconsistent with another fundamental principle elsewhere laid down by Kant ; and (3) so far as it has any force it really tends in my opinion to prove the contrary of the conclusion which Kant draws from it. When I say that it is not strictly true, I mean that there are cases in which, so far as I can perform the psycho- logical experiment suggested, it does seem to me that Space is eliminated from my consciousness nearly or altogether for brief moments : — e.g. when I am absorbed in listening to music. But I quite admit what I rather understand Kant to mean, that when I turn my attention to Space, I am unable to conceive it annihilated. Only I do not find that this characteristic — inconceivability of annihilation — Ill 'EXPOSITION' OF SPACE AND TIME 43 distinguishes Space from Matter, as Kant affirms : I do not find that I can readily think of Space as empty of material things : — i.e. not all Space of all Mattel-. Such a complete emptying of Space is no less impossible to me than, the complete elimination of Space from my thought. And further, I should have supposed that Kant would have found the same impossibility, since he elsewhere ^ gives as a synthetical a priori cognition " that the quantum of substance in Nature can neither be increased or diminished." He holds this to have been admitted in all ages by men of common understanding no less than by philosophers, and expressed in the ancient Gigni de nihilo nihil, in nihilum nil posse reverti — nothing can be produced from nothing or return into nothing. But how can we readily think Space emptied of all Matter, if the permanence of material substance is a necessary condition of experience ? Whither is the Matter of which Space is emptied conceived to go ; and when it has gone where does the permanent substance hide ? But, lastly, granting it true that I can conceive Matter annihilated, but cannot conceive Space annihilated, the Space that I am unable to conceive annihilated is not conceived by me as a form of my cognition, or of human cognition, but as some- thing that exists independently of my cognition of it. Now, I concede to the Empiricists that we cannot infer with absolute certitude the existence 1 Watson's Selections, " First Analogy of Experience," p. 106. Cf. Prolego- rmna, § 15, p. 65. 44 THK MK.rArilVSK'S OV KANT lkct. of iiuv thing from the impossibility of conceiving it non -existent. At tlie s;ime time, I tliink the 'inconceivability of the opposite' is of some value jis a test of truth. But surely, if it is legitimate to infer anything from the inconceivability of annihilat- ini]j Space, it is the necessary existence of Space apart from my sensibility ; for it is that Space that I cannot conceive annihilated and not Space regarded as a form of my sensibility. For this — being a notion I never found till I came across Kant — is one of which I can get rid with the utmost ease. It would take too long to go through in the same way the metaphysical exposition of Time. I think it will be found that the reasoning I have employed in criticising the metaphysical exposition of Space applies, rtiutatis mutandis, to that of Time. I now pass to the ' Transcendental Exposition.' I have tried to show that the arguments Kant uses in his ' Metaphysical Exposition,' viz. that the notion of Space is presupposed in external perception, and that it is a necessary notion which we cannot by any effort think away, are ineffective to prove that Space is a form of human sensibility and not a determination that belongs to objects when abstrac- tion is made from our subjective conditions of perception. I ought, however, to say that I do not think Kant would have regarded them as effective, apart from an assumption which lies at the basis of the Transcendental Exposition. This is the assump- tion that I could not have universal knowledge, universal synthetic — not merely analytical — jvidg- Ill 'EXPOSITION' OF SPACE AND TIME 45 ments with regard to Space and its properties, if Space existed independently of my (or any human) perception ; but only if it be regarded as a form and subjective condition of such perception. Now doubtless geometry as commonly accepted does give us such synthetical universals : I know that all triangles inscribed in a semicircle must be right- angled triangles, and I could not obtain this knowledge by mere analysis of the notions of 'semicircle' and 'right-angled triangle.' But why am I to infer from this that the proposition is not true of a real extended world existing as such, independently of human cognition ? Kant's answer to this question is perhaps most clearly given in the Prolegomena^^ 9, p. 43. He there says : (1) " I can only know what is contained in the object itself when it is present and given to me"; and (2) "Even then it is incomprehensible how the intuition of a present thing should make me know the thing as it is in itself, since its properties cannot migrate into my faculty of cognition (Vorstellungskraft)." The second of these arguments, if valid at all, would render it unnecessary to consider the first or talk any further about things as existing apart from my perception. For if I cannot have immediate knowledge of any entity, because it cannot migrate into my faculty of cognition, it must surely for the same reason be impossible to have mediate knowledge of it or any rational conviction with regard to its existence : so that Rational Cosmology and Theology 46 TllK MKrArilVSlCS OK KANT lkct. would vuiiish ill a t winkliiii;, Icaviui:; iiotliiiiii' tor the Critical IMiilosopliv to ronlutc. Hut with tlicui also wouKl vanish the coiuH'ptiou ot" tiic reality of things lu tlu'iusclves, anil Ivant must inevitahly Tail iuto the lilealisin that he repudiates. But this is not all : not only would material things in themselves be thus eliminated, but all knowledge of other minds would equally be eut off: for another mind cannot migrate into my faculty of cognition any more than anything else. If the mind can only know what can get into the mind, then, as I certainly cannot he. anything except myself, I cannot know anything except myself. We are thus reduced from Ideidism to Solipsism : and the Critical Philosophy is thereby rendered absurd ; for what is the meaning of suspending all meta- physicians from their business and appealing to the ' uncontested ' position of Mathematics, if I do not know whether there are any metaphysicians or mathematicians except myself? This short-cut to agnosticism which has tempted others besides Kant — the strange dogma that in order to know a thing I have to he it — has thus led us into a quagmire of absurdities and inconsistencies. Let us abandon it once for all, and pass to the other contention, that I cannot know a thing unless it is ' present and given ' to me. This at first sight seems more plausible : but on looking closer, I think it will be found to involve a confusion between physical and psychical fact. It surreptitiously transforms a merely empirically known condition of bodies acting on bodies, into a condition, Ill 'EXPOSITION' OF SPACE AND TIME 47 dogmatically assumed, of a purely mental function. In our ordinary experience of material changes, the bodies that appear to act on other bodies appear generally to be locally contiguous with them. It is true that gravitation constitutes a vast 'prima facie exception to this generalisation : but efforts have been made to explain away this exception, and it is possible that they may some day succeed. But what then ? How can this physical generalisation as to the causation of motion justify us in dogmatically limiting the possibilities of the purely psychical fact that we call knowledge of Matter or Space ? Kant certainly does not mean to materialise mind so far as to localise it : and if not, the object of knowledge can never be properly said to be in local contiguity to the knowing mind. What meaning, then, can be attached to the statement that the mind can only know what is ' present and given to it,' except that it can only know — in fact what there is to be known ? It may be said that our apparent particular knowledge of the relations in space of particular things is scientifically known to be obtained only through a chain of movements between the things and our brains, throughout which contiguity of moving particles is always a condition of the transmission of motion. But granting this, how can we legitimately infer from this empirical generalisation the impossi- bility of obtaining by reflection universal knowledge of the spatial relations of real things ? To the ordinary geometer it undoubtedly appears that certain universal spatial relations, applicable to a 48 TllK MKr.\rilVSlCS OF KANT i.kct. rt';il oxtcnial world, are proscntcd to his iniiul as nocossai y : siiivly the assuni]>tioii that this is impossible is a iiiere doi^nia, which eaiiuot be justified bv anv enipirieal ujencralisation b;ised on our empirical knowledge of tlu^ particular spatial relations of ]mrticular things. But, further, if 1 could have no universal know- ledge of anything except the forms of my own sensibility, why should I suppose that I can have universally valid synthetic judgments with regard to these ? This is a question which Kant never seems to have asked himself: but it is of fundamental importance to examine it, when we are considering the pros and cons of the question as to the subjectivity or objectivity of Space and Time. If I can only know — or let us say " only know with the certainty that Mathematics claims " — what is * present and given,' surely I can only thus know the form of my sensibility as it is here and now : I cannot know what it has been in the past, nor what it will be in the future : I cannot know that it has not changed, or that it will not change : still less can I know that it is precisely similar to the forms of sensibility of other human minds. But if this is so, what can possibly be gained for the explanation of the universal validity of our geometrical cognitions by transferring Space from the non-ego to the ego ? I have gone into this at some length because the view to which I am replying is a part of Kant's doctrine which has been more widely accepted than many other parts. In pursuing this argument so Ill 'EXPOSITION' OF SPACE AND TIME 49 far, I have followed Kant in assuming that the synthetic universals of Pure Mathematics depend on intuition ; and therefore that the objects of mathe- matical cognition cannot be merely thought but require to be constructed in concreto. This is the distinction which Kant draws between mathemati- cal cognition and philosophical (under which term he includes both Physics and Metaphysics). " Philo- sophical cognition," he says, " is the rational cognition obtained from concepts, mathematical that obtained from the construction of concepts. . . . By constructing a concept I mean representing a priori the intuition belonging to it. For the construction of a concept, therefore, a non-empirical intuition is required which as an intuition is a single object, though as the construction of a concept or general notion it must express relations generally valid for all possible intuitions that come under the same concept." He takes the instance of a triangle : in order to reason about triangles generally I ' con- struct ' the concept either by representing a particular triangle "by mere imagination in pure intuition, or after this upon paper also in empirical intuition, in both cases however a priori, without borrowing the pattern for it from any experience." ^ Now no doubt what Kant says here is broadly true of ordinary geometry : when we reason about triangles, or squares, or circles we do draw in imagination or on paper particular triangles, etc. ' Kritik der reinen Vemun/t, Hartenstein's edition, p. 478 [M. Miiller's translation (emended), p. 611.] 50 THE MKTAl'HVSICS OK KANT lect. It seoins io inc. liowever, bold to atliriii tliat these simple figures are not borrowed from experience. We had got empirical ideas of these before our earliest stutlies in geometry : we callfd a plate circular, and the sides of dice and boxes square, and the tlaps of envelopes triangular : and when we came to get more precise ideas of these from Euclid, and to be introduced to unfamiliar figures — such as the rhombus — they were always drawn for us on paper before we represented them in imagination. Doubtless, as we came to understand geometrical reasoning we realised that the square we reasoned about was not the square we drew : for first, the latter was a particular square of a particular size on a particular piece of paper, whereas our reasoning was about any square of any size anywhere ; and secondly, the lines of the drawn square were slightly w^abbly and unequal, while the square of our thought was a perfect square. This distinction between the real general object of geometrical thought and the imperfect particular copy that we use to aid that thought has been a starting-point for philosophical Idealism since Plato : but this imperfection and this particularity belong no less to any square I may imagine, if I try to solve a geometrical problem in my head. Indeed, in my poor experience, the circle of my imagination is much inferior to that which I draw as a representation of the general or abstract circle about which I think : the circle I draw is not quite round, but it is clear and stable, whereas the circle I imagine is dim and fluctuating. It seems to me Ill 'EXPOSITION' OF SPACE AND TIME 51 indubitable that the latter is a copy of the drawn circle, and that there is nothing of pure intuition about it. No doubt, as my geometrical faculty develops, I can imagine more or less definitely new figures, even surfaces of complicated convolution which I could not draw on paper. But I see no diff"erence in this respect between geometrical and mechanical reasoning. The inventor of a machine imagines new combinations of wheels, levers, screws, cranks, etc., varying the data of his mechanical experience to produce a novel result ; in the same way proceeds the geometer, whose imagination, guided by and aiding his thought, constructs {e.g.) a pseudo- spherical surface. I do not see why a construction in ' pure intuition ' should be interpolated between the thought and the empirically developed imagina- tion in the case of the geometer any more than in that of the mechanician. So far I have been considering, as Kant is, elementary geometry. But it seems to me important to note that, when we have learnt to apply analytical methods to algebraic figures and quantities, our thought is to an important extent able to dispense with the aid to reasoning furnished by the particular concrete specimens — drawn or imagined — of its general notions. It is able to grasp the law of construction of a regular curve, never presented or represented before, to know it to be possible and to deduce important properties of it without constructing any specimen of it at all, either in imagination or on paper. And, speaking from my THK MKTAIMIYSICS OV KANT lkct. own rxjHM'iriico, wIumi in my stiulios of analytical i;i'onu'trv I c'anit> to conatnict tliese unfamiliar curves: — the catenary, tlie cycloid and epicycloid, cissoid, conchoid, cti-.— 1 could never trust imagina- tion in the Icjist to construct the curve as a whole. This had always to be done on paper : the imagination was reduced to the humble role of interpreting various simple cases of the general equation to the curve in the terms of very familiar relations of position and quantity. But refleetion on advanced geometrical reasoning introduces us to another notion, which establishes a still more striking exception to Kant's universal statement as to the dependence of mathematics on intuition : I mean the notion of a limit, to which certain varying quantitative relations approximate, as the quantities related are conceived to become very large or very small ; though the limit is never attained, so long as the quantities in question have a finite value. Well, in geometrical reasoning beyond the most elementary, this conception of a limit is continually introduced. For example, in measuring the area of a circle, we suppose a regular polygon inscribed in it and a similar polygon circumscribed : it is easy to see that the area of the inscribed polygon is smaller than the area of the circle, and the area of the circumscribed polygon larger. So far intuition carries us : and also the judgment that the larger we make the number of sides of the two polygons, the smaller becomes- the difiference between the two areas, and therefore the difference between either and the Ill 'EXPOSITION' OF SPACE AND TIME 53 area of the circle, is also intuitive up to a certain point : but in the final conclusion that by increasing sufficiently the number of sides of the two polygons, the difference between their two areas, and between either area and that of the circle may be made less than any assignable quantity, so that the area of the circle may be measured to any degree of exactness — this final step in ^the reasoning cannot be realised intuitively or imaginatively, any more than it can be drawn on paper : the notion of a difference less than any assignable quantity is one in which geometrical reason goes clearly beyond geometrical intuition. And this case is all the more important, because of the resemblance between this mathematical reasoning — as uncontested in validity as any other — and the philosophical reasoning in the department of Rational Cosmology which Kant criticises in the third part of his transcendental philosophy.^ So far I have been considering the case of geometry, and I have tried to show that the ' pure intuition' which Kant considers as indispensable to geometrical reasoning is not really to be found at any point of the development of the reasoning in question. For (1) in the more elementary stages, while we certainly rely on the aid of individual con- crete specimens — or rather approximate though im- perfect copies — of the ideal objects of thought whose relations we are examining, yet the imagination ' Further, mobility is commonly assumed in geometrical demonstrations, though 'motion ' is not a 'pure' conception according to Kant. 54 TllK MKI'AIMIVSICS OF KANT lect. or jMjrceptiun tli;it aids the reasoning seems to be as enipirieal in the case of geometrical lus it is in the closely analogous case of physical reasoning. While (i!) in the more advanced stages of geometry our reason emancipates itself from this dependence on intuition, to an important extent; ascertains {<'■. 65. 58 LECT. IV THE TEANSCENDENTAL ANALYTIC 59 experience, and is assumed in order to explain these — atoms and molecules, e.g., are such objects of possible experience. But this does not, in Kant's view, exhaust the common meaning of the term. It is implied, he thinks, in the conception of Nature — it is certainly implied in the conception of a Science of Nature — that this complex of objects and changes, in spite of the manifold diversity it exhibits, is subject to general laws : and the aim of Science — as distinguished from mere natural history — is to ascertain these laws. Hence Kant regards this (uncontested) subjection to law of all objects of experience as the formal aspect or meaning of the term Nature : and includes it in his original definition of ' Nature ' ^ as " the existence of things so far as it is determined according to universal laws " — ' things ' being afterwards limited to ' objects of experience.' A science of Nature, then, is under- stood to mean systematic knowledge of the laws by which the complex of empirical things and events is governed. We know from our previous discussion that ' the jpure Science of Nature ' denotes the non-empirical element of this knowledge, the universal law^s that may be known independently of particular ex- periences. But are there such laws ? We cannot find them — at any rate without further analysis — even in such principles of wide application as the laws of motion. For the concept of ' motion ' is not a pure concept ; it could not be formed apart from experience. Also, ' Prolegomena, § 14, p. (53. 60 rilK MKTArUVSK'S OF KANT lkct. Kant lioKls iliat siviclly a prion and univorsal laws of nature must relate to all ohjocts of ex})erieiice, ' iniuM-' no loss than ' outer.' Still liiere are principles that liave the required universality ; e.g. the [)rin- c'iple that " substance is permanent" and the principle that "everything that happens is predetermined by causes according to fixed laws." ^ Observe that Kant thus gives the proposition that " substance is permanent" a wider scope than current science commonly assigns to it. We now regard it as a proposition belonging to physics as distinct from psychology. Thus the * conservation of mass ' in all transformations of matter' is empirically proved or confirmed by weighing the products of any such transformation, and comparing them with the weights of the matter previous to transformation. But for us the proposition has no direct application to psychical experience. The wider scope that Kant gives it he found in the system of Wolff: in this system not only was everything in the material world conceived to consist ultimately of simple indestructible sub- stances (atoms), but human souls were also such simple substances naturally indestructible and there- fore immortal. This conception, with the momentous inference from simplicity to immortality, Kant afterwards assails with j^^reat force : he holds that mind, as an object of experience, cannot be speculat- ively known as having a permanent substance distinct from the substance that has to be conceived as ' Prolegomena, § 15, p. 65. Compare "Analogies of Experience," Watson's Selections, pp. 106, 110. IV THE TRANSCENDENTAL ANALYTIC 61 underlying all transformations of matter.^ He thus reduces the proposition " substance is permanent " to its present purely physical scope : while still main- taining it as a universal law of Nature in general, i.e. of the whole complex of objects of experience. The main problem, then, of the second part of Transcendental Philosophy is to show how this a priori element in our scientific knowledge of Nature is possible : which, in Kant's view, is equivalent to showing how it follows necessarily from the constitu- tion of the mind — the laws of thought acting on the data of sensibility. But as this a pinori element is not, in the pursuit of physical science, clearly distinguished from the empirical element, it is also a part of the task of Transcendental Philosophy to give it in the requisite systematic form. But another problem, which may be partly dis- tinguished from this — though the answer to the one, as we shall see, involves the answer to the other — is presented for Kant's solution, when he approaches this second part of his philosophy from the Transcend- ental ^Esthetic, which formed the first, the problem, namely, How there comes to be a world of objects of experience for human minds at all ? The Common Sense answer to this question is that this physical world has gradually come to be known through an innumerable mass of particular cognitions of material things, cognised as they exist apart from human minds ; — such cognitions being remembered, recorded, ' The immortality of the soul he maintained only as a postulate of the Practical Reason. 62 TlIK MKTArHYSlCS OF KANT i.Kc-r. c't»mnuiiU('Jiti'd. conihiiieil, ami liiially rectified and •^[enenilised by Scieuce. But from any such answer Kiint is altogether precluded by the conclusions of the first part of his Transcendental Philosophy. For this world of empirical objects is certainly ordinarily conceived to exist in Space and Time : all our detinite knowledge of it involves and is insepar- able from spatial and temporal determinations. But Kant has already arrived at the conclusion that Space and Time do not belong to the world of reality, as it exists apart from human cognition, to ' transcendental reality ' as we may call it. All that we know of this transcendental reality rests on im- pressions produced by it in human minds : and these, so far as yet analysed, consist of a manifold of sensations received in the two fundamental forms of human sensibility. Space and Time. But this result is obviously very unlike our common world of material things in complex motion. How then did we ever oret from the one to the other ? What is the transition from a mass of formed sensations to a world of matter in motion ? This is a question which Kant must answer — and indeed every one must answer who rejects the Common Sense assump- tion that we can know things as they exist apart from our cognition. As I have said, Kant's answer to the two questions that I have just distinguished is the same : and indeed, it is in this identity that its interest and persuasiveness lies. It is, according to Kant, the synthetic or unifying action of the Understanding IV THE TRANSCENDENTAL ANALYTIC 63 that converts the data of sense-perception into objects of experience : and it is because this is so, that we are able to lay down a priori certain fundamental laws to which experience and all objects of experience must conform. When I say that ' the synthetic action of the Understanding ' converts our sensations or sense -perceptions into experience of objects, I ought to explain that the understanding alone — according to Kant's view of the faculties of the human mind — could not produce the result. The forms of intellectual synthesis which Kant calls categories are too heterogeneous from the data of sense -perception to be applied to them directly. "There must be some third element which is homo- geneous on the one hand with the category, and on the other hand with the data of sense, so as to render possible the application of the one to the other." ^ This mediating element is furnished by Imagination, the faculty whose ordinary empirical use is to re- produce the data of sense. But Imagination is also capable of a pure or non-empirical exercise, in which its only matter is drawn from the pure form of all sensation and of all the empirical facts of conscious- ness — viz. Time. It is Pure Imagination influenced by Understanding which supplies what Kant calls the transcendental 'schemata.' These are the time- determinations which fit the categories of the Under- standing to be applied in connecting the data of sense, and so enable the Mind to lay down principles to which all objects of sensible experience must conform. ' [Cf. Watson's Selections, "The Schcinatism of the Categories," p. 85.] 64 TllK MKrAi'llVSlCS OF KANT lkct. This complic'Hted operation of faculties — though I have not as yet given its full eomplexity — is some- what ditUeult to grasp in this general presentation ol' it. 1 will therefore illustrate by applying it to the two categories used in the a 2^^ori principles that I before quoted : the principle that ' Substance ' in Nature is permanent and that every event is determined by antecedent ' Causes ' of which it is the necessary consequent. Here the notions of Substance and Cause correspond respectively to the forms of the understanding which logicians distinguish as the Relations exhibited in the Categorical and Hypothetical judgment respectively. The relation in the categorical judgment is that of subject to predicate : this, applied to connect the data of sense into objects, becomes the relation of substance and attribute. ' Substance ' so conceived as a pure category of thought, and applied to sensible data, is that in the object of experience which can only be thought as 'subject' and not as 'predicate' — the data connected with it would all be possible predi- cates. But there is nothing in the data of sense, so long as we consider them apart from the pure form of time, to which this conception of ' necessary subject ' could be applied. Every empirical datum of sense that forms an element of the notion of a material thing can be and commonly is regarded as an attribute of the thing : and yet if our under- standing is to think the thing at all, we must apply the notion of substance somehow, otherwise the requisite connexion or combination of data w^ill not IV THE TEANSCENDENTAL ANALYTIC 65 be effected. Here, then, the pure imagination comes in, and gives the rule for the application of this connecting form of subject and predicate, by the time -determination of 'permanence' or 'duration.' Substance is now recognised as that in objects of experience which remains permanent while their sensible qualities change : and this, and this alone, being the significance of ' substance ' as applied to data of sense — so far as it is more than logical 'subject' — we can, Kant holds, lay down a priori that the substance in Nature is permanent amid all changes of phenomena, and that its quantum neither increases nor decreases. Let us turn now to the notion of Cause. This has a special historical interest, because it was Hume's criticism of the supposed necessary con- nexion of causes and effects — as a truth evident to reason — that woke up Kant from his ' dogmatic slumber.' Here again we have the pure category distinct in the logical form of the Hypothetical judgment " If A is, B is," which expresses a rational dependence of B on A : but in this form, it is a purely rational dependence with no refer- ence to time. And here again Imagination and the pure form of time render this form of thought- synthesis applicable to the data of sense, by the time- determinations of antecedence and consequence ; and so enable us to define Cause as that in the phenomena of sensible experience which must come before the efiect, and after which the effect must ensue. ^ ^ I have had two reasons for the selection of these illustrations of the F 66 THE MKlWrilYSK'S OF KANT i.kct. Wo liiivc traced the I'liiietions of Uiulcrstaiuling and Imagination in su})plying connexion to the data of sense : but we must now go deeper, and penetrate to the root of this comi)lcx operation. Synthesis or combination, as \vc have seen, is the essential function of the understanding : it is a function which reflective analysis of any ordinary conception shows us to have been exercised in the framing of such conception. It is not only that every general notion combines the similar elements of an indefinite number of particulars. Take the notion of any individual material thing, obtained through sense-perception : we find in it elements derived from diflfereut senses which must have been somehow put together. But that is not all : take the sensible quality belonging to one sense, e.g. vision, it is commonly a manifold : different parts of a coloured surface may be differently coloured, in conceiving it as *a surface' we have unified the manifold. Even if it be perfectly uniform in colour, still as an extended surface it is analysable into parts which must — Kant says — have been put together. Nor is this true only of empirical notions. syntheses of Undersunding and Imagination, operating on the pure form of Time, thereby generating the fundamental connective elements in our common thought of objects of experience, and furnishing the a priori constituents, the necessary universal truths, that are the basis of our scientific knowledge of the empirical world. First, in these notions of Substance and Cause — especially the latter — and in the principles in which they are employed (Substance necessarily unchangeable and Causation necessarily universal) we have the historical starting-point of the Transcendental Philosophy ; as is shown by what Kant tells us of his relation to Hume. Again, this always seems to me the most impressive and plausible part of Kant's elaborated system— 'forms of pure thought' applied through 'time-determinations' to sensory data, whereby an empirical world, a nature of things, is built up, the fundamental laws of which we may lay down a priori. IV THE TEANSCENDENTAL ANALYTIC 67 Take the simplest conception used in a geometrical proposition, take a straight line : in * drawing ' it, even in imagination, we put together its parts into a whole. And observe, the unity that results from this synthesis is not the category of unity as opposed to plurality (with which I shall deal later on) ; for it is found just as clearly in the notion of a 'number of things ' as in the notion of ' one thing ' — though in the case of ' a number ' the synthesis is of a kind that keeps the parts put together distinctly before the mind. Now this combination, which we find everywhere, which all our analysis presupposes, cannot be referred to mere sensibility. So far as the mind is merely passive, merely recipient, the context of perception is mere diversity. Sensation gives us a manifold of qualities : we have no sensation of oneness. The combination and the resulting unity must be referred to the mind qua active : it is not something that comes from without and is merely passively appre- hended. Yet again this connexion that we find everywhere is not arbitrarily introduced by thought : it cannot but find it everywhere. We cannot con- ceive a datum of sense, a feeling of any quality, absolutely isolated, unconnected, unrelated : while at the same time we cannot, Kant holds, conceive this relation and connexion as merely given, merely passively apprehended. How are we to explain this universality of connectedness in the data of sense, which yet mere sense cannot give ? Kant finds the explanation in ()S TllH MKrArilVSlCS OK KANT i.kct. what he oalls * the original synthetic unity or the tmnscendentiil unity of self-consciousness or apper- ception ' — the necessary reference of all the data of experience to one identical experiencing subject. He calls it 'transcendental' because it is not merely an empirical fact that I do refer all my sensations to one identical self, but I know that they must be so referred in order to be elements of experience at all. In fact, I am not always actually conscious of self-identity, at least not clearly conscious in every moment of sensible experience : still every datum of sense that can form an element of an object of experience for me, must be capable of being: tliouo;ht of as mine, must belong to one identical percipient self, though I may not actually be conscious of this reference in having the perception. This fundamental unity of self-consciousness, ' trans- cendental ' because knowable a priori as necessary, is the root or basis of all the complex synthesis of Understanding and Imagination combined, of which the Transcendental Analytic gives the detail. It is because this self is an intelligent, not merely a percipient self, because the activity exercised in its synthesis of the data of sense is the activity of Thought or Understanding, that we find necessary thought-elements, forms of thought applied a^^WoW through time -determinations, in our notions of empirical objects, and are able to lay down a priori laws to which such objects and their relations and changes must conform. The detail of the system I shall examine in the IV THE TKANSCENDENTAL ANALYTIC 69 next lecture : but before we enter upon it, it seems necessary to understand more clearly and exactly Kant's use of the terms ' Object,' ' objective.' What precisely does Kant mean by the 'object' or complex of objects, of which he proposes to determine the necessary conditions ? In the first place, as we know, he does not mean things in themselves : he sometimes speaks of these as * objects per se ' and even as ' objects of the understanding [Noumena) ' ; but in calling them by either name he is usually careful to explain that we can know nothing at all determinate about them. ' Object ' in the sense in which it is used in the Analytic — especially if used in connexion with 'objective' — is always * object of possible experience.' The objects of which Nature is the complex are solely such empirical objects. But what are objects of experience ? In the first place, Kant does not include under this term all that in a wider sense we are accustomed to call objects of thought or knowledge.^ I hardly think that Kant bears this limitation always in mind, when he expressly restricts the application of his categories to objects of experience : but his language leaves no doubt about it, and it seems to me very important to make it clear. There are two kinds of objects of Knowledge expressly excluded : (l) Forms of thought, considered as not applied to things ; (2) Elements of sensation, considered otherwise than as elements of material things. As regards the first, ^ Kant uses 'knowledge' (Erkenntniss) in a narrower sense. Cf. Critique, 2nd edn. ; Analytic, § 24, end. 70 THK METAPHYSICS OF KANT lect. Kant's language is quite deeisive. "General Logic," which deals with the forms of thoutrht and reasoninc: in ahstracto, " abstracts from all distinction of objects" and ''from all relation of knowledge to its objects."^ The forms of thought, therefore, with which Logic deals are not to be considered objects for the purposes of the present discussion : though they must be not only objects of thought, but — as Logic is a Wissenschaft — objects of scientific knowledge. And in fact the forms of judgment and reasoning of which logicians treat obviously admit of being compared and classified, made the subjects of judgments universal and particular, athrmative and negative. Indeed, if we are to think about thinking, as the logician does, we must apply the forms of thought, the fundamental categories of thought, to the forms of thought themselves : and Kant himself does this here very definitely when he presents us with a table of twelve categories divided into two groups, each group subdivided into two classes, and explains that the third category in each class arises from a union of the second category with the first." All this, as I shall hereafter point out, seems to me difficult to reconcile with Kant's view of Number as a Time-determination. However, the forms of thought are not ' objects ' in the sense in which Kant's Transcendental Analysis employs the term. The object in this signification must have elements supplied by Sense. But again, we cannot say that any feeling, or * Watson's Selections, pp. 41 f. - Watson's Selections, pp. 51-53. IV THE TEANSCENDENTAL ANALYTIC 71 kind of feeling, or even any combination of feelings thought under one notion, can be an ' object ' in this narrower Kantian meaning. It is true that in one place he defines 'Object' as "that in the conception of which the manifold of a given intuition is united." ^ But he explains in more than one passage that such judgments as " the room is warm, sugar sweet, and wormwood bitter " ^ are merely subjectively valid : adding in a note that " because they refer merely to feeling which can never be attributed to the object," such "judgments can never become objective, even though a concept of the understanding were superadded." So again he speaks of the " fine flavour of the wine as not belonging to the objective characteristics of the wine, even con- sidered as a phenomenal object."^ But obviously sweetness, bitterness, and flavours generally — like the forms of thought — may become objects of thought, be compared and classified, and otherwise subjected to the application of the categories and forms of judgment. Thus I may judge that some or all flavours of wine are agreeable, that the flavour of whisky is pronounced but not delicate, that if the flavour of port is combined with that of olives the pleasure is heightened, that the flavour of champagne is either sweet or dry, etc. : — and thus apply in turn all the logical functions of judgment and the pure concepts of the understanding in Kant's table. Still, as I take it, the judgment would not be objective ^ Prolegomena, § 19, pp. 70, 71. " Watson's Selections, p. 58. ^ Kritik der reinen Vernunfl, Hartenstein, p. 63, M. Miiller, p. 25 [a passage omitted in the 2nd edition]. 72 rill': MKl'AniVSICS of KANT lect. or rohito to objei-ls m ilio siL,Miiti(.'alii.)ii Ivaiit here uses. Aiul in Kant's view this would be true also of combinalions of sounds and eolours,' How thou are we to distinguish the kind of sensible manifold of whieh the combination con- stitutes an object for Kant ? So far as I can see, we might suthciently distinguish it by the characteristic that in ordinary thought, or ordinary thought rectified by physical science, it is conceived to exist as we perceive it independently of our perceptions. This is what we commonly mean by a ' thing ' or ' reality ' when we use the word carefully : and it is such a ' thing ' that Kant means by his ' object ' here.^ But this characteristic, as we know, Kant declares to be illusory : what I call objects are nothing but modifications of my — or some other man's — sensibility, they are merely 'in us ' : and yet in thinking of them, we inevitably think of them as independent of the sensibility of which they are modifications, I do not think that Kant is definitely aware that his 'implicit' definition of objects attributes to them a characteristic which his system withdraws from them, and declares to be illusory. Indeed, in important parts of his argument he appears to me to forget that it is an illusion, in spite of the explicit language in which he has elsewhere characterised it as such. For we find among the characteristics of ' Cf. Transcendental Esthetic, I.e. Hart. p. 63, M. Mhller, p. 25. Sensa- tions of colours, sounds, heat do not in themselves help us to know any object. * Cf. Second Analogy of Experience, Hart. pp. 175, 176, M. Miiller, pp. 166 f. IV THE TKANSCENDENTAL ANALYTIC 73 empirical objects laid down as a 'priori cognisable, that they must contain a (phenomenal) substance that is thought of as remaining unchanged amid all phenomenal change : but it seems impossible to think this and at the same time to think of all phenomena as merely modifications of my sensibility. Yet Kant nowhere seems conscious of this prima facie contra- diction, or makes any efi'ort to explain it. It seems to him absurd that "the thing -in -itself" should " wander into my consciousness " ; yet, so far as I can see, neither he nor his English expositors find any difficulty in conceiving the phenomenal thing to wander out of it. Both he and they seem to hold that I can know objects to be merely modifications of my sensibility, combined in certain ways by my understanding ; while at the same time I also conceive them as difierent from the modifications of my sensibility and as perduring when the latter cease. Indeed, this unconscious contradiction seems to run through Kant's use of his cardinal term ' presentation ' ( Vorstellung) : the ' Vorstellung ' is now identified with its object, and now again contrasted with it, without any attempt at reconciling the two incom- patible views. At one time we are told that " out- ward things are nothing but mere Vorstellungen," ^ while again it is declared that " the determination of my existence in time is only possible through the existence of real things which I perceive outside me, and not through the mere Vorstellu7ig of a thing outside me."^ Will it be said that these really > "^Esthetic," § 3, p. 64. ^ i„ ^^^ "Refutation of Idealism, " p. 198. 74 THK MKrArilVSICS OF KANT lkct.iv existent phoiioiiKMuil things, though indepemlent ot' my consciousness, are implicitly thought by nie to be iu rehition to 'consciousness in general,' and that it is this relation which gives them their permanence, when they cease to be modifications of my scnsil)ility ? This — which resembles the Berkeleyau mode of reconciling Idealism and Common Sense — is an explanation certainly suggested by some passages in our recent English expositors of Kant. Thus {e.g.) Mr. Caird says,^ that by the recognition of the data of sense as objective " the data of sense are taken out of their mere singularity as feelings, and made elements in a universal consciousness, in ' conscious- ness in general ' ; or, to put the same thing in another way, they are related to a consciousness, which the individual has, not as a mere individual, but as a universal subject of knowledge." But whatever happens to the data of sense in Kant's psychological laboratory, it is at any rate certain that they do not cease to be modifications of sensi- bility. Hence in order to explain how phenomenal things can be conceived to exist independently of my — or any other man's — sensibility, we should have to suppose not merely a rational consciousness which all men share, but a universal quasi-human sensibility, modified similarly to the human ; and I need hardly describe the emphasis with which any such chimera would be repudiated by Kant. ^ Philosophy of Kant [1st edn.], c. viiL p. 341. [The latter part of this lecture is taken from an article "A Criticism of the Critical Philosophy" (Mind, 1883, O.S. viii. pp. 318 f.), written before Dr. Caird's second edition had appeared.] LECTURE V THE MATHEMATICAL CATEGORIES AND PRINCIPLES At the close of the last lecture I was discussing Kant's use of the term 'object' — in the sense in which the word is commonly used by him, when it is used without qualification — i.e. for the empirical or phenomenal object as distinct from the * object per se.' On the one hand, it is ' altogether in me,' consists of modifications of my sensibility unified by my understanding ; on the other hand, its elements have to be somehow distinguished from other data of sense — colours, flavours, sounds, heat — which, as Kant says, cannot form part of an object even though a concept of the understanding were superadded. It seems to me that the sensible elements of the object can only be distinguished by the characteristic that in ordinary thought duly instructed by science they are conceived to exist apart from my sensibility, i.e. by a characteristic which Kant's Philosophy regards as illusory. This is a serious objection. We have now to observe that one result of the view Kant takes of objects is a change in the meaning of * objective ' ; and herein is to be found 75 To rilK MKlWrilVSlCS OF KANT lect. one ex])l;uiatioii o( his unconsciousness of the pecu- liarity in liis implied definition of * object ' which I have pointed out- — namely, that an object of experi- ence is prima facie distinguished from what is not an object of experience, by tlie characteristic of being commonly believed to have an existence independent of the mintl, an existence which, however, the philosopher knows it not to have. The change is expressed in the following passage of the Prolego- mena: — "All our judgments are at first mere per- ceptive judgments, they hold good merely for us (that is, for our subject), and we do not till afterwards give them a new reference (to an object), intending that they shall always hold good alike for us and for every one else ; for if a judgment agrees with an object, then all judgments [our own and those of others] concerning the same object must likewise agree among themselves ; and thus the objective validity of the judgment of experience signifies nothing else than its necessary univer- sality."^ And this, accordingly, is the meaning that in Kant's philosophy is chiefly attached to the terms 'object,' 'objective,' except w^hen the former is qualified by per se. Thus, while in the more ordinary use the signification of the noun is prior and that of the adjective secondary, in this new Kantian meaning the relation is reversed and the notion of 'object' is now determined by reference to this new meaning of ' objective.' Objective, that is to say, means what is necessarily thought by all ^ Prolegomena, § 18, p. 69. V MATHEMATICAL CATEGOKIES & PRINCIPLES 77 minds. Object means that the existence of which is so thought, even though the elements of such object are only in us. In fact, the antithesis of subjec- tive and objective is quite changed. Subjective as opposed to objective is now not used of elements of thought derived from the judging minds, for these so far as they spring from the nature of the mind have objective validity. It is now used of what belongs only to the thought or feeling of particular subjects. In consequence of this new antithesis, the same notions and judgments — for example, the notion of Space and the synthetic judgments of Geometry, are sometimes spoken of as subjective — when tYL^iv source is the point considered ; and sometimes as objective — when stress is laid on their universal validity. Still there often seems to me a hopeless confusion in what Kant says of objectivity and object, owing to the conceptions of object per se and empirical object falling into one in his mind. AVe now pass to examine in detail the contribution of Pure Thought — that is, of Thought considered apart from the data of sense and the forms in which the human mind receives them — to our conceptions of empirical objects. In virtue of this contribution we are able to lay down a priori — independently of particular experiences — the fundamental laws to which the complex of empirical objects which we call Nature must conform. The ascertainment of this contribution, in an abstract form, is, in Kant's :S rilK METArilVSU'S of KANT i.k.t. view, not diliioult : fi)r tlic WDrk, in the main, is found tilroady jiorformed by the science which, as ' common ' ov ' general ' or ' formal ' Logic, he dis- tinguishes from Transcendental Logic, as the science that deals with the manner in wliich tlicse forms determine our conceptions of empirical objects and their connexion. The general function of the Under- standing, as we have seen, is Synthesis or Combina- tion, In our conceptions of empirical objects and their connexion in experience, the results of this Synthesis are implicit or latent, and only discover- able by analysis. But the forms implicit in our conceptions of objects become explicit and manifest in our judgments about them. Accordingly Common or Formal Logic, concentrating attention on the formal rules of judgment and reasoning, and abstract- ing altogether from the content of knowledge (the objects about w^hich we judge and reason), has already classified and systematised the universal forms of thought made explicit in judgments. The acceptance by Kant of the results of Formal Logic is — with one or two qualifications to be presently noticed — complete and noteworthy. He considers that Logic — so far at least as the forms of judgment are concerned — was created in sub- stantial completeness by Aristotle,^ and that from his time it has not had to retrace a single step, of material importance, nor has it from his day been ' Aristotle did not, he thinks, find the right princijile for making a system of fundamental categories, and consequently mixed in spatial, temporal, and empirical notions. V MATHEMATICAL CATEGOEIES & PEINCIPLES 79 able to make one step forward. Indeed it is, I think, the example of this completeness attained at one stroke by Formal Logic which encourages Kant to hope that the work of Transcendental Analysis, and the true metaphysic in which it is ultimately to result — the systematic exposition of the a 'priori elements in our thought about the world — may attain completeness and fixity with almost equal rapidity.^ Now, as I have before said, Kant's historical knowledge is seldom distinguished by thoroughness and accuracy : but in the present case his misconception of historical facts is very remark- able. If we look at his Table of Judgments, classified according to logical form and the strictly correspond- ing Table of pure concepts or categories,^ we see that there are twelve forms classified under the four heads of Quantity, Quality, Relation, Modality. Under the head of Relation we find the concepts of Substance and Cause : we have already seen that the principles based on these are selected by Kant himself as examples of strictly a priori principles in the science of Nature. I think his exposition of these is the most interesting and important part of his account of these a priori principles ; for, as we saw,^ it was the new view of Causality, attained by meditation on Hume's penetrating criticism of the older view, which was the historical starting-point of Transcendental Philosophy. We may say, then, that the categories of Relation have a special importance ^ Gf. Prolegomena, p. 177. ^ Watson's Selectioiis, pp. 48, 51 ; Prolegomena, § 21, pp. 76 f. '' Cf. above, p. 65 n. 80 rilK METAPHYSICS OK KANT lect. ill the 'rraiisceiulontal systcMii : ami, as we see, the three categories uiulor this lioad, Substance, Cause, Community (or Reciprocal Action), arc derived from the logical classification of Judgments as Categorical, Hypothetical, and Disjunctive. It is, therefore, really remarkable that this triple classification is not Aristotelian : Aristotle does not analyse the hypothetical form of judgments nor expressly the disjunctive, though he lays down the general formula for strict disjunctions in the principle of the excluded middle, but he only worked out a scheme of cate- gorical syllogisms. Here, then, Formal Logic, as conceived by Kant, has taken a step forward since Aristotle. But this is not all : in respect of the fourth class — ^judgments and categories of modality — Logic has had, in Kant's phrase, to " retrace the step " taken by the founder. I do not mean that modern logicians are agreed to exclude the topic of modality altogether : but there is certainly no consensus in favour of including it, still less as to the view w^hich Formal Logic ought to take of modal distinctions.^ My aim now^ is to show that the two last out of the four heads in Kant's tables represent one a step forward, and the other a step, if not exactly backward, at least on one side, from the Aristotelian view of the forms of judgment : and if so, Kant's confidence in the completeness and fixity of his systematic tables is certainly not justified on the historical ground on ' Cf. Keynes, Formal Logic, Srd edii. pp. 76-78 ; taking the discussion in Sigwart's Logic, pt. i. ch. vi., as a basis. V MATHEMATICAL CATEGORIES & PRINCIPLES 81 which he is inclined to base it. But has he any other ground for this confidence ? He seems to think that because he has shown the Understandinor o to be essentially a faculty of Synthesis or Combina- tion, having its root in the transcendental unity of conscious experience as referred to a self-conscious subject, therefore its fundamental forms have been obtained from a common principle, and therefore systematically, and therefore completely. But I cannot see that he has established any rational relation between the unity of a self-conscious in- telligence and the multiplicity of the recognised logical forms of judgment : he has not shown — I do not see that he has even tried to show — that there must be just these forms and no more : the categories are no more systematised by being referred to one understanding or faculty of synthesis than beads are systematised by being strung on one string. But having signalised this defect in Kant's demonstration, I pass on. Our general view of philosophy and its problems is very different from Kant's, recognising the slow and gradual evolution of human knowledge in the past, and not expecting any part of Thought to be free from it. Logic did not spring from Aristotle's brain, like Pallas from the brain of Zeus, as Kant seems to have supposed : but we will assume that the labours of formal logicians have had some result and that it is worth examining, without making any assumption as to its completeness and finality. And in this examination, in the present lecture. 82 THK MKTArilYSICS OF KANT lect. I shall ooutine myself to the first two heads of the t-iible, Quantity aiul Quality : as the distinctions here taken are certainly Aristotelian, anin in the world iipparently known to us through experience mid in the wider wi^rid ot" man's reasoned thouijht — to say this is to say that Kant has a genius for philosophy. But a man's forte is often also his foible : and Kant's genius for finding true system, discoveriuor true relations and connexions of funda- mental thoughts, carried with it a temptation to invent false system, and impose a fictitious and misleading appearance of symmetry on thoughts the true relations of which are only obscured by it. And this, I think, is what has happened in the case of this doctrine of ' schematism,' i.e. of the limitation by time-determination of forms of thought which the forms of judgment exhibit as independent of time. As I have said, this part of Kant's philosophical construction started with the categories placed under the third head — the categories of * Sub- stance ' and ' Cause ' — first ' Cause ' then ' Substance ' : and here the notion of schematism shows a philo- sophical insight whicli appears the more brilliant and penetrating, the more we study the efi'orts of previous thinkers to grasp the true significance of these fundamental notions. The relation of the notion of Subject in a categorical judgment to the predicate affirmed of it is clearly the relation of the substance of material things, that we conceive to remain unchanged amid the changes of their phenomenal attributes, to those changing attributes : only that in the thought of Substance we have added V MATHEMATICAL CATEGOEIES & PRINCIPLES 87 to, blended with, the logical notion of subject the time-determination of 'permanence.' So again the relation of dependence between two judgments expressed in the mere form of the hypothetical judgment " If A is, B is," is quite apart from any reference to time : when we judge '* If virtue is involuntary so is vice," reference to Time does not come in at all : while, again, in scientifically judging physical phenomena to be connected as cause and effect, we have this same relation of thought applied to, blended with, limited by, a time-determination : the cause on which the effect depends must come before it in time : it is that phenomenon or complex of phenomena after which, as we conceive, the effect must come. In both these cases, the blending of time-determinations with thought-relations that have a wider scope is clear and unmistakable, however we may ultimately interpret it. Well, then, Kant, having, as I said, by a brilliant and original stroke of philosophic insight, found this connexion between logical forms of judgment and time-determinations in the case of these fundamental notions of substance and cause, is irresistibly tempted to system-making on the strength of this discovery. He thinks that he has here the key to the whole matter, the explanation of our whole conception of empirical objects and their connexion, and of the principles that can be laid down a 'priori with regard to them : and therefore he determines to find a similar ' schematism ' everywhere, to drive it through the whole table of logical forms and cate- SS THE MHTAFIIYSICS OF KANT lect. gories. Let us iu)w examine the fallacious results of iliis luistakoii system-making: and in so doing, re-establish the true relations and distinctions of thought which Kant is forced to pervert or ignore, in order to obtain his false appearance of symmetry. 1 begin with Quantity. The ' schema ' of Quantity — the time-determination by which tlie application of the logical category of Quantity to empirical phenomena is supposed to be regulated — Kant declares to be Number, w^hich is said to be the "generation (synthesis) of Time itself in the suc- cessive apprehension of an object." ^ And on this application of the logical category to Time is said to depend the a priori principle that "all perceptions (objects of perceptions qua perceived) are extensive magnitudes." ^ Now there is, no doubt, an important difterence between logical quantity and number : in passing from the former to the latter we pass from the merely indefinite plurality, involved in the relation between a class-notion and the individuals included in the class, to a perfectly definite plurality. But I cannot see that the transition introduces a time-determination. A number, as I conceive it, is the conception of a whole of like parts, considered simply as at once like and distinguishable. It does not matter in what their likeness consists ; and, as we can apply the category of unity to any fact or aspect of fact which we make an object of thought, we can similarly apply number everywhere — counting ' [Watson's Selections, p. 90 ; Critiqiie, M. Miiller, p. 128.] 2 [Watson, p. 92 ; M. Miiller, p. 143.] V MATHEMATICAL CATEGOKIES & PKINCIPLES 89 together objects of thought that are only alike in being objects of thought — though we apply it most naturally and easily to things markedly alike and so naturally classified together. But it seems to me quite arbitrary, to limit the primary application of Number to successive phenomena, regarded as successive, and to regard Number accordingly as a Temporal notion. Kant's only argument for this seems to be that it takes time to count. I do not think this true of very small numbers where the things numbered are markedly alike : looking at my bookshelves, I perceive the volumes of different works to be two, three, or four respectively, by apparently single acts of attentive perception. But granting that it always takes time to count — as it certainly does in forming or applying the notions of larsjer numbers — it also takes time to draw a logical conclusion from premises : but it would be obviously absurd to say that therefore the thought of the conclusion involves a time-determination. Indeed, I cannot see how this view of Number can be made consistent with Kant's fundamental distinction between forms of pure thought and the data and forms of sense. He tells us that the categories, the pure conceptions of the understanding have their origin in the understanding alone, indepen- dent of all sensibility : and he expressly says that these pure conceptions are, considered in themselves, free from all limitation by human conditions of sensibility : ^ and potentially applicable to perception ' Cf. "Transcendental Analytic," § 22, Watson's Selections, p. 75. 90 THE METAPHYj^ICS OF KANT lect. of ;iiiy kind, wlu'tlu'r like or unlike ours, if only it is sensuous : only that such an application must beyb?* lis empty and fruitless. But if this be so, the system or table of categories must surely be conceivable apart from any reference to Time ; and if conceivable at all, it must surely be conceivable as a table of twelve catetTories : the characteristic of being twelve must therefore be as independent of time as any other characteristics of the categories. In short, the parts of any whole, whether logical or physical, to which we apply the idea of number, are commonly conceived so far as numbered, without any reference to time : and though the parts of a physical thing must be conceived as coexisting in time, this is not the case, according to Kant's express and repeated statement, with the parts of a logical whole or system. And this leads me to another point, which Kant overlooks and which is inconsistent with his view of number : viz. that the notion of number does not necessarily involve any notion of extensive magni- tude ; the scientifically fruitful and important applica- tion of number is, of course, to such magnitudes : but it is not necessarily involved in the very idea of number. If I judge that there are four cardinal virtues and seven deadly sins, I do not in so judging even suggest to myself that there is more deadly sin than virtue in the world — though this may be an unhappy fact. This brings us to the a priori principle or law of the empirical object, which Kant connects with the V MATHEMATICAL CATEGORIES & PRINCIPLES 91 ' schema ' of quantity, namely, that " all intuitions, or all phenomena as far as perceived, are extensive magnitudes." ^ Now, firstly, we observe that this principle, so far at least as its more obvious application to spatial magnitude is concerned, follows at once as an immediate inference from the propositions main- tained in the ^Esthetic. It was there maintained expressly (l) that Space is the form of all the phenomena of the outer sense, i.e. of all objects externally perceived ; and (2) that Space is an unlimited given magnitude : all phenomena or objects perceived in Space must thus have the characteristic of being spatial magnitudes. The introduction of the notion of number is, then, not required for this conclusion. On the other hand, its introduction leads Kant into serious errors. It leads him to ignore the important distinction between the discreteness of the parts of number and the continuity of spatial magnitude. Number, in fact, is not applicable to spatial magnitude simply and immediately, but only through the medium of the assumption that the magnitude is divisible into equal parts : and, con- sequently as we know, some of the most familiar relations of spatial magnitudes — e.g. the relation between the magnitude of the circumference of a o circle and the magnitude of its diameter — are not perfectly expressible by definite numbers. But [secondly] Kant does worse than ignore this distinction and relation between the notions of discrete ' [The principle is differently stated in the two editions : the two state- ments are here combined.] 9'2 THK MKrvrUVSlCS OF KANT lect. mill of continuous quantity. Ho is lod by ignoring it into the serious error of saying that " an extensive magnitude is one in which the idea of the parts necessarily precedes and makes possible the idea of the whole." ' He expressly applies this to all the parts, " I cannot," he says, " have the idea of a line, however small it may be, without producing all its parts one after the other" : and "similarly with any, even the smallest, portion of time." Now in the very same passage he lays stress on the infinite divisibility of spatial magnitudes. Surely Kant's acumen could not have failed to see — had he not been temporarily obfus- cated by his unhappy schematism — that it is impos- sible to hold at once that a spatial magnitude is infinitely divisible, and that a distinct idea of the parts of this magnitude as parts has necessarily pre- ceded the idea of the whole. For of however many parts we may be definitely conscious in forming the idea of a given line or a given portion of time, as all these parts are themselves extended magnitudes, they must be conceived as in their turn divisible into parts of which no definite consciousness can have preceded. I have laid stress on this palpable inconsistency, because it affords a clear illustration of what I regard as erroneous in Kant's general assumption that the understanding " cannot separate what it has not previously bound together," "^ especially in its applica- tion to phenomenal objects. In my view there is no foundation for this assumption : the essential function of thought, in all its departments, is not primarily ^ [Watson's Selections, p. 92.] "^ [Cf. Watson's Selections, p. 64.] V MATHEMATICAL CATEGOEIES & PEINCIPLES 93 or mainly the binding together into a whole of elements previously separate : but rather a process by which we pass from the consciousness of a vague manifold, of which the elements are obscurely thought, and even may have a merely potential existence, to a consciousness of the same manifold as not only more connected, but also more distinct in its parts or elements, and not only more distinct but fuller. The schematism of the categories of Quantity, therefore, seems to me a mere illusion that leads Kant into a quagmire of fallacies. But if the schematism of the categories of Quantity breaks down, that of Quality fares no better : indeed, I think that the forced and fictitious character of the construction is even more palpable in this case. For, first, the a "prion principle at which he arrives is more startlingly aloof from the logical forms he professes to apply. ' Quality ' of a judgment or proposition in common Logic signifies the distinction between affirmative and negative judgments. Now there is nothing more evident about this logical antithesis, when abstractly contemplated, than its absoluteness, and the apparent absence of any possible mediation or transition between the two. And this is a point on which Logic had been clear and decisive from Aristotle's time to Kant's : the Law of the Excluded Middle ' that A must be either B or not B ' is the one germ of the subsequently developed topic of disjunctive judgments and reasonings that we do find in Aristotle. The one fact, therefore, which is most alien to this antithesis is the continuity of transition from non-existence to 94 THE MKTArHYSICS OF KANT lect. existeuce which we actually tiiul in sensible experience. But Kant is determined to bahmce liis a piriori princi{ile that all j>lienc>nu'ua have extensive magni- tude, l»v a corresponding principle relative to intensive magnitude or degree ; and symmetry requires him to connect this with logical Quality. lie has therefore to invent a 'schema' for Reality and Negation, and he accordingly invents the notion of a " continuous and uniform generation of reality from nothing to a definite degree " ^ — reality being conceived as that in phenomena which corresponds to sensation. Now, in the first place, the notion of continuity in the gradations of intensity manifested by the sensible qualities of empirical objects is not a time-determina- tion. No doubt, as sensations and empirical objects must exist in time, the continuity in variations of intensity which they exhibit must be manifested in time, but the notion itself has nothing to do wdth time. Secondly, we suddenly find here a new meaning given to reality. So far we have come across a Transcendental Reality which we cannot know, and an empirical reality which Kant repeatedly attributes to Space and Time : but now we are sud- denly told that " reality is that in phenomena which corresponds to sensation . . . the transcendental matter of all objects."- Why should reality be thus equated to matter alone, instead of to form and matter combined ? Only, I venture to think, from the unfortunate necessities of symmetrical schematism : ' Cf. Watson, Selections, p. 88. ' [Watson's Selections, p. 88 ; M. Miiller, p. 126.] V MATHEMATICAL CATEGOKIES & PKINCIPLES 95 for only this would have turned Quality into Intensive Quantity. But, thirdly, how are we to reconcile this correspondence of Reality to Sensation with Kant's view before quoted, that colours, sounds, etc., "being merely sensations and not intuitions, do not help us by themselves to know any object " ? ^ How then can even empirical reality correspond to them ? Surely Kant here gets confused between the popular and the scientific conception of an object. But turning to the philosophical question raised, Can we lay down a priori that every sensible quality must have a definite degree ? Observe, degree belongs also to spatial magnitudes, but not to all. There are degrees of curvedness of lines but not of straightness ; of obtuseness and acuteness of angles but not of rightness ; of oblongness in rect- angular figures but not of squareness ; of ellipticality but not of circularity. In the case of the material world, we do commonly assume that sensible qualities vary continuously upwards from the lowest per- ceptible degree. Not less remarkable is the deduction which Kant makes from his principle of the ' Anticipations of Perception,' viz. that we cannot have experience of a vacuum. We are first told that reality corresponds to sensation, and negation to absence of sensation ; and the possible continuous diminution of the real down to zero is inferred as corresponding to a similar diminution of sensation. But then we suddenly find that we somehow know a priori that 1 Cf. above, p. 71. 06 THK MKTArUYSICS OF KANT lkct. " every sense must have ;i iletiiiite degree of reeeptivitv," ' and aeeordingly that below the point at which any kind of sensation stops — below what we may rail tlie sensible zero — the transcendental matter corresponding to such sensation must be still conceived as possibly existing, in any one of an indefinite number of continually diminishing degrees. Thus " we see that experience can never supply a proof of empty space or empty time, because the total absence of reality in a sensuous intuition can never be perceived, neither can it be deduced from any single phenomenon, and from the difference of degree in their reality ; nor ought it ever to be admitted in explanation of them " : ^ and thus the schematism of the category of Negation seems to end by demonstrating its strict inapplica- bility to phenomenal reality. I hardly know where to begin to criticise this singular argument. (1) If the matter of all phe- nomenal objects consists of mere modifications of our sensibility, how can we consistently suppose a phenomenal object to exist corresponding to modifications which, by the very nature of our sensibility, cannot possibly occur? And (2), if we could suppose this, by what transcendental intuition do we know that our senses must be incapable of perceiving phenomenal reality below a certain degree ? And (3), even granting that we must suppose as possibly existent a phenomenon that cannot possibly appear, and therefore that we can > [Wataon, p. 99 ; M. Muller, p. 151.] ^ [m. Midler, p. 152.] V MATHEMATICAL CATEGORIES & PRINCIPLES 97 never have direct experience of void space and time, it still is not clear why the assumption of such a void can never be admitted as an explanation of phenomena : for, granting that an apparent void cannot be known to be real, it does not surely follow that it must be known to be merely apparent. And, finally, it seems to me that this corollary from the ' Anticipations of Perception ' must land us in serious difficulties when we try to make it consistent with Kant's express interpretation of the first ' Analogy of Experience ' — to the discussion of which I will now proceed/ ^ [The last two paragraphs are from the article "A Criticiam of the Critical Philosophy," Mind, O.S. 1883, vol. viii. pp. 333 f.] H LECTURE VI SUBSTANCE We have now discussed the two first heads of Kant's table of forms of judgment and thought, regarded as applied in the constitution and connexion of empirical objects — the concepts and principles, that is to say, which come under the heads of Quantity and Quality. In passing from this I propose that we dismiss the forced and fallacious schematism, and merely carry with us the a priori principles that all objects of sense -perception must have extensive magnitude,^ and intensive magnitude or degree. Kant calls these mathematical principles, " to indicate that they justify the application of Mathematics to objects of sense-perception." ^ • This seems to me to follow from the Transcendental iEsthetic. - Watson, Selections, p. 102. I may observe — what Kant indeed sees — that in respect of this application of Matlieniatics the first jirinciple is of more fundamental importance than the second : since we can only apply !^lathe- matics to the intensive magnitude of sense-percepts by interpreting it in terms of extensive magnitude. Thus we measure weight, which has empiri- cally only intensive magnitude, by its tendency to produce motion, whicli has extensive magnitude. But I cannot see the a priori certainty that every quality of an empirical thing has a degree. Feeling we do assume to have a degree : also sensible qualities, But tlien, what of the objectivity of these according to Kant? To matter, according to the common view, degree is not ascribed. 98 LECT.vi SUBSTANCE 99 These principles he also calls * constitutive ' of phenomena : through them we know a priori what phenomena will be like in certain important respects. The principles, on the other hand, which Kant connects with the forms of judgment and thought classified under the head of Relation, — the forms explicit in the categorical, hypothetical, and dis- junctive judgment respectively, — he distinguishes as regulative : i.e. they do not tell us what phenomena must be like in any respect, but only give us rules that determine their relations of existence. Thus the a priori principle that every event must have a cause does not tell us in the least what the cause will be like, but only directs us to find something antecedent to the event in time, after which it must follow. This distinction between 'constitutive' and ' regula- tive' a priori principles is, I think, quite clear in the case of the Principle of Causality. The distinc- tion is not quite so clear in the case of the Principle of the Permanence of Substance, which 1 will take first. If we can say a priori of every empirical or phenomenal thing that there must be somewhat in it which remains permanent while other phenomenal elements of the thing change, why is not this principle ' constitutive ' of the object ? I confess that I am rather inclined to think Kant would so have regarded it, if the requirements of symmetry had not forced him to class it with the principle of Causality. However, passing from this for the present, let us consider how the principle is established. 100 rilK MK'IWI'UVSICS of KANT lkct. First let Us note lliat in ost;il>lisliiiig this principle Kant uses, in at loast a I'lcurcr form than in the previous (.-jises, the transcendental method of proof, of whii'h he is the inventor, lie distinguishes this method cari'fully and emphatically at once from the ilnnoiistration of mathemati(\s and from the meta- physical method previously current, which, not adequately distinguishing synthetical from analytical universals, confusedly tried to derive from mere abstract conceptions propositions really synthetical.^ Thus ' substance is permanent ' is such a proposition, if we mean by substance that in a thing or things which cannot be thought as predicate or attribute of some other subject. And if this proposition is not to be merely explicative (' permanence ' being already thought as part of the meaning of Substance) we must mean this by Substance. Now this pro- position taken abstractly is not self-evident and cannot be demonstrated : we can only, Kant holds, establish its truth by showing that experience and objects of experience are only possible, if we assume this principle, and not otherwise. This is the Transcendental method. Let us examine carefully its application to the Principle of the Permanence of Substance. Briefly, the argument is that our common conception of experience, as the apprehension of a complex of things as undergoing change or alteration in time, requires the notion of a permanent somewhat of which the phenomena — in the succession of which ' Cf. Watson, Sdections, p. 105. VI SUBSTANCE 101 change as merely perceived consists — must be thought as successive attributes, or modes of its existence. Indeed Kant goes so far as to say that without this conception of a permanent somewhat the relations of change and coexistence would not be possible, that is to say, they could not be attributed to the manifold as object of experience ; for in mere per- ception the manifold of phenomena is always merely successive. ' Substance,' in fact, stands in our thought for the unchangedness of Time ; for Time itself does not change, but all change has to be thought in it. As Time by itself cannot be perceived, there must be in objects something to represent Time, something unchanging, and of which all change can only be thought as a determination. This is ' Substance ' ; and as it cannot change, its quantum cannot be decreased or increased. Now, first, it does not seem to me true — I mean not truly to represent our common thought about Time as expressed in common language — to lay down in this unqualified manner that " Time does not change." For motion is a form of change ; and Time is certainly thought to move : it seems to us as true to say that " Time flies " as that " Time abides." In short, as I have said, change and permanence, succession and duration, seem to be inextricably combined in our common notion of Time : which, therefore, can only be properly imaged not by a line but by a point, the Present, passing along a line. However, I will not dwell on this, as I am quite 102 riiK Mirr.vi'iivsics of kant lkct. {)ro{>;ireil to ailmit tliat 1 cuiiuot conceive clianoe, at least of an object or tiling, without the conception ot' somewhat that pcnlures in or through Time. Hut 1 do not see that this jx^rduring somewhat need be conceived as absolutely unchangeable. Sup- pose a manifold presented consisting of elements A B C D : it seems to me perfectly possible to conceive change to go on in it, in respect of one element after another, so that ultimately an entirely new manifold E F G H is found to have substituted itself for the other : and yet I can at any point of the process conceive the manifold as a changing thing, provided BCD remain unchanged while A is turning into E, etc. It may be said that my supposition assumes a presentation of coexistent elements, w^hereas Kant declares that " our apprehension of the manifold of phenomena is always successive," and that " as con- tained in a single moment, each presentation cannot ev^er be anything but an absolute unity." ^ However, I do not know how Kant supposes himself to know this ' synthetic a priori judgment ' : so far as my experience goes, I should say that I am continually conscious of a quite simultaneous manifold of sensa- tions and sense-perceptions. But even granting that apprehension strictly speaking is always serial, it is enough for the purposes of the above objection, if I am allowed to be somehow conscious of a simul- taneous manifold, whether strictly presented or ' Cr. Watson, Selections, p. 57. [M. Miiller, p. 88. The italics are Kant" s, though omitted by Watson and M. Muller. ] VI SUBSTANCE 103 partially represented, and this seems to me quite undeniable. The notion, then, of an absolutely permanent sub- stance does not appear to me necessary for the conception of change in empirical objects, as the transcendental argument requires : relative perma- nence would suffice. There are, however, other difficulties in the argument. The necessity of finding substance in objects seems to be regarded as following from the fact that " Time cannot be perceived by itself " : ^ but the consequence would seem to fail unless substance can be 'perceived by itself: and yet the whole argument implies that this character- istic cannot be attributed to Substance — ' the sub- stratum of all the real' — any more than to time. According to the argument, what we perceive is what is attributed to the substance, not the substance itself: the changeable in things, not the unchangeable. But again: in the statement that "the quantum of substance in nature neither increases nor decreases," we seem to have a synthetic a priori proposition not warranted either by the logical category ' subject that is never predicate ' or the time-determination of permanence. There seems to be a gulf not bridged over between the transcendental explanation and the fundamental assumption of physical science that has to be explained. Granting that we must think the known (mutable) qualities predicated of empirical things as qualities of a subject that cannot itself be thought as a predicate, and granting that this subject ' Fiir Hich vjahrgenommen. Cf. Watson's Selections, p. 107. 104 PHK METAPHYSICS OF KANT lkct. iiiust ho tlu>uglit as j>onii;nu'iit, wliy must it also be thought as having dofiuite quantity ? No doubt we do think thus of tlu' mutter wliicli we conceive as identical throughout the processes of change occurring in inorganic things : but when we turn from these and examine our thoughts of organic things or of persons, we find no similar need of quantifying, no disposition to quantify, that which we conceive to remain identical amid change. Take the idea of an animal — a dog : we have in the notion of a dog the conception of something that remains identical from birth to death, through a varying complex of phe- nomenal change to which it furnishes the bond of unity : but we do not in this case quantify the iden- tical somewhat : the idea of a quantum of ' caninity ' that is not increased or decreased is absurd to us. Take, again, the idea of a person : contemplate a life in its psychical aspect. We have a stream of consciousness varying in volume, and in parts vary- ing markedly in intensity : and we conceive the mind that is the subject of all this experience as having faculties and emotional susceptibilities that grow and decay : but to the person, the self-conscious self that remains identical through these varied changes, we cannot without absurdity attribute quantity. I submit therefore that this notion of an ' unchangeable quantum ' must not be allowed to slip in, as involved in the notion of a permanent subject of mutable phenomenal predicates : it demands a transcendental explanation on its own account, and I cannot see that Kant tries to give this, or where his system could VI SUBSTANCE 105 get it. Nay, could he conceivably show this consist- ently ? For it was ' subject that cannot be thought as predicate ' that was argued to be necessary, but by quantifying we surely give it a predicate ! Other difficulties arise, when we ask which kind of quantity Kant means to attribute to his permanent substance, extensive or intensive, or both. There seems no, doubt that he conceives his Substance as extended in space, for he identifies it with the Matter which physicists assume to be permanent. It remains, therefore, to ask whether the parts of this extended substance differ in their intensive quantity or not. He has already, in discussing the 'Anticipations of Perception,' rejected the assumption that " das Reale im Raume allerwarts einerlei sei ": ^ hence we must suppose that the parts of his Substance have different intensive quantities. But thus his Substance turns out to be an aggregate of heterogeneous substances : and yet, as the ground for assuming its existence was that we might have something to represent, in Mr, Caird's words, the "unity or self-identity of time itself," this hetero- geneity is surely a very singular and inappropriate characteristic. ^ [M. Mliller, pp. 152 f. : "The 7-eal in space must always be the same."] LROTURH VII CAUSALITY, COMMUNITY, MODALITY I PASS to the Second Principle under this head, that " all changes take place in conformity with the law of connexion of cause and effect," I have already explained the ' schematism of the category ' here in- volved, by which the abstract notion of ' dependence in thought ' of ' Reason and Consequent ' is at once limited and rendered applicable to phenomena by the time-determination of sequence : so that ' Cause ' as applied in physical science means not only " that on which the effect depends," but " that antecedent phenomenon or complex of phenomena after which the effect must follow." It is certainly with this definite temporal meaning that modern science has investigated causes ; — since final causes, in accordance with Bacon's witty suggestion, have been consecrated to heaven, as holy virgins, unfruitful through their very holiness. I took this [principle of causality] as the leading illustration of Kant's Schematism,^ because in this case the distinction and relation of category and schema is as intelligible, natural, and 1 [Cf. above, pp. 65 f.] 106 LT. VII CAUSALITY, COMMUNITY, MODALITY 107 helpful as in some other cases it appears to me forced and misleading. In what I have now to say, there- fore, I shall concentrate attention on the transcend- ental proof offered of the principle. This is at once simple and ingenious : and if it is — as I hold it to be — unsound, it is only on account of the fundamental error of the whole attempt to explain our apparent knowledge of Nature as a complex of changing things, while denying the Common Sense assumption that things other than the mind knowing, if rightly known, are known as they are independently of such cognition. Kant starts with the assumption, before referred to, that the reception and apprehension of the sense- percepts, through which we know or seem to know material things, is always successive — whether the phenomenal characteristics of the object are known as coexisting with (relative) stability, or as ob- jectively successive, following each other in the object. Thus the apprehension of the various elements of the manifold contained in the perception of a house is successive, no less than the apprehension of a ship moving down stream : although in the former case the successive perceptions correspond to objective characteristics conceived as (for the time) stably coexisting, whereas in the latter case an objective succession of phenomena corresponds to the subjective succession. This being so, a further comparison of the two cases shows that in the case of the house the succession of perceptions is arbitrary, need not conform to any fixed order : " my appre- lOS THK MKrATHVSlCS OF KANT i.kct. honsion niiglit bi' right."' But in the case of the ship moving down stream, the order in which the perceptions follow one another in my apprehension is unalterably fixed and determined: I first perceive it higher up the stream and then lower down, and the order of these perceptions is inconvertible. Here, then, we see the conception that we must apply — what we must think — in order clearly to conceive the difference between the merely subjective succession of perceptions, which is universal in our apprehension of any phenomenal fact, and that succession of perceptions by which we apprehend objective changes. We cannot find the difference in the phenomenality of the subjective succession, as contrasted with the reality of the objective changes — if we mean by ' reality ' that the changes occur in things as they exist apart from their perception by human minds : for the notion of change cannot, any more than the notion of time which it involves, be applied to this extra-phenomenal existence. We can therefore only find it in the fixed and determinate order w^hich, as we have seen, must characterise the succession of phenomena when thought as objective. That is, the objective sequence of the phenomena A, B, C, must be distinguished from their merely sub- ' Watson's Selections, p. 113. VII CAUSALITY, COMMUNITY, MODALITY 109 jective sequence as perceptions by the characteristic that A must be thought as necessarily antecedent to B, and B to C, and B as necessarily coming after A, and C after B. But in this thought of necessary sequence, we have the thought of causality : for the idea of a phenomenal cause is the idea of a phenomenon after which another phenomenon which we call the effect must come. If therefore we are to conceive of Nature on the one hand as phenomenal — which sound philosophy requires — and on the other hand as a complex of objects undergoing objective changes — as Common Sense and Physical Science do and must conceive it — we must think all phenomena of change, all events, as subject to a fundamental law of necessary sequence : a law by which any event B is thought as necessarily coming after an antecedent event or group of events A. If it be asked, " Will it not suffice if an objective change must be thought as occurring at a definite point of time, without connecting it with events that have previously happened ? " — the answer is that in pure Time there are no points to which anything can be attached : the difference between one part of Time and another lies solely in the changes that take place in time. If therefore you are to fix the occurrence of a phenomenon to a definite point of time, you can only do this by attaching it to ante- cedent phenomena and thinking it as necessarily coming after them : there is no other way of fixing. In order, therefore, that the conception of objective change — experience of objective change — may be no THK MKTArilYSK'S OV KANT i.kct. possible, a univt'isal priiiciplo ol iiecessaiy coiiuexion o{ all events witli events antecedent in time must be admitted. And this is the principle of Causality. The argument that I have just L!;iven in outline is, 1 think, the most brilliant and persuasive example of the transcendental metliod that tlie constructive part of Kant's treatise affords. There seem to me, however, to be im})ortant reasons for not accepting this exposition of Causality, as an adequate explana- tion of the conception as used in modern physical science ; — still less as establishing its validity if questioned by empiricists or sceptics. Firstly, the necessity of a connexion between an event and its antecedents, which it is thus argued is implied in the conception of objective change — that, being objective, must be fixed at a definite point of time — does not carry with it any explanation of the uniformity which is found in our common conception of empirical causation. Yet this uniformity is indis- pensable if the scientific ascertainment of causes is to be practically serviceable for the relief of man's estate. To ensure any practical result, the ascertainment of causes must enable us to predict : but, as a basis for prediction, we recjuire not merely the principle that every event must have a cause in the sense of necessary connexion with antecedent events, but also the principle that similar causes will have similar effects. It may be said that this is implied in Kant's statement of the principle — as it no doubt is in the common statement of it. In saying that every event has a cause, wx commonly mean to imply that the VII CAUSALITY, COMMUNITY, MODALITY 111 complex of antecedent conditions with which we thus connect the event may recur, and that if it recurs the event will follow : we mean, in short, to signify a uniform connexion of similar pairs of phenomena. And it may be said that Kant no doubt means this, and has a right to assume it, so far as there are recurrent phenomena in nature : for if at any point of time, a given event is conceived to follow necessarily from a certain complex of antecedents, we cannot conceive that if the complex of ante- cedents recurred, its necessary consequent would not recur. The necessary connexion, it may be said, cannot be affected by a mere consideration of the point of time or space at which the events in question happen ; since there can be nothing in mere time [or space] that can affect it. But we may, I think, turn the point of this defence against itself. Position in mere time cannot affect any necessity of connexion between two kinds of phenomena that we have any ground for laying down a priori : but then — as we have just seen — there is no such thing conceivable as position in mere time. The connexion with antecedents that we necessarily give to any objective change by fixing it to a point of time, is a connexion with the whole aggregate of immediately antecedent changes, not with any one part of this antecedent complex of change more than any other part : and we have no ground, empirical or a priori, for supposing that this whole complex antecedent will ever recur. And any special connexion that we have empirical grounds for 11 -J rilK MKTAril VSRiS OF KANT i,kct. coiuriving iK'tweeii the evont in (juoation niid any particular part of its immediato antecedents can have nothing to do with the necessary fixity involved in objectivity of change. In fact, in fixing the position oi' any event in time we most commonly connect it with antecedents to which we do not conceive it to be causally related : e.g. we fix the death of a murtlored man at a particular point in the series of continuous and repeated revolutions of the earth round its axis and of its continuously repeated revolutions round the sun : but we do not usually regard the antecedent part of the earth's movement as having any causal connexion with the murder. It would seem, therefore, that these special causal connexions cannot be deduced from or subsumed under the principle of causality as stated by Kant : this may explain the general necessity of conceiving a causal connexion, but not the complex uniformities that Physical Science is believed to have ascertained. It may indeed be said that as science recognises that every portion of the physical world is connected through gravitation with every other portion, the concept of every event as necessarily connected with the whole complex of antecedent events must be admitted by science to be strictly speaking the true conception. I grant this, but my point is that the principle as so conceived, however incontrovertible, is useless for the discovery of the more special uniformities, by which alone the predictive power of Science is attained : not only cannot these sub- ordinate laws of Nature be laid down a priori, but VII CAUSALITY, COMMUNITY, MODALITY 113 the more general fact that there are such laws cannot be thus laid down : at least Kant's transcendental proof has not shown the possibility of knowing it, apart from specific experience. This last consideration afibrds a transition to "the third ' Analogy of Experience/ [the ' Principle of Eeciprocal Action' : — ] "All substances, in so far as they can be observed to coexist in Space, are in thoroughgoing reciprocity." Now, when we consider this principle in relation to Kant's systematic explanation of the a priori element in our knowledge of phenomenal objects — i.e. as resulting from the application, through a time -determination or schema, of one of the forms of thought manifested in the logical table of judgments, to the empirical data given in sense - perception — we fall back again into the bad system- making of which the earlier mathematical principles supplied unmistakable instances. But in this case we may say that the forced schematism is harmless : that is, I cannot see that Kant even attempts to make it plausible. His ingenuity is not stimulated to invent fallacies such as that of regarding Number as a time -determination, and extracting continuous variation in intensive quantity out of the simple logical opposition between affirmation and negation. But in the category of ' Community ' (interpreted as reciprocal action) there is a violent leap of thought from the form of disjunctive judgment. For the predicates of a disjunctive judgment (A is either B or C) are not mutually dependent in any positive way. 114 THK MKTArilVSR'S OF KANT lkct. tliev are luutuiiUy exclusive : the atHrmatiou of either B or C involves the simple negation of the other as a possible predicate of A. The analogy between this relation, and the mutual dependence of two objects in respect of certain positive characteristics — which is involved in the idea of reciprocal action — is surely faint and far-fetched. The transcendental proof of the principle may, however, be examined apart from this forced and invalid connexion with the form of the disjunctive judgment. The first paragraph of the proof runs thus: — "Things are coexistent which exist at one and the same time. But how do we know that they exist at one and the same time ? Only if in the synthesis of apprehension the order in which the various determinations arise in consciousness is in- different, or can go either from A through B, C, D, to E, or conversely from E to A. Were the determina- tions actually to follow one another in time, that is, in an order that began with A and ended with E, it would be impossible for apprehension to start with E and go backwards to A ; for A would in that case belong to a time that was past, and therefore could no longer be an object of apprehension." ^ From this it would seem that the fact that if, in any apprehension of sense-data, the order in which we pass from one to another of a group of sense-percepts A, B, C, D, E, is so far indifferent that we can either have them so or reversely E, D, C, B, A, then we can know that they coexist. But the proof seems ^ Watson's SdeclUms, p. 118. VII CAUSALITY, COMMUNITY, MODALITY 115 obviously inadequate : for we can experience a series of sounds in one order, and then of similar sounds in the reverse order, without making, or having a right to make, this inference of coexistence. So of pains — e.g. if diflferent parts of the body are pricked first in one order and then in the reverse order. We must take the proposition with the restriction that the percepts in question are regarded as objective ; and being so regarded are assumed to have existed in the interval between the perception {e.g.) of A in the first series and its perception in the second. Suppose we make this assumption explicitly. Then it would seem that coexistence — and, therefore, relative per- manence — prima facie is proved by the two series of experiences. And it does not seem that, so far as co- existence in time is concerned, there is any necessity to assume that the substance to which percept A is referred is in a relation of reciprocal action with the substance to which percept B is referred. The notion of ' action ' seems only to come in when change is experienced : but in the case supposed we seem to assume an absence of change. And in fact it will be seen, from the statement of the principle, that Kant is obliged to introduce the condition of coexistence in space : and the argument to prove that coexistence involves reciprocal action requires this condition. "Now suppose" — Kant's proof continues — "that a number of substances could be observed, each of which was so completely isolated from the rest that none acted upon any other or was itself acted upon ; then I say that these objects could not possibly be 116 Till-: MKTArilYSlCS OF KANT lkct. ol):v I'rofessor Watson — slunvs tliat Kant means that only this can he known a priori to be possible : e.g. as his illustra- tions indicate, the possibility of spatial figures, of continuous magnitudes, of permanent substances, of causal connexions, etc. Such things we can afhrm to be possible : thougli (except in the case of spatial figures) it would be an understatement for Kant, who claims to have proved them to be necessary. But as regards any other connexion of ideas in thought, we cannot lay down a priori that it is possible in fact, but only a posteriori, that is to say, on empirical grounds. Thus the notion of a faculty of prophecy, and of what we now call telepathy, are in Kant's view " concepts the possibility of which has nothing to rest on, because it is not founded on experience and its known laws." ^ Kant, however, does not seem to say that he has thus given us two distinct conceptions of ' possibility ' in its empirical use : (a) that which agrees with the formal conditions of experience ; {h) that which accords with empirical analogies.^ Of course prediction and telepathy in no way disagree with the formal conditions of experience. They are only not, according to Kant, founded on empirical laws scientifically ascertained. We certainly require a distinction between the idea of contradiction (or non - contradiction) of necessary a p>riori laws of Nature and the idea of ' [M. Miiller, p. 194.] ' I say 'analogies,' because if the object in question was something of which the existence could be strictly inferred from empirical facts, it would be actnal. VII CAUSALITY, COMMUNITY, MODALITY 125 correspondence (or non- correspondence) to merely empirically ascertained uniformities ; and I should have thought it more in accordance with Kant's system — and the words of the first postulate — to limit ' impossibility ' to the former. And indeed he does not exactly say that telepathy, etc, is impossible, but only that its possibility has nothing to rest on and cannot be tested. So that it is not to be called possible. It may be admitted that there is no scientific use in discussing conceptions to which only possibility in the wider sense attaches : i.e. for the actuality of which no empirical evidence is adduced. But then, as Kant quite well knew, that was not the case with prophecy or telepathy.^ I cannot but think that some other word than ' impossible ^ would be more appropriate to signify things or processes which have no analogy in scientifically ascertained empirical eff'ects. And when this was recognised it would be seen that this narrower notion of ' empirical possi- bility ' is necessarily vague and indeterminate. Who is to say what is ' possible ' in the narrower sense ? ^ I now turn to the other two postulates. It would appear from the second postulate that the two notions ' actual ' and ' necessary ' involve each other (so far at least as the larger part of the application of the notion ' actual ' is concerned). For we are told that " what is hound up with sensation is actual," but what does ' bound up with ' {zusammenhdngt) mean ? If anything is rightly judged to be ' actual ' which is not a datum of sense-perception, what can 1 Swedeuborg ! "^ Cf. Mr. Wells's War of the Worlds. 126 lllK MKrArilVSU'S OK KAXT lkct. it be except sumellniig i>f wliicli tlie existence is co«^entlv inforreil from siieli data ? And Kant him- self says that, in onler rightly to judge anything to be actual, " we must be aware of its connexion with some actual perception according to the analogies of experience which represent all real connexion in an experience."^ But the third postulate defines the necessary as " that which, in its connexion with the actual, is determined in accordance with universal conditions of experience." It is rather difficult to see how the two can be distinguished : and, in fact, the distinction Kant seems to draw between them does not seem to me satisfactory. He says that no exist- ence can be known to be (even ' conditionally ') necessary, " except the existence of effects following from given causes in conformity with laws of causality. It is, therefore, not the existence of things or sub- stances that we can know to be necessary, but only the existence of their state. . . . Substances can never be regarded as empirical effects."- On the other hand, he seems to hold that we may know that a thing ^ actually exists, though it cannot be perceived, when it is inseparably " related to certain perceptions [mit einigen Wahrnehmungen zusammenhdngt). . . . Thus from observation of the attraction of iron filings we know that a magnetic matter pervades all bodies, although our organs of sense are so constituted that we cannot perceive it." But how can this be known except by inferring a causal connexion, and if the ^ Watson's Selections, p. 124. ^ Watson's Selections, pp. 125 f. ^ Substance, I suppose. VII CAUSALITY, COMMUNITY, MODALITY 127 connexion is not necessary, why are we said to know "that a magnetic matter pervades all bodies"?^ It seems evident that Kant accepts as certain infer- ences that are based on a connexion not necessary but only empirical : but how he would justify this I do not know. ^ lu fact, it was a somewhat hasty inference. LECTURE VIII THE TRANSCENDENTAL DIALECTIC So far I have been examining, impartially but closely, the main constructive part of Kant's philo- sophy, i.e. his account of the a priori elements in our commonly accepted knowledge of the physical world. I turn now to consider that part of his work which is primarily, though not solely destructive : destructive, that is, of the ' bad metaphysics ' which attempts to attain, professes to expound, * rational ' as contrasted with empirical knowledge — knowledge, that is to say, going beyond the limits of possible experience. Before we examine Kant's destructive work, it will be well to have before us in outline the soi-disant system of knowledge that it was designed to shatter. The detail of it I propose to reserve in order that it may precede, in each case, the detail of the criticism. But it will be well to have the general plan of it before our minds, before considering the general plan of Kant's attack. From the starting-point of modern metaphysics in Descartes a definite triplicity of Being had always 128 LT.viii THE TKANSCENDENTAL DIALECTIC 129 occupied the thought of the metaphysician. The nature of minds, of the world of matter, the relation between the two, the relation of both to God, and the philosophical conception of God, as distinct from the revelational conception — these were throughout the main topics of speculation. The centuries of mediaeval social life in which religioas conceptions had swayed men's thought with overwhelming predominance had — in spite of the new sway of intellectual interest towards the comprehension of the laws of the physical world — still left the Christian conception of the soul, the world, and God, as that which ordinarily deter- mined the ground-plan of the metaphysician's thought. For him, as for the vast majority of his educated contemporaries, the idea of the world with which he started was that of an aggregate or system of material things, created by the primal and eternal Intelligence for a finite existence. In this world men, composites of soul and body, were placed, but to it they only partially and temporarily belonged in respect of their souls, which, unlike the world, were created for endless existence. As a philosopher, he sought to obtain reasoned convictions independent of revelation, on the matters to which these Christian beliefs related : but, however sincere his love of truth, he was — especi- ally in Germany up to Kant's time — under strong moral pressure to arrive at conclusions in harmony with the established religious beliefs : at any rate, he could hardly think without serious regard to them, or diverge from them without anxious consideration. Now it soon appeared that the questions prominently K 130 TllK MKlArilVSlCS OF KANT lect. siiiijgestod from tliis point of view wore not questions which experience iilone, however systematised, enabled the thinker to answer. As regards the primal Divine Being, and his rela- tions to the world and to human souls, this conclusion [as to the insulHciency of experience] seems to have been accepted unhesitatingly. The philosopher found current a Revelational Theology, based on the state- ments of Scripture and the decisions of ecclesiastical authorities ; but he did not find an experimental theology. There was no body of systematic know- ledge, professing to be derived from observation of the Divine nature and action. The philosopher's Theology, therefore — however important the data it might draw from experience, — must be in the main worked out by processes of abstract reasoning. But even in the case of Minds and the Material world, though empirical study supplied him with a mass of knowledge of particular facts and law^s, it did not take him far towards an answer to the fundamental questions above indicated. Could man know, apart from revelation, that his soul was naturally immortal, and therefore fundamentally independent in existence of the material organism with which it was temporarily so mysteriously connected ? At any rate, experience could not tell him this directly or taken alone : since, however complete the diversity between psychical facts — thoughts, feelings, etc. — and the movements of organic matter that seem to be causally connected with them, we have no experience of the former except as accompanied by the latter. Similarly as VIII THE TKANSCENDENTAL DIALECTIC 131 regards the material world, when we seek to penetrate its ultimate constitution, we have to go beyond experience. We know, indeed, from experience, the compositeness of ordinary empirical things, because we have experience of their breaking up into parts : but we soon convince ourselves that these parts are not ultimate, and that if we are to attain a true conception of the ultimate constituents of the physical world, it must be by processes of abstract reasoning.- And it is still more evident that if we raise the obvious questions as regards the physical world as a whole, whether it had a beginning or has always existed, whether it has bounds or is infinitely extended, ex- perience cannot furnish answers. As little can experience help us, if we raise questions as to the ultimate explanation of what I may call the par- ticularity of the material world. In the view of Physical Science, through the work chiefly of Coper- nicus, Galileo, Kepler, and Newton, the particular state of the material world at any time had come to be definitely and scientifically thought as the necessary consequence of the particular state at any preceding time, the changes between the two times being explained by the operation of the universal laws of matter in motion. But however far back we trace this sequence in thought, the collocation of the parts of matter with which we leave off" seems as arbitrary and contingent — something that might just as well have been otherwise, — as much, therefore, needing a reason for its existence, as the present state. Could philosophy acquiesce in conceiving the process of the 132 IHK MKrAl'lIVSUS OF KANT lect. world us an oiulloss chain i)f arbitrary and contingent facts ? If not, what cxi)hination could be found ? The explanation suggested by traditional theology referred the original collocation of matter in space to the wise creative choice of the primal Being, God. And the purposes determining this choice were usually taken to relate to the life of humanity. Could philosophy convert this into a cogently reasoned explanation ; and if so, could it deal with the question lying behind, as to the existence of the primal Being, God, by conceiving his existence as intrinsically necessary ? Or if not, what other system of thought on the subject could satisfy reason ? However these questions were answered, it seemed obvious that they carried the thinker beyond the limits of experience. In this way the subject-matter of metaphysical inquiry came to present itself — especially to the orderly and systematic mind of Wolff — as naturally divided into three branches, Rational Psychology, Rational Cosmology, and Rational Theo- logy. In the two former cases a distinction was drawn between the Rational and Empirical methods as applied to minds and to the material world respect- ively. Not that Wolff aimed at an absolute separation between the two, as in both cases important data for the rational studies are derived from experience : but it was the rational studies as distinct from the empirical that supplied answers to the (metaphysical) questions above indicated. What these answers were, we will consider more closely in later lectures. It is enough now to say VIII THE TRANSCENDENTAL DIALECTIC 133 that the mind was conceived as a substance, simple not composite, and therefore naturally indestructible and imperishable, — since only the composite is de- structible.^ The material world was conceived as an aggregate of simple substances, which were not actually extended so as to fill space — since what is extended must, it seemed, be divisible — but were arranged in an order which is confusedly perceived as continuous extension. This world was conceived as having; come into existence through the creative act of a supremely perfect Being, whose existence is necessarily ' involved in its essence,' i.e. a supremely perfect non-existent Being was held (as by Descartes) to be a contradiction in terms. These, then, are the chief metaphysical conclusions — or at any rate leading examples of the conclusions — which Kant in this third, destructive part of his Transcendental Philosophy set out to expose, as attained by an illegitimate and illusory exercise of the reason. For his general negative conclusion we are, I think, fully prepared, if we have followed with assent the arguments contained in the two first parts. If Time does not belong to reality, as it exists apart from our consciousness, but is only a form in which things appear to the human mind, it is clear that the very question {e.g.) whether the world — the real world — had or had not a beginning in time is as unmeaning as {e.g.) the question whether the soul is square or oblong. ' Experience seemed to show tliat 'destruction' was really changing rela- tion of parts, 'breaking up.' i:U Tin: MKTArUVSlCS OF KANT i.kct. lUit Kant is not content to cut oil' bad metaphysics l)v sucli a simple and sweeping inference as this. He iiesires to exhibit in detail the fallacies into which the human reason is inevitably led, when it seeks know- ledge beyond the limits of possible experience; partly because, as he tells us, these transcendental fallacies do not vanish when once refuted — as ordinary logical or formal fallacies do, should we happen (c.ff.) inadvertently to perpetrate a syllogism with an undistributed middle term. The * transcendental illusion ' continually recurs and has continually to be corrected. But why are we thus condemned to this continual recurrence of error ? The explanation according to Kant is this : Deep in our reason lie fundamental rules for its use, of the highest value in its empirical employment, but having inevitably a tendency to present themselves as objective principles for determining the characteristics of ' things in themselves,' though their proper applica- tion is merely to produce a certain systematisation of our conceptions, to aid our intellect in a compre- hensive grasp of experience. What, then, are these ideas of the Keason, and how are we to obtain a systematic view of them ? We shall expect, from the method used in the second part to obtain a systematic view of the pure concepts of the understanding, that common logic will again furnish the plan of the system, i.e. that the ideas of the pure Reason will be correlated with the logical forms of reasoning. For the ordinary operation of Reason — in the narrower sense, in which VIII THE TKANSCENDENTAL DIALECTIC 135 it is distinguished from understanding and sensibility, regarded as forms of a priori knowledge — is shown in reasoning on empirical matter, that is, in processes of mediate inference, in which rational conclusions are drawn by a combination of judgments in syllo- gisms or, in most cases, in a series of syllogisms. Now as Kant found in common Logic three forms of reasoning, categorical, hypothetical, and disjunctive ; and as the illusory Metaphysics which he set out to criticise — the professed knowledge transcending possible experience — was divided, as we have seen, under three heads. Psychology, Cosmology, and Theology, the correlation of the two triplicities was irresistible. But here — as in the case of the more elaborate system of the pure understanding — the correlation is not equally satisfactory throughout : the symmetry is partly forced, and so far as it is forced, it obscures rather than illuminates the matter on which it is imposed. This I shall try to show more clearly in the sequel : but it is needful to state it, in giving a preliminary view of the system. The general function of the understanding is, as we saw, com- bination or synthesis of phenomena, constituting empirical objects and connecting them as possible objects of experience. But it does not aim at putting together these objects of experience into a whole. The idea of an absolute totality of possible experience — which the mind finds itself compelled to form — is an idea of the Reason in the special sense in which its function is distinguished from that of Understanding 136 THE MKTAl'UVSirs OK KANT lkct. niul that, of iSeiitiibility. hi tliis special sense the loi/ica! fuuction of the Reason is to combine the juilijjments of the understanding, as the understanding combines the percepts of sense. Now the absolute totality of all possible experience cannot itself be ex[)erienccd : and therefore in seeking to know this absolute totality Reason inevitably aims, and must aim, at transcending Experience. In speaking of ' absolute totality of experience ' I have used the phrase of the Prolegomena. In the Kritik Kant makes the idea more definite : — " A transcendental conception of reason is . . . just the conception of the totality of conditions of anything that is given as conditioned. Now the unconditioned alone makes a totality of conditions possible, while conversely the totality of conditions is always itself unconditioned : hence a pure conception of reason may be defined generally as a conception of the unconditioned, in so far as it contains a ground for the synthesis of the conditioned." ^ Now, what is the exact meaning of the " unconditioned which alone makes a totality of conditions possible and is itself unconditioned " ? The example of Kant's meaning which is certainly most easily intelligible is to be found in the cosmological ideas. Let us take the case of the synthesis of cause and effect — which we have already seen reason to regard as supplying the germ of Kant's system of concepts of pure understanding. The principle of causality is that any event presupposes an antecedent event ' Watson's Selections, p. 141. VIII THE TEANSCENDENTAL DIALECTIC 137 as its necessary condition or cause, but this cause, according to the same law, must have an earlier antecedent cause, and this again a still earlier, and so on, through a retrogressive series. Now reason demands that the totality of this series should be, not indeed specifically known — reason is not so exacting — but thought as a reality. It must therefore admit of being thought without contra- diction. The demand that the series be completed, that is to say, forces reason to the conception of an uncaused or free cause, which yet — we find — cannot be thought without violating the principle of causality that leads to the series : and thus we get an ' Anti- nomy.' Similarly, a material thing existing in space is conditioned (determined in its position) by the coexisting matter with which, according to Kant, it is necessarily connected by reciprocal action and reaction. And here again Reason passes through the series of conditions to the unconditioned, and raises the question of the relation of the material world as a whole — which is necessarily unconditioned by any coexisting matter — to Space, and asks whether the world is finite or infinite in exten- sion. But it is not so easy to apply this idea of the totality of conditions, involving the idea of the unconditioned, to the principles and reasonings either of Rational Psychology or of Rational Theology. I think that in both these cases the conception of a totality of con- ditions or unconditioned is somewhat strained. And I am confirmed in this view by comparing the Pro- UiS TllK MK'rArilVSICS OF KANT lkct. letjomrnn witli the Kritik : ' it will be observed that in the J^rolei/ouwna * coiulitiona ' arc only spoken of in the case of the cosmoloj^ical idea. I do not mean that Kant's view was altered, or that he had abandoned the extended conception of the Kritik when he came to write the Prolegomena, but only that the more limited use of the notion of * totality of conditions ' as equivalent to ' unconditioned ' is the narrower use. He therefore naturally fell back to it in what was designed to be a more popular exposition of his view. However, no doubt, we must take, and try to understand, Kant as extending the conception of conditioned and unconditioned to include all three cases. In the case of Rational Psychology he con- siders that the psychological idea is an * unconditioned of the categorical synthesis in a subject,' or otherwise * the absolute or unconditioned unity of the thinking subject.' To understand this, we must examine more closely what he calls the * Paralogisms ' of Rational Psychology. One disadvantage of the forced symmetry of this part of Kant's system — the correlation of the three branches of fallacious Metaphysics, Rational Psy- chology, Cosmology, and Theology, with the three logical forms of reasoning, categorical, hypothetical, and disjunctive — is that too complete a separation 1 In the former we find "first, the Idea of the complete subject (the sub- stantial) ; secondly, the Idea of the complete series of conditions ; thirdly, the determination of all concepts in the Idea of a complete complex of [all] possible [being]" (MaliafTy's edition of the Prolegomena, § 43, p. 119). In the latter "we have, first, the unconditioned of the categorical synthesis in a su^ect ; secondly, the unconditioned of the hypothetical synthesis of the members of a series ; thirdly, the unconditioned of the disjunctive synthesis of the parts of a system " (Watson's Selections, p. 141). VIII THE TKANSCENDENTAL DIALECTIC 139 is established between the three branches. This will, I think, appear clearly hereafter, as regards the relation of the cosmological ideas to theology : i.e. it will appear that the questions raised in the two later of Kant's four antinomies — the question (1) whether to explain the whole series of caused changes of w^hich the world-process consists, we have to suppose a free causality ; and (2) whether to explain the same series of contingent facts, regarded as contingent, we have to suppose an absolutely necessary Being — are prima facie theological ques- tions. I do not mean that Kant is not justified in giving them a cosmological form : that we will hereafter consider. But certainly the affirmative answers given to these questions in the pre-Kantian metaphysics, which Kant is attacking as fallacious, were theological answers : the free causality supposed at the beginning of the causal changes of the (supposed) created world was the causality exer- cised by the primal Being God, and it was this primal Being who was conceived to be absolutely necessary, A somewhat similar forced separation occurs between the Rational Psychology and the Rational Cosmology : and it is noteworthy that this forced separation increases the difficulty of accepting Kant's account of the transcendental idea corresponding to categorical reasoning. "We have," he says, "firstly the unconditioned of the categorical synthesis in a subject"; and soon after he identifies this idea with "the absolute or 140 THK Mi:rAriIVSK:S OF KANT lkct. uncoiulitioned unity of the tliinkiiiir subject."' But, if we regard the idea thus delined as representing what reason tries to lind in empirical objects — rather than wliat she professes to succeed in finding — and that is the aspect in wliieh the Transcendental Pliilosophy presents the ideas of reason, why should she confine her attempt to Minds or Thinking Subjects ? Every material thing, no less than every thinking person, presents to our thought, when logically analysed, the synthesis of predicates or qualities inhering in a subject or substance. Why should not reason seek for the real substance at bottom — the subject pure and simple that cannot be thought as a predicate — in the case of material things, no less than in the case of persons. And, indeed, was it not evident that Reason — the reason of metaphysicians — had occupied itself with this question.' And in fact, in the Prolegomena, Kant fully recognises this : and thus begins his account of the Psychological idea : — " It has been long since observed, that in all substances the proper subject, that which remains after all the accidents (as predicates) are abstracted, consequently that which is itself substantial, is unknown. . . ." ^ Here we have, I conceive, the critical view of the proper use and application of what I may call the categorical idea, the idea of the absolute subject in anything : i.e. that in it which cannot be thought as a predicate. ' Watson's Selections, pp. 141, 142. ' Cf. Locke, Essay concerning Human Under starviing, Bk. ii. ch. xiii. § 19 ; ch. xxiiL §§ 1, 2. ' Prolegomena, § 46, Mahaffy's ed. p. \2Z fin. VIII THE TEANSCENDENTAL DIALECTIC 141 The only reason for confining his treatment of it to the Thinking Subject, is that only in this case did Reason appear to him to have deceived itself into thinking that it had found the absolute subject it sought for.^ But the fallacious Metaphysics that Kant goes on to attack, here and in the Kritik, does not content itself with affirming that in the Ego we find an absolute subject — i.e. a subject which cannot be thought as a predicate : its fallacy rather lies in a further assumption that the Ego is a simple substance and therefore indestructible. But similar simple, indestructible substances had also been not only sought but (believed to be) found by the reason of metaphysicians in the material world : and the search for such permanent indestructibles was by no means identical with the effort to find a subject which could not be thought as a predicate. For the simple elements of which AVolff", for example, supposed material things to be composed must, if definitely conceived at all, be conceived not as subjects without, but as subjects with predicates. In popular physics and by physicists generally, these ultimates were conceived as having extension, in- volving size, shape, and absolute incompressibility {i.e. absolute resistance to any forces tending to annihilate their extension). And if, to avoid the difficulty of conceiving anything extended and yet not composed of parts, we follow Wolff* in supposing the ultimate elements of matter to be unextended and without size and shape, still the unextended alone ' Cf. Prolegomena, Mahaffy's ed. p. 124 /n. l-rJ THK MKTAl'llVSlCS OF KANT i.kct. viii must liuvo some predicates : and in fact WolfV's atom had not only an essential force or principle of activity by which it was distinguished from other atoms, but also some passive force by whicli the phenomenon of inertia in composite bodies was explained. Now the question of tlie ultimate elements of matter Kant treats under the head of Cosmology, and the symmetry of his system requires this. But by thus separating the question whether the soul is a simple indestructible substance entirely from the question whether such substances are to be found as elements of the material world, he certainly divided questions which had been connected in the thought of the philosophers he was attacking : and it is important to notice this because it is, I think, this previous connexion of the questions which furnishes the real explanation of what Kant calls a Paralogism. Had Wolff and others merely considered the Ego from a psychological point of view, and merely found in it the ' subject incapable of being thought as predicate ' which Reason is said by Kant to seek, I cannot see why they should have immedi- ately attributed to it the predicates of permanence and indestructibility : the inconsistency would have been too palpable. It was because their Rational Cosmology influenced their Ontology, and, through this, their Psychology, that they were misled into attributing to the Ego the characteristics which a cosmological line of thought led them to attribute to the ultimates of matter. LECTURE IX RATIONAL PSYCHOLOGY The general view that I take of the part of Kant's discussion which he calls the * Paralogisms of the Pure Reason ' may be briefly and simply expounded. I am convinced of the truth of this general conclusion, that the propositions of the older Rational Psychology which he attacks — viz. that the human soul is a simple substance, in its nature indestructible and therefore naturally immortal, and having relations, represented in our thought as spatial, with the other simple substances which are the elements of the material world — that these doctrines, regarded as synthetic propositions a jpriori, are invalid and illegitimately assumed. And further, that they must be known a priori if at all, i.e. by considering the general notion of a self-conscious being : they cannot be proved from our experience of human minds, as known to us from introspection, or what Kant calls the * inner sense.' These con- clusions I accept as true : and of their negative importance there can be no doubt. But the reason- ing by which Kant tries to prove them seems to 143 l-M THH MKrArilVSU'S OF KANT i.ect. mo only partiiilly souiul : I tliiiik it contains a tiuulaniontul misajtpielion.sion of the knowledge of self which we obtain througli self-consciousness. My grounds for this double conclusion I will now briefly state. But tirst, I would again point out that (as explained at the close of the last lecture) the separation which the plan of Kant's system — his correlation of the triplicity of the transcendental ideas of the Reason with the triplicity of the logical forms of reasoning — leads him to make between Rational Psychology and Rational Cosmology, puts us at the wrong point of view for understanding how the doctrines that he is assailing were arrived at. It was precisely because Leibniz, Wolff, and their followers did not completely separate Psychology from Cosmology, did not regard the investigation of the nature of mind as quite apart from the investigation of the ultimate nature of the material world with which we find minds mysteriously con- nected, but on the contrary regarded minds as subject to the fundamental laws of the material world as rationally comprehended — it is, I say, just because of this that they inferred from the essential unity of the self - conscious self its substantial simplicity and therefore its natural permanence and indestructibility. I have said that the threefold division of Meta- physics into Psychology, Cosmology, and Theology is adopted by Kant from Wolff — by whom, I think, it was first explicitly introduced. But Wolfi" prefixed to these three branch-studies a general study of the IX EATIONAL PSYCHOLOGY 145 characteristics of Being and our knowledge of Being, which he calls Ontology. The conceptions and propositions of Ontology he regards as applicable, from their general character, to all the branch-studies : and it is here that we find the conception of a simple being, contrasted with a composite as essentially indivisible, and therefore — in contrast with composites — as incapable of coming into existence except by creation or ceasing to exist except through a correspondingly supernatural fiat of God. This notion, then, he and his followers applied to psychical as well as to physical facts. The self is recognised as such a simple being, unextended and therefore without parts, and so naturally indestructible. It accordingly takes its place among the ultimate elements of the physical world, which are similarly conceived to remain, indestructible and physically immutable,^ through all the processes of physical change in which the composite matter that we empirically know is con- tinually being broken up into parts which enter into new composite substances. The validity of this inference from mutable and destructible composites to indestructible and physically immutable elements, we shall presently have to examine in its application to the material world : for this, in fact, constitutes the second of Kant's cosmological problems, that lead, according to him, to antinomies. My point now is that the ^ In saying that these substances were physically immutable, I mean that if they changed they changed from an inner necessity of development, not from the operation of external forces. L 146 THK MKrArUVSlCS OF KANT lkct. separation, which Kant's system imposes on him, between the question as to the simple substantiality of Mind and the question as to the simple substances underlying Matter, is a forced separation. And I may add that 1 find striking evidence of this in a passage — which Watson has not selected — as to the ' interest of reason ' in these antinomies. For in speaking of the antithesis in the second antinomy ' he characterises the antithesis as apparently hostile to morality " if our soul shares the same divisibility and perishableness with matter " ■ — thus fusing the fundamental question of Rational Psychology with the cosmological question relating to completeness of division of a material object. Of this more presently. In any case I agree with Kant in regarding as illegitimate the transfer of the predicate of natural indestructibility to the self-conscious mind, as though it were somehow necessarily connected with the notions of unity and identity of the self- conscious self. I find no such necessary connexion, and therefore find any reasoning in which it is assumed fallacious : — you may call it, if you like, a paralogism. At the same time, Kant's exhibition of the fallacy does not seem to me exactly to hit the right point. He admits, of course, the essential unity of the self- conscious mind, as a perceiving and knowing subject : indeed we may say that it is the special characteristic * In this antinomy the thesis is : "Everything in the world consists of simple parts " ; and the antithesis : ' ' There is nothing simple, but everything is composite." ' [Max iliiller's translation, p. 408.] IX EATIONAL PSYCHOLOGY 147 of his system of philosophy to lay stress on this. The unity of self-consciousness is for him the source of all unity, all synthesis or connexion in the objects of empirical knowledge, as constructed by the under- standing out of the data of sense. The essential function, as we saw, of the understanding, in the application of all its forms or categories, is synthesis, unification : and the root of this synthesis, what renders this unification of sense-perceptions possible, is the necessity that every sense-perception should be referrible to a self, and capable of being thought — if not always actually thought — as its perception. It is because all objects of experience are thus necessarily objects of the possible experience of a single conscious percipient mind, that Kant holds us to have the a priori knowledge of their con- stitution and relations set forth in the second part of his treatise. But it is just because of this startling extension of the meaning and function of self- consciousness in Kant's Philosophy, that he is disposed to draw a sharp line between the unity and identity of self as a subject of knowledge and its unity and permanence as an object of knowledge : and to regard as a paralogism the inference of the latter from the former. He does not deny that ' I ' stands in our ordinary thought for Self as an object of thought, no less than for Self as a necessary subject of thought. He expressly says that the transcendental conception 'I think' — the common ' vehicle ' of all transcendental conceptions — thougli as transcendental it is free from all empirical 14S Till': MKl'Al'llVbiCS Ui"' KAM' i.KCT. oloiiionts, "ycL serves to distiiiguisli l)ct\vcoii two (litVoreiit kiiuls of ohjocts, from the ditfereiit ways ill wliich they are rehited to consciousness. /, as tliinking, am an object of the inner sense." ' Admitting this, he argues, i)i tJic Kr'ttih, that Ivational Psychology is bound to obtain its alleged synthetical truths strictly a priori, i.e. without reference to the inner experience in which I know myself as an object : and that if we examine its propositions with this strict condition in our minds, we find that it has made an illegitimate transition from the characteristics of the self as subject to characteristics which can only belong to self as object. I partly agree with this : but I think that in his exhibition of the paralogism Kant does not state the illicit transition quite correctly : I think the illegitimate inference of the Leibniz-Wolflian meta- physics is not simply from ' subject ' to ' substance,' but from ' 07ie subject' to ' simiDle substance.' In short, the Wolffians might answer truly that this rigid separa- tion between Rational and Empirical Psychology was Kant's and not theirs, and was an unthinkable separa- tion. For identity amid change, and therefore relative permanence, appears to me essential to the thought of a subject : and if Kant says ' permanence belongs to it qua object,' the answer is that it is essentially an object to itself. And Kant's argument seems to me self-contradictory in its subtlety of division of 'subject' and 'object.' "The conception of a thing that can exist by itself as a subject," he says, ' ^&\;&on!& Sdectivns, p. I4l> fin. IX KATIONAL PSYCHOLOGY 149 "does not carry with it objective reality . . . because we cannot understand how an object of that sort could exist at all." ^ " We cannot understand" — but the 'we' are existent 'we's,' subjects that must conceive themselves as existing objects. But if the Wolffians were to make this answer, if they were to admit that the conception of Self as sub- stantial was derived — and must be derived — from that empirical cognition of Self as an object of introspection or inner sense, which is necessarily involved in the most purely speculative thought, I should then urge that from this empirical cognition of Self as object of introspection we can only be justified in attributing to it permanence during the psychical life of the individual : and not in attributing to it the absolute permanence and indestructibility — unless annihilated by creative fiat — which constitutes the important dogma of the Eational Psychology here assailed. And this line of argument is adopted by Kant in the Prolegomena} But in the Kritih he seems to go further and expressly deny the application of the predicate of permanence — even to the limited extent to which experience justifies it — to the Self or Eoro.^ Kant seems to have been led to this view o ' Watson's Selections, p. \b'2fin. ^ Cf. § 48 init. (Mahaffy's trans, p. 126): — "If therefore from the con- cept of the soul as a substance, we would infer its permanence, this can hold good as regards possible experience only, not [of the soul] as a thing in itself and beyond all possible experience." ^ " Now in inner perception there is nothing permanent, for the / is merely the consciousness of my thinking. So long, therefore, as we limit ourselves to mere thinking, we are without the necessary condition for the application of the concei)tion of substance to the self as a thinking being ; we are unable, in other words, to say that the self is an independent subject." — Watson's Selections, p. 153. 150 THK MKrAl'llVSlCS OF KANT lf.ct. hy the roniark;il)le l)arrennes.s of content of the notion of Self. "The simple idea /," he says, "is so completely empty of all content, that it cannot he called even a conception, hut merely a conscious- ness which accompanies all conceptions. This / or he or it, this thing that thinks, is nothing but the idea of a transcendental subject of thought = x, which is known only through the thouglits that are its predicates, and which apart from them cannot be conceived at all."^ He afterwards speaks of it as "the very poorest of all our ideas (Vorstellungen)." '^ Now perhaps this language is justifiable if the ' I ' of the thought * I think ' is treated as strictly transcendental and examined in rigorous abstraction from experience. But in saying that "in inner perception there is nothing permanent, for the ' I ' is simply the consciousness of my thinking," ^ Kant has abandoned the transcendental ground ; and here I think he is guilty of a transition as illegitimate as that which he rightly attributes to his opponents, although in an opposite direction. That is, he tries to reduce the notion of Self as object of inner experience to the meagreness of the ' I ' of transcend- ental thought. Now of the self which introspection presents to us as a thinking thing, introspection doubtless tells us little enough: all the particularity of the mind, all that interests us in our thought of ourselves and other minds as relatively permanent objects of thought in contrast with the more transient » Watson's Selections, p. 148. 2 [-/^-^^ p_ i_;;o.] » [Ibid. p. 153.] IX EATIONAL PSYCHOLOGY 151 states of consciousness, we only know by inference from tlie transient and ever -varying element of inner experience. But still it is going too far to say that the self presented in inner experience is merely thought as a logical subject ivithout predicates. However little ' I ' know of ' myself in introspection, I still know myself as one and identical, perduring through the empirical stream of thoughts, feelings, and volitions. This cognition may be liable to error — I find infallibility nowhere in human thought — or again it may seem unimportant : but it is presented as imme- diate and is as certain as any empirical cognition, and in it I certainly find ' given ' — if anything is ever * given ' — the empirical permanence which Kant — in the Kritik — denies. LECTURE X THE MATHEMATICAL ANTINOMIES I NOW pass to the Cosmological Idea and the Anti- nomies of Pure Reason. The Antinomy, as Kant sometimes calls it, sub- divides itself into four antinomies correlated — not without something of the violence to which we are now used — to the four logical categories of Quantity, Quality, Relation, and Modality. In the present lecture I shall confine myself to the first two anti- nomies, and shall begin with the second, because it refers to the inference from the composite to the simple, of which I have already had to speak. And not only for this reason : but also because the dogma of Wolffs metaphysics, that the ultimate elements of the physical world are simple substances — which Kant here presents as in irreconcilable conflict with an equally tenable opposite dogma — was the cardinal doctrine of Wolff's Rational Cosmology, so far as it is distinguished by him from Theology. The ques- tions raised in the first antinomy, as to the finity or infinity of the world in Time and Space, are, as I shall presently explain, less inevitable and funda- mental from Wolff^'s point of view. 152 LECT.x THE MATHEMATICAL ANTINOMIES 153 But before we proceed to the particular case, let us contemplate for a moment the Cosmological Idea, and the Antinomy or conflict to which it leads, in a general form. For here in the symmetrical exposi- tion of the fourfold conflict to which the human reason is reduced, if it clings to the illusion that it can know things in themselves, and in the double solution that the Critical Philosophy afibrds of the conflict — explaining the two first cases by showing that neither of the conflicting conclusions is true, and explaining the two last by showing that both may be true — we have not only the most brilliant product of Kant's genius for system-making, but also, as he claims, the most persuasive.^ This is, I think, true. If anything can persuade a man that the proper task of man's understanding and reasoning is not to know reality as it is, but to systematise the impressions it makes on our sensibility, this will persuade him. For if he takes the sensible world — the world of things as Sense and Understanding, in their ordinary empirical operation, present it — to be a real world, and tries to form a consistent conception of it as a whole, he finds himself environed on all sides with overwhelming and inevitable contradictions : whereas if he will only be content to regard it as a pheno- menal world, Kant assures him that the contradic- tions all vanish, and his reason, accepting its limita- tions, is at peace with itself. To illustrate the conflict let us take first, as I proposed, the second case ; in which the human mind ^ Cf. Prolegomena, § 50 init., Mahaffy, p. 131. i:.4 Till'. MKrAPTrv^^Trs or kant lkc-v. attempts to u^rasp coniplt'U'lv aiul roncoive consist- ently the constitution of matter, by reasoning from tlio composite substances presented in experience to tlie ultimate elements of which they are composed. Take any portion of sensible matter, we can usually break it into parts and these again into further parts. Even if we find it too hanl actually to break up, we can alter it by pressure, heat, or chemical methods, so as to convince ourselves that it is actually com- posed of a vast number of insensible parts that change their relative position when the whole is thus modified. But these parts as we commonly conceive them are not absolutely ultimate elements : when we reflect on any such part, we find that, since we conceive it as occupying space, we must conceive it as extended, and therefore as ideally divisible into further parts. Such a part, then, is not the ultimate reality of which we are in search, that would still exist if all composition were removed. But where then is this ultimate reality ? Yes, says the Tran- scendental Philosopher, who is contemplating this metaphysical process from his position of critical aloofness, where is it indeed? You cannot find it, and yet you must find it, unless you will consent to learn the lesson of criticism. You cannot find it, because however far you go in your process of imaginary division, the ideal result of division at which you stop must still be extended or it is no longer matter ; and yet if it is extended it must consist of parts, and the division has to begin again. At the same time you must find it, if you cling to X THE MATHEMATICAL ANTINOMIES 155 your belief that your reason in this process is dealing with a real and not a phenomenal world. For if you once admit that you cannot find it and that the division has to go on for ever, then the answer to the question what matter ultimately consists of must be "nothing at all." For, in Kant's words, " assume that composite substances are not made up of simple parts. Then, if we think all composition away, no composite part will be left. And, by hypothesis, there is no simple part. Hence nothing at all will remain." ^ The dilemma is efiectively pressed home : and there is no doubt a strong temptation to relieve our minds of it by adopting the critical position, and accepting it as the business of our understanding and reason, in their empirical and scientific use, to systematise the phenomenal data of sense, and the business of the metaphysician merely to understand the way that the understanding and reason do this and must do it. For from this point of view the dilemma vanishes. A merely phenomenal object must indeed be con- ceived as infinitely divisible, but this does not mean, in the case of the phenomenon, that it actually consists of parts infinite in number. For the parts of the phenomenal object do not exist as parts prior to our thinking them : they are constituted for our thought through our thinking them ; in short, the phenomenal object is infinitely divisible but not infinitely divided. And this illustrates the general 1 Watsou's Selections, p. 160. 156 THK Ml'MWrilVSKN OV KAXT i.kct. explanation wliich Kant ujives o( tlic cnttcul solution ot' those antinomios: viz. that wo havo to conceive as nioroly ros^ulativc tlio idoa of tho reason, which demands coni{)letion in a rational process through a series of conditioned objects to an unconditioned — in this case demands that the series of parts of parts of parts of a material thing, each part being found by reflection to be necessarily composite, be brought to a termination somehow. In saying that we have to conceive the idea as merely regulative, I mean, as Kant says, that " the principle of reason serves as a rule which postulates what must take place, if we make the regress " — from the conditioned object or event to its condition, which we find also conditioned, and then to a further condition lying behind that and so on, — " but does not anticipate what is present before the regress is made, in the object as it is in itself."* For example, in the case we are contemplating the unsatisfiable demand of the uncritical reason for the unconditioned, for an absolutely partless atom of matter, becomes for the reason duly self- critical a postulate of infinite divisihilitij, which carefully avoids any affirmation of actual infinite dividedness. In this way it claims room for any degree of fine- ness of division, w^hich Science, working on the data of sense, may find needful for a consistent theory of the phenomenal world : and at the same time shows us why we must not trouble ourselves, have no rational ground for troubling ourselves, with the question whether the smallest atom, which ' Vfaitaon'i Selections, p. 174. X THE MATHEMATICAL ANTINOMIES 157 science requires to suppose, consists of parts and how these parts are to be conceived. The escape thus offered from our dilemma is, as I said, certainly attractive. But the dilemma was not a new one : and Leibniz and Wolff had found a way out of it, which Kant does not here adequately deal with ^ — though what he would have said about it may be inferred from a general criticism of the Leibniz- Wolffian philosophy appended to the second part of his work. But we ought briefly to notice this other way out of the dilemma of the second antinomy, because it is perhaps the only other way out of it, if we insist, as Kant insists, that Eeason is to answer somehow all the questions that Eeason finds itself disposed to ask about the world — i.e. all the questions that refer to the possible and the necessary, for the actual could of course only be learnt in detail from experience. Briefly, the reason of Leibniz and Wolff found ultimately simple elements of the composite matter which experience presented : but the ultimates were unextended. " The elements," says Wolff, "of material things, are not extended, have no shape or size, and fill no space." The question of course arises : How then can they exist in Space, if they do not fill any part of it, and how can solid matter be composed of them ? The answer that Leibniz and Wolff gave to this was that though the things which sense perceived as spatial were not ultimately phenomenal but real, their spatiality and ' He makes a contemptuous reference to the 'Monadists' in his remarks on the second Antithesis : but does not appear to me to deal adequately with their position. [Cf. M. Muller's translation, p. 381.] 158 Till-: MKI'ArilYSh'S OF KANT lkct. continuous extension were phtiionienal. Space as an object of the umlerstaiuling was an order of coexist- ence of unextended entities, confusedly apprehended iu external sense-perception, and its apparent lionio- geueous unbroken continuity is due to this confusion. For Wolti' tlien externality or outsideness has two meanings, viz. (1) real externality as diversity or otherness of existence ; and (2) spatial externality as the confused appearance of this. Keal extension is the union, the coexistence as united, of a number of different things, which, as different from each other, are mutually external.^ Our notion of pure space, however, as an extended continuous immovable entity, in which real things are and move, is imaginary: real space is the order of things coexisting, regarded as coexistent : but our imaginary notion of space may be usefully taken as representative of real space, when we are only considering and comparing bodies in respect of magnitude. Similarly, real time is the order of continuously successive things. Now w^hat is Kant's arcjument against this ? We see at once that it seems to him a confusion between the thing as it is apart from our apprehension and the phenomenal thing. The thing in itself was rightly conceived as unextended, but the phenomenal thing must be conceived as extended, and Wolff's process from the composite to the simple appears to him to jump from the one to the other. But suppose Wolff were to answer : Certainly there is such a * Wolff accordiugly, as Kant after him, but with more systematic consistency, rejects the Idealism that denies all reality to tlie matter we perceive. X THE MATHEMATICAL ANTINOMIES 159 jump somewhere. Reason, arriving at the end of the regress from the composite empirical object to its ulti- mate element, has somewhere dropped the character- istics of continuous extension, size and shape. But that is because, in arriving at this conclusion, it has got out of the disturbing influence of sense. What would Kant have answered ? I imagine he would have pointed to Geometry as a proof that the pure notion of space as continuous and extended was not confused but remarkably clear. I admit the force of this, but if it is admitted, is not the whole success of physical science in understanding the laws of the physical world similarly an argument against the complete phenomenality attributed by Kant to the empirical object ? Let us turn back now to the first antinomy. In the conflict that we have just discussed, the series of conditioned objects which Reason tries to carry in thought beyond experience to the unconditioned is a series of continual division and diminution — we try to pass from the thing made up of parts which them- selves are made up of parts, to the ultimate element, whose existence is not conditioned by the prior exist- ence of parts that make it up. In the conflict to which we now pass — and that Kant puts first — the series, on the contrary, is one of addition and enlargement. We find that the existences of which we have experi- ence have things existing beyond them in space, and have had previous states of existence in time : and our Reason asks : When in thought we put these existences all together into a world, are we to conceive 160 TlIK .Mi;i Al'll\si('S OF KANT i.ect. this woiKl ay unlimited or limited in extension and in piust duniliou ? Now according to Kant we seem able to prove, with equal irresistibility, ou the one hand that the world had a beginning in time and is limited in space, and on the other hand that the world had no begin- ning in time and has no limits in space. But, though the conclusiojis are thus symmetrically opposed, this is only partially the case with the reasonings : and it will conduce to clearness to take the question as to time apart from the question as to space. One difference is that in the case of time, but not of space, theological considerations naturally come in. (And this is another reason why I took the second antinomy first : because of all the cosmological con- flicts, it is the only one that can be quite separated from theology.) For traditional theology conceived the world as coming into existence through a creative act of God : and this, for ordinary thought, involved the conception of the Creator as existing before the creative act, and therefore of the world beginning in time. Again, the conception of the creative act as wise seemed to require, and experience seemed to confirm, the conception of the process of the world in time as not merely a process of change but a progress towards perfection : and this seemed to exclude the notion of an eternity already past in the process. Even Leibniz, the creator of the differential calculus, says " if the nature of things in the whole is to grow uniformly in perfection, the world of created things must have had a beonnninor."^ ^ Fifth Letter in the Correspondence with Clarke, § 74. X THE MATHEMATICAL ANTINOMIES 161 However, Kant's argument for the thesis that the world has had a beginning in Time, keeps clear of theology. It is, simply, that a series of changes at once past and infinite — a completed unending series, — is inconceivable. The series cannot be thought as both endless and over and done. I admit the difficulty of thinking this : but it seems to me to depend on the nature of time, and not on the nature of an infinite series — as Kant suggests. For I find no difficulty in the case of space in conceiving infinite extension — e.g. of a line — limited at one end : so far as I can think of infinity at all, I can conceive an infinite number of infinite lines in diff'erent directions starting from a given point in space. Nor, as we shall see, does Kant urge the inconceivability of a bounded infinite in spatial extension as a reason for regarding the world as limited in space. In the case, however, of space, I also find the argument for the thesis devoid of cogency. Kant argues that if I think the world infinite in space, I must suppose " the successive synthesis of the parts of an infinite world to have been com- pleted," and that " this is the same as saying that an infinite time must have elapsed during the summation of the totality of existing things," which " is impos- sible."^ I deny this necessity. It is true that I cannot conceive myself as experiencing the boundlessness of space except in an infinite time : but I require no such time to negate the idea of a limit of space. ' Watson's Selections, p. 159. M LECTURE XI THE 'dynamical' ANTINOMIES AVe come now to the two later Antinomies — or con- victs in which speculation is involved when it tries to pass through the series of conditioned objects or events, which experience as grasped by the under- standing presents, to the unconditioned ultimate which seems needed to satisfy the reason. These Kant distinguishes as ' dynamical ' from the two * mathematical ' antinomies which we considered last time. But first, I must complete what I had to say on these earlier antinomies ; and in so doing I shall point out a difficulty which attaches to Kant's separation between the two pairs. In examining the first pair, I inverted Kant's order. I did so for two reasons. First, in the metaphysical view which Kant is pri- marily assailing, the positive conclusion of the second antinomy is intimately connected with the conclusions of Rational Psychology. The simple subject of psychical predicates, the permanent thing with which the varying elements of psychical life were connected as attributes, was not, in the view of Leibniz and 162 LECT.xi THE 'DYNAMICAL' ANTINOMIES 163 Wolff, an entity disparate and to be kept apart in thought from the simple substances grasped by- thought as the ultimate realities underlying empirical matter. On the contrary, the former was conceived so far as possible as analogous to the latter. I say * so far as possible ' because the extent of analogy varied : Wolff's common sense declined to follow Leibniz in attributing appetition and perception to the ultimates of inorganic matter. Still, for Wolff no less than for Leibniz, Minds took their place side by side with the elements of material things among the simple substances of which the world was composed. Secondly — We have now to observe that the argument of the first antinomy, so far as it relates to the past duration of the world, is not in Kant's own treatment clearly separated from the argument of the third. This does not appear in the argument for the thesis — that the world has a beginning in time — which rests on the inconceivability of an endless series over and done. But the argument for the contradictory proposition (the antithesis) that the world cannot have begun in time appears to me not quite distinctly separated from the argument against an uncaused event in the third antinomy. For what Kant here argues is that " nothing can come into being in an empty time, because no part of an empty time has in it any condition decisive of existence rather than non-existence, which dis- tinguishes it from any other part."^ That is to say, he seems to argue that nothing can come into ' Watson's Selections, p. 158 ^w. 161 THK METAlMTV^^irS OF KANT lkct. being in empty time, because thorc can l)e no cause lor its cominix into being at one time rather than another. l>ut this seems to assume that it cannot come into being through theyWr, uncaused volition of tlie Creator : that is, it assumes the question argued in the third antinomy. Now it" we kee}) the questions distinct — as the articuhition of Kant's system certainly seems to require — the argument must take a somewhat different form ; and I cannot find any form in which it appears conclusive. It may be said — as Leibniz urges against Clarke ' — that if we are to conceive the world as beginning in time we must conceive it beginning at some definite point of hitherto empty time, and that this is impossible, because there is nothing in empty time to distinguish one point of time from another. But the first premise cannot be granted : the conception of the besrinnins: of the world in time does not necessarily involve a dating of the beginning in relation to empty time. It is quite sufficient if we date it in reference to the time with which we are familiar. Suppose the process of the physical world is like that of a clock running down : and that physicists could time it so exactly as to know that a hundred millions of years ago some initial event must have occurred analogous to the winding up of the clock. We can obviously conceive this initial event to have occurred a hundred millions of years ago, and to have begun the particular process in which we now are, without defining further its relation to antecedent time : and > Fifth Letter, §§ 55, 56. XI THE 'DYNAMICAL' ANTINOMIES 165 if we can do this as regards this relative beginning, I do not see — apart from the causal difficulties con- nected with beginning — why we cannot similarly conceive an absolute beginning of the world, without dating it in reference to pre-mundane time. I think, therefore, that the argument for the antithesis in the first antinomy, if rigidly separated from the argument of the third, lacks cogency. And the same may be said of the similar argument as regards limits in space. Kant argues that a bounded world in an unbounded space must be related to empty space, and that there is nothing in empty space to relate it to, no means of dis- tinguishing one part of space from another. I quite admit that we cannot assign to the world a definite position in space, and that such questions as Where is it in space : is it moving or at rest ? are questions to which we can conceive no answer having any relation to possible experience. But I do not think that this applies to the mere question whether it has or has not limits : we can conceive it limited, and therefore having empty space beyond it, without raising the question where it is in space. I have said that the argument of the first antinomy, so far as it relates to the past duration of the world, is not, in Kant's own treatment, clearly separated from the argument of the third. I must now point out that it is hardly possible to separate the two questions, so long as we accept the principle of causality with the interpretation which Kant has given to it in the second part of the treatise. For if 160 THK MKrArHVST(\^ OF KANT lect. every event must li;iv«' a cuuse, i.e. an antecedent event after which it must come — and it', us Kant has argued, we cannot conceive an event as objective and tlierefore happening at a fixed point of time, without conceiving it as in this sense caused — then we clearly cannot ask whether the world has had a bee:inninf!: in time, without seeing at once that an affirmative answer brings us into conflict with the principle of causality. Why, then, does Kant separate the two questions so decisively in his arrangement, if not in his argu- ment ? Partly, I think, on account of the entirely different ansivers which his philosophy leads him to give to them. The same confusion of thought be- tween empirical or phenomenal objects, and things as they exist independently of human perception, occurs in all the antinomies : but the confusion leads to quite different results in the case of the first pair and the second pair respectively. In the case of the two first antinomies the apparently contradictory conclusions are found to be both false, when we get rid of the confusion of thought which has led to them. They are false equally — though for different reasons — whether we regard the conclusions as relating to the phenomenal world or to the world of independent realities. If I inquire about the extension in space or duration in time of the phenomenal world (the world con- stituted by putting together the objects of sensible experience), it is, Kant says, " equally impossible to declare it infinite or to declare it finite " ; because XI THE 'DYNAMICAL' ANTINOMIES 167 " experience either of an infinite space or of an infinite time elapsed, or, again, of the limitation of the world by a void space or antecedent void time, is impossible." ^ Similarly, it is false to say of pheno- menal matter — matter as an empirical object — either that it actually consists of an infinite number of parts, or of a finite number of indivisible parts. What we ought to conclude, according to the Critical Philosophy, is (l) that the magnitude of the world may be extended indefinitely in space and time, so far as we have empirical grounds for conceiving it extended : it can never be a rational objection to any physical hypothesis adequately supported on other grounds, that it requires too much time or too much space. And similarly (2) that any given quantity of matter is indefinitely divisible, though not infinitely divided : we may assume molecules or atoms as small as we please, so far as we have scientific grounds for assuming them. In short, the true Metaphysics, according to Kant, gives Physical Science a licence to assume the material world as large and the parts of matter as small as it likes, on the simple conditions of calling the world phenomenal and never pretending to have reached a maximum or a minimum. As we saw, Kant does not maintain that Physical Science required the licence ; and, in fact, it is pretty certain that it would go on just the same, if the licence were not granted. But the vogue of Kantism is partly due to the fact that many students of physical science, with a philosophical turn, have considered ^ Prolegomena, § 52 c, Mahaffy, p. 137. 168 TlIK MKrArilVF^TCS OF KANT i.ect. the licence cheap at ihc price, aud accepted the terras. So mucli for the soliiti(»n of the inatlieiimticiil aiiti- noinies. But in tlie case of the dyiiainical antinomies, thougli tlie fundamental confusion from which the apparent contradictions spring is the same, the solution is of an opposite kind. When the questions are raised ( 1 ) whether or not there is a ' free causality,' besides the natural causality (interpreted as necessary sequence) ; and (i2) wliether or not there is a necessary being, the affirmative and negative answers are, when the confusion between phenomenon and independent reality is removed, found to be both possibly true. That is to say, if we take the ' cause ' to be a phenomenal cause — an event in time — then we can admit no other kind of causality. For even extending the notion of ' cause ' to the phenomenal thing that is conceived as * agent ' or ' efficient,' it still must remain true, as Kant says, that " the determina- tion of the cause to act must have originated among phenomena, and must consequently, as well as its etfect, be an event which must again have its cause, and so on : hence natural necessity must be the condition on which efficient causes, so far as phenomena, are determined." ^ Thus the conclusion of the antithesis " that all that comes to be in the world takes place entirely in accordance with the laws of Nature " ^ is true, if the world be understood as phenomenal. But at the ' Prolegomena, § 52 c, Mahaffy, p. 140. '^ [Watson's Selections, p. 162 ^ra.] XI THE 'DYNAMICAL' ANTINOMIES 169 same time the argument of the thesis that the phenomenal world as an effect is not adequately- accounted for by an endless series of causes which must themselves be regarded as effects, is not answered. We may, however, find the answer in the relation between phenomena and things per se — when we have once clearly distinguished the two : and there is nothing to prevent us from applying to this relation the conception of Freedom. Thus, as Kant says: "Nature and Freedom can without contradiction be attributed to the very same thing but in different relations, on one side as a pheno- menon, on the other as a thing per se." ^ Observe that Kant does not affirm that we rnust attribute free causality to the thing per se, just as we must think all the changes in phenomenal objects as necessary consequences of antecedent changes. All that he regards as established by the critical solution of the antinomy in which the Speculative Eeason is involved by trying to reach through the series of conditioned causes a cause that is unconditioned and not in turn an effect, is (1) that the principle of Natural Causality cannot completely satisfy our demand for an adequate cause of the phenomenal world ; and (2) that there is no reason why free causality should not be attributed to a ' thing in itself,' if we have other grounds for attributing it. Now in the case of human beings he holds that our moral consciousness gives us practical grounds for attributing to ourselves such free causality : that ^ Prolegomena, § 53, Mahaffy, \>. 141. 170 THK MKTArUVSK'S OK KAXT lkct. our apparent cognition that something in tlic eye of reason * onght to l)e ' necessitates the assumption that what ought to he can he, and that reason there- fore can have causality in respect to ])henomena. This part of Kant's doctrine, so far as it rehites to human freedom, 1 have already examined in a lecture of the ethical course.^ Here I have only to point out that we must distinguish the ' practical freedom ' which rests on ethical data, from the ' transcendental freedom,' or ' freedom, in the cosmo- logical sense,' by which, as Kant explains, is merely meant * the faculty ' or ' power ' of ' beginning a state spontaneously ' " — a kind of causality which is not subject to the necessity imposed by the principle of natural causality on all phenomenal causes : i.e. of being also effects. I have said that Kant's critical explanation, dis- tinguishing phenomena from extra -empirical or transcendental realities, shows both affirmative and negative answers to the third — and fourth — Antinomy to be i^ossihly true. He does not intend to prove the actual truth of both the [seemingly] contradictory conclusions. With regard to freedom this is most emphatically stated. " We have had no intention of proving that there actually is freedom, and that it is one of the faculties which contain the cause of the phenomena of our world of sense. . . . All that we have been able, or wished, to prove is that nature does not contradict the causality of freedom."^ The * [Cf. Methods of Ethics, 6th edn. Bk. i. chap, v., and App. ] 2 Das Vermugen, einen Zustand von selbd anzufangen. Cf. Watson's Selections, p. 182. ^ 'Wdttson's Selections, p. \%Q fin. XI THE * DYNAMICAL' ANTINOMIES 171 critical solution therefore does not treat the thesis and the antithesis similarly. When the confusion between the phenomenon and the thing in itself is done away with, the argument and conclusion of the antithesis are completely validated so far as phenomena are concerned : it is entirely true that "all that we conceive to happen in the phenomenal world we must conceive as entirely conformed to the law of natural causality." But the argument and conclusion of the thesis are not similarly affirmed as valid with regard to the real world. The critical philosophy does not warrant us — so far as the cosmological argument goes — in laying down that there must be a free causality attaching to, exercised by, things in themselves ; but only that there may be. This ' lopsided ' result is quite natural : since in Kant's view our faculties are made to know phenomena and are not made to know things in themselves. But the question still may be raised, Is not the negative argument in favour of the thesis still valid, in a sense ? Does it not remain true that ' natural causality ' does not afford a complete explanation of phenomena ? and if so, must we not find that explanation in the realities of which the objects of experience are the phenomena ? Yes, answers Kant, " phenomena must have their source in that which is not a phenomenon." ^ That step beyond experience Kant definitely affirms. There must be Keality if ^ [Watson's Selections, p. 184. Kant, however, says not 'source' but 'grounds.'] 172 TIIK MKTArilVSTrs OF KANT r.F.CT. there are Apj>t'araiK'os : aiitl in Reality, if we only knew it. we should find the explanation of experience. But we eannot know it, and tlierefore can form no positive eonception of the ex])lanation. The world is ratictnal : hut not for us : it is not theoretically kn(nval)le as such. We have now to observe a flaw in the symmetry of Kant's system. His interest in the question of human freedom has led him to make the freedom of inan prominent in the discussion of the third antinomy. But the kind of Transcendental Freedom which the argument for the thesis naturally suggests is not human freedom, an uncaused beginning of the various particular series of effects that w^e attribute to human volition : but an uncaused beginning of the whole complex process of cosmical change. Human freedom is certainly not enough, as the effort to find an unconditioned cause to explain Nature can certainly not be satisfied by finding a free causality for human volition. And since, in Kant's view — by the application of the category that he calls ' community ' — the whole aggregate of empirical objects that make up the physical world must be conceived as connected by actions and re- actions, reciprocally determining each other's changes, the complex of natural change has to be thought as one connected whole. Hence a spontaneous causality adequate to satisfy the demand of Reason, and enable us to think the reosition of an absolutely free causality, in the fourth the same line of thought is supposed to drive the Reason to the assumption of an absolutely necessary Being. In short, it would seem that, if Kant's system had only permitted, he might have represented the thesis of the third and that of the fourth antinomy as together forming a single antinomy, of which the two conflicting conclusions were the affirmations of Freedom and Necessity. Reference, however, to the solution of the fourth antinomy shows that there is in Kant's view a difference in the lines of thought pursued in arguing the third and fourth thesis respectively, which he certainly has not clearly expressed in expounding the antinomies. " In what immediately precedes," he says, " we have considered the changes of the world of sense in their dynamical series — a series each member of which stands under another as its cause. We shall now take this series of states as our guide in the search for an existence that may serve as the supreme condition of all that changes ; that is, in our search for the necessary being. Here we have to deal not with an un- conditioned causality, but with the unconditioned existence of substance itself."^ That is, in the third antinomy attention is fixed on the changes in em- pirical things ; in the fourth, on the changing things. The reason why the two arguments look so much ' Watson's Selections, p. 191. XI THE 'DYNAMICAL' ANTINOMIES 175 alike is that, in Kant's view, the ' contingency ' of the empirical thing seems to depend on its changeability. He says : "It is easy to see that, as every object in the totality of phenomena is changeable, and there- fore is conditioned in its existence, no member of the series of dependent existence can possibly be uncon- ditioned : in other words, we cannot regard the existence of any member of the series as absolutely necessary." ^ It is because it is changeable that it is ' conditioned in its existence,' and therefore, however far back we retrace in thought the existence of phenomenal things, we cannot find necessity : though, when we have clearly distinguished phenomena from things in themselves, the existence of such a neces- sary Being is seen to be possible, but only as an ' extra-mundane being ' entirely outside the series of the sensible world. I think, however, that Kant is wrong in thus con- necting the contingency of the things that constitute the sensible world, as ordinarily conceived, with its mutability. To show this, suppose we assume — what we ordinarily do assume in trying to conceive physical and chemical changes — that the ultimate parts of matter only change in their relations to other parts, and remain in other respects unchanged. Kant must admit this conception, according to the ' first analogy of experience ' : viz. that ' Substance is permanent and its quantum in nature neither in- creases nor decreases.' Then let us trace back in thought the changes in the physical world-processes : ^ Watson's Selections, p. 191. ITu THK MKTAI'HVSrrS OF KANT lkct. at any point al wliirh we .stop, the positions in which we leave the ultimate parts of matter seem to us no loss arhitrary and eontiiiixent than the positions in which we now tind tliem. That is, we see no reason whv their coUocatiou in space should not have been ditlerent. But, it may be asked, with regard either to my supposition or Kant's, how does the introduction of a Necessary Being help the matter ? For if we conceive it in time, as the argument for the thesis contends,' we have still to understand hoiv a Neces- sary Being in time can be the cause of a contingent : and I know" no way in which this transition can be made to appear rational, nor does Kant's argument suggest any. But again, if we take the critical solution, and suppose the necessary, uncaused Being, out of time, the difficulty still remains : how comes a Necessary Being to cause a contingent being ? It seems to me impossible to conceive the contingent as the necessary consequence of the necessary. I draw attention to this difficulty, because it appears to me that the solution of the third antinomy has to be combined wqth that of the fourth, in order to afford to the Speculative Reason that moderate amount of satisfaction which is all that the critical philosophy professes to afford to it. That is, we have to suppose, in order to explain the series of the sensible world — whether we regard that as a series of changes or a series of changing and contingent ^ "The causality of the necessary cause of the changes, and therefore also the cause itself, must belong to time and to phenomena in time." — Watson's Selections, p. 166. XI THE 'DYNAMICAL' ANTINOMIES 177 existences — not only a transcendental and free causality, but also a necessary Being to which this free causality is attributed. We have to suppose this, in the case of the third antinomy no less than in that of the fourth ; for the transcendental causality which is supposed to explain the series of natural phenomena must be the causality of something : and if the being that exercises it is not conceived as necessary and therefore uncaused, its existence will require a cause no less than the series of phenomenal existence. It may be said, that on the principles of the Critical Philosophy, we cannot thus apply the con- ception of causal dependence to things in themselves, since that conception has only a legitimate applica- tion to empirical objects. I admit the force of the argument : and can only answer that Kant repeatedly applies it so himself.^ Further, if the Critical Philo- sophy rigidly abstains from this extended application of the category of Causality, its so-called critical solution of the conflicts of reason becomes illusory. That is, it amounts only to saying that besides the necessary sequence of natural or phenomenal causality, by which we can never really explain any pheno- menal effect, because the series cannot be completed, we may also suppose an unknown relation to an ^ For example, his refutation of Idealism (as expounded in the Prolego- tnena, § 13, Remark ii. Maliatfy, pp. 53 ff. ) involves this 'transcendent' application of the notion of cause. And also expressly his solution of the third antinomy: "phenomena must have their source in that which is not a phenomenon." [Watson's Selections, p. 184. For a fuller discus- sion of this topic by Professor Sidgwick the reader is referred to 3Iind, O.S. iv. pp. 408 ff.; v. pp. Ill ff.] N ITS TlIK ^n^T.\^lTY^iTO^^ OF KANT ikot. xi unknown entity wliidi is n(»t a plicnoniciion, wliich mi. 530.] 2 Cf. Watson's Selections, pp. 219 ff. 190 rilK MKrAl'IIYSICS OF KANT lect. ;ui;i]>tatioii of nieaiis to a ilefinile end : the more he kiiDNVs o( tlie natures of finite tilings and the unifor- mities of tlieir behaviour, tlie more difficult it seems to regard this adaptation as the unpurposed result of natural laws. He cannot but refer the unmistakably planned result to designing intelligence : he cannot but infer from the systematic unity of the plan the unity of the intelligent cause. But, granting all this, the argument proves, as Kant says, an Architect, not a Creator of the world : it is the origin of the form and order in the physical world that it explains, not the origin of its matter or substance. For in the human adaptation of means to ends on which the argument rests, the matter is always given to the designing mind, not made by it : not an atom of the material of the watch derives its existence from the watchmaker. To justify us in conceiving the matter of the world as created by God, we have to introduce a new arirument : we have to fall back on the con- tingency of every finite thing and all finite things. The physical world, in all its parts and all stages of its process, presents itself to our thought as something that might have been otherwise, i.e. granting that we find necessary connexions in the coexistence and sequence of its parts, the necessity thus found is always a conditioned necessity and leaves the whole still continsfent ; our reason therefore still demands a cause why the whole physical world and its history is and has been what it is and has been. Even if, under the guidance of speculative astronomy and physics, we suppose our world and planetary system as it is XII KATIONAL THEOLOGY 191 to be the necessary result of the nature and collocation of material particles in an original nebula, that nature and collocation still present themselves to thought as no less contingent and arbitrary than the particulari- ties of our actual globe and planetary system. Our reason must still seek for an explanation, a cause of this contingency and particularity : and we can only find it in a necessary being, something of which we cannot think that it might have been otherwise, because it is inconceivable that it should not exist as we conceive it. And this necessary being must be the Ens Kealissimum : for we must conceive it as completely determined from a logical point of view : i.e. it must be either A or not- A, B or not-B, and so on through the whole series of possible predicates, and in each case we must think it as having the positive predicate — for if any real positive quality were denied of it, the manifestation of that quality in the world of finite things would remain unexplained. Now we already know from the fourth antinomy that Kant cannot regard this line of argument as valid : the solution of the fourth antinomy was that while nothing in the world of phenomena can be thought as unconditionally necessary, there may be an absolutely necessary Being in the world of things as they are apart from our sense-perceptions ; but we cannot afiirm that there must be : our ignorance of things in themselves is too complete to allow of this assertion. But, even if we grant the inference from the contingent to the necessary, he holds that our reason cannot identify this necessary being with 19 2 rHK MKl'ArilVSK'S OF KANT lkct. I ho ICns Kcalissiuuiiii, unless \vc cau prove iu some other way that the Ens llealissimum must necessarily be. For, without this, we cannot be certain a priori that the existence of Unite things may not be unconditionally necessary, although we could not infer this necessity from our conception of finite things. Thus the cosmological proof, when strictly examined, is found to require the ontological proof for its validity. Here, then, lies the final and central issue for rational or speculative theology. Is this proof cogent ? Well, allow me to suspend for a moment this great question, and answer a somewhat easier one : Is Kant's argument against it, which appears to have convinced many generations of thoughtful persons, itself cogent? To give it in Kant's words. " If," he says, " I take the term God, and say, there is a God, I do not enlarge the con- ception of God by a new predicate, I merely posit the subject itself with all its predicates, as an object corresponding to my conception. The content of the object and of my conception must be precisely the same : the real contains no more than the possible. A hundred real dollars do not contain a halfpenny mcfre than a hundred possible dollars : — If the object contained more than the conception, the conception would not express the whole object, and would therefore be an inadequate conception." ^ I have tried, by selection of phrases, to put the argument as plausibly as possible : I must regard ' Cf. Watson's Selections, pp. 208 f. [The translation is amended.] XII EATIONAL THEOLOGY 193 it as plausible, as it has satisfied so many people. But I confess it seems to me to involve an intolerable paradox. That my conception of anything — say 100 dollars — which I do not think as actually existing is precisely the same as my conception of it as actually existing seems to me quite unthinkable. Kant says that 100 real dollars do not contain a halfpenny more than 100 dollars not thought as existent : but the remark seems to me an uncon- sciously crafty suggestion to throw the reader's mind on a wrong track. Certainly the difference is nothing like a halfpenny : the question is whether it may not amount to 100 dollars I Look at it thus. If the predication of existence makes no difference to the conception, it must be equally true that the predication of non-existence makes no difference to it : therefore there can be no difference between the thought of a hundred dollars as non-existent and the thought of a hundred dollars as existent. Is it not, on the contrary, palpable that there is just a hundred dollars difference ? It is not, therefore, because the conception of a thing as existing is not different from the conception of a thing precisely similar but not thought to exist, that I fail to find cogency in the Ontological proof: but rather because the two conceptions seem to me not only distinguish- able, but when distinguished equally possible, in the case of the Ens Realissimum, no less than in the case of other objects. So far as I am able at all to conceive an individual being having all positive predicates, I am able to conceive it as including all ll>4 Till-: MKTArilVSlCS OF KANT lect. positive predicates exeept existence : and when I have so conceived it, I am conscious of no rational necessity compelling me to add the predicate of existence rather than the predicate of non-existence. The proposition tliat tlie Ens Realissimum thus conceived exists seems to me no more necessarily true than the projiosition that it does not exist, — so long as I try to settle the question by mere reflection on my abstract ideas. But I have a prior ditHculty, as regards the formation of the notion of an Ens Realissimum : viz. I do not know that all positive predicates are really compatible, as attributes of the same being. For this is certainly not the case as regards objects of empirical thought : positive predicates are fre- quently incompatible, as straight and curved of a line, square and round of a figure, blue and yellow of the same surface.^ And this is especially important, when I consider that this notion of Ens Realissimum is to be identified with the theological notion of God, and to have all the moral attributes of Deity. For thus viewed, we see that the assumption of the com- patibility of all positive predicates, made in the formation of this transcendental Ideal, requires us to hold — what Leibniz, of course, did hold — that ' Evil ' moral and physical is a merely negative attribute. But I can see no reason to suppose this. Physical ^ Wolffs exclusion of ' phenomena ' is meant to get rid of these analogies ; but I do not know that the same incompatibility is not true of the qualities without limit attributed to the Ens Realissimum. XII EATIOKAL THEOLOGY 195 pain seems to me as positive as pleasure : and, though much moral evil is no doubt analysable into mere defects or negations of positive quality, I do not find this conceivable in all cases, as, for example, in the case of pure malevolence. APPENDIX TO LECTURE XII infinite and absolute or unconditioned (infinite-absolute) These terms for nearly half a century — second and third quarter of the nineteenth century — were leading terms in English metaphysical controversy. The period begins with Hamilton's article on the " Philosophy of the Unconditioned," ^ and it may perhaps be taken to end gradually with the decline of the influence of Mill and Spencer on English metaphysical thought, which I place about forty years later, attributing it primarily to the teaching of Caird and Green. In the current controversy between Empiricism and Tran- scendentalism these notions have somewhat fallen into the background : I think partly from policy. Transcendentalism, endeavouring to persuade a world largely dominated by Empiricism, thought it best to come forward in an Episte- mological rather than an Ontological garb : and to transcend experience — if I may so say — without the waving of flags so conspicuous as these words had come to be. But it still remains, I think, important that we should obtain as clear and complete a grasp of them as we can : and for this purpose we may still derive some instruction from the con- troversy to which I have referred. First, I ought to say that in this controversy, as regards the main question at issue, the English writers — keenly as they * Edinburgh Review, Oct. 1829 [republished in DiscussioTis on Philosophy and Literature, 1852]. 196 APPENDIX 197 disputed with each other — were all on one side. The common enemy was the post-Kantian philosophy of Germany. This philosophy — especially as taught by Schelling and Hegel — was held to maintain the cognisability of what Hamilton called the ' Infinite- Absolute ' or the ' Unconditioned.' In Hamilton's language, " Kant had annihilated the older Metaphysic, but the germ of a more visionary doctrine of the Absolute (Infinito- Absolute) than any of those refuted, was contained in the bosom of his own philosophy. He had slain the body, but not exor- cised the spectre, of the Absolute ; and this spectre has continued to haunt the schools of Germany even to the present day. . . . The theories of Eeinhold, of Fichte, of Schelling, Hegel, are just so many endeavours to fix the Absolute in knowledge." ^ And indeed this knowledge is conceived by them as the special aim of Philosophy. As Hamilton says, expressing Schelling's view, " While the lower sciences are of the relative and conditioned. Philosophy, as the science of sciences, must be of the Absolute — the Unconditioned." This view, then, the leading English thinkers for the half century indicated — however widely they differed — agreed in rejecting. They argue that "the Absolute cannot in any manner or degree be known, in the strict sense of knowing " : - though Hamilton holds that " we are, in the very consciousness of our inability to conceive aught above the relative and finite, inspired with a belief in the existence of something unconditioned beyond the sphere of all comprehensible reality " ; ^ and Mr. Spencer holds that we necessarily affirm its existence as logically implied in the existence of the relative and the finite, and have an indefinite consciousness of it : though at the same time it is rightly described as unknowable. Indeed he goes so far as to say that this indefinite consciousness of the Absolute and Un- limited itself exists absolutely in our minds.* And Mill, too, speaking as then the leading representative of English Empiricism — though disagreeing entirely with Hamilton's arguments — has no doubt that he has "established the futility of all speculations ^ [Discussions, p. 18.] - [Spencer, First Principles, 3rd edn. § 27, p. 98. Omitted in the last edition.] * [Discussions, p. 15.] ^ Cf. his First Principles, chap. iv. 198 TllK MKTArUVblt'S 01' KANT rcs|^H5cting those meaningless iibstractions Mho Intiiiite' ami 'the Ahsolute,' notions to which no oorrespontlirig otitities ilo or can exist."' The grounds on which Mill holds this may be briefly summed up as the acceptance uf the doctrine of the 'Kelativify of ilumau knowledge' in its widest sense : — "the entire inacces- sibility to our faculties of any other knowledge of Things than that of the impressions which they produce in our mental consciousness." - On the whole, then, we may say that the prevalent view of English Philosophy in the middle half of the nineteenth century, in spite of all its internal controversies, was in conscious, uncom- promising antagonism to the doctrine that the Absolute or Unconditioned or the Infinite-Absolute was knowable, and that it was the special business of Philosophy, as distinguished from empirical sciences, to know it. At the same time it recognised that in holding this view it was in opposition, not only to the post-Kantian philosophy of Germany, but to the general drift and aim of metaphysical speculation from its earliest appearance in the development of European thought — as Hamilton puts it — " from the dawn of philosophy in the .school of Elea," at the end of the sixth century B.C. " Metaphysic," he says, "strictly so denominated is virtually the doctrine of the Unconditioned. From Xenophanes to Leibnitz [before Kant, no less than from Fichte to Hegel after Kant] the Infinite, the Absolute, the Unconditioned, formed the highest principle of Speculation " ; but, he adds, "until the rise of the Kantian Philosophy, no serious attempt was made to investigate the nature and origin of this notion."^ But in saying this last, Hamilton does not go far enough. Speaking of Modern Philosophy,^ from Descartes onward, we may say that though ' Infinite ' is an essential attribute of the primal Being which the metaphysician calls God, the notions of ' Unconditioned ' and ' Absolute ' are not applied by them to this Being. They speak of God, the source ' {^Examination of Sir W. Hamilton s Philosophy, 3rd edn. p. 70.] * [Op. cit. p. 13.] •* [Discussion on Philosophy, etc., 3rd. edn. p. 15.] * And especially e.xcluding Plato, as I have no time to digress into a dis- cussion, how far the first principle, the 'something not hypothetical,' wliich philosophy, according to Plato, seek.s, may be pioperly interi)reted as ' Uncon- ditioned' or 'Absolute.' APPENDIX 199 of all finite Being, as original Being, most real Being, Higliest Being, Infinite and All-perfect, comprehending all realities or perfections, perfections which are thought as Infinite : but they do not apply to this original or primary Being the conceptions of ' Absolute ' or ' Unconditioned.' On the other hand, in the post-Kantian philosophy of Germany, ' the Absolute ' — though conceived in a fundamentally diflferent way by Fichte, Schelling, and Hegel respectively — is undoubtedly throughout the rapid and remarkable evolution of thought which these names repre- sent the leading conception of the chief object of philosophical inquiry. It may be said that the difference is merely one of words : but to discuss this would involve a discussion of the whole course of Modern Philosophy, which is necessarily beyond my scope. I am concerned with making as clear and precise as possible the conceptions in which the great issue between English and Germans was formulated in the century now closing ; and in order to do that, I must confine myself to the thinkers in whose exposition the terms in question are leading terms. But I am not undertaking to give a summary account even of post -Kantian philosophy. I am only trying to help towards an answer to the questions : — What do the post-Kantian thinkers mean by the terms Absolute, Unconditioned (I take these rather than Infinite, as that, as we have seen, is equally characteristic of pre-Kantian thought) ; What place does the notion Absolute or Unconditioned occupy in their philosophy ; and How came it there ? Well, the answer to the third question is pretty evident from what I have said. Comparing pre-Kantian with post- Kantian philo- sophical terminology, it undoubtedly com es there through the epoch- making influence of Kant. ' Absolute,' however, is not a leading notion in Kant's philosophy, except as qualifying the necessity of the Necessary Being sought in the fourth Antinomy — he only uses the term in the subordinate manner of pre-Kantian thinkers ; but ' Unconditioned,' as we know, is a very important term in his system, and I think that the post-Kantian term Absolute, whatever else it means, always has a certain correspondence to Kant's term Unconditioned. Let us, then, examine the Kantian use of this term. "The Unconditioned," as we have seen, is a general term for what the Reason seeks but cannot find, when 200 Till': MKTArilWSlCS OF KANT it aims, on difVorent linos of thdui^lit, at putting tof:;cther into a ciinij»K'te whole that conni'ctcil knowledge of empirical objects wliiih unilorstaiuling anil imagination, combining the data of soiiso, supply, and the physical sciences present in a systematic form. Trying to think the empirical world as a whole, Speculative Reason asks questions which experience obviously cannot answer, but which a natural and inevitable confusion between objects of experience and " things-in-tht-mselves " misleads Reason into supposing answerable a piori. Whether the world had or had not a beginning in time, has or has not limits in space : whether the substances that make it up have or have not indivisible ultimate elements : whether the necessary sequence of causal events which we must find everywhere in tracing back the world-process, terminates anywhere in an uncaused event — such as a ' free ' volition would be — or must be thought as endless : whether, — from the contingency which belongs to all empirical facts, which, though necessary results of other facts, are only conditionally necessary, — we can or cannot infer the existence of an absolutely and unconditionally necessary Being : — if valid answ^ers to these questions were really attainable, they would, according to Kant, give us under each head knowledge of the Unconditioned.^ But, as we know, valid answers cannot be obtained ; so long as we confound phenomenal things with realities existing independently of our sensibility and thought, the contradictory answers under each head are found equally untenable, and yet there is no conceivable third answer. When, however, we get rid of this confusion, we find that under the first two heads the questions are such as ought not to be asked : for they cannot relate to real things existing out of Time and Space ; while, as regards phenomena, they are meaningless in the form originally asked. For, e.g. even to ask whether a merely phenomenal world had or had not a beginning in time implies that Time is real, otherwise than as the form of our ^ According to Hamilton's use of Absolute, one of the alternatives in each case is the Unconditioned Infinite, another the Unconditioned Absolute. ' Absolute,' he notes, is used in a wider sense= 'aloof from relation, condition, dependence,' and a narrower= ' finished, perfect, completed ': in the narrower, the Unconditioned is a genus of which the Absolute is a species. I shall not use the term in the second sense except I .so state. APPENDIX 201 Sensibility. All we can reasonably ask is, * How far back may we go in time, in our scientific synthesis of phenomena ' : to which the answer is : ' As far back as you have empirical grounds for going ' ; and similarly as regards spatial extension and divisibility. That is the idea of the Unconditioned, under these heads, has a merely regulative use, in that synthesis of objects of experience which is Reason's proper task. On the other hand, under the last two heads both answers may be true. Here again, so far as experience and the empirical world are concerned, the use of the idea of the Unconditioned is purely regulative : it entitles and directs us to seek without limit empirical causes and conditions for all empirical facts. At the same time the free causality and the necessary existence affirmed in the theses may be attributed to the extra-cognitional Reality or Thing-in-itself. The conclusion of the Speculative Reason is thus that there may be in the world of Noumena a free causality and may be an unconditionally necessary Being : but we cannot know positively that they are. But though this is the conclusion of the Specu- lative Mind, you must never forget that it is not the conclusion of the Kantian Philosophy. For, on the basis of ethical con- viction, and for the purposes of practice, we have to postulate the free causality of the human will, and the existence of God. For man, as a rational agent in the world, must (1) recognise the moral law as 'absolutely' and 'unconditionally' binding; (2) aim at realising his ' highest good.' But this ' highest good,' in Kant's view, does not consist in Virtue only. " Virtue or the worthiness to be happy is the ' supreme good,' ' the supreme con- dition ' of a rational pursuit of happiness : but it is not the whole or complete good ; ... in the highest good which is practical for us, that is, which is to be realised through our will, virtue and happiness are conceived as necessarily united." ^ But " a rational agent who is also a part of the world of nature and dependent on it," and has " no power to bring nature into complete harmony with his principles of action," has no reason to expect that nature as such will realise the required connexion between morality and happiness : still, since as a rational being he ought to seek to promote the highest good, the highest good ^ Watson's Selections, pp. 291 f. 202 TUH MKTAI'HVSICS OF KANT must ho attaumhle. "Ho must thorcforo postulate a cause of nature jis a whole, distinct from nature, with at once power anil will to connoct morality and happiness in exact liarmony with each other": ' it. God, as CJod is conceived in what Kant dis- tinguished as Moral Theology. But now, when we try to put together the results of the criticism of the Speculative Reason,- with the results of the examination of the Practical Reason, we find that the negative result-3 of the former are importantly modified. For the Specu- lative Reason, though it could not prove the existence of an original, unconditionally necessary I>eing, yet was not critically barren of valid results. It showed the possibility of such a Being outside nature and its Supreme cause : it showed how God must be conceived if a pi'oof of the reality of His existence could be obtained on any other line of thought. " The Supreme Being," said Kant, " is for purely speculative reason a mere ideal, but still a perfectly faultless ideal, which completes and crowns the whole of human knowledge. And if it should turn out that there is a moral theology, which is able to supply what is deficient in speculative theology, we should then find that transcendental theology is no longer merely problematic, but is indispensable in the determination of the conception of a Supreme Being, and in the continual criticism of reason, which is so often deluded by sense and is not always in harmony even with its own ideas. Necessity, infinity, unity, existence apart from the world (not as a soul of the world), eternity as free from conditions of time, omnipresence as unaffected by conditions of space, etc., are purely transcendental predicates, the purified conception of which, essential as it is to every theology, can be derived only from a transcendental theology."^ Having given this brief summary of Kant's complex view, let us now consider it in relation to the issue before-mentioned raised between English philosophy of the central half of the nineteenth century and the post-Kantian philosophy of Germany. Does Kiint hold that the Absolute or Unconditioned can be cognised or conceived, and if so, what is it, what are we to say of it ? Now to these questions very various answers have been ^ Watson's Selections, pp. 296 f. ' Cf. Watson's SelerMons, p. 221. " Watson's Selections, p. 222. APPENDIX 203 given ; and it will be instructive to compare them, not merely for the light they throw on Kant's system, but also for the difference of meanings which they show to exist in the use of the terra 'Absolute.' According to Hamilton the Unconditioned, for Kant, is not an object of knowledge : but its notion, as a regulative principle of the mind, is more than a mere negation.^ Now this answer is not wrong, in my view ; but it is not luminous : it does not give Kant's view, because it does not introduce his distinction between phenomena and Things in themselves — things as they are apart from human apprehension. When we take this dis- tinction, we see that a double answer is required, because there are two questions — one relating to phenomena, the other to things in themselves. As regards the phenomenal world, the Unconditioned is not to be found, in any of the cases in which the uncritical reason seeks to find it, not on account of the limitation of our faculties of cognition, but simply because it is not there. But if this is what becomes of the idea of the Unconditioned in its application to the phenomenal world, what are we to say of its application to Things in themselves ? Now, if I under- stand Hamilton, his view of Kant's answer to this question, simply identifies ' the Unconditioned ' with ' Things in themselves,' and declares it unknowable, because the human mind can only know, not the things themselves, but their effects on our senses. To quote Hamilton's words : " Things in themselves, Matter, Mind, God — all in short that is not finite, relative and pheno- menal, as bearing no analogy to our faculties, is beyond the verge of our knowledge. . . . Thus ... a knowledge of the Unconditioned is declared impossible."- I think this entirely misrepresents Kant's view. Kant certainly does not hold that Things in themselves, realities as existing out of relation to human experience, are one and all Unconditioned : nay, he does not even know speculatively that any of them are Unconditioned. I will not speak of Things in themselves other than thinking beings : because, though in the Prolegomena, repudiating " Ideal- ism," Kant certainly affirms the existence of this class of Things in themselves, in the Critique he seems to treat their existence ^ [Cf. Discussions, p. 27.] ^ [Op. cit. p. 16.] L'04 rilK MKIA I'M VSU'S OF KANT :vs problomaticivl.^ Lot us then (.■onfuie ourselves to thinking bcinj^s : tlu'so certainly are for Kant Things in thcmsclvos. Kant expressly says of the human subject that ho is "conscious of himself as a tiling in himself"; and, on the basis of the postu- lates of the Practicjil Reason, he conceives such subjects as creatures, created indeed timclossly in a timeless act, but still created by an Original Being of Infinite Power, Wisdom, and Goodness. It is true that ho attributes to them, as rational beings, a free causality : and it is important to lay stress on this, because this is the main starting - point in the Kantian system for Fichte's doctrine of the Absolute Ego, which begins the evolution of the post-Kantian Metaphysic of the Absolute. But though he conceives them as having a free causality, he conceives them as essentially finite and imperfect : indeed it is on this conception that the postulate of immortality depends, because a " finite rational being is capable onl}' of an infinite moral progress from lower to higher stages of moral perfection." ^ Well, then, beings whom we cannot but think as created finite, imperfect, we obv'iously cannot but think as conditioned ; even though we can have no speculative knowledge of the conditions of their existence, except on its moral side. How far, then, does Kant apply the idea of the Unconditioned to Things in themselves ? Well the answer, from what has been just said, is surely clear. He can apply it only to God the Original Being ; and the postulates of the Practical Reason compel us to think of God as a First Being all-wise, all-good, all-powerful, the cause of nature but not a part of nature. To such a being we must apply in practical thought, and in theology (in Avhich practical thought and speculative thought blend, though the former is predominant), the conception of uncon- ditioned necessity of existence, Avhich the critical discussion of the fourth antinomy left as possibly applicable in the world of things in themselves though not in the phenomenal world. Here, then, according to my view, is Kant's final answer to the questions, ' Can the Absolute be known, and how far can it be known ' 1 — meaning by the Absolute, ' Unconditioned Realit^^' ' Cf. Watson's Selections, " On the Distinction of Phenomena and Xoumena," pp. 129-134. "^ [Watson's Selections, p. 295.] APPENDIX 205 Kant's ' Absolute ' is God : His existence cannot be speculatively known, but for practical reasons He must be thought to exist, as the First Cause of the World, with infinite power, wisdom, and goodness ; and being so thought, He cannot but be identified with the unconditionally necessary Being which the critical solution of the fourth antinomy showed to be possible, though it could not prove it to be actual. We have examined two views of Kant's Absolute, the difference of which depends on the difference of meaning attached to the term. (1) If " Absolute " = non-relative = non-pheno- menal (according to a prevalent view of Relativity of Human Knowledge), then, no doubt, Kant's Absolute = Things in themselves. (2) But 'Absolute' is not an important Kantian term : ^ its importance, as I have said, is post-Kantian : and if we take Absolute = Reality, that is. Unconditioned (this latter being a leading term with Kant), Kant's answer to the inquiry concerning our knowledge of the Absolute must, I think, be that I have given. But there is another view of Kant's Absolute that is given by Fichte, the first and nearest to Kant of the three leading Teutonic thinkers who worked out the doctrine of the Absolute against which the English mind rebelled ; it is thus important as throwing light, if not on Kant, at any rate on these further developments. In a remarkable passage in a course of lectures delivered towards the close of his life — in 1813, not more than a year before his death — Fichte says that though Kant com- prehended the Transcendental Ego as the union of inseparable Being and Thought, he did not comprehend it in its pure inde- pendence, but only as the common fundamental characteristic of its three for him original modes x, y, z : and thus he " had really three Absolutes, while the one true Absolute was reduced to their common characteristic." ^ These ' three Absolutes,' according to Fichte, are to be found in the three Kritiken, the Critique of Pure Reason, the Critique of Practical Reason, and the Critique of Judgment. With the fundamental doctrines of the two former we are already familiar ; with regard to the third, I Avill only say that it only aims at a kind of Mediation, through the notion ' It does not occur, for example, in Watson's index. ^ [Fichte's Nachyelassene Werke, ii. pp. 103 f.] 206 rilK MKl'AriIYSl("S OK KANT of Kml, hotwoen tho conclusions of the Speculative Kcison in the first Critique as to tho world of Exporicnco or Nature, and the view put forward in the second ('riti<|uc as to the super- sensible world of free rational beings. Fichte's accoiuit of the ' three Absolutes,' then, is as follows : " In the Critique of Pure Jicnsan sense-experience was for him the Absolute (x) : and of the Ideas — the higher, purely Spiritual World — he speaks in truth in a very deprecatory way. One might conclude from his earlier works, and from certain hints thrown out in the Critique itself, that in his own view the matter could not be left so : but I would undertake to prove that these hints are only one more inconsistency : for if the principles there laid down were carried to their logical con- clusions, the Supersensible world must entirely vanish, leaving as the only Noumenon the ' is ' to be realised in experience." ^ But the lofty morality of the man " corrected the error of the Philosopher, and the Critique of Practical Reason appeared. In it was manifested, through the indwelling notion of the Cate- gorical Imperative, the Ego as something in itself, which it could not be in the Critique of Pure Reason, where its only basis is the Empirical ' is ' : so we get a second Absolute, a moral world = z." He then goes on to say that in the Critique of Judgment it was acknowledged that the Supersensible and the Sensible Worlds must have some common though quite in- scrutable root, which would be the third Absolute = y. Overlooking this third, let us ask what Fichte means by the two distinct Absolutes found respectively in the Critique of Pure Peason and the Critique of Practical Peason. First, I must explain that Fichte's development of Kantism — ignoring or overriding, as Modern English Transcendentalists ignore or override, the Refutation of Idealism in the Prolegomena — discarded altogether the concejjtion of Things in themselves other than Thinking beings. Accordingly the points in Kant's doctrine that are fundamentallj'^ important for Fichte are (1) the conception of self-consciousness as making nature in the Critique of Pure Reason, i.e. as the source of all Synthesis and all form in the world of Empirical objects; and (2) the conception of * Fichte means the hond of Syntliesis between sensible data, supplied by the Transcendental Ego, and expressed by the copula "is." APPENDIX 207 independent rational activity in the moral world, the essence of all thought of duty and moral action. But these two, he con- siders, Kant ought to have conceived as essentially one and brought into intelligible relation : he ought to have seen that it is the same rational self-conscious activity that makes nature and makes duty and is at once the source and explanation of all knowledge and all duty : and he ought to have effected a rational systematisation of the two functions — Avhich in his system as expounded by him are apparently so diverse, and deduced them from a common principle, a primary activity of the Transcendental Ego. Had he done this, the 'one true Absolute ' would have been revealed in this primary activity, the first source and condition of all else in consciousness, there- fore of all else in the universe. But as he did not do this, we are left — so far as the two treatises on the Pure and Practical Reason go — with two different Absolutes.^ In the Critique of Pure Reason — according to Fichte's trenchant but one-sided account of it — if its line of thought were consistently carried out, the higher spiritual world would have no place. The only Noumenon, the only Reality as distinct from appearance or the phenomenal (the sham Noumena, i.e. all Things in themselves other than Self - conscious Egos, being abolished) would be merely the Transcendental Ego as the source of Synthesis of Empirical elements, of such Synthesis as is expressed in the copula ' is ' in any Empirical judgment. In the Moral World shown us in the Critique of Practical Reason, on the other hand, the Reality is the Activity of rational, free, self-determining Will. I think that this application of the notion of Absolute to Kant's system is quite legitimate, when we regard the system from Fichte's point of view, and as partially transformed by his mind ; though it is certain that no such application was ever made, or would have been admitted by Kant himself. ^ 'Absolute,' I think, means here primarily Reality as contrasted with jihenomena (but also with the attribute of being unconditioned). THE METAPHYSICS OF T. H. GREEN LECTURE I SUMMARY ACCOUNT I CAN perhaps most easily show the difference between my point of view and that of Green by examining closely the language of the first page of his Metaphysics of Knowledge} Now what we are supposed to admit is, I presume, the general con- clusions of Psychophysiology, the dependence, that is to say, of the series of feelings, thoughts, etc., which constitutes our mental life, on another series of changes, viz. changes in the nerve-matter of our brain. The question still remains : " how there come to be for us those objects of consciousness, called matter and motion, on which we suppose the operations of sense and desire and thought to be dependent." Now the phrase ''he for us' is am- biguous. It may mean (1) how we come to con- ceive, (2) how we come to conceive rightly or to know, those objects called matter and motion. The first question would be purely psychological or psychogonical : it would not raise any question as ' Prolegomena to Ethics, vol. i. ch. i. § 9, p. 13. 209 P 210 THK MKTArilVSlCS OF T. U. OKEEN lect. to the validity ot" the notions. liut it seems clear that the second nicanini^ is what we require. For when we admit the functions of the soul to be dependent on material processes, we mean on the reallv occurring processes of really existing matter, not on our thoughts of these processes. In the vast majority of cases these processes occur when no one perceives or thinks of them : and they occurred, as we believe, in just the same way in the ages when no one thought of them, or when they were wrongly thought of — for example, when the heart, not the brain, was supposed to be the seat of emotion or intelligence. This is important when we come to the next sentence : " If it could be admitted that matter and motion had an existence in themselves, or otherwise than as related to a consciousness, it would still not be by such matter and motion, but by the matter and motion which we know, that the function of the soul, or anything else, can for U3 be explained. Nothing can be known by help of reference to the unknown." Now in this sentence there is a certain danger of confusion between the view of Kantian or Spencerian Agnosticism and the view of Common Sense and ordinary physical science. If by ' existing otherwise than as related to a con- sciousness ' Green means ' existing so as to be incapable of being known ' it is obviously true that matter and motion as so existing cannot furnish an explanation of the functions of the soul or anything else. And, according to Kant, ' matter in itself ' is essentially I SUMMARY ACCOUNT 211 unknowable : according to Spencer, qua agnostic : "the reality underlying appearances is totally and for ever inconceivable to us." If, then, when we say that the functions of the soul are materially con- ditioned, we mean that they are conditioned by an unknown = x,l agree that the affirmation is certainly not an explanation. But if by ' existing otherwise than as related to a consciousness ' we mean ' existing when no one is conscious of them ' — that is obviously what taught by physiology we do hold. The move- ments of nerve-particles on which we believe thought and feeling to be dependent are movements that we believe to have gone on for long ages before any one knew anything about them. In this sense we must and do conceive matter as existing in itself — capable of being known but not known. But the phrase ' otherwise than ' is confusing. For know- ledge implies that the thing known exists as it is known : so far as our conception of a thing is different from the reality, that thing is not truly known. " But," Green goes on, " matter and motion, just so far as known, consist in or are determined by relations between the objects of that connected con- ciousness which we call experience." Here again we have to disentangle and distinguish incontrovertible truth from mentalistic paradox. No doubt our common conception of matter and motion is a concep- tion of related fact : the extension, even the position in space of a thing, involves relation to all else that is extended or placed in space. The effort to conceive •21 •: THK MKTArHVi=;irS OF T TI OREEN lect. of aiivthing not related to something else would be a futile otlort. Hut the words 'consist in relations' seem to imply the absurdity that relations are conceivable without things related. And the first question that we have to put is, Does Green mean this ? This question we must put, because his language repeatedly seems to mean it : yet I think we must answer in the negative : and understand the next sentence accordingly. He does not mean to reduce matter and motion — the physical world gener- ally — to relations alone, but to relations and related feelings. " If," he says, " we take any definition of matter, any account of its ' necessary qualities,' and abstract from it all that consists in a statement of relations between facts in the way of feeling, or between objects that we present to ourselves as sources of feeling, we shall find that there is nothing left." Now so stated — apparently as a result of direct reflective analysis — I have to meet this proposition by a simple denial. But a simple denial is uninstruc- tive : let us try to explain it. The source of the error, in my opinion, lies in the fact that imaginary sensation accompanies conceptions when we dwell on them, just as sensation accompanies perception. In ordinary perception of an object external to my body I appear to cognise — and, according to Common Sense and Science, in most cases really do cognise — a portion of matter really existing (though not precisely as I conceive it) in the world known to me in experience. But along with, empirically inseparable I SUMMAKY ACCOUNT 213 from, the perception occurs feeling of various kinds : and in ordinary thought about matter elements of feeling (colour, etc.) are undoubtedly mixed. Accord- ing to me, however, reflection aided by science separates these elements, and the notion of matter in space, as used in scientific thought, is not reducible to feelings.^ Here I would ask those who hold the other view to state what feelings and relations the motions of nerve-particles conceived as concomitants of our states of consciousness mean to them : and what they mean by regarding such feelings as having existed, or to what substitutes for such feelings they attribute reality. I have never seen an answer to these questions that will stand examination. Mentalists commonly avoid the diflaculty by saying that in speaking of nerve-particles and nervous processes, or any other kind of matter in motion, they use, and have a right to use, popular language — as an instructed person does in speaking of the sun rising and setting : he knows all the time that the earth moves round the sun, and misleads no one. Similarly, they know all the time that what is called matter is really analysable into feelings and relations, and therefore with this explanation should be allowed to use the language of Common Sense freely. Now I quite admit that it would be absurd to dispute the mentalists' right to use popular terminology in merely popular discourse or writing : just as it would be pedantic to object to a modern astronomer for talking ' Cf. Philosophy, its Scope and Relations, pp. 63 ff. 214 TllK MKrArilVSlCS OF T. H. CKKKN i.kct. of the sun risiiit:; or sottiiiij;, tlu)ugli such hmjjjuage, strictly taken, implies the geoceutric view, lint my ohjectioii is not to the mentnlist's using in ordinary discourse language that implies assumptions con- tradictory of his express conclusions, but to his using such language in the professedly scientific reasonings by which the conclusions are reached. What would be thought of an astronomer who in a scientific treatise began by apparently assuming that the sun went round the earth, and carried the apparent assumption through the very arguments by which he leads us to the conclusion that the earth goes round the sun ? Surely we should require that he at least altered his terminology : we should challenge him to throw his argument in a form which avoided assumptions contradicted . by his con- clusions. That, then, is my challenge to the mental- ists who trace psychophysiologically the process by which the notion of matter in space is alleged to be compounded of feelings visual and tactual. Having said this much, I now propose to accept, for the sake of discussion. Green's mentalistic starting- point, and see how he proceeds to work out his system. The argument has two steps, one dealing with knowledge and one dealing with nature. First, we are told that * the knowledge of nature ' can only be explained by a principle w^hich is not part of nature. For knowledge of nature is knowledge of the relations of the content of experience, through which alone that content possesses a definite character and be- I SUMMARY ACCOUNT 215 comes a connected whole. The source of this knowledge of relations, of this connected experience which thus combines, unifies, organises these rela- tions, cannot itself be conditioned by them. It is commonly granted that we can only know pheno- mena : that what we call an ' objective ' world is only a phenomenal world. Still we make, and have to explain, the distinction between ' appearance or illusion ' and ' reality ' in this phenomenal world. We shall find that the terms ' real ' and ' objective ' have no meaning except for a consciousness which conceives a single and unalterable order of relations determining its experiences, an order with which, as each experience occurs, the temporary presentation of the relations determining it may be contrasted. When we make a mistake — e.g. of vision — we conceive phenomena as related in a manner incom- patible with this single system of relations. This conception of a system of relations is pre- supposed in all conscious experience : for conscious experience involves consciousness of change ; and consciousness of change involves ' consciousness of events as a related series.' Now a consciousness of events as a related series cannot be one or any number of the series of events, nor a product that supervenes after some of the events have elapsed — since "it must be equally present to all the events of which it is the consciousness." ^ Nor will it solve but only throw back the problem to say that such consciousness is a product of previous events ; unless we say that it is ' [Op. cit. § 16, p. 21.] 21 G THK Mi:r.\riIVSirS of T. II OIvKKX lect. producod i)y a series of events of which there is no consciousness. And tliat is inconceivable. In short, tlien, experience, in the sense of 'a con- sciousness of events as a related series ' — experience as the source of ii knowledge of the order of nature — cannot itself be explained by any natural history. "It would seem to follow that a form of con- sciousness, which we cannot explain as of natural origin, is necessary to our conceiving an order of nature, an objective world of fact from which illusion may be distinguished. In other words, an under- standing — for that term seems as fit as any other to denote the principle of consciousness in question — irreducible to anything else, ' makes nature ' for us, in the sense of enabling us to conceive that there is such a thing." ^ Let us assume, then, that in order to conceive experience — the very experience to which the naturalist appeals as the basis of his knowledge — we must conceive a continuing and unifying principle that is not natural, but that distinguishes itself from nature, and in knowing nature, knows itself other than nature, a consciousness which cannot be con- ceived as the product of nature, or explained by any natural history ; because it is implied in the experience through which our conception and knowledge of nature is attained. The next question is, whether ' Understanding ' can be held to ' make nature ' in the further sense that it is a source or condition of there being these relations — not only of our conceiving 1 Op. cit. § 19, p. 22. I SUMMAEY ACCOUNT 217 them. Can we hold that " the understanding which presents an order of nature to us is in principle one with an understanding which constitutes that order itself ? " ^ The common sense objections to this are not really valid. Briefly it comes to this. 'Common Sense' is supposed to hold that the relations — say of order in space, causation, resemblance — by which the mind puts together its notions of things into a coherent system, are merely notional and not real : fictions of the mind not in the things. Against this view it is easy for Green to show that if we try to conceive the things without this relation, we' fail: the things vanish.^ I pass over this for the present, because I cannot follow Green ^ in accepting Locke as a representative of the 'traditional philosophy of Common Sense.' Locke no doubt did not intend to diverge from Common Sense ; but he did diverge from it fundamentally, and thus led — as all histories of philosophy recognise — to the mentalistic paradoxes of Berkeley and the sceptical paradoxes of Hume. It was the task of Reid to trace this divergence to its source, get rid of the radical error in analysis that led to it, and thus found the Philosophy of Common Sense. But to discuss this adequately would take us too far afield. [Let us pass then to the second step in the argument.] Here Green takes as a point of departure that Nature means to us a single, unalterable, all-inclusive ' [Op. cit. § 19, p. 23.] 2 Q^ ciC. § 23, p. 26. 3 Op. cit. § 20, p. 23. 218 TllK MKTAlMIVSirs OF T. H. ClIEEN lect. sysU-m of rolalioiis." It means, even ucccording to him, something more and diHerent : a system not of rehitions onlv but of rehited facts (say feelin^i^s) — facts not 'unalterable' but in continual change, though, no doubt, such change is subject to invariable laws. But of the latter point more presently : let us now assume provisionally, and concentrate attention on, the ' single unalterable system of relations.' What is implied in such a system ? What is the condition of its possibility ? I must quote Green's answer at length, because I must confess my inability to follow his argument :-- " Whether we say that a related thing is one in itself, manifold in respect of its relations, or that there is one relation between manifold things, e.g. the relation of mutual attraction between bodies — and one expression or the other we must employ in stating the simplest facts — we are equally affirming the unity of the manifold. Abstract the many relations from the one thing, and there is nothing. They, being many, determine or constitute its definite unity. It is not the case that it first exists in its unity, and then is brought into various relations. Without the relations it would not exist at all. In like manner the one relation is a unity of the many things. They, in their manifold being, make the one relation. If these relations really exist, there is a real unity of the manifold, a real multiplicity of that which is one. But a plurality of things cannot of themselves unite in one relation, nor can a single thing of itself bring itself into a multitude of relations. 1 Op. cit. § 26, p. 29. I SUMMAEY ACCOUNT 219 It is true, as we have said, that the single things are nothing except as determined by relations which are the negation of their singleness, but they do not therefore cease to be single things. Their common being is not something into which their several existences disappear. On the contrary, if they did not survive in their singleness, there could be no relation between them — nothing but a blank, feature- less identity. There must, then, be something other than the manifold things themselves, which combines them without effacing their severalty."^ I grant that " relation involves the existence of many in one " — that what we conceive as one thing we, in so conceiving it, necessarily conceive as having many relations, and that any one relation must be a relation which connects a plurality (two at least) of objects related. But what is the meaning of saying that " a plurality of things cannot of themselves unite in one relation, nor can a single thing of itself bring itself into a multitude of relations " ? I thought the aim of the preceding argument was to show that they are in the relation and cannot be conceived out of it. What, then, is the meaning of the phrase ' cannot of themselves unite ' ? In order even to ask the question. Can they of themselves unite ? etc., we must conceive them out of the relation : whereas Green's point is they cannot be so conceived. This, indeed, he dimly sees [as is evident] from what he goes on to say : "It is true . . . that the single things are nothing except as determined by relations which are the 1 Op. cit. § 28, p. 31. 220 TllK .MKIAI'HVSICS OF T. H. CJKEEN lect. negations of their singleness, but they do not therefore cease to be single things. ... On the contrary, if they did not survive in their singleness, there could bo no relation between them — nothing but a bhiuk, featureless identity." But the fact that they survive in their singleness does not show that they need something other than themselves to make them so survive. However, let us grant that unless we " deny the reality of relations and treat them as fictions of our combining intelligence " we must suppose them to be held together by something other than themselves. Then, as in the world of experience, the world as presented to sense and represented in thought, we find on reflection that the unifying principle is a conscious intelligence, so we must suppose that in the world of reality there is an analogous principle. " If we suppose them (the relations) to be real other- wise than merely as for us, otherwise than in the ' cosmos of our experience,' we must recognise as the condition of this reality the action of some unifying principle analogous to our understanding." ^ At this point Green takes some pains to deal with the doctrine of Kant which distinguishes [between ' phenomenal reality ' and] ' reality in some absolute sense.' Into this argument I do not now propose to enter. I have already given my own criticism of Kant ; and the doctrine of ' unknowable things in themselves,' though I will not say that it is not held by scientific men,' is at any rate not one in which ' Op. cit. § 29, p. 32. ^ Mr. Spencer in a sense holds it. I SUMMARY ACCOUNT 221 scientific men as such take much interest. If there is a world of such uuknowables, it is at any rate not the world of past and present reality into which science is ardently inquiring — with a firm conviction of its power of distinguishing the real from the unreal, truth from error, with regard to it. Let us keep ourselves to this world, and ask what is required to make it conceivable. LECTUKE II THE SPIRITUAL PRINCIPLE IN KNOWLEDGE AND IN NATURE I PROPOSE to begin the present lecture with a brief examination of the fundamental points of the doctrine of Green's chap, i., so far as I gave a summary account of them in the last ; and then to proceed with the critical exposition from the point where I left off. The conception of Knoivledge we all agree requires a knowing mind : the main drift of the chapter is to show that the conception of Nature involves it equally. The argument might be put in two sentences : — (1) Nature as known and as knowable is a system of objects related to a subject or knowing mind, and related to each other through their relation to the subject. (2) No other Nature is conceivable. Materialists — and Common Sense so far as the physical world is concerned — think of Nature as matter in motion. But Matter means " relations between facts in the way of feeling, or between objects that we present to ourselves as sources of feeling," and Motion similarly " has no meaning except ... as expressing relations of what is con- 222 II SPIKITUAL PKINCIPLE IN KNOWLEDGE 223 tained in experience." ^ Also " it is an accepted doctrine of modern philosophy " that " knowledge is only of phenomena," and that " nothing can enter into knowledge that is unrelated to consciousness/' ^ which Green takes as meaning that " relation to a subject is necessary to make an object, so that an object which no consciousness presented to itself would not be an object at all." But the last two propositions are not necessarily identical : since the former may be held, and has been widely held, by Mentalists of a different type from Green : those whom in previous lectures I distinguished as Sensationalists.^ There is in fact an ambiguity in it depending on an ambiguity in the word ' consciousness.' This word is sometimes used as equivalent to, or definitely including 'self- consciousness,' the reference, i.e., to a permanent identical self or subject of the stream of transient changing psychical fact which constitutes the varying element of the mind's empirical life ; but sometimes again it is used for this varying element itself. In this latter sense, the proposition that " nothing can enter into knowledge that is unrelated to conscious- ness" may be held by a Sensationalist, who agrees with Hume that when he observes himself he cannot find anything but a changing complex of transient facts, ultimately analysable into Sensations or Feelings. I note this, because Green's polemic appears to be primarily directed against this view, and not 1 [Op. cit. § 9, p. 13.] 2 [Qp_ cit. § 10, p. 14.] ' Cf. Philosophy, its Scope and Itelatioiis, p. 52. 224 THK MKTAriTVSICS; OF T. H. GREEN lect. against the philosophy of Common Sense or Natural Dualism. But no philosophy can ever ignore Natural Hualism. The result is that there is a kind of ' triangular duel ' : a contest in which three distinct views are involved ; each of the two opposed to Green's partially agrees and partially disagrees with his view. The Sensationalist and the Idealistic Spiritualist (Green) agree in being raentalistic : i.e. in reducing the material world, at least as known and knowable, to mental fact of some kind : the Natural Dualist and Green agree as against the Sensationalist in recognising reference to an identical self as an essential and permanent element of consciousness. Availing himself of this division of his opponents, Green puts together the views in which each agrees with him, and takes the world as known to, and believed to exist by, each individual, as a world of essentially mental fact, every part and element of which is necessarily related to a conscious subject. But this does not yet bring us to Green's characteristic doctrine. The elements of this empirical world of each individual are not only essentially related to a conscious subject : they are essentially related to each other, each to all, — related through position in time, position in space, resemblance, causal connexion, reciprocal action. No object of experience is conceivable apart from a whole complex of such relations. Nature then, no less than the experience of each individual, is for Green a connected system of objects of consciousness, which are what they are — when we rightly conceive them — through the relations II SPIEITUAL PKINCIPLE IN KNOWLEDGE 225 that connect them. I have said ' objects of conscious- ness.' This appears to be for Green equivalent to related ' feelings,' with the proviso that the difference between one feeling and another consists in its different relations. This appears from his analysis of error in empirical cognition. The question " whether any impression is or represents anything real or objective " is a question " whether a given feeling is what it is taken to be," that is, " whether it is related as it seems to be related," for "a particular feeling is [merely] a feeling related in a certain way." Error in empirical cognition, accordingly, consists in a mistake as to the relations of what is felt — in con- ceiving a certain set of relations so that they are incapable of combining into a system with other recognised relations. Or, as the Table of Contents says, " The question, Is anything real or not 1 means Is it, or is it not, related as it seems to be related ? " It thus " implies the conception of reality or nature as a single unalterable order of relations."^ Thus the essence of ' Nature ' is for Green ' an order of Nature. On the other hand, " Nature with all that belongs to it is a process of change : change on a uniform method, but change still. All the relations under which we know it are relations in the way of change, or by which change is determined."^ The question. How can Nature be at once ' unalterable ' and a ' process of change' seems to require more consideration than Green vouchsafes it ; but what he means is that the real world, though perpetually changing, is changing 1 [Op. cit. p. X., pp. 16 ff.] "^ Op. cit. § 18, p. 22. 226 THK MKTArilVSlCS OF T. TI. (;REEN lkct. accordin*; to uiiolianuiii^ laws. On this more presently. However, taking this view of Nature as essentially a single connected all-inclusive system of relations, in a sense unalterable, let us now examine the non-natural principle which it necessarily implies. (1) Why is this 'non-natural,' and (2) what is its relation to Nature ? In the earlier part of the chapter, in which Green is considering the " Spiritual Principle in [empirical] knowledge," his answer seems clear. ' The relations of the experienced ' must have a ' source,' a ' principle of union,' from whose ' com- bining and unifying action ' they ' result ' : and this Principle " being that which so organises experience that the relations . . . arise therein," cannot " itself be determined by those relations." ^ But why do the relations want a source ? Why cannot they get on without one ? These questions are answered in a passage (§ 28) to which I directed special atten- tion at the close of the last lecture.^ As I said, the argument appears to me invalid on Green's premises ; because, according to him, we cannot even conceive the manifold things out of the relations : and, there- fore, cannot even raise the question whether, if we could conceive them out of the relations, they would be seen to require something other than themselves to bring them into the relations. We must conceive the real world as a system, having unity and con- nexion as well as manifoldness and diversity ; but I cannot see why we should therefore suppose a special source for the unity ; or why " either we must deny the ' [Op. fit. § 9, p. 14.] 2 Qf above, p. 219. II SPIRITUAL PRINCIPLE IN KNOWLEDGE 227 reality of relations altogether and treat them as fictions of our combining intelligence," or we must suppose the manifold things to be held together by something other than themselves.^ But still, granting the fundamental assumption of Mentalism, I admit the force of the argument which Green urges, from the analogy between the world of each one's experience and the ' real ' world, common to us all, of which the world of each one's experience and thought is an indefinite fragment. If the aggregate of thoughts and feelings into which the world as empirically known to me is analysable has every element of it connected by reference to a self-conscious subject, we may argue from analogy that there must be such a subject similarly related to the Universe. Before I proceed to examine further Green's con- ception of this universal principle or non- natural subject, I must say a word on his relation to Kant, as explained by himself (§§ 31-41). It will be seen that he is arguing for some time on Kant's side {§§ 31-37) in favour of the doctrine which they agree in holding, viz. that what Green calls a ' principle of consciousness,' Kant a ' synthetic unity of appercep- tion,' is the source of form, relation, and connexion in the world of empirical reality. Observe that Green does not in this argument distino-uish forms of sensi- bility from forms of thought : and that he abstracts from the difierence between Kant's phenomenalism and his own mentalism. For the ' Nature,' that in Kant's view is made by the Understanding and J [Cf. op. cit. § 29, p. 32.] 228 THE MKTArTrVSTCS OF T. H. GREEN lkct. Imagiiiatit)!! detonniiiinii; tlie toriii of time in which the data of sense are apprehended — this is merely the systematised appearance of a really real world not existing under the conditions of Time and Space. The objective empirical world, in Kant's view, is therefore only a world common to human subjects, and gives no ground for Green's supposition of a Universal Subject of the Universe of Reality, being, as I said, for Kant independent of the forms of human sensibility and understanding. Then (§§ 38-41) Green argues against Kant's unknowable world of things-in- themselves. In § 38 he states fairly the difference between Kant's view and his own, — though keeping in the background the complexity of the Kantian psychology, the threefold distinction between forms of sensibility, forms of thought, and ideas of reason. But in his argument against this view there seems to me a certain mis- apprehension of Kant. Green says that Kant's distinction between ' form ' and ' matter ' implies " that phenomena have a real nature as effects of things-in-themselves other than that which they have as related to each other in the universe of our ex- perience. And not only so, it puts the two natures in a position towards each other of mere negation and separation, of such a kind that any correspond- ence between them, any dependence of one upon the other, is impossible. As efifects of things-in-them- selves, phenomena ^ are supposed to have a nature of ' By the way, ' feelings ' or ' sensations ' is more appropriate than ' phenomena ' for Kant's view of the matter of empirical objects. II SPIRITUAL PRINCIPLE IN KNOWLEDGE 229 their own, but they cannot, according to Kant's doctrine, be supposed to carry any of that nature with them into experience."^ (l) The first sentence and the last suggest that in Kant's view the data of sense, by synthesis of which empirical objects are formed, could actually exist apart from the forms (of sensibility and thought) due to the constitution of the human mind. But I know no ground for attributing this view to Kant. (2) The second sentence altogether ignores Kant's view that the forms of the Understanding were applied to the data of Sense through the schemata or time-determi- nations due to the exercise of pure or productive imagination. These schemata, e.g. * 'permanent,' ' mutable,' for subject, predicate ; antecedent, tem- porally consequent, for reason, logically consequent, bring about, in Kant's view, just the correspondence required for the synthesis of form and matter in knowledge of empirical objects. But the criticism of Kantism in the following passage seems to me to hit the mark, still with the partial misunderstanding in one sentence that I have just indicated : — " The ' cosmos of our experience ' and the order of things-in-themselves will be two wholly unrelated worlds,^ of which, however, each deter- mines the same sensations. All that determination of a sensible occurrence which can be the object of possible experience or inferred as an explanation of experience — its simple position of antecedence or ^ Op. cit. § 39, p. 41. '^ Op. cit. § 39, p. 42. ' Unrelated ' — in the one causes, effects in the other. 2:>0 THK MKTArHVSirS OF T H. GREEN lkct. sequence in time to other occurrences, as well as its relation to conditions which regulate that position and determine its sensil)lo nature — will belons: to one world of which a unifying self-consciousness is the organising principle : while the very same occurrence, as an effect of thinijs-in-themselves, will belonor to another world . . ." ' So again, the objection in § 41 to the causal rela- tion which Kant assumes to exist between ' things- in-themselves ' and their effects on sensibility seems to me sound ; as according to Kant we have no warrant for extending the application of the category of causality, in any -positive way, beyond the limits of experience. The assumption, therefore, that 'things- in-themselves ' (other than thinking beings) are causes of phenomena contradicts the principles of Kant's Analytik. I have said enouorh on Green's relation to Kant : I return to the exposition of his own system. Dropping ' things-in -themselves,' he conceives the real world, the only real world, the Universe, as a connected order of knowable facts, and therefore essentially a ' single, unalterable, all-inclusive system of relations.' This real world, therefore, presupposes, just as the experience of each finite mind presupposes, a combining, unifying, self-distinguishing principle or subject which by its synthetic action constitutes the relations that determine phenomena. It is a principle other than nature ; for " the relations by which, ^ ' Belong to another world ' — according to Kant it only belongs to this in respect of its causation, not in respect of its quality. Otherwise Green's objection to the double determinatioa of the phenomenon seems to me sound. II SPIRITUAL PRINCIPLE IN KNOWLEDGE 231 through its action, phenomena are determined are not relations of it — not relations by which it is itself determined. They arise out of its presence to pheno- mena, but the very condition of their thus arising is that the unifying consciousness which constitutes them should not itself be one of the objects so related." ^ This principle, therefore, is not in time, nor in space. It is not material, nor subject to motion, for matter and motion are merely names of relation ; it is not a substance, for ' substance ' is only a correlative of change, has no meaning or conceivable existence apart from change. This One Subject, therefore, is not to be conceived as the substance of the world, for " that connexion of all phenomena as changes of one world which is implied in the unity of intelligent experience cannot be the work of anything which the substance qualified by those changes."^ Such a non-natural self-conscious subject is what Green means by a Spiritual Principle. Let us look a little closer at this strange entity. It is not, as Green has before explained (§ 41), a cause of which nature is the efi'ect, for " causation has no meaning except as an unalterable connexion between changes in the world of our experience." ^ But what then is meant by saying that this non- natural principle is a ' source ' of relations, that they " result from its combining and unifying action " ? "* Surely this is only saying in other words that they are efi"ects of which it is a cause. Green seems to I Of. cit. § 52, pp. 54, 55. * q^ ^it. § 53, p. 56. 3 Op. cit. § 41, p. 44. * [Op. cit. § 9, p. 14.] USJ THK MKTAI'HYSICS OF T. H. CJKEEN lkct. admit the ditiiculty, luul to answer by saying that this languafj^e is ' meta})horical.' ' But surely it is a weak position when such fundamental notions us 'source,' 'action,' 'agency,' etc., are admitted to be used ' metaphorically,' and yet no attempt is made to justify or explain the 'metaphor' by st)me clear and precise statement of the truth it adumbrates. But let us suj)pose that these terms, apparently implying a causal relation, really mean something else. The fundamental difficulty is not removed. Green's argument was that this principle of union cannot be conditioned by any of the relations that result from its combining and unifying. How then are we to obtain a conception of its relation to nature ? for any such conception must have a ' unifying ' effect : it must enable us to form a coherent view of Nature and Spirit taken together. There is, indeed, one conception which is at least free from the special objections urged against the notions of ' action,' 'agency,' ' source,' ' results,' etc., as applied to the relation between Spirit and Nature : and this is the conception which the main line of Green's argument suggests. That is the relation of subject and object in knowledge, in its simplest form — the cognition of an object by a subject, or the pre- sentation of an object to a subject. For this relation is disparate from, unlike any relation among objects; and thus this relation comes to the front in the next chapter On the Relation of Man, as Intelligence, to the Spiritual Principle in Nature, to which I pass. ' Op. cit. § 54, p. 57 init. II SPIRITUAL PRINCIPLE IN KNOWLEDGE 233 Here we are led to analyse more closely the fact of human knowledge. Man, as a being that knows, is not a mere series of events : human consciousness is not a mere stream or succession of changing states. Knowledge is of related facts : it is essential to every act of knowledge that the related parts of the object known should be present together to the knowing mind. " The acts of consciousness in which the several members are apprehended, as forming a know- ledge, are a many in one. None is before or after another. This is equally the case whether the know- ledge is of successive events or of the ' uniformities ' which are said to constitute a law of nature." ^ As an instance Green takes " a man's knowledge of a pro- position in Euclid. This means a relation in his con- sciousness of certain parts of a figure determined by the relation of these parts to certain other parts. The knowledge is made up of those relations as in con- sciousness. Now it is obvious that there is no lapse of time, however minute, no antecedence and consequence, between the constituent relations of the consciousness so composed " — in this I quite agree. But Green goes on — " nor between the complex formed by these constituent relations and anything else." ^ If ' the complex formed,' etc., means — as the words seem to mean — the whole state of consciousness, this state- ment cannot be accepted. It seems clear, on the contrary, that there is the most definite time-relation of ' antecedence and consequence ' between the com- plex consciousness which constitutes the knowledge ' Of. dt. § 56, p. 61. 2 Qp cit. § 57, pp. 61, 62. 234 THE ^^KTA^HYSTC^^ OV T. H. GREEN lkct. of a demonstrated conclusion and the intellectual apprehension of the successive steps of the demon- stration. This is so plain, that when Green draws the inference that this knowing consciousness is not a ' phenomenon,' not an ' event in the individual's history,' he seems to be confounding the knowing • ' Do consciousness with the object known. It is no doubt true that when we consider the object of any one's knowledge — say a proposition of Euclid — the system of relations of which it consists is independent of time. Though complex, there is no succession, no lapse of time between its parts : and it is — so far as it is true knowledge — unalterable, the same at one period as at another. It is not affected by the fact that A knew it yesterday, B knows it now, and C will know it to-morrow. And this is also true, as Green points out, when the knowledge is of successive events. Take, e.g., the knowledge that I have gone through half a dozen steps of reasoning in learning a proposition of Euclid. It is true of this knowledge, no less than of knowledge of a demonstrated con- clusion, that it does not itself consist of successive steps, but is a single apprehension of such successive steps ; and it is no less true that so far as this complex fact is truly known, it may be equally well known by any one else at any subsequent time. All this is true : but it does not justify the inference that this single apprehension of a complex truth — whether geometrical or biographical — is not an event in my mental history. If, as Green says, in learning a II SPIEITUAL PRINCIPLE IN KNOWLEDGE 235 proposition of Euclid, a series of events takes place, surely we must recognise the conscious know- ing of the proposition as the final event of the series. We cannot allow him to pass from the " conscious- ness which constitutes a knowledge " to " the content of such consciousness " as though they were identical conceptions. We may admit the content not to be an event in time, but we cannot admit that the knowing of it is not an event in time. Indeed when Green tells us that " a known object is a related whole, of which . . . the members are necessarily present together," he seems to mean that they are present simultaneously ; his argument, in fact, has no force unless he means this. But what is simul- taneously present must be present at some particular instant — or during some particular period — to some particular knowing subject. And if so, surely what thus happens or begins to happen must be an event in the history of this person ! The truth appears to be that Green is so concerned to lay stress on the points (1) that knowledge is not a succession of states of consciousness, and (2) that the complex relations that make up the extent of any act of knowledge are present together, in this act, to the knowing mind, that he allows himself to be carried along to the paradox of asserting that the act of knowledge itself is not an event in the mental history of this mind. Now a knower who knows, but does not know at any point of time or through any period of time, is absolutely inconceivable to me, and 23r. THE MKTArHYSTCS OF T. TT. CliEEN lect. nothing in my I'XjH'rienco helps rac towards con- oeiving it. Mere I may conveniently develop another criticism, brieHy noticed before. In several passages of Book I. Green speaks of the real in human experience as a "single unalterable system of relations." I let this pass provisionally, because there is a sense in which the epithets may stand. If we assume that all events are completely determined by their antecedents, then the whole process of change in which our minds live, and which it is the effort of the study of nature to know, is in a sense unalterable : — i.e. from a complete knowledge of the [physical] world at any point, in- cluding all physical laws of change, we could infer the past, as far back, and the future, as far forward, as we choose to follow it. But because the process of cosmic change is deter- mined, and in this sense unalterable, it does not therefore cease to be a process of change, of which it is an essential condition that it takes place in time. Accordingly when Green concludes that any act of knowledge — even the " ordinary perception of sensible things or matters of fact" — involves "the determina- tion of a sensible process which is in time by an agency which is not in time," we have to point out that this sensible process must be a part of the whole cosmic process — of the " single and unalterable system of relations," and must as a part of this be completely determined ; so that there would seem to be no room for any other determination. To this I shall return. However, Green's conclusion is that the knowledge II SPIEITUAL PEINCIPLE IN KNOWLEDGE 237 we arrive at through sensation and sense-perception is not itself in time — though the sensation is — and implies the presence of an agent not subject to the conditions of time, an ' eternal ' and ' spiritual ' principle. LECTURE III THE RELATION OF MAN TO THE SPIRITUAL PRINCIPLE IN NATURE Before passing to the end of chap. ii. and chap. iii. let us review the ground so far traversed, and try- to make clear to ourselves the results attained. First, I may again remark that the controversy between Green's Idealism and the Philosophy of Common Sense or Natural Dualism — to which I provisionally adhere — is never prominent or important in Green's argument. As we saw, he treats this line of thought so slightly that he takes Locke as a representative of what is called Common Sense and does not even allude to Reid, and obviously knows nothing of his work. His chief controversy accordingly is not with Natural Dualism which maintains the current distinction between mind and matter, accepted by Common Sense and Physical Science, including Psychophysiology. It is not this view which he conceives his opponent to hold, but rather a species of what I call Mentalism — the philosophical view that resolves matter altogether into mental elements. The species is that which I think it convenient to distinguish from Idealism as 238 LECT.iii SPIKITUAL PEINCIPLE IN NATURE 239 Sensationalism — which resolves matter into Feelings, and is mainly English. He also has in view what I may perhaps distinguish as Phenomenalism, the doctrine that resolves matter as known into elements of feelings while recognising an unknown external matter whose action on us causes these feelings. To this view he naturally opposes, in a great measure, the same arguments that he opposes to the more paradoxical Sensationalism ; but he has to add, in dealing with Phenomenalism, a confutation of the supposed unknown substratum of matter, the ' thing by itself,' that has the support of Kant's authority. Into this dispute between schools, to neither of which I belong, I have entered but slightly : my concern has been with Green's own system and its construction. This construction, however, is influenced by the system of thought that he con- ceives to be opposed to him. The adversary is supposed to hold that the world is composed of feelings as elements. Against this Green's point is that " feelings without relation are nothing to us as thinking beings," and that the concatenation of objects which make up for each mind its experience or its empirically known world consists essentially of relations, of which as thought-relations some principle other than any or all objects, some thinking principle, must be conceived as the source. The source of these thought-relations, that which combines, unifies, organises experience, cannot be conditioned by the relations, and therefore cannot be conceived as a part of the empirical object world that exists LMO TIIK METAPirYSTCS OF T. U. OREEN lkct. for each. T\nn\ piissing from the world of each one's experieuce to the hirger common object world that each conceives to exist for all, we find ourselves led to postulate a similar non- natural principle for Nature. Nature is thus conceived as essentially a single unalterable all-inclusive system of relations, by which all phenomena are combined into a systematic whole : and the source of connexion, the combiner, the unifyer, must be a non - natural or Spiritual Principle. Here I made one criticism. How, as no element of Nature is conceivable out of relation, can we conceive it as requiring a non -natural principle to bring it into relation ? It seems that in order to exhibit the evidence for a non-natural principle Green has first to conceive Nature as analysed into elements ; yet this in the same breath he declares to be irrational and inconceivable ! Surely this will not do. But further difiiculties appear when we examine Green's fundamental doctrine that the relations by which through the action of this non -natural principle "phenomena are determined are not relations of it — not relations by which it is itself determined."^ First, it is difficult to under- stand how this universal Principle is, on Green's premises, conceivable. He has laid down that everything which is an object of thought to me must be determined by relations which my thought supplied : and that this eternal self- distinguishing consciousness cannot be conceived as determined 1 [Op. cit. § 52, p. 54.] Ill SPIEITUAL PEINCIPLE IN NATURE 241 by the relations of which its activity is a source : "the very condition of their thus arising is that the unifying consciousness which constitutes them should not be one of the objects so related." ^ But if it cannot be thought under its own relations, surely it can no more be thought under the relations which are the product of my intellectual activity — since, as we noted last time, I am qua thinker, a mere limited reproduction of the eternal consciousness. But if so, how can this eternal consciousness be an object of thought at all to me, consistently with Green's general view of thought and its objects ? Similarly, how can other human beings, conceived as self-conscious selves, be such objects? Finally, how can I myself be properly an object of my own thought ? The difficulty takes another form when we examine the relation of the non- natural principle to Nature. It is not in Time, not in Space, not a Substance, not a Cause — for " Causation has no meaning except as an unalterable connexion between changes in the world of our experience."^ But then what is meant by saying that it is a source of relations, and that they * result from ' its combining and unifying action ? To this, as we have seen,^ Green only gives the singularly weak answer that this language must be taken to be 'metaphorical.' I do not debar a philosopher from the use of meta- phor by way of illustration : but I think he is bound also to state his meaning in unmetaphorical language : 1 {Op. cit. § 52, p. 54 fin.] 2 ^Qp ^it. § 4I, p. 44.] 3 Cf. above, p. 232. R 24-J TllK MKl'VPHVSKN OV T. II. CKKKN i.kct. and this Green does not seem ever to do as regards the important point that we are discussing. There is another inconsistency in his conception of Nature. It is a " single, unalterable, all-inclusive system of relations."^ But why not 'related feelings ' — frrantintj the negation of thinf^s other than feelings — surely relation must relate something ! Green's only answer is that " feelings without relations are nothing to us as thinking beings." But that is his answer. The question is whether thought- relations are not equally inconceivable without feelings. Green to our surprise ultimately admits this : he is as willing to deny that there can be * mere thought ' as ' mere feeling,' he declares *' feeling and thought to be inseparable and mutually dependent " : and yet, having admitted this, goes on speaking of Nature as essentially a " single, unalter- able, all-inclusive system of relations" ; and throughout his discussion, seems to ignore feelings completely in his account of the real world. I cannot refrain from conjecturing that in this Green has been unconsciously influenced by the desire to avoid attributing feelings to his universal self-distinguishing consciousness ; as this would clash with the tradi- tional philosophical conception of the Divine Mind as Rational but not Sentient.^ However, we seem to be left with (1) Nature as a single unalterable — though, by the way, continually changing — system ' [Op. cit. §29, p. 30.] ^ This difficulty seems to me to attach to all Neo- Kantian attempts to reconstruct the Kantian view of the world without 'Things in them- selves.' Ill SPIRITUAL PEINCIPLE IN NATURE 243 of thought- relations, with feelings admitted to be somehow inseparable from thought, though ignored in the definition; and (2) with a spiritual principle which ' acts ' without causality, unifies and combines what is inconceivable otherwise than in combination, and, in short, of which we can form no distinct conception except that it is a subject related to the world of objects as each one's intellect is related to the objects of his own experience. The one positive conception which he does give of spirit is more closely contemplated in chap, ii., in which the relation of man as intelligence to the Spiritual Principle in nature is considered. Here again it should be observed that Green's antagonism is primarily to a sensationalist explanation of know- ledge which professed to resolve an act of cognition into a series of feelings. Against this he urges well and forcibly that in the knowledge of any complex object — whether a succession of past events, or the uniformities of nature, or a geometrical proposition — all the relations of the parts of the object known must be apprehended by the mind in a single act. He also urges truly that the content of any cognitive consciousness — so far as it is truly knowledge — is unaffected by the time at which (or the knower by whom) it is known. It is the same yesterday and to-morrow, for me and for you — otherwise it would not be true knowledge. This is the part of the argument which I consider most sound and con- structively important. But from these sound premises Green draws the 244 THK MKlArilVSKS OF T. H. (UlEEN lect. startlinn foiiolusioii that tliis cojiiiition is not ' an event in the imlividiiars history.' If, however, loaruing a proposition of Euclid is — as Green says — a series of events in the individual's history, it is absurd to refuse to recognise the conscious knowing of the proposition as the final event in the series. Also, when he says that the members of a known object are " necessarily present together . . . none before or after another," he in effect says that they are present simultaneously. But simultaneity is a time-determination as well as succession : what is simultaneously present must be present at a par- ticular point of time, or through a period of time — through all time if we like. In fact, however, Green recognises that ' our perceiving consciousness ' has itself apparently a history in time. The solution [of this seeming inconsistency] I must give again ^ in his own words, as it is one of the cardinal points in his teaching : *' Our consciousness may mean either of two things : either a function of the animal organism, which is being made, gradually and with interruptions, a vehicle of the eternal consciousness ; or that eternal consciousness itself, as making the animal organism its vehicle and subject to certain limitations in so doing, but retaining its essential characteristic as independent of time, as the determinant of becoming, which has not and does not itself become."^ He afterwards speaks of the eternal consciousness as a ''system of thought and knowledge which realises ' Cf. above, p. 235 fin. - Op. cit. § 67, p. 72. Ill SPIEITUAL PEINCIPLE IN NATUEE 245 or reproduces itself" in the individual.^ Let us consider these ' two meanings ' of consciousness. They seem to be two very different things : (l) a function of an animal organism ; (2) an eternal consciousness limiting itself and making the animal organism its vehicle. The conceptions seem as clearly distinct as can be ; but what then becomes of the 'self in this doubleness? Oh! Green assures us " our consciousness is one indivisible reality " ^ of which these are two aspects, the same thing regarded from two different points of view. But then there would seem to be a third meaning of ' conscious- ness': it is the (3) "indivisible reality" of which (l) and (2) are aspects, but it is also both of these. I confess I find it difficult to conceive God as an aspect of myself, and yet God existing already and eternally as all -knowing. Green adduces the old simile of the two sides of a shield. But we can see clearly how the two sides of a shield are united into a continuous surface by the rim ; we surely cannot similarly see how 'one indivisible self should result from an eternal consciousness limiting itself and using the animal organism as its vehicle. ^ Op. cit. § 68, p. 74. Observe the alternatives offered, ' realises ' or 'reproduces.' If 'realises' is the right word, then the eternal consciousness is only potentially, if 'reproduces,' then it is actually, existent apart from the finite individual. Surely a little more ought to be said on these alternatives, as the difference is, from a theological point of view, immense. It is, in fact, the issue between Hegelian Theism [Right] and Hegelian Atheism [Left] that is thus slurred over. But I think Green must be taken theistically : and therefore to mean 'reproduces' rather than 'realises,' as he holds (§ 69. p. 75) that "there is a consciousness for which the relations of fact, that form the object of our gradually attained knowledge, already and eternallj'' exist " — though how anything can exist already for a subject out of time, he does not explain. 24 'I'llK M I:T A I'll VSU;S OK '!". II. (IHKKN lkct. The explaiKitioii is: "The conscioiisiicss wliicli varies from moment to moment, which is in suc- cession, anil of which eacli successive state depends on a series of 'external and internal' events, is cousciousness in the former sense. It consists in what ma}' properly be called phenomena ; in suc- cessive modifications of the animal organism, which would not, it is true, be what they are if they were not media for the realisation of an eternal conscious- ness, but which are not this consciousness. On the other hand, it is this latter consciousness as so far realised in or communicated to us through modifica- tion of the animal organism, that constitutes our knowledge, with the relations, characteristic of knowledge, into which time does not enter, which are not in becoming but are once for all what they are." ^ But does not this, Green supposes his reader to ask, " involve the impossible supposition that there is a double consciousness in man ? No, we reply, not that there is a double consciousness, but that the one indivisible reality of our consciousness cannot be comprehended in a single conception. In seeking to understand its reality we have to look at it from two different points of view . . ."^ Here and elsewhere Gjeen is so much occupied with distinguishing intellect and knowledf^je from mere feelings that he is led to obliterate the distinction between 'psychical' and 'physical' phenomena. He seems to say that the "consciousness that varies from moment to ' Op. cit. § 67, p. 72. 2 Op. cit. § 68, p. 73. Ill SPIEITUAL PEINCIPLE IN NATUKE 247 moment . . . consists in successive modifications of the animal organism," but surely my conscious- ness which varies from moment to moment is a stream of psychical facts, distinct from modifica- tions of a material organism, however these may be mentalistically interpreted. But I will not dwell on this. Let us assume with him that the important distinction is between ' know- ledge ' and ' sentient life,' and not between ' feeling ' and 'matter in motion.' The difiiculty is not the least reduced. One of the things I am most certain of is the unity of myself. Green says that (1)1 am really two things, so disparate as an eternal con- sciousness out of time, and a function of an animal organism changing in time ; and yet at the same time that (2) I am one indivisible reality contemplated from two difi'erent points of view. I submit that Green is bound to reconcile this contradiction, which he does not do by simply stating that both con- tradictory propositions are true. As it is, his doctrine is rather like the theological doctrine of the Athanasian Creed, only the Athanasian Creed does not profess to give an intelligible account of the mysteries that it formulates. But apart from this there is a further difiiculty, or rather the old difficulty of chap. i. — the difiiculty of conceiving the eternal subject, according to Green's view of it, as a cause of which anything in the world is the effect. For it will be observed that, in the later pages of chap. ii. (§§ 67-73), these causal terms recur. The eternal consciousness " makes the animal •2-ts Tin: MKrAPHY.^irs of t n huekn lect. organism its vehicle," it is the " dotermiiiant of becom- ing": it is "operative" througliout the succession of events wliich constitute the growth of the individual mind : it " acts on the sentient life of the soul " and *' uses it" as its organ. These are all terms which imply the causality of the eternal subject, in special relation to a certain })art of the world in space and time, in the most definite and unmistakable way. They are just as irreconcilable as the terms used in chap. i. with the statement repeatedly and emphatically made that the relations by which the non-natural subject unifies Nature are not predicable of it, the subject : and in particular with the statement that " causation has no meaning except as a connexion between changes in the world of our experience."* Are they also ' meta- phorical,' and if so, what becomes of the whole view if metaphor is discarded ? This question may per- haps find an answer in chap, iii., to which I now pass. It is in virtue of this " self-realisation or reproduc- tion in the human consciousness of an eternal consciousness not existinfj in time, but the condition of there being an order in time, and an intelligent experience," that we are entitled to say that " man is a free cause." ^ The term ' cause ' is, indeed, not strictly appropriate, since, though this ' eternal consciousness ' or ' unifying principle ' distinguishes itself from the manifold which it unifies, it must not be supposed that it has " another nature of its own apart from 1 [Op. cU. § 41, p. 44.] 2 ^Qp cit. § 74, p. 79.] Ill SPIEITUAL PKINCIPLE IN NATUEE 249 what it does in relation to the manifold world." ^ But what is meant by calling it a cause and what does it cause ? Green says that " but for our own exercise of causality" in knowing the statement would have no meaning.^ We know the action of our own minds in knowledge, we infer thus the action of the self-originating mind in the universe. How we can apply the notion of causation in any sense consistently with what is stated in chap. i. as to the im- possibility of applying thought-relations to the source of these relations is not explained. But I will not dwell further on that. Let us try to get a clear idea of what the action is. The agent is said to give the world its character: — that would seem to mean creates it. But then the agent must have a determined character apart from the world ; and that, as we have seen, is denied : " it must not be supposed that the unifying principle has another nature of its own apart from what it does in relation to the mani- fold world." It seems, in short, to be a cause that is nothing apart from its effect. Green is perhaps aware of the obscurity of his statement, for he takes pains to repudiate any notion of explaining with any detail what the work is as a whole. ^ Perhaps if this declaration of philosophic impotence had been made at the outset, the reader would have read with less ardour. But however little Green offers an explanation of the world, at any rate he offers an account of it : and it seems not unreason- ' Op. cit. § 75, p. 80. 2 [Op. cit. § 77, p. 82 inil.] ^ Cf. op. cit. § 82, p. 86. 250 rHK MKTArHVSI(\^ OF T. 11. (IKKEN lv.ct. able to tloin.nid tliat tlio account sliould not contain inconsistent conceptions. Wluit are we to make of a subject out of Time, to wiiich objects are ah'eady present — a sul)ject to which we cannot apply any thouujht-rehitions, because it is tlie source of all such, yet which we must think as making, determining, openly acting : a subject which gives the world its character, and yet has no nature of its own apart from it ? And what again of a world composed of thought -relations, admitted to be inseparable from feelings related, and yet of which the thought - relations are given by a subject to which feeling is never attributed ? However, let us take it that this agent is a free cause, and man, as knowing, is similarly free so far as his consciousness is " identified by this eternal consciousness with itself, or made the subject of its self-communication," — so far, in short, as it is a ' reproduction ' of the eternal mind. It is true that " man's attainment of knowledge is conditional on processes in time and on the fufilment of strictly natural functions." But even these functions, " which would be those of a natural or animal life if they were not organic to the end consisting in knowledge, just because they are so organic, are not in their full reality natural functions, though the purposes of detailed investigation of them — perhaps the purpose of improving man's estate — may be best served by so treating them. For one who could comprehend the whole state of the case, even a digestion that served to nourish a brain, which was in turn organic to know- Ill SPIEITUAL PEINCIPLE IN NATURE 251 ledge, would be essentially different from digestion in an animal incapable of knowledge." ^ This seems to me a bold assertion. Wliy should it be ' essentially ' different ? No doubt brain affects digestion. But why ' essentially ' ? By this Green means a great deal, as we see from what follows. For we may say, he holds, that " in strict truth the man who knows, so far from being an animal altogether, is not an animal at all or even in part." He has only to add that he is not an eternal consciousness at all : and the fasciculus of contradictions would be symmetrical and complete. However, we need not pause on this hard saying. At any rate, in Green's view the " inquiry as to what man in himself is, must refer . . . to the character which he has as consciously dis- tinguishing himself from all that happens to him."^ "We are entitled to say," he tells us, "that in him- self, i.e. in respect of that principle through which he is at once a self and distinguishes himself as such, he exerts a free activity — an activity which is not in time, not a link in the chain of natural becoming, which has no antecedents other than itself but is self- originated." Or — which Green apparently regards as a convertible statement — is originated " by the action of an eternal consciousness, which uses them ['the processes of brain and nerve and tissue, all the func- tions of life and sense '1 as its organs and reproduces itself through them." ^ Now, in order to examine closely this attribution of ' Op. cit. § 79, p. 84. •^ Op. cit. § 80, p. 85. » Op. cit. § 82, p. 86. 2o2 THK Ml-yrATHYSICS OF T. II. (IREEN i,kct. ' freedom ' to human intelligence, let us grant — what 1 have above strenuously denied — that an act of human knowledge is not an event in time, and also that the relation of knowledge to its object is entirely unlike any other relation within the known world, and in- capable of being developed out of any concatenation of such relations. But it must remain true that in human minds knowledge is partial and changing : some know some things, others other things, and the knowledge that any one has at one time of his life is different from the knowledge he has at other times. I suppose Green does not intend to deny that of these differences and variations there is a natural explana- tion to be given, since he says " why any detail of the world should be what it is we can explain by reference to other details which determine it'V ''^-^d surely the exact degree of finiteness, the limitations, the particularity, in the knowledge of any finite mind at any particular time is a ' detail of the world,' and its variations must come into and form part of the process of cosmic change. If so, we must conclude that the ' freedom ' of intelligence has no particular or practical application. For if intelligence is 'free,' still the particularity of the intelligence of any particular mind must be as much caused as anything else in the world. That it knows at all may defy a natural explanation : but that it knows this or that, so much and no more, must be com- pletely so explicable. We may remember that the same statement was found to apply to Green's 1 [Op. cit. § 82, p. 86.] Ill SPIEITUAL PEINCIPLE IN NATUEE 253 'freedom' in the ethical sense. ^ But there is a want of complete correspondence between the two which it is desirable to note. Green, it will be seen, treats the notion of Freedom under two heads in chap. iii. of Book I. and in chap. i. of Book 11. — as the " freedom of man as intelligence " and as the " freedom of the will." It is fundament- ally important, in understanding Green's ' freedom,' to keep this double use of the notion in mind. For in his view 'free' simply means 'not natural,' not explicable by natural causality ; and that, in either case, means only that in human intelligence and human volition alike — so far as the two are dis- tinguishable — a self- distinguishing, self- objectifying consciousness is necessary. At the same time, there is a considerable difference between Green's treatment of the two cases. For in considering the "freedom of man as intelligence " I do not find that he gives any explanation of — or even takes any notice of — the fact of Error. He has, as we saw, previously given an account of error as conceiving a phenomenon in relations inconsistent with the single unalterable system. But he does not consider how man's ' free intelligence ' can do this. The self-distinguishing con- sciousness in chap. iii. to which ' free causality ' is attributed is always (so far as I can see) conceived to exercise its freedom so as to attain or produce knowledge — real knowledge — not illusion and error. But in the case of the 'free will' the distinction between 'virtuous' and 'vicious' choice — choice of ' Of. the author's Ethics of Green, SiKiicer, and MartineaAt, pp. 16 ff. 25-1 TlIK METAl'llVSirs OK T. IT. (ITJEEN i.Kvr. true gooil and choice of mere plcasuit' — appears in the forefront of tlie discussion. Still, ctjually in both cases, the deflection from truth and right on the part of the self-distinguishin;^ consciousness is inexplicahle. Why does the eternal spirit, reproducing itself so many million times in connexion with so many organisms, produce so much error and so much vice ? I find no serious attempt to answer this in Green. But he seems practically to admit in both cases that the particularity of the in- dividual's cocjnition or volition — the difference between A, who discovers truth, and B, who produces chimeras, between A, who makes a right choice and seeks his true good, and B, who makes a wrong choice and seeks his self-satisfaction in pleasures that do not satisfy — these particularities and differences are to be explained by differences in the previous histories of A and B. For he says — as already quoted — " why any detail of the world is what it is, we can explain by reference to other details which determine it " : ^ and the ignorance and errors of some, the particularities and limitations of the knowledge of others, are certainlv ' details ' of the world. So ao^ain " the form in which the self or ego at any time presents the highest good to itself — and it is on this presentation that its conduct depends — is due to the past history of its inner life . . . The particular modes in which I now feel, desire, and think, arise out of the modes in which I have previously done so." " He lays stress, indeed, on the fact that in all cases a self-distinguishing I Cf. above, p. 252. ^ Op. cil. § 101, p. 105. Ill SPIKITUAL PEINCIPLE IN NATURE 255 consciousness has been operative throughout this his- tory : but as this is a similar element in all the different cases — for it is the eternal consciousness reproducing itself in all — it cannot possibly furnish an explanation of the differences. Summary of Green's Metaphysical View (1) Everything that is or can be an object of thought is constituted by relations : — relations of its elements to each other, of itself to other objects, and to the whole of nature ; it also involves a self- conscious, self-distinguishing thinker or subject, apart from which any object is inconceivable. (2) If it be said *' but besides relations there must be feelings related" — the answer is that feelings without relations are nothing for us as thinking beings. (3) Relations are results of the activity of thought, combining and unifying : thus the world of each man's experience is in some sense produced by the activity of each intelligent self (4) But the distinction between truth and seeming, between impressions that correspond to objective reality and mere subjective illusion, involves the conception of a single unalterable system of relations — for error and illusion lie merely in conceiving relations wrongly, i.e. otherwise than they are in this single system. (5) This single unalterable system of relations must therefore be referred to a universal self or '256 THK AIKTAPirVSirS OF T. H. GREEN lt. hi ego : and its juirtial apprehension l>y liunian minds must be explained by supposing this universal ego to [*' realise or "] reproduce itself in individual human beings gradually, making the func^tion of au animal organism its vehicle. (6) The acts of knowledge of human beings — acts in which the knowledge eternally possessed by the universal ego is reproduced in the human mind — are " out of time," though the process of attaining know- ledge is, no doubt, a process carried on in time. APPENDIX TO THE LECTURES ON THE METAPHYSICS OF GREEN [The chief part of a lecture entitled "The Philosophy of T. H. Green," which Professor Sidgwick delivered at Oxford shortly before his death, is here reprinted from Mind, N.S. vol. x. 1901, pp. 18 ff. The lecture was never revised, but — as it was written some time after those that now precede it — it is inserted as supplementary to them.] I can now, I hope, state both briefly and clearly my view of Green's Metaphysical System. First, it is a species of Mentalism. Nattire, or the world of space and time, is con- ceived as a single, unalterable, all-inclusive system of relations : and these relations are thought-relations ; they result from the activity of thought. So again, so far as this conception of Nature goes, the system is clearly the species I have called Idealism. If Nature is essentially a system of thought-relations, Reality is — so far — Thought. And if Thought was conceived as simply fur sich bestehend ^ — as Green had conceived it some years before — the whole system might have been purely Idealistic. Thought would then not only have made Nature, but have completed itself — its system of relating and related notions — in Spirit : so that the Universe of Reality would have been truly thought as Thought itself. But this is not Green's view in the Prolegomena : on the contrary, it is a view that he decidedly and emphatically excludes. The single all-inclusive system of thought-relations which constitutes nature, " implies something other than itself, as a condition of its being what it is." - It presupposes the activity of a thinking being, a " self-distinguishing, self-objecti- fying, unifying, combining consciousness " whose synthetic activity is the source of the relations by which the knowable ^ Cf. Works, vol. ii. p. 11 note. '*' Prolegomena, § 52 f. 257 S 258 TlIK MI-.TArilVSU'S OK T. II. CKKKN world is iiiiilioil ; aiiil wo ;iic odlilk-il to say of this entity, that the rolalioiis wliiili ri'sult from its synthetic action are not pre^he(l tliom — really thinkable? ami docs Groon rejilly succeed in thinking it ? I am compelled to answer both questions in the negative, but 1 shall devote my own dis- cussion chieHy to the second (juestion. Lot us first tiikti Green's positive account of Spirit, and ask, j>oint by point, whether wf can definitely think the qualities or functions he attributes to it, without, in so thinking, predicating of it some of the relations which, according to Green, result from its combining and unifying activity, and are therefore not properly predicable of it. First, he conceives it as one and many : one Divine Mind and many reproductions of it ; here we have relations of number. Secondly, the human spirit is identical with the Divine : — the latter is stxid to be a "spirit which we ourselves are" : yet again it is a " reproduction " of it and a reproduction is different from the original. Here we have a peculiar and difficult combination of the relations of identity and difference. Again, a Spirit is a " self-distinguishing " consciousness : that means, I suppose, that it attributes to itself unity, identity, difference from nature and, I suppose, from other spirits. But again it is a " self-objectifying " consciousness : that is, it con- ceives itself as an object : and therefore in a relation of similarity with nature, so far as both spirit and nature must be thought as haWng whatever attributes are connoted by the word " object." Finally, it is a "unifying" and "combining" consciousness : but by each of these terms its function is conceived in a relation of similarity to processes that we conceive as occurring in Nature ; Nature is continually presenting to us combinations and unifications, as well as separations and divnsions. In short, taking Green's descriptive terms, and endeavouring to think by means of them, we find that we are inevitably conceiving Spirit as conditioned or determined by the very same relations that we use in determining phenomena. Tuin now to the negative characterisation that he gives of Spirit, to emphasise and impre.ss on us its non-naturalness. It is, he says, not in time, not in space, not a substance, not a cause. But can he really think it thus 1 Let us see. First, the Spirit is " not in Time." If so, we are to under- APPENDIX 261 stand not merely that it does not change but that it does not perdure ; since changing and perduring are equally time- determinations. Hence when Green speaks of the Divine Spirit as " eternal," we must understand him to intend to mean not " everlasting," but merely the same as when he speaks of it as " not in time." But can we conceive this to be his meaning when he speaks of it as " a consciousness for which the relations of fact that form the object of our gradually attained knowledge already and eternally exist " ; or when he speaks of the " best state of man as already present to a divine Consciousness " 1 Must we not think of the divine Consciousness as " in time " if we think of it as " already " such and such. So again, when speaking of the problem suggested by the constant spectacle of unfulfilled human promise, he says " we may content ourselves with saying that the personal self-conscious being, which comes from God, is for ever continued in God " : — surely here God is conceived as eternal in the sense of abiding " for ever," Again, it is because the divine mind reproduces itself in the human soul that that soul is said to have a " spiritual " demand for an "abiding satisfaction of an abiding self" ; but how could this be legitimately inferred unless the Divine Mind itself were con- ceived as abiding and perduring through Time ? But if " in time," why not a substance, since substance is for Green the permanent correlate of change 1 and can we avoid thinking of the Eternal Mind as the permanent correlate of the processes of change and development essential to finite minds ? Finally, can we conceive the Eternal Consciousness — following Green's thought — as not a cause 1 He tells us that it is a " source " of the relations which constitute Nature ; that they " result from " its combining and unifying action ; that it "makes the animal organism its vehicle" ; that it "is operative" throughout the succession of events which constitute the growth of the individual mind; that it "acts on the sentient life of the soul " and " uses it " as its organ. Are not these all terms implying causality 1 And yet he says — arguing against Kant — that " causation has no meaning except as an unalterable con- nexion between changes in the world of our experience." Green ultimately sees the inconsistency, — though I think he carries the exposition of the Metaphysics of Knowledge much 2(;-J rHK MKrAl'HVSICS (M- T. 11. CKKKN Uio lar witht'Ut hinliiii; ;it it. Mul i will not ili<;ress on this point. Lot us nithor try to undcrst.vnd the explan.-ition that he nltimat^dy gives. It is, 1 think, tiic most ilitlicult passai^c in the Prolf'ijiuufiKt : — "When we transfer the tciiu 'cause' from the relation between t)iie thinij and another within the determined worhl to the relation between that worlil and the agent implied in its existence, wo must understand that there is no separate ])articu- larity in the agent, on the one side, and the determined woild as a whole on the other. . . . The agent must act absolutely from itself in the action through which that world is — not as does everything within the world, under determination by some- thing else. The world has no character but that given it by this action ; the agent no character but that which it gives itself in this action." ^ It should be added that the " action," in the same passage, is stated to be " that inner determination of all contained in the manifold world by mutual relation, which is due to the action of the unifying principle." It appears, then, that Green idtimately attributes to God Causality : but endeavours to establish an essential difference between Divine and Natural Causality : viz. that the Eternal Consciousness, as unifying ])rinciple, has " no separate particu- larity " apart from the manifold world, " no character but that which it gives itself in ' its unifying ' action " — although it " must act absolutely from itself in the action through which the world is." Now I cannot myself conceive these character- istics united : I cannot conceive anything " acting absolutely from itself " and yet having " no character but that which it gives itself in this action." But, waiving this objection now, I admit that this negation of " character other than that which it gives itself in the action " differentiates the Causality of the Divine Mind profoundly from Natural Causality : but I think it does this at great cost to the system as a whole. For, first, if God is thus reduced to a mere unifying principle, having no character except that which it gives itself in synthesi.sing the manifold of nature, I do not see how the conception can be made to include the content which the ethical ' Prolegomena, Metaphysics of KnotvleJgc, p. 81. APPENDIX 263 part of Green's doctrine requires. It is because there is a Divine Consciousness realising or reproducing itself in man that the true good of man is argued to be not Pleasure, but Virtue or Perfection, and Perfection is held to consist in the realisation of capabilities already realised in the Divine Existence : briefly put, man's true good is development in the direction of becoming liker to God. But this whole conception implies that God has what Mr. Balfour calls a ' Preferential Will ' in relation to human life and action ; and that this Will is realised in man's choice of Virtue in a sense in which it is not realised in his choice of sensual pleasure. Well, I do not see how this concep- tion can be maintained if God is also conceived as having no character except that self-given in unifying the manifold of nature : for this unification is surely equally effected in the lives of sinners and in the lives of saints, as both are equally capable of being scientifically known. In short, this conception of the relation of God to the world seems to me to constitute a gulf between Green's Metaphysics and his Ethics which cannot be bridged over. If, on the other hand, we leave Ethics aside, and confine ourselves to the conception of the Divine Spirit regarded as belonging to the Metaphysics of Knoivledge, it seems to me that this eternal consciousness, characterless apart from its unifying action, is a rather insignificant entity : whose existence is not only difficult to establish logically, but not much worth estab- lishing. The conception, indeed, of the world as a systematic whole, having unity and order through the complex relations of its parts, as well as infinite plurality and diversity ; and the conception of the progress of knowledge as consisting in the continual discovery of order, system, and unity in what at first presents itself as an almost chaotic diversity — these are con- ceptions of the highest value. But when they are grasped, what is the further gain to knowledge in referring the unity and system to a unifying principle as its source, if that principle is to have no other character except what it gives itself in its unifying action. Is there any hope that such a conception can in any way help us to grasp the unity, the system of relations, more fully and truly ? Nay, must not the notion of a Divine Minil if reduced so far, inevitably dwindle still further, and •J04 rilK MKlArilYSU'S OF T. ]{. (IKKKN revwil itjit'lf Jis meroly u.siioss, Ijut in Lho o:ise also of " tho world-conscioiisnoss of which ouis is a liiuiti'd mmlo. " Hut if this be so, I do not see how (Iret'ii is justified — or thinks liinjsclf justified — in inaUinn the thouglit olonient so proniinont, ami the feeling element so subordinate in his account of Nature : or in speaking of Nature as a system of relations, instead of related feelings ; or in resolving — as we saw — the particularity of a feeling entirely into relations. And finally, if "mutual independence of thought and feeling has no place in the world-consciousness," difiicult questions arise to which Green suggests no answ^er. For instance, if any feeling is attributed to the world-consciousness, must not all feeling in the world be so attributed ? or how are we to distinguish 1 Does God then feel the pleasure and the pain of the whole animal kingdom ? And if so, is not the ground cut from under the anti-hedonistic positions of Green's Ethics 1 But I perceive that this topic will introduce so great a wave of discourse — as Plato says — that I must reluctantly abandon it, and apologise for the extent to which I have already tried your patience. THE PHILOSOPHY OF MR. HERBERT SPENCER INTRODUCTORY : KANTIAN INFLUENCE IN ENGLAND AGNOSTICISM AND RELATIVISM In the lectures on Green I have endeavoured to characterise and to criticise elements of actual philosophical thought derived from Kant's Transcendental Philosophy vieived on its constructive side : i.e. viewed as an attempt to exhibit systematically those factors of our conception and knowledge of the empirical world which are cognisable a priori, either as forms of sensibility or as forms of intellectual synthesis, otherwise termed fundamental concepts or categories. But this is only one side or aspect either of the Kantian system itself or of its influence on English thought ; nor is it the side or aspect which was at first clearly the most prominent. It is true that, as I say in my Outlines of the History of Ethics,^ the thinker Avho in the first third of the nineteenth century was commonly regarded as the representative of German tendencies in philosophy — namely, Coleridge — transmitted the influence ^ of Kant as apprehended through the medium of post- Kantian thought and especially the thought of Schelliug. Thus, as I have said {Outlines, I.e.), " the Kant partially assimilated by Coleridge was a Kant who could not be believed ' to have nieant more by his Noumenon or Thing in itself than his mere words express ' ; ^ who, in fact, must Ije believed to have attained, 1 P. 271. - Cf. J. S. Mill's essay(1840), "Geriuauo-Coleridgian doctriue," " Coleridge and the Germans." •^ Coleridge, BiogiMphia LUeraria, vol. i. pp. 145 f. 2(>7 2GS rHlI.(^i=^0?nV OF ITKnr.KRT SPKNOETl through his pnictionl oonvirtions of duly uiul freedom, that spooulativo oomprohcnsion of the cssentijd spirituality of human nature which his language appeared to repudiate. Thus viewed on its motaphysical side, the Cierman intliicncc ohsciu'cly com- municated to the English mind through (^)lcridgc was rather post-Kantian than Kantian, though the same cannot be said of it3 strictly ethical side.'' ' Hut the Kantism transmitted through Coleriilge was but very i)artially assimilated. And in the more important examples of Kantian influence in the second third, or rather more, of the century, we find Kant's doctrine assimilated more on its nega- tive and destructive than on its positive side. The two main points of the doctrine so assimilated may be characterised respec- tively as Agnosticism, or the unknowableness of the Absolute or Unconditioned ; and Relativism, that is, the ' relativity of human knowledge.' The Agnosticism, however, in the case of the two leading examples of this influence — Sir W. Hamilton and Dean Mansel — was combined with theological orthodoxy ; and the Relativism is somehow reconciled with Natural Dualism. Before I pass to examine the form which each of these two doctrines assumes in the philosophy of Mr. Spencer, I will explain them briefly in the form in which they are presented by Hamilton — since the influence of Kant comes to Spencer entirely through Hamilton and his disciple Mansel, and not directly. I begin with Hamilton's ' Philosophy of the Condi- tioned ' as Mansel calls it. Briefly the 'Law of the Conditioned' is : " All positive thought lies between two extremes, neither of which we can conceive as possible ; and yet, as they are mutual contradictories, we must recognise the one or the other as necessary." - Or, as Hamilton more fully explains, taking as an illustration our quantitative notions of space and time, all that we positively conceive lies between two poles [the maximal and the minimal], and at either pole — where our thought comes upon the unconditioned — we find two pairs of contradictory inconceivables, one of which must be true, though we can con- ceive neither. So again, we cannot conceive the will to be free, as that w^ould involve an uncaused event, an absolute commence- ^ I.e. p. 277. - [Hamilton's edition of Reid's Works, p. 911.] KANTIAN INFLUENCE IN ENGLAND 269 ment of existence ; at the same time, we cannot conceive an infinite regress from eflfect to cause. Here we have obviously a reproduction of the three first of Kant's cosmological antinomies ; but it is a reproduction with important modifications. For Kant does not argue that infinite time or infinite space is inconceivable. On the contrary, he makes in the Esthetic the remarkable statement that space is presented as an * infinite given magnitude ' {unendliche gegebene Grijsse) ; and in arguing the thesis of the first antinomy it is not infinite time but infinite past time which he argues to be incon- ceivable : for " the infinity of a series consists just in this, that the series can never be completed in a successive synthesis," hence we cannot conceive an "infinite series of states to have passed away in the world." ^ Similarly, Kant argues — ingeniously — that we must think the world limited in space, because " in order to think the world which fills all space as a whole, we must suppose the successive synthesis of the parts of an infinite world to be completed." Finally, Kant never questions the infinite divisi- bility of Space ; it is infinitely divisible Substance which seems to him an unthinkable notion : because if we suppose that any composite substance is not ultimately resolvable into simple parts, " then, if we think all composition away, no composite part will be left ; and as by hypothesis there is no simple part, nothing at all will remain." - The difference, it will be said, is that in the case of Substance — as Kant with those he is arguing against assumes — the simple is necessarily thought as prior to the composite ; but we cannot similarly conceive the parts of Space as prior to the one Space of which they are parts. So again Kant has no difficulty in conceiving Infinity as an attribute of the Divine Being ; indeed he thinks it an indispensable notion ; what he questions is tlie possibility of giving a speculative proof of the existence of such a being. Hamilton's Philosophy of the Conditioned, therefore, diverges widely from Kant, in respect of the notion of the Infinite. And here I agree with Kant : I find no difficulty in conceiving Infinite Time or Infinite Space as such ; but there certainly is a difficulty in conceiving a completed Infinite and therefore a ^ [Watson's Selections, \>\). 158 f. ] - {Op. cit. p. 160.] •J70 rHii.()S(MMiv or iiKi;iu-:i:r si'Hnckk p.-st Intiiiilo. ll is partly Hue that, as Hamilton s;iys, the notion of Inliuito Qiiaiitity is negativo ; that is, when we try to conceive Infinite Magnitude positively otherwise than negatively, we can only conceive it as " greater than any assignable magin- tude" ; and it is with that meaning that we employ the notion in mathematical reasoning. The notion of Infinite, so far as it me^ns more than this — and it certainly seems to mean more — is no doubt negative — negative of limit : but that does not seem to me to justify the assertion that the Infinite is inconceivable. But there is another fundamental diflercnce between Hamil- ton's and Ivant's method of dealing with the dilemmas which Kant calls antinomies. Their solutions are entirely different. Hamilton's conclusion is agnostic. " One or other of two alter- natives is true, but we cannot say which " (except in the case of Free Will, when he follows Kant in deciding for Freedom on moral grounds).^ But Kant's critical conclusion is a solution of the difficulty by means of the distinction between phenomena and things per se. For example as regards Time : once grasp that Time is not a form of real existence but only of human percej)- tion, and the difficulty of an infinite Past vanishes : the series of past Time is not a series that really has existed, but only one that we must think. The true critical conclusion is that in system- atising experience we may carry back the regress of Time as far as we like ; and similarly of Space. But Hamilton is too much of a Natural Kealist to accept the transcendental Ideality of Time and Space. With regard to Space, he expressly main- tains that "we at once must and do think Space as a necessary notion, and do perceive the extended in Space as an actual fact " : and if he makes no corresponding assertion with regard to Time, I think it is only because it seems superfluous. This leads me to the ' Relativity of Human Knowledge.' For, as I have said, it is characteristic of Hamilton's Meta- physic to endeavour to combine — on the question on which Natural Dualism, Materialism, and Mentalism diverge — or to effect a compromise between, the position of Natural Dualism and the position of Kant as defined in the Prolegomena ; i.e. Kant- ism, taken in its Realistic attitude, its attitude of opposition to "all Idealism." ^ [Cf. Metaphysics, ii. pp. 410 ft., 542 f.] KANTIAN INFLUENCE IN ENGLAND 271 On the one hand, Hamilton, developing the old distinction of Primary and Secondary Qualities of matter into a threefold classification of Primary, Secundo-primary, and Secondary, gives as the characteristic of Primary Qualities — of which the most fundamental are ' Trinal extension ' and ' Ultimate incompres- sibility ' — that we " apprehend them as they are in bodies," " as modes of the non-ego, . . . clearly conceive how they must exist in bodies, in knowing what they are objectively in themselves " ; while the Secondary Qualities — colour, sound, flavour, etc. — are apprehended "as they are in us," "as modes of the ego," as " subjective cognitions " or " sensations proper," and " not in propriety, qualities of body at all." This is the old distinction of Locke. But Hamilton's development, as I said, includes also an intermediate kind of qualities, " Secundo-primary " — such as the various modes of gravity, cohesion, and the like, known as heavy, light, hard, soft, rigid, flexible, rough, smooth, etc. — which also fall under the ' category of Resistance or Pressure,' and liave the metaphysical characteristics of both the other classes. That is, we apprehend them both as they are in bodies and as they are in us: both "immediately in themselves" and " mediately in their effects on us " ; " in their Primary or objec- tive phase they manifest themselves as degrees of resistance opposed to our locomotive energy," and are so far quasi-primary : but this '■ objective element " is always accompanied by a secondary quality or affection of our sentient organism. Well, all this — developed at great length by Hamilton ' — is or appears to be ' Natural Dualism ' pure and simple. If we had only this part of his doctrine before us we should never dream of attributing to him the view explicitly stated by Kant {Prolegomena) that " the qualities of body which are called primary " — no less than the secondary — " belong not to the things in themselves but to their phenomena," and " have no proper existence outside our repre- sentation." ^ Yet elsewhere Hamilton's language seems thoroughly Kantian. "Our whole knowledge of mind and matter is relative; ... of things in themselves, be they external, be they internal, we know nothing or know them only as incognisable " ; . . . " all that we ^ [Cf. Dissertations iu his edition of Reid's Works, pp. 845 ff.] ^ [Cf. Prolegomena, ilahafly's trans, p. 55.] 272 rHII.OSOlMIY OF IIKRBEKT SPENCER know is pluMioiiioiial, unci plieiioinenul of tho iiiiknowii." ' At an early stai;t> of his lectures on Metaphysics he states and explains "the great axiom that all human knowledgo is only of tho reliitivo or phenomenal." lie explains that " Matter, so far as it is a name for something known," is "a conunon name for a cerUiin series or aggregate of appearances or phenomena mani- fested in coexistence," which by the constitution " of our natuie we are compelled to think conjoined in and by something " ; . . . but this something absolutely in itself, i.f. considered "apart from its phenomena, is to us as zero." Similarly " in so far as mind is the common name for the states of knowing, willing, feeling, desiring ... it is only the name for a certain series of connected phenomena." But "so far as it denotes the subject in which the phenomena of knowing, willing, etc. inhere, it expresses what in itself or in its absolute existence is unknown. . . . Our whole knowledge of mind and matter is thus only relative : of existence absolutely and in itself we know nothing." - It is somewhat surprising to find these two lines of thought so vigorously pursued and expressed by the same thinker ; and certainly when one now reads the lectures and articles of the most distinguished academic teacher of Philosophy in Great Britain in the first half of the century, it does seem that the two streams of metaphysical thought which meet in him — the tradi- tional Scottish Philosophy of Common Sense, and Kantism — do not properly blend. The explanation is that ' Relativity of Knowledge ' is a complex and ambiguous term : there are various significations which it may bear, and which it does bear for Hamilton : some elements of its meaning are quite compatible with the Natural Dualism to which his doctrine of Primary Qualities belongs, while other elements are not ; and Hamilton's defect lies in not clearly distinguishing these different elements. 1. The assertion that knowledge is relative may mean no more than that it is a relation between the knower and the known : and therefore between two things distinct in existence. This meaning is, of course, quite compatible with knowing qualities of matter as they are in bodies. In fact knowledge, ' [JJiscv^sions on Fhilosojjhy, p. 639.] - [Lectures on Metaphy ides, ii. pp. 136-138.] KANTIAN INFLUENCE IN ENGLAND 273 true knowledge, as we commonly conceive it, is a relation which does not modify the qualities of the known. In this sense, however, the assertion that we can only know the relative is insignificant : for it simply means that we cannot know anything without knowing it ; and similarly the assertion that we cannot know the Absolute — if * absolute ' is understood as meaning ' out of the relation.' For this proposition again simply means that we cannot know anything without knowing : and this would be equally true if we had suddenly revealed to us the most perfect knowledge of God and the Universe as they were independently of our knowledge. 2. A more important meaning, but still perfectly compatible Avith Hamilton's theory of Primary Qualities, is that which refers to relations among objects known, not to the relation between knowing subject and known object. It is undoubtedly true, and epistemologically important, that we never cognise, nor can we really conceive ourselves cognising, an object that is not in rela- tion to other objects : especially in perceiving any part of matter, or the non-ego, we perceive it in spatial relations to other parts ; and again, in judging that it possesses such and such qualities, we attribute to it implicitly relations of resemblance to other things having the same qualities, and relations of difference from other things having different qualities. Relativity in this sense is of course quite consistent with our knowing — objectively, and as they are in the real things — the size, shape, divisibility, incom- pressibility, density, rarity, situation, and change of situation of matter. 3. But there is a third meaning. Though we cannot, speaking generally, resolve ' quality ' into ' relation,' yet many qualities are found by reflection to be essentially relational ; and this is the case with the Primary Qualities of Matter. They are all, as Hamilton says, " evolved from the two universal conditions of occupying space and being contained in space." But reflection shows each of these to be relational : for e.g. what does " occu- pying space " mean except that if another portion of matter moves in the direction of the space said to be occupied, it will at a certain point of its course find an obstacle to its moving further. 'Occupation,' in fact, in its physical as well as its general meaning, implies a relation, actual or potential, T 274 rHll.(^S(iriIV OK HKKHKKT STENrKR to somolhiug t'lso that attoiniils to — or iniglit siiuihuly — occupy. l>ut (4) the }yltifums tlms iniplioiialities are — in the case of Primary Qualities arlnalli/ percdiwl — in part rehitions to the percipient organism. And, in Hamilton's view, my ' immediate knowledge ' of matter must be knowledge of matter actually in contact with my organism, and 80 in definite spatial relation with it. This he expressly says in one passage. "The Primary are the qualities of body in relation to our organism as a body simply, — the Secundo-primary arc the qualities of body in relation to our organism as a propelling, resisting, cohesive body," etc.^ 5. Finally, there is the Relativity of Qualities to Substances and Substances to Qualities. I think that all these different meanings were more or less in Hamilton's view when he affirmed Relativity of Knowledge ; but not adequately distinguished from the meaning which the phrase ordinarily carries with it in Philosophy — a meaning incompatible with Natural Dualism or with his view of Primary Qualities, i.e. the meaning which involves denial of our know- ledge of things as they are independently of our cognition. ^ [Reid's JForks, Dissertations, p. 857.] LECTURE I METAPHYSICAL DOCTRINES I PROPOSE to give a critical exposition of Mr. Spencer's metaphysical and epistemological doctrines — his view of the Universe, so far as known and knowable, and his theory of the criterion or method for distinguish- ing truth from error. I ought to say that he does not himself use either of these technical terms to denote any part of his doctrine. He does not seem to have heard of ' Epistemology,' and he employs the term ' Metaphysician ' exclusively to designate a class of thinkers who have followed an o erroneous method to untenable conclusions. Still he has a very definite epistemology, which he regards as fundamentally important. And he has a metaphysical system — a systematic view of the nature and relations of finite minds to the material world, and to the Primal Being or Ultimate Ground of Being — of the coherence of which he is strongly convinced.^ ' This ' system ' indeed is nowhere systematically expounded : the exposi- tion of it is to be found only in fragments scattered through the three volumes of his First Principles and Priiiciplcs of Psychology — chiefly in Part 1. and the earlier chapters of Part II. of the former, and in Part VII. of the latter ; also in chap. x. at the ond of vol. i. of the Psychology, and in the closing paragraphs of the First Principles. o7r. •j:6 rHiiA»soi'iiv (U' iii:kiU':i;t stknckk lkct. I take tirst tluit part ot" Mr. Spencer's philosophy in whieli the intluencc of Kant through Hamilton and Mansel is most manifest — his doctrine of 'the Un- knowable.'' His avowed object, in this part of his work, is to reconcile the ' antagonism between Religion and Science,' which is, he tells us, " of all antagonisms of belief, the oldest, the widest, the most profound and the most important." With this aim he proposes to " contemplate the two sides of this great controversy," preserving an " impartial attitude." Accordingly, in chap, ii., he gives us a discussion of 'Ultimate Reliorious Ideas.' But what are ' relimous ideas ' ? A o o little discussion of religion would have been in place in this part of Mr. Spencer's treatise. He appears to assume that inquiries concerning " the origin and hidden nature of surrounding things " are as such religious. But though the answers to such questions may be religious — if they affirm that the existence of surrounding things originated in and is sustained by the Will of a Being to whom worship is due — it does not appear that the questions as such are religious any more than scientific or philosophical. When Thales taught that " water is the original source of all things," w^hen Epicurus taught that earth and stars were formed by the collisions and combinations of primordial atoms, they were surely speculating about the ' origin and nature of surrounding things,' but it would be absurd to call their doctrines religious. And we remember that in Kant's system, the question whether the world has had a beginning is classed as ' i.e. First Principles, Part I. I METAPHYSICAL DOCTEINES 277 primarily a cosmological, not a theological question. It especially concerns us to note this, because it is by arguments — to an important extent — derived from Kant through Hamilton as well as from the line of English Empiricism that Mr. Spencer proves his agnostic conclusion that "no tenable hypothesis can be formed as to the origin or nature of the Universe regarded as a whole." He takes the ' origin ' first. There are, he says, three verbally intelligible suppositions : we may either assert that the Universe is self-existent, or self-created, or created by an external agency.^ Now I submit that it is only the third of these hypotheses that can be called ' religious,' and even this only if the external agency is a Divine Mind. The general question, therefore, is philosophical, not theological : accordingly, in the present discussion I shall treat Mr. Spencer's agnostic conclusion as philo- sophical agnosticism, reserving the specially theo- logical or religious aspect of it for consideration later. His conclusion is that none of these verbally intelligible suppositions is really conceivable. As regards the first — ' Self-existence ' can only mean existence with- out a beginning, and we cannot conceive existence without a beginning : for we cannot conceive infinite past time. This, in Mr. Spencer's view, appears to be simply because " unlimited duration is incon- ceivable " ^ — an argument whose apparent force seems to me due to a want of distinction between imagina- ' First Principles, g 11, ji. 30. [Quotations tliroughout from the 3r(l (stereotyjied) edition.] - Op. cit. § 11, p. 36. J •J7S riUlAtsornV of IlKUr.KKT Sl'KNCEIt i.K.rr. tioii and c'oiu'e}>ti()ii — it is not based on the certainly more forcible argument of Kant that infinite past time involves a contradiction, because it is the essence of an infinite series that it should not be completed.' The second hypothesis — self-creatiou — need not de- tain us long. Prima facie, the notion involves a contradiction, and I know no thinker of importance who has maintained it. But, for a reason that will subsequently appear, it is worth while to note ]\Ir. Spencer's method of disposing of it. lie says that "really to conceive self - creation, is to conceive potential existence passing into actual by some inherent necessity " : but we cannot do this, as " we cannot form any idea of a potential existence of the Universe, — as distinguished from its actual existence." For " if represented in thought at all, potential existence must be represented as an actual existence."'^ Noting this, let us pass to the third — " the commonly conceived or theistic hypothesis — creation by external agency." Here, however, it is at once obvious, Mr. Spencer holds, that, even if the hypothesis be accepted, the question is only pushed a step backward : we shall have to inquire into the origin of the existence of the external agency, and the alleged impossibility of conceiving infinite past time must apply equally to that. Besides this, Mr. Spencer urges that no analogy with a human artificer enables us to conceive the production of matter out of nothing : and even * Kant does Twt affirm, as Spencer seems to do, that infinite progress is an impossible notion : and I find no inconceivability in it, tliouj;li I admit it t .» be unimaginable. - {Op. cit. § 11, p. Z2.] I METAPHYSICAL DOCTEINES 279 if we could conceive this, there would remain the impossibility of conceiving space so produced. Mr. Spencer then turns to the question as to the nature of the world. " When we inquire," he says, " what is the meaning of the various effects produced upon our senses . . . impressions of sounds, of colours, of tastes, and of those various attributes which we ascribe to bodies, we are compelled to regard them as the effects of some cause . . . and we cannot carry out an inquiry concerning their causation without inevitably committing ourselves to the hypothesis of a First Cause." ^ But, since the common notion of * cause ' implies antecedence in time, the inquiry after a first cause of the effects on our consciousness, would seem to carry us back to the inquiry into the origin of the world. It seems, however, that Mr. Spencer means not merely something prior in time to the states of consciousness in question, or to the matter in motion which now apparently operates on our senses ; but something on the present existence of which this consciousness or this matter in motion depends for its existence. And this so-called First Cause, as there can be no cause limiting it, must be Infinite and Absolute. Here Mr. Spencer — largely with the aid of arguments derived from Hamilton and Mansel's Philosophy of the Conditioned — arrives at the con- clusion that while we cannot but assume a First Cause for the phenomena of our own consciousness, and " regard this first cause as Infinite and Absolute," ^ [Op. cit. % 12, pp. 36 f.] 280 rillLOSOl'llV OF lIKliHKUT SPENCER lkct. still the jirn^ument.s wliifh force on us these inferences are illusive, and the conclusions themselves conse- quently fallacious. 1 will not examine the argument in detail, but will only say that it seems to me confused and vitiated by the ambiguity of meaning of ' First Cause.' Let me briefly explain this. Mr. Spencer starts with a plurality of finite minds — his own and his readers' — each knowing immediately the transient facts of his own consciousness. He finds that he must suppose * some cause ' of these facts in the sense of some presently existing entity not himself, on which these facts depend for their existence. Then, he argues, this entity must either be the first cause or " have a cause behind it w4iich thus becomes the real cause of the efi'ect." But this can only mean that the entity in question must either be dependent on something else or independent : and if we grant that it is dependent on something else and so on, it does not follow that we shall ever come to a part of the whole universe which is not dependent for its existence on some other part ; for the parts may be mutually dependent (as the parts of an organism) and only the whole independent. But if we take this view^ the difficulty of conceiving the whole as Absolute and Infinite would seem to be avoided ; unless we assume that whatever exists in independence of anything else can have no necessary relation within itself. This, however, Mr. Spencer does assume : but surely it is an arbitrary assumption. He seems to think that a 'necessary' relation within the whole I METAPHYSICAL DOCTRINES 281 must be " inspired by something else" : but I find no such implication. What I conceive to exist necessarily 1 simply conceive as something that could not be otherwise. The idea involves no relation to anything outside. However, Mr. Spencer's conclusion is — as before stated — agnosticism both as regards the origin and the nature of the universe : and these being in his view the chief 'religious questions/ the only religious truth that Mr. Spencer can recognise is that there is a Power manifested to us by the universe, but that that Power is utterly inscrutable. But this conclusion he also arrives at by an exam- ination of * Ultimate Scientific Ideas,' which forms the latter half of his professedly impartial examina- tion of Science and Religion. (The consideration of this I defer for the present.) And this identical result of the two examinations he ofiers as the ' Supreme Verity' in which the reconciliation of Religion and Science is to be found, viz. that " The reality under- lying appearances is totally and for ever inconceivable to us . . . but we are obliged to regard every pheno- menon as the manifestation of an incomprehensible power, called Omnipresent from inability to assign its limits, though Omnipresence is unthinkable." ^ And this agnostic conclusion is proclaimed not only finally and decisively but solemnly and triumphantly. A ' high merit ' is attributed to Religion for having dimly discerned from the beginning, and continually insisted on this sublime verity : for the guardianship ' [Op. cit. § 27, pp. 98 f.] 282 riiTLOsoniv OF nKin^Kirr spknc^ki: i.kct. and (litVusion oi' wliicli Iluiiianity ever has been antl t'ver must be Religion's debtor. At the same time Mr. Speueer feels bound to point out that Religion herself has been ' partially irreligious ' through not being eonsistently and eompletely agnostic, but asserting that " the cause of all things possesses such and such attributes," ^ As to one part, then, of the funda- mental questions of Ontology or Metaphysics in the narrower sense as I have defined them — the nature of the Divine or Primal being, and its relation to finite minds and the material world — Mr. Spencer's answer is simple. All we know is a Power totally inscrutable and unknowable, whose existence is apprehended by a consciousness which though indestructible is per- fectly indefinite and undifferentiated. " Our consci- ousness of the unconditioned," he says, " is literally the unconditioned consciousness " or " raw material of thought to which in thinking we give definite forms." " This ' Supreme Verity ' is the residuum to which Theology is reduced in Mr. Spencer's philo- sophical laboratory. Let us now leave Theology aside, and turn to the chief metaphysical question or group of questions which remain — those presented by the nature and relation of finite human minds to the material world which is their common object. But even here we cannot leave on one side Mr. Spencer's ' Unknowable.' For, as I said, the existence of this is not only the Ultimate Verity of Religion : it is no less the Ulti- mate Verity for Science. When we try to understand 1 [Op. cU. § 28, p. 101.] 2 [Op. cU. § 26, p. 96.] I METAPHYSICAL DOCTEINES 283 Time, Space, Matter, Force, Consciousness — no less than when we try to understand God and His relation to the finite world, — we are equally driven to the conclusion that the " reality underlying appearances is and must be totally and for ever inconceivable by us." Hence, in dealing with the conception of (finite) minds and matter, no less than in dealing with the conception of God, " he repudiates as impos- sible the Philosophy which professes to formulate Being as distinguished from Appearance." For him Philosophy, like the sciences which it systematises, is concerned throughout with ' appearances ' or ' pheno- mena ' or ' manifestations. ' ^ When, however, with this general characterisation of the object of philosophical knowledge, we apply to it the distinctions of metaphysical schools already dis- cussed,^ it would seem at first sight that the positive element of Mr. Spencer's metaphysics must be indis- tinguishable from mentalism. For what do we mean in ordinary thought and discourse by * appearances ' as distinguished from being or reality ? We surely mean modes of consciousness, feelings, or thoughts, or combinations of the two produced in minds. And much of Mr. Spencer's language would support this view. The "manifestations of the Unknowable, con- sidered simply as such," are, he says, " divisible into two great classes called by some impressions and ' This is the aspect of Mr. Spencer's system which led me to call it Phenomenalism in respect of its positive content, and ' Agnosticism ' in respect of its fundamental litigation. - [Here, as in the note above. Professor Sidgwick is referring to unpublished lectures. But some account of his views concerning these distinctions will be found in Philosophy, its Scope and Jlelations, by consulting the index. J •JSt riIll,(>S(^riIV OF IIKIIHKKT SrKXCKK LF.iT. tilt-US. 'V\\c term scns(t(io)i, too, [beiiip^ also] com- monly used as the eijuivalent of impression, and state, of consciousness as sitrnifvinc^ either an impression or an idea." And though he finds objections to all these terms, it is not on account of their purely mentalistic import : it is because they carry with them implications which he would avoid at the out- set — implications of something impressing, of "a sensitive organism and something acting on it," of " something of which a state of consciousness is a state, and which is capable of different states." He therefore classes the manifestations as * vivid ' and * faint ' respectively, using terms that obviously denote purely mental facts, modes of consciousness.' The vivid manifestations are sensations, or sensa- tional feelings, or sense -percepts — either pains or sights, sounds, tastes and smells, or percepts of the tactual and muscular senses : the faint manifesta- tions are images or thoughts which are, he tells us, " imperfect and feeble repetitions " of the vivid, — what we call ' ideal ' sights and sounds, etc., in con- trast with real. He describes how the stream of vivid manifestations flows, in the conscious life of each of us, side by side with the stream of faint manifestations, sometimes one predominating, some- times the other. Both streams appear to be never broken, the members of each cohere with one another : but the " great body of the vivid current is abso- lutely unmodifiable by the faint, and the faint may ^ Op. cit. § 43, p. 143. Cf. Principles of Pnyclwlogu, Part VII. chaps, xvi. and xvii. I METAPHYSICAL DOCTEINES 285 become almost separate from the vivid." ^ The chief exceptions to this separation between the two cur- rents are (1) that the vivid manifestations which we distinguish as sensations of muscular tension have as their conditions of occurrence ideas of muscular action ; and (2) that the emotions, though vivid manifestations, are produced by and classed with faint manifestations. Well, is not all this pure unadulterated mentalism, so far as the knowable world goes : i.e. an elaborate and emphatic reduction of the material world as commonly conceived, into mental elements ? Sights, sounds, tastes, smells, sensations of pressure, muscular tension — these along with " intense pains " (and I suppose pleasures of sense) are described as making up the main stream of vivid manifestations : emotions though vivid being, as said, connected and classed with faint manifestations : and " all things known to us " being divisible into the two classes. These and the Unknowable Reality underlying them would seem to make up the universe, which might therefore be ex- pressed by the formula : vivid consciousness + faint consciousness + X. The system thus presented might be called Mentalistic Agnosticism or Agnostic Men- talism. Nor is this impression of the system at first altered when we find how Mr. Spencer applies this view to the interpretation of Natural Dualism. I will give it in his own words : — " What is the division equivalent to ? Obviously it corresponds to the division between object and subject. This profound- ' Cf. First Principles, § 43, p. 153. 286 rim.t^soriiv of hkim'.kkt spfa'cki; i.kct. est of distinctions among tlie niunifestations of tlio Uuknowftble, we rocognisc 1)}' grouping them into self juul not-sclj. Tliose faint manifcstatiouH, form- ing a continuous whole ditlVring from tlio other in the quantity, (quality, cohesion, antl conditions of existence of its parts, we call the r(j(> ; and these vivid manifestations, indissolubly bound together in relatively -immense masses, and having independent conditions of existence, we call the non-eyo. Or rather, more truly — each order of manifestations carries with it the irresistible implication of some power that manifests itself: and by the words ego and non-ego respectively, we mean the power that manifests itself in the faint forms, and the power that manifests itself in the vivid forms." ' This is the ultimate division the afiirmation of which, according to Mr. Spencer, is ' postulated ' as the " primordial proposition which Philosophy requires as a datum." ^ I confess that these summary equations, "vivid manifestations " = Non-ego, "faint manifestations " = Ego, are by no means " obvious to me." Indeed it would rather have seemed obvious that — in ordinary thought — sounds, tastes, smells, sensations of muscular tension, etc., belong to the Ego no less than thoughts and emotions. But no doubt they are more difficult to disentangle from our ordi- nary conception of the material world : and we are familiar, from Berkeley and others, with the view that our common notion of matter is made up of and exhaustively analysable into elements of this kind. ' Op. cit. § 44, ji. 154. * Op. cil. § 45, i.. 156. I METAPHYSICAL DOCTEINES 287 This passage, therefore, by itself would not have altered my view of Mr. Spencer's Mentalism ; though it might perhaps have led me to doubt the rigour of his Agnosticism. For the last sentence seems to in- terpret our conceptions of Ego and Non-ego as imply- ing not merely a duality in the manifestations of the Unknowable Power but also a duality in the Power itself. And as he immediately goes on to say that these conceptions "have for their explanation" an " ultimate law of thought that is beyond appeal " he seems to acquiesce in this dualism. But surely if, as we were before told, the deepest verity both of science and of religion is given by an indefinite con- sciousness of an utterly unknowable reality, it can- not also be right to have a definite conception of it as two powers, manifesting themselves respectively in vivid and faint consciousness. And indeed — in spite of the " law of thought that is beyond appeal " — this dualism is expressly repudi- ated by Mr. Spencer in a later passage : " The true conclusion implied throughout the foregoing pages is that it is one and the same Ultimate Reality that is manifested to us subjectively and objectively." ^ The antithesis of Subject and Object, of Ego and Non- ego, belongs to Appearance and not to Reality. In- deed it is just the inevitability of this antithesis, combined with the philosophical conviction that it is not valid, if taken as representing Reality, that is the ' Principles of Psychology, vol. i. § 273, p. 627. [Quotations throughout from the 3rd edition.] In view of this sentence, and the preceding section, Kiilpe is douhtless right in regarding Spencer as a Monist of what he calls an ' abstract ' type. 288 rillLOSorilV nv HKKHERT spencer lkct. deepest Ikisis of Mr. Sponcor's Ai^iiosticism. *' The antithesis," lie says, " of subjeet aiul object, never to be transeendeil while consciousness lasts, renders impossible all knowledge of that ultimate reality in which subject and object are united," ' But this does not meet the ditticulty of consist- ently athrming knowledge of the utterly unknow- able : since the affirmation that it is one imports just as definite a piece of metaphysical knowledge as the affirmation that it is two or more. But 1 will not dwell on this now : as we shall find later on that Mr. Spencer seems to have a much more extensive and complex knowledge of his Unknowable. I will rather point out that ' the ultimate law of thought ' which he goes on to explain {First Principles, § 44) hardly seems to me to justify even his phenomenal duality. He says that the " primordial division of self from not- self is a cumulative result of persistent con- sciousnesses of likenesses and difierences among manifestations." But though the two groups of manifestations are internally alike, and unlike each other, in being respectively ' vivid ' and ' faint,' it hardly seems that the unlikeness is sufficient even to suggest their reference to difi'erent powers, when we consider that the ' faint ' are said by Mr. Spencer to be copies or repetitions of the * vivid.' If the copies called ideal sounds resemble the copies called ideal smells in being faint, they resemble on the other hand the vivid manifestations called real sounds in quality, and, so far as the latter resemblance goes,. ' Op. dt. p. 627, end of § 272. 1 METAPHYSICAL DOCTEINES 289 would be naturally referred to the same cause as the real sounds, operating more feebly. However, according to Mr. Spencer the group of ' vivid ' manifestations, carrying with it the implica- tion of a manifesting power and excluding emotions (which are rather summarily thrown over to the ' faint ' manifestations), is the non-ego. But our common notion of the non-ego implies existence distinct from and independent of the ego : indeed Mr. Spencer goes on to say that the primordial datum of Philosophy is " the postulate that the mani- festations of the Unknowable fall into two separate aggregates constituting respectively the world of consciousness and the world beyond consciousness." ^ But how does the ' vivid ' element, or aggregate of elements, in the stream of our conscious experience — our sensations and sense-perceptions, sights, sounds, tastes, smells, touches, pressures, muscular tensions — how does all this become a " world beyond con- sciousness " ? He has admitted that we commonly think and speak of the ' vivid ' manifestations as states of consciousness : and when he comes to the Principles of Psychology he gives these sensations — distinguished as ' feelings peripherally initiated ' — a leading place among the elements of Mind or Con- sciousness. How then is it that, in First Principles, they come to be an " aggregate of manifestations con- stituting the world beyond consciousness " ? Mr. Spencer's answer is as follows (pp. 155, 156) : — " We continually learn that while the conditions ^ First Principles, p. 15G. 290 rini.()S(U'HV ok HKKHKKT STKNCER lkct. of ooi'unoiu'c t»f laiiit inaiiitrstatiitns arc alwaVH to be loiiiul, tlio i-Diulit ions of (uciincnct' of vivid luani- festations aiv oft»n not to \)c loinul. We also con- tinually leiirn that vivid manifestations wliich have no ])crc('ivable antecedents among the vivid mani- festations, are like certain preceding ones which had perceivable antecedents among the vivid manifesta- tions. Joining these two experiences together, there results the irresistible conception that some vivid manifestations have conditions of occurrence existing out of the current of vivid manifestations — existing as potential vivid manifestations capable of becoming actual. And so we are made vaguely conscious of an indefinitely -extended region of power or being, not merely separate from the current of faint manifesta- tions constituting the ego, but lying beyond the current of vivid manifestations constituting the immediately-present portion of the non-ego ^ It would seem from this that the manifestations that properly constitute ' the world beyond con- sciousness ' — since it is too paradoxical to put beyond consciousness my present sensations of sight, sound, etc. — are merely ' potential manifestations capable of becominf; actual ' : i.e. sensations that we might have but actually do not have. At any rate these merely potential manifestations are a main part of the ' world beyond consciousness.' But how can we conceive merely potential manifestations existing as the con- ditions of occurrence of actual manifestations ? Surely the conditions must be as ' actual ' as the manifesta- tions that they condition ! Moreover, before, in deal- I METAPHYSICAL DOCTKINES 291 ing with the pantheistic hypothesis of self-creation, Mr. Spencer has laid down that " we cannot form any idea of a potential existence of a universe as dis- tinguished from an actual existence : if represented in thought at all, potential existence must be repre- sented as actual existence." ^ Well, what is sauce for Pantheism must be sauce for Phenomenalism : the potential vivid manifestations must be thought as actual : but if thought as actual, how can they be thought as beyond consciousness ? There seems to be a dilemma. If the " vivid mani- festations indissolubly bound together which we call the non-ego " are actual, they cannot constitute a " world beyond consciousness." They must be within consciousness, elements of consciousness in the sense in which Mr. Spencer conceives consciousness when he distinguishes ' the thoughts and feelings which con- stitute a consciousness' — with which Subjective Psy- chology is concerned — from the existences with which the rest of the sciences deal. They must therefore belong to Mind, in the sense in which Mind is regarded as " something totally without kinship with other things " : ^ that is to say, they must belong to the ego, not to the non-ego. If, on the other hand, the vivid manifestations are conceived as merely potential, they cannot constitute an actual non-ego, an actual world beyond consciousness ; and it is an actual, not a potential world, which Common Sense and physical science require. ^ First Principles, § 11, p. 32. 2 Principles of Psychology, vol. i, § 56, p. 140. 29 -J I'lllLOStU'llV ol' HKKUKKT SPENCER lect. In tho faro oi' tlu.s ililfiniua will .Mr. Spencer ulti- mately ilecicK' ti> K't till' ' vivid inaiiirestatious ' go to the Ego or subject^ He certainly seems to do this iu some jmssagos : and is indeed led to this by another line of thought, developed in the Psychology (Pari II. chap, iii.), iu which he gives an elaborate psycho- physiological proof, in liis best manner, of the pro- position that " though internal feeling habitually depends upon external agents, yet there is no like- ness between them either in kind or degree." The feeling, he argues, is an etl'ect which varies, qualita- tively and quantitatively, according to the specific structure of the sentient organism, its individual structure, the part affected, the condition and motion of that part, etc., while the cause all through remains the same. " Thus," he says, " we are brought to the conclusion that what we are conscious of as properties of matter, even down to its weight and resistance, are but subjective afiections produced by objective agencies that are unknown and unknowable. All the sensations produced in us by environing things are but symbols of actions out of ourselves, the natures of which we cannot conceive."^ And what is here said of ' Relativity of Feelings ' is said in the next chapter of ' Relations between Feelings ' : it is simi- larly shown that no relation in consciousness can " resemble or be in any way akin to its source beyond consciousness," it can only symbolise something un- known beyond consciousness. Accordingly the con- clusion that he calls Transfigured Realism is thus ' Principles of Psychology, vol. i. § 86, p. 206. I METAPHYSICAL DOCTRINES 293 stated : " While some objective existence, manifested under some conditions, remains as the final necessity of thought, there does not remain the implication that this existence and these conditions are more to us than the unknown correlatives of our feelings and the relations among our feelings. The Realism we are committed to is one which simply asserts objective existence as separate from, and independent of, sub- jective existence. But it affirms neither that any one mode of this objective existence is in reality that which it seems, nor that the connexions among its modes are objectively what they seem."^ But, if this be so, if the ' vivid manifestations ' are not properly thought as elements of the objective existence beyond consciousness, but only symbols of such existence, which they do not resemble and to which they are not in any way akin, what becomes of that differentiation of subject and object, elaborately expounded in First Principles, and expounded again more fully in The Principles oj Psychology ? For by this differentiation, owing to the accumulated differences between ' vivid ' and ' faint ' manifestations, the former are shown as aggregated into the Non-ego and the latter into the Ego : so that the funda- mental antithesis between the two appears to be the necessary result of psychological laws. But this necessary result, this conclusion that we are irresist- ibly led to think, is surely the conclusion of Crude Realism, not of Transfigured Realism. In describing it Mr, Spencer continually talks of sights, sounds, ^ Principles of Psychology, vol. ii. § 472, p. 494. 204 ruiLitsornv i)\' HKKiu-iirr spenckk i.kct. odours, pressures, sensations of cokl, etc., as the leadinu; examples oi' viviject and ()l)ject,' all the ' vivid manifestations,' even the primordial consciousness of resistance, have to l)e recognised as subjective and altogether unlike anything in the ' world beyond consciousness ' ; in fact this world, as an object of Mr. Spencer's thought, is merely an Unknown Cause and Unconditioned Reality — though, as he atHrms it to ' persist ' and ' to be without beginning or end,' I suppose it must be conceived to be in time. But still this mentalistic agnosticism does not ex- press his final view, in spite of the vigour with which some of his arguments lead to it. He still holds to his phenomenal dualism — for reasons which I will presently examine — and he holds to a knowledge of the world beyond consciousness, which I at least find it difficult to reconcile with his agnostic utterances. Take, for example, besides all that is said of ' the differentiation of vivid from faint manifestations,' the interesting discussion of the scope of Logic which we find in his Principles of Psychology {§§ 302-305). Here, speaking as one of those who "acknowledge that subject and object are separate realities," he states as the distinctive characteristic of the science of Looric — as distinoruished from an ' account of the process of reasoning' — that "Logic formulates the most general law-s of correlation among existences I METAPHYSICAL DOCTRINES 299 considered as objective . . . contemplates in its pro- positions certain connexions predicated, which are necessarily involved with certain other connexions given : regarding all these connexions as existing in the non-ego — not, it may he, under the form in which we know them, hut in some form.'' '^ Here we appear to know, as existing beyond consciousness, the same connexions which we know in the world of conscious- ness — e.g. relations of number — although we do not know that they exist ' under the form in which we know them.' This seems difficult to reconcile with the proposition that " no relation in consciousness can resemble or be in any way akin to its source beyond consciousness":^ for in the passage describing the scope of Logic, there seems to be not only affinity but some sort of identity between the connexions we contem- plate within consciousness and those that we may believe to exist really in the non-ego. And Mr. Spencer's whole view of Logic is difficult to reconcile with the position that the non-ego or object- world is strictly an unknown and unknowable reality, appre- hended in an indefinite consciousness. The final expression of Mr. Spencer's view is to be found in the chapter entitled Transfigured Realism, where he tries to illustrate it by a diagram, showing the projection of a cube on a cylinder, made by lines radiating from a point behind the cube. The cube represents the objective reality ; the cylinder " stands for the receptive area of consciousness"; the " pro- ^ Principles of Psychology, vol. ii. § 302, p. 87. Italics mine. ^ O}). cii. vol. ii. § 472, p. 494 init. 300 riiii.<>s(»riiv i^' iii:i;iu<:kt spencer lect. jootcil figure stniuls for that stiitc lA' conscionsncss wo i-all a |>orfO|>titm of llic objci't."' Tlu' illustration is worth studying ti> undcrstaiul Mr. Spencer's meta- pliysieal view ; Init it luis a niisleaclinii; element, since cube, evlinder. and projected figure have all in common tlie important attrihnte of extension : so that they are fundamentally more alike than Subject, Percept, and * Reality out of Consciousness ' are held to be by Mr. Spencer. I am not sure that Mr. Spencer sees this : still his application avoids the misleading suggestion. " We may understand," he says, " very clearly how it becomes possible that a ])lexus of objective phenomena may be so represented by the plexus of subjective eflfects produced, that though the effects are totally unlike their causes, and though the relations among the effects are totally unlike the relations among their causes, and though the laws of variation in the one set of relations differ entirely from those in the other ; yet the two may correspond in such a way that each change in the objective reality causes in the subjective state a change exactly answering to it : so constituting^ what we call a cognition of it — a relative knowledge of it." (Jn this I will make now two remarks. First as to ' plexus of objective phenomenal' But what can ' phenomena ' mean here ? The cube, I understand, stands for what I call extra-cognitional fact, the world out of consciousness : ' phenomena,' then, must surely mean the effects on consciousness of such fact. I can- not help thinking that Mr. Spencer is here confusedly ' Principles of Psychology, vol. ii. §473, pp. 496 ff. I METAPHYSICAL DOCTEINES 301 carrying the antithesis of phenomenon and reality outside the sphere within which it belongs. Next it will be observed that we come ultimately to ' what we call a cognition of it [the objective reality] — a relative knowledge of it.' But how is this reconcilable with the assertion in First Principles of the utter incon- ceivability of the underlying reality ? Of this more in the next lecture, when we shall have to examine Mr. Spencer's epistemological prin- ciples. LECTURE 11 METAPHYSICAL AND EPISTEMOLOGICAL DOCTRINES In the last lecture, after explaining Spencer's philo- sophical Agnosticism and its grounds, I passed to the more difficult task of ascertaining the exact relation of this Asrnosticism to the Natural Dualism which he regards as the primordial datum of Philosophy, as systematised or unified knowledge of the knowable. This affirms the profoundest distinction among pheno- mena or manifestations to be that between ego and non-ego, or perhaps rather between ' Mind ' and ' Matter ' (as Mr. Spencer contemplates throughout a plurality of conscious minds). I directed attention to the diversity and contradiction of the conclusions to which we seem to be led when we examine his conception of Matter or the non-ego. To this point I shall return presently. But first I propose to com- plete the discussion of the Metaphysical question (in the narrower sense) by trying to ascertain similarly his view of the Nature of Mind : I shall then pass to his epistemological doctrine. As regards the Nature of Mind we find — as we found regarding the Nature of Matter — that results 302 LECT.ii METAPHYSICAL DOCTKINES 303 reached by different lines of thought are difficult to put together. The process of differentiation of subject and object, which we have examined in considering Mr. Spencer's notion of matter, leads primarily to the conclusion that the Ego is a term for one of the two great aggregates of ' states of consciousness ' — i.e. for the aggregate or series of ' faint states ' (thoughts and imaginations) as contrasted with the aggregate or series of ' vivid states,' ^ which are distinguished as Non-ego. This, as I said, I find irreconcilable with the view of Common Sense — accepted elsewhere by Mr. Spencer — that sensations, colours, sounds, touches, etc. are among the feelings which constitute a con- sciousness or mind, and this, being ' a something with- out any kinship ' with the nervous actions from which those feelings are inseparable, renders the Psychology which studies them a " totally unique science, . . . antithetically opposed to all other sciences whatever." ^ These vivid states of consciousness are also described as ' peripherally initiated feelings,' and as such form one of the primary divisions of " components of mind." ^ So much for the varying and transient psychical facts which Mr. Spencer calls states of consciousness. But in ordinary thought Mind or Ego does not denote an aggregate of these states; but (l) a per- manent identical something of which they are states, ' Vivid states, it will be remembered, are briefly sensations and sense- perceptions, because emotions, though 'vivid states,' are handed over to the ' faint aggregate. ' - Principles of Psychology, vol. i. § 56, p. 140. " Ofi. cil. § 66, p. 166. 304 rilll.osoi'HV nv llKi;i;KKT srENCEU lkct. aiul (-) is conceived tt) l)e differently related to different states, active in some, passive in others. Now, though Mr. Spencer allows himself to speak of the " faint aggregate which I cull my mind," ' he does in his own way recognise that I do not apply the term to anything 1 conceive as merely an aggre- gate. Thus, as regards (2), he rather startles us by referring — as though it needed no explanation — to " the fact that the faint series has a power of chang- ing its own order." ' But surely that is unthinkable. How can the series have the power ? The past states cannot be thought to have it, as being past they are not actual ; still less the future ; and even if we could think the present state of our consciousness as having the power of changing, it is not the series. This ' power of changing,' in short, if attributed to mind at all, must be attributed to it not as a series of changing states, but as something that remains permanent through the series. And in fact Mr. Spencer eventually gives us a new view of " the Subject as the unknown permanent nexus which is never itself a state of consciousness, but which holds states of consciousness together." ^ This would seem to be what we ordinarily call Self or Ego, considered as supplying the element of continuity in our con- scious life. [But then at an earlier stage of his work Mr. Spencer has demonstrated that the substance of Mind cannot be known*], and by this we should understand » Op. cit. vol. ii. § 462, p. 472/n. ^ Qp ^ y^i jj § 455^ p ^qq » Op. cit. vol. ii. § 469, p. 484. * Op. cit. vol. i. § 59, p. 146. II METAPHYSICAL DOCTEINES 305 him to mean that we can know no more about it than this — that it is not and cannot be known. Yet now, to our surprise, this permanent nexus is treated as material ! Even ' self-analysis,' he says, would show the subject "that this nexus forms part of the nexus to that peculiar vivid aggregate he dis- tinguishes as his body " ; and psycho-physiology will enable him to see that it is a set of nervous plexuses. "For, ... an idea," he continues, "is the psychical side of what on its physical side is an involved set of molecular changes propagated through an involved set of nervous plexuses. That which makes possible this idea is the pre-existence of these plexuses, so organised that a wave of molecular motion dif- fused through them will produce, as its psychical correlative, the components of the conception in due order and degree. This idea lasts while the waves of molecular motion last, ceasing when they cease ; but that which remains is the set of plexuses. These constitute the potentiality of the idea, and make possible future ideas like it. Each such set of plexuses, perpetually modified in detail by perpetual new actions ; capable of entering into countless combinations with others, just as the objects thought of entered into countless combinations ; and capable of having its several parts variously excited, just as the external object presents its combined attributes in various ways, is thus the permanent internal nexus for ideas, answering to the permanent external nexus for phenomena." But what then becomes of the ' unknownness ' of the substance of X 306 rilll-OSorHV ok UKKBKKT spencer lkct. Miml ? Mr. JSpont'or is aware tliat lio has to answer this question ; and his answer seems to he tluit the set of nervous j^lexuses is itself only a mental symbol of an unknowable reality. For " our ideas of matter and motion, merely symbolic of unknowable realities, are complex states of consciousness built out of units of feeling."^ Although the set of plexuses appears when we take a psycho-physiological view as the " permanent internal nexus . . . which continues to exist amid transitory ideas" — each idea being only the psychical side of an involved set of molecular motions propagated through the set of nervous plex- uses — this relative permanence of the material sub- stratum of mental phenomena vanishes again when w^e turn to analyse our concept of a nervous plexus. For then we see that our concept of this or any other complex modification of matter " is but the symbol of some form of Power absolutely and for ever unknown to us ; and a symbol which we cannot suppose to be like the reality without involving our- selves in contradictions." ^ *' See then our predicament," he says : " we can think of Matter only in terms of Mind. We can think of Mind only in terms of Matter. When we have pushed our explorations of the first to the uttermost limit, we are referred to the second for a final answer ; and when we have got the final answer of the second we are referred back to the first for an interpretation of it. We find the value of ^ Principles of Psychology, vol. i. § 63, p. 150. 2 Op. cit. vol. i. § 63, p. 159 init. II METAPHYSICAL DOCTEINES 307 X in terms of y ; then we find the value of y in terms of x ; and so on we may continue for ever without coming nearer to a solution. The antithesis of subject and object, never to be transcended while consciousness lasts, renders impossible all knowledge of that Ultimate Reality in which subject and object are united. And this brings us to the true conclusion implied throughout the foregoing pages — the conclu- sion that it is one and the same Ultimate Reality which is manifested to us subjectively and objectively. For while the nature of that which is manifested under either form proves to be inscrutable, the order of its manifestations throughout all mental phenomena proves to be the same as the order of its manifesta- tions throughout all material phenomena," ^ It would seem, therefore, that the 'power' which the * faint series ' has of changing ' its own order ' is after all only the power of our old friend the Unknowable to produce faint manifestations ; and that though by inevitable laws of thought we are led to contrast the power manifested by faint feelings = Ego, with the power manifested in vivid sensations = Non-ego, the ' true conclusion ' is that the same power is mani- fested in both. How are we to put together this complicated set of inconsistencies ? Mr. Spencer's agnostic conclusion doubtless seems to him sufficiently humble ; but I am not satisfied with it. To me it seems misleading for him to say that ' Principles of Psychology, vol. i. § 273, p. 627. :u)8 I'liiLosoriiv o\- iiKi:i;Ki;'r spkncek lect. tlic antithesis of sul)jo('t and ohject is " lu-vor to be tninsceiuleil wliilo (•onsciousiioss lasts." If ilmt were so we surely could not think of (Mthcr in U-rms of the other; but, aceording to his argument, this is just what we can do, and we can do nothing else ; even although at the same time we have to think of mind as something totally without kinship with matter in motion. And out of this medley of oscil- lating contradictions it seems to me to result not only that knowledge of the Ultimate Reality itself is im- possible, but that philosophical knowledge even of its manifestations is — I will not say ' impossible ' — but is as yet unattained l)y Mr. Spencer. For he has told us that the task of Philosophy is to co-ordinate, unify, systematise the results of the particular sciences ; but a systematisation that leaves such fundamental inconsistencies ought surely to admit that it has failed to accomplish its task. I now turn to ]\Ir. Spencer's Epistemology as set forth in his Principles of Psychology, Part VII. chaps, ix.-xiii. I must begin with a brief account of the earlier chapters. In the first he explains that, having " provisionally assumed certain fundamental intui- tions," we have now to "prove their congruity with the other dicta of consciousness ... in other words, we have to take up the vexed question of subject and object. The relation between these, as antithetically opposed divisions of the entire assem- blage of manifestations of the unknowable, was our II EPISTEMOLOGICAL DOCTKINES 309 datum." ^ In chap. ii. commences an attack on ' Metaphysicians ' continued through four chapters. The root-error attributed to metaphysicians is a faith in Reasoning * greatly in excess of that which is its due,' an ' unbounded confidence in it.' ^ Reasoning, says Mr. Spencer, has done so much for us, that we have been led to a superstitious awe of Reason as against Perception, i.e. to an 'unwarranted belief in the superiority of ' the deliverances of consciousness reached through mediate processes to the deliverances of consciousness reached through immediate processes.' He observes, however, that men of science are not apt to fall into this superstition : if experience (or 'reasoning so automatic as to be no longer called reasoning ') conflicts with calculation, they prefer experience. It is metaphysicians who tacitly assume that 'beliefs reached through complex intellectual processes ' are superior in authority ' to beliefs reached through simple intellectual processes,' and, "setting out with this as their postulate, seem unconscious that they have postulated anything."^ But, asks Mr. Spencer, ' how can Reason claim superior trustworthi- ness in the trial of Reason versus Perception ' ? '' But ' Priticiplcs of Psychology, vol. ii. §§ 386 f. pp. 310 f. - As Hume is one of the metaphysicians contemplated, I may remark that it shows a curious ignorance of Hume to attribute to him an unbounded con- fidence in the reasoning process. I suppose Mr. Spencer has never read or has forgotten the first section in Part IV. of the Treatise on Human Nature entitled " Of Scepticism, with regard to Reason." •' Op. cit. § 391, p. 316. * Op. cit. § 391, p. 317. Before going further, may I say that I rather object to all controversies carried on against a class of people holding such various doctrines as 'metaphysicians.' It reminds me of tlie vulgar view of Greek sophists ; and indeed Mr. Spencer is not quite free from unworthy appeal to vulgar dislike of metaphysicians. 310 I'lirLosoniv of hki:p.kkt si-kxcku i.kct. it" wo tako Kiii::lisli IMiilosophy, Locko lias no idea that tlu'iv is any conflict between his philosophical retisoning ami Conimon Sense ; and P>crkeley, who sees the conflict, seriously puts forward Ids ' Idealism' as a moile of reconciling Common Sense and Thilo- sophy. In fact, this reconciliatiou is what nearly every eminent English metaphysician (since it was seen that reconciliation was needed) has been trying to etl'ect, Berkeley as much as Reid, and Brown no less than Hamilton. Hume is the conspicuous excep- tion ; but Hume, while declaring the conflict irrecon- cilable, does not sum up in favour of Reason : that is just what he does not do. JMr. Spencer's reply would be that he is defending ' Realism ' ; and that meta- physicians generally are opponents of true Realism, if they are not all Idealists and Sceptics. Here it becomes obvious to ask : What does Mr, Spencer understand by ' Realism ' ? Well, he adopts the rather inconvenient course of going on for a long time without any definition ; but in the course of the argument it gradually comes to be defined by im- plication. Thus in chap, iii,, on the 'Words of Metaphysicians,' though the main aim is to show that " language absolutely refuses to express the idealistic and sceptical hypotheses," the final positive conclusion is that the words used by metaphysicians " separately and jointly imply existence beyond consciousness ; " ^ e.g. that the word ' impression ' only * remains intelli- gible ' when I understand it as connoting the * in- dependent existence ' of something that impresses, as ' [Principles of Psychohriy, vol. ii. § 39't, p. 335.] II EPISTEMOLOGICAL DOCTEINES 311 well as something — mind — that is impressed/ But other conclusions appear to be arrived at, which are not expressly formulated in the final summary of the chapter. P'or example, that the " word hrown is meaningless unless space of three dimensions ... is simultaneously conceived." ^ So, in chap. iv. on ' the Eeasonings of Metaphysicians,' arguing against the Kantian view that time and space are ' subjective forms,' his conclusion is that it is impossible to separate space from the objective world. It would seem then inferrible from chaps, iii. and iv. that the Realism which Mr. Spencer is concerned to defend is the belief in the existence of an objective world in space of three dimensions. Of this belief he proceeds to give (chaps, v.-viii.) what he calls a 'negative justification': i.e. a "proof that Realism rests on evidence having a greater validity than the evidence on which any counter - hypothesis rests." This negative justification consists of three arguments, drawn respectively from the priority, the simplicity, and the distinctness of the realistic belief. The argument from priority affirms that, in what we commonly regard as sensation or sense-perception of external objects, " the thing primarily known is not that a sensation has been experienced, but that there exists an outer object," and even that "the existence of a sensation is a hypothesis that cannot be framed until external existence is known." ^ By ' primarily known ' Mr. Spencer seems to mean that ^ [Principles of Psychology, vol. ii. § 394, p. 334.] 2 Op. cit. § 392, p. 320. ^ Qp_ cit. § 404, p. 369. ?.\-2 nill.t^SOrilV OF IIKKI'.KUT SPENCER lkct. the (lefinito i-oiit'option of an txttrnal object cornea, in the developnu'iit of (he individual human mind, earlier than the detinile couception of one's own feelings as one's own. And it is important to note how, accord- ing to his view, this external object was primarily conceived. " Even the metaphysician," he says, '• will not fail to remember that originally he regarded colours as inherent in the substances distinguished by them ; tliat sweetness was an intrinsic property of sugar ; and that hardness and softness were supposed actually to dwell in stones and in flesh." ^ But this ' priority ' to sensation of the cognition of matter as coloured, sweet, etc. is importantly qualified. For Mr. Spencer distinguishes * having a sensation ' — which he even calls ' the simple consciousness of sensation ' (!) — from 'being conscious of having a sensation,' and admits not only that the former fact is prior to the cognition of the external object, but also that the ' conception of the outer agent eventually framed is framed out of such sensations ' which are rightly regarded as the * things originally given.' His point, in short, is simply that these sensations existed before there was " any consciousness of subject or object."^ But, thus qualified, the 'argument from priority ' has no force against that species of Mental- ism which I have distinguished as Sensationalism : it concedes all the priority of Sensation to Percep- tion which the Sensationalist — as distinct from the Spiritualist — is concerned to claim. The ' argument from simplicity ' affirms that " the 1 Principles of Tsycholoyy, vol. ii. § 404, p. 372. ^ Qp cU. § 405, p. 373. II EPISTEMOLOGICAL DOCTEINES 313 deliverance of Consciousness which yields Eealism," i.e. the apparent cognition of an external object, is either immediate or — granting it to be inferential — is reached by a single act of inference ; whereas the con- clusion of either ' Idealism or Scepticism ' ^ is reached by a long complex process of inference : the latter, therefore, from its mere length and complexity, in- volves more danger of error. Finally, the ' argument from distinctness '" affirms that "the one proposition of Realism is presented in vivid terms " ; '" while " each of the many propositions of Idealism or Scepticism is represented in faint terms." ^ Therefore the Realistic proposition is prima facie more trustworthy. Surely there is some con- fusion here, due to the fact that Mr. Spencer has not defined the 'proposition of Realism.' Doubtless the elements of the external object as perceived are ' vivid ' ; but the question at issue between Realism and Mentalism does not involve any difference as to these : the question is whether this object has an existence independent of consciousness ; and surely * existence independent of consciousness ' which Realism predicates of the object is a term exactly as faint as ' existence dependent on consciousness.' In short, whatever else in the object as commonly appre- hended is ' vivid and definite ' it is certainly not its objectivity ! Coming now to the main Epistemological doctrine, ^ Or, as I should say, 'Mentalism or Scepticism.' 2 < Vivid ' is not the same as ' distinct.' •' [Principles of Psychology, vol. ii. § 410, p. 380.] 314 rilll.oSi^rilY (»K IIKKI'.KKT SPENCER lect. I ptiss ovor tlio fliarartoristic oiuleavour to prove that thcro miu^t he a Criterion of Tnitli and lu'ror, rimst be an answer to tlie question, " What is it which makes one deliveranee of consciousness preferable to anotlier?" must be " somewhere some fundamental act of thought by which the validities of other acts are to be deter- mined." * It seems to me that j\Ir. Spencer's attempt to demonstrate this necessity is manifestly fallacious : he tries to show that " a certainty greater than that which any reasoning can yield has to be recognised at the outset of all reasoning " ; " but as the demonstra- tion is itself a process of reasoning, it could surely only establish its conclusion by a self-contradiction. With this preliminary remark I pass to the discussion of the criterion that jNFr. Spencer actually proposes. But before proposing it Mr. Spencer first shows by a loose induction that complex propositions are more liable to error than simple ones : he does this in order to lay down that, before applying the criterion, we must ** resolve each complex proposition into the simple propositions composing it," and then test each simple proposition separately.^ He next proceeds to classify propositions *' according as their terms are real or ideal, or partly the one and partly the other." He shows how cognitions may be ' presentative,' * repre- sentative,' and ' re-representative,' or partly one, partly another of these : how they become ' constructively compound' when — remaining particular — they pass into the representative and re - representative ; and ' Principles of Psychology, vol. ii. §416, p. 389. 2 Op. cit. § 417, p. 390 /n. ' [Op. cit. § 422, p. 399.] II EPISTEMOLOGICAL DOCTRINES 315 ' cumulatively compound ' when they are generalised from particular cases. ^ Finally, he passes to the epistemological classifica- tion to which this is preliminary. The fundamental distinction is between (1) propo- sitions " of which the predicates always exist [in consciousness] along with their subjects " ; and (2) propositions " of which the predicates do not always exist [in consciousness] along with their subjects. Those of the first class express cognition such that the thing alleged continues before consciousness as long as the thing of which it is alleged continues before consciousness ; and those of the second class express cognition such that the thing alleged may disappear from consciousness while the thing of which it is alleged may remain. These are respectively the cognitions we necessarily accept and the cognitions we do not necessarily accept." ^ Class (1) is again subdivided into : (a) " Cognitions in which the coexistence of the two terms is but temporarily absolute," such as ' simple cognitions of the presentative order,' as " I perceive light as long as I gaze at the sun " ; and ' certain presentative- representative cognitions,' such as the proposition that a body has extension as long as its resistance is being felt. (6) " Cognitions in which the union of subject and predicate is permanently absolute," such as the axioms of Mathematics, and other " cognitions which contain abstract relations, quantitative or ^ Principles of Psychology, vol. ii. § 423, p. 400. 2 Op. cit. § 425, p. 402. ;uo riiiLosoriiv of hkkrkkt spencer lkct. (|iialit:itivo,'" « .7. tlie "most abstract cognitions which Logic IVuinulatos." ' I>ut "otic inoro important distinction remains to be noticed. " In tlic simpleM propositions of any of these subclasses " the connexion of predicate with its subject is so close that its coexistence cannot be kept out of consciousness." In other cases — e.g. in the " cumulatively -representative cognitions which Logic formulates " — the " invariable coexistence predi- cated is often inconspicuous, and may be overlooked. ... It exists in consciousness but implicitly, and not explicitly. It may not be sought for, and in some cases search may fail to disentangle it."^ In chap. xi. we come at length to the Criterion, to which the previous discussion has been leading up. " The inconceivableness of its negation is that which shows a cognition to possess the highest rank — is the criterion by which its insurpassable validity is known."" ^ Or, in the more psychological language of the preceding paragraph, " to ascertain whether along with a certain subject a certain predicate invariably exists," we have to try " to replace this invariably existing predicate by some other or to suppress it altogether without replacing it." If the negation of a proposition is inconceivable — i.e. if its "terms cannot by any effort be brought before conscious- ness in that relation which the proposition asserts ' Principles of Psychology, vol. ii. § 425, pp. 403 f. "^ Op. cit. § 425, pp. 404 f. This very iraportant remark seems to me to involve Mr. Spencer's view of necessary truth in something like a contradic- tion. For how can it be said that a relation between two terms exists in consciousness when we are not conscious of it, or that we are conscious of it when we overlook it and fail to find it. " Op. cit. § 426, p. 407. n EPISTEMOLOGICAL DOCTEINES 317 between them" — we are "at once under the psycho- logical necessity of thinking it, and have the highest possible logical justification for holding it to be unquestionable." ^ This is, in Mr. Spencer's view, the simple and universal criterion of truth, the ' universal postulate,' on the validity of which the validity of all reasoning depends. Before I examine the criterion, the meaning of the term " inconceivable " requires some discussion. In the controversy between Mr. Spencer and J. S. Mill, to which reference is made in Mr. Spencer's chap. xi. and Mill's Logic, Bk. II. chap, vii., we find both admitting that, in ordinary use, ' inconceivable ' has two meanings, one of which is ' incredible ' : and both equally regard this latter meaning as improper. Mill, however, holds that Mr. Spencer has been some- what hasty in repudiating the meaning so far as his use of the term is concerned. I have said Mr. Spencer intends the criterion to guarantee propositions that represent particular facts, no less than propositions of universal import — e.g. the proposition ' I feel cold,' or * I perceive light ' when I am gazing at the sun — and Mill urges that if I say that the opposite of such a proposition is inconceivable, I must mean incredible ; for it would not be true to say, in the strict sense of ' conceive,' ' I cannot conceive myself not feeling cold.' We can say, " I cannot conceive that I am not feeling cold," but then we have passed from con- ception to belief. Mr. Spencer, as I understand, maintains that in this case the coexistence of the ' Principles of Psyclwlogxj, vol. ii. § 426, p. 407. :u8 riiii.osoriiv ok hkkkkkt srENCEii lkct. predicAte-notion ' foelinix coKl ' wltli tlie subject-notion 'self is 'temporarily absolute,' but only ' tempor- arilv.' But is this so? Only, I think, in extreme oases of very intense sensation or ronce[)ti()n. Shake- speare says — No man can hold a (iro in liis liand By tliinking on the frosty Caucasus. And though I have never tried this painful experi- ment, I think it probable that it would exclude even the imajiination of Caucasian frost. But that would not be the case with a milder degree of disagreeable heat. I find, indeed, that disagreeable sensations, when not too violent, even tend to provoke the imagination of their opposites, e.g. great thirst con- tinually excites the image of cool spring water gur- gling down my throat, etc. I cannot, therefore, agree that the utmost certainty in a proposition representing a transient particular fact involves the inconceivability of its negation, except in the special sense of inconceivability, in which it is indistinguish- able not from ' incredibility ' unqualified, but from intuitive incredibility. This particular species of in- credibility Mr. Spencer does not take account of in his distinction. It is not ' intuitively incredible ' that a cannon- ball should be fired from England to America ; though, as Mr. Spencer says, it is 'unbelievable.'^ But my refusal to believe it cannot be justified by a mere examination of the terms of the proposition : it requires me to recall what I know of the experi- ' IFrinciples of Psychology, vol. ii. § 427, p. 408.] 11 EPISTEMOLOGICAL DOCTRINES 319 enced range of cannons. In this meaning I agree with Mr. Spencer in regarding ' inconceivability of negation ' as a universal characteristic of propositions which present themselves as self - evident truths. But I do not hold this * Intuitive Criterion,' as I €all it, to be infallible — any more than the Cartesian form of the criterion.^ Let us now observe the limitations with which Mr. Spencer affirms the validity of his criterion : "That some propositions," he says, "have been wrongly accepted as true, because their negations were supposed inconceivable when they were not, does not disprove the validity of the test, for these reasons: — (1) That they were complex propositions, not to be established by a test applicable only to propositions no further decomposable ; (2) that this test, in common with any test, is liable to yield un- true results, either from incapacity or from careless- ness in those who use it."^ These two qualifications surely reduce very much the practical value of the criterion. For how are we to proceed if philosophers disagree about the application of the criteria ? How are we to test ' undecomposability ' ? For notions which on first reflection appear to us simple are so often found on further reflective analysis to be composite. Which conclusion, then, are we to trust, the earlier or later ? This seems to me a serious dilemma for Mr. Spencer ; whichever way he answers he is in a difficulty. If he says the earlier, then I do not see how he can meet Mill's example of the dis- ' Of. below, pp. 461 f. " Principles of Psychology, vol. ii. § 433, j). 425. :\'2o rim.osoruv ov iiKKr.KUT sncNCPni lect. beliof in tlio existence of antipodes, for the proposi- tion that 'heavy tliin Xoyrp opcpavco ovri. eXirep yap 6 Trarrjp avrov e^Tj, iroWd 7 av ^fjivve. The line marked out for such a rejoinder will appear more clearly from a brief notice of the steps of the controversy. The old view of the Sophists was that they were a set of charlatans who appeared in Greece in the fifth century, and earned an ample livelihood by imposing on public credulity : profess- ing to teach virtue, they really taught the art of fallacious discourse, and meanwhile propagated ' [In an edition published subsequently Sir Alexander Grant modified his view to some extent.] THE SOPHISTS 325 immoral practical doctrines. That, gravitating to Athens as the UpvTaveiov of Greece, they were there met and overthrown by Socrates, who exposed the hollowness of their rhetoric, turned their quibbles inside out, and triumphantly defended sound ethical principles against their plausible pernicious sophistries. That they thus, after a brief success, fell into well- merited contempt, so that their name became a byword for succeeding generations. Against this Grote argues: (l) That the Sophists were not a sect but a profession : and that there is no ground for attributing to them any agreement as to doctrines. That, in fact, the word Sophist was applied in Plato's time in a more extensive sense than that in which he uses it : so as to include Socrates and his disciples, as well as Protagoras and his congeners. So that, as far as the term carried with it a certain invidious sense, this must be attributed to the vague dislike felt by people generally ignorant towards those who profess wisdom above the common : a dislike which would fall on Plato and the Philosophers as well as on the paid teachers whom he called Sophists : though no doubt the fact of taking pay would draw on the latter a double measure of the invidious sentiment. (2) That as regards the teaching of immoral doctrines, even Plato (whose statements we must take cum grano) does not bring this as a charge against the principal Sophists, Protagoras, Prodicus, Hippias, Gorgias : that it is a priori improbable that any public teachers should propound doctrines so offensive to the common 326 Till-: soriirsTs seiitimonts ot" inankiiitl : that, therefore, we can soan'oly suppose that 'rhiasyinaclius so propounded the iinti-social theory of justice attributed to him by Phito in the Republic ; and that even if he did, we cannot infer from this anything as to the other Sophists. On this second point Grote is chiefly at issue with the German writers (with whom Sir A. Grant substantially agrees). It is on the first head that Mr. Jowett joins issue, and to this I shall at present restrict myself. JNIr. Jowett urges that though the meaning of the word Sophist has no doubt varied, and has been successively contracted and enlarged, yet that there is a specific bad sense in which any intelligent Athenian would have applied the term to certain contemporaries of Socrates, and not to Socrates himself, nor to Plato. Wherever the word is applied to these latter, " the application is made by an enemy of Socrates and Plato, or in a neutral sense." In support of this he points out that " Plato, Xenophon, Isocrates, Aristotle " all give a bad import to the word : and the Sophists are " regarded as a separate class in all of them.' Now, first, I should have thought that we might say of any term denoting a man's walk in life, and connoting doubtfully an invidious sentiment, that it is either applied in a neutral sense or by an enemy, i.e. with polemical intent. Even the slightest flavour of dislike is enough to make the man himself, and his friends, avoid such a word : as we see in the common use of the terms 'attorney' and 'solicitor.' THE SOPHISTS 327 Therefore, that disciples of the martyred sage, and those who learnt from them, never called Socrates a Sophist is very certain. But that the Athenian public considered him as such, whether intelligently or not, is surely undeniable. Mr. Jowett says that Aristophanes may have identified Socrates with the Sophists " for the purposes of comedy." But the purposes of comedy are surely not served by satire that does not fall in with common conceptions. The Athenians looked on Socrates as the most popular and remarkable of the teachers to whom young men resorted with the avowed object of learning virtue or the art of conduct, and the more evident result of learning a dangerous dexterity in discourse ; and as such they called him a Sophist. The differences between him and such men as Protagoras would appear to them less important than the resemblances. The charges brought against him by his accusers express just the general grounds of suspicion felt against both alike. "Whether a man corrupted youth rhetorically or dialectically, whether he made the worse case appear the better by Declamation or Disputation, would seem to them quite a secondary matter. That this view involved a profound mis- apprehension, I do not of course deny : but all evidence seems to me to show that the misapprehen- sion was wide -spread and permanent. More than half a century afterwards, ^schines (who can scarcely be regarded as * an enemy '), when pleading for another example of salutary severity, reminds the Athenians how they had put to death the Sophist 328 THE SOPHISTS Socrates. Ai\ocr6(f)oiiplus(cs uiul caricatured in the Euthydcmus. The difference of type is most striking. The Sophist's manner of discourse is no longer sharply contrasted with that of Socrates : it is rather, as Professor Campbell says, " the ape of the Socratic Elcnchus." A shifty disputer has taken the place of the windy declaimer of the other dialogues : instead of pre- tentious and hollow rhetoric we have perverse and fallacious dialectic. The Sophist of the Protagoras and Gorgias has close affinity to the p^rwp and is with difficulty distinguished from him : in fact, Plato can only distinguish them by restricting the sphere of pT}TopLKri to forensic speaking : this, he tells us, is a quackery that simulates justice, while the Sophists are more ambitious quacks who mimic the art of legislation. These latter, then, correspond to the teachers of ttoXctikoI Xo-yot among whom Isocrates classes himself — strongly objecting to be confounded with those who merely wrote and taught for the law- courts — except that the latter carefully avoids the more vague and extravagant professions which Pro- tagoras and others probably made : he still, however, maintains that in so far as Virtue, Practical Wisdom, and Political Science can be taught, the teaching of them is involved in and bound up with the art of public speaking, his own ^iXocro^ia- This, he claims, does impart to Xiyeiv ev Koi (f)povetv in so far as these are not gifts of nature and effects of practice : and as making this claim he is distinctly Plato's Sophist of the first type. Still this restriction of pr,TopiKrj to its THE SOPHISTS 335 forensic application is somewhat forced : both Sophist and Rhetor would be popularly regarded as professing the art of declamatory or rhetorical discourse and so naturally classed together and confounded : as Plato himself tells us in the Gorgias, vpovTai, ev to3 avrw Kol irepi ra avrd. But the Sophistes of the dialogue so called is expressly contrasted with both the Statesman and the Rhetor : he is the Professor of Disputation, of the art of question and answer according to rules, ipio-TLK^, — thus exhibiting exactly the character which Isocrates tries to fix upon Plato. Further, we are told that this Sophist claims to deliver men from groundless conceit of their own knowledge by cross- examining them and pointing out their inconsistencies : the special function of Socrates. Of course Plato does not admit that the Sophist is the true Dialectician : but he resembles him as a wolf does a dog. He is a tremendous argufier, and able to impart to others the argumentative art. The difference between him and Socrates is that his effect is purely negative : he begins and ends with captious disputation, his skill is simply to bewilder and perplex : he is not, as Socrates, a midwife of true knowledge. It is just this difference which is dramatically exhibited in the Euthydemus, with much broad drollery of caricature. Here a couple of Sophists of the eristical sort are seen exercising their art on an intelligent youth. They put captious questions to him and entangle him in contradictions by means of verbal quibbles, until he does not know whether he is r.:?0 TIIK SOPHISTS staiuling on his lu\ul or liis hcols. Socrates then Uikvs him in li;uul and, by gentler (jueationiug, ultimately draws out of him answers of remarkable point and pregnancy ; and so the true Dialectic is contrasted with its counterfeit Eristic. The ditl'erence is clear enough to us, who arc accustomed to trace the whole growth of philosophy from the fertile germ of Socratic disputation. But we can see even from Plato himself that it would be much less clear to unphilosophic contemporaries : that the effect of the Socratic interrogations on a plain man would be just this bewilderment and perplexity and sense that he had been taken in l)y verlml quibbling, which Plato describes as the effect of Eristic Sophistry. At any rate, the Sophist of the Sophistes and the Euthydemus is much more like the disciples of Socrates than he is like the Sophist of the Protagoras and the Gorgias. And therefore, while the uninstructed public, as we have seen, would lump Declaimers and Disputers together as Professors of the Art of Discourse, I think Mr. Jowett's " intelligent Athenian " would be much more certain to grasp the distinction between the teachers of public speaking who more or less claimed to impart political wisdom on the one hand, and the teachers of disputa- tion and ethics on the other, than he would be to appreciate the finer differences that separated Euthydemus and Dionysodorus from the Socratic Schools. But we may go further than this. Plato himself does his best to obliterate these latter differences : THE SOPHISTS 337 not of course as far as his own teaching is concerned, but certainly in respect of his brother Socratics. Even the received Histories of Philosophy do not altogether conceal this fact from the student. It is true that he reads in one place of Sophistical Eristic, which he is led to look on as a part of the charlatan's stock-in-trade : and in another place of Megarian Eristic, which he regards as a development of philo- sophy. But he can get no clear notion of the difference between the two : and when he comes to the Euthydenius he finds them indistinguishably blended in the object of Plato's polemic.^ Not only is the whole manner and method of the Sophists in this dialogue a manifest caricature of the manner and method of Socrates — the Sophists profess et? aperrj^ eTn^eXeiav irpoTpk^ai by means of dialogue : they challenge the interlocutor v'ire')(etv \67; — he says — 8vo \o^ouhilosophical originality and his moral earnestness must have expressed themselves in some quite different manner. But Socrates once there, appearing to the public as the Arch -Sophist, who overcame all rivals in wordy fight, and by his greater impressiveness and attractiveness to youth threw them all into the shade, so that comedians naturally selected him to represent the class — wdiat could be more natural than that he should have a host of imitators ? Indeed Xenophon expressly tells us of such men who, from the free and abundant banquet of Socratic discourse, carried away fragments which they sold for money. The question then is, Would Plato call such men Sophists ? It must be borne in mind that a Sophist, in Plato's peculiar use of the term, combined two attributes : he taught for pay, and he taught sham knowledge : and the term might seem to be applicable wherever these attributes were found in combination. If then there were among the disciples of Socrates men who taught for pay, not having private fortunes like Plato, and who taught sham knowledge, i.e. doctrines with which Plato disagrreed : how was he to regard them? I imagine he w^ould be puzzled, and would make dis- tinctions among them. There might be some like o o Euthydemus and Dionysiodorus, in wdiom he would THE SOPHISTS 343 feel an absolute want of philosophic earnestness : with these, whether they had or had not formed part of the — no doubt varying and irregular — circle who listened to Socrates, he would recognise no tie of brotherhood : and would not hesitate, if occasion oifered, to satirise them under the invidious term. There would be others like Aristippus, who certainly took money for his teaching, and against whose theory and practice Plato would feel a strong aversion : but who was yet a man of convictions, and a man of speculative force and originality. He would be difficult to class. And in fact, though Aristotle speaks of him as a Sophist, Plato never does, never indeed mentions him personally, though he is under- stood to be directly controverting his theories in two dialogues. If, again, there were also members of the School of Megara, with which Plato had at first felt the closest affinity, and from which his divergence had been slow and gradual : if these undoubted Socratics had fallen away into the wickedness of taking fees, while their dialectical method degenerated more and more into captious and purely negative disputation : Plato, we may suppose, would be pained and perplexed. But he might gradually come to recognise that these men, even though they might be old friends and actual co-disciples of Socrates, were yet essentially Sophists, and their teaching Sophistry. I conceive, then, that Socrates was seed and source of a new kind of Sophistry, the post-Socratic Sophistry, as we may call it : which it was extremely difficult for the subtlest mind to distinguish from the profes- 3\\ THE SOPHISTS sion of Socratic philosophy. Or may wo not say, that the distinction wouUl be properly impossible, conjecturing that the proper positive and negative characteristics of the Sophist, presence of fees and absence of philosophic earnestness, would not be found together ? It is clear that Plato's conception of a Sophist involves the — I trust — groundless assump- tion that " the man who takes fees must be a quack " : and if he found men taking fees, whom he would shrink from calling quacks, though he might deplore their philosophic aberrations, he would be in a dilemma as to the employment of the term. At this point, one wants to know exactly how far the Socratic principle of not taking fees was carried out in what we are accustomed to call the Socratic schools, intensively and extensively : how many acted on it, and how strictly. No doubt all true disciples of Socrates would be reluctant to abandon the principle, and to give for gold what gold should never buy.^ But il faut vivre : and what were men to do who had neither the ainapKeia of Antisthenes nor the fortune of Plato ? To the latter, indeed, who is described to us as consuming his full share of ra e^co dyaOd, such men might fairly say, in the words of Euripides — TTjOO? Tcbv e^ovTcov Tov vofMOv tl9i]<;. Then, again, there are different ways of effecting the transfer of commodities : one may veil or attenuate the repulsiveness of the transaction in various degrees. Even the virtue of Socrates is said to have gone out ' Cf. Memorabilia, I. c, vi. § 13. THE SOPHISTS 345 frequently to dinner : Quintilian, indeed, reports a tradition that ' Socrati coUatum sit ad victum.' ^ Plato was, as I have said, well-born, and probably well-to-do : but even he, if we may trust the Epistles, did not disdain presents from Dionysius and other friends. Poorer Socratics, one may surely assume, would take similar presents with less scruple, and the practice would gradually become regular. At this stage it would be difficult to distinguish presents from fees : especially from fees claimed in the magni- ficent manner of Protagoras. I observe that Dr. Thompson has no hesitation in identifying the dis- putatious Sophists of Isocrates, who imparted virtue for four or five minae, with " some of the minor Socratics " : and it seems probable that the number of such paid Socratics would increase as time went on and the personal influence of the master declined. In fact, the principle of gratuitous teaching was so impracticable, that it must be given up : until the community generally saw the propriety of supporting philosophers, as in Plato's model state, they must get a livelihood out of society somehow. Meanwhile, I think, we may assume that the first type of Sophist was declining : or rather was gradu- ally shrinking back into the rhetorician out of which he had expanded. The new dialectical method had the attraction of novelty : and at the same time all the nobler element of the strong and wide -spread influence which had thronged the lectures of Pro- ' The same authority adds that Zeno, Cleanthes, and Chrysippus mereedes ncceptaverint : so that the principle appears to have been altogether aban- doned by the severest of the post- Aristotelian schools. 346 TlIK SOl'HISTS tagttms aiul llippias, tlio oiithusiasm for wisdom aiitl virtuo, the iVarloss aspiration and tlie sublime credulity of youth, would be attracted and absorbed by the new teaching. laocrates, no doubt, with his " philosophy " represents in a manner the old Sophists : but in his profession of practical wisdom there was but a meajrre residuum of the ma ovre o olSev ovre o f^rj olSev. This latter position is examined at length in the Theaetetus, which I consider to belong to a group of dialogues later than any yet mentioned. This group is defined in my view by two characteristics. (1) The concentration on ethical and political in- terests, due to the influence of Socrates, has ceased : Plato's attention is fixed on questions from a social point of view more narrow and professional, from a philosophical point of view more central and fundamental — on knowledge : its nature, object, and method. He has passed definitely from the market- place into the school ; and as an indication of this (2) he is now engaged in controversies with other philosophers : an element absent from the earlier dialogues — even from the Republic. When he takes up ethical questions again, as in the Philebus, the more scholastic and technical treatment is strikinor. o Now in the Theaetetus perverse dialectic is noticed, though not by the name of Eristic, but by that of Sophistic, which here bears its later meaning. "If," says Socrates, "you and I were engaged in Sophistic logomachy {^weXOovre^ ao^iaTiK(o\v tliat tlic EuthydcmiLS bus gener- ally been phiced earlier : l)ut I think this is due to a mistaken inference from the style. The extreme dit^erence of form has blinded readers to the sub- stantial affinity of its polemic with that of the Sophistes. I am aware that any argument which depends on an assumption as to the order of Plato's dialogues is insecure, on account of the diflference of opinion that exists on the subject. In particular, many would dispute the place I assign to the Theaetetus. But most, I think, would allow at any rate that there was a time at which Plato attacked as Sophists rhetorical moralists and politicians, a later time at which he defined a Sophist as a perverse disputer, and a time betw^een the two at which he contended against the same sort of perverse disputations with- out identifying it wdth Sophistry. And this seems strongly confirmatory of my view that this kind of disputatious Sophistry is post-Socratic and a degener- ate ofi"shoot of Socratic method. THE SOPHISTS n (Reprinted from the Journal of Philology, vol. v. No. 9, 1873.) In the last number of this Journal I argued in favour of the view put forward by Grote as to the common acceptation, in the age of Socrates and Plato, of the term Sophist. I tried to show, that even after it had partly lost its vaguer and wider signification — inclusive of Masters of any Arts, Poets and literati generally — it still was not restricted to teachers of a particular sect or school, having common doctrines, or even a similar philosophic tendency : but was applied to all whom the vulgar regarded as teaching \6y(ov Texi'v^, whether they were rhetoricians and declaimers like Gorgias and Protagoras, or arguers and disputers, after the fashion that Socrates brought into vogue. It comprehended, therefore, several classes of persons besides the Professors of the Art of Conduct with whom Socrates is contrasted in the earlier Platonic dialogues. It included, for example, Rhetoricians generally, even though like Gorgias they disclaimed altogether the teaching of Virtue : in fact, it is evident from Plato's Gorgias that the 351 352 THE SOPHISTS (listiiu'tion which \ic thnv tries to draw botwecu Sophist aiul Rhetor is but vaguely apprehended by the popular mind. It ineluded also (as I was cliieHy coiicernod to show) Socrates and his disciples : who were considered — by all except themselves — as Sophists of the Disputatious, as distinct from the Declamatory, species. In fact, even Plato, in his later works, and Aristotle, show us, under the title of Sophist, a professor of quasi-Socratic argumenta- tion : quite unlike the rhetorical lecturers on Conduct whom Socrates confutes in the earlier dialofjues. We may perhaps distinguish three stages in the signification of the term : or rather (as they are not strictly successive) three areas of an application narrowing gradually, but not uniformly, so that at any time the class would be conceived with consider- able vagueness, and very differently by different persons. (1) Even after the o-o(/)ta which a Sophist professed w^as generally understood to be something higher than mere technical skill in any department, still an eminent specialist who made any pretensions to general enlightenment might easily be called a Sophist : and so the term would be applied, by many persons, to such professors of music as Damon and Pythoclides, to Hippodamus the architect and Meton the astronomer. Then (2) I conceive that for about the period 450- 350 B.C. the word was commonly used to denote all who professed, as Xenophon says, \6ywv Texvrjv : including both the rhetorical and dialectical pro- THE SOPHISTS 353 fessors of the Art of Conduct (which the vulgar would persist in regarding as an Art of talking about conduct), and also rhetoricians like Gorgias, Polus, etc., down to Isocrates : not that the line between the two was very clearly drawn, as Isocrates claimed that his ' Philosophy ' really involved instruction in morals, and it was matter of debate down to the time of Cicero whether the true orator must not necessarily possess a knowledge of things in general. However, during the latter half of this period, after the death of Socrates, the appellation, being an invidious one, was probably repudiated with equal vigour and ultimate success by Rhetoricians and Philosophers. But (3) we need not doubt that the still stricter manner in which Plato (in the Gorgias) conceives the class of ao(f)iarat, distinguishing them from the prjTope'^, was at least partially current in the time of Socrates. For when once cultivated society in Greece had become persuaded that aperr] — excellence of character and conduct — could really be imparted in lectures, and were willing to pay large sums for obtaining it : naturally the professors of this Ars Artium would be regarded as in a special sense Professors of Wisdom, ao(f)iaTaL And it is such men as these that the term always suggests to readers of Greek history, however they may be vaguely conscious of its wider usage. The fresh light in which he placed the ethical teaching of these men was the most important result of G rote's discussion. If his argument had appeared generally so over- 2 A 354 THK SOl'HISTS whelniini,' as it scema to myself, the present paper NvouM not li;ivo hvrn written : hut since the contrary view is still supported hy tlie whole prestige of (iernian erudition, I shall endeavour to re -state Grote's ease in such a manner as to show most clearly on what a curious combiuatiou of mis- represented historical evidences, and misconceived philosophical probabilities, the opposite theory rests. But before doing this, 1 wish to notice one or two points in which I cannot follow Grote, and by which he seems to me to have prejudiced unnecessarily the general acceptance of his theory. Although one mav fairly say that to a mind like Grote's scarcely anything could be more antipathetic than the manner of Protagoras and his followers : and although it is evident to careful readers of his Plato, that he had the deepest enthusiasm for the spirit that dwelt in Socrates, and reigned over the golden age of Greek philosophy : still the intensity of his historical realisation has made him appear as an advocate of the pre-dialectical teachers. He seems always to be pleading at the bar of erudite opinion for a reversal of the sentence on certain eminent Hellenes. Now with this attitude of mind I have no sympathy. There was at any rate enough of charlatanism in Protagoras and Hippias to prevent any ardour for their historical reputation — even though we may believe (as I do) that they were no worse than the average popular preacher, or professional journalist, of our own day. One might more easily feel moved to take up the cudgels for Prodicus, resenting the THE SOPHISTS 355 refined barbarity with which Plato has satirised the poor invalid professor shivering under his sheepskins. But justice has been done to Prodicus by the very German erudition against w^hich I have here to contend. And as for the class generally — they had in their lifetime more success than they deserved, and many better men have been worse handled by posterity. It is only because they represent the first stage of ethical reflection in Greece, and there- fore the springs and sources of European moral philosophy, that one is concerned to conceive as exactly as possible the character of their teaching. The antagonism to that teaching, which developed the genius of Socrates, constitutes really so intimate a relation that we cannot understand him if we misunderstand ' Sophistik.' But again, in his anxiety to do justice to the Sophist, Grote laid more stress than is at all necessary on the partisanship of Plato. No doubt there is an element of even extravagant caricature in the Platonic drama : and the stupidity of commentators like Stallbaum, who treat their author as if he was a short-hand reporter of actual dialogues, is provoking. Still, one always feels that the satirical humour of Plato was balanced and counteracted by the astonish- ing versatility of his intellectual sympathy. And the strength of Grote's case lies in what Plato actually does say of the Sophists, and not in suggestions of what he may have said untruly. Before examining the evidence, it may be well to state clearly the conclusions commonly drawn from it ;k'»g thk sophists which I reijjanl as erroneous. What does ii writer mean wlien he speaks of ' Sophistical ethics,' ■ Sophist ieiil theories on Law and Morality'? As far as 1 can see, he always means speculative moral scepticism leadinj^ to pure egoism in practice. He means a denial of the intrinsic validity of all traditional social restraints, and a recommendation to each individual to do exactly what he finds most con- venient for himself. That nothing is really proscribed or forbidden to any man, except what he chooses to think so : that Nature directs us to the unrestrained pursuit of pleasure, and that the seeming -strong moral barriers to this pursuit become mere cobwebs to enlightened reflection : that " Justice is good for others " than the just man, and that the belief that it is good for him to be just is kept up by these others in their own interest — this is supposed to be the teaching which the youth of Athens thronged to hear. Whatever speculative and rhetorical garnish the Sophists may have added, this was " der langen Rede kurzer Sinn." I might have abstracted this statement from almost any of the German writers w^hose works are text-books in our universities : but I will choose as my authority the generally judicious and moderate Zeller. He speaks of "Sophistik" as " Moralische Skepsis": of the " Sophistische Theorie des Egoismus," the sophistical " Grundsatz dass fiir jeden recht i^ei, was ihm niitzlich," the sophistical " Satz von der Naturwidrigkeit des bestehenden Rechts " : to the Sophists, he says, " das naturliche Gesetz schien nur THE SOPHISTS 357 in der Berechtigung der Willkiir, in der Herrschaft des subjectiven Beliebens und Vortheils zu bestehen " : "das Sophistische Ideal" was "die unbescliriinkte Herrschermacht. " I need not multiply quotations : and perhaps even these are superfluous. In Schwegler's smaller treatise, in Erdmann's more recent handbook, in the popular history of Curtius, views substantially the same are put forward. Now I would not deny that licentious talk of this kind was probably very prevalent in the polite society of Athens during the age of Socrates and Plato. But the precise point which I, after Grote, maintain, is that such was not the professional teaching of those Professors of the Art of Conduct whom it fell to Socrates to weigh in his formidable balance : that it was not for this that he found them wanting : and that it is a grave misapprehension of his relation to them to conceive him as shielding morality from their destructive analysis, and reaffirm- ing the objectivity of duty in opposition to their " Absolute Subjektivitat." The indictment thus sweepingly drawn against a profession proceeds upon two lines of argument. It appeals to the evidence of contemporary authority, especially Plato : and it is further supported on a presumption drawn from the metaphysical doctrines believed to have been held by the Sophists. It will be convenient to take the two arguments separately : accordingly, in the present paper, I shall confine my- self entirely to the first. The only testimony which it is worth our while to 358 TiiK sorinsTs considor :it loiii^th is that of IMato. Aristotle's kiiowloilij:*' «»f tlio coiitemporiiries of Socrates must have been entirely second-hand : and indeed what he says of the Sophists must be taken to refer chiefly to what I have ventured to call post-Socratic Sophistry — the Eristical disputation which I conceive to have been chiefly imitated iVoni Socrates, and to have borne at any rate less resemblance to the rhetorical moralising of Prota2;oras and Prodicus than it did to the dialectic of Socrates. Obviously we can make no use of the evidence of writers like Aristophanes and Isocrates, who lump Socrates and his opponents together under the same notion. And though Xenophon does not, of course, do this : still his conception of sophistical teaching is evidently of the vaguest kind. He probably would have included under the term physical theorists like Anaxagoras, for we find him speaking of "the Cosmos, as the Sophists call it." So that we cannot refer with any confidence to his description of the class generally, but only to the notices that he gives of particular individuals. The most important of these is an account of a dialogue between Socrates and Hippias, which is noticed below : he further represents his master as borrowing from Prodicus the well-known fable of the Choice of Hercules : and this together with other testimonies has led to the general acquittal of Prodicus from the charges brought against his colleagues. But the main part of our historical investigation must turn upon the Platonic dialogues. Those in which the Professors of Conduct THE SOPHISTS 359 appear or are discussed are chiefly the Hippias Major and Mino7' (if we admit the genuineness — or veri- similitude — of the former), and the Protagoras : the Meno, Gorgias, and Republic. I have tried to show that in the Sophista and Euthydemus the Sophist is a teacher of an entirely different type. And of the six dialogues above mentioned I think it may be fairly contended that the three former are most likely to represent the actual relation of Socrates to the ethical teachers of his age ; for they are no doubt the earlier, and the obvious aim of each of them is to exhibit Socrates in controversy with Sophists : whereas in the Meno the Sophists are only mentioned incidentally ; the polemic of the Gorgias is directed primarily against Rhetoricians, and the Republic is chiefly constructive and expository. Now suppose a person to know no more than that there were in Athens certain clever men whose teaching was dangerous, as being subversive of the commonly received rules of morality, and tending to establish egoistic maxims of conduct : and suppose that with this information he is set down to read the three first -mentioned dialogues. He is introduced to Hippias, Protagoras, and Socrates. Hippias has com- posed an apologue in which he makes Nestor recommend to Neoptolemus the different kinds of conduct that are considered Noble or Beautiful : Socrates, by ingenious questioning, reduces him to helpless bewilderment as to the true definition of the term koXov. Again, Hippias has lectured on the contrast between the veracious Achilles and the mendacious Ulysses : 300 'I'HK SOVHISTS Sornitos with similar ingenuity urgues lliat wilful meutlacity or wilful wrong-doiug generally is better than ignorance and involuntary error: Hippias pro- testing against the dangerous paradox. Again, he finds Protagoras explaining how it is that any plain man i.s, to a certain extent, a teacher of Virtue, having knowledge of the chief excellencies of conduct, and being able to communicate them to others : a Professor of Conduct is only a man who knows and teaches what all plain men know and teach, in a some- what more complete and skilful manner. Socrates, on the other hand, argues that all Virtue resolves itself into a method of calculating and providing the greatest possible pleasure and the least possible pain for the virtuous agent. Can any one doubt that such an unprejudiced reader would rise from his perusal of the three dialogues with the conviction that Socrates was the Sophist as commonly conceived, the egoist, the ingenious subverter of the plain rules of morality? And though perhaps even at this point of his studies (and certainly when he had read a little further) he would decide that Socrates was not really a '' corrupter of youth," he would see no reason to transfer the charge to Protagoras or Hippias. He would see that Socrates attacked their doctrines not as novel or dangerous, but as superficial and common- place. Impostors they might be, in so far as they pretended to teach men what they knew no better than their pupils : but if they knew no better, they knew no worse : they merely accepted and developed the commonly received principles. Ajid thus — to THE SOPHISTS 361 come to the later dialogues to which I have referred — one finds that Socrates even half defends them in the Meno against the popular odium which he shared with them : Anytus is made to confess, that whatever blame they may deserve, his own abuse of them has been uttered in mere ignorance. So again in the Republic, where Plato's satire takes a bolder sweep, there is a sort of indirect and latent defence of the Sophists against the charge on which Socrates sufifered as their representative. Plato clearly feels, that whatever quarrel Philosophy might have with the Sophists, Demos had no right to turn upon them : Demos himself was the arch-Sophist and had corrupted his own youth : the poor Professors had but taught what he wanted them to teach, had but conformed to the common manner and tone of thought, accepted and formulated common opinion. Nor is the view of ' Sophistik ' presented in the Gorgias really diflferent, though it has been differently understood. No doubt it is a ' sham Art of Legislation,' it does not give the true principles on which a sound social order is to be constructed : but that is not because it propounds anti-social paradoxes : rather, it ofiers seeming-true principles, which fit in with the common sense of practical men. It is said, however, that there are other passages in Plato which clearly exhibit the anti-social tend- encies of the Sophistic teaching : and that especially in the last two dialogues to which I have referred such evidence is to be found. Let us proceed to examine these passages in detail. 362 THK SOl'HISTS Tlio most oomprrluMisivo luul pregnant fonmil:i in whii'h this anli - social teaching is tlionght to be suninuHl up. is tliat to ^iKaioi', justice, or social duty genenilly, exists v6/j.(p only, and not <\)v(T€i. It is clear from the references in his Ethics, etc., that Aristotle found this doctrine very widely held by his predecessors : and we should draw a similar inference from a well-known passage in Plato's Laws (x. pp. 889, 890), where he speaks of " the wisest of all doctrines in the opinion of many . . . that the honourable is one thing by nature and another thing by law, and that the principles of justice have no existence at all in nature, but that mankind are always disputing about them and altering them." The commentators do not hesitate to treat thciSe passages as referring to the Sophists : in fact, they make the reference in such a matter-of-course manner, that one is startled to find how entirely unauthorised it is. Aristotle's allusions are quite general : and Plato simply says that these are " the sayings of wise men, poets as well as prose-writers." This no doubt does not prove that he is not referring to the Sophists : but when we consider that it is the great assailant of Sophistry who is speaking, it seems pretty strong negative evidence. It is said, how^ever, that other passages in Plato show so clearly that the doctrine was actually held by the Sophists, that there was no reason why he should mention them by name in the Laws. It is said (l) that Hippias in the Protagoras draws precisely the same distinction between v6fjbo<; and (f)vac^, and that Plato's testimony is here confirmed THE SOPHISTS 363 by Xenophon (Mem. iv. c. 4) ; (2) that Callicles in the Gorgias employs the same antithesis as a quasi- philosuphical defence of his cynically avowed immor- ality : (3) that Thrasymachus in the Republic puts forward a view of justice coinciding substantially with that of Callicles, though not couched in the same language. This cumulative evidence seems at first sight very strong : but I think that on a closer exam- ination every part of it will be found to break down. In the first place, it must be observed that the mere adoption or bringing into prominence of the distinction between the ' conventional ' and the ' natural ' as applied to the laws and usages of society is no evidence of egoistic, anti-social disposition or convictions. Rather, we may say, is the recognition of such a distinction an obvious and inevitable incident of the first beginnings of philosophical reflection upon society, especially in an age of free and active mutual communication among a crowd of little States differ- ently organised and mostly in a state of rapid change. And the natural effect of such recognition upon an ordinary mind, sharing in the ordinary manner the current moral sentiments and habits of its society, is rather an endeavour to separate the really sacred and stringent bonds, the fundamental and immutable principles of social behaviour, from what is conven- tional and arbitrary in positive law and custom. And it is just in this attitude of mind that Hippias appears in the dialogue with Socrates that Xenophon records. After some characteristic sparring, Socrates has defined the Just to be the Lawful. This surprises Hippias, :}64 TllK SOPHISTS '■ I>i) yon uwiui they are identical?" he answers, "1 do not quite understaml how you use the words . . . how ean one attribute niueh intrinsic worth to laws, when their makers are continually changing them ? " That is, Justice in Ilippias' view is therefore not to vofu/xov, because it must be a-vovBaLorepov TTpdyfia. And the few sentences in the Protagoras in which the Professor's style of lecturing is somewhat broadly caricatured are quite in harmony with Xenophon's account : and indeed would suggest this view rather than the other if taken alone. With Callicles the case is quite different. His use of the antithesis of ^uat? and i^o/xo? is no doubt flagrantly immoral : an open justification of the most sensual egoism. The only lacuna in the argument here — and it seems to me a sufficiently large one — is that Callicles is not a Sophist, and has no obvious connexion with Sophists. " No matter," say Zeller and others, " he must be reckoned a representative of the Sophistische Bildung." Now here a distinction must be taken, the importance of which I shall presently urge at more length. If by ' Sophistische Bildung ' is merely meant what German writers com- monly call the ' Aufklarung,' or rather the frivolous and demoralising phase of the ' Enlightenment ' diffused through polite society in this age, the negative and corrosive influence which semi-philoso- phical reflection upon morality has always been found to exert — this is no doubt represented in Callicles. But if it is meant that Plato intended to exhibit in Callicles the result, direct or indirect, of the teaching THE SOPHISTS 365 of our Professors of Conduct : then I can only say that he dissembled his intention in a way which contrasts strikingly with the directness of his attack in other dialogues. For Callicles is not only nowhere described as a friend or pupil of Sophists : but he is actually made to express the extremest contempt for them. " You know the claims," says Socrates, " of those people who profess to train men to virtue." " Yes, but why speak of these empty impostors {dvdpcoTTociv ov8evo<; d^icov)," replies Callicles. Certainly we have here a most unconscious ' representative.' It is said, however, that Aristotle speaks of Callicles as a Sophist, or at least as a Sophistical arguer : and that, in respect of his use of this very antithesis. The passage referred to is Sophist. Elench. xii. 6. Both Sir A. Grant and Mr. Cope interpret it in this way : and as Aristotle's authority on such a point cannot be disregarded, we must consider the passage carefully. Sir A. Grant introduces it as follows : ^ — " One of the most celebrated ' points of view ' of the Sophists was the opposition between nature and convention. Aristotle speaks of this opposi- tion in a way which represents it to have been in use among them merely as a mode of arguing, not as a definite opinion about morals. He says {Soph. El. xii. 6), ' The topic most in vogue for reducing your adversary to admit paradoxes, is that which Callicles is described in the Gorgias as making use of, and which was a universal mode of arguing with the ancients, — namely, the opposition of ' nature ' and ' Ethics of Aristotle, vol. i. p. 107 [2nd edn. 1866 (p. 148, 3rd edii. 1874)]. 3t)C THE SOPHISTS * foii volition ; for tlirsc arc maintained to be con- traries, ami thus justice is riii;lit according to conven- tion, but not according to nature. Ilence they say, when a man is speaking witli reference to nature, you should meet him with conventional considerations ; when he means ' conventionally,' you should twist round the point of view to * naturally.' In both ways you make him utter paradoxes.'" Now the words which are here rendered " that which Callicles is described in the Gorgias as making use of" are uxnrep koI 6 KaXXt«:X>}? €v rw Topyla yiypairraL Xeycov. But what is " Callicles in the Gorgias described as saying " ? Is he " reducing his adversary to admit paradoxes " ? On the contrary, he is complaining of this procedure on the part of Socrates. to? ra ttoWcl 8e, he says, TaOra ivavTia aXXi]Xoi<; iarlv, >; re vai<; koI o vofiof;. iav ovv Tt? ala^vvrjTai kol firj roXfia Xeyetv anep voei, avayKa^erai evavTLa XeyeLV. o S?) Kal av tovto to aotpov Kara- vevorjKOi's KaKovpy€i<; ev Tolva€o)<;, ra tov vofiov. It is Socrates who is the Sophist, or at least is charged with Sophistry : and Aristotle, intent on his subject, and not thinking of the reputa- tion of Socrates, has simply quoted the passage as a good illustration of a particular sophistical topic. This piece of evidence therefore turns out most unfor- tunately for our opponents. It incidentally illustrates that close affinity between the later, Eristic, Sophistry, and the teaching of Socrates, which it was the object of my former paper to exhibit : but it has nothing THE SOPHISTS 367 whatever to do with the morals of Callicles or their origin. When we attempt to speak exactly of the relation of Callicles to ' The Sophists,' the necessity of dis- tinguishing the different meanings of the term So^L(nrjSL' ixi'()iiKi;i:x('i<: OF KMriKUWL I'liiLosoriiv absoluto univorsality wii li wliidi lliey arc comnioiily accepted, :iri' lortanily not based upon expcrieuee. I do not now dispute the empirical arguments used to prove that these beliefs, when duly restricted, have really a solid empirical basis — as, for instance, if we believe not (as common sense holds) that a straight line is always the shortest line between its extremities, but merely that it is so in the space with which we are familiar. But such modifications of current beliefs implicitly accuse common sense of error too extensive to leave its guarantee philosophically trust- worthy : so that it becomes impossible in strict philosophical reasoning for an Empiricist to start with assuming the validity of what is commonly taken as knowledge. We may allow him to accept for practical purposes whatever is believed by "every sensible man " or " every one with the least know- ledge of physical science " ; but he must not introduce in philosophising propositions guaranteed by this kind of warrant alone. This seems so plain that I need not enter into further difficulties involved in the acceptance of the criterion of General Consent, — as that the consent of the majority to science and history is ignorantly given, or not really given at all ; that the consent of one age and country differs from that of another, and that in past ages the criterion would have certified many doctrines that we now reject as erroneous and superstitious, etc., — especially since these con- siderations have been forcibly urged by more than one empirical philosopher. In fact, empirical philo- INCOHERENCE OF EMPIRICAL PHILOSOPHY 383 sophers do not, for the most part, appeal expressly to the criterion of General Consent, so far as their philosophical procedure is concerned. If formally- asked what the cognitions are which they assume to be true in the reasoning by which they establish the empirical criterion, they would usually answer (2) that they assume, first, what is immediately known, or what we are immediately conscious of, and, secondly, whatever may be cogently inferred from this. The second part of this answer has been frequently attacked ; and it certainly appears to me that no perfectly cogent inference is possible on strictly empirical principles ; because no cogent inference is possible without assuming some general truth, the validity of which cannot itself be guaranteed by any canon of cogent inference. But the assumption of the validity of immediate cognitions seems to me equally open to attack ; and it is to this that I now wish specially to direct attention. I must begin by removing an ambiguity in the term 'immediate.' When an Empiricist speaks of a cognition as ' im- mediate ' he must not be understood to mean that it has not among its causes some antecedent psychical or physical phenomena — some feelings, or some move- ments of the matter of the organism of the cognising individual ; for no empiricists maintain that any cognitions or any other mental phenomena are un- caused ; and if they are caused at all, they must stand in the relation of eflfect either to psychical or physical phenomena, or to both combined. The 384 IXCOHEKENCE OF EMPIRICAL PIIILOSOPKY ' medititiou ' that is excluded by terming any cog- nition * immediate ' must therefore be logical media- tion or inference. If then it be asked, why should we make the general assumption that error is absent from non- inferred cognitions and from these alone, the answer would seem to be, first, that immediate knowledge carries with it its own warrant ; that when we immediately know we also, by a secondary insepar- able act of the mind, — generally latent but becoming explicit if any doubt is raised, — know that we know certainly ; and, secondly, that we have no experience of error in non-inferred cognitions ; error being always found to come in through inference. But it is practically of no avail to say that immediate cognition is infallible, unless we have a no less infallible criterion for ascertaining what cognitions are immediate : and the difHculties of ascertaining this are profound and complicated. Are we to accept each man's own view of what he im- mediately knows ? This certainly seems in accord- ance with empirical principles, as all experience must be primarily the experience of individual minds. But if we take, unsifted and uncriticised, what any human being is satisfied that he or she immediately knows^ we open the door to all sorts of mal-observation in material matters, and to all sorts of superstition in spiritual matters, — as superstitious beliefs commonly rest, in a great measure, upon what certain persons believe themselves to have seen, heard, or otherwise personally experienced. And, in fact, no empiricist INCOHEKENCE OF EMPIRICAL PHILOSOPHY 385 adopts this alternative ; there is no point upon which empirical philosophers are more agreed than on the incapacity of ordinary persons to distinguish their immediate from their mediate knowledge. Shall we, then, say that we take each man's experience so far as it commends itself to other men ? But if we mean ' other men generally,' this is only our old criterion of General Consent, in a negative instead of a positive aspect, and the acceptance of it would therefore bring us round again to the difficulties already discussed ; with this further difficulty, that it is hard to see why, on empirical principles, any one man's experience stands in need of being confirmed by that of others. I do not see what right an empiricist has to assume that one man's immediate cognitions ought to coin- cide with the immediate cognitions of others ; still less, that they ought to coincide with their inferences. And if empiricists do not trust common men's judg- ment as to their own immediate knowledge, they can hardly put them forward as trustworthy judges of the immediate knowledge of others. It may, however, be said that to distinguish accurately immediate from mediate cognitions re- quires a skill beyond that of ordinary men, only attainable by training and practice : that, in short, it requires the intervention of psychological experts. This seems to be the doctrine of James and John Mill, and, in the main, of the school of which they, with Mr. Bain, are the founders ; but, in my opinion, it is open to several fatal objections. In the first place, I do not see how even an expert can claim to 2c ;>8o INOOlIKlJKXnE OF KMrilUCAL PlTTLOSOrHY kiuuv allot luT man's ininiccliate kuowledixc; without assuming that till human minds arc similarly consti- tuted, in respect of immediate cognition ; and 1 do not see how this assumption is legitimate on empirical principles. And this difficulty is increased when we consider that the psychological expert, if he is an Empiricist, has to throw aside as untrustworthy the affirmations, as to their own immediate knowledge, of thoughtful persons who have given much attention to the subject — I mean the Intuitional Metaphysicians, who say that they immediately know universal truths. If we admit these to be experts, I do not see how we can hope to establish the cardinal doctrine of Empiri- cism. Yet how can w^e exclude them, except by assuming the empirical philosophers to be the only real experts? — and this seems hardly a legitimate assumption in an argument that aims at proving the empirical philosophy to be true. Nor is it any answer to this objection to show that Intuitional Metaphysicians have in certain cases affirmed as immediately known propositions that are not true ; since the question is not whether error is incident to non-empirical cognitions, but whether we may legiti- mately assume that it is not incident to empirical cognitions. But further, even supposing that we only recognise, as experts in discriminating immediate knowledge, persons who will not allow anything to be immedi- ately known, except particular facts, serious difficul- ties still remain ; because we find that these experts disagree profoundly among themselves. We find — INCOHERENCE OF EMPIRICAL PHILOSOPHY 387 not to speak of minor divergences — that there is a fundamental disagreement between two lines of empirical thought which — if I may com a word for clearness' sake — I will call respectively materialistic and tnetitalistic. When a Materialistic Empiricist affirms that physical science is based upon experience he means that it is based on immediate knowledge of particular portions of something solid and extended, definitely shaped and sized, moving about in space of three dimensions. Whether he regards this matter as also coloured, resonant, and odorous, is a more doubtful question ; but probably he would say that colour, sound, and odour are effects on the mind — or perhaps on the brain ?— of the molecular movements of material particles. I can hardly profess to give a consistent account of his views on this point, if he is a thorough-going materialist, but it is enough for my present purpose that he at any rate believes himself to know immediately — through touch, if in no other way — matter with the qualities first mentioned. The Mentalistic Empiricist, on the other hand, maintains that nothing can be immediately known except mental facts, consciousness or feeling of some kind ; and that if we are right in assuming a non- mental cause of these mental facts — which he is generally inclined to doubt — we must at any rate regard this cause as unknown in every respect except its mere existence, and this last as only known by inference. How, then, is Empiricism to deal with this dis- agreement ? It cannot be denied to be rather serious; .SSS INC'OIIKRKNC'K OF KMIMKKWL I'HILOSOrHY siiu'O, tluniijjh nintorialisni has [)l(Mity of support among philosophising men of st-ienco, the tendency of the main line of English cMnpirical philosophy, from Locke downwards, is definitely towards Mentalism. [ may observe that the more thoughtful Materialists, like Dr. Maudsley, do not exactly say that there are uo mental facts which we may contemplate introspec- tively. But they hold that no scientific results have ever been reached by such contemplation ; and they say very truly that physical science has always progressed by taking the materialistic point of view, and that there is no admitted progressive science of psychology, proceeding by the introspective method, which can be set beside the physical sciences. Hence they boldly infer that there never will be such a science ; and in fact, they are inclined to lump the Mentalists along with Transcendentalists and others, under the common notion of " Metaphysicians " (used as a term of abuse), and to charge them all together with using the Subjective Method, con- demned as fruitless by experience. The Mentalists do not quite reply in the same strain ; indeed, they have rather a tenderness for the Materialists, whose aid, as against Transcendentalism and Superstition, is not to be despised. But they say that the Materialists are inexpert in psychological analysis, and that what they call " matter " is really, when analysed, a complex mental fact, of which some elements are immediately known and others added by inference. In so saying, the Mentalists appear to me to use the term " inference " loosely, and also to INCOHERENCE OF EMPIRICAL PHILOSOPHY 389 fall into the confusion before pointed out between the antecedents (or concomitants) and the elements of a cognition. Certainly I find myself unable to analyse my notion or perception of matter into feelings or ideas of feelings, tactual, visual, or muscular ; though I do find that such sensation-elements present them- selves as inseparable accompaniments of my notion or perception of matter, when attention is directed to it introspectively. But my object now is not so much to enter into this controversy between two sets of Empiricists, as to point out the serious obstacle it opposes to a satisfactory determination of the question what is immediate cognition. Let us suppose, however, that this controversy has been settled to the satisfaction of both parties, in the manner in which some empiricists have tried to settle it. Let us suppose that both Materialists and Mentalists agree to afiirm (l) that we immediately know the external world, so far as it is necessary to know it for the purpose of constructing physical science ; (2) that we immediately know nothing but our own consciousness ; and (3) that these two state- ments are perfectly consistent. It still remains to ask who are the " we " who have this knowledge. Each one of us can only have experience of a very small portion of this world ; and if we abstract what is known through memory, and therefore mediately, the portion becomes small indeed. In order to get to what "we" conceive "ourselves" to know as "matter of fact" respecting the world, as extended in space and time — to such merely historical know- 300 TXrOTIKKEXOK OF K^fPlTJHWI, rillLOSOrHY ledfje as wi* (•onnnoiilv nuMnl not as "rcstinjjf" on expericiu'o, but as fonstitutiiig the exj)ericncc on whioli scionce rests — we must assume the general trustworthiness oi" memory, and the general trust- worthiness of testimony under proper limitations and conditions. 1 do not for a moment sav that we have no right to make these assumptions ; 1 only do not see how we can prove that we have such a right, from what we immediately know. At this point of the argument Empiricists some- times reply that these and similar assumptions are continually " verified " by experience. But what does " verified " exactly mean ? If it means " proved true," I challenge any one to construct the proof, or even to advance a step in it, without assuming one or more of the propositions that are to be verified. What Empiricists really mean, I conceive, by " veri- fication " in this case is that these assumptions are accompanied by anticipations of feelings or percep- tions which are continually found to resemble or agree with — though not identical with — the more vivid feelings of perceptions which constitute the main stream of consciousness. Now, orrantino^ that such resemblance or agreement may be immediately known, I yet cannot see that anything is gained towards the establishment of the cardinal doctrine of Empiricism. For there is a similar agreement between actual experience and the anticipations accompanying all the general propositions — mathe- matical, logical, or physical — which philosophers of a different school affirm themselves to know im- INCOHERENCE OF EMPIRICAL PHILOSOPHY 391 mediately ; so that this " verification " can hardly justify one set of assumptions, as against the other. If, finally, the reader who has got through this paper should say that my cavils cannot shake his confidence in experience, or in the aggregate of modern knowledge that has progressed and stiJl progresses by accumulating, sifting, and system- atising experience — I can only answer that my own confidence is equally unshaken. The question that I wish to raise is not as to the validity of received scientific methods, but as to the general epistemological inferences that may legitimately be drawn from the assumption of their validity. It is possible to combine a practically complete trust in the procedure and results of empirical science with a profound distrust in the procedure and conclusions — especially the negative conclusions — of Empirical Philosophy. A DIALOGUE ON TIME AND COMMON SENSE (Reprinted from Mhxd, vol. iii. N.S., October 1894.) I WAS interested in a conversation that I had, a short time ago, with a Russian Professor of Philosophy — who, I ought to say, spoke English with a fluency rarely attained out of Russia. What interested me in our talk, when I came to think it over, was the peculiarity that while it ranged rather widely it was almost entirely occupied with the effort to explain our views each to the other, with hardly any aim at either confuting or convincing, and no sense of a cause that had to be defended or a school that might be attacked. He had never read my books and I had never read his : he was on his travels, curious to know what we thought in England : I was also curious — though perhaps not equally — to know what was thought in Russia : time was short, and as I have never myself been convinced of anything im- portant in half an hour, I never expect to convince any one else in that limited space. But when I tried to write down the talk I found I had forgotten too much of it : if I aimed at exactness, the result would 392 DIALOGUE ON TIME AND COMMON SENSE 393 be meagre and uninteresting ; so in what follows I have allowed imagination to supplement the defects of memory, merely trying to preserve the general attitude of our minds towards each other, and the general impression that my visitor had given of his philosophical position. The talk bes^an with an account of his recent visit to America, where he had been for some months : he had been much impressed with the activity with which philosophical and psychological studies were being developed there, and the wide range and diversity of their development. One set of minds were working with transatlantic energy at the minutest problems of psychophysics, in the psychological laboratories that have sprung up like mushrooms during the last ten years or so : another set were agitating the largest questions of speculative philo- sophy : and my visitor's admiration seemed to be equally divided between metaphysicians and experi- mental psychologists. While we were thus chatting about academic institutions and persons in America, he suddenly said, " Excuse me, but there is a question I always ask of a philosopher, which perhaps you will not mind answering. What do you think really exists ? " My first impulse was to borrow Hegel's famous answer to Cousin, when the Frenchman asked him for a succinct account of Hegelianism. But I remembered that earlier in our talk my guest had permitted himself a mild complaint of the reserve of Englishmen, as contrasted with the communicative- :?9-i i>iAi.<)(;rK ox timk and common sknsk noss of his Ainoricaii iViciuls. So, Irelino- tli:it our reputation tor iiitcniat ioiial conlialilv was at stake, my secoiul impulse was to gain time "No doubt," I said, "you ])iit this cpustion to your American friends." ■'Oh yes," said he, " And wliat tlid tliey answer?" '' Well," he said, " it is difficult to remember all their answers. But I think that a majority of those whom I persuaded to take an interest in the question were of opinion that God is the one ultimate reality." " But did they all mean the same thing," said I, '" or may we not rather invert the oft-quoted Greek phrase — TToWmV OVOfXaTCOl^ /jL0p(f>7] /Mia — and say that, in current thought, ' God ' is one name for many and diverse ideas ? " I thought this might be a successful diversion, as the topic seemed both wide and attractive. But I had overshot my mark ; it was too obvious an invitation to go off into infinite space ; and declining this, he returned to the charge and reminded me that I had not answered his question. Well, there was no help for it, but I thought I saw still a way of gaining time. " Do you mean," I said, " what really exists now ? or do you include what has existed and what will exist ? " " Ah," said he, " but that is a part of the question I am asking you. Do you think that the past really exists ? " DIALOGUE ON TIME AND COMMON SENSE 395 " Well," I said, " one has to distinguish different modes of real existence. It would be absurd to say that the great study of History is not conversant with reality. So far as the historian attains truth — as doubtless he does in some degree — the past exists for him as an object of thought and investigation : but so far as it is past it has ceased to exist in the sense in which the present exists." "Ah," said he, brightening, "then in spite of Kant you think Time really exists as a condition of tilings, and not merely as a form of perception. Why, I thought that even your empiricists and your scientists all held now that science only deals with phenomena, and that Time is only a sum of relations among phenomena." " I think," I said, " that you must not take our men of science too much au pied de lettre when they talk of a 'phenomenon.' For instance, I was referring to a text-book on physics the other day, and I found ' a phenomenon ' defined as ' any change that takes place in the condition of a body.' I think scientific men commonly mean by ' phenomenon ' a real event that occurs in real time : they call it a phenomenon, only because the real event as conceived by their science is something other and more than the event as first perceived through the senses." " Then," he said, " you think Time really exists, and you can conceive Time pure and simple, apart from the changes that make up experience." " I have not said that," I replied, " but I certainly distinguish it in thought from the changes : — for I ;^96 DIALOGUE ON TIME AND COMMON SENSE lun conceive any particular scries of changes going quicker or slower, and occupying more or less time : and that conception would be impossible if I did not distinguish the course of time from the course of change." " Well," he said, " I have no wish to prove Time unreal : for the most real thing to me is my own existence : and though as a thinking, knowing being 1 can think myself out of Time, I admit that I can form no idea of myself as a living, feeling being except under the condition of Time. And perhaps my life is, on the whole, more interesting to me than my knowledge. But still — there are the antinomies. How do you get over the antinomies ? Can you help me to conceive either a beginninf]: of Time or an infinite past — a ' finished infinite ' as Kant says, — or any tertium quid ? " " No," I said, " I am afraid I cannot help you over that stile. I admit that these alternatives are at present both inevitable and inconceivable to me, and 1 infer from this that I do not comprehend past time as a whole. But to conclude, therefore, that Time is unreal seems to me — what is the German phrase ? — to be 'throwing out the child in emptying the bath.' If Time is unreal, succession is unreal : and if succession is unreal, the interest of the study of the past is destroyed." "Are you not forgetting," he said, "that Kant's solution of the antinomies is critical and not sceptical, and leaves ample room for the scientific study of past experience, in order to discover the general DIALOGUE ON TIME AND COMMON SENSE 397 laws of the empirical world? Surely the particular succession of past events is of no interest except as a basis for scientific generalisation : the study of them is only of practical value, so far as it enables us to grasp the present and foresee the future by the ascertainment of general laws. And surely, so far as we get hold of these general laws, we have a grasp of reality which remains unimpaired, even if we grant that the element of Time in our conception of these laws is due to the necessary form of our apprehension and does not belong to the reality of things." " I admit the force of what you say," I replied, " so far as the empirical laws with which physics and chemistry deal are concerned ; though by the way I do not think the Kantian theory will explain why we succeed — so far as we do succeed — in discoverinsf these laws. Kant explains ingeniously why we inevitably seek for the causes of phenomenal change, but not why we find them. However, putting this aside, and granting all you say, I do not think the interest of human history is saved by it. For the interest of human history lies not merely in the general laws of change that we can discover in it, but in the general fact of progress through stages each different from the one before. If Time is unreal progress is unreal, and if progress is unreal the interest goes." " Still surely," he said, " the important point for practice is that we should discover the general laws of social change and be able to foresee what is coming." :?9s i>iAi.(u";rE o\ TiAFK AND (Mi:\r:\r()X sknsh •' Wi'll. " 1 said, " 1 \vill follow vou into the rcL^ion o( pnu-tioe. Surely all our notions of practice become unnieaninu^ if you suppose Time to be unreal — a mere form of our apprehension. 1 always feel this in reading Kant. So long as he is engaged with his destructive work I can get on with his ' things in themselves': but when he tries to become construc- tive on the basis of moral experience I feel that all the fundamental conceptions he uses — the conceptions of rational action, springs of action, means and ends and so forth — become altogether unmeaning if his view of Time be accepted. The real man, in Kant's practical philosophy, seems to me a being who, in an unintelligible position out of Time, makes an absolutely incomprehensible and unaccountable cJioice of partial irrationality. A more uuexplanatory explanation of the mystery of our fallen nature it is impossible to conceive." " I agree," said he, "that Time is indispensable to my notion of human action — and human life gener- ally. But the case seems to me quite otherwise with knowledge. The knowing subject, that combines experiences in Time and Space and so makes a world — surely we necessarily conceive that out of Time. Time belongs to the object of knowledge, and there- fore not to the knowing subject as such." " Let me see," I said : " Time is an object of my thought, therefore the subject of thought is not in Time. Is that the argument ? " " Something like it," he said ; " an object or condition of the object." DIALOGUE ON TIME AND COMMON SENSE 399 "Suppose," I said, "that we consult your American friends who say that the ultimate reality is God. God then is an object of thought — the object of thought — to each of these philosophers ; yet surely no one would say that he was therefore out of God. You, on the other hand, say that self is to you the most real existence ; in thinking this you make yourself an object of thought, but you are not, therefore, out of yourself Why are you any more out of Time ? " " I don't think the cases are analogous," said he : " at any rate, I do not find that your argument convinces me. For my own part, I am not a Pantheist, because — as I said — what is most certain to me is my own existence as an individual ; and though I know I am not the whole of things, I cannot feel sure that all the rest is God. But still less am I an atheist : for when I consider my relation as a thinking being to Truth, I find myself irresistibly led through Finite Thought to the conception of Infinite Thought, and so to an Infinite Thinker of Infinite Truth, of which the truth apprehended by me is only an infinitesimal part. Now truth is essentially unchangeable, otherwise it would not be truth — though it may relate to things subject to change, — hence as Time is essentially changing, in laying hold of truth I carry myself out of Time, and accordingly I have to conceive God, the Infinite Thinker, as essentially out of Time." While he was speaking, I took out my watch. "You say," I answered, "that you are more certain too 1)IAL()(UTK OX TIME AND COMMON SENSE of your own existence than of Hiiythino- elsr. Well, 1 am as rertaiii as I am of my owu existeuce that my ideas about Truth, Infinite Thought, Infinite Thinker, as avowed by your words, have occurred in succession between five and six minutes past three on the 20th of April 1894 — or at some other definite point of time, for my watch is not infallible, — and, further, that these ideas would not have been what they actually were, had they not had as essential ante- cedents other ideas which have occurred before at definite points of time. Granting that Truth is not subject to change, my intellectual life is as much subject to it as any other element of my life." "Well, but," said he, "what do you say of God's existence ? " " I say as little as I can," I replied, " under this head ; since the relation of God to Time is one of the things that I do not understand." " In short," he said, " you do not believe in a Divine Being out of Time." " I have not said that," I rejoined ; " I am led by the same consideration of Truth that you gave just now — but especially by a consideration of ethical Truth — to regard a belief in a Divine Being as indispensable to a normal human mind ; and though I may not always keep this in mind in philosophical speculation, I was a man before I became a philosopher, and I do not forget it for long " " Well," he said, interrupting, " I have no wish to dispute the correctness of your attitude as a man and DIALOGUE ON TIME AND COMMON SENSE 401 a citizen. But we are talking philosophy now, we are not talking about beliefs practically necessary for the plain man or the good citizen ; and in any case you can hardly say that it is normal to humanity to believe in a God out of Time. The good people who go to church believe in an everlasting deity, enduring through Time, not out of Time." " Yes," I replied, " but I understand that the better opinion — as lawyers say — among students of theology is that the efflux and succession of Time takes place only for finite beings and is not a condition of Divine existence ; and I respect this preponderant opinion, — although I am unable to share it, because what it affirms is to me inconceiv- able. I follow these theologians in conceiving the past and the future as simultaneously present in knowledge to the Divine Mind ; but I am forced to conceive this presence of all the known to the Infinite Kjiower as perpetual, if I would avoid conceiving it at a point of time." " You will pardon me," he said, " the question I am about to ask ; I know some of you English philosophers are anxious to keep in touch with orthodoxy — I found this also in America — and I do not wish to be indiscreet. But, between ourselves, do you think the theologians really know anything about the matter ? " " You need not be afraid of indiscretion," I said, laughing. " For if I were more concerned about my reputation for orthodoxy than is in fact the case, I could still answer your question in the negative and 2d 402 niAlAXUK ON TIMK AND C'OiMMON SKNSE vot claiin tlio support of many hiujiily tulliodox persons, who would cniphatically and piously declare that the human mind was not intended to find an answer to such questions as these, and that to ask tliem was a sign of idle — and perhaps worse than idle — curiosity. Indeed, 1 think the prevailing opinion of theologians at the present time would be in favour of giving these transcendental inquiries a wide berth." " I thought," he replied, " you said that the pre- ponderant opinion was inclined to regard the Divine existence as independent of Time." "I meant," said I, "the preponderant opinion of persons who had thought seriously about the matter ; I never attach importance to a man's judgment on questions he does not care to consider." " Well, but," he said, "you seem to attach import- ance to the movement of what you call the normal mind in these matters ; and if the normal mind of religious persons is moving away from certain questions — it would not aflfect me in the least, but ought it not to influence you ? " " I think it w^ould affect me more," I answered, "if I had not observed that the normal mind seems to move about these questions in a spiral way ; so that the philosopher may avoid too wide a divergence from it, and save himself unnecessary motion, by keeping nearer the axis of the spiral." " That depends," said he, " on the goal he wants to reach." " I think we are agreed," I said, " on his goal, which can be nothing less than to understand the DIALOGUE ON TIME AND COMMON SENSE 403 whole of things. To do this I think he must try to get the whole of our normal thought free from con- fusion and contradiction ; and therefore not ignore the answers given by Theology to any questions he is led to ask, any more than he ignores the answers given by physicists to questions about the material world. For Theology is the result of the eflforts of generations to understand the universe as manifested in the religious consciousness, just as sciences are the results of the similar efifort to understand it as apprehended through sense-perception." " But surely if one finds the answers of Theology confused and contradictory, it is a sign that the method is altogether wrong. You would not surely maintain that there is similar confusion and contra- diction in the fundamental conceptions and methods of physical science "? " "Your former question," I said, smiling, "was not indiscreet, but this one, I am afraid, is ; or is it with deliberate malice that you are tempting me to pro- voke more formidable antagonists — at the present time — than theologians ? But I think I see a pacific way of answering. I think we shall agree that two centuries ago — or perhaps even a century ago — the fundamental notions and methods of natural science had not been brought to the condition of clearness and consistency that they have now reached ; yet surely it would have been unphilosophical then to throw their methods and conclusions aside, and not rather to endeavour to aid in clearing them from confusion and contradiction. And that is how I would 40-1 DIALOiilK OX TIMK AN'D COMMON SENSE tlo;il witli 'riioology now, imd witli other subjects besides 'riuH)logy — for instaiiee, Ktliics and Politics.'' " 1 am not sure," said he, " tluit I understand your view of philosophy. You think it the business of philosophy to put together a number of ditlerent sciences and arts — or whatever you call them. But will they not be an aggregate rather than a whole, and the student a polymath — as we call it — rather than a philosopher ? " " I should not exactly say ' put together,' " I re- plied, " as that would imply that they were not already in intimate and essential relation — and if that were so, the task of the philosopher would doubtless be impossible. I should rather say ' exhibit the essential coherence which is now somewhat latent and obscured in their relations.' The philosopher may not succeed in this, but the polymath — as you call liim — does not try." "Well," he said, "I rather fear that your philosopher will get bewildered and lost in the multiplicity of the bits of his puzzle. I had rather aim directly at the whole : find out and make clear the fundamental conditions of its being a whole for me — my whole, my universe — since I must begin from myself ; and having made this out, then descend to particulars and connect them while distinguishing them by their varying relation to these fundamental conditions." " Well," I said, " the world is wide both for living and for philosophising. I am glad you feel energy enough for this adventure, which grows more daring as the world grows older. Ex Oriente Lux ! " DIALOGUE ON TIME AND COMMON SENSE 405 He looked dreamy but hopeful. Then a thought struck him, and he said, " But I do not see that you have, after all, told me what you think really exists." " Do you not think," I replied, " that it is now time for you to go and ask this question of some other Cambridge philosopher ? " He looked at his watch and assented ; we rose and went downstairs : and as we bent our steps westward through the grounds of the college, I occupied his mind with a series of questions about the academic institutions of Kussia. THE PHILOSOPHY OF COiMMON SENSE AX ADDRESS DELIVERED TO THE GLASGOW PHILOSOPHICAL SOCIETY ON JANUARY 10, 1895 (Reprinted from Mind, vol. iv. N.S., April 1895.) When I received, some months ago, the invitation to address your society, my mind was carried irresistibly back to a period in the last century, in which, through my study of three eminent teachers whose works have had a permanent influence on my thought, I seem to feel more at home in the intellectual life of your famous University than in that even of my own. It is a period of about fifty years ; beginning in 1730, when Francis Hutcheson was summoned from Dublin to fill in Glasgow the chair now worthily occupied by my friend Professor Jones ; and ending in 1781, when Thomas Reid retired from the same chair to put into final literary form the teaching that he had given here for seventeen years. Between the two, as the immedi- ate predecessor of Reid, though not the immediate suc- cessor of Hutcheson, stands the greater name of Adam Smith. I felt " in private duty bound " to select the work of one of the three as the theme of my address : the difficulty was to choose. I should have much 406 THE PHILOSOPHY OF COMMON SENSE 407 liked to try to explain the attraction which the refine- ment, balance, and comprehensiveness of Hutcheson's ethical views have always had for me ; but on such an occasion it seemed prudent to defer to the some- times capricious judgment of history : and in face of that judgment, I felt diffident of my power of per- suading you to regard Hutcheson's system with more than antiquarian interest. With Adam Smith, as I need hardly say, the case was altogether different. His doctrine has gone out into all lands, and his words unto the ends of the world : and hardly a year passes without some attempt being made somewhere to extract fresh instruction from his epoch-making work, or to throw fresh light on its method or its relations. But for this very reason I doubted whether I should not seem superfluous in adding my pebble to the imposing cairn of literary products that has thus been raised to his memory. The intermediate position of Reid, unquestionably a more important leader of thought than Hutcheson, unquestionably less familiar to current thought than Adam Smith, seemed on the whole to fit the opportunity best : I propose therefore this evening to present to you — not with the fulness and exactness of a critical historian, but in the lighter and more selective style allowed to an occasional utterance — such features of Reid's philosophical work as appear to me of most enduring interest. I will begin by endeavouring to remove a prejudice, which perhaps my very title may have produced. " The Philosophy of Common Sense," you may say, " is not this, after all, an intellectual monstrosity ? ■\os Till-: rim,()S()riiv ok common sknse l^hilosoj)hy is a vrood tliiiii^, and Common Sense in its place is a good thini;- too : l)ut they are both better kept apart. If we mix llicm, shall we not find ourselves cutting blocks with a scalpel, and using a garden-knife for the finer processes of scientific dissection ? " And I am the more afraid of this prejudgment, because in the only passage of Kant's works in which he speaks of Reid's philosophical labours, it is this antithesis that he applies in condemnation of them : and, speaking as I do in a University where the leading expositor of Kant, to Englishmen as well as Scotsmen of our age, has taught for so many years, I cannot but feel this condemnation a formidable obstacle to my efforts to claim your sympathy for Reid. The passage I refer to is that in Kant's Prolego- mena to any Future Metaphysic (1783) in which he " considers with a sense of pain " how completely Hume's opponents, " Reid, Oswald, Beattie, and even Priestley," missed the point of Hume's problem. Instead of answering Hume's sceptical reasoning by " probing more deeply into the nature of reason," as Kant believed himself to have done, " they discovered a more convenient means of putting on a bold face without any proper insight into the question, by appealing to the common sense of mankind ... a subtle discovery for enabling the most vapid babbler " without a " particle of insight " to hold his own against the most penetrating thinker. The censure, you see, is strong : but is it thoroughly THE PHILOSOPHY OF COMMON SENSE 409 intelligent ? Reid, says the critic, has not caught Hume's point. Has Kant caught Reid's '? I venture to doubt whether he ever gave himself a chance of catching it. This for two reasons. First, look at the names he puts together, " Reid, Oswald, Beattie " ; — the first a thinker of indubitable originality ; the third a man of real, but chiefly literary, ability, a poet by choice and a philosopher from a sense of duty ; the second a theological pamphleteer. Is it likely that Kant would have thus bracketed the three, if he had really read them ? How came he then to put them on a par ? That is easily explained. He had doubtless read Priestley's examination which treats the three together, and which, written as it was primarily from a theological point of view, gives even a larger space to Oswald. This explains Kant's odd conjunction of names, " Reid, Oswald, Beattie, and even Priestley," — even, that is, their critic Priestley. 1 imagine Kant was on general grounds more likely to be attracted by Priestley's book than by Reid's, since he had a keen interest in the progress of contemporary physical science, and Priestley had here a well -deserved reputation : and certainly the Reid who appears in Priestley's pages, misquoted, misrepresented, and mis- understood, was likely enough to be regarded as another Oswald. My second reason is that if Kant had ever studied Reid's Inquiry into the Human Mind he could hardly have failed to extend his studies to the Hume to whom Reid was replying. This may startle you. -no TiiK riiii.osoriiv ok common sense " \\ hat, " yt)ii mav say, " Kant, not ivad ilunie : wliy, any shillin*; liaiulbook of the history of pliilosophy will tt>ll vou that Hume's scepticism woke up Kant from his dogmatic slumbers." Certainly, but it was not the same scepticism as that which woke up Reid to construct the Philosophy of Common Sense : it was the veiled, limited, and guarded scepticism of the Inquiry into the Human Understanding, not the frank, comprehensive, and uncompromising scepticism of the Treatise on Human Nature. Kant's Hume is a sceptic who ventures modestly to point out the absence of a rational ground for his expectation that the future will resemble the past, while in the same breath hastening to assure the reader that his expectation remains unshaken by his arguments. Reid's Hume is a sceptic who boldly denies the infinite divisibility of space, who professes to have in his intellectual laboratory a solvent powerful enough to destroy the force of the most cogent demonstration, and who ventures to tell his fellow- men plainly that they are each and. all " nothing but bundles of different perceptions, succeeding each other with inconceivable rapidity." I think that if Kant had even looked into Reid's Inquiry, the difference between the earlier and the later Hume must have struck him, and he must have been led on to read the Treatise on Human Nature ; whereas it is evident and admitted that he never did read it. Do you still want proof that Kant did not catch Reid's point ? I have a witness to bring forward whom Kant himself would have allowed to be a good THE PHILOSOPHY OF COMMON SENSE 411 witness — Mr. David Hume : who was persuaded by a common friend to peruse parts of Eeid's work before it appeared, and to write his view of them to the author. Hume did not much like the task in prospect. " I wish," he grumbles to the common friend, " that the parsons would confine themselves to their old occupation of worrying one another, and leave philosophers to argue with moderation, temper, and good manners." In fact, he expects another Warburton : but when he has read the MS. his tone changes. "It is certainly very rare," he writes to Reid, " that a piece so deeply philosophical is wrote with so much spirit, and affords so much entertain- ment to the reader. . . . There are some objections," he goes on, " that I would propose, but I will forbear till the whole can be brought before me. I will only say that if you have been able to clear up these abstruse and important topics, instead of being mortified, I shall be so vain as to pretend to a share of the praise : and shall think that my errors, by having at least some coherence, had led you to make a strict review of my principles, which were the common ones, and to perceive their futility." Well, I think you will agree with me that this is a charmingly urbane letter, from a freethinker of established literary reputation to a parson turned professor, as yet hardly known in the world of letters, who had hit him some smart blows and ventured to laugh at him a little as well as argue with him. But Hume recognises that the parson unexpectedly writes like a philosopher : and Hume, 412 TllK IMIII.OSOIMIV OV COMMON SENSE as we saw, lias a lii^li iil<'al of tlic maimer in wliicli philosophers should ooiuluct their debates; and it is a pleasure to liiid him acting up to his ideal, a pleasure all the greater from the rarity with which it is aflbrded to the student of i)hilosophical controversy. But it was not on Hume's urbanity that I wished now to dwell : I wished to jmint out that it never occurs to Hume that Reid has appealed from the expert to the vulgar, and endeavoured to avoid his conclusions without answering; his arojuments. What rather strikes Hume is the philosophic depth that his antagonist has shown in attacking his fundamental assumptions ; — which were, as he says, the common ones, and which Reid accordingly had traced back through Berkeley and Locke to the start of modern philosophy in Descartes. It is difficult, I think, for us to appreciate equally the penetration shown in this historical apergu, because the connexion of ideas that Reid makes apparent now seems to us so obvious and patent. But this is the case with many important steps in the development of philosophical thought : w^hen once the step has been taken, it appears so simple and inevitable that we can hardly feel that it required intellectual force and originality to take it. You remember, perhaps, the depreciatory remark made on Christopher Columbus by a schoolboy who " didn't see why so much fuss should be made about his discovery of America, since, if he went that way at all, he could not well miss it." Similarly it now seems to us that if Reid " went that way at all " he THE PHILOSOPHY OF COMMON SENSE 413 could not fail to find the source of the Idealism of Berkeley and the pulverising scepticism of Hume in Locke's assumption that the immediate object of the mind in external perception is its own ideas : and that finding this view equally in Malebranche, he could not fail to trace it to Descartes. His merit lay in the independence of thought required to free him- self from this assumption, question it, and hunt it home : and this merit Hume evidently recognised. And now, perhaps, I may have persuaded some of my hearers that Kant entirely failed to see what Reid and his followers were driving at. But if so, I have gone too far, and persuaded them of more than I intended. The appeal to vulgar common sense has an important place in Reid's doctrine : he does rely on it : nor can I defend him from the charge that he relies on it too much. He does hold that the mere ridiculousness of Hume's conclusions is a good reason for disbelieving them : and even in his later and maturer treatise he speaks of the sense of the ridiculous as a guide to philosophic truth, in language that lacks his usual circumspection. For our sense of the ridiculous is manifestly stirred by the mere incongruity of an opinion with our intellectual habits : a strange truth is no less apt to excite it than a strange error. When the Copernican theory was slowly winning its way to acceptance, even the grave Milton allowed himself a jest on " the new carmen who drive the earth about " : and I can remember how, when the Darwinian theory was new, persons of the highest culture cracked their jokes on 41 I THK rillLOSOrTTY OF COMMON SKNSK llu> zoologist's supposed private reasons Tor tlu' alisurd couclusioii that liis ancestor was a monkey. Ami this is doubtless all for the best : laui;liter is a natural and valuable relief in many perplexities and disturbanees of life, and 1 do not see why it should not relieve the disturbance eausetl by the collision of new opinions with old : only let us remember that it is evidence of nothing except the mere fact of collision. But, though Reid does rely more than he ought on the argumentum ad visum, he is not so stupid as to think that a volume is required to exhibit this argument. He does say to the plain man, " If philosophy befools her votaries, and leads them into these quagmires of absurdity, bew^are of her as an ignis fatuus " : but he immediately adds, " Is it, however, certain that this fair lady is of the party ? Is it not possible that she may have been misrepre- sented ? " and that she has been misrepresented is the thesis which he aims at proving. In the course of the proof, no doubt, he leads us again to Common Sense, as the source and warrant of certain primary data of knowledge at once unreasoned and indubitable : but the Common Sense to which we are thus led is not that of the vulgar as contrasted with the philosopher : Reid's point is that the philosopher inevitably shares it with the vulgar. Whether a philosopher has been developed out of a monkey may possibly be still an open question ; but there can be no doubt that he is developed out of a man ; and if we consider his intellectual life as a whole, we may surmise that the THE PHILOSOPHY OF COMMON SENSE 415 larger part of it is occupied with the beliefs that he still shares with the unphilosophical majority of his contemporaries. It is on this fact that Reid's appeal to him is based. He refers to Hume's account of the manner in which, after solitary reflection has environed him with the clouds and darkness of doubt, the genial influence of " dinner, backgammon, and social talk " dispels these doubts and restores his belief in the world without and the self within : and Reid takes his stand with those who are " so weak as to imagine that they ought to have the same belief in solitude and in company." His essential demand, therefore, on the philosopher, is not primarily that he should make his beliefs consistent with those of the vulgar, but that he should make them con- sistent with his own ; and the legitimacy of the demand becomes, I think, more apparent, when we regard it as made in the name of Philosophy rather than in the name of Common Sense, For when we reflect on plain Common Sense, — on the body of unreasoned principles of judgment which we and other men are in the habit of applying in ordinary thought and discourse, — we find it certainly to some extent confused and inconsistent : but it is not clear that it is the business of Common Sense to get rid of these confusions and inconsistencies, so long as they do not give trouble in the ordinary conduct of life : at any rate it is not its most pressing business, since system -making is not its afl"air. But system -making is pre-eminently the affair of Philosophy, and it cannot willingly tolerate -no iiii: rini.osoriiv of (^(nniox sense inconsistciiciea : at least it it lias to tolerate them, as 1 sadly fear that it has, it can only tolerate them as a phvsieian tolerates a chronic imperfectiou of health, which he can only hope to mitigate and uot completely to cure. Accordingly, in Reid's view it is the duty of a philosopher — his duty as a jthilosopher — to aim steadily and persistently at bringing the common human element of his intellectual life into clear consistency with the special philosophic element. And Reid is on the whole perfectly aware — though his language occasionally ignores it — that for every part of this task the special training and intellectual habits of the philosopher are required. For the fundamental beliefs which the philosopher shares with the plain man can only be defined with clearness and precision by one who has reflected systematically, as an ordinary man does not reflect, on the operations of his own mind ; even the elementary distinction between sensation and per- ception is, Reid admits, only apprehended by the plain man in a confused form. To bring the distinc- tion into clear consciousness, to attend to "sensation and perception each by itself, and to attribute nothing to one which belongs to the other," requires, he tells us, " a degree of attention to what passes in our own minds, and a talent for distinguishing things that difier, which is not to be expected in the vulgar." The philosopher alone can do it : but in order to do it, he must partially divest himself of his philosophic peculiarities ; that is, he must temporarily put out THE PHILOSOPHY OF COMMON SENSE 417 of his mind the conclusions of any system he may have learnt or adopted, and merely bring his trained faculty of reflective attention to the observation and analysis of the common human element of his thought. But if it be admitted that the philosopher alone is capable of the steady and clear attention required to ascertain the fundamental beliefs of Common Sense, what valid evidence is there of the general assent to these beliefs on which Reid lays stress, and which, indeed, the term implies ? He seems to be in a dilemma ; either the many must be held capable of reflective analysis, or the decision on questions of fundamental belief must after all be limited to the expert few. The difticulty is partly met by pointing out that the philosophical faculty required to dis- tinguish and state such beliefs with precision much exceeds that required to judge of such a statement when made ; just as few of us could have found out the axioms required in the study of geometry, but we could easily see the truth of Euclid's at a very early age. Still, granting this, I think that Reid presses too far the competence of plain men even to judge of philosophical first principles. It is true, as he urges, that this judgment requires no more than a " sound mind free from prejudice and a distinct con- ception of the questions " : but it does not follow, as Reid seems to think, that " every man is a competent judge, the learned and unlearned, the philosopher and day-labourer alike " : because a good deal of the painful process we call ' learning ' is normally needed to realise these apparently simple requirements, 2E 4 IS TTTK TITTI.OSOT'TTY OF rO^fAfON RENSE free(.li>in i'vom pivjiulice jiikI distiiutiu'ss of con- ception. 1 will not allinn tliat no day-labourer could attain a distinct conception of the positions that Reid is defending against Berkeley and llumc : but I venture to think that a day-labourer who could convince us that he had attained it would be at once recognised as a born philosopher, incontrovertibly qualified by native genius for membership of the society that I have the honour to address. At the same time, 1 cannot think Reid wrong in holding that the propositions he is most concerned to maintain as first principles are implicitly assented to by men in general. That for ordinary men sense- perception involves a belief in the existence of a thing perceived, independent of the perception : that similarly consciousness involves a belief in the exist- ence of a permanent identical subject of changing conscious states : that ordinary moral judgment involves the belief in a real right and wrong in human action, capable of being known by a moral agent and distinct in idea from what conduces to his interest : that in ordinary thought about ex- perience we find implicit the unreasoned assumption that every change must have a cause, and a cause adequate to the effect, — all this, I think, will hardly be denied by any one who approaches the question with a fair mind. He may, of course, still regard it as unphilosophical to rest the validity of these beliefs on the fact of their general acceptance. But here again it must be said that Reid's own deference to general assent is of a strictly limited and subordinate THE PHILOSOPHY OF COMMON SENSE 419 kind. He is far from wishing truth to be determined by votes : he only urges that " authority, though tyrannical as a mistress, is useful as a handmaid to private judgment." He points out that even in the exactest sciences authority actually has this place : even a mathematician who has demonstrated a novel conclusion is strengthened in his belief in it by the assent of other mathematical experts who have ex- amined his demonstration, and is " reduced to a kind of suspense " by their dissent. This is, I think, undeniable : and perhaps we may separate Reid's just and moderate statement of the claims of Authority from his exaggerated view of the competence of untrained intellects to deal with philo- sophical first principles ; and simply take it as a cardinal point in the philosophy of Common Sense that a difference in judgment from another whom he has no reason to regard as less competent to judge than himself, naturally and properly reduces a thinker to a " kind of suspense." When the conflict relates to a demonstrated conclusion, it leads him to search for a flaw in the opponent's demonstration ; but when it relates to a first principle, primary datum, or funda- mental assumption, this resource appears to be excluded : and then, perhaps, when he has done all that he can to remove any misunderstanding of the question at issue, the Common Sense philosopher may be allowed to derive some support from the thought that his own conviction is shared by the great majority of those whose judgments have built up and continually sustain the living fabric of our 4 20 TlIK rilllA^SOrilV OF (H)MMON SENSE common thoiiLilit ;iiul kno\vlcd«re. And this, I tliiiik, is all that Reiil really means to claini. I have now, I hope, succeeded in making clear the general relation which Rcid's epistemology bears to his psychology. 1 have not used these modern terms, because Reid himself blends the two subjects under the single notion of " Philosophy of the Human Mind": but it is necessary, in any careful estimate of his work, to distinguish the process of psycho- logical distinction and analysis through which the fundamental beliefs of Common Sense are ascertained, from the arguments by which their validity is justi- fied. I do not propose to enter into the details of Reid's psychological view, which has largely become antiquated through the progress of mental science. But if Locke is the first founder of the distinctively British study. Empirical Psychology, of which the primary method is introspective observation and analysis, I think Reid has a fair claim to be regarded as a second founder : and even now his psychological work may be studied with interest, from the patient fidelity of his self-observation, the acumen of his reflective analysis, and, especially, his entire freedom from the vague materialism that, in spite of Descartes, still hung about the current philosophical conception of Mind and its operations. It is, indeed, in the task of exposing the unwarrant- able assumptions generated by this vague materialism that the force and penetration of Reid's intellect is most conspicuously shown. Let me briefly note this in the case of the beliefs THE PHILOSOPHY OF COMMON SENSE 421 involved in ordinary sense -perception, since this problem occupies a leading place in his discussion. Not, I ought to say, that he is specially interested in this problem on its own account : he makes it quite clear that it is on far greater issues that his thought is really set. God, Freedom, Duty, the spirituality of human nature, — these are, for Reid as for Kant, the grave matters really at stake in the •epistemological controversy. But these greater matters, for the very reason of their supreme importance, are apt to stir our deepest emotions so strongly as to render difficult the passionless precision of analysis and reasoning which Reid rightly held to be needful for the attainment of philosophical truth : while at the same time it is clear to him that all the questions hang together, and that the decision of one in the sense that he claims will carry with it the similar determination of the rest. Accepting this view then, and remembering that in a trivial case we are trying no trivial issue, let us examine his treatment of the cognition by Mind of particular material things. Here Reid's task, as he ultimately saw, was merely carrying further the work of Descartes. By clearly distinguishing the motions of material particles antecedent to perception from perception itself as a psychical fact, Descartes had got rid of the old psychophysical muddle, by which forms or semblances of things perceived by the senses were supposed somehow to get into the brain through the ' animal spirits ' and so into the mind. But he had not equally got rid of the view that perception was -122 THE rillLOSorilV OF COMMON SENSE tlie uottinn of an iilo;i in the mind, from which tlie cxisteuce of a tiling outside tlic mind like the idoa had to be someliow inferred. Tliis view is definitely held, not only by his disciple Malebruuche but by his independent successor Locke. They do not see what Raid came to see, that the normal perception of an external object presents itself to introspection as an immediate cognition : that is, as a cognition which has no psychical mediation, no inference in it. AVhat prevented them and others from seeing this was, mainly, a naive assumption that the mind can only know immediately what is ' present ' to it, and that things outside the body cannot be thus present ; as the mind cannot go out to them and they cannot get into the mind, only the ideas of them can get in. It was reserved to Reid to point out the illegitimacy of this assumption, and to derive it from a confused, half- unconscious transfer to Mind and its function of cognition, of the conditions under which body acts on body in ordinary physical experience. When the assumption is made explicit and traced to its source, it loses, I think, all appearance of validity. It is to be observed, that in athrming external perception to be an immediate cognition, Reid does not of course mean that it is physically uncaused. He only means that the perceiving mind has not a double object, its own percept and a non-mental thing like its percept : and accordingly that our normal conviction of the present existence of the non-mental thing perceived is not a judgment attained by reasoning, Ijut a primary datum of knowledge. He THE PHILOSOPHY OF COMMON SENSE 423 recognises like his predecessors that it has physical antecedents, movements of material particles both without and within the organism. And he recognises, more distinctly than his predecessors, that it has psychical antecedents and concomitants, i.e. sensations which he carefully distinguishes from the perception that they suggest and accompany. A consideration of these antecedents may possibly affect our reflective confidence in the cognition that follows them, — that question I will deal with presently, — but at any rate it cannot properly modify our view of the content of this cognition as ascertained by introspec- tive observation. This, I think, remains true after duly taking account of the valuable work that has been done since Reid's time, in ascertaining more accurately the antecedents and concomitants of our common perceptions of extended matter. Whatever view we may take on the interesting but still disputed questions as to the precise manner in which visual, tactual, and muscular feelings have historically been combined in the genesis of our particular perceptions and general notions of matter and space, — there can still be no doubt of the fundamental difference in our present consciousness between these perceptions or notions and any combinations of muscular, tactual, and visual feelings. It has indeed been held, by an influential school of British psychologists, that this manifest difference is merely apparent and illusory : it has been held that by a process of " mental chemistry " sensations and images of sensation have been "compounded" into 42-j iiii: riiii.osoriiv ot co\n\ny skxse what NVi' MOW ilistin«;uisli as perceptions ami concep- tions of matter in space, and tliat tlic latter really consist of sensations and iniatj^es of sensation, just as water really consists of oxygen and h3'drogen. Hut this view involves a second illegitimate transfer of physical conditions to psychical facts ; and Reid would certainly have rejected 'mental chemistry' in this application as unhesitatingly as he does reject it when applied to support the conclusion that a " cluster of the ideas of sense, properly combined, may make up the idea of a mind." He would have rejected it for the simple reason that we have no ground for holding any fact of consciousness to be other than careful introspection declares it to be. In the case of material chemistry, the inference that a compound consists of certain elements depends on experimental proof that we can not only make the compound out of the elements, but can also make the elements again out of the compound. But even if we grant that our cognitions of Matter and Space, of Self and Duty, are derived from more elementary feelings, it is certain that no psychical experiment \vill enable us to turn them into such feelings again : the later phenomena, if products, are biological not chemical products, resulting from evolution, not from mere composition. Still, it may be said, granting the existence of cognitions and beliefs that cannot now be resolved into more elementary feelings, and that present them- selves in ordinary thought with the character of unreasoned certitude, systematic reflection on these beliefs and their antecedents must render it impossible THE PHILOSOrHY OF COMMON SENSE 425 to accept them as trustworthy premises for philo- sophical reasoning. It is a commonplace that the senses deceive, and the more we learn of the psychophysical process of sense-perception, the more clear it becomes why and how they must deceive. Even apart from cases of admitted illusion, philo- sophical reflection on normal perception continually shows us, as Hume urges, a manifest diff'erence between the actual percept and what we commonly regard as the real thing perceived. Thus, Hume says, " the table which we see seems to diminish as we remove farther from it : but the real table which exists independent of us suffers no alteration. It was, therefore, nothing but its image which was present to the mind. These are the obvious dictates of reason." In answering this line of objection Reid partly relies on a weak distinction between original and acquired perception, which the progress of science has rendered clearly untenable and irrelevant. Apart from this his really efi'ective reply is twofold. First he points out that the very evidence relied upon to show the unreality of sense -percepts really affords striking testimony to the general validity of the belief in an independent reality known through sense- perception. It is by trusting, not by distrusting, this fundamental belief that Common Sense organised into Science continually at once corrects and confirms crude Common Sense. Take Hume's case of the table. If nothing but images were present to the mind, how could we ever know that there exists a real table which does not alter while the visible 4l'G I'HK 1'HI1.(>S(HM1V OV COMMON SKXSK mat^iuliulo chiinges with its ilistaiici' iVoiii us? Tlie plain man knows this ihrou^li an actjuirtMl porccptiou, !)>• wliiih lie hahitually jutl^fos of real niaL^nitiule from visilile nppearaiioes : but science carries the knowledge furtluT, enabling us to predict exactly what a])pcarance a given portion of extended matter will exhibit at any given distance fnnn the spectators. Now all this coherent, precise, and unerring prediction rests upon innumerable sense-perceptions ; and the scientific pro- cesses which have made it possible have been carried on throuf^hout on the basis of the vulgar belief in the independent existence of the matter perceived. ** Is it not absurd," Reid asks, " to suppose that a false supposition of the rude vulgar has been so lucky in solving an infinite number of phenomena of nature ? " Suppose, however, that the opponent resists this argument : suppose he maintains that, though physical science may find the independent existence of matter a convenient fiction, — as mathematicians find it convenient to feign that they can extract the square root of negative quantities, — still in truth Mind can only know mental facts — feelings and thoughts. Suppose he further urges that the common belief in the independent existence of the object of perception is found on reflection to have no claim to philosophic acceptance, because while admittedly un- reasoned it cannot be said to be strictly intuitive : — granted that I may directly perceive the table before me, I cannot directly perceive that it exists independ- ently of my perception. To this line of argument Reid has another line of reply. He points out to the THE PHILOSOPHY OF COMMON SENSE 427 Idealist that he does not escape from this kind of unreasoned belief by refusing to recognise a reality beyond consciousness. He has still to rely on data of knowledge which are open to the same objections as the belief in the independent existence of matter. For instance, he has to rely on memory. If sense- perception is fallible, memory is surely more fallible ; if we do not know intuitively and cannot prove that what we perceive really exists independently of our perception, still less can we either know intuitively or prove that what we recollect really happened : if on reflection we find it difficult to conceive how the Non-ego can be known by the Ego, there is surely an equal difficulty in understanding how the Present Ego can know the Past, And yet once cease to rely on memory, and intellectual life becomes impossible : even in reasoning beyond the very simplest we have to rely on our recollection of previous steps of reason- ing. A pure system of truths reasoned throughout from rational intuitions may be the philosophic ideal : but it is as true of the intellectual as of the physical life that living somehow is prior to living ideally well : and if we are to live at all, we must accept some beliefs that cannot claim Reason for their source. Is it not then, Reid urges, arbitrary and unphiloso- phical to acquiesce tranquilly in some of these beliefs of Common Sense, and yet obstinately to fight against others that have an equal warrant of spontaneous certitude ? May we not rather say that it is the duty of a philosopher to give impartially a provisional acceptance to all such beliefs, and then set himself to ■i2S IHK riHLO.^OrTTV OF (^(niMOX SENSE I'larifv tliem by ivtlection, ivinovo iiKidvcrtoiuios, confusions, and contnulictions, and aa far as possible build together the purcjod results into ;in ordered and harmonious system of thought ? If, finally, the opposing philosojdier answers that he eannot be satisfied by any system that is not perfectly transparent to reason, licid does not alto- gether refuse him his sympathy, though he cannot encourage him to hope. " I confess," he says, " after all that the evidence of reasoning, and of necessary and self-evident truths, seems to be the least mysterious and the most perfectly comprehended . . . the light of truth so fills my mind in these cases that I can neither conceive nor desire anything more satisfying. On the other hand, when I remember distinctly a past event, or see an object before my eyes," though " this commands my belief no less than an axiom . . . I seem to want that evidence which I can best com- prehend and which gives perfect satisfaction to an inquisitive mind." And ''to a philosopher who has V>een accustomed to think that the treasure of his knowledge is the acquisition of his reason, it is no doubt humiliating to find" that "his knowledge of what really exists or did exist comes by another channel," and that " he is led to it " as it were " in the dark." "It is no wonder" then "that some philosophers should invent vain theories to account tor this knowledge " : while others " spurn at a know- ledge they cannot account for and vainly attempt to throw it off." But all such " attempts," he holds, are as impracticable as " an attempt to fly." THE PHILOSOPHY OF COMMON SENSE 429 The passage from which I have quoted was pub- lished in 1 785, when Keid was seventy-five years of age. Even before it was published attempts at aerial navi- gation had suddenly come to seem less chimerical in the physical world ; and before the end of the century, in the world of thought, attempts to tran- scend and rationally account for the beliefs of Common Sense — more remarkable than any dreamt of by Reid — had begun to excite some interest even in our insular mind. The nineteenth century is now drawing to its close ; and these attempts to fiy are still going on, both in the physical and in the intellectual world ; but in neither region, according ta my information, have they yet attained a triumphant success. At the same time our age, which has seen so many things achieved that were once thought impossible, may without presumption contemplate such attempts in a somewhat more hopeful spirit than was possible to Reid : and I should be sorry to say anything here to damp the noble ardour or to depress the high aspirations that ought to animate a society like yours. But if there should be any one among you who, desirous to philosophise and yet fearing the fate of Icarus, may prefer to walk in the dimness and twilight of the lower region in which my discourse has moved, — then I venture to think that he may even now find profit in communing with the earnest, patient, lucid, and discerning intellect of the thinker who, in the history of modern speculation, has connected the name of Scotland with the Philo- sophy of Common Sense. CRITKRIA OF TRUTH AND ERROR (Reprinted from Mitui, vol. ix. N.S., January 1900.) The present essay is a partial discussion of what I regard as the central problem of Epistemology. In order that its drift may be clearly seen from the outset, I will begin by explaining briefly — without argument — my view of Philosophy, Epistemology, and their relation. I take it to be the business of Philosophy — in ^Ir. Spencer's words — to ' unify ' or systematise as completely as possible our common thought, which it finds partially systematised in a number of different sciences and studies. Now before attempting this unification, we must wish to be somehow assured that the thoughts or beliefs which we seek to systematise completely are true and valid. This is obvious ; no rational being with his eyes open would try to work up a mixture of truth and error into a coherent system without some attempt to eliminate the error. It is prima facie necessary, therefore, as a pre- liminary to the task of bringing into — or exhibiting in — coherent relation the different bodies of sys- tematic thought which furnish the matter for Philo- 430 CEITEEIA OF TKUTH AND EEEOE 431 sophy, to have some criteria for distinguishing truth from error. It may, however, be thought that this need — though undeniably urgent in the case of such studies as, e.g., Politics and Theology — will not be practically presented, so long as the philosopher's work is confined to the positive sciences. The pre- valence of error in Politics is kept prominently before our minds by the system of party government ; and the effective working of this system almost requires the conviction on either side that the political pro- gramme of the other party — unhappily often in a majority — is a tissue of errors. So again in Theology, it is the established belief of average members of any religious denomination that the whole world outside the pale of the denomination lies in the darkness of error on some fundamental points ; and even within the pale, the wide -spread existence of right-hand backslidings and left - hand defections from the standard of orthodoxy is continually attracting the attention of the newspapers. But no doubt, in elementary study of the positive sciences, error is commonly only brought before our minds in the strictly limited form of slight discrepancy in the results of observation, as something reducible to a minimum by an application of the theory of proba- bilities. Still the danger of error is only thus kept in the background, so long as we confine our attention to the more settled parts of the established sciences in their present condition. Around and beneath these more settled portions, in the region where knowledge 432 C'KITEKIA OF Till TH AND KIMM)!: is growinj^ in range ordeplli, ami ilu- liimian inldlect eudcavourin«; to solve new (juestions, or penetrate to IX more solid basis of jirineipK's, we lind eontinually eontliet and eontroversy as ti> the truth of new eon- elusions — whieli a})j)ear estal)lislicd and demonstrated to the adventurous minds that liave worked them out — as to the h'gitimacy of new hypotheses, and the validity of new methods ; and wherever we find sueh coufliet and eontroversy, there must be error on one side or the other, or possibly on both. And the fact of error is still more prominently brought before our minds when we turn from the present to the past, and retrace the history of the now established sciences : since we find that in almost all cases human knowledge has progressed not merely by adding newly ascertained facts to facts previously ascertained, but also, to an important extent, by questioning and correcting or discarding beliefs — often whole systems of connected beliefs — previously held on insufficient grounds. In this way, convinced by Copernicus, the human mind dropped the Ptolemaic astronomy and reconstructed its view of the planetary and celestial motions on the helio- centric hypothesis ; convinced by Galileo, it discarded the fundamental errors of Aristotle's view of matter ; convinced by Lavoisier, it rectified its conception of chemical elements, and relegated the remarkable sub- stance ' phlogiston ' — that had enjoyed an imaginary existence for something like a century — to the limbo of recognised non-entities ; convinced by Darwin, it abandoned its fundamental notion of the fixity of CEITEKIA OF TEUTH AND EEEOK 433 organic species, and accepted a revolution in morpho- logical method. Now the student of science is ordinarily not much disturbed by this evidence that his class forms no exception to Pope's oft -quoted characterisation of man as " sole judge of truth, in endless error hurled." When, in the progress of thought, any prevalent scientific belief is recognised as erroneous, he simply discards this — with more or less endeavour to ascer- tain the particular causes of error and guard against their recurrence, — and, on the whole, continues his natural processes of acquiring, evolving, systematis- ing beliefs with undiminished confidence. But to the philosophical mind the ascertained erroneousness of some beliefs is apt to suggest the possible erroneousness of all. If a belief that I once held to be certainlv true has turned out to be false, what guarantees me against a similar discovery in respect of any other belief which I am now holding to be true ? The mind is thus overspread with a general and sweeping distrust of the processes of ordinary thinking, which is not exactly to be called philo- sophical scepticism — since this usually presents itself as systematically deduced from premises accepted by philosophers, — but is rather to be conceived as the naive, untechnical scepticism of a philosophic mind, which may turn out to be (as in the classical case of Descartes) a mere stage in its progress toward a dogmatic system. At any rate, it is the removal of this philosophic uncertainty — in respect of beliefs that, in ordinary thought, arc commonly assumed 2 V 431 (MJITKKIA OF TKl'Tll AND KKROR to \h' triif- -lli.'it I rci^ard as tlie jtrimary aim ol" Kpisteint)l()rry. 1 liavo said that tliis tusk lies in \hv way of philosopliy ; but I ouulit to add tluit it dors not appear to lio in the way of all pliilosophors. Some of those who have devoted their minds to the solution of philosophical problems seem hardly to have contemplated error except as a kind of misconduct into which the rest of the human race — and especially other philosophers — are inexcusably prone to fall. It is, indeed, a common experience of mankind in all departments of theory and practice that the liability to error is more equally distributed among human beings than the consciousness of such liability. But the variations of self-confidence that we find among persons who have devoted themselves to the business of philosophy are perhaps less than elsewhere to be attributed to differences of individual temperament : it would rather seem that in the social movement of philosophic thought there are general ebbs and flows ; an age of confidence followed by an age of diffidence. It is partly the fact that the philosophic mind of the modern world is now rather at the ebb, with its constructive impulses compara- tively feeble, which explains the development and the prominence that the epistemological aspect or function of philosophy is now receiving ; and has accordingly led to the composition of the present paper. I will begin by somewhat limiting my subject for clearness of discussion. I have contrasted ordinarv CEITEEIA OF TEUTH AND EEEOE 435 certitude with philosophic doubt ; but even the plain man is not always cocksure. Sometimes he even doubts and suspends his judgment ; but even when he believes and positively affirms, many of his beliefs and affirmations — most of those relating to the future — are intended to be taken as not certain but prob- able. By a ' probable ' belief I do not now mean a belief relating to probabilities ; for this may be as certain as any other — as for instance the belief that the chances are even that a penny I toss will come down tails. The theory of chances has been described as a method of extracting knowledge out of ignorance ; it is undoubtedly a method of converting probable judgments into certain ones — though the certainty is of a peculiar kind, and its verification presents a special epistemological problem of some interest. But the probable beliefs that I now wish to dis- tinguish from certain ones are beliefs which involve no attempt at a quantitative estimate of ' amount of probability ' ; and they are often* in form of ex- pression indistinguishable from beliefs held with certitude : — thus when a man affirms in conversation that the new plan of international arbitration will have no practical effect, or that the Liberal Party must return to power after the next general election, it will be generally understood that though the speaker may appear to express certitude on these points, he only means that the events are extremely probable. I draw attention to this ambiguity of expression, because it facilitates an indeterminateneps of thought, of which we have to take note in applying -i:^G CINTKK'IA OK 'IMM'Tll AND KKKOi; tl»o tlistiiu-tioii tli;it I now draw liclwccii ' ccrlaiii ' juitl • pn>l)al)li' ' l»olit>fs. ()t'ltii in onlinary llioui^lil. we di) Mot know wIuMImt wc aiv. sum ol" what wr atlirm unloss wo an* ltd to reflect on tlio jxiiiit ; sometimes we ilo not know alter lellection ; some- times we are conscious of elements of nncertainty whirh we ilcM-ide to disreii^anl, and then wc say thai we arc 'morally certain' — meaning that we should unhesitatingly act as if we were certain. This last. state of mind I shall consider hereafter; at present I wish to confine attention to beliefs which present themselves in ordinary thought as certain without 4ualifieation. Of these I may roughly distinguish three chief classes : (l) particular beliefs about the present and recent past of the changing world of which wc are part; (2) general beliefs more or less systematised in the sciences, especially the exact sciences, which we may happen to know ; (3) beliefs that prima facie relate not to mere matters of fact but to moral or aesthetic valuation — to what we ought to do as individuals, or what government ought to do, or what is good and bad in manners, literature, and art. Of course in these latter regions of belief any educated person is aware that there is much doubt and controversy ; still there are plenty of propositions in each of the regions indicated, which it would seem in ordinary thought as absurd to dispute or (jualify as propositions with regard to the most familiar matters of fact. When C'harles Lamb took a candle to examine the cerebral bumps of the soap-boiler who affirmed that Shakespeare was a CEITEKIA OF TEUTH AND EEROK 437 first-rate dramatic writer, it was, I suppose, because the irrefragable certainty of the proposition seemed to render its express statement absurdly superfluous. Concentrating attention, then, on beliefs that in ordinary thought are certain in the sense explained, let us — with a view to a necessary limitation of our inquiry — take a second distinction. Eeflecting upon the beliefs, of the truth of which I have no doubt, I perceive that some of them (e.g. the propositions of Euclid) have only derivative or dependent certainty — my belief in them rests on my belief in some other proposition or propositions ; while in other cases {e.g. most of the axioms of Euclid) my certitude may be distinguished as primary or independent. In the instance given — as I have personally followed the reasonings of Euclid and satisfied myself as to their cogency — I might employ a clearer antithesis, and say that some of my geometrical beliefs have ' intuitive ' and others demonstrative certainty. But this an- tithesis is too narrow for my present purpose. For, firstly, I do not profess to have intuitive certainty with regard to all beliefs for which proof does not seem to be required. I am certain that I read through the three first pages of this essay before I sat down to write the fourth half an hour ago ; but it would be contrary to usage to call this certaint}' ' intuitive,' though the belief does not present itself to me as requiring proof. Secondly, I wish to include among beliefs with derivative certainty that: comparatively large body of scientific conclusions which I believe to have been scientifically proved, 438 CKIl'KKIA OF TIU Til AND KKKOli though not ti> lut', and wliith 1 acronlingly accept on tho authority of oiic or nunc otlnT {)(>rsoiis. Of oourso, ill a wide scnso. of the word, u stalenuMiL of my grounds for trust in^x J^ny conclusion arrived at hy some other mind niii^ht lie caUcd niv 'j>roof'of llic proposition ; hut at any rate it wouhl not he scientific; demonstration, and it wouhl l)e odd to call the certainty of any such belief to me ' demonstrative certainty.' For simplicity, let us here provisionally disregard any doubts of the authority of others as others: then the distinction will be between beliefs which requiring proof seem to have obtained it, and beliefs which do not seem to require it. Now the errors due to taking invalid proof for valid are the special subject of investigation in the science of Logic ; and it is widely held that the labours of logicians have j^rovided adequate criteria for excluding them : that they have discovered by analysis certain forms of reasoning into one or other of which any cogent inference may be thrown, and by the application of which the validity or invalidity of any process of inference may be made manifest. Suppose we grant this : then our epistemological problem is solved in respect of dependent or in- ferential beliefs — so far as the process of inference by which they are reached is capable of being thrown into a logically cogent form. That is, I can in this way obtain assurance that all my apparently proved beliefs are true if the premises from which they are inferred are true : and if these premises are them- selves arrived at by inference I can similarly apply CEITEEIA OF TRUTH AND ERROR 439 the test to the proof of them — and so on till we come to the ultimate premises. I propose to assume for tlie purpose of this paper that Logic has done satis- factorily what it commonly professes to have done ; and that our task, accordingly, may be limited to the verification of ultimate premises, or beliefs that are in ordinary thought accepted as not requiring proof. The importance of the task thus limited has been fully recognised by some philosophers. J. S. Mill, indeed, seems disposed to bestow on this inquiry the venerable name of " Metaphysics." " The grand question," he says, " of what is called Metaphysics is ' what are the propositions that may reasonably be received without proof ? ' " And it is, I suppose, to propositions of this kind that Descartes' famous criterion — expressed in the formula " that all the things which we very clearly and distinctly conceive are true " — was primarily designed to apply. On the other hand, it seems to be also primarily to this class of propositions that Kant's unqualified rejection of "a general criterion of truth" applies^ — since Kant regards Logic as having adequately furnished criteria of formal truth, and therefore of all kinds of inference. In fact, Kant's condemnation of the task on which I am engaged is so strong and sweeping that I think it well to examine his argu- ments before proceeding further. I give it somewhat abbreviated. ^ See § 3 of the Introducticm to Transcendental Logic {Kntik der reirun Vernunft. Hart. p. 86), 440 CKITKKIA <>K TKI"1'I1 AND KKIJOK " ll' truth ((insist.-^ — as is adniith'd — in llu- a<2;ree- iiuMit of a cogiiitioii willi its object, tliat object must, Ity the true eognitiou, be tlistinguished fioiu some other object or objects. Nt)\v it is implied in tlic idea t)f a general criterion of trutli that it is valid with regard to every kind of cognition, whatever the objects cognised may be. But then, as such a criterion must abstract from the particular contents of particular cognitions, whereas, as we have seen, truth concerns those very contents, it is im})ossible and absurd to suppose that such a general criterion can give us a sign of the truth of cognition in respect of its content or matter. Therefore a sufficient and at the same time general criterion of truth cannot possibly be found." In examining this passage I may begin by pointing out that Kant's view of truth as ' consisting in the agreement of cognition with its object' — which he takes as universally accepted — cannot be applied to all propositions without a difficult extension of the notion of 'object' (Gegenstand). This will appear, if we try to apply it to strictly hypothetical pro- positions, or to categorical propositions of ethical import. To this consideration I shall hereafter return ; meanwhile, in discussing Kant's definition, I shall assume for clearness, that we are dealing with judgments that are intended to represent some fact, past, present, or future, particular or general. Thus restricted, Kant's argument is simple and at first sight plausible ; but 1 think it contains a petitio CEITERIA OF TRUTH AND ERROR 441 principii. For it proceeds on the assumption that true cognitions cannot as such have any common characteristic, except that of agreeing with their objects ; but that is surely to assume the very point in question. To illustrate this, let us take Descartes' criterion before referred to, as the first that comes to hand in the history of modern philosophy. How can the diversity of the objects of cognition be a logical ground for denying that "what is clearly and distinctly conceived" is necessarily true ? — since the distinction between clear and obscure, and between distinct and confused conception, does not become less applicable when we pass from one kind of object to another. It may be answered on Kant's behalf that "clear- ness and distinctness of conception " belong to the form of thought, not to its matter ; that clearness and distinctness of conception may prevent us from attributing to any subject an incompatible predicate, but not from attributing a predicate that though compatible does not actually belong to the subject. But it is just this dogmatic separation of form from matter that I regard as an unproved assumption. It is surely conceivable that the relation of the knowing mind to knowable things — to the whole realm of possible objects of knowledge — is such that, whenever any matter of thought is clearly and distinctly conceived, the immediate judgments which the mind unhesitatingly affirms with regard to it are always true. As will presently appear, I do not hold a brief for the Cartesian criterion ; on 442 CKI'l'KKIA (>K Till Til AND KKKOR tlio I'ontrarv, 1 linvf no doulit wliatcvcr that tlu' Cjirtosiaii i-rittM-iou takrn hy itsolf is inadcquatt.'. All I urge is tliat its iiia(lec|uucy is not cstablisluHl by Kant's sumniarv arijunient. Lot us turn to consider KanLs s\vei'[)in<^ negation in rclfttion to a ditVereut criterion, laid down by Empiricists. I tiike the principle of Empiricism, as an epistemo- logical doctrine, to be that the ultimately valid premises of all scientific reasonings are cognitions of particular facts ; all the generalisations of science being held to be obtained from these particular cognitions by induction, and. to depend u})on these for their validity. I do not accept this principle ; I think it impossible to establish the general truths of the accepted sciences by processes of cogent inference on the basis of merely particular premises ; and I think the chief service that J. S. Mill rendered to philosophy, by his elaborate attempt to perform this task, was to make this impossibility as clear as day. But I wish now to avoid this controversy ; and, in order to avoid it, I shall take the Empirical criterion as relating only to particular cognitions ; leaving open the question how far we also require universal premises in the construction of science. The criterion is briefly discussed by Mill [Logic, Book IV. chap. i. §§ 1, 2). It being understood that the validity of the general truths of the sciences depends on the correctness of induction from correct observation of particular facts, the question is what coiarantee there is of the correctness of the observa- o CEITEEIA OF TEUTH AND EEEOE 443 tions ? — in Mill's words " we have to consider what is needful in order that the fact supposed to be observed may safely be received as true." The answer is, "in its first aspect," very simple. "The sole condition is that what is supposed to have been observed shall really have been observed ; that it be an observation — not an inference." The fulfilment, indeed, of this sole and simple condition is not — as Mill goes on to explain — so easy as it may appear ; " for in almost every act of our perceiving faculties, observation and inference are intimately blended ; what we are said to observe is usually a compound result of which one -tenth may be observation and nine- tenths inference." E.g. I affirm that I saw my brother at a certain hour this morning ; this would commonly be said to be a fact known through the direct testimony of my senses. But the truth. Mill explains, is far otherwise ; for I might have had visual sensations so similar as to be indistinguishable from those I actually had without my brother being there ; I might have seen some one very like him, or it might have been a dream, or a waking hallucination ; and if I had the ordinary evidence that my brother was dead, or in India, I should probably adopt one or other of these suppositions without hesitation. Now, obviously, " if any of these suppositions had been true, the affirmation that I saw my brother would have been erroneous " ; but this does not, in Mill's view, invalidate the Empirical criterion, for "whatever was matter of direct perception, namely, the visual sensations, 444 CKITKKIA oK TRUTTr AND EKROll would Iwivo lu'iMi n';il " ; uiv ajipait'iit cognition of this reality (ho lucilly ussuines) woiiKl have been a true and valid coLcnition. in short, only separate observation from inference, and observation — or apparent knowledge obtained through observation — is absolutely valid and trustworthy ; the idea that these are 'errors of sense' is itself a vulgar error, or at least a loose thought or phrase ; there are no errors in direct sense-perception, but only erroneous inferences from sense. Now 1 shall presently consider how far this criterion, taken in any sense in which it would be available for its purpose, is completely trustworthy. But, however that may be, it seems to me that Kant's sweeping negative argument — which we are now examining — has really no force against its validity. No doubt, according to Kant's general view of the form and matter of thought, this criterion, like the other, relates primarily to the form ; for it rests on the distinction between two different functions of the knowing mind — Observation or Perception and Inference. But I see no reason to infer that it is therefore incapable of guaranteeing the material truth of Empirical cognition ; or that the relation of the knowable world to the knowing mind cannot possibly be what Empiricism affirms it to be. If now we contemplate together the two criteria that have Vjeen examined — the Cartesian and the Empirical — it is evident that, at least in its primary intention, neither alone covers the whole ground of CKITEEIA OF TRUTH AND ERROR 445 the premises for which verification is pi^ma facie required. The Empirical criterion only verifies particular premises, and the Cartesian appears to be applied by its author primarily to universals — to what is " clearly and distinctly conceived by the pure understanding." This leads me to suggest that Kant has perhaps taken too strictly the demand for a ' universal ' (allgemein) criterion of truth. He has understood it to be a demand for some ascertainable characteristic — other than truth — always found to belong to valid cognitions, and never found in invalid ones. And no doubt a criterion of this scope is what any philosopher would like to get ; but any one who has realised the slow, prolonged, tortuous process by which the human intellect has attained such truth as it has now got, will thankfully accept something less complete. If {(^'.g.) any epistemological doctrine offers, among the commonly accepted premises of scientific reasoning, to mark out a substantial portion to which the stamp of philosophic certainty may be affixed ; or if, again, it offers to cut out a class of invalid and untrustworthy affirmations, to warn us off a region in which our natural impulse to affirm or believe must, if indulged, produce mere illusion and semblance of knowledge — then, if either offer is made good, we shall gratefully accept it as a philosophic gain. Now it is remarkable that in both these ways, but especially in the latter way, Kant undoubtedly does offer general criteria of truth which, if valid, 446 (MMTKIUA OF TKl'Tll AND KIMIOK are of immeiiso iiuportaiico. liuloetl, it is tlu' very aim ami purpi>.se of liis Critical PhdosopJiij — as its name indicates — to estahlisli such criteria : it is its aim, by a critical examination of our faculties of knowledge, to cut oti' and stamp as manifest illusion the whole mass of beliefs and alKrmations with regard to 'things in themselves' wliicli common sense naively makes, and which — or some of which — previous dogmatic philosophers had accepted as valid. At the same time, by the same critical analysis, Kant seeks to stamp with philosophic precision and certitude the fundamental principles of physical knowledge — as that every event has a cause, and the quantum of substance in the physical world is unchangeable — while restricting the application of these principles to phenomena. And here 1 would remark that the main import- ance for philosophy of the epistemological question brought into prominence by Kantian Criticism — the question as to the Limits of human knowledge — seems to depend upon its connexion with the question with which we are now concerned — the inquiry after criteria. For our interest in Kant's inquiry into the limits of knowledge certainly depends on the fact that the limits which the critical thinker aims at establishing have been actually transgressed by other thinkers. It therefore implies an actual claim to validity on behalf of assertions transgressing the limits which the criticist denies : so that he may be viewed as propounding in respect of these assertions a criterion for distinguishing CKITEEIA OF TEUTH AND EEEOE 447 truth from error, which stamps them as error. It is true that as regards a part of the assertions he discusses — e.g. as to the infinity or finiteness of Space and Time, or the infinite or finite divisibility of matter — the criticist finds a controversy going on which implies error on one side or the other : but by his criterion he decides that there is error on both sides, the ' antinomy ' which leads to controversy in each case arising from a fundamental misconception common to both sides. It is no part of my plan to criticise Kant's episte- mology : what I am rather concerned to point out is that his system is embarrassed in a quite special manner by the difiiculty that besets every construc- tive epistemology — the difiiculty of finding a satis- factory answer to the question, * Quis custodiet custodem ? ' For the claim of Criticism is to establish the limits of human knowledge by an examination of man's faculties of knowledge : but the proposition that we have faculties of cognition so and so consti- tuted can only be an inference from the proposition that we have such and such valid cognitions. It would thus seem that the Critical procedure must presuppose that truth adequately distinguished from error has already been certainly obtained in some departments. And in fact this presupposition is frankly made by Kant so far as Mathematics and Physical Science are concerned. He expressly takes their validity as a datum. Mathematics, he tells us {Proleg. § 40), "rests on its own evidence," and Physical Science " on experience and its thorough- 4-lS ('KITi:i:i.\ OK TIM'Tll AND KIIKOK going oontinnatidii " : iicillicr study stands in need of Criticism " tor its own sntVtv and ctMtainty." And lie similarly assumes the validity tuid compleLeness of Formal Logic as the starting- ])oint for his Trnn- scriuh-ntal Analytic. if, therefore, we ask for a criterion of truth and error in Mathematical and Loj/ical Judfrments — and error undenialily occurs in hoth — or in the Empirical cognitions which confirm the general propositions of physical science, we cannot obtain this from Kantian criticism without involving the latter in a circulus in pi'ohamh. We are therefore pritna facie thrown back in the former case on the Cartesian or some similar criterion for guaranteeing ' truths of reason,' in the latter case on some Empirical criterion for guaranteeinof ' truths of fact.' I turn, therefore, to examine more closely these two criteria. With regard to the former, however, it may be thought that such examination is now super- fluous, since the historic failure of Descartes' attempt to extend the evidence of mathematics to his physical and metaphysical principles has sufKciently shown its invalidity. ^' Securus judical orhis terrarum'' \ and the inadequacy of the Cartesian criterion may be thought to be now ^ res judicata.' On the other hand, Mr. Spencer has in recent times put forward a criterion which, so far as it relates to universal cognitions, has at least a close afhnity to the Cartesian. I propose, therefore, to begin by some consideration of the earlier proposition. I may begin by saying that Descartes' statement CKITEEIA OF TKUTH AND EKEOE 449 of his criterion hardly satisfies his own requirements, i.e. it is not quite clear what he means by the ' clear- ness ' of a notion. I think that it will render Descartes' meaning with sufficient precision to drop the word ' clear,' keeping ' distinct ' (which, he says, involves ' clear '), and explain a distinct notion of any object to be one that is not liable to be confounded with that of any different object — 'object' being taken to denote any distinguishable element or aspect of Being, in the sense in which Descartes uses ' Being ' as a wider term than Existence, and includes under it the objects of mathematical thought. One further modification of Descartes' statement seems expedient : Descartes applies the term ' clear ' (or ' distinct ') * conception ' to the cognition of the connexion of subject and predicate in a true judgment, as well as to the notions taken separately. But it seems desirable to make more explicit the distinction between the two ; since the indistinctness that causes error may be held to lie not in the latter but in the former. We may state our question, then, as follows : " Is error in universal judgments certainly excluded by a distinct conception of the subject and predicate of the judgment and of their connexion ? " But this at once suggests a second question : " Why does Descartes hold it to be excluded ? " And here it is noteworthy that he nowhere affirms the infallibility of his criterion to be intuitively known. He seems to have three ways of establishing it : (l) He presents it as implied in the certainty of his conscious existence (Meth. iv. 2G 4r»o rrjTKKiA of tki'tii and KKlJOi: ami Mril. iii.); ('J) ho prcsonls it iis a dciluction from the viTHcitv of (lixl [Pnnc. xxix., xxx.) ; {'A) lie rests it on an appeal to the cxperienee of his readers {Rt'ponscs (tux 11^" Ohjectiouii, Deniande vii.). The first two procedures appear to me obviously unsatis- factory ; ' I therefore propose only to consider the Empirical basis of the criterion. Let us ask, then, whether, when error occurs and we are convinced of it, in mathematical or lof tkuth and kkkoi: wise (lenionstrntt'il to ho erroneous. In the case of perception Oescartes expressly recognises tliis ; he speaks (.l/c(/. iii.) of the existence of things outside him exactly like his ideas as somethinf!: whicli " I thought I perceived very clearly, though in reality 1 did not perceive it all." In this case, however, the Empirical criteriou offers a guarantee against error l>y the rigorous separation of observation from in- ference. This guarantee I will now proceed to examine. I may begin by remarking a curious interchange of rdles between Rationalism and Empiricism as regards the evidence claimed for their respective criteria. While the Rationalist's criterion is partly supported, as we have seen, on an appeal to experi- ence, the validity of the Empirical criterion appears to be treated as self-evident. At least this seems to be implied in Mill's language before referred to ; where, after pointing out various possible sources of error in the affirmation that " I saw my brother this morning," he says that if any of these possibilities had been realised, " the affirmation that I saw my brother would have been erroneous : but whatever was matter of direct perception, namely, the visual sensations, ivould have been real." For his argument requires us to understand the last sentence as meaning not merely that there would have been sensations for me to perceive, but that my perception of them would certainly have been free from error : and as no empirical proof is offered of this last proposition, it seems to have been regarded as not requiring proof. CEITERIA OF TEUTH AND EEEOK 453 But — even if we assume, to limit the discussion, that a man cannot, strictly speaking, observe anything except his own states of consciousness — it still seems paradoxical to affirm that the elimination of all infer- ence from such observation would leave a residuum of certainly true cognition : considering the numerous philosophical disputes that have arisen from the con- flicting views taken by different thinkers of psychical experiences supposed to be similar. Take {e.g.) the controversy since Hume about the impossibility of finding a self in the stream of psychical experience, or that as to the consciousness of free-will, or the disinterestedness of moral choice, or the feeling-tone of desire ; surely in view of these and other contro- versies it would be extraordinarily rash to claim freedom from error for our cognitions of psychical fact, let them be never so rigorously purged of inference. The truth seems to be that the indubitable certainty of the judgment 'I am conscious' has been rather hastily extended by Empiricists to judgments affirm- ing that my present consciousness is such and such. But these latter judgments necessarily involve an implicit comparison and classification of the present consciousness with elements of past conscious experi- ence recalled in memory : and the implied classification may obviously be erroneous either through inaccuracy of memory or a mistake in the comparative judgment. And the risk of error cannot well be avoided by eliminating along with inference this implicit classi- fication : for the psychical fact observed cannot be 4r.-l C'KITKKIA OF Till Til AND KIHIOK distinctlv thouijht at nil wilhoiit it : if wc riiiorouslv purge it away, tluTc will \)c iiotliiuL; left save the cognitiou of self and of we eamiot say what psychical fact. Nay, it is douhtfiil whether even this nuich will be left for the Kmpiricist's observation : since he may ahare Hume's inability to find a self in the stream of psychical experience, or to maintain a clear distinction between psychical and material fact. Thus the Empiricist criterion, if extended to purge away com- parison as well as inference, may leave us nothing free from error but the bare atlirmatiou of Fact not further definable. Here again I am far from denying the value of the Empirical criterion. I have no doubt of the importance of distinguishing the inferential element in our apparently immediate judgments as far as we can, with a view to the elimination of error. Only the assertion that we can by this procedure obtain a residuum of certainly true cognition seems to mo neither self-evident nor confirmed by experience. I pass to examine the criterion propounded by Mr. Herbert Spencer in his Princij'yles of Psychology (part vii. chaps, ix.-xii.) : which, in his view, is appli- cable equally to particular and universal cognitions. It is there laid down that " the inconceivableness of its negation is that which shows a cognition to possess the highest rank — is the criterion by which its un- surpassable validity is known." . . . " If the negation of a proposition is inconceivable " — i.e. " if its terms cannot by any effort be brought before consciousness in that relation which the proposition asserts between CEITEEIA OF TKUTH AND EKEOK 455 them" — we "have the highest possible logical justi- fication for holding it to be unquestionable." This is, in Mr. Spencer's view, the Universal Postulate, on the validity of which the validity of all reasoning depends. Before we examine the validity of the criterion, the meaning of the term ' inconceivable ' requires some discussion. In replying to a criticism by J. S. Mill, Mr. Spencer — while recognising that * in- conceivable ' is sometimes loosely used in the sense of ' incredible ' — repudiates this meaning for his own use. But I agree with Mill in regarding this re- pudiation as hasty, so far as the criterion is applied to propositions that represent particular facts — e.g. " I feel cold." For in most cases in which such a state- ment is made it would not be true to say '* I cannot conceive myself not feeling cold," since only very intense sensation excludes the imagination or con- ception of a feeling opposite in quality. We might, no doubt, say, " I cannot conceive that I am not feeling cold " : but the form of this sentence shows that I have passed from conception, strictly taken, to belief. Spencer's contention that in this case the connexion of the predicate-notion " feeling cold " with the subject-notion "self" is for the time "absolute," though only " temporarily," seems to me to ignore the complexity of consciousness. According to my experience, disagreeable sensations, when not too violent, even tend to excite the opposite imagina- tion, e.g. great thirst is apt to be attended by a recurrent imagination of cool spring water gurgling 4r>G CKITKIMA OV TIM'TII AND I-IKKOK down luv ihrtMt. I cannot t luTi'Toro a^^rrc thai tlio utmost oortaintv in a ]>riij)()siti(»n representing a transient onipirieal lact involves the 'inconceivability' of its negation — except in a j)eculiar sense of the lerni in wliieh it is e(|uivalent lo ' intnitive in- credibility. It is, no doubt, otherwise in the case of universal propositions intuitively known — or, in Mr. Spencer's phrase, " cognitions in which the union of subject and predicate is permanently absolute." I cannot imagine or conceive two straight lines enclosing a space : here ' intuitive incredibility ' coincides with ' inconceiva- bility ' in the strict sense ; only either attribute must be taken with the qualification that I can suppose ray inability to conceive or believe to be due to a defect of my intellect. With this explanation, I shall allow myself to use Mr. Spencer's term in a stricter or looser sense, according as the cognition in question is universal or particular. I have no doubt that * inconceivability of negation,' so understood, is normally an attribute of propositions that appear self-evident truths; I think that, in trying to comprehend distinctly the degree of certainty attaching to any such proposition, we commonly do apply — more or less consciously — Mr. Spencer's test, and that a systematic application of it is a useful protection against error. But I think that the objection before urged against the infallibility of the Cartesian criterion applies equally to Mr. Spencer's. Indeed he admits " that some pro- positions have been wrongly accepted as true, because CEITEKIA OF TEUTH AND EEROE 457 their negations were supposed inconceivable when they were not." But he argues that this " does not disprove the vahdity of the test" ; chiefly because (l) *' they were complex propositions, not to be established by a test applicable only to propositions no further decomposable"; and (2) this test, like any other, is liable to yield untrue results, " either from incapacity or from carelessness in those who use it." The force of the second admission depends on the extension given to * incapacity.' Casual and transient in- capacity — similar to the occasional logical fallacies that occur in ordinary reasoning — would not seriously impair the value of the criterion ; but how if the historical divergences of thought indicate obstinate and wide-spread incapacity 1 Mr. Spencer seems to hold that this is not the case if we limit the applica- tion of the criterion to simple propositions ; thus he contrasts the complexity of the erroneous proposition maintained by those who regarded the existence of antipodes as inconceivable with the simplicity of the propositions that " embody the ultimate relations of space." But the proposition that " heavy things must fall downward" is apparently as simple as the pro- position that " two straight lines cannot enclose a space " ; and if analysis reveals complexity in the notions connected in the former proposition, this is equally the case with the latter, according to Spencer's own account of spatial perception : since, in his view, any perception of space involves " an aggregate of simultaneous states of consciousness symbolising a series of states to which it is found equivalent." 45S riMTKUIA OK TRUTH A X I > KllUOll The (litlioulty of applyiiiLi: tliis triterion is forcibly preseutod when wo cxiimiiir the jthilosophical doctrine to support wliicli it is esperially propounded. For Mr. iSpencer's primary aim in establishinre, that Mr. Spencer's Universal Postulate is inadequate to guarantee even the pri- mordial datum of his own philosophy ; and, on the whole, that — however useful it may be in certain cases — it will not, any more than the criteria before examined, provide the bulwark against scepticism of wliich we are in search. With this negative con- clusion I must here end. In a later article I hope' to treat the problem with which I have been dealing in a somewhat more positive manner. ' [Owing to the illness and death of the author some months later this hope was never realised ; but appended is the concluding portion of the second of two lectures entitled Verification of Beliefs, which probably furnishes in rough outline some part of what the later article would have contained. The lectures belong to a course on Metaphysics.] APPENDIX TO THE PRECEDING ESSAY On the whole, then, I have to reject the claims of Empiricism no less than of Rationalism to put forward a simple infallible criterion for the kind of knowledge which is to be taken as the ultimately valid basis of all else that is commonly taken for know- ledge. I regard both criteria as useful, as a means of guarding against error, but neither as infallible. I propose, then, to turn from infallible criteria to what I call methods of verification : from the search after an absolute test of truth to the humbler task of excluding error. One of these methods I call the Intuitive Verification. It includes as two species the Eationalist and the Empiricist criteria somewhat modified. They may be regarded as two applications of a wider rule : Assure yourself of the self-evidence of what appears self-evident, by careful examination. As regards universals, especially scrutinise both the clearness and distinct- ness of the notions connected in a judgment, and the intuitive certainty of their connexion. As regards particular judgments, especially purge observation of inference so far as reflection enables you to do this. These, I think, are valuable rules ; but even after they have been observed as carefully as they can be observed, we may be convinced of error through conflict of the judgment thus appar- ently guaranteed with some other judgment relating to the same matter which is equally strongly affirmed by us. This indeed is the most common way in which error is discovered. Such conflict does occur, even as regards the universal intuitions of reason or the conclusions demonstrated from them : indeed in this region it is sometimes obstinate and is then called an ' antinomy.' It is more familiar in the case of 461 462 t'KITKKlA (U' TKl'IIl AND KKHOR particular judgmenU — wbothor reluting to luiittor or tn niiiui. Hill ptMhaps tlu' most iniportaiit cuso of tin' kiinl is ii conflict hotwron a universal jiulgnient accepted as self evident, and tho jKirticular juiignionts of perception, or iMferenee from these. The fate of the belief that "a thing cannot act where it is not" may illustrate this. It was found to conflict apparently with tho hypothesis of universal gravitation, which rested on a multitude of ]>aiticular observations of the position of the heavenly bodies; and this h;ui, I think, destroyed any appearance of intuitive cer- tainty in it for most of us. And I may illustrate it further by the method by which in my work on Kthics Common Souse is led to UtiliUirianism.' This was, indeed, suggested by tho method of Socrates, whose ethical discussion brought to light latent con- flicts of this kind. It was evident (e.g.) to Polemarchus that 'it was just to give every man his own ' ; but being convinced that it is not just to restore to a mad friend his own sword, his faith in his universal maxim was shaken. - Now it is possible that what I have called the Intuitive Veri- fication might exclude error in some of these cases, one of the conflicting intuitions being due to inadvertence. If we had examined more carefully the supposed universal truth, or the supposed particular fact of observation, we might have detected the inadvertence, or at any rate have seen that we had mis- taken for an intuition what was merely inference or belief accepted on authority. But the history of thought shows that I cannot completely rely upon the Intuitive Verification alone. It seems, then, that the Intuitive or Cartesian Verification needs to be supplemented by a second, which I will call the Discursive Verification, the object of which is to exclude the danger of the kind of conflict I have indicated. It consists in contemplating the belief that appears intuitively certain along with other beliefs which may possibly be found to conflict with it. Of course we are always liable to obtain new beliefs which \\ill conflict with old ones ; therefore this verification is neces- sarily fallible. Still we may reduce the danger of failure by carefully grouping the intuitions that we see to be related, and ' Cf. Methods of Ethics, Bk. III. chaps, iii.-xi. '•' Cf. Plato's Republic, Bk. I. p. 331. APPENDIX 463 surveying them together in the most systematic order possible. It would, I think, be a gain if ethical and metaphysical writers would take more pains to state implicitly in the best attainable order the propositions they ask the reader to accept without proof. I may observe that among the chief of our par- ticular beliefs which we commonly regard as intuitively certain — those relating to the External World — there is a natural concatenation which enables us to dispense with an artificial one ; we may trust our ordinary physical beliefs with regard to the [roughly measured] size, shape, and relative position of familiar objects, because if we made a mistake we should find it out. The most noteworthy application of the Discursive Verification is to the relations between universal propositions which appear self-evident, and the particular beliefs which they implicitly include. We continually have this verification in the case of Mathematics, though in the case of Geometry only indirectly and approximately. We see universally and necessarily that two straight lines cannot enclose a space ; the lines we meet with in experience as boundaries are not exactly straight, but the more nearly straight they are the less space is it possible for two such lines to include, if they meet in two points. We might call this case of the Discursive Verification, Inductive Verification : it may be applied either to intuitive beliefs directly, or to beliefs demonstratively inferred from them. Comparing the Intuitive and Discursive Verifications, we see that while the former lays stress on the need of clearness, distinctness, precision, in our thought, the latter — the Discursive — brings into prominence the value of system. The gain of system in any part of our thought is not merely (1) that it enables us to grasp a large and complicated mass of cognitions, or even (2) that it prevents our overlooking any hiatus, or lapse through forgetful ness, which may be either important in itself or in its bearing on other cognition, but (3) that it provides against the kind of error which the conflict of beliefs reveals. And this, I may say, is the kind of service which Philosophy may be expected to render to the sciences. I have spoken of the history of thought as revealing dis- 464 CKITKKIA OF TlMTIl AXD KKllOR crojviucy In'twi'i'ii tlir intuit iuris of ono iigo and those of ;i suKsosovirv of rant ' i'ouvpiitioii ' lUiU ' Nuturo ' (f6fioi kiul ^>7i(\ 365 ; ilistiuotioii li«:i i'.niMli. 393 rnt.ri.«: of Truth, 4, 5, 8, 430467 ; .HUiuiiiary of, 466 ; the central prob- lem of Kpi.steinology, 430 ; of Gfiu-ral Con5»'nt, iliflicnltios involved in, 381 f., 385 ; of Imnie.lijuy, 383, an.l appeal to expt-rt-s on lH.-half of, 385, and olij.vtions to. 383 f., 390 ; Carteisinn (Kationalist) and Knipiri- cist, evidence for, 45'2 ; ditTero-nt 8C0i>e of the Empiricist and Cartesian, 444, 445 ; Empiricist, iis given by J. S. Mill. 44-2 f. ; of inferre.l beliefs, provided by Logic, 438 f. ; Intuitive, 318-319 Critical standpoint, 1 flT. CYude Realism, 293, 320 Curtius, 323, 357 Cynics, 337 Darwin, 432 Darwinian Theory, 413 Descartes, 11, 12, 178, 184 f., 198, 373, 112, 413, 420. 421, 433, 448, 449 f., 452 ; his Criterion of Truth, 319, 439, 441 ; this not disposed of by Kant's argument, 441, 442 ; his Disantrte on Method, 7, 449 ; his Meditations, 450, 452 ; his Prin- ciples of Philosophy, 450 ; his Answers to Objections, 450 Diogenes Laertius, 339 f., 341 n. Dionysodorus, 336, 342 Divine Mind (or Spirit) : for Green, ' 242, 244 f., 258-266, jB7 Eristic, 336 f., 339, 347, 348 ; only one kinil in Plato's view, 338 ; Plato's changing views of, 346 ; and Dialectic, 336 f., 346 f. ; and Sopliistic, 349, and Plato's use of tlie-se two terms, 346, 347 Euclid, 437 Euripides, 344 Euthydemus, 342 ' Experience ' : not explicable by ' Natural' history, 214 f . ; ways in which knowledge may lie 'founded on,' 374 ff. ; world of, How does it come to be for human ndnds ? Com- mon Sense and Kantian answers, 61 f. Externality and a priority, ambiguity of these notions, 38, 39, 40, 158 Fichte, his ' Absolute Ego,' 204, 206, 207 Finite minds and their relation to the material world, 282 f. First Cause, 279 f. ' Force,' change in meaning of, 378 f. Freedom, 169, 180 f. ; and moral con- sciousness, 169, 170 ; Divine, 173, 178 ; human, 170 ; practical and transcendental, 170, 172 Geometrical knowledge, Kant's view ol its relation to intuition, examined, 49-54 God: metaphysical idea of, 184 f. ; Existence of: ontological proof of, 184, 186 ; Kant's argument against onto- logical proof of, stated, 192, and ex- amined, 193-195 ; cosmological proof of, 188 f. ; physico-theological proof of, 189 ; speculative and practical proofs for, 180 f. ; Freedom and Im- mortality, 180, 182, and their relation to speculative and practical reason, 17-20, 23 (cf. 36, 37, Freedom) Gorgias, 325 f., 332 f., 347, 348, 351, 353, 367, 370 INDEX 471 Grant, Sir A., 324, 365 ; his Ethics of Aristotle, 365 n. Green, T. H., 1, 196, 209-266, jscwsm ; and Locke and Common Sense, 217, 238, 265 ; controversy of, with Sen- sationalism and Phenomenalism, 238, 239,243, 265 ; his treatment of Error, 225, 253 f. ; his relation to Kant, 227-230; his Spiritual (or Non- natural, or Eternal) Principle (or One Subject), 230 f. ; his SpiritualPrinciple (or Conscious Intelligence) criticised, 231 f., 250, etc., 264 ; this Conscious Intelligence the unifying principle in the world of reality, as well as in the Cosmos of Experience, 220, 222 f., 230 f., 240, 242, 243 f., 255, 258-266, passim ; his Metaphysical System is Idealistic and Spiritualistic Men- talism, 257 if. ; polemical aspect of his Metaphysical System, 265 ; in his view, Man a Free Cause, 248 f., 251 f., and self-conscious, 245, 251, 253, 264 f., and a composite or dual being, 258 f. ; his view of the relation of God (the Spiritual Principle) to man, 222 f., 243 f., 258, and to the world, 262 f. ; his Metaphysics and his Ethics cannot be reconciled, 263 ; can we really accept his account of Spirit, and does Green himself suc- ceed in thinking it ? 260 f. Grote, G., 323-371, jsassi/n. ; his Plato, 354 Hamilton, Sir W., 1, 196 f., 203, 268 and n., 276 ; and Mansel, 268, 279 ; his acceptance of Free- Will, 270 ; his Agnosticism, 268 f., 270 f. ; his metaphysical compromise, 270 ; his Natural Realism, 270, 271 ; his philo- sophical inconsistency, 272 ; his ' Philosophy of the Conditioned,' 268 and n., 279 ; his Primary, Secundo- primary, and Secondary qualities, 271 ; his edition of Reid's Works, 268 n. ; his Discussions on Philosophy, 272 n. ; his Disserta- tions in his edition of Reid's Works, 271 and n., 274 and n. ; his Lectures on Metaphysics, 270 n., 272 n. Hegel, 197, 198, 199, 393 Human consciousness and Eternal consciousness, relation between, in Green's view, 244 f., 245 n., 250 ; this view criticised, 245 f. Human sensibility, fundamental forms of, 62 Hume, D., 32, 217, 223, 408, 409, 410 f., 418, 425, 453, 454 ; and Spencer, 309 n. ; his treatment of Cause, 11, 79 ; Kant's estimation of, 11 ; his Inquiry into the Human Under- standing, 410 ; his Treatise on Human Nature, 309 n., 410, 415 Hutcheson, Francis, 406, 407 Idealism : repudiated by Kant, 203, 206 ; problematical and dogmatic, 30 Imagination, function of, 63 f. ' Immediate,' ambiguity of, 383 'Inconceivable,' meaning of, 317, 455 f. Inconsistency commonest sign of error, 461 (cf. Antinomy) Independent, 280 Infinite, 269 f. ; and Absolute (or Un- conditioned), \^&-2Q7 , passim, 279 f. Intuitional Metaphysicians, 386 Isocrates, 334, 345 f., 353, 358 ; his Encomium of Helen, Kara tCiv 'Lo(pifetaphvsical system,' meaning of, 275 Metaphysics: for Kant, 13, 14, 16, 21 ff. ; and Criticism, 6 ; Dogmatic, 22 ; chief question of, 282 ; com- pared with mathematical and phy- sical science, 4, 5. 6. 7, 8, 9, 22, 24, 25, 49 ; compared with Pure Mathematics, 54 f. ; criterion of, for Kant, 9. 10, 11 ; limitation of, for Kant, 9, 10, 11 Methods ol Verification : as opposed to infallible criteria, 461 f. ; questions in connexion with, 465 f. ; summed up, 465 Mill, James, 385 Mill, John Stuart, 196, 198, 385 ; his Ej-ximination of Sir ir. Hamil- ton's Philosophy, 198 n. ; his System of L»g%c, 317, 442 Milton, 413 Mind and Matter, relation of, 32, 33 Modality, 1 1(M27. ;«iMi'/i ; categories of ( I'oMtibility, .\cluftl Exi>tencc. nud Nciossity^ 11(5127 Moral Thcolojty, 202 ' Morally certain," 426 Nature, 'a prot-ess of change.' 225- 226 ; (for (Jreen). an ordered systeni of objecta. 224, 'a Mingle unalter- able all ■ inclusive .•■■ysteni of rela- tions,' 226, and iinplie.s a 'non- natural ' i>riiu'ij)lc, 226 f. ; (.Common Sense as-siinipfiou concerning know- le.-«: (for Kuitk 91. '169 f. ; (riaI), liS; wid Time, 31 f. ; ' MeUpliy- mcaI oxpiviiuou' of, 'JC f., SS f . ; ' lY»ji!k-enil««ut«l tfxpo.iitiiin ' of. 44 f. ; n«nii!ii»'!<>ti «■>( . .t> Ml, n.ert : 1, 196 f., 430. 4S5 f. ; *nni, '217 ; and Hunto, 309 n., 310 ; and Kant. 268, '217 ; *tid Monism, 287 and n., 307 ; and llealism, 310 f., 313, 320 f. ; and Sir W. Hamilton, '268 f., 277 ; his Agnosticism, 277 ff. ; and MenUlism, 285, 287, 298 ; and Natural Dualism, 285 f., 302 ; his Criterion (Universal Postulate), 296, 314. 316 f., 448 f. ; its inadequacy, 458-460 ; his Dualism, 283 ff. ; his doctrine of First Cause, 279 f., and of 'the Unknowable,' 276, 282 f., 286, 288 ; his Epistemological doc- trines, 308-321 ; his Metaphysical doctrines, 275-308 ; his Philosophi- cal datum, '2^6, 289, 458 ; his Philo- sophy, its scope and relations, 283 n. ; his 'Supreme or Ultimate Verity," 207. 281, 282, 287, 297-298 ; his use of Self, Ego, Not-self, and Non- ego, 458 f., and of the term 'Meta- physician,' 275 ; his \'iew of Force, 295 f., of Matter (Non-ego), 282 f., and Mind (Ego), 302 f., and of Logic, 289 f. ; his ' Transtigured Realism, ' 292 f., 299 f. ; his view of 'Reli- gious Ideas,' etc., 276 f., 281 f. ; on Space and Time, 32, 33 ; his ' vivid ' and 'faint' manifestations or states, and their equation to ' Object ' and 'Subject,' Non-ego and Ego, 284 f., 285-286, 292 f., 296 f., 298 f., 303 f. ; Philosophy in his view con- cerned with phenomena, 283 ; why his system is called Phenomenal- ism, and Agnosticism by the author, 283 and n. ; his First Principles, 275 ff. ; his Principles of Psychology, 303 ff. Spinoza, 11 Stallbaum, 355 Subject and Object in cognition, 232 f. ; Green's view of, criticised, 233 f. Subjective Method, 388 SulMt*nce,64,98 f. ; atidCiiuso, schoma- tisin of. 64 f., 86 f. ; primiplo of tilt" jteriiiniuMu'p of. 1*9 105 Sui-ccsxiou, oljfctive iiud subjective, 108, 109 Succeiuiive apprehuusiou of phenomena, 102, 107 f. Tlialfs 276 Thi'ology, 403 f. ; and practical in- terests, 179 f. ; and the thesi.H of the Antinomies, 182 ThiugiuiUsclf, 73, 201, 203 f. Thompson. Dr., 323 f., 345 ; his Oor- gias, 367 n. Thought : ami Feeling. Green's view of the relation between, criticised, 265, '266 ; aud Reality, relation of, 185 f. "nme, 101, 269 f."; (real). 158; and Change, 109 ; and Common Sense, 392-405 ; and Numl^er, 56, 57 ; and Space, Are tliey entities of relational quality, or merely forms of sensi- bility '( 33, 34, 35 ; consequence of regarding it a.s a form of human sensibility, 35, 36 Transcendental ..Esthetic, 21 f. ; the two main points of, 31 TransceudenUl Analytic, 24, 25, 26, 28, 58 f. ; subject and scope of, 57 f., 61 ; problem of, 58, 61 Transcendental Dialectic, 26, 128-142 ; its aim, 23 Transcemlental Ideal, the, 187 f. Transcendental Idesis of Reason, the, 134 ff., 153 Transcendental Illusion, the, 134 Transcendental Reality, 62 Transcendental schematism of the Categories, 60 f., 68, 85, 86 Transcendentalism, 196 Ultimate beliefs, verification of, 439 f. Ulysses, 359 Unconditioned (or Absolute), notion of : not applied to God in pre-Kantian philosophy, 198-199; in Kantian thinkers, 199 ; in Kant's philosophy, 199 f. ; its speculative use only regulative, 200, 201 Understanding, function of, 63 f., 66 f., 78, 135 ; Kant's forms of, 26, 63 f. Unity of Apj)erception, Transcendental : its function, 67, 68 ; its import- ance, in Kant's view, 146, 147 Universe, origin of, 277 f. Verification, methods of, 461 ff. ; In- INDEX 475 tuitive or Cartesian, 461 f. ; Dis- cursive (including Inductive), 462 f. ; Social or Oecumenical, 464 f. Volitionism, 265 Vorstellung, 73 Warburton, 411 Watson, P*rof. : his Philosophy of Kant KS contained in Extracts from his aion. Writings ( = Selections from Kant), 22, etc. Welcker, 323, 346 Wolff, 141, 142, 144, 148, 152, 157. 158. 163. 183, 184 ; and Kant, 11, 12 ; his PhDosophic system, 132 f. Xenophon, 328, 329, 358, 363. 364 Zeller, 356. 370 THE END UC SOUTHERN REGIONAL LIBRARY FACILITY A A 001 416 065 CEMRAL UNIVERSITY LIBRARY University of California, San Diego DATE DUE i 7 1972