6 BOUGHT WITH THE INCOME FROM THE SAGE ENDOWMENT FUND TEfE GIFT OF 1891 Mk5M3.. M/A/..lcf..m 5474 r DECi3i^4g/(| My » '40 ^^8'^ l^e ^f^'V. 2 1950 * MHm ■^'^^^^T^^iP^ AUG 3 lS4o ^y- 6 1948 M MAY2 0m8 li Cornell University Library B808 .F62 olin 3 1924 032 325 437 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924032325437 AGNOSTICISM pw^-vv BY THE SAME AUTHOR Agnosticism. Crown 8vo. la. 00 net. Postage, 18 cents. Sermons and Addresses. Svo. $2.50. Theism. The Baird Lecture for 18T6. Seventh Edition, Eevised. 12mo. $1.60 net. Anti-Theistic Theories. l2mo. $4.20 net. History of the Philosophy of History. Historical Philosophy in France and French Belgium and Switzerland. Svo. $4.00. AGNOSTICISM BY ROBERT FLINT D.D., LL.D., F.H-.S.E. CORKESPONDING MEMBER OP THE INSTITUTE (SF PRANOB H02T0BABY MEMBER OF THE ROYAL SOCIETY OF PALERMO ; ANI> PROFEBSOB IN THE UNIVERBITY OF BDINBDBQH NEW YORK CHARLES SCRIBNER'S SONS 1903 T 2'/ B F62- Copyright, 1902, 1903, bt CHAKLES SCRIBNER'S SONS All rights reserved Published, January, 1903 THOW DirEOTORY PRINTINQ AND BOOKBINDINQ COMPANY NEW YORK CONTENTS Chapter I.— THE NATURE OF AGNOSTICISM. I. Origin, Original Application, and Defects op the Term . 1-10 Hutton and Huxley on origin of the term 1-3 Critical remarks on the term and its applicatio n . 3-6 FaTourable reception of the word : its causes 6-9 Tribute to Huxley ... 10 II. Agnosticism an Epistemological Theory. Equiva- lent TO Philosophical Scepticism . 10-24 Questions as to limits of knowledge . 11-U On the differentia of the term agnosticism 15 Likeness to the term socialism 16 Definition must be relative .... 20 The duty thus implied ..... . 20-22 Correct use of the term .... . 22-24 III. Agnosticism and Gnosticism .... . 24-37 How related ... ... . 24-26 Merits and defects of both .... . 26-27 Truthfulness . 28-30 Faith and doubt . 30-34 Function of doubt . 34-35 Our knowledge imperfect .... . 35-36 Lesson for us . 36-37 vii CONTENTS Chatter II.— EEEONEOUS VIEWS OF AGNOS- TICISM. PAGES I. Not Equivalent to Truth— Search ok Honesty in Investigation 38-46 Owen on dogmatism and scepticism . . . 38-42 Huxley on agnosticism and creed . . . 42-46 II. Not Equivalent to Know-Nothingism. Relation OF Agnosticism to the Theory of Nescience 46-50 Agnosticism and doctrine of nescience . . . 46-48 Agnosticism only a special theory of nescience . 48-50 III. Not Necessarily Atheism, although Atheism is OFTEN Agnosticism. Dr. Bithell's Position 50-54 Agnosticism not atheism ..... 50-52 Atheist's relation to agnosticism : opinion of Dr. Bithell 52-54 IV. Not to be Identified with Positivism. Pro- fessor Eraser 64-58 Agnosticism not positivism 54 Professor Eraser's view. Note .... 55 Positivism and theory of knowledge . . . 56-58 V. Not to be Identified with Denial of the Cog- NOSCIBILITY, accompanied WITH AFFIRMATION OF THE EeALITY, OF THE ABSOLUTE. PRO- FESSOR Caldekwood 58-62 Professor Calderwood's view .... 59-62 VI. Roberty's Views on the Nature of Agnosticism Stated and Criticised 62-71 General statement of his views .... 62-68 Originality and ingenuity of them . , . 68-69 Defects -71 VII. Criticism op Leslie Stephen's Views op Agnos- ticism 71_80 His application of the name 71-72 Its unfairness ....... 72-76 On misuse of certain terms 76-80 viii CONTENTS Chapter III.— HISTOEY OF AGNOSTICISM. I. Introductory. Oriental Agnosticism Agnosticism should be studied historically Origin and early stages of agnosticism Oriental agnosticism— Hebrew, Chinese, Hindu II. Greco-Roman Agnosticism. Pre-Socratic or Preliminary Period Iniluence of Eleatic, Heraclitean, and Material^ istic philosophy on Greek scepticism The sophists and agnosticism III. Post-Socratic or Developed Period Pyrrho, Arcesilaos, and Carneades as sceptics JEnesidemus and the ten tropes . JEnesidemus and causality Sextus Empiricus and the Pyrrhonic Institutes Greek scepticism not . specially antagonistic to religion or morality .... PAGES 81-88 81-82 82-84 81-88 88-94 88-92 92-94 95-108 95-101 102-103 104-105 105-106 106-108 IV. Middle Ages 108-111 How far agnosticism was present in the medie- val world 108-110 V. First Period of Modern Agnosticism. Causes and Characteristics 111-116 Causes of agnosticism in the period . . . 111-113 Characteristics of the agnosticism of the period 113-116 VI. Representative Agnostics of the Transition Period 116-135 Agrippa of Nettesheim 116-118 Montaigne 119-124 Charron , 124-125 Sanchez . 125-128 La Mothe Le Vayer 128-129 Glanville, Hirnhaim, and Huet .... 129-132 Pascal 132-133 Bayle 133-135 ix CONTENTS Chapter IV.— AGNOSTICISM OP HUME AND KANT. I. Hume : Prefatory as to his Agnosticism Influence of Hume. How Huxley, Riehl, and Compayre have overlooked his scepticism Sources of Hume's philosophy and scepticism Relation of his agnosticism to his psychology, Hume and Locke II. Hume's Agkosticism in General His seusism and views on substances Hume on consciousness of self and causality III. HnME's Agnosticism in Religion His personal attitude towards religion Thoroughness of his anti-theological seep His Natural History of Religion His Essays on Miracles Final issue of his scepticism IV. Kant's Answer to Hume Relation of Kant to Hume Kant's Transcendental Esthetic stated Criticism of it : its defects Kant's Transcendental Logic ; (.4) Analytic General statement of ... Hume and Kant compared. Superiority of latter Criticism of Kant's criticism of knowledge Kant's Transcendental Logic : (B) Dialectic His doctrine of pure reason stated and criticised His " Rational Psychology " partly true, partly erroneous .... His " Rational Cosmology " in the main a failure His " Rational Theology " ingenious and self-con- sistent but inconclusive . . His criticism of the theistic arguments examined Its general defects Worth of the cry, " Back to Kant " X PAGES 136-143 136-138 138-140 140-143 148-155 143-150 150-155 155-168 155-158 159-161 161-162 162-166 166-168 168-238 168-170 170-175 175-184 184-198 184-188 188-190 190-198 198-237 198-206 206-210 210-216 216-230 230-232 232-235 235-238 CONTENTS Chaptek v.— complete OE ABSOLUTE AGNOSTI- CISM. PAGES I. Agnosticism not Exactly Divisible. Its Gen- eral Divisions 239-243 No strictly absolute or total agnosticism . . 239-240 No merely partial or modified agnosticism . 240-241 Modification and limitation imply each other . 241-243 II. Why' Complete Agnosticism Requires to be Discussed. Criticism or the Views of Padlsen 244-253 Complete agnosticism should be first considered 244 Paulsen's reasons for disbelief in a sceptical theory of knowledge ... . 245-247 They are contradicted by the history of philosophy 247-249 Hume misunderstood 250-251 Absolute agnosticism not extinct . . . 251-252 Is a distinct form of solving the problem of knowledge 252-253 III. Species or Complete Agnosticism. Inconsist- ency OP Systematic and Universal Doubt 254-265 The agnosticism of universal doubt self- contra- dictory 254-261 Relationship of doubt to belief and disbelief. Implies the existence of a self-contradictory world 261-265 IV. Inconsistency of Systematic and Universal Dis- belief 265-268 Warrants the inference only of negations . 266-268 V. Absolute Agnosticism and First Principles . 269-279 Antagonistic .... . . 269-270 Its relation to philosophy and to first principles and necessary laws of thought . . 270-275 Its claim to be incapable of refutation . . 276-278 Assent to first principles not mere belief but ac- ceptance of self-evidence . . 278-279 VI. Absolute Agnosticism and Practical Life . 279-286 The former inconsistent with the requirements of the latter 279-284 Is ethically unsatisfactory .... 284-286 xi CONTENTS Chapter VI.— ON MITIGATED AND PARTIAL AGN08 TICISM AND THEIE FOEMS. Absolute agnosticism a false and unattainable ideal 287-288 I. Mitigated Agnosticism. Its Underlying As- sumptions . . ... 288-300 Has always been effected through an illegitimate combination of dogmatism with scepticism -. 288-290 That shown in Pyrrhonism. Its inconsistency 290-295 Its separation and contrast of reality and appear- ance, knowledge and opinion . . . 295-298 Pyrrho on reasoning. His scepticism and dog- matism 298-300 II. Hume on Mitigated and Absolute Scepticism 300-309 Professedly an Academic not a Pyrrhonian scep- tic. His view of their relationship . . 300-304 Eeid's position towards scepticism more reason- able 304-306 Hume clearly recognised an absolute scepticism to be ruinous, but erroneously imagined it might lead to a useful mitigated scepticism . 306-309 III. Partial ok Limited Scepticism : its Fokms and their Inter-Relations .... 809-334 Classification of agnosticisms .... 309-312 Religious and anti-religious agnosticism . . 312-314 How the agnosticism which originates in faulty theorising bears on religion .... 314-316 Mathematics and agnosticism .... 316-318 Scepticism as to physical science . . . 318-319 Historical scepticism ..... 319-322 Ethical agnosticism 322-324 Metaphysical agnosticism 324-328 How it bears on theology ..... 328 How its forms may be grouped with reference to the powers or principles of mind . . 328-329 Scepticism as to the testimony of the senses . 329-330 Scepticism as to memory 330-333 Scepticism as to reason 333-334: xii CONTENTS Ohaptee VII. — PARTIAL OR LIMITED AGNOSTI- CISM AS TO ULTIMATE OBJECTS OP KNOWL- EDGE. PAGES Prefatory Remarks 335-336 I. On Assigning Limits to Knowledge . . . 336-355 Not to be ascertained a priori or by mere intro- spection 336-338 The proper place and function of epistemology 338-340 Scepticism as to philosophy .... 340-341 Scepticism as to physical science . . 341-343 Philosophy should provide on epistemology . 343-346 Claims of theology to fair treatment . . . 346-348 Truths as to the limitation of knowledge . . 348-350 Dr. Bithell on the subject . . . 350-353 On a favourite quotation from Tennyson's An- cient Sage 353-355 n. The Ultimate Objects op Knowledge . 355-369 " Self," " World," and " God " not unambiguous terms,— especially " Self " and " World " . 355-358 Perhaps " Space " and "Time " should be added as ultimates of knowledge .... 359-369 m. Agnosticism and the Self .^69-388 Agnosticism impossible within the sphere of self- consciousness strictly so-called, but that sphere is narrow . 369-371 Yet without self-consciousness we could have no knowledge even of our own minds, nor of other selves, nor of comparative psychology . 371-374 Hume's sceptical treatment of consciousness . 374-375 Also H. Spencer's 375-378 Broussais' polemic against consciousness . . 379-381 Exaggeration of the dependence of consciousness on physiological conditions . . . 381-382 Alleged loss of consciousness of self-identity 383-384 Knowledge even of self is limited and defective and involved in mysteries like those connected with a knowledge of God and of nature . . 384-388 xiii CONTENTS PAGES IV. Agnosticism as to the World .... 388-417 Greatness and littleness of nature . . . 388-390 Are other Worlds than the earth inhabited . 391-393 Nature-worship and Hylozoism .... 393-394 No absolute knowledge of matter . . . 394^396 Agnosticism as to an external world . . . 396-402 Inconsistency of materialism .... 402-404 Knowledge of matter not proved to be the most certain . 404-407 Parallelism between knowledge of God and of the world. Thompson's Christian Theism . 407-408 Huxley and Spencer not materialists . 408-410 Objections against knowledge of God and knowl- edge of matter substantially identical . . 410-411 Imperfections of physical science . . . 411-414 Reasons for accepting it 414-417 Chapter VIII.— AGNOSTICISM AS TO GOD. I. Introductory Remarks on Agnosticism as to God . ... . . 418-427 All agnosticism as to religion refers to the object of rehgion .... . 418-422 May be religious, anti-religious, or non-religious 422-423 How those forms are related .... 423-427 II. Prevalence of Anti-Religious Agnosticism . 427-485 Contrast of past and present in the history of agnosticism . . .... 427-429 The change largely unjustified .... 429-433 Mr. A. Balfour's demands on so-called scientific agnostics essentially just . . . . 433-435 III. Some Causes of Prevalence op Anti-Religious Agnosticism . .... 435-451 Critical temper and scientific spirit of the age to some .extent a cause ..... 435-438 Also want of religious susceptibility and disci- pline 438-443 xiv CONTENTS PAGES The preyalence of anti-religious feelings and passions 443-44:5 That evil has. often favoured the spread of relig- ion of a kind must be granted . . . 445_446 A main cause of anti-religious agnosticism is non-recognition of the practical and ethical character of religion 447-451 IV. " The Will " and " the Wish " to Believe . 451-456 The views of William James and Wilfrid Ward not the full truth 461-455 Chapteb IX. — AGNOSTICISM AS TO RELIGIOUS BELIEF. Power of belief 457 I. Theories as to Belief ..... 458-477 Importance to the theologian of a theory of be- lief 458-460 What belief is and is not 460-463 Hume's theory of belief 463-465 James and J. S. Mill on belief .... 466-469 Dr. Bain's theory of belief stated and criticised 469-477 II. The Sphere of Belief 478-486 III. Christian Faith in Relation to Belief . . 486-491 IV. Why Belief as to Religion is so often False 492-499 It is to a vast extent false 492-494 Some of the chief causes why it is so indicated . 494-499 V. The Sceptical Inference from Prevalence OF False Religious Belief Erroneous . 499-510 The sceptical inference indicated . . . 499-500 As relevant against doubt and scepticism as against belief and dogmatism. Also an exces- sive generalisation 500-501 The distinction betweeen reasons and causes of belief is an indefinite and confused one. Re- marks on the point ..... 502-503 XV CONTENTS Reasons are also causes of belief, good reasons its only true causes ..... 604-505 Eeasons which are the causes of the progressive truthfulness of religion, and causes which are conditions of its reasonableness . . . 505-508 The Supreme reason 608-510 VI. True Inferences from Prevai-ence of Fai.se Keligious Belief 610-516 Need for caution in the formation of religious beliefs and for guarding against unreasonable doubt or excessire unbelief .... 510-512 Also need of concerning oneself chiefly with what is essential and vital in religion . . 512-513 What is meant by the reason to which belief ought to be conformed 514-515 VII. Bases of Agnostic Religious Belief in Christianity ... ... 616-526 History of opinion as to relationship between rea- son and faith 516-518 Religious belief and Christian faith based on various grounds ...... 618-521 On attempts to rest belief on itself . . . 622-523 Storrs Turner's view 523-526 VIII. Religioos Belief and Transmitted Common Doctrine and General Consent . . 526-536 Significance of the traditional factor in belief . 526-528 Leslie Stephen, J. H. Newman, and H. Spen- cer's views on subject 528-530 Protestant and Catholic estimates of tradition . 530-533 Theories of De Bonald and De Lamennais . 533-536 rX. Relation of Character to History of Belief 636-539 X. Belief in Relation to Authority. Forms of Religious Authority 539-551 Nature, value, and influence of authority . . 639-640 Personal authority ... ... 641-543 Church authority 643-546 Authority of Bible and external evidences . 646-551 xvi CONTENTS Chapter X.— AGNOSTICISM AS TO KNOWLEDGE OP GOD. PAGE I. A Glance AT THE HisTOKT OF Religious Knowledge 552-577 Religion as subjective fact and historical phe- nomenon 652-555 Dirersity of opinions as to starting-point and earliest stages 555-556 Polytheistic religions 556-558 Monistic religions (Egyptian, Chinese,and Hindu) 558-564 Dualistic religion (the Mazdean) . . . 564-566 Monotheistic religions (Jewish, Christian, and Mohammedan) 566-573 Lessons from the history 573-577 II. In what Senses Knowledge op God is not At- tainable 578-585 ComprehensiTe knowledge not attainable . 578-580 Nor non-relative knowledge .... 580-583 Nor apart from divine self-manifestations . 583-585 III. Agnostic Positions Relative to Knowledge oe God 585-603 That such knowledge is only to be attained through special revelation (Bibliolatry) . . 585-589 Not warranted by the Bible . . . 590 That religious knowledge is unlike all other real knowledge (Symbolism, Ritschlianism, &c.) . 590-596 Sabatier's " critical theory of religious knowl- edge " 596-598 His distinction between natural and religious knowledge not valid 598-599 Nor his distinction between scientific and relig- ious knowledge 599-603 IV. The Agnosticism op Hamilton, Mansel, and Spencer 604-639 I. The Agnosticism of Hamilton . . . 604-621 Relation of Hamilton to Kant . . . 604-607 Criticism of his fundamental principles . 607-621 xvii CONTENTS PAGES Relativity of knowledge — how far true, how far erroneous . . . ' . . 607-610 To think is to condition — how far true, how far erroneous ...... 610-613 Incognoscibility of the absolute and infinite not justified 614-621 II. Agnosticism of Maneel .... 621-629 Relation to Hamilton 621-622 His chief errors ...... 622-629 III. Agnosticism of Spencer .... 629-639 Relation to Mansel 629-630 Defects of his first principles and of his infer- ences from them 630-639 Present Work Part of a System of Natural Theology . 640-664 Influence of Kantian agnosticism on theology . 641-643 Indirect influence of 643-644 Merits and defects of the philosophy of the ab- solute 644-651 The proofs and bases of theistic belief . . 651-656 Advance of trinitarian and unitarian theism . 656-660 Place of the idea of God in human thought and experience 660-663 Agnosticism as to God likely to be long prevalent 663-664 AGNOSTICISM CHAPTEK I THE NATURE OF AGNOSTICISM I. ORIGIN, ORIGINAL APPLICATION, AND DEFECTS OF THE TEEM OuE study of agnosticism may appropriately begin with an inquiry as to the nature or kind of thought so designated. What, then, ought we to mean by agnosticism? The name itself should so far help us to an answer; and even if it be found not directly of itself to aid us much, we may be indirectly profited by an ex- amination of it. It is a comparatively new term, being little more than thirty years old. , It was preceded by the word " agnostic," as to the date of the invention of which we have very precise information. According to Mr. E. H. Hutton, this latter word was "suggested by Professor Huxley, at a party held previous to the formation of the now defunct Metaphysical Society, at Mr. James Knowles's house on Clapham Common, one evening in 1869, in my 1 THE NATUEE OF AGNOSTICISM hearing. He took it from St. Paul's mention of the altar to ' the unknown God.' " ^ /'Professor Huxley's own account of the matter is as follows : " When I reached intellectual maturity and began to ask myself whether I was an atheist, a theist, or a pantheist ; a materialist or an idealist ; a Christian or a freethinker, I found that the more I learned and reflected the less ready was the answer, until at last I came to the conclusion that I had neither art nor part with any of these de- nominations except the last. The one thing in which most of these good ' people were agreed was the one thing in which I differed from them. They were quite sure that they had attained a certain ' gnosis ' — ^had, more or less successfully, solved the problem of existence; while I was quite sure that I had not, and had a pretty strong conviction that the problem was insoluble. And with Plume and Kant on my side, I could not think myself presumptuous in holding fast by that opinion. . . . This was my situation when I had the good fortune to find a place among the members of that remarkable confraternity of antagonists, long since deceased, but of green and pious memory, the Metaphysical Society. Every variety of philosophical and theological opinion was represented there, and expressed itself with entire openness; most of my colleagues were -ists of one sort or another ; and however kind and friendly they might be, I, the man without a rag of label to cover himself with, could not fail to have some of the ' Murray's New English Dictionary, s.v. 2 HISTORY OF THE TERM uneasy feelings which must have beset the historical fox when, after leaving the trap in which his tail remained, he presented himself to his normally elon- gated companions. So I took thought, and invented what I conceived to be the appropriate title of ' agnostic' It came into my head as suggestively antithetic to the ' gnostic ' of Church history, who professed to know so much about the very things of which I was ignorant ; and I took the earliest oppor- tunity of parading it at our Society, to show that I, too, had a tail, like the other foxes. To my great sat- isfaction, the term took ; and when the Spectator had stood godfather to it, any suspicion in the minds of respectable people, that its parentage might have awakened, was, of course, completely lulled. That is the history of the terms ' agnostic ' and ' agnosti- cism.' " ' ^ The foregoing statements of Mr. Hutton and Pro- fessor Huxley well deserve to be borne in mind, but they may also perhaps be usefully supplemented by the following remarks. 1. When Professor Huxley took the term " agnos- tic " from St. Paul's mention of the altar to " the un- known God," he did not adhere very closely to the original. That was arfvaxn^ Bern, not oifvmaTiKtp 0ew.. There is a Greek adjective yvma-TiKoi;, but not an arfveoaTiKo\^ .'; ,. , ,. ■-■ ■ ,[. i • 1. The agnostic is sometimes described as one ivJio does not hnow, and agnosticism as Jinciv-nothingism, " a know-nothing creed." This account of the agnos- tic and agnosticism is not unfrequently to be met 46 AGNOSTICISM AND NESCIENCE with in a certain kind of religious literature. It is nevertheless a misrepresentation and caricature. Not to know is merely to be ignorant, and to know nothing is merely to be completely ignorant. But merely to be ignorant is not to be an agnostic ; igno- rance, even if it were complete, would not be agnosti- cism. The new-born child is ignorant, but it is not an agnostic. The agnostic is not only one who does not know, but one who has convinced himself that the human mind lacks the powers necessary to enable it to know. Agnosticism is a learned ignorance based on self-knowledge and philosophical reflection. Besides, there are very few agnostics who profess to know nothing, and to be unable to know anything, unless it be in a special sense of the word " know," which so alters the meaning of the statement as to make it harmless or even true. There is a sense in which no man does know anything. lie knows only as a man may know ; he does not know as God knows ; he does not know completely, or with a full and infal- lible knowledge, anything in its whole nature and en- tire relationships. " If any man think," says St. Paul, " that he knoweth anything, he knoweth noth- ing yet as he ought to know " (1 Cor. viii. 2). There we have St. Paul declaring that any true knowledge we can have — the knowledge that we all ought to have — is a consciousness of knowing nothing. But he did not thereby proclaim himself an agnostic, anx- ious to convert all men to agnosticism. He only ex- pressed his sense of the imperfection of human knowl- edge, and his desire that all men humbly so felt its 47 EKEONBOUS VIEWS OP AGNOSTICISM imperfection as not to be unduly proud of it, which they are very apt to be, and indeed generally the more apt the more imperfect it is. Accordingly his remark- able and profound declaration was immediately pre- ceded by the words — " We know that we all have knowledge. Knowledge puffeth up, but love edi- fieth." Agnosticism and the profession of complete inabil- ity to know are, then, not to be identified. But neither are they to be entirely separated. The com- pletest agnosticism must be that which allows to the mind least capacity for knowledge. An agnosticism which succeeded in showing that man is wholly des- titute of power to know would be perfect as agnosti- cism. But such perfection agnosticism has never at- tained, nor can reasonably hope ever to attain. A demonstrated ignorance is at least not ignorance so far as the demonstration is concerned. As all prov- ing involves knowing, the proving that there is no knowing is a sort of proving which is inherently self- contradictory. The necessity of self -justification is for agnosticism a necessity of self-limitation. 2. It is desirable to have a clear view of the rela- tion of agnosticism (scepticism) to the doctrine of nescience. The relation is not unfrequently one of identity with what is called the doctrine of nescience, the designation often meaning merely the agnostic or sceptical doctrine of nescience, — such a doctrine as undertakes to show that what all except agnostics (sceptics) suppose to be knowledge (science) is really ignorance (nescience), — unsupported and unverified 48 DOCTEINES OP NESCIENCE belief, a blind faith produced by non-rational causes. The doctrine of nescience of Aenesidemus and Sextus Empiricus, of Ilirnhaim and Iluet, of Hume and Maimon, for example, is often called a doctrine of universal nescience, and it is so, but only in the sense indicated, — one in which it is plainly identical with universal scepticism, absolute agnosticism. So the doctrine of nescience of Auguste Comte and Thomas Huxley, which while professing to prove man's neces- sary ignorance of all that lies beyond the sphere of sense-perception admits a possible knowledge of all that lies within it, is at once a doctrine of partial nes- cience and a form of partial agnosticism or scepti- cism, for the simple reason that it is an agnostic or sceptical doctrine of nescience. But a doctrine of nescience has no exclusive or special connection with agnosticism or scepticism. The most extreme gnosticism, the most uncompromis- ing dogmatism, implies a doctrine of nescience no less than does the most thorough agnosticism, the most resolute scepticism. A doctrine of nescience may, in a word, be either gnostic or agnostic, or neither gnos- tic nor agnostic. And, it should be observed, that in endeavouring to reach a true doctrine of nescience we should prosecute our investigations unbiassed by a desire that it should be either the one or the other, or the one more than the other. In itself a doctrine of nescience is simply a reasoned answer to the ques- tion. What are the limits beyond which, and the con- ditions failing to comply with which, the mind of man necessarily wanders in ignorance and error ? In 49 ERRONEOUS VIEWS OF AGNOSTICISM other words, it is an essential part or necessary com- plement of the theory of knowledge, or, as it is com- monly called, epistemology. Thus understood, ag- nosiology or agnoiology is a sufficiently appropriate designation for a doctrine of nescience, but agnosti- cism is a very misleading one. Agnosticism is only a special theory of nescience, the sceptical theory. If, in this special sense, it is a doctrine of universal nescience, it is complete agnos- ticism, but if a doctrine of nescience only within a particular sphere of belief or inquiry, it is partial agnosticism. It would be decidedly advantageous, however, if by the doctrine of nescience were always meant not agnosticism but agnoiology ; — not a delib- erate endeavour to prove knowledge in whole or in part unattainable, but an impartial inquiry as to when and where it ceases to be attainable. III. N^OT NECESSARILY ATHEISM^ ALTHOUGH ATHEISM IS OFTEN AGNOSTICISM. DE. BITHELL^S POSITION Agnosticism is not to be identified with a know- nothing position in religion or as to the object of religious faith and worship. This is often done in popular religious discourse and literature, but it is an error in defence of which little can be relevantly said. Agnosticism is properly a theory about knowledge, not about religion. It may be about religion, for it may doubt or deny that we can Imow spiritual truth ; it may even be exclusively about religion, for it may doubt or deny the attainability of no other kind of 50 AGNOSTICISM NOT ATHEISM truth than spiritual truth. Recent agnosticism has been in a large measure agnosticism only as to the truth implied in religion and indispensable to its vindication. But religion may be held to be the one thing which may be best known, or even the only thing which can be truly known; all else, it may be contended, is illusion and error. In India philo- sophic thought has been agnostic in hardly any other sense than this. The Greek sceptics were not more sceptical as to religious than as to empirical or ethi- cal truth : their agnosticism was universal, or nearly so, — not specially anti-religious. It has often been attempted to show that nature and reason are untrust- worthy, with a view to induce men to accept revela- tion with unquestioning faith. This procedure is none the less agnostic because undertaken in support of religious authority. It follows, even from what has just been said, that agnosticism is not atheism. Agnosticism is some- times spoken of as only another name for atheism, or as a kind of atheism. This should never be done. Agnosticism may be combined with atheism as it may with Christianity, but it is no more atheism, or a kind of atheism, than it is Christianity, or a kind of Chris- tianity. A theist and a Christian may be an agnostic; an atheist may not be an agnostic. A man who believes that God can be known, but not that an external world can be known, is as much an agnostic as a man who believes that an external world can be known, but not that God can be known. An atheist may deny 51 EEKONBOUS VIEWS OF AGNOSTICISM that there is a God, and in this case his atheism is dog- matic, not agnostic ; or he may refuse to acknowledge that there is a God simply on the ground that he per- ceives no evidence for His existence, and finds the arguments which have been advanced in proof of it invalid: and in this case his atheism is critical, not ag- nostic. It consequently always shows want of clear- ness of mind, and sometimes, it is to be feared, it shows uncharitableness of heart, to treat agnosticism as equivalent to atheism. The atheist may however be, and not unf requently is, an agnostic. There is an agnostic atheism or atheistic agnosticism, and the combination of atheism with agnosticism which may be so named is not an uncommon one. While, therefore, it is unwarranta- ble and unjust to identify agnosticism and atheism, the accuracy of a passage like the following, taken from an exceptionally interesting agnostic treatise, cannot be admitted : " An agnostic is not an atheist. Positive, dogmatic atheism is as repugnant to the sentiments of the true agnostic as any of the false cer- titudes embodied in the professions of religious sects. He usually knows quite as much of God, immortality, the soul, as most other men; but he does not pre- tend to know what he does not and cannot know, nor does he dignify with the name of knowledge what is perhaps at best a mere traditional belief, inca- pable of proof, and unverified by experience. The atheist does the contrary of this. The man who says, ' There is no God,' makes a universal declara- tion which assumes an amount of knowledge, and 53 ATHEISM AND AGNOSTICISM knowledge of such a kind as never was possessed by any human mortal." ^ l^ow such an account of the atheist is just as much a misrepresentation as is that of the agnostic to which objection is taken. The atheist is not necessarily a man who says " There is no God." What is called positive or dogmatic atheism, so far from being the only kind of atheism, is the rarest of all kinds. It has often been questioned whether there is any such thing. But every man is an atheist who does not believe that there is a God, although his want of be- lief may not be rested on any allegation of positive knowledge that there is no God, but simply on one of want of knowledge that there is a God. If a man have failed to find any good reason for believing that there is a God, it is perfectly natural and rational that he should not believe that there is a God ; and if so, he is an atheist, although he assume no superhu- man knowledge, but merely the ordinary human pow- er of judging of evidence. If he go farther, and, after an investigation into the nature and reach of human knowledge, ending in the conclusion that the existence of God is incapable of proof, cease to believe in it on the ground that he cannot know it to be true, he is an agnostic and also an atheist, an agnostic- atheist — an atheist because an agnostic. There are unquestionably many such atheists. Agnosticism is among the commonest of apologies for atheism. While, then, it is erroneous to identify agnosticism ' The Creed of a Modern Agnostic. By Richard Bithell, B.Sc, Ph.D., pp. 12, 13. London, 1883. 53 ERRONEOUS VIEWS OF AGNOSTICISM and atheism, it is equally erroneous so to separate them as if the one were exclusive of the other : that they are frequently combined is an unquestionable fact. The author of the passage to which I am referring seems to suppose that a man may believe that there is a God, and at the same time believe that he has no knowledge that there is a God, and that his belief that there is a God is " perhaps at best a mere tradi- tional belief, incapable of proof, and unverified by ex- perience." If there be any such man, I grant that in virtue of his belief that there is a God he is entitled to be called a theist and wronged if he be called an atheist. But I confess I seriously doubt his exist- ence. Belief which is fully conscious of being mere belief, without any true knowledge of its object or any good reason for itself, and without any ca- pability of proof or verification, is, it appears to me, self-contradictory belief, and a psychologically impossible state of mind. Why I think so will be in- dicated at a later stage of this inquiry. IV. NOT TO BE IDENTIFIED WITH POSITIVISM. PEOFESSOE FEASEE A very common misconception as to agnosticism is that it is identical with positivism, phenomenalism, empiricism — with that system according to which knowledge is limited to what is sensibly apprehended or immediately felt, to appearances, to perceptions or modes of consciousness. This view has commended itself to many philosophical thinkers. Thus, to give 54 PR0FE8S0K FRASER'S VIEW only one instance, Professor Eraser has written as follows : " One of the chief intellectual formations, in the interval since Hume, has been what is now called positive or agnostic philosophy. In this pan- phenomenalism, knowledge is limited to physically produced beliefs in coexistences and successions — ex- tended by ' inferences from particulars to particu- lars ' — all at last regarded as an evolution, through habit and association, individual and inherited. With regard to everything beyond, this sort of philosophy is professedly antagonistic. Agnosticism must be distinguished from the universal scepticism that does not admit either of proof or disproof. The latter dissolves the cement of all belief, even beliefs in relations of coexistence or succession among phe- nomena. The former only alleges that outside the coexisting and successive phenomena of sense there is nothing to be cemented — ^that all assertions or denials about supposed realities beyond the range of natural science are illusions." '^ • Berkeley (in Blackwoods' Philosophical Classics), p. 226. The quotation is from the first edition. In " a new edition, amended," the corresponding passage runs thus : " One of the chief intel- lectual formations, in the interval since Hume, has been what is sometimes called Naturalism. In Naturalism, knowledge is sup- posed to be limited to physically produced beliefs — extended by ' inference from particulars to particulars' — all regarded as issue of blind erolution, through habit and association, individual or inherit- ed. With regard to everything beyond, this philosophy is professedly agnostic," &c. To that view my criticism is not meant to apply, and I am happy to find myself in agreement with the esteemed and hon- oured author. I have not deemed it necessary, however, to alter what was not only written but in print before his second edition appeared, as the whole section is as relevant now as then against the very prevalent confusion of agnosticism with positivism. 55 EEEONEOUS VIEWS OF AGNOSTICISM Xow, positivism may, perhaps, be correctly held to imply agnosticism, but it should not be identified with it. In all that it affirms positivism is the contrary of agnosticism, the limitation and exclusion of agnosti- cism. It is the concession that all phenomena are knowable, and so far is non-agnostic. But there have been forms of philosophy directly opposed to positivism — idealistic, ontological, speculative forms of philosophy — which made no such concession to sense and ordinary experience, but held, on the con- trary, that these were the special haunts of uncer- tainty and falsehood, and that truth was only to be found in the regions of pure thought and absolute be- ing. So far as regards sense and phenomena, it is plainly such forms of philosophy which are agnostic, and the varieties of positivism which are non-agnos- tic. "When positivism denies that we can know any- thing beyond what it calls experience and phenomena, the denial seems clearly to require for its vindication a theory of knowledge, and one which, if the denial be legitimate, must be of an agnostic kind. The pos- itivist may or may not, however, have such a theory ; and although he may be inconsistent without it, he may be not more so than with it. Irrationality is before him either way. It is obviously unsatisfactory to define the limits of knowledge without any investigation of the nature of knowledge. The positivism which merely " alleges " that the mind can know nothing except the coexisting and successive phenomena of sense is not entitled to 56 AGNOSTICISM AND POSITIVISM be called agnosticism, because it is not philosophy. It has an unreasoned belief and makes an arbitrary assertion regarding knowledge, but it has no critical or philosophical theory regarding knowledge; and where there is no such theory to speak of agnosticism is out of place. On the other hand, how, consistently with the gen- eral theory of positivism, can a theory of knowledge be attained which will justify agnosticism ? How from actual experience alone can the limits of possi- ble experience be determined ? It would seem as if, in order to attempt, with any reasonable hopes of suc- cess, to ascertain the range of man's capability of knowledge, we must inquire into the nature of his powers of knowledge, and not merely make a survey of what, in our opinion, he actually knows. And yet it is very difficult to see how positivism can afford to acknowledge this; for it means that so far from ex- perience exclusively limiting thought, thought still more limits experience — that knowledge itself is not to be studied merely in the phenomena of knowledge — ^that even to attempt to cast, out the Beelzebub of metaphysics we must begin by invoking his aid. In a word, while the negations of the positivist as to the spiritual and the supernatural must appear un- warranted assertions until based on some agnostic theory of the nature and conditions of cognition, in order to establish such a tlieory the positivist must sacrifice his positivism. Hence many positivists evade the task of inquiring into the limits of human knowledge, and simply assert that nothing is known 57 EKEONEOUS VIEWS OF AGNOSTICISM beyond phenomena, on the ground that experience and history testify that all attempts to know more than phenomena have hitherto been failures, and that it may, consequently, be held that all similar attempts will equally be failures. That this is not self-con- sistent or logical may readily be granted ; but positiv- ism cannot be self -consistent and logical, either when superficial and dogmatic, or when more profound through alliance with agnosticism. The preceding considerations may suffice to show that positivism ought not to be identified with agnos- ticism, although it has an agnostic aspect or involves agnosticism. It has to be added, that there is no need for this new name of agnosticism merely to designate the system called positivism, phenomenalism, empiri- cism, sensationism. These other and older terms are amply sufficient. ISTone of them may be free from defects, but the most faulty of them is a more appro- priate appellation than agnosticism of the doctrine to which they are applied. V. NOT TO BE IDENTIFIED WITH DENIAL OF THE COG- NOSCIBILITT, ACCOMPANIED WITH AFFIRMATION OF THE REALITY, OF THE ABSOLUTE. PROFESSOR CALDERWOOD Another mode of employing the word agnosticism is the restriction of it to.a denial of the cognoseibility of the absolute, when the denial is associated with an admission that the absolute, although imknown and unknowable, certainly exists, and is a legitimate and 58 AGNOSTICISM AND THE ABSOLUTE even necessary object of belief. Agnosticism thus understood is deemed of modern growth, and traced to Kant's theory of knowledge. It is, indeed, virt- ually identified with the doctrine of Hamilton, Man- sel, and Spencer as to the unconditioned.^ Is it desirable to take this limited view of it ? I think not. If it may be thus restricted, why not still further ? Why not define it, for example, as the doc- trine which teaches that the absolute cannot be known, and is to be believed in only as the cause everywhere present, and manifesting itself in all phenomena ? You will thereby be freed from the necessity of treat- ing Christian theists, like Hamilton and Mansel, as agnostics, and will mean by agnosticism a definite individual theory- — that of Spencer as to the im- knowable. It will be said that such definiteness and restriction would be the reverse of merits; that by exclusively applying an essentially general name to the particular theory of knowledge held by Mr. Spencer, the inti- mate afiinity of his theory with that of Hamilton and Mansel would be ignored or concealed; that it is sufficient to say " the agnosticism of Spencer," when- ever this theory is meant, but very inexpedient on any occasion to represent Mr. Spencer and his fol- lowers as the only agnostics. And all that is true, and quite conclusive against identifying agnosticism with Spencerian agnosticism. It applies also, however, against restricting the name ' See the article Agnosticism by Professor Calderwood in Relig- ious Encyclopedia, edited by Dr. SchafE. 59 EEEONEOUS VIEWS OF AGNOSTICISM agnosticism even to the whole movement of specula- tion as to the incognoscibility yet credibility of the absolute with which the names of Hamilton, Mansel, and Spencer are familiarly associated. The entire doctrine which these authors hold in common is but a stage or form of a far older and broader doctrine, a portion of a whole from which it cannot without vio- lence and violation of nature be severed. In the negative and only properly agnostic element of it there is nothing original. The cognoseibility of the absolute has been denied from the very commence- ment of the history of philosophical scepticism; by Protagoras and Pyrrho not less than by Hamilton or Spencer, although in a different manner and for dif- ferent reasons. On the mere ground of that denial, therefore, it is unreasonable to confine the name of agnosticism to a dass of thinkers who have lived after Kant. Is it said that these thinkers, while denying the possibility of knowing the absolute, have yet affirmed the necessity of believing in its existence either as personality or cause, as God or force ? But this af- firmation also is not original or distinctive. It had been maintained by theologians ages before Kant and Hamilton associated their names with it. It was even more generally approved among the philosophical sceptics of the sixteenth and seventeenth centuries than it has been among those of the nineteenth. Besides, it does not seem to be an appropriate rea- son for calling a man an agnostic that he holds him- self bound to believe more than he can know. For ob- 60 NOT MERELY A MODEEN THEORY viously it implies that if a man hold that the absolute cannot he known and ought not to be believed in he is not an agnostic, but if he hold that the absolute cannot be Tcnown yet ought to be believed in he is an agnos- tic; in other words, it makes the distinctive charac- teristic of agnosticism to lie not in its restriction of the sphere of knowledge, but in its extension of the sphere of belief. But to do so is to sever the connec- tion between the term agnosticism and its etymology, and to that extent an abuse of language. All who hold the same theory of knowledge should obviously, when viewed in respect to it, be called by the same name; and all who claim extraordinary rights or powers of belief should, when that claim is in ques- tion, be designated by some name indicative of its nature. Agnosticism is an appropriate name for a certain theory of knowledge, but one altogether in- appropriate for any theory of belief. Hamilton, Mansel, Spencer, and other supporters of that theorv of nescience which found in Professor Calderwood one of its most acute and careful critics, may justly be called agnostics on account of their denial of the cognoscibility of the Absolute or God, just as those who deny the cognoscibility of the Kela- tive, whether World or Self, may be fairly so desig- nated. But, it seems to me, one cannot consistently limit the name of agnostic to those who deny the cog- noscibility of the Absolute, and still less to those who, while denying its cognoscibility, affirm their faith in its reality. A philosophically maintained belief in the incognoscibility of the Absolute is not coextensive 61 EEKONBOUS VIEWS OF AGNOSTICISM with the theory or doctrine of nescience, but only with a theory or doctrine of a certain kind of nescience. The term agnostotheism might, perhaps, be an appro- priate term for the theory which denies the cognosci- bility of God, and agnostotheists for its upholders. My Greek does not suggest to me a suitable designa- tion for the theory which at once denies knowledge of and affirms faith in God. Possibly even the Greek language may not be found readily to supply a con- venient term for such a tenet as that what is entirely unknowable ought nevertheless to be believed. VI. EOBEETy's views on the nature OB AG- NOSTICISM STATED AND CRITICISED An able French publicist — M. E. de Eoberty — has during recent years treated of agnostisicm in a very ingenious and independent manner in several works the titles of which are given below. '^ His peculiar view of its character is naturally and logically de- pendent on the thoroughness of his empiricism. For M. de Koberty, as for Professor Calderwood, agnosticism is " the doctrine of the unknowable " ; but he holds that the doctrine of the unknowable is in- clusive of " the whole of religion and the whole of - metaphysics," and not merely of such phases of belief regarding the unknowable as the so-called critical ag- ' L'Ancienne et la Nbuvelle Philosophic, 1887. L' InconnaissaUe, Sa Metaphysique, Sa Psychologic, 1889. La Philosophic du Siecle- Cf'itidsmc-Positivisme-Evolutionisme, 1891. Agnostidsmc. Essai sur quelqucs theories pessimistcs de la connaissance, 1892, La Recherche de I'v/nite^ 1893. Auguste Comte et Herbert Spencer., 1894. 6^ DE KOBERT^ OJS'^ AGNOSTICISM nosticism of Kant, positivist agnosticism of Comte, conditional agnosticism of liamilton and Mansel, or evolutionist agnosticism of Spencer. The latter, according to M. de Eoberty, far from really being what their adherents, the advanced spir- its of our epoch, suppose them to be — the last and highest results of a long legitimate evolution, or re- cent and valuable acquisitions of philosophy, or direct negations of all religion and metaphysics — are only nineteenth century phases of a process of illusion which goes back to, and is essentially one with, prim- itive fetichism. Agnosticism, he holds, is a very complex illusion, which has its roots in a great num- ber of similar illusions, some of which are of a psychi- cal and others of a social nature ; and when this com- plex phenomenon is analysed there is found to be a perfect identity between the central conception of the most primitive religions, or of the most outgrown metaphysical systems, and the notion of the unknow- able. Religious faiths, metaphysical doctrines, and agnostic beliefs are, in his view, perfectly homologous groups of sociological phenomena, fulfilling essential- ly the same functions and following the same laws of metamorphosis. The unknowable plurality of inae- cessibles accepted by Comte is akin to polytheism; the Unknowable in the singular revered by Mr. Spen- cer is akin to theological monism ; the faith of Ham- ilton in an unknowable Unconditioned is a revival of the belief in the supernatural characteristic of the primitive state of humanity — a case of intellectual atavism. Supernatural and unknowable are only dif- 63 EEEONEOUS VIEWS OF AGNOSTICISM f erent names for the same thing ; and in all religions there is the same supernaturalism, in all philosophies the same agnosticism. Eeligion is an unconscious ag- nosticism; metaphysics a semi-unconscious agnosti- cism, varying according to the epoch of time and the type of system — materialistic, idealistic, or sensual- istic — to which it belongs; and the avowed agnosti- cism of to-day is a stage of the same process. So far from being the formal negation of theology and its eldest daughter metaphysics, it is simply their mod- ern form, their direct descendant and legitimate heir- ess. M. de Koberty foresees that the view which he gives of agnosticism will be objected to on the ground, that whereas what he calls ancient agnosticism (re- ligion and metaphysics) pursued the search of the unknowable with faith and hope, modern agnosticism deliberately renounces such search and expressly ac- knowledges that the absolute cannot be known, — that all quest of first and final causes must fail. He deals, however, with the objection, and concludes that it is worthless. Those who pursue objects which turn out to be absurdities, and those who renounce pursuit of them because they always so turn out, yet continue to theorise on them as unattainable, as unknowable, are, he holds, in the same self -contradictory position and labouring under an essentially identical delusion. Between the unknowable which one seeks to render knowable by extra-scientific processes, and the un- knowable which one cannot know by the methods of science and consequently abandons to methods of spec- 64 BELIEF IN THE UNKNOWABLE Illation which science forbids, there is only the slight- est and most shadowy of distinctions. All forms of belief in the unknowable, — although so many meta- physicians regard them as irreducible, just as zoolo- gists so regarded animal species in the days of Cuvier, — are of the same nature and stages of the same evolutionary process. Eoberty denies the legitimacy of belief in the un- knowable in all its forms, but combats it chiefly in such as are characteristic of the present age. Mod- ern agnosticism he recognises, indeed, to be incapable of acting on humanity either for good or evil with anything like the power of the older agnostic systems. It seems to him to be even in the forms which have been given to it by Kant, Comte, and Spencer, un- worthy of attention for any intrinsic merits. But he deems it to be of prime interest notwithstanding its inherent weakness and poverty, inasmuch as it is " the last citadel of metaphysics," — " almost the only phantom of the theological past of humanity which has not been exorcised by science," — " the only sur- viving chief of what M. Taine calls ' the army of ver- bal entities which had formerly invaded all provinces of nature, and which during three hundred years the progress of science had been overthrowing one by one.' " When it is universally recognised to be a pseudo-concept, a merely " verbal entity," and think- ers cease to occupy their minds with it, then, he holds, all science falsely so called will have at length come to be disowned, and all theological and metaphysical rubbish swept away. Positive science will receive 65 EERONEOUS VIEWS OF AGNOSTICISM the honour due to it, and a scientific philosophy capa- ble of serving as an adequate basis to aesthetic and in- dustrial art will begin to be constituted. The state of positive science reached in any age has always been, according to Roberty, the determining cause of the character of the philosophy of the age. Religion was evolved from least knowledge, meta- physics through a further growth of knowledge, and contemporary agnosticism testifies by its very vague- ness and emptiness to the pressure and predominance of science. But religion, metaphysics, and contem- porary agnosticism are not stages of theorising which lead up to or pass into science. There is no natural or logical transition from the unknowable to the known. There has always been knowledge, and knowledge has always been positive, or in other words, of the nature of science, for there is no other knowl- edge. Contemporary agnosticism is no more occu- pied with an object of knowledge, and has no more a scientific character, than fetichism. The unknowa- ble is altogether an illusion, and when examined al- ways vanishes in the unknown. We know absolutely nothing of the limits which separate the certain do- main of the unknown from the problematical domain of the unknowable. Such is the general view of agnosticism presented in the able and suggestive works of M. de Roberty. It is just the view which we should naturally expect to be given by one who surveys the realm of knowledge from the particular intellectual standpoint which he occupies. Being not only an independent and cou- 66 HIS USE OF THE TEEM rageous but an exceptionally consistent and logical thinker, he is generally able in criticising the agnos- tic doctrines of the present day to show that those who propound them are not as faithful to their own prin- ciples as he himself is, but have involved themselves in contradictions which they should have, and which he has, avoided. His attacks on these doctrines are made from the same position on which their defenders stand, but which he easily proves that they have no right to occupy unless they surrender them; and so clearly has he, on the whole, consistency and reason on his side, that professed agnostics are much more likely to say nothing regarding his assaults than to attempt to repel them. Probably no one else has given so extensive a sig- nification to the word agnosticism as he has done, but the way in which he employs it cannot be denied to be in entire accordance with his philosophical stand- point and principles. These being what they are, he is clearly entitled to regard all religionists and meta- physicians as well as all professors of the creed of nescience as agnostics. He has as much right to use the word in the very wide sense which he attaches to it as I have to use it in a much more restricted one. We both employ it in the same way, namely, with reference to what is deemed unknowable. While differing widely we differ only as to the limits within which knowledge lies. He denies and I affirm that men can attain, and have attained, to a knowledge of theological and metaphysical truths. To me there seems to be hardly any fact of which we may be, and 67 ERRONEOUS VIEWS OF AGNOSTICISM ought to be, so certain as of the existence and govern- ment of a Supreme Being, omnipotent and omnipres- ent, omniscient and righteous. For thinking so M. de Roberty must include me among agnostics, seeing, that he supposes that belief in God is never knowledge b\it always illusion. I, on the other hand, just be- cause he thinks so — just because he deems to be neces- sarily illusion what I hold to be adequately evidenced truth — am compelled to consider him to be the real agnostic; one who would extrude from the realm of knowledge a province which rightfully belongs to it. This shows how relative and personal our views of agnosticism and applications of the term agnostic are, but it does not imply injustice on either side, or tend to obliterate differences, or to conceal or confuse any issues involved. M. de Roberty has given expression to many orig- inal ideas, and formulated many interesting generali- sations. He has traced with searching vision the main currents of human thought, and set in a fresh light the interrelations of the chief systems of specu- lation. He is especially instructive when he treats of the philosophical strivings of the present time, and has, perhaps, successfully shown that very much of what has been written about the unknowable by con- temporary agnostics is as nonsensical as anything of an analogous kind which can be laid to the charge of medieval scholastics. What alone concerns us here, however, is the question, Has he made out that all metaphysics and all theology are of an agnostic char- acter ? And that question 1 can only answer in the negative. 68 DE ROBEETY AND METAPHYSICS As to metaphysics, I can nowhere find that he clearly tells us what he means by it. That it was incumbent on him to do seeing that there are various and conflicting conceptions as to its subject and lim- its. It can surely not be held that in no sense which can reasonably be given to the word will it designate a section or province of real knowledge. And even should M. de Eoberty be of that opinion he has not shown its correctness. To most of his readers his own works will assuredly be thought to consist largely of metaphysical reflections. The positions from which he reasons and the results at which he arrives are rarely the data or the generalisations of physical science; they are, in the plain etymological sense of the term, metaphysical views, although they may have a reference to physical facts. How any sort of theorising as to the attainability of knowledge or as to the merits or demerits of knowledge can be other than metaphysical in character is so difficult to under- stand that it should not be left unexplained by one who believes in its possibility — which I do not. Wherever thought is, even although it be thought about objects of sense, there is something, and even much, which is real, and yet not physical but meta- physical. As to theology also M. de Roberty seems to me to have wholly failed to justify his inclusion of it in ag- nosticism, although his consistency in regarding arid describing it as agnosticism be unquestionable. Anti- theological agnostics would almost seem to have en- tered into a conspiracy not to adopt the only method 69 EERONEOUS VIEWS OF AGNOSTICISM of establishing the truth of their own doctrine which can possibly be satisfactory. Instead of applying themselves to show that the alleged rational bases of theology are unsound, they, with comparatively few exceptions, prefer to adduce reasons for declining the task so obviously incumbent on them. They too often deem it sufficient to assume that it is one which is unnecessary in the present enlightened age, or to assert that there can be no knowledge except within empirical laws. M. de Eoberty attempts to do more, but to little purpose. He lays down as a psychologi- cal discovery of his OAvn what he calls " the law of the identity of super-abstract contraries." What he means thereby is that such lofty abstract correlatives as God and the universe, noumenon and phenomenon, infinite and finite, absolute and relative, although ap- parently opposed, are really equivalent and synony- mous. And from this law he concludes that the word God signifies only the universe or an abstract idea of it, the infinite only the pure or abstract finite, &c. Of course, were it so theology could only be a science falsely so called, one exclusively occupied with illu- sions generated by the inherent weakness and falla- ciousness of human thought. But is it so ? Kather, is not the alleged law a mere paradox ? Our author, at least, has not yet shown it to be anything else. Certainly he has in no way proved it, and it may well be doubted if he has per- suaded a single individual to believe in the truth of it. So long as he has not proved it, or shown theo- logians that what they consider to be evidences of 70 ME. LESLIE STEPHEN'S VIEW God's agency in the physical universe, in historical development, and' spiritual experience have been mis- interpreted by them, he cannot be held to have made out that theology is a species of agnosticism. VII. CRITICISM OF LESLIE STEPHEN^S VIEWS OF AGNOSTICISM. There is yet another view of agnosticism which it appears to me ought to be rejected. It proceeds on the assumption that the attitude of the mind to knowl- edge may be fairly described as either gnostic or ag- nostic; that every individual thinker who is not an agnostic must be a gnostic. This view Mr. Leslie Stephen has adopted. Hence his " Agnostic's Apol- ogy " begins thus : — " The name Agnostic, originally coined by Professor Huxley about 1869, has gained general acceptance. It is sometimes used to indicate the philosophical theory which Mr. Herbert Spencer, as he tells us, developed from the doc- trine of Hamilton and Mansel. Upon that theory I ex- press no opinion. I take the word in a vaguer sense, and am glad to believe that its use indicates an advance in the courtesies of controversy. The old theological phrase for an intellectual opponent was Atheist — a name which still retains a certain flavour as of the stake in this world and hell-flre in the next, and which, moreover, implies an in- accuracy of some importance. Dogmatic Atheism — the doctrine that there is no God, whatever may be meant by God — is, to say the least, a rare phase of opinion. The word Agnosticism, on the other hand, seems to imply a fairly accurate representation of a form of creed already common and daily spreading. The Agnostic is one who as- serts — what no one denies — that there are limits to the sphere of human intelligence. He asserts, further, what 71 EKRONEOUS VIEWS OF AGNOSTICISM many theologians have expressly maintained, that those limits are such as to exclude at least what Lewes calls ' metempirical ' knowledge. But he goes further and as- serts, in opposition to theologians, that theology lies with- in this forbidden sphere. This last assertion raises the im- portant issue ; and, though I have no pretension to invent an opposition nickname, I may venture, for the purposes of this article, to describe the rival school as Gnostics. The Gnostic holds that our reason can, in some sense, transcend the narrow limits of experience. He holds that we can attain truths not capable of verification, and not needing verification, by actual experiment or verification. He holds, further, that a knowledge of those truths is es- sential to the highest interests of mankind, and enables us in some sort to solve the dark riddle of the universe. ' ' ' With much that is said there I cannot agree. The substitution of the name agnostic for atheist may in- dicate no advance in the courtesies of controversy. The application of the term gnostic to all who are not atheistic may be deemed to indicate the opposite. There may have been a lack of courtesy shown by the early Christian writers who turned into a nickname the name of gnostic which some of their adversaries applied to themselves as a title of honour ; but surely to call all who are not atheists gnostics, a name which has never been so used before, and which has been in bad repute among Christians almost since the days of St. Paul, is still less courteous. The term atheist must be admitted to have been often applied in a reckless and unjust way, but to say that it was " the old theo- logical phrase for an intellectual opponent " is itself not an accurate or fair statement. The " flavour " ' An Agnostic's Apology and Other Essays, pp. 1, 2. 73 MR STEPHEN'S VIEW EXAMINED of which Mr. Stephen speaks is not inherent in the word, and is felt only by the vulgar, to whom the term agnostic, when employed as equivalent to atheist, will have just the same flavour. The word atheist is a thoroughly honest, unambig- uous term. It means one who does not believe in God, and it means neither more nor less. It implies neither blame nor approval, neither desert of punish- ment nor of reward. If a purely dogmatic atheism be a rare phase of opinion critical atheism is a very common one, and there is also a form of atheism not uncommon which is professedly sceptical or agnostic, but often in reality dogmatic or gnostic. So far from the word agnosticism, on the other hand, implying, as Mr. Stephen says, a fairly accu- rate representation of a creed which asserts that theol- ogy falls without the sphere of knowledge, it has no special reference whatever to theology. It denotes merely a theory of knowledge, and so may apply to any or every sphere of conceivable existence; but it no more implies theology to be beyond the limits of human intelligence than physiology or psychology. An agnostic may be either a theist or an atheist. There are theological as well as anti-theological ag- nostics ; and to call the former gnostics is 3,3 manifest- ly an abuse of language as it would be to call the latter so. Mr. Stephen's attempted delineation of a gnostic is not a recognisable likeness. " The gnostic holds that our reason can, in some measure, transcend the narrow limits of experience." And so do many ag- 73 BREONEOUS VIEWS OF AGNOSTICISM nostics, including, I imagine, Mr. Stephen himself. Experience ! What sort of experience ? Does Mr. Stephen hold that human reason cannot transcend the narrow limits of sense-experience ? If so, it is clear that he ought to sacrifice to his agnosticism mathe- matical, mental, and moral science, as they all tran- scend the narrow limits of such experience. If not, he is not entitled to assume that religion and theology themselves may not lie within the limits of a real al- though non-sensuous experience. No cautious think- er will affirm that reason can transcend the limits of all experience, seeing that the only known limits of universal experience are the laws of reason itself. Ex- perience extends just so far as reason can go without violating its own laws, and so ceasing to be reasona- ble. A man who simply asserts that reason cannot transcend this or that species of experience is a pure dogmatist ; he may call himself in good faith an ag- nostic, but is really a gnostic, so befogged as not to know what or where he is. "He" (the gnostic), Mr. Stephen further says, " holds that we can attain truths not capable of veri- fication, and not needing verification, by actual exper- iment or verification." This trait also is not dis- tinctive of theologians, Mr. Stephen's so-called gnos- tics. Speaking generally, they neither hold religious truth to be incapable of verification nor to be without need of it. They liold, on the contrary, that religious truth can and ought to be verified. They have always done so more or less ; and at the present day their best representatives are characterised by the earnestness 74 HIS INJUSTICE TO THEOLOGIANS with which they insist on the importance of verifica- tion in religion. But, of course, they maintain at the same time that the verification must be of an ap- propriate kind — one which has a real and intelligible relation to the nature of religious truth and of relig- ious experience. If the verification demanded be that of physical sensible experience, then the de- ductions of the mathematician and the inductions of the historian are unverifiable, and all that claims to be mental or moral truth must be rejected by science. Colours are not to be discriminated by the same organ and processes as sounds ; physics and chemistry apply different standards and tests ; and religion is in like manner to be judged by criteria which can be reason- ably applied to it. To ask that spiritual truth should be verified by a sensible experimental proof is to ask what is self-contradictory — namely, that such truth should be both what it is and is not, both spirit- ual and physical. As spiritual it can only be verified by spiritual beings through spiritual perceptions and experiences. That it cannot be verified at all is a mere dogmatic assertion. No proof or verification has ever been given of that assertion. Mr. Stephen adds: "he" (the gnostic) "holds, further, that a knowledge of those truths " (" metem- pirieal " truths) " is essential to the highest interests of mankind, and enables us in some sort to solve the dark riddle of the universe." The addition is not an improvement. The sole essential difference of opin- ion between the agnostic and his opponent is as to the attainability or unattainability of truth beyond cer- 75 EEBOiJEOUS VIEWS OF AGNOSTICISM tain limits ; the value of truth is not in question. The agnostic does not deny that a reliable knowledge of God, were it attained, would be of high value, and could not fail to dispel much darkness. Keal knowl- edge of every kind is useful and enlightening. Belief in the value of truth is common to agnostic and gnos- tic, and hence should have had no place assigned to it in Mr. Stephen's definition of a gnostic. The intro- duction of it serves no legitimate end, although it may give some slight colour of relevancy to various assertions and reasonings which are really irrelevant in " An Agnostic's Apology." What I wish, however, chiefly to emphasise in con- nection with the view under consideration is that the terms agnosticism and gnosticism can only be rea- sonably understood by the generality of thinkers as of the same character as, for example, empiricism and rationalism, individualism and socialism, scepticism and dogmatism. That is to say, they belong to the class of words which denote extreme and contrary ten- dencies, widely divergent and opposed schemes of thought, the narrowness and exclusiveness of which wise men endeavour to avoid. It is between such an- tithetic extremes as such words denote that the gen- eral course of belief, and the main movements of thought, and far the larger portion of knowledge ac- quired by speculation and research, are to be found. The philosophical world is happily not divided into empiricists and rationalists, — those who would evolve all knowledge out of sensation and those who would resolve it all into reason. Hardly any are purely em- 76 MISUSE OF TERMS piricists or exclusively rationalists. Many, indeed, ascribe so much to sense and so little to reason, and many others so much to reason and so little to sense, that they can without injustice be characterised as empiricists and rationalists respectively, if it be suf- ficiently recognised that they alike allow to some ex- tent both sense and reason to be constituents of knowledge. The great majority of philosophers, however, attach so much weight to both the empirical and the rationalist elements of knowledge that to de- scribe them as either empiricists or rationalists is manifestly unfair. The social world, in like manner, cannot be rea- sonably divided, as so many socialists would have us do, into socialists and individualists, — themselves and others, — the sheep and goats of humanity. Those who call themselves, or can justly be called, individ- ualists are few; and of those who call themselves so- cialists a considerable number appeal more to indi- vidual selfishness than those whom they denounce as individualists, and an even greater number designate themselves socialists largely from aversion to being designated by others individualists. The pretenders to the name of socialists outnumber those who are entitled to it, and of those who are entitled to it com- paratively few are students of social or any other science. The real students of the social sciences, for the most part, regard both individualism and social- ism as irrational and dangerous aberrations. At the present day many profess to be agnostics, but no one will allow that he is a gnostic. The latter 77 EERONEOUS VIEWS OF AGNOSTICISM designation is old, and it early ceased to be regarded as complimentary. The former being of recent in- vention, is as yet comparatively unsuggestive of obnoxious associations such as cling to the terms gnos- tic, sceptic, and dogmatist. It is to this circumstance, and especially to the discredit into which the term sceptic has fallen, not to its own merits, that it owes most of what popularity it possesses. Inevitably, however, disagreeable associations will in course of time attach themselves also to it. The inherent de- fects of agnosticism are sufficient of themselves to ensure this. The eagerness of atheists to exchange their own name for that of agnostics must hasten the degradation of the latter term. A reckless applica- tion of the term gnostics to theists can only tend to the same end. It is desirable that the term agnostic should be as long as possible kept as pure as possible. Those who feel so will not, 1 think, approve of Mr. Leslie Stephen's use of it. The antithesis of scepticism and dogmatism coin- cides to a great extent with that of agnosticism and gnosticism. The former refers more directly to the subjective and the latter to the objective side' or as- pect of the same contrast ; the one more to the dispo- sition and attitude of the mind towards knowledge, and the other more to the range and limits of knowl- edge in relation to the mind. Whenever they are usefully employed both sets of terms imply the same antithesis and denote the same extremes. The words scepticism and dogmatism in themselves imply nothing excessive, defective, or blamable. Ee- 78 SCEPTICISM AND DOGMATISM garded simply from an etymological point of view, scepticism may quite reasonably be defined as the search for truth and dogmatism as the holding of truth. Unfortunately when so defined they are use- less. They indicate no contrast ; seeking truth is not the antithesis but the condition of finding it. And, further, the history of the words has made it impos- sible for us so to employ them. Although 807- fiaTi^€iv and Sojij,aTiK6<; did not originally suggest intellectual rashness, opinionativeness, over-confi- dence, and therefore did not signify what we now mean by dogmatising and dogmatical, nor did o-zcei^t? and ff/cfiTTTtKo? imply excessive doubt of the existence or attainability of truth or aversion to recognise evi- dence, and therefore did not mean scepticism or scep- tical in their current sense, they naturally and inevi- tably acquired those unfavourable implications, and had their significations determined accordingly. The majority of the Greek philosophers of post- Socratic times were characterised by all that is im- plied in the worst sense of the word dogmatism. They were divided into contentious, self-assertive, proselytising schools, each so very sure of possessing the whole truth, and so unwilling to allow that others might have a share of it, that many persons felt doubtful if there were any such thing as truth, and at least if truth were discoverable. Hence the rise of a school of reasoners against reason, ready to dispute everything, and professing either to be certain only that nothing was certain or that not even that was cer- tain. Hence, also, the words dogmatism and dog- 79 EERONEOUS VIEWS OP AGNOSTICISM niatic, scepticism and sceptical, came to denote two opposed extremes of philosophical temper, tendency, and opinion. They are of service to denote the ex- tremes ; but it is unwarrantable to represent every philosophical system as a form either of dogmatism or of scepticism and all philosophers as either dog- matists or sceptics. Could that be done with justice all philosophy would be abnormal and extravagant. Every cautious, circumspect, essentially sane and catholic philosophy is neither dogmatic nor sceptical. 80 CI-IAPTEK III HISTORY OF AGNOSTICISM I. INTEODTJCTOEY. ORIENTAL AGNOSTICISM Agnosticism isjiot merely a Jiind pXtheorising, but also a Eistorical fact, and one of considerable m agniz, ■ tude, imBortaJicP; and interest. I t is older than l Ch ristianity or than any European nation , and has followed a course just as real^nd traceable as that of I a religion or a kingdom. It has passed jthrough a , variety of stages, assumed many forms, been at sun- dry times prevalent, and shows^at^ the present day_no_ signs ol exhaustion.__ It has commended itself to men of very different types of character, and its leading representatives have been distinguished in philoso- phy, theology, science, literature, and even in politics and other spheres of practical life. It is clearly not a mere creation of human wilfulness or exemplifica- tion ot human lolly, but a something deep root edln the nature of the human spirit, a nd hen ce al so a social ^ _^orcej_a power c apable of moulding though t, influenc- 1 ^5S_J£ii2S*~3j[5£ti£g^S_S^^^™^ course of man's de-1 velopment, and serving providential ends. Hence it is only by the uhrS^ectihglEatlFwill be contemptu- ously, impatiently, or wrathfully treated; from oth- ers who feel called to deal with it, even on the whole 81 HISTOEY OP AGNOSTICISM unfavourably, it will receive careful and respectful consideration. Throughout the present work it will be constantly necessary to refer to historical forms of agnosticism — to the views and tenets of individual agnostics or of particular agnostic schools. To discuss the sub- ject of it in a merely indefinite and general way would be useless and unjust. Agnosticism is SQ _EagBe-~arBd- variable that to attempt to reason on it in itself, apart from its actual manifestations, must be futile. It *V ^^^Jjow seems requisite, therefore, to cast a rapid glance h^tx-ottV over the History of agnosticism, to note its chief stages, and to name or characterise the more famous of its representatives. Even a mere outline of the kind will be better than none. It must help the reader to form a fairly adequate idea of what agnosti- cism is as here understood ; let him know who its chief advocates have been, so that he may make himself acquainted with their pleadings if so inclined; and indicate to him what was the place occupied in agnos- tic history by those agnostic theorists whose views he finds subjected in our pages to special criticism. While agnosticism is old, it is far from as old as thought, or even as old as either religion or philoso- phy. Man is naturally less critical or sceptical than imaginative and credulous. He readily satisfies his curiosity with conjectures, and is apt to believe what- ever he is told. The lower religions manifest the extraordinary credulity of those who accept them. It is only at a comparatively advanced stage that relig- ious beliefs are seriously tested with reference to 82 EAELIEST PHILOSOPHIES their truth or falsehood. Before there arises an earn- est demand for rationality and evidence there must be the felt want of them which springs from doubt: hence the spiritual necessity, the religious impor- tance, of doubt in beings so constituted and circum- stanced as men are. The oldest historical forms of philosophy similarly exhibit the most evident marks of having originated in a reason too easily satisfied and ovorweeningly con- fident in its own strength. " Had men," says Comte, " not begun by an exaggerated estimate of what they could do, they would never have done all they were ca- pable of. It has to be added that their pride was chief- ly due to their inevitable ignorance — their excessive confidence to their defective experience. If they had been critical or sceptical — if they had clearly seen how difficult were the problems with which they pro- posed to deal and how inadequate for their solution were the means at their disposal — they would certain- ly never have begun to philosophise at all; but this they could not be, could not see, the humility and the knowledge which it implied being only attainable through the experience acquired in the course of con- tinuous philosophising itself. They began in the only way in which they could begin with such knowl- edge and methods as they possessed. The earliest philosophies were those which most boldly undertook to explain mysteries the most pro- found and to grapple with questions the most inacces- sible; and it was through this boldness that they came into conflict with contemporaneous religions. 83 HISTOEY OF AGNOSTICISM They had the courage to assail and the ambition to seek to supply the place of these religions. Hence intellectual struggles which led to doubts of the truth both of religion and of philosophy. This result was, of course, sooner reached where philosophy started, as in Greece, from an independent rational basis, than where, as in India, it grew directly out of religion. The conflicts and contradictions of philosophical systems, all largely at variance with experience, great- ly contributed to the rise of scepticism. Indeed, it was only when philosophical systems of the too vent- uresome and speculative type had discredited one an- other that doubt or disbelief of a properly sceptical or agnostic nature could arise. Doubt and disbelief are only sceptical or agnostic when they attempt to justify themselves by a distinctive kind of philosophic theorising. The history of agnosticism has been divided into three periods — the Oriental, the Classical, and the Modern. The division is a convenient one ; but the first period was only of a rudimentary character. It presents us merely with approximations to agnosti- cism, not with distinct forms of it. Palestine, China, and India are the oriental lands in which the closest approximations of the kind appeared. So far as has yet been shown, the question. What are the limits of human knowledge ? was not specially discussed, or even distinctly raised, by any ancient Egyptian, Chal- dean, or Persian sage, deeply impressed although many of them cannot fail to have been with the little- ness of their own knowledge and the uncertainty of 84 HEBREW SCEPTICISM much which passed among their contemporaries as knowledge. Tlie Hebrews had no philosophy, and consequently no philosophical scepticism, no scepticism in the sense of agnosticism ; but in the post-exilian period of their history scepticism in a more general sense — a scepti- cism of a spiritual and practical, not speculative and theoretical kind, which expressed itself in the most earnest questionings and gravest doubts as to the rela- tion of sin and suffering and the consistency of the facts of life with Divine goodness and justice — was far from unknown among them. Their dim and du- bious views of a future existence caused suffering virt- ue and prosperous wickedness to be peculiarly inex- plicable and harassing facts even to the most pious among them. These facts gave rise to almost all that can be called even in popular language scepticism in the Bible, — such scepticism as found utterance for itself in Psalms Ixxiii., Ixxxviii., and Ixxxix., in sun- dry sentences of Jeremiah, Ezekiel, and Malachi, and in the Books of Job and Ecclesiastes. In the Book of Job all the theories of providence and TPtriKntir.n o-n-,--. rent among t he Hebrews are seriously examined and t heir weaknesses boldly exposed. Ecclesiastes (Kohe- leth) is more pessimistic than sceptical, but its pes- simism springs from a keen sense of the feebleness and fallibility of human reason and of the complex- ity, mysteriousness, and apparent confusion and plan- lessness of nature and history. " The Preacher " perceives in all spheres* of existence, in all apparent good, in all human aims and efforts, self-contradicto- 85 HISTOEY OF AGNOSTICISM riness, deceptiveness, fruitlessness, and, in a word, proofs and illustrations of his text, — " Vanity of vanities, all is vanity." Chinese scepticism as found in Confucianism somewhat resembles that of modern Positivism, be- ing what is negative in a system of ethical naturalism ; in Taoism it is a denial of the possibility of knowing an Absolute Personal Cause; and in Pohism it has the character proper to Buddhism. So far as it pre- sents itself in what the Chinese regard as their classi- cal writings, it cannot be properly described, I think, as agnostic. Even in India agnosticism did not attain to a dis- tinct and separate form of existence, but grew out of the dogmatic idealisms which sprang up in that land and remained always dependent on them. It is in the writings of the Vedanta school of philosophy that it is most conspicuous ; and the Vedanta philosophy is the most developed and influential of the Hindu philosophies. It rests on the idea that there is but one existence, the universal soul; and to defend this assumption it has to maintain that all the objects of the material world and all separate souls are illusions produced by ignorance or false conceptions: in other words, it is a pantheism which issues in acosmism, and makes use of a partial agnosticism to protect and justify itself. All that the great majority of modern agnostics accept as the only region within which knowledge is attainable, Vedantists consider to be en- tirely the territory of ignorance. A follower of Vyasa and Sankara can only view the exactest observations 86 AGNOSTICISM IN^VEDANTISM of modern science as false conceptions, and the discov- eries of -which it is proudest as vain illusions. Buddhism in its original form was more imbued with the agnostic spirit than any other religion has been. It recognised and appreciated only a kind of knowledge which involved the negation and repudia- tion of all other knowledge. It virtually identified true knowledge with what it inculcated as saving faith. Eight beliefs, according to Buddha, were just right views, — those which when truly appropriated through the personal effort and contemplation of the believer naturally led to right words, right feelings, right acts, right dispositions, and all else that is right, and so led to the chief good, — deliverance from all that is temporal and phenomenal, from birth and death, desire and pain, individuality, consciousness, and change. Its pessimistic conception of life was conjoined with the agnostic conviction that insight into the nothingness of existence is the absolute truth, the sum of truth, and that ordinary knowledge and so-called science are a portion of the burden of false- hood and vanity from which deliverance is to be gained by following the " noble path " revealed by Buddha. While identifying faith and knowledge Buddhism assigned to both a singularly contracted sphere; while a severe ethical rationalism it was ag- nostic and pessimistic in its attitude towards all that constitutes and characterises existence and life. This view of Buddhism, it must be observed, is meant to apply only to its original and philosophical form, — one widely different in various respects from modern 87 HISTORY OF AGNOSTICISM German agnostic pessimism, yet almost certainly more like to it than to the modern Buddhistic relig- ions of the East. With the myths and legends, fic- tions and dogmas of the latter, the historian of agnos- ticism has no concern/ II. GKBCO-EOMAIT AGNOSTICISM. PEE-SOCKATIO OB PKELIMINAEY PEEIOD We now pass to the Greek or Greco-Eoman period of agnostic history. The agnosticism of this period Avas almost entirely a product of the Greek intellect, an outgrowth of Greek philosophy, although the prin- ciples and conclusions of it came to be known and to have their influence felt throughout the Roman world. It owed its being and form, its ingenuity, thorough- ' Owen treats of Hebrew, Hindu, and Buddhist scepticism in Evenings with the Skeptics, vol. i. pp. 367-450. In connection with a sketch like the present it would be inappropriate, I think, to give references to the literature regarding so-called Chinese, Hindu, or Buddhist scepticism. As to so-called " Hebrew Scepticism," it may suffice to mention the following English works : (1) Wright, Book of Koheleth (Donellan Lecture, 1883) ; (2) Cheyne, Job and Solo- w,on, 1887; (3) Plumptre, Ecclesiastes (Camb. Bib. Series), 1892; (+) A. B. Davidson, The Book of Job (C.B.S.), 1893; (5) Momerie, Agnosticism (Part II. Ecclesiastes), 4th ed. ; (6) Wcnley, Aspects of Pessimism (Jewish Pessimism, pp. 1-50), 1894; (7) Dillon, The Sceptics of the Old Testament, 1895. The reason why Dr. Dillon pronounces "Job," "Koheleth," and "Agur" to be "sceptics" is that " all three reject the dogma of retribution, the doctrine of eternal lifQ, and belief in the coming of a Messiah, over and above which they at times strip the notion of God of its most essential attributes, reducing it to the shadow of a mere intellectual abstraction " (p. 1 0). The word "reject" is too strong; and even mere reycrftow, however explicit, of the "dogma," "doctrine," and "belief" mentioned would not be scepticism in the special sense of the term. PRB-SOCEATIC PHILOSOPHY ness, and comprehensiveness, to the love of inquiry j and the speculative qualities of the Grecian mind, I although a long course of historical preparation and a variety of occasional causes concurred with these to secure and perfect its solution. Its history may be divided into Pre-Socratic or Preliminary and Post- Socratic or Developed. It is a history which has been the subject of an immense amount of disquisition and research. In Greece, as everywhere else, agnosticism was pre- ceded by dogmatism. The earliest Greek philosophers were cosmologists. They began with external nature ; sought to find out what was the primary substance of the world ; and tried to explain how the world came to attain its present condition and contents. Their aim was not only legitimate but grand, and their ef- forts to attain it proved wonderfully inspiring. But their own systems were necessarily crude and conject- ural, discordant and contradictory. Hence although they were neither agnostic in themselves nor directly tended to agnosticism, they indirectly led to it both by their one-sidedness and by their conflicting findings. The immediate successors of these philosophers were forced to be more critical, and especially compelled to inquire how appearance and reality are to be distin- guished and how they are related. This, in turn, raised the question how knowledge and opin ion differ^ if _Jhey -differ. A most formidable question! It could not be got rid of ; the adherents of all systems felt vitally interested in finding an answer to it ; yet no one did answer it in a way which commanded gen- 89 HISTORY OF AGNOSTICISM eral assent. Hence a sense of the difficulty of deter- mining the true sphere of knowledge increasingly deepened and spread among Greek thinkers. Hence also the later Pre-Socratic systems of Greek philoso- phy mostly tended directly to generate agnosticism. Greek Eleatic philosophy involved agnosticism in the same way that Hindu Vedantic philosophy did so. Its doctrine of unity implied the impossibility of plurality and change, the unreality of space and time and motion, the non-existence of material ob- jects, and the delusiveness of the senses. All these conclusions Parmenides actually deduced from it and expressly inculcated. And one of his disciples, Zeno of Elea, argued so ingeniously against the possibility of plurality and motion, that although many of the ablest logicians from Aristotle to the present day have undertaken to show the fallaciousness of his reason- ings, there is even yet no general agreement as to wherein their fallaciousness lies, and not a few of those who have treated of them have come to the con- clusion that they cannot be answered. Sir William Plamilton, for example, says " that they at least show that the possibility of motion, however certain as a fact, cannot be conceived possible, as it ' involves a contradiction." '^ If they really proved that, they must also, it seems to me, have proved that mo- tion itself is neither possible nor certain. But that Sir William Hamilton could imagine them to have ir- refutably proved so much may help us to realise what a great advance towards scepticism proper Zeno must ' Lectures on Metaphysics, vol. ii. p. 373. 90 THE SOPHISTS AND AGNOSTICISM have made. Tlie Eleatic philosophy, like the Ve- dantist philosophy, clearly shows that such scepti- cism is not exclusively dependent, as Saisset and oth- ers have affirmed, on sensualism; it may spring as directly and necessarily from idealism and ontolo- gism. The Heraclitean philosophy was essentially antag- onistic to the Eleatic, but not less exclusive- or less favourable to scepticism. What it recognised every- where was not being but becoming, not unity but plu- rality, not immobility but ceaseless motion. It de- nied what Eleaticism. affirmed, and affirmed what it denied ; but it denied as much ; its negations and doubts were as fundamental and comprehensive, -^nesidemus, one of the most renowned of the Greek sceptics, is reported to have attached himself to the Heraclitean system in his later years ; and this might well be, as the Heraclitean tenets of a perpetual flux, and of a self-contradictoriness inherent alike in all things and in all thoughts, are thoroughly sceptical. Some students of Greek scepticism consider that the materialistic philosophy propoimded by Democritus exerted an even greater influence in its formation and development than either the Eleatic or the Heracli- tean. It may have been so ; but it is enough here to say merely that its influence on them was undoubtedly very great. When a professed materialist like De- mocritus, who explained all things by the intercon- nection and interaction of physical atoms, also taught that the senses are incapable of apprehending truth, and that nothing is known of reality, the true nature 91 HISTORY OP AGNOSTICISM even of empirical objects, he did more to discredit sensuous knowledge at least than those who endeav- oured to theorise with less reference to it. He has been ranked both among the sophists and among the sceptics of Greece, erroneously indeed but not inex- plicably ; and it is certain that alike the sophist Pro- tagoras and the sceptic jEnesidemus were largely his disciples. The last stage of Greek Pre-Socratic philosophy was that which is known as the age of the sophists. It was an age of great and varied intellectual and practical activity ; an age of high culture, of famous men, and of brilliant achievements in policy, war, and art; but also an age in which the Greeks had grown dissatisfied with tradition and authority in matters of morals and religion, and had likewise lost confidence in the power of reason and of philosophy to replace them and to supply their defects ; in which self-interest, vanity, and ambition were the ruling motives of action, while self-sacrifice and the pure love of virtue were rare ; and in which the clever ad- vocacy of a bad cause was more admired than the most honest truth-search. It was natural that the sophists should appear and flourish in such an age. They ex- emplified instead of opposing its predominant evil tendencies. They ministered to some of its real wants and rendered considerable services to learning and culture. Their want of faith, however, in any abso- lute truth or goodness made them all the readier and abler to supply reasons for or against any opinion whatever. 93 SOPHISTS AND SCEPTICS Were the sophists sceptics proper, genuine agnos- tics ? Certainly not in so far as they were insincere and dishonest in their professions of doubt or unbe- lief. No real sceptic should be iflentified with a sophist in the discreditable sense acquired by the term. The sophist is a man who does not care for truth, and so is readyto argue~either Ior~or against any thesis or cause. The scepticreally doubts or di? believes the possibility of attaining truth, and argues on behalf of such doubt or disbelief. But the Greek sophists were manifestly the precursors of the Greek sceptics. They so combated the conclusions of each school of philosophy by the arguments of another as to produce the impression that all philosophy was a de- ception ; so exaggerated the relativity alike of sense and of thought as to leave no room for a reasonable trust in the certainty of any kind of knowledge. They appropriated and popularised whatever was sceptical in the teaching of the earlier philosophers, and em- ployed all that was favourable to scepticism in their logical methods. Further, some of the Greek soph- ists seem to have been almost, if not altogether, indis- tinguishable from real sceptics. A Protagoras and a Gorgias, for example, appear to have been about as thoroughly agnostic as human nature has allowed almost any human beings to be. There is not suf- ficient evidence to prove them to have been insincere ; and it is difficult to see that the respects in which their teaching differed from that of the Pyrrhonian, Aca- demic, or Empiricist sceptics ought to prevent us from regarding it as truly sceptical. 93 HISTORY OP AGNOSTICISM There are times when philosophy appears to die, and to rise again out of its own ashes. Its epochs of renascence have been generally preceded by a wide diffusion of indifferentism and of scepticism. The mind cannot rest in doubt, and so is impelled by its pressure to seek the more earnestly for certitude. It perceives that the deepest doubts do not disprove the existence of truth, but merely the depth of the well in which it is contained. Hence there appeared amidst the sophists a Socrates to inaugurate a new era of philosophy, in which almost all the great questions with which human thought has since been occupied were to be raised and discvTSsed in a way which has profoundly influenced the spirit and life of mankind. Plato followed up the movement with wonderful gen- ius and effectiveness. Aristotle made the first and perhaps the most remarkable of attempts to elaborate a universal system of science on philosophical prin- ciples. Various schools of philosophy arose, the dis- ciples of which actively and successfully propagated their respective tenets as to God, nature, man, the laws of reason and of morals, the chief good and how to attain it. But there were few subjects on which general agreement of opinion was reached ; the new philosophies proved as discordant and conflicting as those which preceded them had been ; and so scepti- cism reappeared, and at length assumed its proper or strictly agnostic form. 94 PYEEHONISM III. POST-SOCEATIO OR DEVELOPED PERIOD Pyrrho of Elis, a contemporary of Alexander the Grreat and of Aristotle, is generally regarded as the founder of Greek theoretical scepticism. From him; Pyrrhonism became the ordinary Greek, medieval,- and even, until the close of last century, modern des- ignation for such scepticism. He left no written ex- position of his views, but his disciple Timon of Phlius transmitted to the world what little is known of them. The deep impression which Pyrrho made by main- taining them is only explicable by their having been clearly thought out and ingeniously defended. His philosophy centred in the belief that nothing can be known, and that nothing should be either affirmed or denied, regarding the natures of things, not even whether they exist or not. It was one not of the nega- tion of a knowledge of things but simply of doubt ; it' was one, however, of complete doubt, of entire sus- pense of judgment, as to what things are or whether things are or are not. It did not, of course, exclude' assent to phenomena or appearances considered mere- ly as states of consciousness. Among immediate disci- ples of Pyrrho were, in addition to Timon, Eurylo- chus, Philo of Athens, Nausiphanes, and Hecatseus of Abdera; and among immediate disciples of Timon, Dioscurides of Cyprus, jSTicolochus of Ehodes, Uphre- nor of Seleucus, Praylus, and Xanthus. Arcesilaos (b.c. 316-240) introduced into the Pla- tonic school a scepticism closely akin to that of Pyr- rho, and thereby founded the so-called Second or Mid- 95 HISTORY OF AGNOSTICISM die Academy. He not only began, as Socrates and Plato had done, with doubt, but ended with it, which they did not. The starting-point of his scep- ticism seems to have been opposition to the Stoic view of the criterion of truth; but he was led on to deny that there could be any criterion of truth, or any certitude. Like Pyrrho, he doubted the possibility of knowledge, and inculcated as to all that pretended to be knowledge a total suspension of judgment. He said " he knew nothing absolutely, not even that he knew nothing." He felt, however, that his agnosti- cism required mitigation so far as ordinary life was concerned ; that a distinction must be drawn between speculation and practice, and that, whatever be the conclusions of the former, grounds must be found for satisfying the claims of the latter. Hence while holding that we cannot triily know aught about the natures of things, he argued that we are not thereby reduced either to entire or to irrational activity, see- ing that among the apparent grounds for choosing and rejecting actions or courses of action there is enough of difference to enable us to rule our choices and re- fusals and our conduct generally in a wise and pru- dent way, — or, so as to act rightly and be happy. What he thus regarded as the guide of life he called the reasonable {to evXo^ov), which has been generally identified with the probable {to -KiOavov) of Carne- ades. That they were not identical seems to have been satisfactorily proved by Hirzel; but we may, perhaps, still regard Arcesilaos as the originator of the doctrine of probabilism. The reasonableness 96 CARNEADES which he accepted while denying knowledge and cer- tainty necessarily implied that probability was the guide of life. Arcesilaos was succeeded in the direction of his school by Lacydes, Laeydes by Evander, Evander by Hegesinus, and ITegesinus by Carneades. Of the first three we know almost nothing except the names and the names of some of their disciples. But it is far otherwise as regards the fourth. He was not only the most distinguished successor of Arcesilaos, but himself a still more remarkable and celebrated man ; and Cicero and Sextus Empiricus have made us fairly acquainted with his opinions. Carneades possessed talents of a high order, a mind of amazing vigour and versatility. He was a great orator, a consummate dialectician, a singularly in- genious and subtle critic, and almost irresistible in debate. We have no reason to suppose that his gen- ius was unfitted for the work of construction, but it was specially fitted for the word of destruction, and into that he threw himself with all the energy and ardour of his strong and vehement nature. ITeces- sarily the incessant assaults of such a man on the dogmatic systems and tendencies of his time greatly influenced the minds of his contemporaries, and even those of subsequent thinkers. Carneades endeavoured to confirm and develop the doctrine of Arcesilaos as to the criterion of evidence. He assailed the various hypotheses maintained by the dogmatists of his day on that subject, and laboured to prove that neither sense nor reason supplies any sure 97 HISTORY OF AGNOSTICISM sign of truth, any reliable test by which we can cer- tainly distinguish between the semblance and the real- ity of knowledge. He was, it must be further noted, one of the few Greek sceptics known to have occupied themselves specially with investigation into the grounds of relig- ion. He subjected them to a serious criticism, and one not unsuccessful in so far as it dealt with Stoic and Epicurean opinions. He attempted to refute the ar- gument in favour of religion drawn from its univer- sality, and entirely rejected the theory of final causes. He sought to show that the idea of God is a self -con- tradictory one, seeing that God can only be believed in as a moral being, yet cannot be conceived of as such, since morality implies imperfection overcome, and cannot be thought of as either finite or infinite, al- though He must be either the one or the other. The most important portion of his theological argumenta- tion was his adverse criticism of the doctrine of providence. It contained almost all the weightier of the objections which have since been urged against it. Carneades greatly developed the doctrine of proba- bilism. While denying the possibility of attaining to certainty, he maintained that a measure of proba- I bility may be reached sufiieient for the regulation of practical life. The source of such probability, he ar- gued, could not be in the object, for that is unknown ; but must be in the subject, the mind which thinks it knows. According as the mind is more or less viv- idly impressed, or apprehends appearances as accord- 98 HIS DOCTEINE OF PEOBABILISM aiit or discordant, as permanent or evanescent, it will naturally and reasonably place more or less trust in its sensations and perceptions — in other words, will regard them as more or less probable, and will act on them with more or less confidence. Carneades seems to have been the first to endeavour to determine what were the conditions and degrees of probability. He represented the degrees as corresponding to the condi- tions, the lowest degree being that in which only a single condition is fulfilled, and the highest that in which all the conditions are fulfilled. The highest degree of probability is the best attainable criterion of belief and the best attainable rule of action. Car- neades elaborated his doctrine of probability in order to meet objections which were waged against his de- nial of certainty, and so to give plausibility to his scepticism. He rendered by it, however, good ser- vice to philosophy. The subject of probability is a very important one both in logic and in ethics. The scepticism of Carneades lay in his teaching regard- ing certainty, not in his teaching regarding probabil- ity. His successors, Clitomachus, Charmidas, and Philo of Larissa, were much inferior to him, and carried on the war against dogmatism in a languid and ineffec- tive way. With Antiochus of Askelon scepticism even ceased to be dominant in the teaching of the Acad- emy, and became subordinate to eclecticism. Dog- matism in the form of Stoicism acquired ascendency in the Greco-Roman world. But its triumph was not complete. It even gave rise to a revival of Pyrrhon- 99 HISTOEY OF AGNOSTICISM ism, — to an attempt to develop a decided and thor- oughly consistent scepticism. ^nesidemus of Cnossus, a man of very acute and subtle intellect, was the originator of the movement, and so is knovsm as the founder of the New Sceptical School. There is great difference of opinion as to the time at which he lived. The best supported view is, I think, that which would make him a contempo- rary of Cicero. He taught at Alexandria. ^-Enesidemus restated and defended the Pyrrhonic as opposed to the Academic position. He maintained that we are not entitled either to affirm or deny any- thing regarding things ; that we have no right to do the one more than the other ; and that the Academi- cians, when they pronounced things to be incompre- hensible, and the knowledge of them to be unattaina- ble, erred and showed themselves to be not genuine sceptics. A consistent sceptic can affirm nothing as to the truth of which he is always in search. Such was the way in which ^Enesidemus and the neo-sceptics distinguished their philosophical point of view from that of the Academics. And the validity of the distinction has been generally conceded. To me it seems null or deceptive. The philosophical standpoint of ^Enesidemus, iVgrippa, and Sextus Em- piricus was, I hold, not essentially different from that of such Academicians as Arcesilaos and Carneades. The sceptical Academicians affirmed truth to be in- comprehensible, knowledge to be unattainable, and thereby exposed themselves to the charge of incon- sistency and self-contradiction brought against them 100 ^NESIDEMUS AND THE NEO-SCEPTICS both by the dogmatists and the neo-sceptics of their time, — the charge of declaring a universal proposi- tion to be true, and known to be true, the subject of which they nevertheless asserted to be incomprehensi- ble and unknowable. The answer which they gave to it was that they did not know 'even that they knew nothing; that their universal proposition itself was not to be taken dogmatically, not as real and certain, but only as relative and problematic. It was the best answer which they could give, yet one cannot wonder that it failed to give satisfaction to any but them- selves. Whether their general assertion, however, was itself consistent or not, it was indispensable as a justi- fication of their refusal either to affirm or deny the truth of any particular proposition as to the nature of things. And it was as indispensable to the Pyrrhon- ists as to themselves. These would-be thorough scep- tics professed to be always seekers, on the ground that they never found. But why did they suppose that they never found ? How did they think them- selves always entitled to declare that truth, if it existed, had eluded them? Only because, in their opinion, there were no means of finding what was sought, no reliable organs or criteria by which to as- certain truth. But this was just the same assump- tion to which the sceptical Academicians gave expres- sion. The neo-sceptics refrained from giving it ex- pression, but they constantly implied and acted on it. They were less explicit than the Academicians, and therefore in appearance more consistent, but not more so in reality. 101 HISTORY OP AGNOSTICISM ^Enesidemus very considerably improved the theory of scepticism by classifying and describing the va- rious ways in which what claims to be truth and cer- tainty may be discredited and doubt superinduced. He was the first to arrange the arguments on behalf of scepticism under the heads known as the ten tropes {rpoTToi). Although he originated none of them, he collected and grouped them, and thereby showed the strength of the case for scepticism more effectively than had previously been done. His arrangement of them cannot be justly praised as clear or natural, but even such as it was it marked an important ad- vance. The ten tropes corresponded to the grounds on which they were based, and these were the following : (1) the diversities of the organisation and constitu- tion of the various kinds of animals; (2) the diversi- ties of the organisation and constitution of human in- dividuals; (3) the diversity of the senses even in the same individual ; (4) the variableness of our physical and mental conditions and circumstances and their effects on our perceptions and judgments; (5) the influence of distance, place, and position on the ap- pearance of objects; (6) the way in which our views of objects are affected by their connections with oth- ers; (7) the extent to which the characters of things are altered by changes of quantity and composition ; (8) the relativity of all things to one another and to their percipient subjects; (9) the degree to which men's notions of phenomena aro depei:dent on their frequency or rarity; and (10) the divergences of 102 iENESIDBMUS AND CAUSALITY moral and religious belief and practice, of customs, laws, rites, institutions, and opinions, among different peoples. These tropes show that ^nesidemus chal-l lenged the credibility of all our immediate percep-! tions and all our ordinary judgments, as well as of all the philosophical theories which rest on such percep- ' tions and judgments. His criticism of the notion of causation must not be forgotten; indeed, it was the most original and suggestive portion of his argumentation against the validity of human knowledge. By it he remarkably anticipated the views as to causality reached by Hume and Kant, while he yet strikingly differed from both. He denied to the belief in causality all objective legit- imacy, and on at least two distinguishable grounds. First, the belief has no warrant in the notion of cau- sality. The notion of a cause is a relative notion, the notion of a relation, a cause not being conceivable without that which it causes. But no relation can be shown to have any objective legitimacy, any existence except in thought. Thought relationships belong, or may belong, only to thought. Further, the notion of causality, according to ^nesidemus, is so inherently perplexing and inconsistent as to be unworthy of cre- dence. It involves insuperable difficulties. A cause cannot be rationally thought of as either synchronous with, antecedent to, or consequent on its alleged effect. Not as synchronous with it, for then cause and effect would be so indistinguishable that each might as well be either cause or effect as the other ; not as antecedent to it, for nothing can be the cause of anything- until 103 HISTOEY OP AGNOSTICISM its effect exists ; not as consequent on it, for what pro- duces cannot be subsequent to what is produced. The point of view from which ^nesidemus crit- icised belief in causality seems to have been the De- mocritean or Heraclitean — one virtually materialis- tic. What his criticism showed was that the belief could not be justified from that standpoint. Hume, starting from the principles of sensationism taught by Locke, deduced from them scepticism on the strength of one having no perception of the connection of cause and effect in the external world. Kant pro- fessed to refute Hume by arguing that causality is a condition, and a necessary condition, of thought. What ^Enesidemus contended was that as causality could not be shown to be other or more than thought, pure scepticism, Pyrrhonism, was justified. Ob- viously he would have regarded Kant's attempted an- swer to Hume as not a refutation but a confirmation of scepticism. Of his immediate successors in the direction of the school the names merely are known. The only mem- ber of it recorded to have made any considerable change on its doctrine was Agrippa, as to the precise time of whose teaching there is much uncertainty. For the ten tropes of zEnesidemus he substituted five which were wider and deeper in their range, as well as more closely and logically connected, one naturally leading up and lending support to another from the first to the last. His first trope he described as rest- ing on the contrariety of opinions the truth or falsity of which there are no satisfactory means of determin- 104 SEXTUS BMPIRIGUS ing; the second, on the inevitable necessity of prov- ing every proof ad infinitum; the third, on the rela- tivity of all objects of sense and intelligence; the fourth, on the impossibility of carrying on any in- vestigation or demonstration without making assump- tions which themselves need to be established; and ' the fifth, the trope designated d StaWijXo?, on the en- deavour to verify sense by reason and reason by sense — i.e., by a circular or alternative process, which is fallacious, inasmuch as the truthfulness of both sense and reason is challenged by the sceptic. These were, according to Agrippa, the species or means of produc- ing doubt best fitted to show that no one was entitled to deem himself certain of any truth, possessed of any indubitable knowledge. Ancient scepticism had Sextus, surnamed Empiri- cus from adhering to the school or sect of physicians called Empirical, for its last literary representative and expositor. He was a Greek and flourished about A.D. 200. It is not known where he was born or where he taught. It has been conjectured that he lived for some time at Athens, at Alexandria, and at Rome. He wrote " Medical Memoirs," a treatise " On the Soul," and perhaps other works, which have been lost. The two works which have come down to us are the " Pyrrhonic Institutes," consisting of three books, and the treatise " Against the Dogmatists," which comprises eleven books. On them his fame will rest securely so long as the history of philosophy is a subject of human interest. All Greek scepti- cism, all that was most important in the most thorough 105 HISTORY OF AGNOSTICISM and consistent developi.ient of agnosticism which has appeared in the world, seems to have been preserved in them, and would certainly have been in great part lost if they had not survived. They have had a very great influence even on modern thought and philoso- phy. The scepticism which prevailed in Europe from the beginning of the sixteenth to about the close of the eighteenth century drew its inspiration, its principles, and its methods largely from the writings of Sextus. Montaigne and Huet, Bayle and Hume, borrowed as freely from him as he himself had done from Arcesilaos and Cameades, ^nesidemus and Agrippa. Probably he originated no absolutely new agnostic idea or argument, but has transmitted to us only thoughts and reasonings which he derived to some extent from his Greek predecessors; probably also, however, there is scarcely any absolutely new ag- nostic idea or argument in all modern literature, scarcely any even which are not to be found indicated to some extent in the pages of Sextus. Greco-Roman philosophical scepticism began its course in the latter half of the fourth century b.c.j and became extinct about the commencement of the third century a.d. It does not appear to have been at any time widely accepted in the classical world, and certainly never enjoyed the popularity of such dog- matic systems as Epicureanism and Stoicism. Prob- ably it was never much more prevalent than it was desirable it should be as a counteractive to philosophi- cal dogmatism. It dealt, of course, with religion and morality in 106 GREGO-ROMAN SCEPTICISM the same spirit and fashion as with all other things maintained to be objects of knowledge. But it was not specially antagonistic to them. Its adherents showed no predilection for attacking religion or mo- rality ; on the contrary, even when arguing that there was no real knowledge possible of divine things or moral distinctions, they professed to hold the common faith regarding them. Carneades keenly criticised religious beliefs and represented their grounds as non- rational or irrational, but he did not pretend that his own reasoning did more than show the unsatisf actor i- ness of the reasoning to which it was opposed. He ridiculed various aspects and portions of the popular religion, but he did not infer that it was not to be ac- cepted. So Sextus professed his faith in the gods and providence even while he argued against its reasona- bleness. The philosophical sceptics of Greece and Rome had little proselytising zeal. Professedly regarding in- dividual imperturbability as the chief good, they did not aim at either destroying or reforming religion, and still less at revolutionising society, but were con- tent to influence only cultured and ingenious minds. Those who attribute to them the ruin of religious faith in the ancient world take insufficient account of the fact that philosophical scepticism died out of that world and was succeeded by a great dogmatic reaction both in philosophy and religion. The centuries which immediately preceded the definitive triumph of Christianity were characterised not by excessive doubt but " by excessive faith. In those centuries 107 HISTOKY OF AGNOSTICISM philosophical scepticism was extinct. It had worked out its own destruction, its endeavours to prove by reason that truth could not be found by reason hav- ing tended to make men seek it by other means, — faith, feeling, mystic vision, abnormal spiritual proc- esses, tradition, authority, revelation. Hence it so far prepared the way for Neo-Platonism and Chris- tianity, and inevitably disappeared before them.^ IV. MIDDLE AGES Between the disappearance of ancient and the rise of modern agnosticism there intervened a period of about fourteen hundred years, during which agnosti- cism had no distinct existence as a special and pecul- iar kind of philosophic thought. This is not equiva- lent to saying that it was wholly absent. That would not be a correct assertion. Agnostic elements may easily be detected in various medieval systems. An agnostic spirit was even very prevalent during the last two centuries of medieval history. Several learned Christian agnostics of modern times have, in part perhaps to justify their own prac- tice, represented the Christian Fathers generally as advocating the cavise of the Gospel by arguments drawn from the Greek Pyrrhonists and Academi- ' The general accounts of Greek Scepticism in Zeller, Owen, Brochard {Philosophic des Orecs), and Credaro (Lo Scetticismo degli Accademici) are detailed and erudite. Hirzel's Untersuchungen and Natorp's Forschungen are indispensable to those who would en- ter on a thorough study of the subject. Raid's edition of Cicero's Academics' is valuable. It is greatly to be regretted that there is no adequate edition of Sextus Empiricus. 108 MEDIEVAL TIMES cians. The evidence warrants no such view. It shows only that fervidness of temperament and dog- matic narrowness led some of the Fathers so to glory in faith and Scripture as to think and speak at times unworthily of reason and nature. There was a kind of agnosticism which passed into the theology of the early and medieval Church, chieily through the channel of Neo-Platonic philosophy. Wtfin-^ee;Platoiiism taught that God was wholly un- approachable_by_reason ; that He must be reached throu gh faith, or ecstasy, or the self-surrend er_of in- d ividual consciousness ; that He w as so^ essentially one and so entirely indeterminaj^e^to bs^oihoJlLiiis.- tmctions or attributes, without power, knowledge, love, jiistice, or exceirence, in any sense intelligible by man ; it was in one essential respect clearly agnostic, although in other respects conspicuously gnostic. But tero'Mng'of alike kind is" to be found in the writings of St. Augustine and the Pseudo-Dionysius, and of John Scotus Erigena, Bonaventura, Eckhart, Nico- laus of Cusa, and others. The authors named incul- cated in express terms the doctrine of " learned igno- rance " (docta ignorantia), on which Hamilton and Mansel have laid so much stress, and, indeed, made the corner-stone of their agnosticism. They held that all positive knowledge of the Self-existent Being, the Unconditioned, is impossible, and that a thought- ful acquiescence in this fact, a carefully acquired con- viction of inevitable nescience as regards ultimate reality, is the consummation of human science. Hence it may be maintained with a certain measure of truth 109 HISTORY OF AGNOSTICISM that Greek agnosticism was rather absorbed into Christian thought than extinguished by it. Among medieval thinkers the nominalists were the most sceptical and negative, and in the two last cen- turies of the medieval era there lived many nominal- ists. They started with empiricist preconceptions, and were disposed to deny that there. could be any knowledge except of individual objects of sense. This sort of philosophising leads necessarily to some kind of agnosticism ; in the Middle Ages it naturally led to a theological agnosticism, l^ominalists like William of Occam and Peter D'Ailly may not inappropriately be described as theological agnostics, seeing that al- though they accepted the doctrines of the Church as articles of faith imposed by legitimate authority, they relegated theology to the sphere of the unprovable and unknowable. The nominalists generally so sev- ered and opposed faith and reason that they could claim to be rigidly orthodox, while holding the human mind to be incapable of finding valid reasons for be- lief in the existence of God or in any other supersen- suous verity. From this unnaturally divided root the doctrine of a twofold truth, — the doctrine that equally valid dicta of faith and reason may not only be dis- tinct but contradictory, — that what is true in theology may be false in philosophy, and vice versa, — was a very natural outgrowth. And the doctrine found ac- ceptance. It had its strongholds in the Universities of Paris and of Pavia. In the thirteenth century eccle- siastical censures were pronounced against it, but it was, perhaps, more prevalent three centuries later. 110 TRANSITION TO MODERN TIMES There were many ready to say with Pomponazzi — " I believe as a Christian what 1 cannot believe as a philosopher." To explain fully how agnosticism was suppressed to the extent that it was during fourteen hundred years, and yet how the suppression instead of being completely and permanently effective prepared the advent and influenced the development of a new era of a most powerful agnosticism, which still shows no signs of decadence, would require a philosophical sur- vey of medieval history, showing how Christianity came to be accepted ; how the Church became subject to a priestly hierarchy ; how theology was shaped into system, and all science brought under its control; how society was organised by ecclesiastical authority ; and then how a reaction of thought set in; how the general mind of Europe, influenced by various causes and circumstances, ceased to be satisfied with its con- dition, began to regard with a critical and even hos- tile disposition the powers which claimed lordship over it, and learned to cherish aspirations and hopes which had been previously unfelt or stifled or con- cealed. This, of course, cannot be here attempted, and so we pass at once to outline the history of Mod- ern Agnosticism. V. FIEST PEEIOD OF MODERN AGNOSTICISM. CAUSES AND CHAEACTEEISTICS The history of Modern Agnosticism may be divided into two periods: the first extending from about the 111 HISTORY OF AGNOSTICISM beginning of the sixteenth century to about the close of the fourth decade of the eighteenth century; the second comprehending the time from the commence- ment of Hume's philosophical career to the present day. There were a variety of causes favourable to agnos- ticism operative throughout the first of these periods. The dominant and most comprehensive one vsras that just referred to — ^the general change in the European mind from submissiveness to authority and acquies- cence in dogma to an independent and critical dispo- sition of spirit. The struggles of the conflicting forces in the medieval vrorld, the nevi^ experiences which the course of time had brought with it, a mul- titude of notable events, and even the efforts of scho- lasticism itself to extend its own sway and to promote by argument the cause of authority and faith, had all concurred in bringing about that profound and com- prehensive change, and giving rise to the modern world, which, as contrasted with the medieval world, has for its distinctive characteristic the independent exercise of reason. Then special causes, which were, however, closely connected with the general cause, strongly favoured the diffusion of the agnostic spirit in the period indicated. Thus there was such a cause in the state into which philosophy lapsed when scho- lasticism broke up. It was a state of chaos in which all the ancient systems and a multitude of new ones, hastily extemporised, struggled with each other, and sought in vain for general recognition. It would have been strange if scepticism had not been among 112 MODERN AGNOSTICISM them. The strife occasioned by difFerences of opin- ion as to religion was probably an even more powerful cause of scepticism than the struggle of philosophies. [ Its violence and unscrupulousness, and the wicked deeds and horrible wars which it produced, directly tended to discredit both religion and human nature, and to make men disbelieve in truth and morality. The combination of intellectual culture and of moral and religious corruption, widely prevalent in the epoch of transition from medieval to modern times, worked in the same direction. Further, the special sciences and professional studies were in a condition much more fitted to foster and confirm than to re- strain and correct the sceptical spirit. The conject- ural and the false in them largely predominated over the certain and the true. To be led to consider them with a critical mind was to be subjected to the temp- tation to regard all science as vanity and delusion. Scepticism even in the form which may be called agnosticism was, accordingly, prevalent in the six- teenth and seventeenth centuries. But it was, of/ course, considerably different from the agnosticism/| of Hume and Kant, and of our contemporaries. It had its own characteristics, derived from the causes which originated it and the circumstances in which it appeared. One of these characteristics was imperfect develop- ment. It did not rest on any searching or compre- hensive criticism of the powers of the human intellect. It did not attain in the writings of any of its repre- sentatives a properly philosophical character. It was 113 HISTOEY OF AGNOSTICISM mainly the expression of an exaggerated depreciation of knowledge or of a despair of acquiring knowledge, due to the real or imagined detection of the uncer- tainty of what passed for science and of the aberra- tions of what was called reason at the time in which it prevailed. It moralised and preached; satirised, jested, and declaimed; cultivated belles lettres and availed itself of the resources of erudition favourable to its ends; but it shunned the arduous labours of real philosophising, and neglected exact analyses, se- vere argumentation, logical precision, and verbal ac- curacy. It was a superficial, popular philosophy; not a solidly founded or carefully built up speculative system. I Another characteristic of the first phase of modern I agnosticism was absence of essential originality. It I was in the main a revival of Greek agnosticism. Its weapons of warfare were drawn almost entirely from the arsenals of ancient scepticism, and especially from the works of Sextus. The only originality of its champions lay in their mode of handling those weapons. Even in the scepticism of Montaigne there is nothing new but the manner of expression, the fresh literary style. The sort of want of originality indi- cated is no reason for depreciating the authors re- ferred to or undervaluing their services, seeing that it was not only compatible with but favourable to orig- inality as regards the expression of their views. Each of them was remarkably successful in presenting a scepticism essentially common to all with a natural- ness and individuality of form which contributed 114 ANCIENT AND MODERN SCEPTICISM greatly to its attractiveness and diffusion. It was no ordinary service vsrhich they rendered to the world when they resuscitated, revivified, and popularised the agnosticism of antiquity among their contempo- raries, and so transmitted it to future generations. But for their comparatively unoriginal and super- ficial scepticism we should probably have had neither the more original and profound scepticism nor the more original and profound positive speculation of later ages. A third characteristic of the agnosticism of the transition period is that it was predominantly relig- ious in aim, and, at least, more reverent towards re- ligion than towards science. It was generally repre- sented by its advocates as the best defence of religion. Only in the sixteenth century did attempts to support religion by philosophical scepticism begin to be made ; only in the seventeenth century did they become com- mon. The ancient sceptics were more consistent than to make such attempts. Their scepticism was all of a piece, so to speak. They saw that if it could be shown that men have no knowledge of objects, it followed that they can have no knowledge of religious objects; that the general includes the particular. Some of the early Christian Fathers were led by their zeal against pagan philosophy to harsh censure of philosophy itself, and to occasional denials of the authority of reason ; but none of them sought to raise scepticism to the rank of a method of producing be- lievers. Of course, the scholastic divines felt no need of such a method. It was only when reason began to 115 HISTORY OF AGNOSTICISM take up an attitude of opposition to religion, and when it began, at least, within the sphere of religion to criticise independently and unfavourably the dog- mas of the Church, that there was evoked an antago- nistic spirit, — a desire to humiliate and discredit reason in order thereby to exalt and glorify faith. VI. REPRESENTATIVE AGNOSTICS OF THE TRANSITION PERIOD The first representative of modern agnosticism was Henry Cornelius Agrippa of ISTettesheim (1486- 1535). His career was of the most diversified and romantic character. He lived in many lands, ac- quired many languages, studied a,ll kinds of subjects, and passed through the most varied experiences. As early as the twentieth year of his age he was striving to fathom the secrets of theosophy, alchemy, astrol- ogy, and magic, and interesting himself in the foun- dation of Rosicrucian societies. He was a conspirator in Spain, a soldier in Italy, a courtier in Austria, an ambassador in England, a physician in Switzerland, a theologian at Dole in Burgundy, an advocate at Metz, and served in other capacities in other places. He was knighted on the battlefield ; he was a Doctor of Laws, a Doctor of Medicine, and a Doctor of Divin- ity ; and, in popular reputation, a most powerful sor- cerer and magician. An adventurer he unquestion- ably was, but not an unprincipled one. Although impetuous and imprudent, and sometimes driven into false positions, he was essentially honest, chivalrous 116 AGRIPPA OF NETTESHEIM and even, notwithstanding his wars with sword and pen, refined and gentle. While he saw clearly the errors of the Church of Eome and condemned them with a boldness which roused against him the wrath of its clergy and monks, and involved him in much suffering, he refused, like Uean Colet, Sir Thomas More, and many other learned and good men of his time, to take part in the disruption of the visible unity of Christendom. His two chief works present him to us in very dif- ferent aspects. In the treatise On Occult Philosophy (written in 1509, and, after being widely circulated in MS., printed in a revised form at Antwerp in 1531), we see him in his eager, credulous, enthusias- tic early manhood, a theosophic mystic, a confident believer in the existence and cognoscibility of magical and marvellous secret powers pervading nature, a man much too ready to accept as science all that claimed to be science. In the work on account of which he is mentioned here, A Declamation on the Uncertainty and Vanity of the Arts and Sciences (written in 1526 and published in 1530), we see the same man, but that man disillusioned, and who has gone, as men of his temperament not infrequently do, to the contrary extreme. The work makes no pretensions to impartiality : it is avowedly a satire, " a cynical declamation." Yet, in the main, it is quite serious and sincere. Its au- thor knew the sciences and arts of his age as scarcely any one else did; but he had come to the conclusion that men enormously overestimated the worth of 117 HISTORY OF AGNOSTICISM them, and felt, in particular, how much he had been himself deceived in regard to them. Hence he now assails them with as much fervour as he had formerly lauded them. Surveying them one after another in a long succession of chapters, he gives prominence to what is weak and uncertain, useless or hurtful, in all human studies and professions, and argues that it is dangerous to trust them, foolish to be proud of them ; that all is dubious except God's Word, and that its truth is accessible to all men by faith in Jesus Christ and the enlightening grace of the Holy Spirit. The scepticism of Agrippa was not a reasoned-out theory or even a definitive intellectual conviction, but a frame of mind, and indeed largely of feeling, of the exaggeration in which he was himself so far conscious, produced by his having so often found what was called science to be conjecture or absurd- ity and the professions esteemed most honourable to be pervaded with deceit and charlatanry. The one thing to which his scepticism did not extend, the one thing in which he felt there was no illusion or falsity, was God's Word received in its plain and simple meaning as revealed in the Gospel.^ • The late Professor Henry Morley's Life of Henry Cornelius Agrippa von Nettesheim, Doctor and Knight, commonly known as u Magician, 2 vols., 1856, is an admirable biography and study, — the first work in which full justice was done to Agrippa, although Naude, Moreri, and Bayle had to some extent shown that he had been greatly calumniated. M. Auguste Frost's Corneille Agrippa, sa vie et ses auvres, 2 vols., 1881-82, is not nearly so vivid and ar- tistic, but he has pushed research further and added considerably to our knowledge of the events and circumstances of Agrippa's ca- reer. Owen treats of Agrippa's agnosticism very fully, but, in my opinion, considerably exaggerates it. 118 MONTAIGNE Michel de Montaigne (1533-1592) did much more to diffuse scepticism. His Essais (1580), owing to qualities on which hundreds of critics and admirers have descanted but on which no word need here be said, have enjoyed immense popularity and exercised immense influence. Scepticism has never appeared in a more generally attractive form : and all ancient scepticism is there, — transmitted, revivified, and modernised in passing through the mind of Mon- taigne into the book which he so truly tells us " is himself." ISTo writer in the whole history of literature, so far as I know, has portrayed his own character with more candour, fulness, and skill than Montaigne has done in his Essais. That character was obviously one constitutionally favourable to the reception of the agnostic spirit, — one to which " not less than knowl- edge doubt was grateful." Montaigne loved dearly his own ease and comfort ; disliked all constraint ; was keenly alive to the hatefulness of intolerance and persecution; was quick to see reasons both for belief and disbelief in all opinions ; and, although very fond of reasoning for the sake of the pleasure of the exer- cise, was too impatient and unsteady to seek truth in a persistent and methodical manner. The almost exclusively classical and humanist education which he received tended to foster his sceptical proclivities. It made him more in sympathy with the pagan than the Christian spirit, and failed to initiate him into any real acquaintance with science. He knew many things, but few well. The one subject which he care- 119 HISTOEY OF AGNOSTICISM fully studied, his owu self-contradictory and change- ful nature, was not suggestive of aught permanent or stable. Further, the character of the period in which he lived must have contributed to evoke and confirm his scepticism. It was not only the time when the conflict between the ruling ideas of scholasticism and the beliefs distinctive of the renaissance was at its height, but also one of the most deplorable epochs of French history: an age of ethical and spiritual as well as of intellectual disintegration, of lax morality, of religious hypocrisy and religious fanaticism, of political unscrupulousness and of continuous and fe- rocious civil war, in which Romanists and Huguenots, Leaguers and Lutherans, alike sought to cloak the most abominable crimes with professions of piety and of patriotism. It is easy to understand how in such evil days a clear-sighted and peace-loving man like Montaigne should have come to form a low estimate of human nature, and to have the most serious doubts of the attainability of truth. The scepticism of Montaigne was of an indulgent, half-pitying, half-contemptuous kind. ~No man could be more tolerant towards all sorts of opinions and ac- tions : their diversity and strangeness were an un- failing source of interest and amusement to him. The contradictions and absurdities of the learned afforded him his favourite argument for representing so-called science as a failure, and the human mind as singularly unreasonable in its reasonings. For his own part, he did not profess to philosophise or even to be consistent ; did not put forth his opinions " as 120 ITS GEECO-EOMAN SOUEOE true but as his " ; and did not formally inculcate scepticism, but so treated, in his own easy and natu- ral, free and familiar way, whatever themes happened to occur to him, as to make them all suggest the vanity of anxious search for truth, and minister to the spirit of doubt. The manner in which he presented his scepticism was entirely his own and inimitable; the scepticism itself he derived from Greco-Roman sources. There was no distinct originality in his point of view or ab- solute novelty in his arguments. On the other hand, there was very little, if anything, of a sceptical nature and tendency which the ancient sceptics are known to have urged that he did not recall and make use of in his own peculiarly fascinating way. His scepticism must be credited with thoroughness. He preferred the Pyrrhonian attitude towards trutht knowledge, and certitude to the Academic. He saw the inconsistency of at once denying those things, and yet admitting, as the Academicians did, " a cer- tain partiality of judgment," " an appearance of like- lihood," in any direction or instance. " What is such an apparent inclination but a recognition of some more apparent truth in this than in that ? " . . . " Why do they (the Academicians) suffer themselves to incline and be swayed by verisimilitude, if they know not the truth ? " The symbol or emblem of his scepticism was a balance perfectly poised as regards truth and falsehood, knowledge and nescience, and therefore liable to be swayed or turned either way as regards belief, feeling, or action by any non-rational 131 HISTOEY or AGNOSTICISM influence, however strange or slight. His motto was not Jene sais pas, but Que sais-je ? Being radical, his scepticism was also naturally universal. The notion that it extended only to meta- physical things or questions, and so was merely a sort of Positivism, has foimd defenders, but is wholly er- roneous. Montaigne troubled himself very little about metaphysical disputes. His doubts were brought to bear on all the apprehensions of sense and all the applications of reason ; they spared nothing in morals and religion. He acknowledged, indeed, that we cannot help assenting to certain perceptions or ap- pearances of sense; but he none the less on that ac- count held that the senses alter and falsify everything that they bring us. Xay, even from our entire de- pendence on them he inferred our entire ignorance not only of their own objects but of the objects of all our other faculties, these being all derived from sense. What are commonly regarded as virtues and as laws of conscience and principles and verities of religion, he regarded as products of custom and other causes wholly independent of truth and knowledge. Yet although his scepticism undoubtedly extended to morality and religion, he did not seek by diffusing it to spread immorality or irreligion, or even moral or religious indifferentism. He wished to make life better and happier, and recognised a Divine excel- lency in Christianity. He deemed it right to reserve for faith a sphere exempt from the intrusive interro- gations of reason. He acknowledged the need of su- pernatural grace to convey Divine truth to the mind 123 CHAEEON and heart of man, and " the need of a Divine basis and foundation on which those who rest will not be shaken as others are by human accidents, the love of novelty, the constraint of. princes, the fortunes of parties, the rash and fortuitous changes of opinions, the subtleties of argument, or the attractions of rhet- oric." ^ Montaigne had in the later years of his life an inti- mate friend and admiring disciple in Peter Charron (1541-1603), a Roman Catholic theologian and cel- ebrated preacher. This divine published in 1593 a work of religious apologetics, Les Trois Verites, in which he defended Theism against atheists, Chris- tianity against idolaters, Jews, and Mohammedans, and Catholicism against Protestants, and sought to establish these three positions: 1st, There is one God whom alone we ought to worship ; 2nd, Of all re- ligions the Christian is the only credible one ; and 3rd, Of all Christian communions the Roman is the only safe one. The orthodoxy of this work passed un- challenged, although its spirit and the method of rea- soning pursued in it were of a decidedly agnostic character. It was not otherwise with the Discours » The following are among the English writing regarding Mon- taigne most worth consulting : (1) Bayle St. John's Montaigne the Essayist, 2 vols. , 1857 ; (2) Dean Church's Article in the Oxford Essays, 1859 ; (3) Collins's Montaigne (Blackwoods' Foreign Clas- sics), 1889; and (4), Owen's Skeptics of the French Renaissance, ch. i., 1893. Bruneti^re {Manuel de I'histoire de la litterature fran- ^aise, pp. 86, 87) gives a judiciously selected list of French writ- ings. Two German authors, H, Thimme (1875) and A Hemming (1879), have published dissertations with the title Der Skepticismus 123 HISTORY OP AGNOSTICISM Chretiens, published in 1600. But the De la Sagesse, which appeared in the following year and revealed the scepticism of its author in a fully developed form, evoked a great storm of wrathful controversy. And from his own day until now many have supposed that the same man could not have honestly preached the " Christian Discourses/' and believed what he wrote in " The Three Truths," and yet entertained the scep- tical views set forth in the " Treatise on Wisdom." But that merely shows that those who have thought so have not understood the character of his scepticism: a scepticism not less sceptical of itself than of other things; a scepticism founded on distrust of reason, yet anxious to draw from the admission of the weak- ness and worthlessness of reason some advantage to the cause of religion and virtue and social peace. The scepticism of Charron was substantially the same as that of Montaigne, but he expounded it in a graver and more systematic form. As Montaigne can no more be justly credited with any essential orig- inality of thought than Charron, and the style and method of the two men are most unlike, it seems un- fair to represent, as is often done, the latter as a mere disciple and copyist of the former. According to Charron, science is unattainable: truth is hid in the bosom of God, and cannot be reached by the natural faculties of men. Education and custom mainly determine what our religion will be. Wisdom — that practical acquaintance with one's own spirit, its limits, weaknesses, and obligations, which displays itself in honouring and serving God, 134 SANCHEZ governing the desires and appetites, conducting one- self moderately and equally in prosperity and adver- sity, obeying the laws, customs, and ceremonies of one's country, trying to do good to others, acting pru- dently in business, being prepared for death, and maintaining peace in one's own heart and conscience — is what we may hope for and should strive for. Laborious efforts to attain science and passionate contentions about religion are alike vain. Reason is one of the feeblest of instruments and the sphere of certainty is of the narrowest range. Charron treats of virtue in a much warmer and worthier tone than Montaigne. He places it above everything else, and ascribes to it absolute dignity and unconditioned value in words which remind us of what Kant has said of the moral law. * But this is not to be regarded as a limitation of his scepticism, un- derstood as equivalent to agnosticism : it is a declara- tion of his faith, not an admission or profession of his knowledge, and therefore quite compatible with agnosticism.^ Trancis Sanchez (1552-1632) was uninfluenced by either Montaigne or Charron. He was of Judeo- Portuguese origin, but spent the greater part of his life in France, and chiefly at Toulouse, where he taught philosophy and also laboured both as a medical professor and practitioner. His writings are all in Latin, and deal mostly with anatomical and medical ' See on Charron (as also on Montaigne) D. Stewart in his Dis- seHation (Collected Works, vol. i. pp. 98-107) ; Owen, Sk. F R. , ch. iii. ; Ste. Beuve, Causeries du Lundi, Tol. xi. ; and Vinet, Mor- alistes fraA';ais au xvi^ siede. 125 HISTORY OF AGNOSTICISM subjects. The work on account of which he has been ranked among sceptics, A Treatise on the Noble and First Universal Science that Nothing is Known, al- though not published until 1581, had been in man- uscript since 1576, four years previous to the appear- ance of Montaigne's Essais. In it he criticises very courageously and, on the whole, very justly, the sci- ence and logic of his age, dwelling especially on the inadequacy of syllogistic rules and processes to the requirements of research; on the fallacious substitu- tion of words for things and of abstractions for facts ; on the folly of inventing imaginary entities and hav- ing recourse to occult qualities for the explanation of experiences instead of directly, patiently, and method- ically studying them ; on the worthlessness of verbal definitions and mere erudition ; and on the pernicious consequences of a servile dependence on authority. All that, however, shows not scepticism but good sense and the intellectual clearness of a naturally scientific mind. And I am not prepared to maintain that San- chez can be fairly classed among philosophical scep- tics. Yet there is, perhaps, no sceptic of the sixteenth or seventeenth century whose language has a more ag- nostic sound than his in many passages. But all his seemingly agnostic utterances may be found on exam- ination to mean comparatively little, and what would be quite harmless did it not suggest to the unwary reader more than it really means. They can all be referred to one and the same cause — a most unfortu- nate and indefensible definition of " knowledge " (scientia). He chooses to mean by it rei perfecta 136 HIS EBLIGIOUS FAITH cognitio, the complete comprehension of a thing both in itself and in all its relationships. Of course, with such a conception of knowledge he could not fail to reach the conclusion quod nihil scitur, and might have reached it by a single step instead of by the lengthened course which he actually followed — -that of showing the various respects in which human cog- nition, as regards alike its object, subject, and nature, falls short of perfection. But who pretends to have " a perfect cognition," an absolute comprehension, of anything? ISTo one. Only an iniinite intelli- gence can have knowledge in that sense even of the least of things. Hence the question as to whether or not Sanchez was an agnostic can only be settled by ascertaining whether or not he denied the attainabil- ity of knowledge in the sense in which other people af- firm its attainability. It seems to me that he did not. He certainly speaks of a scientia, and even of a scien- tia firma, very different from the scientia of his def- inition ; of a scientia quantum possimus, a scientia quantum fragilitas humana patitur; and announced his intention to follow up the Quod Nihil Scitur by another tractate showing how such knowledge or sci- ence may be attained. He has also informed us that he meant his method to begin from the perceptions of sense, the primary data of all knowledge, and to pro- ceed experimentally and critically {per experimentum et per judicium liherum, nan irrationaiile tamen). The book promised unfortunately never appeared ; but, notwithstanding that, we seem entitled to consid- er his so-called scepticism as only the initial and pre- 137 HISTOEY OP AGNOSTICISM paratory stage of his philosophy, and himself not as an agnostic but as an eminent precursor of Bacon and Descartes. He was a devout Roman Catholic. There is no evidence that his religious faith rested on scep- tical foundations.^ La Mothe Le Vayer (1588-1672), a facile and en- tertaining writer, highly reputed in his own age as a scholar, and influential both at Court and in the Academy, was a typical specimen of the seventeenth- century sceptic. A considerable number of the com- positions contained in the collected edition of his CEuvres are illustrative of his agnostic mode of thought and reasoning, but the Cinq Dialogues faits a I'imitation des anciens par Horatius Tuhero (1671) is the most famous and interesting of them in this ref- erence. Le Vayer was thoroughly imbued with the spirit of the old Greek Pyrrhonians and Academi- cians, and constantly used their arguments. Sextus was " his dear patron and venerable master," the Hypotyposes " a golden book, an inestimable and Di- vine writing," and the ten tropes " his decalogue." His motto was the two lines of Spanish verse— " De las eosa,s mas seguras La mas segura es dudar. " " Of things most sure the surest is doubt." He had a great predilection for. what has been called the geographical argument for scepticism. In- deed it was chiefly by dwelling on the different opin- ' See Owen, Sk. F. R., ch. iv. ; and L. Gerkrath, Franz Sanchez, &c., Wien, 1860. 128 LB VAYEE AND GLANVILLE ions and customs prevalent in different lands and ages that he attempted to produce the impression that there was nothing fixed and certain in physics, in logic, in matters of taste, in moral and religious prac- tice, &c. Yet, although nothing has been more varied and conflicting in its forms than religion, he repeat- edly declared that he did not question or doubt the religion founded on revelation, and that scepticism was favourable to true religion. He held himself to be a Christian sceptic, and described Saint Paul as another; but religion, and even morality, were not conspicuous either in his character or writings.^ The worthy English divine Joseph Glanville (1636-1680), author of Scepsis Scientifica and other works, ought not, I think, to be included among scep- tics. He was the enemy of confident dogmatising both in philosophy and in theology : he was the advo- cate of experimental investigation in the former and of the moderate doctrines of " the latitude men " in the latter. But in neither was he properly speaking sceptical. What is called his scepticism is little more than an emphatic dwelling on the uncertainties of what passed among his contemporaries for science. A mere enunciation of the view of causality so effec- tively employed by Hume in the interests of scep- ticism is no evidence of the scepticism of one who made no sceptical application of it.^ ' As regards Le Vayer see Owen {Sk. F. R., eh. r.), and L. Eti- enne, Essai sur la Moihe Le Vayer, 1840. ' There is a fine and just estimate of Glanville's character and position as a thinker in TuUoch's Rational Theology, vol. ii. pp. 443-452 See also Owen's Introductory Essay to his edition of 129 HISTOKY OJF AGNOSTICISM Theological agnosticism, however, was advocated with passionate zeal by a contemporary of Glanville, Jerome Hirnhaim (1637-1679), a dignitary of the Roman Catholic Church in Bohemia. His De Typho Generis Humani (1676) is one of the most violent and extreme attacks on secular science and natural reason. The validity even of the principles of caus- ality, identity, and contradiction is denied in it. All human knowledge is assumed to rest on the testimony of the senses, and that testimony is maintained to be proved untrustworthy by experience and the evi- dence of faith. The dogma of creation discredits the maxim ex nihilo nihil fit. The Incarnation shows that the belief that God cannot be contained in a body is untrue. Transubstantiation disproves the princi- ple that there is no accident without a substance. The Word of God, the revelation confided to the care of the Church, is alone certain ; and the duty of man is to accept it with entire and unquestioning faith. Worldly wisdom and science are error and vanity, and ought to be sacrificed to theology. Divine science. So Hirnhaim taught, — and taught, there can be no doubt, in all sincerity and with excellent intentions. Hirnhaim reminds us in various respects of Lamen- nais and his De Typho Generis Humani of the lat- ter's Essai sur I'lv difference en matiere de religion.^ Glanville's Scepsis Scieniifica, Kegan Paul, 1885. Dr. Ferris Greenslet's Joseph Glanville (New York, 1900) is a comprehensiTe and careful study, issued by the Columbia University, and published by the Macmillan Company. ' C. S. Barach, Sieronymus Jlirnhaim, Wien, 1864. Hirnhaim's book had great influence in Bohemia. 130 HUET Daniel Peter Huet (1630-1721), Bishop of Av- ranches, was a much more widely famed representa- tive of the same school of theological thought. He was a man of versatile genius and vast erudition, and the author of numerous works once celebrated and still occasionally consulted. His Demonstratio Evangeli- ca, 1679, and his Alnetance Qwstiones de Concordia Bationis et, Fidei, 1690, gave him a high reputation among theologians, notwithstanding the agnostic as- sumptions and conclusions to be found in them. In his Traite de la faihlesse de V esprit humain, pub- lished posthumously in 1723, a completely Pyrrhon- istic system is set forth and advocated in the interests of religion. In opposition to the methodical doubt of Descartes he contends for unlimited doubt of nat- ural reason. He represents the writers of the Bible and the Fathers of the Church as cherishing and in- culcating such doubt. By appealing to sundry words of Scripture, and by dwelling on the deceptions which proceed from defects in the senses and intel- lectual powers, from the changes in things, the diversi- ties in men, the want of a certain criterion of truth, the fallacies in reasoning, the dissensions of dogma- tists, &c., he endeavoured to prove that the human un- derstanding is incapable of attaining to certainty by the exercise of its natural faculties. Probability suf- ficient to direct us in the common ailairs of life is, he holds, all that reason can give us. As respects matters of philosophy those who affirm nothing are alone worthy to be esteemed philosophers. The " art of doubting" should be cultivated in order to pre- 131 HISTOKY OP AGNOSTICISM pare the mind to receive the faith. Certainty can only be obtained through recourse to revelation and grace for enlightenment and support. The best foun- dation for Catholic theology is Pyrrhonic philosophy. Thus Huet taught.^ Was the illustrious Pascal (1623-62) — the im- mortal author of the Lettres Provinciales and of the Pensees — also a religious sceptic ? The question has been repeatedly and elaborately discussed, and much unwillingness has been shown to answer it in the af- firmative. It seems to me that it is thus that it must be answered; and, indeed, that Pascal is the most striking example in history of a man Christian to the core and yet thoroughly agnostic in his estimate of natural reason. Certainly he made extraordinary con- cessions to the most absolute scepticism and bestowed on it extravagant praise. He declared Pyrrhonism " the truth," Pyrrho " the only sage before Christ," and that " to mock at philosophy is truly to philoso- phise " ; and although he affirmed the impossibility of universal doubt he said nothing against its reason- ableness, and so was merely sceptical even of his own scepticism. In dwelling on the doctrine of the Pall and its effects, on the weaknesses and inconsistencies of man, on the variations of morality and kindred topics, he forgot measure and proportion. In oppos- ing the head to the heart, understanding to faith, nat- ure to grace, he made sheer and violent contrasts of ' See M. Pattison's Essays, vol. i. ; Jourdain's art ffuet in Franck's Diet. d. Sc. ph. ; and especially C. Bartbolmess, Hud ou (e Scepticisme theologique, 1850. 133 PASCAL AND BAYLB what ought to be closely conjoined. Holding that so far as reason is concerned there are equal grounds for believing and disbelieving in the existence of God, the reality of moral distinctions, and the truth of Christianity, he was reduced to urge that men should act as if they believed in them, and as a means of be- lieving in them, on the same ground that a gambler when the chances are visibly or demonstrably equal bets on the side on which his interests lie; in other words, that they should wager on the side of God, virtue, and the Gospel, because if the result proves them to have been correct their gain will be immense, whereas if it should turn out that they have been mis- taken their loss will be insignificant. An apologetic of Christianity rested on such principles as those in- dicated may well be deemed unsatisfactory. It does not follow that Pascal performed no great service as a Christian apologist. In reality, he rendered by the way in which he applied in the Pensees the psy- chological or experimental method, the method of spiritual verification to the probation of the Chris- tian faith, an inestimable service — one which fully justifies his being regarded as one of the most original and profound of Christian apologists.^ Perhaps the most influential of the sceptics of the seventeenth century was Peter Bayle (164Y-1706), ' For the affirmative view as to Pascal's Scepticism see Cousin (Et'udes sur Pascal), Saisset (Le Scepticisme), and Ovven {Sketpcis, &c., ch. vi.) ; and for the negative view Vinet {Etudes sur Pascal) and Droz (Etude sur le Scepticisme de Pascal). See also Tulloch's Pascal, 1878, and Prof. Grote's Pascal and Montaigne, "Cont. Eev.," vol. XXX , July, 1877. 133 HISTORY OF AGNOSTICISM so widely known by his Historical and Critical Dic- tionary. The relations between his scepticism and the peculiarities of his character, the tendencies and controversies of his age, and his personal experiences are both interesting and easily traceable, but must be left by us unindicated. He had an insatiable and indiscriminate curiosity regarding facts and opin- ions, wonderful logical dexterity, extreme ingenuity in inventing and great fondness for maintaining par- adoxes, only feeble cravings either for fixed princi- ples or for unity and harmony in his speculations, a painful want of moral delicacy, and no depth of re- ligious emotion. His strongest passion was the love of toleration. "While intellectually honest he so keen- ly enjoyed discussion for its own sake as to care too little whether it led to truth or not. He himself called his scepticism " historical Pyrrhonism." It is commonly known as " erudite scepticism." The secret of it consists in so exhibiting the arguments for and against all opinions as to leave the mind puzzled and perplexed, and with neither power nor desire to' form a decision. The scepticism of Bayle was directed even more against theological than against philosophical dog- mas. This is a noteworthy characteristic of it. Scepticism is seen in the writings of Bayle assuming that especially anti-religious attitude so generally taken by it in the eighteenth and nineteenth centuries. Bayle sought, however, partially to conceal the anti- religious nature of his scepticism by arguing that faith and reason are contradictory, and therefore it 134 BAYLB'S AEGUMENTS does not follow that the dogmas of faith are not to be believed even vyhen proved irrational. Even in that case they may have as much right to acceptance as the conclusions of reason. Thus, on the plea of the harmlessness of his procedure, Bayle kept constantly evolving the absurdities v^hich he supposed to be im- plied in the doctrines of religion. He thereby brought the first stage of the movement of modern agnosticism to a natural close. By completing it he abolished it. By generalising its arguments he made evident the futility of its pretensions. Without pro- fessing to do anything of the kind, he really and ef- fectively showed how delusive was the notion that religion could reasonably hope to find a friend in scepticism; how mistaken was the policy of an al- liance of religion with the sense of doubt or nes- cience.^ ' See D. Stewart's Dissertation, pp. 313-324 ; A. Deschamps' Oenese du Scepticisme erudit chez Bayle ; and the articles of M. Pillon in the Annie philosophique, 1895, 1896, and 1897. CHAPTEE IV AGNOSTICISM OF HUME AND KANT I. hume: peefatoey as to his agstosticism The agnosticism of the present day flows directly from Hume and Kant as its two great fountain-heads. Of the two Kant was the greater philosopher but the lesser agnostic. He surpassed Hume in comprehen- siveness, constructiveness, inventiveness, and other qualities, but he did not equal him in critical acute- ness and clearness ; and, one single feature excepted, his whole agnosticism may be found more sharply and finely delineated in the writings of his predeces- sor. Hume is, undoubtedly, one of those " dead but sceptred sovereigns who still rule our spirits from their urns " ; probably he is of all the eminent Scotchmen of the eighteenth century the one who has most affected the general course and character of British and European thought. The influence of Adam Smith, as the author of the Wealth of Nations, has been more definite and visible, but also narrower and not so deep. Carlyle hardly exaggerates when he speaks of him as " the true intellectiial king of the eighteenth century " ; at least if the description be understood to refer not so much to his direct personal sway, to what he achieved himself or to the number 136 HUME^S INFLUEJ^OE of adherents and disciples he gained, but to his indi- rect influence, to what he stimulated or compelled others to do, what he brought to an end and caused to be begun. And his influence, viewed as a whole, may reason- ably be held to have been decidedly for good. Inci- dental and immediate evils of a kind it no doubt had. If the shaking of an unquestioning faith be essentially evil, his whole mode of theorising must have been evil. But if such shaking be far more a good than an evil, any evil Hume did must have been slight compared with the eventual good which he brought about. It is manifest that the latter supposition is the true one, and the one which facts have conflrmed. It was absolutely necessary that the questions which he raised as to the grounds or bases both of knowl- edge and of religion should be put, and that in the un- impassioned and searching way in which he put them. It was an essential condition of the new departure which was needed both in philosophy and in theology that the doubts which he suggested as to the very foundations of both should be propounded, and that by a powerful and constitutionally sceptical intellect. The time called for the man; the man was exactly suited to meet a want of the time. The sceptic and the dogmatist are alike the instruments of prov- idence. Authors like Huxley in England, Riehl in Ger- many, and Compayre in France have given us expo- sitions of Hume's philosophy in which they have ig- nored this aspect of it, or rather this the very essence 137 AGNOSTICISM OF HUME AND KANT of it ; have actually been unable to see any scepticism at all in the theorisings of Hume. How happens it that men so able have given us such a misrepresenta- tion, or at least one-sided representation of his doc- trine ? Partly, must be the answer, because Hume really was a pioneer of experimental science, in which capacity Huxley admirably delineates him ; a precur- sor of the Critical Philosophy, as Eiehl maintains ; and to a large extent a positivist, as shown by Com- payre. Experimentalism, epistemological criticism, and positivism neither exclude one another nor agnos- ticism, and they all coexisted with agnosticism in the mind of Hume, and coexist with it in his writings. It was thus rendered possible to study Hume so one- sidedly as to overlook his scepticism. But the main reason why the writers referred to have actually so erred is that they have substantially adopted his phil- osophical principles without seeing the full bearing of them as regards the validity of knowledge and the suf- ficiency of science. Because they have not seen how deep in their reach and wide in their range the scepti- cal consequences of his principles were, they have sup- posed that he had not seen it. That is a huge mis- take on their part. It is a historical fact that he did see it, and logically certain that in so seeing he saw truly. The agnosticism of Hume was of the very essence of his philosophy, and his philosophy was the natural outcome of a kind of philosophy which preceded it. His agnosticism, in other words, had for its basis modern philosophy so far as modern philosophy had 138 HUME'S PREDECESSOES been agnostic in tendency. In that lay to a great ex- tent the significance and importance of it. It was not merely the scepticism of an individual thinker: it was a scepticism which had been present and oper- ative in the speculations of some generations of thinkers, although it had not previously shown itself in its full force and in the light of open day. Hume evolved and gave admirable expression to the scepti- cism latent in the empirical or sensationist philosophy which, gradually acquiring strength from the days of Lord Bacon downwards, was to become for a time the ruling power in all departments of thought and life. The philosophers to whom he owed most were the ancient sceptics, and Bacon, Locke, and Berkeley. Like draws to like; and there is .abundance of evi- dence, although Hume seldom makes quotations, that at an early period of his philosophical studies he had made himself well acquainted — either at first or sec- ond hand — with the arguments and topics of the scep- tical schools of the ancient world. The Academica, De Natura Deorum, Disputationes Tusculanw, &c., of Cicero, his favourite prose classic, may safely be held to have, from the first awakening of literary am- bition within him, influenced his mode of thought as well as his style of expression. His method of inves- tigation was sincerely meant to be experimental and Baconian, whether or not it was really so. Locke made him a psychologist. His theory of knowledge was a simplification of that of Locke : in a multitude of instances he reaped what Locke had sown. From 139 AGNOSTICISM OF HUME AND KANT Berkeley he derived his views of abstraction, of the distinction between primary and secondary qualities, of the hypothetical character of substance, &c. The scepticism which Berkeley had applied to the outer world of matter Hume supplemented and completed by applying a like scepticism to the inner world of mind. With the spirit of religious doubt so preva- lent in the contemporary literature and society of France he was intimately in sympathy. When first propounded the most distinctive feature of his agnosticism was its claim to be wholly founded on experimental psychology. It was the character of its connection with psychology — inductive mental sci- ence — which gave it its originality, its influence, and such worth as it possessed. Had the psychology with which it was associated been as true as its con- nection therewith was firm and natural, agnosticism would have achieved a decisive victory. This, how- ever, was not the case; and, accordingly, instead of Hume's psychology proving his agnosticism, his ag- nosticism became the reductio ad absurdum of his psychology, and of all psychology of the same kind, — every merely sensationist psychology. We are not to suppose that he himself desired or meant it to be so. He is not to be thought of as starting with a convic- tion of the insufiiciency of the principles he adopted, and then labouring to make their insufficiency appar- ent by exhibiting the consequences to which they led. He accepted his principles in perfect good faith, see- ing no others which seemed to him so good. When he began to form a system they were those generally 140 HUME'S THEOKY OF KNOWLEDGE accepted, and the only ones on which he thought he could found it. After he had constrxicted it, and seen all that he could make of them, he remained unable to detect where they were at fault, and certainly un- prepared to abandon them, although he made no attempts to defend them, and was very indulgent towards those who attacked them. What was the extent of his faith in them is never likely to be determined, but the kind or quality of it was obvi- ously very appropriate to an agnostic. His merit was that, having adopted the principles of a merely empirical philosophy, he tried with rare skill and perfect dispassionateness to bring out of them all that was in them, and only what was in them; saw what that was with a clearness which has never been surpassed ; and did not attempt to conceal either from himself or others what it was — namely, not a satisfactory explanation of the world or a satis- factory foundation of science or conduct, but an indi- cation that knowledge is unattainable, the world an inexplicable enigma, and the state of man whimsically absurd. The theory of knowledge adopted by Hume was, as I have said, a simplification of that of Locke. Ac- cording to both Locke and Hume, all our knowledge is derived from experience, and experience consists of particular states of mind which presuppose no necessary conditions or elements of cognition. The states which compose experience — the contents of con- sciousness — are reduced by Locke to two kinds, — those which are given to us through the external 141 AGNOSTICISM OF HUME AIS-^D KANT sense, and those which are given to us through the soul's internal sense of its own operations; or, as he designates them, to ideas of sensation and ideas of re- flection. What Locke calls ideas Hume calls percep- tions; and perceptions — i.e.j Tasntal states of every kind — he reduces to impressions and ideas, — impres- sions being all those states which are produced in sen- sation, and ideas being the copies or images which the mind takes of them in thinking and reasoning. Im- pressions precede ideas, being the originals from which the latter are taken ; and they are as a rule more forcible and lively. Impressions and ideas differ, however, only in degree — only in strength and vivac- ity. The theory thus virtually is that all mental states may be analysed into mere sensations. What are called ideas are represented as not essentially dis- tinct from what are impressions, and should in con- sistency have been reduced to impressions, or, in other words, to sensations, as these are the only original im- pressions. Given sensations, and we should be able, according to the philosophical theory espoused by Hume, to explain how all knowledge, all minds, and the whole knowable universe have been formed out of them. The theory is of an attractive but delusive simplic- ity. It includes, however, all that is essential in the creed of a self -consistent empiricism ; and Hume's great merit was that, having adopted it, he was so true to it, and so courageously evolved what it im- plied. It was most desirable that there should be a clear exhibition of the consequences which naturally HUME'S SBNSATIONISM follow from the hypothesis that all the contents of consciousness may be traced back to and resolved into sensations, and that thoughtful men should thus be compelled to perceive that the path along vfhich an empirical philosophy sought to lead the human mind was one which must bring it to a bottomless abyss. That service Hume thoroughly accomplished. Grant him his primary psychological assumptions, which are only those which every consistent and coherent form of empirical agnosticism must assume, and the most sweeping of his agnostic inferences plainly follow. II. HUME S AGNOSTICISM IN GENEEAL Let us glance at some of those consequences of Hume's assumptions as to the origin and composition of experience in which the scepticism of his doctrine consists. One of the most obvious is that there can be no such thing as knowledge at all. That all our knowledge is reducible to impressions and ideas means with Hume, as it must mean with every person who expresses his thought correctly in those terms, that we can have no knowledge of other things than impressions and ideas ; no knowledge of an objective world, of a personal self, or of a Supreme Being ; no knowledge of any kind of real existence. But no knowledge of reality is equiv- alent to no real knowledge. Hence the problem of psychology thus viewed is not that of explaining how impressions and ideas come to be a knowledge of real 143 AGNOSTICISM OF HUME AND KANT persons and objects, but how they come to be taken, or rather mistaken, for siich knowledge. And that was the problem Hume grappled with. He asserted that we have no knowledge or experience except what is composed of states of consciousness ; that the only components of consciousness are sensations and ideas (their copies) ; that the belief in realities beyond consciousness, to which impressions and ideas corre- spond, has no discoverable foundation ; and that, con- sequently, all that mental science can be expected to do is to explain how mere states of mind come to ap- pear to be a world of objects, and how the erroneous belief that we have a knowledge of external realities — or, in other words, that our supposed knowledge of them is knowledge a!nd not illusion — takes irresistible possession of us. This is all that Hume has at- tempted to do as regards external realities. He has not sought to show that there are or are not such real- ities as material objects, but to show how, through the influence of custom on transient but recurrent sense-impressions, a belief which has no real or ra- tional warrant in the existence of such objects may be imagined to have grown up. Thus Hume would destroy the world of the ordinary man, of the mate- rialist, and of the realist. Having reduced knowl- edge to the sense-impressions and* traces or images of them in individual minds, he makes it apparent that whatever may be fancied to lie beyond those subjec- tive individual states nothing can be seen or known beyond them. His sensationism thus at once reveals itself as subjective idealism or illusionism, and at 144 HUME'S THEORY EXAMINED every onward step more fully so, until it stands dis- closed as perhaps the eompletest example of such a ■philosophy which has ever appeared. According to Hume we do not know external reali- ties. Does he allow that we can have a real knowl- edge of ourselves ? No. What he affirms is that the mind can know nothing except its own states. Hence it follows that it cannot know external reality. But it equally follows, as Hume rightly perceives, that it cannot know internal reality — i.e., its own self. There can, indeed, be no internal reality in a mind of which the only constituents are mere states, no knowledge of self in a consciousness composed exclu- sively of series or groups of mere impressions. Hume does not deny that we naturally come to think of and believe in what we call the mind or self as an indivis- ible, permanent, and active principle or subject pres- ent in its states ; but he maintains that we are not enti- tled so to think or believe, and that such an idea of mind or self is a " fiction " of the imagination. Of course there can be no knowledge of the reality of what has no reality. But, it may be said, he express- ly allows that the mind can know its own states, al- though only its own states. That is true. But the admission certainly does not come to much. For there is, in the first place, the difficulty of understand- ing how " a fiction " can either have " states " or " know " them. And, waiving that difficulty as much too large a subject for treatment here, there is, in the second place, the very obvious fact that what Hume allows the mind may know is just what it can- 145 AGNOSTICISM OP HUME AND KANT uot possibly know. The mind cannot know only its own states. It cannot know them unless as relating to something. It cannot know them without at least also knowing itself. Along with, or rather as correla- tive with, whatever is known, self or the ego is also known — is the simplest and most indubitable condi- tion of knowledge. It is a law without which con- sciousness is inconceivable, knowledge impossible; and yet Hume will only grant us a consciousness of which every state, a knowledge of which every act, so contradicts this law as to be an unthinkable absurdity. In a word, he does away with all that can properly be spoken of as a knowledge of mind, and allows us to retain as such only what is manifestly unworthy of the name. The character of Hume's theory of knowledge shows itself in its true and clearest light when applied to substances, whether things or persons. It resolves them all into less than dust and ashes, for even in dust and ashes there is something of reality, of being, of conceivability, whereas the elements into which Hume resolves all substances are the mere illusions of dreams which have no dreamers. Substances — all things which claim and seem to be existences — ^he holds are simply products of association and imagi- nation. In his own words, a substance is " a collec- tion of ideas, that are united by the imagination and have a particular name assigned them, by which we are able to recall either to ourselves or others that col- lection." Now, strange as such a view may seem to one who is unacquainted with the history of philoso- 146 ITS APPLICATION TO SUBSTANCES phy, even such an one may, perhaps, without diffi- culty see that Hume could not have consistently sup- posed substances to be more than he has defined them to be ; that his hypothesis as to experience being com- pletely analysable into sensations logically precluded his attributing to them any kind of external reality, independent existence, permanency, self-hood or the like. Indeed, even Hume here went further than he was entitled to go. For, while it is clear that if all that can be known may be resolved into states of mind, all that is known must consist of states of mind, and all so-called substances or things must be merely col- lections of states of mind although they are imagined to be of a very different nature, it is very far from clear how states or perceptions which have neither subject nor object — ^which are originally separate, suc- cessive, and in perpetual flux — can be collected. As nothing is supposed to exist save themselves, it would seem to follow that they must form themselves into those collections of ideas which are mistaken for sub- stances. Hume has not explained himself on that point. Certainly he has not made out that any such wondrous feat was accomplished by mere series of transient sensations as gathering and grouping them- selves into what men call their bodies and minds, the ocean, the earth, and the starry heavens. But he applied his agnosticism as to substances in all directions. Thus he sought to convict material substances of non-existence. Berkeley had already resolved matter into phenomena dependent on the ac- tion and perception of mind, and maintained it to be 147 AGNOSTICISM OF HUME AND KANT essentially nothing more than the sum of its appear- ances to sense. But although he thus exhibited the material universe as merely phenomenal, he did not exhibit it as objectively unreal. He filled up the void left by the abstraction of material substance with active mind. What we call physical phenomena he ascribed to the impressions and suggestions of the Divine Spirit on the spirit of man ; and all the math- ematical relations and natural laws of the universe he represented as simply manifestations of the Su- preme Intelligence. Hume cordially assented to his reasoning so far as it led to a purely negative result — the elimination of " material substance " ; but he de- cidedly rejected what was positive in its conclusion — the reference of sense-appearances to the energy and operation of Deity as their true and sole source. That he pronounced a view " too bold ever to carry conviction with it to a man sufficiently apprised of the weakness of human reason and the narrow limits to which it is confined." Yet the view which he sub- stituted for it was, in reality, far bolder. It was that the whole material universe and all its contents, as apprehended by each individual, is the creation of that individual's imagination and affected by the hy- jDostatising of impressions impressed by nothing ob- jective on nothing subjective. Hume explicitly and completely accepted this view, pronouncing external- ity a fiction due to association, and arguing that space and time are mere ideas which imply nothing exter- nal or real. The hypothesis of Berkeley is a timid and cautious one if compared with the one which 148 HUME ON CONSCIOUSNESS OF SELF Hume would have us accept as specially in accord- ance with the feebleness and narrowness of our rea- sons. The boldness of the idealistic theologian has often been thus exceeded by the professed modesty of the empiricistic sceptic. Hume took a still bolder step. It was one which Berkeley saw might be made, on the supposition that the denial of spixitual substance followed naturally from the same principles which had led himself to the denial of material substance. He protested, however, against its being taken on the ground that we are consciotis of our own being but not of the ex- istence or essence of matter. Hume paid no atten- tion to the protest, and treated mind just as he had treated matter. " What we call mind," he says, " is nothing but a heap or collection of impressions united together by certain relations, and supposed, though falsely, to be endowed with a perfect simplicity and identity." The appeal to consciousness, he main- tains, fails to assure us of the existence of any minds or selves which are not such mere heaps or collections of impressions. And, of course, it is not directly ap- plicable to any mind or self save one's own. Hiime himself, however, showed forgetfulness of the bear- ing of this fact when he wrote : " I venture to affirm of the rest of mankind that they are nothing but a bundle or Qollection of different perceptions which succeed each other with an inconceivable rapidity and are in a perpetual flux and movement. . . . There is properly no simplicity in it [the mind] at one time, nor identity in different: whatever natural propen- 149 AGNOSTICISM OP HUME AND KANT sion we may have to imagine that simplicity and identity." Hume was not entitled to affirm this of the rest of mankind until they had affirmed it of themselves. His inability to see in other men more than such bundles as he describes could only be a proof of the defectiveness of his vision if they recognised in them- selves the unity and permanency, the self-identity, self-consciousness, and self-activity which he denied to them. He could only express directly the testi- mony of his own consciousness. That he has attempted to do, and in doing so he has boldly ventured to deny his having any consciousness of a self. " For my part," he writes, " when I enter most intimately into what I call myself, I always stumble on some particular conception or other — of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time with- out a perception, and never can observe anything but the perception." Can this statement, however, be accepted ? Manifestly not, for it is wholly self con- tradictory. It implicitly affirms what it expressly denies, and implicitly denies what it expressly af- firms. " When I enter into myself I always stumble on some particular perception." Granted; but then in every such case you are there as well as the percep- tion. " T never catch myself at any time without a perception." Nobody supposes you do, but do you not catch yourself with your perceptions? " I never can observe anything but the perception." Oh but that is incorrect even according to your own account 150 HUME ON CAUSALITY seeing that you say it is you who observe the percep- tion. Whoever perceives catches himself perceiving, and therefore himself along with the perception. He never finds a mere perception any more than a mere self in his experience. Hence Hume might as well have denied his perceptions as his self — only in that case, as he allowed of nothing but perceptions, he would have had nothing whatever either on which to base or with which to build up his philosophy. In a word, what Hume tries to represent as the testimony of his consciousness is at this point so preposterously sceptical that his language refuses to convey it. Could any human language have given to it a self- consistent expression? Maintaining that mind was nothing but series or collections of transient perceptions, Hume could not, of course, allow that a Supreme Mind would supply any ground of unity or permanence in the universe. Mind was, according to him, as devoid of unity and permanence as matter. If, however, material objects and human minds alike are nothing but so many bun- dles of particular perceptions, any difficulties which we may have as to the possibility and intelligibility of these bundles will not be removed by reference to a bundle called the Supreme Mind. The scepticism of Hume does not spare even math- ematics. He perceived that it could not consistently coniine itself to what professed to be physical, men- tal, or theological science. From mere sensations it is impossible to derive the universal ideas on which necessary and exact deductions are dependent. If 151 AGNOSTICISM OP HUME AND KANT all knowledge be reducible to contingent and particu- lar sensations, one can establish no right to lay down any proposition as an axiom — as necessarily and universally true. Hume saw this, and therefore de- scribed even geometry as only approximately true. " When geometry," he says, " decides anything con- cerning the proportions of quantity, we ought not to look for the utmost precision and exactness. None of its proofs extend so far. It takes the dimensions and proportions of figures justly, but roughly, and with some liberty. Its errors are never considerable, nor would it err at all did it not aspire to such abso- lute perfection." Such a view of the nature of math- ematical science is a fair inference froin Hume's theory of knowledge, yet to have drawn it is a proof of his candour, since he could not have failed to an- ticipate that his readers would generally regard it as a reason for rejecting any theory or philosophy which implied it. Our author's views as to causality were, perhaps, those which attracted most attention. They are thor- oughly characteristic of his agnosticism. He admits that we believe that every object which begins to ex- ist must have a cause ; he allows that in this our nat- ural belief the idea of necessary connection is in- volved ; and he elaborately shows that the belief is the foundation of all other beliefs and inferences as to matters of fact. His agnosticism, in a word, does not show itself in denial of the idea or belief, but in the full admission of its existence and an emphatic in- sistence on its importance conjoined with a strenuous 152 HIS VIEW AS TO CAUSALITY contention that it has no warrant either in sense or reason. He could have had no objection to any one referring it to instinct, for if all our reasoning as to matters of fact be dependent on an irrational instinct it must, of course, be itself irrational, and that was just what Hume held. He could not have admitted that it was any refutation of him to insist on the ap- parent universality and necessity of the belief, for unless it were seemingly universal and neces- sary he could not infer it to be an invariable and constitutional illusion of the human mind. What he applied himself to establish was that for the sense of universality and necessity insep- arable from the belief in causality no justifica- ) tion was to be found either in reason or experience ; ^ that the only ground for it and for all the reasonings and conclusions dependent on it is custom or associa- tion, — a repetition of successive impressions which produces in us the delusion that one always is and always must be the cause of the other, and even the delusion that every change or event must have a cause. " After a repetition of similar instances, the mind is carried by habit, upon the appear- jance of an event, to expect its usual attendant, jand to believe that it will exist. The connection ' which we feel in the mind, this customary transition i of the imagination from one object to its usual attend- ant, is the sentiment or impression from which we form the idea of power or necessary connection. Nothing farther is in the ease." Yet, according to Hume, the causal belief, although only " the offspring 1S3 AGNOSTICISM OF HUME AND KANT of experience engendered by custom," is the source of all orderly and developed experience; the principle on which it is built up ; the foundation of all reason- ing regarding empirical objects, the only objects to which in his opinion reasoning is at all applicable. In other words, he represents the very basis of all seemingly intelligible experience as an illusion, and its contents when analysed as devoid of a single ele- ment of rationality. Hume's general theory of belief is not less sceptical than his theory of the causal belief. Belief is of its very nature a protest against scepticism, and the scep- tic, in order to vindicate his own consistency, must explain the true nature of it away. This Hume at- tempted. He represented belief as less akin to judg- ment than to imagination, and as indeed only an intenser and livelier form of imagination. He distin- guished belief from imagination not by what really difierentiates them — the fact that the former does and the latter does not imply a real or supposed apprehen- sion of truth — but by the greater vivacity and force of the former as compared with the latter. He thus implicitly denied belief to be what it really is, and ignored the numerous instances in which it is weaker and less vivid than imagination ; but, imquestionably, if he had been able to substantiate his theory of belief he would have gained a decisive victory for his scep- ticism. The agnosticism of Hume, so far as it has up to this point been before us, must be admitted to be both radical and consistent. Thoroughness is its. most 154 HUME'S BELIEF AND SPECULATIONS manifest characteristic. It goes straight to the very bases of belief, to the ultimate foundations of knowl- edge, and does not shrink to draw from its premisses their natural inferences even when most likely to cause unrest and alarm. And in this lies its chief merit, and the reason why it has exerted so great an influence as it has done on the development of philos- ophy and of thought in general. It compelled phi- losophers to concern themselves anew and earnestly with the deepest and most essential questions intelli- gence can raise, and to seek clearness and certainty as to the conditions which underlie all investigations and must determine the worth of all the efforts of reason to reach truth. It thus rendered inevitable a change in philosophic thinking from half ness towards whole- ness, from superficiality towards profundity, which necessarily affected other forms of thinking. III. Hume's agnosticism in eeligiow Holding the views which have been indicated, Hume must have been a singularly inconsistent think- er had he not been an agnostic in religion. In that case he would obviously have been unfaithful to the spirit, the principles, and the conclusions of his philosophy. He can be charged with no such incon- sistency. Pie was as agnostic in religion as in phi- losophy. He has sought to undermine all religious knowledge, all rational faith. Although well en- dowed with natural and social affections, his spiritual susceptibilities were not strong, and hence his scepti- 155 AGNOSTICISM OF HUME AND KANT cal reasonings were little checked or disturbed by his feelings even in the religious sphere. His intellect had little emotional resistance to overcome even when treating the most momentous religious questions as freely and coolly as if they were mere metaphysical puzzles without any practical bearing on life and con- duct. It was thus that he treated them. The de- pendence of religious opinion on philosophical spec- ulation has never been more obvious than in Hume's case. While Hume, however, may be fairly described as not less agnostic in theology than in philosophy, he ought not to be represented as more so. He showed no special desire to throw doubt or discredit on relig- ion. He simply dealt with it on the same principles, in the same spirit, and after the same manner as he dealt with physical nature and the human mind ; that is to say, he was, so far as his speculations were con- cerned, about as consistently and completely agnostic as an agnostic can be in the religious as in other spheres. I repeat, so far as his speculations were concerned. I do not speak of his personal belief, nor do I think that we know exactly what that was either in philosophy or in theology. It clearly did not coin- cide in either with his speculations. He saw that his principles led to conclusions which left no room for science or philosophy and could not be consistently and completely accepted without arresting all thought and action, and he did not pretend so to accept them, although he professed not to see on what other princi- ples he could proceed or what other conclusions he 156 HUME NOT HOSTILE TO RELIGION could deduce from them. We have no reason to sup- pose that it was otherwise as regards religion. There are no traces in Hume's writings, or in his correspondence, or in trustworthy accounts of him, of hostility to religion. He objected to being called a Deist, and manifestly because the name implied an- tagonism to Christianity. He did not directly as- sail Christianity. His reticence in regard to it was in striking contrast to the attitude and con- duct of the English Deists and the French Encyclo- pedists. He was the intimate friend of some of the most eminent divines of Scotland in his day, — able and cultured men, but certainly not sceptical as to the truth of Christianity. In his intercourse with them religious subjects were avoided, plainly by a sort of tacit understanding on both sides, only explicable, I think, by their recognition of the difference between " Davie " Hume the natural man and David Hume the celebrated Academical philosopher, and of the un- reasonableness of expecting that the latter, whatever might be the personal faith of the former, would dis- avow his speculations so long as he did not see that they had been refuted. When his friend Mr. Boyle attribiited the uncommon grief manifested by him on the death of his mother to his having thrown off the principles of religion and so deprived himself of its consolations, his answer, we are told, was, " Though I throw out my speculations to entertain the learned and metaphysical world, yet in. other things I do not think so differently from the rest of the world as you imagine." 157 AGNOSTICISM OF HUME AND KANT Hume did not profess to be a philosopher except when he was philosophising. He did not attempt to conform to his scepticism when he wrote on political subjects, or composed the History of England, or en- joyed the society of his friends. Why should he be supposed to have done so in regard to religion ? Is it because he has occasionally spoken as if his theories merely undermined religion and metaphysical specu- lation, and has even told the students of science that theirs is the only kind of knowledge worth possess- ing? But he has as explicitly told theologians that faith cannot be overthrown by reason and that " Di- vine Revelation is the most solid of foundations." Like so many " academical philosophers " Hume was quite willing to compliment science and reason at the expense of religion and faith, or religion and faith at the expense of science and reason, although aware, or perhaps rather because aware, that reason and faith, science and religion, were alike uncertain, if his scep- ticism were true. If his scepticism were true; he was not unconscious of the if there — not unsceptical of his own scepticism. To what extent he was so we shall probably never know. It was not his business, and still more mani- festly not his interest, to enlighten the world on that point. What he has made clear, however, is that those who adopt his premisses must be prepared to adopt his conclusions, and even must in the main ac- cept them all, seeing that those which bear destruc- tively on ordinary knowledge and science are not less legitimately drawn than those which affect religion. 158 HUME'S AGNOSTICISM IN RELIGION Those who adopt his premisses and draw only con- clusions unfavourable to religion show that their logic is biassed by anti-religious prejudices. In theology the agnosticism of Hume had the same characteristic and merit as in philosophy, and the re- sult was the same. Here too it was thorough; it went to the foundations — passed by all questions of secondary importance, and dealt with those on which the entire fate of religion as a claimant to reason de- pended. And here too this was, on the whole, a de- cided service to religion, the deepest truths of which are only to be conclusively established through exclu- sion of the deepest doubts. The decisive and ulti- mate victories of faith must be those gained over unbelief as to what is absolutely fundamental and es- sential. Hume helped more than any one else of his time to do away with halfness and superficiality in theology no less than in general philosophy. He con- vinced thinkers that the Deistic assumption of the self-evident certainty of so-called natural religion was a mere assuniption ; that natural religion was no more indubitable than revealed religion; that both those who would attack and those who would defend relig- ion must go deeper down than they had been doing. The change introduced by Hume was thus a very great one. It was the agitation, so far as religion was concerned, not merely of the question. What in it is true ? but also of the question, "V\^aether or not there is any truth in it ? His denial of the ability of the mind to rise above sensible experience, and his views of substance, cause, 159 AGNOSTICISM OF HUME AND KANT and personality, left liim no principles on whieli he could justify belief in the Divine existence. And he did not seek to justify it. On the contrary, while he did not openly assail it, he, in his character of philo- sophic sceptic, endeavoured to shov? that what had been regarded as its rational bases were untrust- worthy. He set aside as not deserving of discussion the opin- ion that we know God by intuition. Those who hold that opinion should take note of Hume's estimate of it, instead of merely attaching, as they so often do, an excessive value to his criticisms of the theistic proofs. The a priori argument he rejected without any serious consideration. A^Tiatever had an appearance of a scholastic origin or character got slight justice from him. His treatment of the a priori proof strikingly exemplifies this. Instead of being studied with in- terest or insight, instead of being examined and judged with impartiality and care, it is summarily condemned on the assumption that every matter of fact is a contingent existence — a mere and most doubtful assumption which manifestly begs the whole question at issue. The reasoning by which Hume attempted to get rid of the a posteriori proof is ingenious, and has not un- deservedly attracted much attention. It is entirely founded, however, on his agnostic view of causality, and must appear inconclusive to those who do not accept that view. It is equally in his peculiar view of causality that he finds the principle of his cele- brated argument against the doctrine of a future dis- 160 NATUEAL HISTORY OF RELIGION ti'ibvition of rewards and punishments. The argu- ments just referred to I do not require either to ex- pound or examine. It is sufficient for my purpose to have thus referred to them. It will not be questioned by any one that, if they be valid, belief in God and the immortality of the soul must be without rational war- rant, the so-called " light of nature " an illusion, and all so-called natural religion merely blind instinct, inherited prejudice, caprice, and superstition. The most valuable and interesting, perhaps, of Hume's writings regarding religion is his Natural History of Religion. This treatise had the great merit of initiating that historical method of studying religion which has been found so fruitful. In it Hume very properly distinguished between the rea- sons and the causes of religion — i.e., between the grounds which may be adduced in vindication of it and the motives or influences which may have actually evoked it and made it what it is : and with no less jus- tice showed that in dealing with religion simply as a historical' phenomenon we have only to do with its causes, not with its reasons as such. He likewise quite correctly showed that its causes had often not been reasons but imaginations, feelings, casual occur- rences — illusions produced by fears and desires, ex- ternal causes and circumstances. But when he endeavoured to produce the impres- sion that the reasons of religion were not among its causes, or even that reason had ever been entirely without influence in the formation of religion, his scepticism made itself manifest, and led him to con- 161 AGNOSTICISM OF HUME AND KANT travene and contradict the truth. Thus to dissociate religion from reason was consistent with his agnosti- cism, but it is not warranted by the history of relig- ion when studied in a strictly historical manner. The rational apprehension of religious truth has often been far from the strongest factor in the rise and growth of religion, but it has always been a factor. In keeping it out of sight Hume ignored what alone explains why the history of religion has been the pro- gressive movement which he himself represented it to be. He was candid enough to recognise that the history of religion had, on the whole, from beginning to end, steadily advanced towards reasonableness, growingly increased in consistency. But if so, must not the inspiration and power of reason have pervaded it throughout ? Must not a continuous progress tow- ards truth be one essentially true? Must not the history of religion, even as treated by Hume and by many since Hume, as well as by students of every kind who have shown regard for its facts, be allowed to be one which bears testimony not for but against agnos- ticism as to religion ? Hume dealt with revelation agnostically in his cel- ebrated Essay on Miracles. He assumed revelation to be essentially miraculous, and only provable, if provable at all, by miracles of an external character perceptible by the senses. Many Christian apologists of the present day would decidedly refuse to admit the assumption, or to accept the conceptions of revelation and miracle which it presupposes, but they were uni- versally received by the contemporaries of Hume. 16a "ESSAY ON MIEAOLES" Besides, even although it may be said that his ideas of revelation and miracle belonged to an age which has to a considerable extent vanished, his mode of treating them must be allowed to have been none the less thoroughly characteristic of his agnosticism. He did not question the conceivability of miracles ; he thought he had a distinct enough notion of them to define them as " violations of the laws of nature " — i.e., events brought about not by natural means but by an agency above, beside, or opposed to nature. He did not attempt to prove the impossibility of miracles ; he recognised that that could only be done by disprov- ing the existence of God and of supernatural beings. But he undertook to show the incredibility of miracles — their unprovability to those who have not been wit- nesses of them. Experience, he argued, assures us that the laws of nature are invariable, while human testimony is deceptive, and can never therefore cer- tify a deviation from these laws a miracle. Even if witnesses were always trustworthy, and if there were a full proof from testimony in favour of a miracle, it would only be equal to the full proof from expe- rience which is against it, and consequently could not entitle us to prefer belief in a miracle to belief in the inviolable uniformity of natural law. Hence a mir- acle, even if attested by testimony in the highest pos- sible degree, can never be rendered credible in the lowest degree, but in reality never is so attested, see- ing that testimony is frequently erroneous and men- dacious. Such is the general tenor of his argument — one 163 AGNOSTICISM OF HUME AND KANT which it is not necessary to criticise or to endeavour to show to be fallacious, but of which it may be de- sirable to indicate how agnostic it is. First, then, the argument while denying that a miracle can ever be so proved as to be credible allows that it conceivably might be fully proved, as fully proved as a law of nature. It is only because testi- mony has not been found to be universally true that the proof in favour of a miracle is represented as nec- essarily weaker than the proof in favour of a law of nature. But it is easily supposable that human testimony might have been always veracious and ac- curate. Suppose it to have been so. What would theil, according to Hume's ovvoi account, and if his ar- gument be valid, be the state of a human intellect in the presence of testimony in favour of a miracle ? This : there would be a full proof for and a full proof ■ against the miracle — an equal proof on opposite sides to which no addition could be made, and a perfectly truthful human intellect cognisant of both but utterly incapable of ever coming to a rational decision for or against either the miracle or the related law of nature. It suits admirably an agnostic like Hume to devise an argument which thus implies that the human mind, even at its best estate and in the most favourable cir- cumstances, must be of a whimsical and absurd nat- ure ; but a non-agnostic can hardly fail to regard such an argument as in the highest degree suspicious even prior to logical scrutiny. Further, were the argimient in question valid, a thing might be true, and clearly seen to be tru?, and 164 HIS AEGUMBNT EXAMINED yet evidence of its truth might he impossible to be given to those who were not eye-witnesses of it. It is an argument to the effect that even if a miracle oc- curred, its having occurred could not be made known to any one who did not see it take place. Now, that a thing may be true while one has not sufficient evi- dence of its truth, is an obvious and incontestable ■proposition; but that a thing may be true and have been observed, and yet that no sufficient evidence can be given for it to others than the eye-witnesses, is an assertion of a very different kind, and indeed a para- dox which could only originate in an agnostic imag- ination, and which implies that there is an impassable gulf between man's mind and a certain class of real or at least conceivable facts. Still further, if the principle of Hume's argument be valid it should prove more than he inferred from it ; it should prove that even the eye-witness of a mir- acle could not have sufficient evidence of its existence to make belief of it rational. The ground on which Hume rejects the evidence of testimony when ad- duced in support of a miracle is simply that testimony does not invariably correspond to the truth of facts, while the laws of nature are, it is alleged, invariable. But the testimony of sense itself is not always accord- ant with the truth of facts. We see wrongly accord- ing to the laws of vision as well as correctly. The senses deceive us, and there is no miracle involved in their deceiving us. Hence on Hume's principles even the senses can in no circumstances afford a suf- ficient proof of the occurrence of a miracle. His ar- 165 AGNOSTICISM OP HUME AND KANT gument not only places an impassable barrier between the truth and those who have heard it reported by others, but raises an insurmountable barrier between a man's own mind and what may happen before his eyes. Finally, Hume directly sought by his argument concerning miracles to justify scepticism as to revela- tion, and so regarded the argument was not less rele- vant than ingenious. If valid at all, what it proves can be no less than that God could not make known His character or will to mankind otherwise than through the laws of nature ; that even if He wished to put Himself in direct and special communication with His creatures He could by no means carry His desire into eilect. That is a thoroughly agnostic con- ception, and yet how much gnosticism there is in thus attempting to limit the power of omnipotence. The speculative attitude of Hume towards religion has been thus described by himself when concluding his Treatise on its Natural History. " The whole is a riddle, an enigma, an inexplicable mystery. Doubt, uncertainty, suspense of judgment, appear the only result of our most accurate scrutiny concerning this subject. But such is the frailty of human reason, and such the irresistible contagion of opinion, that even this deliberate doubt could scarcely be upheld, did we not enlarge our view, and, -opposing one species of superstition to another, set them a quarrelling; while we ourselves, during their fury and contention, happily make our escape into the calm, though ob- scure, regions of philosophy." Such by his own con- 166 ISSUE OF HUME'S PHILOSOPHY fession was the final issue of a thorough and com- plete scepticism. But what a dismal, dreadful issue ! For the vast majority of mankind, who certainly can- not escape into the regions of philosophy, no hope, no refuge, only the doom of living and dying in the dark- ness of delusion. For the few who, like Hume him- self, can escape into them, no prospect beyond that of finding them as empty, as unreal, as unsatisfying as he has repeatedly and pathetically confessed them to be, and aa obscure, as enigmatic, as uncertain as the region out of which they had fled. It must now, I think, be apparent that those who have seen no scepticism in the speculations of Hume have not examined them very closely, and that any characterisation of Hume as a philosopher which ig- nores the agnostic in him is quite like an estimate of the play of Hamlet which leaves Hamlet out of ac- count. All else in the mind and activity of Hume can no more make up Hume if his agnosticism be ex- cluded* than the other characters of the drama can make up Hamlet if the Prince of Denmark be omit- ted. It is an injustice to Hume himself not to place his agnosticism in a clear light, for it is above all that which gave him, and still gives him, his eminent place and immense significance in the history of philoso- phy. It was what opened the eyes of Eeid and of his followers to the necessity of seeking anew for the foundations of knowledge and belief. It was what roused Kant, as he himself avowed, out of his " dog- matic slumber," and compelled him to undertake those labours in which all subsequent German philos- 167 AGNOSTICISM OP HUME AND KAiSTT ophy may be held to have originated. It was what directly and immediately evoked the latest great stage or phase of philosophy, the one which influences so powerfully all contemporary thought and life. The scepticism of Hume deservedly made its author's name immortal and his influence enormous. It had all the comprehensiveness and thoroughness appro- priate to a radical scepticism, while easily intelligible and free from all scholastic formalism, technicali- ties, and pedantry. It was singularly bold and un- sparing, and yet skilfully conciliatory. It presented the most subtle thoughts in an attractive form. And, further, it was a really logical deduction from long dominant and widely accepted philosophical princi- ples. As the means of bringing to light the erroneous- ness of those principles it was a needed, a reasonable, and even a providential thing. The justification of it has been ample, being whatever is true and good in the intellectual and spiritual development to which it has given rise. IV. KANt's ANSWEE to HUME The theories of Hume could not fail to be perceived to have an immense significance for philosophy and theology, for science, religion, and morality. They brought fully to light the scepticism latent in the em- piricism derived from Bacon, Gassendi, and Hobbes, and at the same time made manifest that the dogmatic rationalism which had appeared as Cartesianism, Spinosism, and Wolfianism could supply no rational 168 KANT'S ANSWER TO HUME answer to it. In a word, they plainly showed the necessity for a thorough revision not only of British but of European speculative thought. They required a refutation of such a kind as could only be obtained through a reinvestigation of the entire problem of knowledge. Thomas Reid and Immanuel Kant clearly recog- nised the necessity and sought to meet it. Their an- swers to Hume were to a considerable extent identical or accordant, and to that extent they were substantial- ly satisfactory. Kant's answer was reached through a process of investigation much more profound and systematic than Eeid's, but one which often led him to false conclusions, and, indeed, issued at many points in a scepticism as radical as Hume's own. Like Reid, he conclusively showed that knowledge could not be reduced to sensations, and that intelli- gence implied in all its operations necessary condi- tions as well as contingent impressions, and so far he substantially disposed of the scepticism of Hume by proving its dependence on an inadequate and erro- neous psychology. But when he proceeded to argue that the constitutive principles involved in knowledge have to do only with phenomena or states of conscious experience, but are wholly incapable of placing us' face to face with things ; that they have a merely sub- jective and relative value, but give us no information as to external reality ; that while useful in co-ordinat- ing and unifying our perceptions they in no degree justify our affirming that there is anything corre- sponding to these perceptions, — then he virtually un- 169 AGNOSTICISM OF HUME AND KANT did his own work, and became not the conqueror but the lineal successor of Hume. Keid was too single and simple minded thus " both to run with the hare and hunt with the hounds." Hence his work as a philosopher, although far inferior to Kant's in most respects, was greatly superior to it in consistency. It was wholly anti-agnostic. With it, therefore, we need not here further concern ourselves. With Kant's, however, we have still to do, although only in so far as it is agnostic. The limits within which a sketch like the present must be confined forbid my attempting either to de- scribe or refute at length the agnosticism of Kant. I must, in fact, restrict myself to indicating the respects in which I dissent from what is peculiar to it and essential in it. All that is so is contained in one work, The Critique of Pure Reason; and, indeed, is just what is distinctive of the three theories expound- ed in that work and derived by Kant from his exam- ination of the three faculties which in his view have to do with knowledge, namely, sense, understanding, and reason. All else in the Criticism is merely scaf- folding, not building. Hume had explained away everything like neces- sary connection in thought. He had dissolved, by analysis, all apparent knowledge into unintelligibil- ity. He had got rid of all synthetic judgments. That being the general result of his scepticism, the general problem with which Kant had to grapple was to show that there were necessary synthetic judg- ments so rooted in the very constitution of intelli- 170 TEANSCBNDENTAL ESTHETIC gence that they could not be rationally destroyed by any analysis. To solve it he raised three questions : Are synthetic a priori judgments possible in mathe- matics ? Are such judgments possible in physics ? And, are they possible in metaphysics? The first question he answered affirmatively in his critical the- ory of cognition through sense (his transcendental (esthetic) ; the second also afiirmatively in his critical theory of cognition through understanding {tran- scendental analytic) ; and the third negatively by his critical theory as to cognition through reason (tran- scendental dialectic) . § I. Transcendental Esthetic The general scope of Kant's investigation into the capacity through which objects are given and percep- tions furnished to us may be thus stated. The effect of an object upon the senses is a sensation. The sort of perception which relates to an object by means of sensation is an empirical intuition; and the undeter- mined object is a phenomenon. That in the phenom- enon which corresponds to the sensation is (in Kan- tian phraseology) its matter, and that which causes it to be arranged under certain relations its form. The matter comes from without, the form must lie within to receive it. The form regarded wholly apart from the matter is said to be pure; the pure form of sense to be pure intuition. It is with the pure forms of sense or pure intuitions that Transcendental ^Esthetic has to do ; and to accomplish its work it first isolates the 171 AGNOSTICISM OP HUME AND KANT ' sensuous faculty from all other faculties of mind, and then takes away from intuition all that is given through sensuous impression, so that nothing may re- main but pure intuition. The result, according to Kant, is the discovery that there are two, and only two, pure forms of sensuous knowing, viz., space and time — space the form of external sense, and time the form of internal, and, mediately, of external sense. Was Kant entitled to affirm these positions ? Not, it seems to me, at the outset of his critical inquiry; not until he had settled a number of questions which he never even distinctly raised. He assumed, for example, that the sensuous fac- ulty can be isolated from our other faculties of cog- nition. What, then, does the assumption amount to ? I Virtually to assuming the falsity of two doctrines well I entitled to a careful discussion — the doctrine that sen- I sation is the root of all thought, and also the doctrine 'j that thought in its essentials, the reason in its generic [ integrity, is the condition of sensation. If all thought j be, as experientialists hold, involved in and evolved out of sense, the separation of the sensuous faculty ' from other faculties is impossible, as the other f acul- i ties are developments or transformations of sense. The fact that Kant did not adopt the doctrine, but, on the contrary, aimed at definitely refuting it and thoroughly discrediting the sceptical conclusions j which had been deduced from it, only made it so much the more necessary for him not to| assume to be done j what, according to the doctrine in question, could not possibly be done. And the necessity was yet further 173 THBOEIES OF FORM AND MATTER increased by the fact that most even of those who re- ject the em piricist t he ory of knowledge will so far agree with those who maintain it as to jdeny that sense canbg ^separate d from what else is in cognition in the way^Eant aupposes. The assumption that sensuous ' cognition is the result of an impression and a form — sight, for instance, of an impression produced by light without and received into the form of space within — is happily not the only alternative supposition to the view which would make sensuous cognition the result I merely of the external impression. Further, Ka;nt began his investigation by dividing sensuous cognition into matter and form, on the as- sumption that the former comes from without and the latter from within. But was not starting thus, if not begging^ the jguestion in dispute, at least unduly fa- vouring a particular answer ? Was it fair even to suggest at the outset that the form is in any respect more subjective than the matter ? Prima facie it seems just as probable that the form is without and the matter within, or that both form and matter are without or both within, as that the matter is without and the form within. Until proof is produced that space and time are within the mind or subjective, every mode of "expression which implies that they are so may well be deemed objectionable. Kant's account of the matter and the forms of sen- suous cognition, it may be added, implies that the lat- ter are so separate from and independent of the former as to be given in the mind previous to all experience* and to exist in it as pure intuitions. That view, how- 173 AGNOSTICISM OP HUME AND KANT ever, does not seem to be confirmed by the observation i or analysis of the processes of sensuous consciousness. We cannot apprehend space before or apart from experience. Any apprehension of space is already experience. We apprehend bodies as spatial, as ex- ternal and extended, but have not the slightest con- sciousness of being in possession of an intuition of space which we superimpose on bodies and thereby attain to a knowledge of them. The so-called form and the so-called matter of sense-experience condition I each the other, are inseparable, and are not related to I each other as a subjective to an objective constituent. No sufficient reason has been shown for conceiving of space as given in the mind before all actual percep- tions, or for representing its a priori character, as- suming it to be of such a character, as dependent on its being merely subjective, simply a mental form. It would seem to be capable of being described with pro- priety as a form only in the sense that external ob- jects must be apprehended and thought of as in it, and it only as capable of containing such objects and ren- dering possible their groupings and motions. Kant's exposition of space was a remarkable and important piece of work. It opposed to experiential- ist accounts of the cognition of space a nativistic theory of a bold and ingenious character, containing a large amount of important truth, and presented with so much skill as to make an epoch in the history of the doctrine of external perception. For a length- ened period it was very generally regarded as having definitively shown the futility of attempting to trace 174 KANT'S PECULIAR INFERENCES the pi-inciples of mathematics to roots latent in expe- rience. There is less confidence felt in it now among competently informed students of psychology. The problem is still under discussion, and expert opinion is as much divided regarding the solution of it as it ever was. Nevertheless the study of the subject to which it relates has certainly been in various respects much advanced since Kant wrote, and largely so be- cause of the impulse which Kant gave. His views as to space and sense-perception no longer satisfy, but that is owing, perhaps, almost as much to their sug- gestions having been followed up as to their defects having been detected. Kant has quite conclusively shown that the cogni- tion of space is not a general notion, not a concept derived by abstraction and generalisation from a mul- titude of particulars. But he was hasty in inferring that because not a general notion it must be a pure intuition. A cognition may be neither a general no- tion nor a pure intuition. It may be also either a particular notion or an impure intuition. And^ in fact, so far as space is apprehended through sense — and it is largely so apprehended through muscular mobility, touch, and vision — it is no t apprehended by pure intuition. Berkeley and Hume, by showing 1 that we cannot even imagine space apart from colour and figure, had refuted by anticipation Kant's view of the apprehension of pure space through sense-per- ception. And, it may be added, consciousness clearly testifies that in the most abstract, supersensuous, purely rational thought, space can only be cognised 175 AGNOSTICISM OF HUME AND KANT by us as that in which bodies may be contained, in which lines, circles, planes, cubes, &c., may be drawn or conceived to be drawn, and in which motions may take place or be imagined to take place. As to the nature of space Kant draws from his in- vestigations two distinctive and very peculiar infer- ences, both of which seem to me unwarranted. The first is that space represents no thing-in-itself or attribute of a thing-in-itself. Now that inference was manifestly premature unless he himself knew what a thing-in-itself was, and until he had also in- formed his readers what it was. Yet he affirms and insists on the truth of the inference without giving any information as to the thing-in-itself. This is so illogical a procedure that one naturally seeks for an explanation of it ; and that is not difficult to find. It is just that Kant started on his critical investigation with a bias in favour of a particular conclusion and worked steadily under its influence to the close of the investigation. What was that particular conclusion ? This — ^that we neither know nor can know anything whatever about the thing-in-itself. Now, that such was Kant's belief was certainly a sufficient reason for his not explaining to us what a thing-in-itself is. It is rather strange, however, that he should not have seen that it was an equally sufficient reason for his not volunteering to tell us that space, or anything else, is not a thing-in-itself. Where knowledge ceases the right to deny ceases as well as the right to affirm. If we know and can know nothing about things-in-them- selves, we cannot possibly be entitled to say either ire OBJECTIONS TO KANT'S VIEW what belongs to theiii or what does not belong to them. If we know nothing about them, then, for anything we know space may belong to them, or may be one of them, if there be more than one of them. Further, the cause of our inability to know anything about them, and of our consequent inability to affirm or deny anything about them, is a most obvious one. It is that the very conception of the Kantian " Ding an sich " is, as has been said, " ein Unding." It is a pseudo-conception, an inconceivable conception, which owes its existence wholly to unreason. It has been a most disastrous conception, the seed of a vast growth of nonsense which has pretended to be knowl- edge or science or philosophy of the unknowable. One is sorry to have to say it, but Kant may be re- garded as the father of all those who during the last hundred years have vainly laboured to acquire and communicate knowledge of the unknowable. The second of Kant's inferences as to the nature of space is that it is only a subjective condition <)f sense. I admit none of the premisses from which the infer- ence is drawn, and reject the inference itself. If space be not known by us as objective and external, nothing is so known by us, and we can have no intel- ligible and consistent conception of objectivity or ex- ternality. The mind has no consciousness of space as subjective. It knows it only as independent of it- self, as out of itself, as what it and what the objects it knows are in. It knows it not as what is given by the mind, but as what is given to the mind and appre- hended as an external quality. And we have no 177 AGNOSTICISM OF HUME AND KANT right to assume that it is not what it is apprehended as being. The testimony of consciousness must be accepted as true until proved to be illusory. Of proof that it is illusory none has been produced. For setting it aside no weightier reason has been assigned than the mere conjecture, the alleged possibility, that the perceptive faculty might have been so constituted that space and its relations would not have been valid to it. The conjecture has not been shown to be even intelligible, nor the alleged possibility to be, properly speaking, conceivable. The existence of intelligences incapable through their limitations of man's knowl- edge of space can prove nothing against the validity of his knowledge of it. Ignorance is no contradiction of knowledge. If, as some Jewish philosophers have maintained, God is not in space but space in God, — space not the place of God, but God the place of space, — or, if He in any other imaginable or even unimag- inable way transcend space, it cannot be therefrom rationally inferred that man's geometry must be false in God's sight. Omniscience cannot regard any sci- ence as nescience, and still less any truth as an error. That the cognition of space is so far dependent on the constitution of the perceptive faculty may be admit- ted without any concession to the fiction of the sub- jectivity of space or of the possible or partial non- validity of necessary truth. An intellect for which the relations of space were not valid would be an in- tellect of such a kind that although its existence may be verbally affirmed it cannot be truly, i.e., ra- tionally, thought. 178 KANT'S THEOEY EXAMINED To such objections as the foregoing Kant and his disciples can only reply that they do not deny space to be perceived by us as objective and real, and neces- sarily so perceived ; that, on the contrary, they affirm it to be empirically real in the sense that it is objec- tively valid for us, inasmuch as necessarily seeming to contain all that can externally appear to us, and that by maintaining its transcendental ideality, as not be- ing or belonging to any " thing-in-itself," they justify the common consciousness in believing in its empiri- cal reality, which is all that is needed to repel scepti- cism. But to such a defence as that the obvious an- swer is that what they are charged with is precisely what they admit — namely, maintaining that space is real and objective in the sense of necessarily seeming so, and maintaining at the same time that it merely seems to be so, while actually ideal and subjective; and that to do so is not to attempt to repel scepticism but to vindicate it, and is, in fact, virtually to repre- sent the human intellect as self -contradictory and un- trustworthy. Consistently to hold both the empirical reality and the transcendental ideality of space is im- possible. Nothing can be objectively valid for us which can be proved by us to be only subjectively ex- istent. It may be added that if space be merely sub- jective the things perceived in space must be merely subjective also, and the most rational view of the uni- verse will be that it lies, as Schopenhauer maintained, within the brain, or that it is, with all individual brains included, one vast illusory concept. Kant's doctrine of time closely corresponds to his 179 AGNOSTICISM OF HUME AND KANT doctrine of space, and has the same defects, so that it may be left both unexplained and uncriticised. It is even less satisfactory, however, than his doctrine of space, inasmuch as it takes no notice of the differences between the cognition of time and that of space — dif- ferences so radical as to make it doubtful whether time ought not in consistency to have been ranked by Kant, as M. Pillon and other Neo-eriticists hold, rather among the categories of the understanding than among the forms of sense. The latter portion of the Transcendental Esthetic consists of remarks meant to illustrate and enforce those two positions: 1st, Space and time are condi- tions only of phenomena; and 2nd, They are the nec- essary conditions of phenomena. They are the most distinctive positions in Kant's theory of sensuous knowledge, the theory on which his whole philosophy is based. Whoever denies either position must be ranked among his opponents, however highly he may admire his philosophical genius and in whatever other respects he may acknowledge his services. Kant considers that the first of these positions de- livers philosophy from great difiiculties. To regard time and space as conditions only of phenomena dis- poses, he thinks, of all the metaphysical perplexities connected with them. These perplexities arise, in his opinion, simply from our forgetting that time and space are only valid within the sphere of phenom- ena, and cannot be legitimately made use of beyond it. To recognise that time and space are not real ex- istences, but only conditions of sensuous knowing, is 180 KANT'S SOLUTION CONSIDERED sufficient, according to him, to free us at once from the otherwise insuperable difficulty of what he regards as the manifest absurdity of three infinites — space, time, and God. Other deliverance, he holds, there is none. Such is the problem which Kant raises, and such the solution which he gives to it. Now ought he to have presented the problem in that form ? Surely not. He required not merely to assume but to show that it was a real or rational prob- lem. Belief in more infinites than one may be ab- surd, but it is plainly not self -evidently absurd. On the contrary, it is an indubitable fact that the human mind cannot but think of space as unbounded and of time as without beginning or end. That being the case, what is manifestly irrational is to regard them, prior to proof, as inconsistent with each other, or in- consistent with the existence of an Infinite Creative Intelligence. Kant gave no proof; nor has any one else. Grant, however, the rationality of his problem, the reality of his so-called " insuperable difficulty," and consider only his proffered solution. Is it not a mere evasion ? The absurdity, if there be any, which is alleged to constitute the difficulty, lies in our think- ing, and in our being so constituted as to be unable not to think, two or more infinites. From that ab- surdity, however, if it be an absurdity, we can only be freed by being freed from the necessity of thinking those infinites. To make out the objective unreality of the infinites in themselves does not remove, does not diminish, the subjective self-contradiction involved 181 AGNOSTICISM OP HUME -AND KANT in the thinking of them ; nay, it increases it, inasmuch as it removes the only ground on which we can hope to explain what difficulty there may be in conjointly thinking them — namely, that they are mysterious be- cause real and infinite. If not real and infinite, if simply in us, why should they present in appearance and in thought such a pef plexity ? Kant's famed " solution " is quite illusory. His second position — namely, that space and time are the necessary conditions of phenomena — wards off, he thinks, the scepticism which had been based on the theory that all knowledge comes from experience, and establishes the possibility and validity of mathe- matics. And it might have done so had it not been bound and chained to the position already considered — the dogma that they are conditions only of phenom- ena and necessary only so far as our thinking is con- cerned. Conditions necessary only for us are not truly necessary. The notion of a necessity which does not transcend what is contingent and particular is essentially self-contradictory — the notion of a ne- cessity which is not strictly and universally neces- sary. Scepticism does not deny that space and time are apprehended as necessary conditions of phenom- ena. Even the scepticism based on the theory that all knowledge comes from experience does not deny that ; it merely resolves the apprehension into an illu- sion by the way in which it explains its relation to ex- perience. Hume did not deny that time and space appear to human thought as necessary, but, in con- sistency with his general theory of knowledge, he re- 183 KANT'S SOLUTION ILLUSORY fused to recognise that their necessity could be more than an appearance evoked out of sensations and their derivatives. Kant shows that the apprehension of time and space as necessary is not derivable from ex- perience but presupposed by it, and yet argues that it is only in appearance objectively, and in reality mere- ly subjectively valid. Now, that may be considered by some persons to place scepticism on a less easily refutable basis, but it is certainly not a refutation of it. The difference between the conclusion reached by Kant through his alleged refutation of the scepti- cism which founds on the assumption that all knowl- edge is derived from experience and the conclusion of that scepticism itself is not great ; and, what differ- ence there is, is not in its favour. If our apprehen- sion of space and time as necessary and objective be only derivable from experience there may be some slight chance — a very slight one, I admit — of its be- ing legitimately so derivable; but if that apprehen- sion, although a primary element of the constitution of the mind, is not to be accepted as guaranteeing that space and time are what we necessarily believe them to be, the legitimacy of the apprehension is hopelessly beyond possibility of proof. Kant did not attempt to give a comprehensive an- swer to the question raised by him : How is mathemat- ical science possible? He gave a powerful impulse to the study of the theory of mathematical knowledge, but made to it no substantial contribution of his own. By mathematics he virtually meant geometry. And the reasoning by which he attempted to prove geom- 183 AGNOSTICISM OF HUME AND KANT etry to be possible o:aly through space being a priori and subjective was inconclusive. It sufficed to show that the mind is endowed with a power of forming geometrical conceptions and drawing geometrical inferences, and also that that power, which in- cludes various energies of intellect, is a priori in the sense of subjective, but no more. It entirely failed to show what specially required to be shown — namely, that the space which is presupposed in all the operations of geometrical definition, construction, and inference, is a priori in the sense of subjective. It proved that the mind, in order to be able to trace the relations of extension, must have its thorough mas- tery over geometrical conceptions through the posses- sion of the power of constructing them ; but it simply ignored the fact which is the real difficulty to the Kantian hypothesis in question — the fact that the power implied in every concept and process of geom- ■ etry assumes space to be not constructed but given, to be not subjective but objective, to be not ideal merely but real. § II. Tbansckndental Logic : (A) Analytic From Transcendental Jisthetic Kant passes to Transcendental Logic, by which he means not what is commonly called Logic, but a science or exposition of the pure and a priori elements to be found in the constitution and use of thought. Transcendental Logic thus understood he divides into Analytic and Dialectic. 184 KANT'S TEANSCBNDENTAL LOGIC As in the Transcendental Esthetic, he had meta- physically criticised the faculty of sense and attempt- ed to explain the possibility of mathematics; in the Transcendental Analytic he examines in the same way the faculty of understanding, and seeks to show how physical science is possible. In the ^Esthetic he had allowed that phenomena, through being posited and co-ordinated in time and space, — the mental forms of sensibility, — become knowledge, although only knowledge in its lowest and crudest form, " a chaos of blurred perceptions " ; but denied that they are, properly speaking, objects of thought until also oper- ated on by the understanding and subjected to and synthesised by its forms. In order to be knowledge proper there must, he maintained, be the union of intuitions of sense with notions of the understanding. The sensuous faculty cannot think and the judging faculty cannot perceive. Neither faculty can do the work of the other, and consequently they must com- bine and co-operate in order to produce what may be worthy of the name of knowledge. Kant's first endeavour in the Analytic is to bring to light all the a priori elements which the under- standing imposes on the perceptions of sense in order to make them intelligible. Sense he had treated in the Esthetic as essentially passive. In the Analytic he assumes understanding to be essentially active. The intuitions of sense imply the receptivity of im- pression; the notions of the understanding, on the contrary, imply the spontaneity of thought. All the operations of the understanding are reducible to ele- 185 AGNOSTICISM OF HUME AND KANT mentary acts of judgment, and, consequently, in order to know how many primitive pure notions of the un- derstanding, or categories, as Kant calls them, there are, we only require to know how many species or forms of judgment there are. But this, he thinks, we do know. He accepts as firmly demonstrated and virtually complete the traditional doctrine of judg- ment to be found in all the ordinary text-books of Logic. Accordingly he holds that there are just four chief species of judgments — ^those of quantity, qual- ity, relation, and modality — and that each of these has three kinds of judgment under it, so that there are twelve sub-species, neither more nor less, each having a distinct a priori condition underlying it, and only one such, while all are so connected as to constitute a general system of a priori notions regulative of the understanding within the whole sphere of its opera- tions. Having obtained his so-called categories of the un- derstanding, Kant proceeds to what he found to be the most difficult task he ever undertook, the " tran- scendental deduction " of them, or, in simpler terms, the showing that they must apply to objects, and how, and to what extent. He starts by laying stress on an unquestionably and supremely important fact; on a still higher principle than either the forms of sen- suous intuition or the categories of logical judgment — " the original synthetic unity of apperception — the combining self-conscious activity of a self-identical ego, underlying alike all impressions of sense and all operations of judgment." Without such a centre of 186 ALLEGED NECESSITY OP THE CATEGOKIES convergence and basis of permanence no conjunctions of sense and understanding can be supposed to gener- ate knowledge of any kind or degree. There follows what is represented as " the deduc- tion " itself — the alleged proof that the categories are necessary for the determination of objects, and that only objects obtained through sense — contained in sensuous experience — are determined by the catego- ries. The aim of it is obvious enough, but its success, in my opinion, is nil. It is, even in the second edi- tion, the merest semblance of a " deduction." In- stead of a methodical and orderly process of argu- mentation, there is only a diffuse, lumbering, and pointless repetition in uncouth modes of expression of the doctrine to be proved — namely, that the cate- gories, along with the synthetic unity of appercep- tion, while a priori and possessed of a necessary and universal validity so far as sensuous experience is concerned, have no validity beyond it. On the other hand, what Kant has written in the last sections of the Analytic regarding the " Schema- tism of the Categories," " Axioms of Intuition," " Analogies of Experience," and " Postulates of Em- pirical Thought," is, to say the very least of it, most ingenious and suggestive. Kant's exposition of the theory of knowledge strict- ly so called may be considered as coming to a close with the Analytic, seeing that only sense and under- standing are regarded by him as really and directly faculties of knowledge. The reason dealt with in the Dialectic is not such a faculty. It is represent- 187 AGNOSTICISM OF HUME AND KANT ed as indirectly aiding the acquisition of knowledge, but also as contributing nothing of its own to knowl- edge. At this point, therefore, it may be well for me to look back for a moment at the Kantian theory of knowledge and note as briefly as possible some of the chief points or features of it. It has certain obvious merits. Contrast it with the theory of knowledge which it was meant to dis- place, the theory of Hume, and some of them at least at once sautent aux yeux. For example — (a) As re- gards recognition of the complexity of knowledge, the object to be accounted for, Kant and Hume differ greatly, and the difference is wholly in favour of Kant. Hume's reduction of knowledge to isolated and arbitrarily associated impressions of sense must seem a manifest reductio ad ahsurdum of his analysis to every one who really sees what knowledge includes and involves. Its simplicity is sufficient to condemn it. It is not so with Kant's analysis, which is far more adequate. (&) Kant's theory is also vastly su- perior to Hume's as regards recognition of the spon- taneity involved in knowing. It represents the un- derstanding as essentially self-active, and lays stress on an operation of supreme epistemological impor- tance — " the synthetic unity of apperception." That act Hume ignored. He found no place in the cogni- tive process, or elsewhere in the mind, for self-activ- ity, (c) Another respect in which the Kantian the- ory excels the Humian, and all other exclusively empirical theories of cognition, is its exhibition of one large class of the elements or constituents of knowl- 188 HUME AND KANT COMPAKED edge as characterised by necessity and universality, (d) Again, while Hume sought to show that nature and experience are not consistently interpretable in any terms, both the Esthetic and the Analytic of Kant tend throughout to prove that the world as known to man — the world alike of ordinary knowl- edge and of exact science, mathematical and physical — is one which can only be interpreted in terms of mind J a truth of prime importance in the controversy with empiricism, and with the scepticism based on empiricism, (e) And, further, although both Hume and Kant did even more to advance epistemology by stimulating others to inquiry than by what they them- selves discovered, the suggestiveness of the latter's work was of much the higher and richer kind. Hume, a sceptic by temperament as well as in intellect, with all his extraordinary acuteness, clearness, and subtili- ty, was the very genius of negation, but only that ; he was content to bring all knowledge into suspicion, and yet to rest in his scepticism. He compelled attention to be directed to the most radical doubts and terrible questions, but gave no help as to how the doubts were to be removed and the questions answered. The good David did not feel at all called upon to act as " a guide to the perplexed." Kant, although he so far fell into scepticism, being not a sceptic either by temperament or with intention, was earnestly anxious to overcome it, to answer Hume, and to conquer his own deepest doubts. Hence, although he may justly be reckoned as one of the fathers of modern agnosticism, he may be also as fairly credited with having done much tow- 189 AGNOSTICISM OF HUME AND KANT ards tlie refutation of it. His work was as largely constructive as destructive. His suggestiveness has been, not as Hume's, negative, but positive in char- acter. Scarcely another epistemologist has scattered abroad so many seminal thoughts which have taken root and ripened. Many even of his incidental, or at least undeveloped, observations — as, e.g., several of those to be found in his discussion on the categories (in the Analytic) — ^have exercised an extraordinary influence on the development of modern speculation. There were so many excellent, ingenious, and novel ideas in Kant's theory of knowledge that it most nat- urally excited great interest in the philosophical world, and strongly influenced the course of philo- sophical opinion. There are many even now who deem it, on the whole, a satisfactory doctrine. As I do not share that view, I must briefly indicate my ob- jections to it. 1. Kant, in his attempt to explain the possibility of knowledge, tacitly assumed that he required to have to do only with the intellect and its powers. It was an assumption very natural for a man in his time to make, but it was a mere assumption. In conjunction with the crude view of " faculties " prevalent among the psychologists of Kant's day, it led him to treat his whole subject in an artificial and mechanical fashion. He starts on his investigation without any attempt to determine either what knowledge is or how it has be- come what it is. The deepest roots of knowledge may lie far below so-called intellectual powers ; may be the 190 KANT'S THEORY OP KNOWLEDGE earliest and simplest impulses of sentient and voli- tional consciousness ; nay, must be so if there be any truth in the modern doctrine of psychological evolu- tion. In man, as in all earthly beings, learning to know has been chiefly the result of requiring to act. No knowledge of any kind is the product or the prop- erty of any " faculty," or group of faculties, or de- partment of mind. The minimum in knowledge is a self with an object or objects in relation to it. All human knowledge and all growth in knowledge are only possible and intelligible where there are along with objects entire minds, true selves, directed to them, acting on them, and influenced by them. Kant in the conduct of his investigation proceeds on lines quite incompatible with that truth. He isolates in- telligence from mind as a whole, takes account only of theoretical thought on the radically erroneous as- sumption of its being essentially distinct from practi- cal reason, cuts off " the sensuous faculty " from self or mind, and separates it sharply even from " the un- derstanding," the other facvilty of knowledge. In all these respects he seems to me to have been at fault. 2. After detaching and isolating sense in the way described, Kant nevertheless represents it as supply- ing some sort of knowledge, and as even furnishing the whole matter or content of knowledge. Professor Ferrier has so very effectively shown how serious an error it is to regard sense as capable of itself yielding any sort of intelligible data to the mind, and how much depends on making it apparent that matter 191 AGNOSTICISM OF HUME AND KANT per se is contradictory and sensuous perceptions per se nonsensical, that I content myself with a reference to what he has written/ The assumption that the whole matter or content of knowledge originates in sense- perceptions is an error quite as great. The matter or content of experience, no careful study of conscious- ness can fail to inform us, comes to a far greater ex- tent from within than from without. The assump- tion to the contrary is a rashly adopted metaphysi- cal illusion, not an ascertained psychological truth. Kant's acceptance of it made it logically impossible for him to escape from a phenomenalism practically as agnostic as the scepticism of Hume by the intro- duction of any elaborating machinery of forms, cate- gories, and ideals. That he, nevertheless, combated agnostic phenomenalism with ingenuity and profun- dity is also a fact, and one which non-agnostics will gladly acknowledge; but even a happy inconsistency is an inconsistency, and every inconsistency is a weakness. The hypothesis of the " Ding an sich " is itself so nebulous and ambiguous as rather to in- crease than remove or lessen the self-contradictoriness of the general theory. 3. I have previously indicated why I regard Kant's account of " the forms of the sensory " — space and time — as largely erroneous ; his opinion that he either removed the metaphysical difficulties connect- ed with them by arguing that they are onl'y necessary conditions of phenomena, or warded off scepticism by maintaining that they are necessary conditions, as ' Institutes of Metaphysics, pp. 276-282. 193 SENSE AND UNDEKSTANDING not well founded ; and what he called his " Critical Idealism," as far from exempt from the faults which he himself charged on other forms of idealism. Here I would add that the only reason which he has given for regarding the whole matter or content of expe- rience as derived from impressions of sense — namely, that the categories of the understanding are only ap- plicable to the objects of which we gain experience through sense — is one which is not substantiated by any evidence. In reality, the categories are just as applicable to internal states as to external phenom- ena. Mind, in all its phases and processes, so far as these are consciously realised, is not less capable of being thought in the pure immediate cognitions of re- lation which Kant terms categories than Matter and its phenomena. Mind per se, in the Kantian sense, of course cannot, but neither can Matter per se. 4. Objection must be taken both to Kant's mode of separating and of connecting sense and understand- ing. There can no more be perception without the categories of the understanding than without the forms of sense. The former are not merely superim- posed on perceptions in order to transform them into notions; they are implied in their existence and even in their very possibility as perceptions. The so-called forms themselves presuppose the so-called categories. Space cannot be apprehended or thought of as other than quantitative, relative (to its o"\vn parts or to other things), real, and necessary; in other words, ex- cept as presupposing the categories which it is repre- sented as preceding and conditioning. Kant's sepa- 193 AGNOSTICISM OF HITMB AKD KANT ration of sense and understanding, and of the forms of the one and the categories of the other, is psycho- logically unnatural and exaggerated. It is of a one- sidedness and rigidity altogether mechanical. His way of connecting them is equally mechanical, equally of a kind inappropriate to spirit. Hegel sarcastical- ly, yet correctly, described it as " such an external and superiicial union as \rhen a piece of wood and a leg are bound together by a cord." 5. According to Kant sense is essentially passive and understanding essentially active. In thinking so he was, I believe, mistaken. Wholly passive eyes, ears, and iinger-tips, if they see, hear, or feel at all, assuredly see, hear, and feel very little. In order to be media of information the senses must be largely active and operative. The tmderstanding, on the oth- er hand, is not essentially self-active. It must more or less passively receive its matter or content, and be acted on thereby. It is the self alone which is self- active. The understanding is only active in so far as it is actively exercised by the self. Were it es- sentially self-active, however, sense must be so too, inasmuch as every sense-perception includes a judg- ment, an act of understanding. 6. Kant's identification of the understanding with judgment has been allowed to pass almost (not en- tirely) uncriticised. His distribution of judgments into analytic and synthetic, on the other hand, has been much controverted, especially during recent years, with the result that the logic of judgment is far from the point at which it was when Kant wrote, 194 THE UNDERSTANDING AND JUDGMENT without his doctrine of knowledge being, perhaps, greatly affected. He was certainly not happy in his choice of instances of synthetic judgments. Philo- sophical speculation was immensely and beneficially influenced by his doctrine of the categories. But that it was far from being a satisfactory doctrine is now al- most universally recognised. The procedure by which the categories were obtained was perfunctory; and the enumeration, classification, and correlation of them are all liable to obvious objections. Any real " deduction " of them was manifestly impossible if they really were the primary and ultimate modes of judgment which Kant represented them to be. The conclusion of the so-called " deduction " which is given is a conclusion of the very kind which Kant la- bours to prove must be of its very nature unprovable ; a metaphysical conclusion such as he professes to show lies beyond the reach of all possible knowledge. 7. The reason which Kant gave for concluding that the categories must be applicable to the phenomena of sense — namely, that otherwise there would be no orderly, definite, universal experience, and conse- quently no intelligibility in experience — was an ob- vious petitio princi'pii, when employed as the basis of an argument against a scepticism like that of Hume, which professed to have logically reached the conclu- sion that there, in reality, was no intelligibility in experience, and no room for ascribing any more to it than such an illusory appearance of order and ob- jectivity as associations of custom and contingency may produce. He failed carefully to discuss the 195 AGNOSTICISM OF HUME AND KANT question, Can pure a priori notions be reasonably sup- posed to operate on, and to transform and elevate into intelligibility, such confused and chaotic matter as sense-impressions derived from no known or knowa- ble what or where ? Must it not be as unwarranted to bring the categories into connection with such per- ceptions as into connection with things-in-themselves ? One of the earliest and acutest of Kant's critics, Solo- mon Maimon, conclusively showed, I think, that here was a fatally weak point in the Kantian theory ; and that, on Kant's own principles, the sphere of knowl- edge should have been limited to mathematics, and all objective validity and intelligibility denied to the contents of sense, seeing that in them there is no ne- cessity, no universality, nor affinity of any kind to the categories of thought. Maimon did not oppose to the criticism of Kant the scepticism which he himself professed. What he did was to maintain that his scepticism was the only true basis of the criticism at Avhich Kant aimed but failed to reach; that not merely tlie forms but also the objects of knowledge must be a priori in us if we are to be en- titled to ascribe to them objective validity, seeing that objects cannot be generated by thought in the empiri- cal as in the mathematical sphere. His argumenta- tion, it appears to me, was incontrovertible. 8. Kant erred in referring all universality and necessity in cognition to an a priori and subjective origin. It was an error which naturally followed from his assumption that the content of knowledge consists wholly of particular and contingent sense- 196 VALUE OF KANT'S LABOURS perceptions. That left him without any other de- fence against the most absolute scepticism than what he could find in the ego alone. Hence he toiled so earnestly to find in the forms and categories of thought the grounds of assurance in a real validity of knowledge and an at least apparent objectivity in things. His labour was certainly not in vain. It showed more convincingly and comprehensively than had ever been done before how much more is implied even in a knowledge of objects of sense than mere sen- sations ; and in that way and to that extent it effected a satisfactory refutation of sensism and of the scepti- cism which depends on it. It was also, however, la- bour which, instead of confirming, virtually dis- proved the assumption on which it proceeded — the empiricist assumption that the subject and object in cognition are not organically one but mechanically distinct. The chief value of Kant's elaborate proc- ess of investigation and argumentation really lies, paradoxical as the statement may appear, in its be- ing a continuous course of self -refutation. The great conclusion to be drawn from it is not the one which was expressed, but one which is throughout suggested to us by it — namely, the truth that knowledge is a process in which subject and object so correspond, re- ciprocate, and harmonise that each is only knovm in and through the other, and in which what Kant called forms of sense and categories of judgment are simply constitutive conditions of intelligence in virtue of which the knowing subject is able directly and truly to apprehend what actually and truly exists in known 197 AGNOSTICISM OF HtTME AND KANT objects. The phenomenalism and representationism of Kant caused him to ignore that truth, but his tre- mendous yet fruitless efforts to vindicate the validity or show the possibility of knowledge without the ac- ceptance of it are, perhaps, more instructive and con- clusive than his advocacy of it would have been. 9. I shall merely add that Kant in his criticism of knowledge should surely have introduced " the syn- thetic unity of apperception " at a much earlier stage than he did. It is not the cope-stone but the comer- stone of a theory of knowledge, being essential to the very existence and conceivability of knowledge; and the theory of knowledge, as of everything else, should begin with what is primary and fundamental. If Kant had paid due regard to the fact that cognition is in no form or stage conceivable otherwise than as a synthetic act of a self -active subject, he would not have started on an inquiry into the possibility and conditions of knowledge by positing iinknowahles, — with which a theory of knowledge can have nothing to do, — and appearances — of what does not appear; — nor would he have separated in the abstract and me- chanical way which he did noumena and phenomena, matter and form, sense and understanding, experience and reason, Jcnowledge and reality, the sensuous and suprasensuous. § III. TjRANscENDEiirTAL LoGic : (5) Dialectic In passing from Transcendental Analytic to Travr scendental Dialed k Kant passes from the second to 198 KAKT'S TKANSOENDENTAL DIALECTIC the third intellectual faculty, from the understanding to the reason^ taken not in the general sense in which Kant sometimes employs it but in its restricted and distinctive sense. In this latter sense reason, accord- ing to Kant, is the faculty which reduces judgment to unity in virtue of its continually striving to rise above the domain of experience, the sphere of sense, to the suprasensuous and unconditioned. As sense manifests itself in perceptions, and understanding in judgments, so does reason in conclusions. As sense has its forms, and understanding its categories, so has reason its ideas. As the perceptions of sense can only be made subjects of intelligence through the activity of the understanding, so can the axioms of the under- standing only be reduced to unity through the opera- tion of the reason. Reason, — the faculty of the unconditioned, the in- finite, the absolute, — has, according to Kant, tJwee ideas; and, just as he had derived the categories of the understanding from the twelve kinds of judg- ment, so he derives the ideas of reason from the three forms of the syllogism, — the categorical, the hypo- thetical, and the disjunctive. The three ideas are the Soul, the World, and God. And on each of them, he holds, there has been built up by the reason a meta- physical system of doctrine erroneously claiming to be a science : on the idea of the absolute unity of the thinking subject, the soul, the so-called science of Ra- tional Psychology ; on the idea of the absolute totality of phenomena, the universe, the so-called science of Eational-. Cosmology; and on the ideal, of absolute 199 AGNOSTICISM OF HUME AND KANT reality, God, the so-called science of Rational The- ology. That reason, taken in its distinctive sense, pos- sesses those ideas, means, according to Kant, that the mind from the very nature of its intellectual constitu- tion necessarily assumes the unity of the soul, the ex- istence of the universe, and the reality of a First Cause. At the same time, he maintains that those assumptions, although necessary assumptions, are merely assumptions, and not to be accepted as pos- itive truths or to have any objective value assigned to them. We necessarily seem, he thinks, to know what reason compels us to believe, and are in- evitably led to credit its conclusions and to ascribe validity to its arguments; but, in reality, we do not know what we necessarily seem to know and cannot but believe, and the conclusions of reason are all, in fact, illusions^ and its arguments are all, to use his own words, " as regards their result, rather to be termed sophisms than syllogisms, although indeed as regards their origin they are very well entitled to the latter name, inasmuch as they are not fictions or accidental products of reason, hut are necessitated hy its very nature, — sophisms not of men, hut of pure reason itself, from which the wisest cannot free him- self." While Kant represents reason — pure reason — as an essentially illusory faculty, he does not admit it to be, as Sir William Hamilton affirms, " an organ of mere delusion." He expressly denies it to be essen- tially delusive, and maintains that it only becomes a 300 ILLUSIONS AND DELUSIONS source of fallacies and deceptions when not confined to its legitimate sphere. He says in express terms " it must be the mere abuse of the ideas of reason which cause them to generate in our minds a decep- tive appearance ; " and often repeats the statement in substance. He distinguishes between illusions and delusions^ and attributes only the former to the natural operation of reason. The illusions of rea- son, he affirms, although they cannot be prevented from arising, can be detected and prevented from im- posing on us. They are, he contends, liJce the illu- sions of sense. The moon near the horizon seems larger than when overhead. This is an illusion of perception which cannot be got rid of but which can be detected, so that it does not mislead or prove our senses to be deceptive and mendacious. 'Not is rea- son, according to Kant, without a legitimate and use- ful function. On the contrary, he holds that its ideas have a valuable regulative purpose. They call forth and urge on empirical inquiry; and although they impel men to search for the undiscoverable, the un- knowable, the energy and the efforts thus elicited greatly contribute to the extension and organising of human knowledge. The general view taken of reason by Kant has now been stated. Is it a rational one, or has Kant justified it? I answer in the negative, and on such grounds as the following : — 1. The very conception of a special faculty for the production of inevitable illusions is a most unnatural and improbable one. Is there any other faculty of 201 AGNOSTICISM OF HUME AND KANT the kind in the world either of beasts or of men ? Is it not so abnormal and absurd a sort of power as to have a strong prima facie evidence against the as- sumption of its existence ? There is no other mental faculty merely of illusions. There is no other fac- ulty of necessary and constitutional illusions. The so-called illusions of sense are casual or easily explic- able, and most unlike those ascribed by Kant to rea- son. It would be a violation of the laws of optics, a continuous and needless miracle, were the moon not to appear larger on the horizon than when overhead. Obviously Kant required to prove that human knowl- edge could only be unified and systematised by an ex- clusively and inevitably illusory faculty; that sense, understanding, and imagination with its idealising power, impelled by curiosity and the wants of prac- tical life, and controlled and directed by enlightened and energetic will, would not have suificed for the purpose. But that he failed to do. Looked at from a teleological point of view, the pure reason of Kant is plainly an anomaly in the universe. According to his own description of it, it is a power which strives to rise above experience and to rest in the uncondi- tioned, or, in other words, one the aim of which is essentially unattainable, the objects of which can never be discovered to correspond to anything real. Now, we know of no power like that in the universe ; wherever we find a natural power we find also a real and appropriate sphere for its display. The exist- ence of any instinctive craving or constitutional ten- dency is itself a guarantee of the existence of due sat- 202 DOCTRINE OF REASON EXAMINED isfaction for it. If so, Kant had, of course, no right to posit or postulate such a reason as that which he called pure reason. 2. The utility of what Kant calls pure reason ig not satisfactorily established by him. Let us grant all that he has said in its favour. Let us grant that it gives greater unity and completeness to our knowl- edge, and let us estimate the advantage of that as high as we reasonably can. Has it not, however, disad- vantages ? Are there not evils which flow naturally and necessarily from its operation? Yes, and on Kant's own showing, those disadvantages and evils are numerous and enormous. They comprehend all sorts of superstitions and aberrations, all false relig- ions and all false philosophies. Can the good ascribed to pure reason be fairly held to counterbalance or even to equal such a mass of evil ? 3. There are serious intrinsic defects in Kant's doctrine of reason which take away from its credibil- ity: (a) Tor example, it is only as a faculty, not in- deed of delusion, but of illusion, that pure reason is, even according to Kant's account, of any use. Its influence within what he calls its legitimate sphere is due entirely to its operation within what he calls its illegitimate sphere. It is in virtue of the assumption that it possesses what it does not possess, the princi-. pies of a knowledge of the unconditioned, that it per- forms the work on the conditioned which is alone of value, (b) Again, the very existence of pure reason as a faculty depends, according to Kant's view of it, on the illusions which it entertains. Remove them 303 AGNOSTICISM OF HUME AND KANT and you destroy the reason itself. Thoroughly con- vince a man that he can know only the conditioned, which is all according to Kant that he really can know, and reason must vanish along with the illusion of the unconditioned. From that time onward sense and understanding must be the only cognitive powers of that man's mind, (c) And further, what Kant speaks of as the illegitimate sphere of reason is, in reality, and also according to his own account, its only proper sphere. Reason in its distinctive sense as described by himself is only reason when it oper- ates on and with ideas of the unconditioned, or, in other words, with ideas which are illusions and out of which only illusions can be evolved. Its ideas pro- mote the cause of truth only by calling forth efforts of intellect which serve to systematise and develop knowledge; but the sphere of such efforts is plainly not, as Kant says, the legitimate sphere of reason, seeing that, although they have been called forth by illusions as to the unconditioned, they must be kept free .from all such illusions in order to be successful. They must be guided entirely by principles of the conditioned if they are to help to the apprehension of truth. To associate an idea of the reason with expe- rience is, according to Kant's o"wti teaching, to corrupt and destroy knowledge. In other words, the sphere "v^hich Kant is forced to assign to reason as its legiti- mate sphere of action because there is no other credit- able one to which to assign it, belongs wholly to the understanding, and for reason there is reserved only the sphere of illusions. Siich a doctrine of reason 304 KANT'S PURE REASON refutes itself by its inconsistencies, its self-contradic- tions. 4. The so-called pure reason of Kant is a quite imaginary faculty. The human mind has no such faculty. It was the great illusion of Kant to sup- pose that it had. Eeason is the faculty of all intui- tion proper, or of all that is necessary and universal either in perceptive or intellectual cognition. It has no such ideas exclusively inherent in it, however, as the soul, the world, and God. These are the three fundamental objects of thought, the three great reali- ties to which all human knowledge is related. • They are not properly speaking either mere ideas or mere ideals. We may, indeed, speak of God as the idea of ideas, the ideal of ideals, but only intelligently when we then also think of Him as the ens realissi- mum, the source of all existence and energy, truth and goodness. As mere ideas the soul, world, and God are empty notions. Individuals may have fan- cied that they bad one or other or all of these so-called ideas through a transcendent act of a special faculty apart from all experience, but the fancies of a few confused metaphysicians should not be charged upon the reason itself. The so-called ideas of the so-called pure reason of Kant are none of them original ele- ments or first principles of the reason which is really a faculty of the human mind. They are none of them attained independently of experience but in and through experience. The vast majority of psycholo- gists, cosmologists, and theologians have been under no such delusion as that they could raise sciences on 205 AGNOSTICISM OF HUME AND KANT their mere intuitive ideas of the soul, the world, and God. Owing to Kant having conceived of pure or specu- lative reason in the way which he did, his criticism of it is not really a criticism of human reason or of any truly reasonable kind of philosophy, but mainly of Wolfian rationalism, and unfortunately, also a criticism which proceeds to a large extent on the erro- neous principles of that form of rationalism. As I have already said, Kant assigns to reason three ideas and represents it as raising up on each of them a pretended science, — on the idea of the soul Rational Psychology, on the idea of the world Rational Cos- mology, and on the idea of God Rational Theology. He further maintains that in the erection of these speculative structures reason employs as many kinds of inherently vicious arguments as it has ideas' — namely, paralogisms which relate to the psychological idea, antinomies which relate to the cosmological idea, and ideals which relate to the theological idea. To exhibit and expose these paralogisms, antinomies, and ideals, and to destroy the doctrines or systems with which they are associated, is the task which he endeavours to accomplish in his Transcendental Dia- lectic. I. Rational Psychology. Kant undertakes first to show the futility of the inferences as to the nature of the soul which have been drawn from the charac- teristics of consciousness. He finds them all to be -vitiated by confounding a merely logical subject with ^06 KANT'S RATIONAL PSYCHOLOGY a real thing, and so to be paralogisms, or unconscious sophisms. The root of all mental action is represented by him as being the conscious judgment I think. This I think is the expression of pure consciousness, its pri- mary form, and it unifies and renders possible all experience. Wherever thought or consciousness is, there is immediately felt to be an 1 which is (1) a determining subject, (2) simple, (3) self -identical amidst all the variety of mental states, and (4) dis- tinct from all objects external to itself. It is on this basis that Kational Psychology is raised, and the way in which it is raised is, according to Kant, the conversion of the characteristics of the act through which consciousness is realised into ontological predi- cates, and so making of the mere feeling or concep- tion of conscious unity a real soul, a substance, a sim- ple substance, a spiritual substance, an indestructible or immortal substance, and the like, all of which dog- mas are due to transforming the determinations of a merely phenomenal subject into the properties of a transcendental object, and bring the same terms self or soul to denote two entirely distinct entities, a sub- jective and an objective, a logical and a real, ego. Hence all the alleged proofs of a soul or spirit include a quaternio terminorum. In fact, the existence of a soul or spirit, self or ego, distinct from the body or more than a feeling of the unity of consciousness, cannot possibly be either proved or disproved. The criticism of Kant in this portion of his " Transcendental Dialectic " was not without consid- ao7 AGNOSTICISM OF HUME AND KANT erable relevancy against a sort of psychology preva- lent in Germany when he wrote, although even then falling into discredit. It was so far effective against the " Eational Psychology " of Wolf and his follow- ers. And yet it proceeded throughout on the position of a dualism akin to, and at least as irrational as, the dualism which was Wolf's radical error. It assumed two " egos," one a logical subject and the other a real thing, and two " psychologies," one rational and the other empirical, one a pseudo-science and the other a true science. That, however, was an assumption which begged everything, and which Kant had no right to make. It has received no confirmation from himself or from others. His subjective " ego," his merely logical subject, is a figment of abstraction to which nothing corresponds, and his unknowable on- tological " ego " is another of the same. There is in man but a single ego, and that ego is neither of those imagined by Kant, but a real, living, self -perceptive, and self -active agent. It is the ego alike of conscious experience and of true psychological science. There are not two " psychologies." Wolf and Kant both erred in supposing that there were. The Kantian form of the error is no improvement on the Wolfian. Unsatisfactory as may be the notion of a really " ra- tional psychology " distinct from empirical psychol- ogy, the notion of a necessarily illusory one distinct from it must surely be as unsatisfactory. Both notions are, however, forms of the same error, and have been arguments have precisely the same defects as those supported by the same kind of reasoning. Kant's 208 KANT'S TREATMENT OF SENSE which he condemned — namely, a word used in a double sense, and in each syllogism a quatemio termi- norum. The difference between them and those he assailed lay not in their character but in their appli- cation. His criticism should have been applied not only to Wolfian rationalism in regard to the soul but to his own critical doctrine also. Or, as Hegel puts it, " he fell into contradiction from the barbarity of the conceptions which he refutes, and the barbarity of those which he retains from among those that are refuted." He adhered, however, only too consistently to the most fatally erroneous of his principles, denial to the mind of true perceptive power or immediate appre- hension of reality. He began his " Critique " by treating sense as a mere receptacle of impressions, needing to be somehow organised and objectified by mental forms and categories, but the causes of which, if they have causes, are not causes or objects really perceived, but, on the contrary, imperceptible and unknowable. Having begun by thus misrepresenting the testimony of consciousness and sense-perception, it was just what was to be expected that he would treat the testimony of self-consciougness and intro- spection in the same way. Had he allowed that his phenomenal ego was apprehended as an actual self, as more than a feeling or conception to which no real self corresponds, he would have been manifestly in- consistent. And to what could it on his view corre- spond ? Certainly not to his transcendental ego, for that, if there be any such thing, must be a " thing- 309 AGNOSTICISM OF HUME AND KANT in-itself," and unknowable. Such inconsistency he cannot be charged with. He concedes to what we call self-consciousness no intuition of self or genuine knowledge of self as a reality. He represents the ego as being for consciousness a mere feeling of activ- ity ultimately referring to a conception of unity. There is nothing more in it, according to his view, than the feeling and conception. But he thereby re- jected the testimony of consciousness at its clearest, and logically involved himseK in a scepticism the most absolute, from which he had to try to deliver himself by faith, a faith irreconcilable with his crit- ical philosophy. The truth of truths as to knowledge is the one so ingeniously and eloquently expounded and applied by Ferrier in his Institutes of Metaphys- ics, — the truth that, along with whatever is known, self or the ego is also and necessarily known. This knowledge of self inseparable from all knowledge whatsoever, the condition of all human experience, the source of psychology, and the corner-stone of epis- temology, is a knowledge at once real and relative. Kant went so far as to try to show that being always relative to objects it could not be itself knowledge of reality, but his attempt was feeble in the extreme, and fairly deserved Hegel's sarcasm that it amounted to maintaining that a man cannot know himself be- cause he cannot take his self, his ego, in his hands, and see it, and smell it. II. Rational Cosmology. In this section of Tran- scendental Dialectic, Kant endeavours to show that reason founding on the idea of the universe — i.e., of 210 KANT'S KATIONAL COSMOLOGY the absolute totality of phenomena — and seeking to comprehend the world as unconditioned, necessarily builds up a pseudo-science like that rested on the idea of the soul. In the latter, reason had been shown, he held, to involve itself in paralogisms. When occu- pied speculatively with the universe it leads, he ar- gues, to antinomies or dilemmas which can neither be evaded nor dogmatically solved. Of these there are neither more nor less than four, because there are just so many leading categories of the understanding, and each of them is extensible beyond experience. Hence, in quantity the world is either bounded by a limit in time and space or unbounded ; in quality it is either ultimately simple or infinitely divisible ; in relation it is either caused by free activity or made up of an infinite series of mechanical causes ; and in modality it has either an independent cause or is com- posed of mutually dependent members. These are Kant's famous antinomies. And he not only main- tains that the antithetic positions may in each case be equally demonstrated, but shows us how it may be done. In other words, he gives us what he professes to be equally good and complete demonstrations for holding both that the world had and had not a begin- ning in time; both that it is ultimately simple and infinitely divisible; both that it is produced by free agency and by an infinite series of necessitated ante- cedents; and both subject to and exempt from the condition of causality. Were he correct, reason would seem to be left in a very desperate plight. If all its attempts to under- 311 AGNOSTICISM OP HUME AND KANT stand the universe result in contradictory yet equally well-established conclusions, it would itself appear to be essentially self-contradictory and doomed to absolute scepticism. But Kant thought otherwise. He professed, and quite sincerely, to be no sceptic. So, as in duty bound, he undertook to show to reason a way of escape and deliverance, or, in other words, to solve its dilemmas and dispel its antinomies. And he was right in so proceeding, instead of accepting, as Sir William Hamilton, Mansel, and others have done, the antinomies as expressions of a fundamental law of thought, and making no attempt to solve them, but letting faith decide for the term which pleases it, although there is as much reason in favour of the other. Kant was neither so naive nor so arbitrary as that. He professed not only to demonstrate but to solve the antinomies, and to solve them critically, i.e., in accordance with the principles of his own theoreti- cal philosophy. What is his so-called critical solution? It is an application of his distinction between phenomena and things-in-themselves. The antinomies of reason, he affirms, necessarily arise from our inveterate habit of confounding our own laws of thought with inde- pendent existence. If things-in-themselves be sub- jected to the dilemmas raised by reason, absolute scepticism is inevitable. But, according to Kant, it is just in thinking so that our error lies. What is true of the empirical or phenomenal world may not apply to the transcendental or intelligible (noumenal) world. A totality in oiir conceptions is not to 21% ANTINOMIES OP REASON be identified with a totality of things-in-themselves. The objects which we know in experience are not things-in-themselves, but exist for us only as they ap- pear to us in experience. Hence we have no right to affirm anything of them in themselves ; no right, for I example, to affirm either that the world is in itself ' finitely or infinitely extended, for the world in itself is not a world that we know, and the world we know is one which exists only in an experience that is al- ways extending but never completed, so that we can neither pronounce the extension of it finite nor infi- nite. If, however, we thus distinguish between the worlds of existence and of experience, of noumena and of phenomena, and recognise that what is true of the latter need not be true of the former, and that, indeed, thought ceases to be valid beyond experience and phe- nomena, we may fairly hold ourselves entitled to reject both the theses and the antitheses of the first two antinomies, and to accept the theses of the two last as true of the noumenal world and the antitheses as true of the phenomenal world. Such is Kant's solution of the antinomies of " Ra- tional Psychology." Now, any solution is, as I have said, better than none. And Kant's must be admit- ted to be ingenious, and also to have actually proved fruitfully suggestive. Regarded from a logical point of view, however, it is thoroughly futile. There is only one world or universe — the phenomenal world — the universe of real or possible experience. All Kant's antinomies relate only to it. The question whether the so-called transcendental or intelligible 213 AGNOSTICISM OF HUME AND KANT world, the Kantian " world-in-itself," is finitely or infinitely extended, is a question which cannot be in- telligently put or sanely answered in any definite way. Nothing, not even existence, can be attributed to an absolutely unknown and unknowable world. Kant, seeing that he lectured and wrote regarding it, should have had some real thought of it and belief in it; but it is quite certain he had none, nor has any other person had, notwithstanding all that has been talked and printed concerning it. Inasmuch as all knowing of anything is the knowing of it not out of hut in relation to a knowing self, no finite or even in- finite intelligence can be supposed to know any " thing-in-itself." Indeed, a " thing-in-itself " is not only what no intelligence can know but what no intelligence can be ignorant of, for, as Terrier has well shown, " we can be ignorant only of what can possibly be known ; in other words, there can be an ignorance only of that of which there can be a knowl- edge." ^ A " thing " or " world " " in-itself " is as utterly nonsensical as a whole which is smaller than any of its parts. An infinite intelligence seeing the universe through and through could no more have a glimpse of the Kantian so-called " intelligible world " than the dullest human or even animal intelligence. To refer us in any way to such a world as a key to the solution of the antinomies of reason is certainly not to give us any help. The merit of Kant as regards what are termed an- tinomies of reason lay not in resolving them but in ' Institutes of Metaphysics, 404-408. 314 THE ANTINOMIES EXAMINED calling, or rather recalling, attention to them. His whole treatment of them was hazy and superficial. I-Ie dogmatically assigned them to a particular source instead of critically inquiring what their source was. He did not discuss the important questions as to whether or not, or to what extent, they are natural and necessary perplexities of reason itself or the arti- ficial puzzles of a misapplied metaphysical ingenuity. He arbitrarily assumed them to be confined to a spe- cial faculty and a particular pseudo-science, although they are also to be found in logic, mathematics, dy- namics, ethics, &c. He made no general survey of them ; nor did he show that in any instance both terms of an antinomy really appear to have equal claims to acceptance where a synthesis is not only possible but logically demanded. All the antinomies specified by Kant are pervaded by one fundamental antithesis of which only one term is reasonable or provable. One or other or both of his " demonstrations " will be found to be illogical. How he could have fancied, for example, the verbiage attached to the second prop- osition of his first antinomy a " demonstration " is most amazing. Then, the first proposition of his sec- ond antinomy is a truism, while the second is a para- dox, and the whole antinomy is so feebly and confus- edly dealt with that it is difficult to make out what was aimed at. There is no real contradiction involved in the third antinomy, and no proof whatever is given that there is not room in the universe for both free- dom and necessity. In the fourth antinomy the anti- thetic proposition is absurd, and all that Kant says in 216 AGJSrOSTICISM OP HUME AND KANT support of it merely tends to show that the idea of Necessary Being is not distinct and definite like a perception of sense or a mathematical figure, which is, of course, irrelevant to what he required to prove. In short, he has in no wise made out that reason in theorising on the universe necessarily falls into self- contradiction, and has made it apparent that his be- lief in its self-contradictoriness arose largely from the irrational separation of phenomenal and noumenal by which he pretended to solve the imaginary contra- dictions which he ascribed to it. His attempt to do so has been well characterised by Wundt (Log. ii. 376) as a " ScheingefecM." III. Rational Theology. This is, according to Kant, the pseudo-science based on the third idea of pure reason, the highest of its ideas, and therefore, in order to distinguish it from the other ideas, often called by him an ideal. It is the idea or ideal of the totality of possibility, of reality, and of perfection, in- clusive of individuality and personality, or, in a word, the idea or ideal of God. It originates, he holds, so far as it can be traced to the pure or speculative reason, and, indeed, so far as it can be traced to intelligence at' all, in the form of the disjunctive syllogism, — a form which implies the determinability of a thing to the totality of all possible predicates. To know any- thing completely it is necessary for reason to have the idea of the whole of possible, real, and perfect being, and to determine the thing thereby negatively or positively. Further, reason cannot content itself with entertaining the idea as a merely regulative 316 RATIONAL THEOLOGY principle of thought, but must go on to objectify, hy- postatise, and personify it, and to build up on it a system of dogmas, although it is a mere subjective conception without any real basis or content. Hence belief in the existence of an Absolute and Perfect Be- ing, — faith in Godjy'so far as founded on reason, is a dialectical illusion. Such is Kant's view of the rational idea or ideal of God. It is, unquestionably, an ingenious one. I am not aware that any person before him had the thought of tracing the belief in God to the form of the dis- junctive syllogism as its source. Only a very subtle and speculative individual, of a decidedly scholastic turn of mind, would have dreamed of so curious and abstruse an explanation of a universal belief which probably no man who entertained it had ever before rested on the ground indicated. The view was, fur- ther, as consistent as it was ingenious. It was just the view which Kant's system, and especially his the- ory of the faculties and functions of cognition, de- manded. If his plan of the speculative reason were correct, the origin of the belief in God^could not be found among the forms of sense or the categories of the understanding, but only in the operations of the ideas of reason ; not in the region of perceptions or of judgments but only of syllogisms. He found it where he was logically bound to find it. Ingenuity and consistency, however, are the only merits which we can justly attiribute to his hypothe- sis, and even they seem to deselrve but slight admira- tion. The ingenuity should qf itself suggest suspi- 217 AGNOSTICISM OF HUME AND KANT eion. A belief like the belief in God cannot have had the extremely artificial origin which Kant assigns to it. Its real source must be sought for in the reasons which have actually given rise to it in the conscious experience of the race and made its history what we know it to have been. Among such reasons the one alleged by Kant had certainly no place. Unless un- duly influenced by scholastic habits of thought, he could not have assigned to the disjunctive syllogism the part which he did in the origination of the idea of God. His consistency was maintained by the sacri- fice of naturalness and truth. The point of view, theh, from which he criticised " Rational Theology " was highly dubious. His conceptions, I must add, of the nature of. what are called the proofs for the Divine existence were very defective. He regarded them not as the indications of real processes of knowing by which religious expe- rience is attained and extended, but as formal syllo- gisms, each of which must determine in itself whether or not a conclusion is true and certain. But that is an erroneous assumption, a scholastic prejudice which has found its contradiction in the whole history of modern science. Any criticism of the theistie proofs founded on so inaccurate a conception of their character cannot fail, however acute and subtle it may be, to be in the main inconclusive. Were the proofs of positive and inductive science • exhibited and crit- icised in the same abstract and artificial manner as were the proofs of liatural or Rational Theology by Kant, they would fare just as badly. They actually 218 THE ONTOLOGIOAL AEGtJMENT were so criticised in the Middle Ages, and the result was that there was almost no positive or inductive sci- ence in those ages. The world and man were most superficially known because most unwisely studied. Were geologists, biologists, or psychologists required to set forth the proofs of their conclusions in formal syllogistic processes they must abandon their occupa- tions. Reason reaches a knowledge of God in essen- tially the same way as it requires a knowledge of the other great ultimate realities. No object is known to us otherwise than through acquaintance with its qualities or attributes, its powers and manifestations. The dialectical illusion which, according to Kant, originates in the form of the disjunctive syllogism, he represents as requiring to support itself by three arguments — called respectively the Ontological, the Cosmological, and the Physico-Theological — the only three, he maintains, that can be employed by specu- lative reason to prove the existence of God. To the refutation of these arguments he devotes a very inter- esting and important portion of his " Transcendental Dialectic." The objections urged in it against the theistie proofs, although in no instance original, are well selected and well presented. They are the strong- est of the objections that have been urged against the three proofs, which are alone subjected to scrutiny, l^one of them are devoid of a considerable measure of plausibility or relevancy. They are stated clearly, effectively, and in an order which had at least the merit of being the one most suited to attain the end Kant had in view. And, further, they are essential 219 AGNOSTICISM OF HUME AND KANT portions of one of the greatest of philosophical sys- tems, and occupy in it a distinct and highly important position. It is, therefore, perhaps not surprising that during the period when the influence of Kant was at its height his criticism of the theistic proofs should have been widely regarded as decisive. That time has now largely passed away, and those who still believe so may not uncharitably be regarded as belated thinkers or very uncritical criticists. Yet much even of contemporary agnosticism, both of a popular and of a so-called scientific kind, rests largely on an un- tested assumption of the validity of Kant's criticism of Rational Theology; and hence to indicate the grounds on which the assumption may be questioned is still by no means superfluous. 1. The ontological argument. — (a) Kant treats as such what is only one of a class of the so-called a priori theistic proofs. Hence his refutation of it, even if successful, would not be a disproof of a priori, or even of ontological theistic argumentation in gen- eral. It has no reference at all even to the class of a priori proofs best entitled to be called ontological, inasmuch as they start not from the affirmation of an idea but from the affirmation of existence — the affir- mation that something (anything) is, and that that of itself implies that nothing never was, and eternal and necessary being has ever been. It is to be regret- ted that Kant wholly ignored such arguments, as re- flection on them might have at least led him so far into " the Parmenidean way of truth " as to meditate on the significance of It is, until he recognised the 230 THE ONTOLOGICAL AEGUMENT EXAMINED defectiveness of his conceptions of existential judg- ments , and the rashness of asserting that is "is al- ways merely the copula of a judgment." (b) The so-called ontological argument discussed — one of the Cartesian forms of a priori theistic proof — is taken up hy Kant first on the ground that, al- though it has appeared much later than the other two arguments to be examined, it^is presupposed by them. That ground, however, is a false assumption. The argument in question is not first in the natural order of the theistic proofs. Those proofs represent stages of a process of which the last is the apprehen- sion of God as the all-perfect Being. Hegel saw this, and properly placed the ontological argument last, al- though he erroneously treated the other arguments as merely untrue forms of the ontological, which, there- fore, had to establish the truth of the entire thought of God. The argument criticised by Kant proceeds from the idea of an all-perfect Being. But to arrive at such an idea, the elements of it, the perfections in- cluded in it — power, wisdom, goodness, righteousness — must surely have been cognised or believed in as attributes of the Divine. And they could only be cognised or believed in by some such modes of appre- hension or inference as are designated the cosmologi- cal, physico-theological, or moral proofs. To the ex- tent that God is known in any of these ways He is known as existing. All theistic proofs are proofs of God's existence. There is no more need to begin in theology with an ontological proof, merely to prove the existence of God, than there is need to commence 221 AGNOSTICISM OF HUME AND KANT the study of geology or botany with an ontological proof of the existence of stones or plants. It admira- bly suited Kant's purpose, however, to begin as he did. If he could make it appear at the outset that ideas of their very nature cannot imply existence, he Avould only require to affirm it all through to the end of his argumentation in order to save himself much logical labour. Certain it is that his whole criticism is an attempt to cut the connection between thought and existence at the point where it seems to be thin- nest. (c) I do not admit that he^succeeded in his attempt. On the contrary, his criticism of the ontological argu- ment itself seems to me futile in consequence of his assumption that thought and existence are essentially separate, so that even necessary thinking of a being as necessarily existing is no assurance of its existence. The assumption was a natural consequence of the in- coherent idealism and arbitrary dualism which are the chief defects of his philosophy. Solid founda- tion for it there is none ; no ground for believing that there is any such chasm between thought and exist- ence, reason and reality, as is affirmed. On the con- trary, there can be no genuine thinking which is not a thinking of the existent, no reasonableness except in so far as reason apprehends reality. Mere conceiving is not properly thinking; mere imagining is neither reason nor reasoning. According to Kant's own ex- press teaching, we must necessarily, by the very con- stitution of our reason, not only think of God, but think of Him as necessarily self -existing, as otherwise 222 THE COSMOLOGICAL ARGUMENT we do not think of Him at all. And yet he maintains that the existence of God cannot be inferred from the necessary thought of His necessary existence, seeing that " existence cannot be clawed out of thought." Were that really so, all affirmations of existence would be unwarranted. If the transition to existence from the necessary thought of necessary existence be de- nied, much more must transition from particular per- ceptions to contingent existences be denied to us. If self-contradiction be a law of necessary thinking, all thinking may be self-contradictory, all reasoning ir- rational. The argument of Descartes may be valid or the reverse, but Kant's criticism of it is a suicidal sort of reasoning, an argument for absolute scepti- cism. (d) As to that portion of Kant's criticism of the ontologieal argument which takes for granted the pos- sibility of annulling the subject even of necessary thinking, it, too, implies that necessary thinking may not be necessary and may be unveracious, and does so dogmatically and without evidence. It also, there- fore, has to be regarded as a petitio principii in fa- vour of agnosticism. 2. The cosmological argument is naturally the first in order of the theistic proofs, fh^ Divine has every- where been first recognised as power. Hence the ar- gument made its earliest appearance not with the first man who formulated it but with the first man who, in the presence of natural phenomena, saw in them manifestations which he felt constrained to refer to a supernatural power or powers, to a deity or deities. 323 AGNOSTICISM OF HUME AND KANT It may be as old as human reason itself, and is not in the least likely to be ever separated from it. It has been, however, always so far changing in form, and will doubtless continue so to change, as what it rests on is man's entire knowledge of the world in whatever ways gained, and that is an always widening and varying knovrledge. Kant pronounces the argument " a perfect nest of dialectical assumptions " (ein ganses ISTest von dialek- tischen Anmaassungen) ; but the words are really a good description not of the argument but of his own criticism lof it. His objections proceed almost entire- ly from erroneous assumptions as to causality, neces- sity, and experience. I must be content with a very brief statement of my reasons for thinking so. (a) His first, and perhaps chief objection to the argument is that it illegitimately passes by means of the principle of causality from experience to a " thing-in-itself." Causality, he affirms, cannot take us beyond experience. There is a sense in which that is true, most true. That experience extends just as far as causality and similar principles will legiti- mately take us is a great and precious truth. But that is not what Kant means. He means by expe- rience sensuous experience, and would have us to be- lieve that causality only gives order to sensuous im- pressions but can by no means carry us beyond them. Are there no dialectical assumptions of the most erroneous kind in such a view? Why,4hat concep- tion of experience assumes the truth of sensism, and were it correct, Kant had no shadow of right to rep- 234 KANT'S OBJECTIONS resent causality as even subjectively necessary. In that case the agnosticism of sensism, of Hume, must be well-founded, virhile the agnosticism of criticism, of Kant, can have ho true basis. (6) Kant argues as if the cosmological argument represented thought as proceeding along a series of intermediate causes outside of the universe, each of which is contiijigent, to what it at last through sheer weariness arbitrarily pronounces a first and necessary cause. But he thereby caricatures the real process. There is no warrant for assuming intermediate causes at all. If the universe of physical things and finite minds show no traces of necessary self -existent being, and must therefore have a necessary self -existent cause out of itself, our first step of inference beyond the universe will be also the last. We know nothing, and can reasonably believe nothing, about interme- diate beings between the universe and the first cause, the self -existent being, and hence we have nothing to do with them except to show that we have nothing to do with them, and that is done wherever the argu- ment is properly stated. Eeference to the unthinka- bleness of an infinite regress — the incredibility of an infinite series — of finite causes, is only required to show that the-insertion of imaginary intermediate causes would be irrelevant and ineffective. "^ ]|t is not a direct or constitutive part of the argument ; not em- ployed as Kant's criticism must lead unwary readers to imagine. (c) Another objection taken by Kant to the cosmo- logical argument is that it treats the idea of necessity 335 AGNOSTICISM OF HUME AND KANT as a transcendental object of knowledge. He af- firms that because we are under the necessity of think- ing a necessary cause for the world we conclude that there is such a cause ; and, of course, he tells us that we have no right so to convert a necessity of thought into a necessity of existence. But surely that does not advance the cause he pleads. It is merely a reit- eration of his want of faith in the veracity of neces- sary thought — a want of faith, too, inconsistent with all that is positive in his own doctrine. To ascribe truth to what reason must necessarily think is per- fectly legitimate. To deny to it truth implies not only the critical scepticism which Kant advocates, but the absolute scepticism which he repudiates and pretends to refute. (d) A further charge of Kant against the cosmo- logical argument is that it is only the ontological ar- gument in disguise. It was not unnatural that he should think so, as his view of both arguments was far from clear and definite. Yet the arguments are as distinct as two arguments each representing a stage or moment in the same process can well be. They are connected and explanatory, but neither is the other in disguise. Kant's conception of the second as only a veiled form of the first is transparently erroneous, and cannot be acquiesced in by any one who recognises that the rational transition from the world as a known effect to its cause ip the cosmological argument, the effectuation of the transition its distinctive and even sole function, so that when that function is accom- plished its work is done, and those who wish to know I 336 THE PHYSICO-THEOLOGICAL ARGUMENT whether the cause of the world is more than a First Cause, a self-acjtive, all-productive Power, must con- template the world from other standpoints than a merely cosmol6gical, an exclusively setiological one. Indeed, confused as Kant's own criticism is, it dis- proves what it claims to prove, inasmuch as it repre- sents the cosmological argument to have a function distinct from the ontological, although affirming, in- deed, that the basis of the ontological argument is em- ployed in the support of the cosmological argument. Even on his own showing the distinction of the two arguments is as evident as their relationship, and the second is not the first in another name and another garb. \ 3. The physico-theological argument. In this ar- gument the inference is from the evidences of order and purpose in the world to a divine intelligence. Kant pronounces on it a fine and celebrated eulogium. ISTevertheless he declares it to be logically unsatisfac- tory, and urges certain objections against it, which had, however, been far more skilfully and effectively urged by Hume. Presented as they are in all their nakedness by Kant, they ought not to mislead any in- dependent and wakeful mind. (a) The first is that the idea of finality=or design on:which the argument proceeds is of subjective ori- gin, and consequently, like that of cause, invalid when transferred from the experiential to the supra- sensible, or, in other words, when applied to a tran- scendental object. Here again, however, there is ir- relevancy, due to cohfusion in Kant's own thinking. 337 AGNOSTICISM OF HUME AND KANT The_argiiment does not imply that the idea of finality, as revealed in the universe, carries the mind to a Kantian transcendental object or Kantian thing-in- itself. None of the constitutive ideas of reason do that. The act of spiritual apprehension dependent on the principle of finality merely raises the mind to the intelligence which has displayed itself in the order and adaptations of the universe ; merely brings the human mind into contact and communion with the Divine Mind, so as to enlarge and elevate its expe- rience, while not taking it beyond experience or en- abling it to know the unknowable. The principle of finality always carries us, indeed, beyond the phe- nomena of sense to intelligence. Intelligence in our fellow-men is no more a phenomenon of sense than intelligence in God. That finality is of subjective origin in any way which implies that it has no ob- jective application is an untenable hypothesis — one which would require us to disbelieve not only in the existence of the Divine Mind, but of all minds. (fe) The second objection of Kant to the physico- theological argument, as he calls it, but which is now generally and more appropriately designated the te- leological argument, is that it cannot at the utmost prove more than the existence of a world-biiilder of power and wisdom proportioned to the amount of or- der and adaptation displayed in the world; that it leads not to the idea of a creator who originates his materials, and has absolute power over them, but merely to that of an architect whose materials are given him and who shapes and combines them as 238 THE TELEOLOGICAL AEGUMENT bestjie can. But before making that objection, he oiight manifestly to have taken into due account that the argument of which he was treating, the teleologi- cal argument, presupposed of its very nature the cos- mological or setiologicjal argument, the express and sole purpose of which is to trace all the power and efficiency in the universe to an extra^mundune or pri- mal Will. If he thought it did not accomplish that purpose, he should have objected to it on that account ; but the objection is quite out of place when urged in- stead against the teleologieal argument, the express and sole aim of which is to show that the cause or will which is the source of all the power or efficiency in the universe is also the intelligence or reason which accounts for all the order and harmony therein. To object to the latter argurqent on the assumption that it ought not to be supported on the former, that the arguments have not distinctive yet associated func- tions, is not reasdnable criticism, but a cavilling crit- icism which refuses to judge the parts of a rational process in relation to one another and to the whole. (c) The teleologieal argument, Kant further ob- jects, does not prove the Divine Intelligence to be infinite. That must be so far granted, but it is not much to the point. The questions as to the Divine infinity, absoluteness, personality, &c., do not fall to be discussed at the teleologieal stage of the theistic proofs They become relevant only at a later and higher stage — only when philosophical reason has ascended as high as it can without folly — o nly w hen speculatiye__thought comes to deal with the idea of 329 AGNOSTICISM OF HUME AND KANT God as an organic and harmoniovis whole. The teleo- logical argument suffices, however, to prove Divine omniscience in relation to the universe, or, in other words, that God knows all that is to be known in the universe from its centre to its farthest bounds, and in its history through all the ages of its existence. And, further, as it is of the very nature of intelligence to know itself as well as its objects, a most natural corol- lary from the argument is that God must not only fully know the universe He has made but also Him- self, His own boundless being and blessedness, the whole of His powers and perfections. If, then, the teleological argument does not of itself prove the Di- vine intelligence to be infinite, it certainly gives \is no warrant for supposing it limited. Kant should have stated that. The three theistic arguments criticised by Kant in his Critique of Pure Reason were regarded by him as the only arguments which can be urged on behalf of " Rational Theology " or made use of by " theo- retical reason." His examination of them led him, as we have seen, to merely negative results. The so- called arguments he maintained to be sophisms, the reputed science, an illusion, the knowledge of God un- attainable, and a speculative use of reason in the sphere of religion necessarily illegitimate. Yet he did not infer from those fi-ndings that he must deny the Divine existence or even cease to believe in God. Kant was no atheist. He was a believer in God and a truly religious man ; and in that respect his theolog- ical followers have remained more faithful to his doc- 230 ARE THERE TWO KINDS OP REASON? trine and are more akin to him in spirit than the spe- cially so-called philosophical JSTeo-Kantists. To vindicate^ the^ consistency of his attitude tow- ards religion he had recourse to the distinction be- tween theoretical and practical reason. If we can- not^speculatively prove the existence of God, neither can we^he affirmj^rove His non-existence. Theo- retical reason is no more entitled to decide in favour of His nOn=exi«tence~tlrah.'of His existence. Hence while criticising and, rejecting the theistic proofs, it leaves it possible that a belief in God may reasonably originate, in the practical^ reason. For, according to Kant, there are two sorts of reason — the theoretical and the practical. The former alone gives us knowl- edge, but knowledge only of the phenomena of sense. There is no suprasensible knowledge. There is room, however, for a belief in God as a suprasensible reality — for a postulating of His existence as such — capable of satisfying the requirements of duty, the wants of our nature and life. We are entitled to retain faith, although we must forego knowledge, since knowledge is only of things we see. In short, Kant insists that his criticism is not scepticism, and that it only de- stroys a pretendedly scientific certitude in order to clear the ground for a moral certitude such as is alone attainable within the suprasensuous sphere. Thus while he comes to the conclusion that we cannot possibly know God, he fully admits that we are bound by what he calls practical reason to believe in God. A very few words on those views must here suffice. 1. Kant's division of reason into theoretical and 231 AGNOSTICISM OF HUME AND KANT practical is not to be accepted simply on his authority. It requires to be shown that there are two hinds of reason. That there are not two kinds of reason is a quite tenable thesis. No one will deny indeed that reason may be theoretical and practical, in the sense that it may be directed to the acquisition of knowl- edge and also to the attainment of practical results. But two applications of reason are not two kinds of reason ; they are only reason exercised in two ways. Reason may also be said to have distinct functions — noetic, ethic, and lesthetic — according as it discrim- inates between the true and the false, the right and the wrong, or the beautiful and deformed. It does not follow that there are three reasons or three distinct kinds of reason, but merely that there is one and the same reason conversant with three distinct classes of relations. Kant's belief in two reasons is inseparable from his belief in two separate sorts of objects — noumena and phenomena. The one belief depends on the other. Both beliefs are unwarranted, and have done much mischief in philosophy. 2. Kant, when treating of " Rational Theology," did not include the moral proof among theistic argu- ments. That was only natural in one who separated theoretical and practical reason in the way which he did. But whoever examines an appropriate presen- tation of the moral proof must see that it is just as theoretical as the setiological or teleological proofs. It is an argument from the manifestations of God in moral law and moral order just as they are arguments from two other forms of His self-manifestation to 232 KANT'S VIEWS EXAMINED His power and wisdom. In works published subse- quently to the Critique of Pure Reason Kant has pre- sented a moral argument of his own, and, indeed, in two forms, a simpler and a more elaborate/ But in neither form is it a favourable specimen of its class. Admittedly it does not lead to any real knowledge of God. Kant affirmed, indeed, that although all other arguments for the existence of God are delusive, there is given in conscience (the practical reason) a feeling of responsibility and a sense of freedom which com- pel us to believe in One through whom virtue and fort- une, duty and inclination, will be reconciled, and in whom the will will be free to do all that it ought. At the same time, he was too sagacious not to see that all reasoning to that effect would be met with the re- tort and reproach that the same process by which he pretended to have abolished the other arguments was just' as applicable to his new one ; that the ideas of freedom and responsibility, when appealed to in or- der to assure us of reality, might be as delusive as those of causation and design ; that if the latter were mere forms of human thought the former might, with equal reason, be held to be so likewise, and no less in- capable of affording a warrant for belief in God Him- self ; and consequently that the final religious result of his philosophy was not that there is a God, but merely that there is an idea of God which the human mind cannot get rid of but which it is wholly incapa- ble of justifying or verifying. How did he meet such ' See the Kritik der Urtheilskrafi, §§ 86-90, and Kriiik der Praktischen Vernunft, ii. B., 2 H. v-viii. 233 AGNOSTICISM OP HUME AND KANT antagonistic criticism? He did not meet it at all. He evaded it. His reply amounted simply to reaf- firming that we are under the necessity of associating the idea of a Supreme Being with the moral law, and then qualifying the statement by the admission that we can know, however, nothing about that Being, and, indeed, as soon as we try to know anything about Him, make a speculative instead of a practical use of reason, and so fall back into the realm of sophistry and illusion from which the Critical Philosophy is meant to deliver us. In other words, what he tells us is that the argument is good, but only on the condi- tions that it is not to be subjected to rational scrutiny and that no attempt is to be made to determine what its conclusion signifies. On those conditions might he not have found any argument good? Are such conditions not inconsistent with the whole spirit and very existence of any philosophy which claims to be critical? 3. Kant distinguished speculative and practical reason too sharply, and separated them too widely. They are represented by him as more exclusive and antithetic than they really are. Had he not done so he could not have conceived of God as not in some measure an object of knowledge but merely of belief; could not have failed to see that if God be inevitably thought of as morally necessary, even in the way which he himseK describes, God must be to that extent really and necessarily known ; not known, indeed, in the absoluteness, depths, and mysteries of His being, but known in the only way any being can be known 2U "BACK TO KANT" by men as a moral being — viz., through moral expe- rience and moral intuition or inference. To have for belief in God as a moral being the only kind of reason appropriate for such belief is not to have merely belief in God but to have a real knowl- edge of God, a knowledge founded on reason and valid for reason, and not essentially distinct from so-called theoretical knowledge. Kant, in a word, by crudely contrasting theoretical and practical reason, has, of course, not succeeded in establishing any precise distinction between what he calls knowl- edge and belief; on the contrary, he has shown him- self quite unable to maintain any of the distinctions which he has incidentally laid down between them. The fundamental affirmations of the practical reason, even as exhibited by himself, have the characteristics which he would confine to theoretical knowledge. During the last forty years many philosophical writers have been raising the cry of " Back to Kant," and none have done so more loudly than theological and anti-theological agnostics. The cry was far from wholly unreasonable and has been far from unpro- ductive of good. Kant must be acknowledged to have been to recent and present philosophy, as Aristotle was to ancient and medieval and Descartes to modem philosophy, its chief fountainhead ; and numerous as are the rivers and rivulets into which it has parted, all of them have owed much of what they are to what they have derived from him. We have now, and are likely to have for long, abundant reason to " go back 235 AGNOSTICISM OF HUME AND KANT to Kant," but we should certainly not go back to him in a servile and passive but in a free and critical spir- it. In philosophy to call any man master is proof pos- itive that you have no true sense of what philosophy is. " Back to Kant " — yes, but only back to him as to all great philosophical teachers. " Back to Kant " — yes, but to criticise as well as simply to imbibe; to determine what ought to be rejected and combated as well as to ascertain what should be adopted and utilised. In dealing with Kant in the preceding pages I have gone back to him only to criticise. To have done more would have been irrelevant so far as my task is concerned. I have further criticised only those views of his on which recent ag"nosticism has sought to build ; have challenged merely those positions of his 4;heory of knowledge which are sceptical in tendency and have actually been largely made use of for the support of agnostic ends. I have passed no judgment on other principles or portions of his philosophical teaching, or on the developments to which they have given rise. I willingly acknowledge even that should my criticism of Kantian agnosticism be allowed to be relevant and substantially sound, agnosticism may notwithstanding be rightly held to have been enor- mously indebted to Kant. Recent agnosticism cer- tainly owes to him the larger part of what has given it plausibility and attractiveness, very much of what is best in its spirit and strivings, and, in a word, very much of all that constitutes the superiority of recent agnosticism over earlier agnosticism. 336 KANT'S BPISTEMOLOGY Limited, however, as my treatment of Kant has been in scope, it may suffice to show that neither agnostics nor anti-agnostics can rationally go back to the epistemology of Kant as a foun- dation on which to build a philosophy. Anti-ag- nostics cannot, seeing that the Kantian epistemol- ogy is agnostic to the core. To say, as M. Auguste Sabatier does, that " to make Kantism end in scepti- cism shows a lack of intelligence," is to ignore the bases of the Critique of Pure Reason and to betray a strange ignorance of the epistemological doctrine which he so much admires. Intelligent agnostics can no more go back to Kant's theory of knowledge than their opponents, and that for the simple reason that it is in the main not a sure foundation but one of wood, hay, and stubble. They must substitute for it a better if they would not avow utter scepticism and avoid manifest inconsistency. They must not merely " go back " to Kant, but must do all the fun- damental portion of his work over again. A distinguished German !N^eo-Kantist once warned the philosophical world that Kant should only be criticised on the presumption that he was " a genius." Certainly he was " a genius," and a very great ge- nius. But " a genius " should enjoy no immunity from criticism. Indeed, a genius is a man who is just as capable of going farther astray from the truth than other men as he is of making greater progress in it. Plato, Aristotle, Descartes, Spinoza, Schelling, and Hegel were men of rare philosophical genius, but the.ir genius did not preserve them from colossal blun- 337 AGXOSTICISM OF HTME AXD KAXT ders and terrible misadventures. Kant erred as thej did, erred as a man of genius — erred, as Luther rec- ommended Melanchthon to sin. fortiter. It was in sij erring that He affirmed, and tried but failed to prove those dogmata which so many have rashly accepted as the justification of their agnosticism. Hence many have cried " Back to Kant " who. had their intellect- ual vision been dearer, would have seen that they might more consistently ciy, *' Back from Kant to Hume " — ^back to the abyss which Hume revealed, and from the sight of which Kant recoiled, and then strove (largely, alas I in vain) to fill up and bridge over. 238 CHAPTER V COMPLETE OR ABSOLUTE AGNOSTICISM I. AGNOSTICISM NOT EXACTLY DIVISIBLE. ITS GENERAL DIVISIONS Agnosticism has already appeared in many forms, and may yet appear in many more. Being essen- tially indefinite, it is easily and manifoldly variable. Any agnostic thinker of ability may give to his agnos- ticism an original and individual character, although it seems to surpass the ingenuity of man to devise a type of agnosticism at once consistently agnostic and clearly distinctive. Agnosticism is, in fact, never self-consistent, and never exactly this or that, but always relatively a more or less; and, consequently, any mode of division of its forms vphich pretends to absolute logical correctness shows, on the part of its proposer, want of insight into the essentially Protean nature of his subject. Agnosticism, in a word, is not more exactly divisible or distributable than it is ex- actly definable. Hence to attempt any elaborate clas- sification of its species and varieties would necessa- rily involve a waste of labour. Such a classification might possibly be made plausible, but would certainly be untrue. The most current general divisions of agnosticism are into total and partial agnosticism, and into abso- lute and modified agnosticism. And, so far as I am 339 COMPLETE OE ABSOLUTE AGNOSTICISM aware, there are no others which are of any value. But although the best, the most fundamental, and the most instructive of such divisions, — and although no one can relevantly either argue for or against ag- nosticism without having regard to them, — they are rather ideal than real, and, one may almost say, are so general as to apply exactly to no particulars. There never was, is not, and never will be, a total or absolute agnosticism. Man lacks the skill to con- struct and the courage to maintain a system which entirely and expressly disowns and disavows the ra- tionality distinctive of his nature. What we may agree to call total or absolute agnosticism is never strictly either total or absolute, but always so far lini- ited and qualified. While impartially, perhaps, spreading doubt over all things and extending its dis- avowals of knowledge to all the alleged spheres of knowledge, it is under the necessity of drawing lines and of making assumptions of some kind, which in some measure, and, it may be, to no small extent, re- strict its profession of ignorance, and implicitly re- tract its doubts and disavowals. The pressure of physical phenomena, of states of consciousness, and of the necessities of practical life, is of so direct, im- perative, and powerful a kind as inevitably to pre- vent a complete development of the agnostic ideal even in the most sceptically disposed individual. On the other hand, there is no merely partial or modified agnosticism, — no agnosticism which is not inconsistent as partial and done violence to by modi- fication, — ^which does not logically carry with it a 240 NO PAETIAL AGNOSTICISM demand to be completed and rendered thorough. The agnosticism which is explicitly partial and modified is, always and of necessity, implicitly total and abso- lute. For, while the objections which apply to ag- nosticism in general apply also, of course, to its spe- cial forms, — while any inconsistency involved in the very nature of agnosticism must be found in all its particular phases, — partial agnosticism always adds inconsistency of its own to that which is implied in the mere maintenance of agnosticism as such — an in- consistency inseparable from its specialisation. Ag- nosticism has no right to limit itself ; its " thus far, but no farther," is always an arbitrary one. The same kind of argumentation which is held to destroy the credit of one power of mind or department of knowledge would, were it valid at all, be equally de- cisive if directed against other powers of mind and departments of knowledge. We cannot set aside any one real law of thought, except on grounds which, if sufficient, would warrant us to set them all aside. Our rational life is a unity to which all its laws and powers are essential. From the rejection of the least of the laws of mind the rejection of all will logically follow. From the suspension or extension of the humblest of its powers the entire cessation of its in- tellectual activity must be a necessary consequence. The divisions of agnosticism into total and partial, absolute and modified, may coincide, but are not identical. The one is a quantitative and the other a qualitative division. The one rests on difference of extension and the other on difference of nature. 241 COMPLETE OR ABSOLUTE AGNOSTICISM They are two forms of the complete and incomplete, but distinct forms. Much which is true of the one may be true of the other, and scrupulously to distin- guish them on all occasions may often be pedantic, and even impossible. But it is necessary to know how and when to distinguish them. Far from being always coincident, they may at times be contraries. Total agnosticism may be modi- fied, and partial agnosticism may be absolute. The doubt or disbelief which is unlimited as to extent may be qualified in its nature ; the doubt or disbelief which has a limited sphere assigned to it may have that sphere given wholly over to it. It is likewise to be noted, however, that modifica- tion and limitation imply each other and are insepa- rable. Universal doubt or disbelief may be modified, and yet as modified extended to all things ; but so far as modified it is limited throughout, although not limited to a particular sphere. The doubt or disbe- lief, on the other hand, which is limited to a particu- lar sphere must be so limited because it differs in nat- ure as in legitimacy from doubt or disbelief within the spheres in which certainty and knowledge are at- tainable. In like manner the absolute doubt or dis- belief which is not total cannot be strictly absolute, and the total doubt which is not absolute cannot be strictly total. Totality as to extension and absolute- ness as to nature are alike requisite to completeness. The partial and the modified are alike forms of in- completeness. Agnosticism, then, is divisible into complete and 242 COMPLETE AND INCOMPLETE AGNOSTICISM incomplete as regards either nature or extension. The former refuses to admit that there is any certainty of knowledge, and questions the veracity of every prin- ciple and povsrer of cognition. The latter refuses, with more or less of qualification, to admit certainty of knowledge, and questions the reality or veracity of some particular principle or principles, power or pow- ers, of cognition. The latter is, of course, much the more common. And it has a great variety of forms, seeing that dis- trust of any law of thought or faculty of mind leads to a partial agnosticism, even although due confidence be reposed in all the other laws and faculties of the mind. Thus there is an agnosticism which rejects the testimony of the senses while admitting that of reason, and an agnosticism which rejects the testi- mony of reason while admitting that of the senses. Then there is an agnosticism which holds religious truth to be unattainable but passes unchallenged the findings of philosophy and science, and an agnosticism which combines philosophical doubt with theological dogmatism. And so on. For the ends which the present writer has in view the most suitable classifi- cation of the incomplete or partial forms of agnosti- cism will be into non-religious, anti-religious, and re- ligious. It is chiefly with anti-religious and religious agnosticism that he is in this work concerned. On non-religious partial agnosticism it will be unneces- sary for him to say more than suffices to indicate its bearings on the agnosticism which deals adversely or favourably with religion. 343 COMPLETE OR ABSOLUTE AGNOSTICISM II. WHY COJIPLETE AG^^OSTICISM EEQUIEES TO BE DIS- CirSSED. CEITICISM OF THE VIEWS OF PAULSEN" Complete agnosticism must be first considered. Until it be disposed of we cannot reasonably proceed to judge of any other kind of agnosticism. The gen- eral includes the particular. If there be no certainty or knowledge there can be no religious certainty or knowledge. If the conclusion to which total or ab- solute agnosticism comes can be successfully main- tained, all views to the contrary which men entertain regarding God, spiritual things, and theology must obviously be surrendered, equally with those which relate to nature and man, ordinary knowledge, physi- cal science, and philosophy. We clearly cannot afford to grant that it is a warrantable conclusion, and hence must criticise the claims of the theory which has com- mitted itself to its support. And yet it is very probable that to most persons at first sight absolute agnosticism, universal scepticism, will appear too extravagant and incredible a scheme of thought to call for any discussion. They may doubt whether it has ever been seriously entertained or propounded, and think that to make any attempt ■to refute it, or to take any special notice of it, is to do it too much honour and to assign it too much impor- tance. iKor is this view confined to those who are not conversant with philosophy and its history : it has been maintained by philosophical writers of good re- pute. Therefore it must not be quite ignored, 244 PAULSEN ON THE THEORY OP KNOWLEDGE Professor Paulsen of Berlin may be selected as the spokesman of those who hold it. Treating of the theory of knowledge, or what is commonly called epis- temology, he afSrms that its fundamental problem must be answered in one or other of four distinct ways. These are : — (1) We know things as they are in themselves through perception. This is the answer of Realistic Empiricism, the view which comes nearest to the naiive of common conception. (2) We know things as they are but only through reason, not through the senses. This is the answer of Realistic Rationalism, the one re- turned by Plato, Spinoza, Hegel, and other great metaphysical system-builders. (3) We know about things only through percep- tion, yet certainly attain thereby no ade- quate knowledge. This is the answer of Idealistic Empiricism, and Hume may be regarded as its most resolutely logical advo- cate. (4) We know reality a priori through pure rea- son, yet certainly not as it is in itself, but only as it appears to us, and indeed only according to the forms of our intuition. This is the answer of Idealistic Rational- ism, the view of Kant. Dr. Paulsen then proceeds to say: " The historians of philosophy are wont to bring before us yet another form of theory of knowledge, scepticism, which af- firms that we can have no knowledge. Here and 245 COMPLETE OE ABSOLUTE AGNOSTICISM there some one even takes the trouble to contradict this view. It seems to me to be superfluous trouble. If there were ever real scepticism it has died out in modern times. No modern philosopher has doubted that there is real knowledge, distinguishable from nescience. It is customary to refer to Hume as the representative of scepticism. And, sure enough, Hume plays with the designation. Tor this he has been sufficiently punished by the consequent miscon- ceptions of his meaning. But it never occurred to him to maintain that there is no such thing as science. He merely maintained, on the one hand, that natural theology with its proofs of the existence of God and of the immortality of the soul is not science ; and, on the other hand, that any knowledge attainable regard- ing facts must be acquired through experience, and is not of universal and necessary validity. It was Kant who stamped Hume as a sceptic, whom he had to oppose in order to save the sciences and show the possibility of metaphysics, physics, and even mathe- matics. As regards pure mathematics Kant's jiidg- ment on Hume's scepticism rests on pure misunder- standing ; as regards metaphysics he himself, not less than Hume, rejects rational theology, cosmology, and psychology. There remains physics: here both ad- mit that there is such a science; they difFer only in their views of the form and nature of the certainty of its propositions. Kant thinks that some among them are absolutely universal and necessary (a priori synthetic judgments), while Hume regards even its axioms as only presumptively general propositions 246 PAULSEN ON PHILOSOPHICAL SCEPTICISM dependent on experience, — a difference of opinion which cannot be fitly expressed by saying that Hume denies the possibility of physics. " So far as I see, it is the same with other sceptics. They do not deny the possibility or the existence of the sciences, but only emphasise the limitedness and uncertainty of human science compared with an ideal of knowledge such as may possibly be realised in a divine mind. The only scepticism to be found in modern philosophy is one which opposes the preten- sions of transcendental speculation; it shows a two- fold aspect, inasmuch as it defends either religious faith or empirical research against the usurpations of spfeculation." ^ Now, such statements as these are so apt to mislead that they cannot prudently be passed over in silence. They are, for the most part, very inaccurate. The historians of philosophy are quite justified in bringing philosophical scepticism before us in the way which they generally do. They could neither reasonably ignore so remarkable a phase of philoso- phy, nor could they give any substantially different account of it than that which they present. They could only adopt Dr. Paulsen's view of it by disre- garding or misinterpreting the sources of information relative to it. However, it is inaccurate to say that they represent scepticism as " a kind of theory of knowledge which maintains that we can have no knowledge." This they are not " wont to do." On the contrary, all historians of philosophy of good re- ' EinleUwng in die Philosophie, 352-353. 247 COMPLETE OE ABSOLUTE AGNOSTIOISM pute represent scepticism as a kind of theory of knowledge the holders of which, if not invariably at least as a rule, content themselves with maintaining that those who profess to have knowledge, — those whom they regard as dogmatists, — have not, in their judgment, succeeded in showing that they have ra- tional grounds for their profession, or for the belief which it implies. Dr. Paulsen pronounces it "superfluous trouble " to contradict the sceptical theory of knowledge. But were no one to contravert it, and to show grounds for rejecting it, would the need for any other theory of knowledge be made out ? Would not, in that case, all search for another theory of knowledge be justly cen- surable as " superfluous trouble " ? The actual historical existence of philosophical scepticism, both in ancient and in modern times, is about as certain as anything historical can be. There can only be reasonable difference of opinion as to whether or not there has appeared an absolutely com- plete, fully and self-consistently, evolved scepticism in the course of the history of philosophy. We are ready to grant that there has not; that such scepti- cism is not only a rare phenomenon in history, but an unknown and indeed an impossible and inconceiv- able one. Are not, however, all the specifically dis- tinct theories of knowledge in this respect on the same level? Are not realism and idealism, empiri- cism and rationalism, equally with scepticism, theo- ries of knowledge which have only attained in history an incomplete and inconsistent manifestation ? Has 248 PHILOSOPHICAL SCEPTICISM there, for instance, been any one who, fully realising what he meant, affirmed that he knew things as they are in themselves through^ and only through, percep- tion? Has there ever been a man so naif or such a philosophical simpleton as to be a mere and complete realistic empiricist ? Scepticism — i.e., universal scepticism or absolute agnosticism — -stands on the same footing as other theories of knowledge in being rather an ideal than a reality; and in its contradic- tion being rather an argument against a general spec- ulative tendency than against the doctrine of any par- ticular person, even the most sceptical. On the other hand, philosophical scepticism has often advanced very far towards completeness, — so far that the discussion of it as complete is legitimate and necessary. In the course of the history of specu- lation many thinkers have appeared whose views as to knowledge left hardly any room for belief in its reality. In antiquity, Pyrrho and his followers — ^the founders and disciples of the Middle Academy — and the members of the later sceptical schools, such as ^^nesidemus, Agrippa, and Sextus Empiricus, — all refused to admit that any proposition as to the reality of things or as to real truth could be known or proved with certainty, and held that, as to the truth or falsity of such propositions, suspension of judg- ment was the appropriate state of mind. Those who went thus far were surely very nearly complete scep- tics, although they could not avoid making some con- cessions inconsistent therewith, but without which they could never have justified their reasoning or act- 349 COMPLETE OE ABSOLUTE AGNOSTICISM ing on any subject or occasion whatever. Then, the theological agnostics who, from the Renascence to the present day, have laboured to discredit natural reason in order to induce men to put their trust in supernat- ural grace or the guidance of external authority, are to be accounted, to all intents and purposes, sceptics as to knowledge and science in general. Now, what of Hume ? " It never occurred to him," says Dr. Paulsen, " to maintain that there is no such thing as science." No, and that is not what has been attributed to him. He neither denied that there was any such thing as science, nor professed to disbelieve what either science or sense taught. What he did was to undertake and carry out ingenious investiga- tions which served inevitably to lead to the conclusion that those who believed the teaching of either sense or science, experience or reason, had no logically valid grounds for doing so. His scepticism, in other words, did not appear in a direct denial of the existence of knowledge, but in an elaborate reduction of sub- stances to collections of ideas, of time and space to subjective conceptions, of the causal connection to habitual association, of reason to custom, and the like. And a very thorough scepticism it was. If the con- clusions to which it led were well-founded, no kind of knowledge was well-founded; if it proved any- thing, it proved that perception, experience, and rea- soning proved nothing. It was concentrated in the metempirical criticism which he applied to the bases of all knowledge. That criticism was subversive of all science and pholosophy; as subversive of mathe- 350 HUME'S SCEPTICISM matics and physics as of theology and metaphysics. Hume would have been not less of a sceptic if he had never vreitten a sentence about natural theology. It is as erroneous to say that he was an unbeliever in the existence of God or in the immortality of the soul as that he was an unbeliever in the law of gravitation ; and, on the other hand, it is as erroneous to represent him as recognising that the law of gravitation can be any more rationally proved or known than the exist- ence of God or the immortality of the soul. Says Mr. Balfour not less justly than forcibly : " Nothing in the history of speculation is more astonishing; nothing — if I am to speak my whole mind — is more absurd than the way in which Hume's philosophic progeny — a most distinguished race — have, in spite of all their differences, yet been able to agree, ioth that experience is essentially as Hume described it, and that from such an experience can be rationally extracted anything even in the remotest degree re- sembling the existing system of the natural sci- ences." ^ That Hume is brought before us by the historians of philosophy as a sceptic is certainly not owing to Kant having " stamped " him as such. He presented himself to the world as a sceptic, the author of a philosophy akin to the Greek sceptical philosophy. He pleaded " the privileges of a sceptic " for just the kind of reasonings on account of which alone fair ex- positors of his views designate him a sceptic. He wrote and spoke familiarly of " his scepticism " as 1 Hie Fovmdations of Belief, p. 103 (8th ed.) 251 COMPLETE OE ABSOLUTE AGNOSTICISM " his philosophy " ; took no objection to his most dis- tinctive speculations being characterised and criti- cised as " sceptical " ; and was universally recog- nised, both in Britain and on the Continent, as a sceptic, yea, the coryphaeus scepticorum of modern times, as soon as his philosophical writings became known. Kant was not at fault in attributing to Hume a scepticism as sweepingly destructive as that of a Cameades or ^nesidemus: his mistake lay in supposing that his own Critical Philosophy was an antidote to it. We have so recently had Mr. Arthur Balfour, Dr. Gordy, the Abbe Martin, and others arguing in the most explicit manner that the foundations on which physical science rests are not rational grounds of conviction, but non-rational impressions, impulses, or inclinations, that it is impossible for us to admit that even the latest scepticism, the scepticism of to-day, merely emphasises the limitations and uncertainties of human science or restricts itself to opposing the pretensions of transcendental speculation. It does that, but it also unquestionably does more. It like- wise challenges all so-called positive science to show that its principles are not merely assumptions which have no other guarantee than the fact that they are believed, and that the processes through which its results are obtained are not logically illicit and in- conclusive inferences. What it avowedly and ex- pressly seeks to show is that the foundations of the creed of science are just of the same character as those on account of which so many scientists deem 252 COMPLETE AGNOSTICISM themselves entitled to refuse serious consideration to any religious creed. The interest and value of the latest defences of " philosophic doubt " depend mainly on their being a criticism of the claims of science and of the pretensions which have been based on science. The agnostic solution of the problem of knowledge, however strange or unsatisfactory it may seem to us, is obviously a distinct form of solving it, and one of a thoroughly radical and comprehensive kind. Hence it should not be ignored by us, but examined as to what it essentially and distinctively is, although it may never have been fully realised as such in any one historical system. To refuse to do this on the ground that it has never been so realised is a clear evasion of logical duty, and much more convenient than com- mendable. A complete agnosticism is, indeed, no- where to be found, and consequently a refutation of it cannot apply strictly and immediately or with full force to the teaching of any actual individual agnos- tic. It is, therefore, of less practical use Aan it would be were complete agnosticism prevalent. It leaves a special examination and refutation of each form of agnostic doctrine still necessary. It is, how- ever, useful, and even essential, in its place. It ap- plies indirectly and in some measure to all that is truly agnosticism, and, indeed, applies to it precisely to the extent to which it is truly agnosticism. All incomplete agnosticism tends to completion, and must be so far judged of by what it would be if com- plete. All actual agnosticism must be viewed in re- lation to that absolute agnosticism which is simply 253 COMPLETE OE ABSOLUTE AGNOSTICISM the full natural and logical development of agnosti- cism. III. SPECIES OF COMPLETE AGNOSTICISM. INCONSIST- ENCY OF SYSTEMATIC AND UNIVEESAL DOUBT The agnosticism which professes to he complete, or at least aims at completeness, hoth as to extent and nature, may be received in various aspects or rela- tions. Let us consider it first as to self-consistency or rational self-coherence. 1. Complete agnosticism is either systematic and universal doubt or systematic and universal disbelief. This is not the view commonly taken. Complete ag- nosticism is generally considered to be merely system- atic and universal doubt. It is said that to disbelieve is, in reality, to believe that what is disbelieved is er- roneous, and, therefore, that it is incompatible with the force or philosophical scepticism which is desig- nated agnosticism. But to doubt is also in the same way to believe; it is to believe that there is no warrant for a decision, — ^that there is such a want of evidence in regard to a proposition, or that the evidence for and the evidence against a proposition are so nearly balanced that we are not entitled either to affirm or deny that proposi- tion. The opposite of belief is neither disbelief nor doubt but the absence of belief. An element of belief can no more be eliminated from doubt than from dis- belief. If this be inconsistent with absolute philo- sophical scepticism or complete agnosticism, inas- 254 ITS SELF-CONTRADICTOEINESS much as it means that neither by doubting or denying can belief be entirely got rid of, it is only because such scepticism or agnosticism is not, and cannot be, self -consistent. A man believing nothing except that he knows nothing still believes something. He does not believe more, however, than a man who doubts whether he knows anything or not ; on the contrary, more belief is reserved in the doubt of the latter than in the dis- belief of the former. A man who disbelieves his senses on the ground that the senses are unveracious, is at least as much of a sceptic and agnostic as one who cannot decide whether to believe them or not. No doubt can be more sweepingly sceptical or agnostic than an absolute denial of the possibility of knowl- edge. The utmost extreme and extravagance of ag- nosticism is to be attained not through mere doubt but through a double negation, which, by first denying all things and then denying itself, leaves reason ob- jectless and powerless — a double negation for which Areesilaos may have found the formula, although not meaning to convey by it the signification, when, in opposition to the " I know nothing, except that I know nothing," of Socrates, he said, " I know noth- ing, not even that I know nothing." Agnosticism, then, may be regarded as either doubt or disbelief of the attainability of knowledge and truth, and the question now before us is, Can a com- plete or absolute agnosticism be self -consistent ? It is a question which I can only answer in the negative. While agnosticism must be inconsistent so long as it is 255 COMPLETE OK ABSOLUTE AGNOSTICISM not complete, it cannot be consistent when it is com- plete. It is of its very nature inconsistent and self- contradictory. It is so, alike as universal disbelief and as universal doubt. Both of these states of mind are essentially irrational. And the irrational can- not become rational by logical development; incon- sistency cannot be transformed into consistency by being completed. To make manifest the self-contradictoriness latent in the strictly absolute agnosticism alike of doubt and of disbelief is the task now immediately before us. Agnosticism, then, cannot be self-consistent in the form of systematic and universal doubt. It supposes even in this form a power of weighing evidence which is irreconcilable with the absolute distrust and inde- cision which it inculcates. Men do not doubt, any more than they believe or disbelieve, what they have no evidence either for or against, and know nothing about. So far from implying an entire absence of judgment, doubt is a suspension of judgment based on the judgment that neither an affirmative nor a neg- ative judgment would be warranted in the circum- stances. Where evidence and knowledge are wholly wanting, belief, disbelief, and doubt are alike out of ■ place. The mind is then a blank, unintelligent and unconscious; but this state of mere blankness or emptiness, if it can be called a state, is wholly differ- ent from doubt. Doubt is an actual or positive condition of mind, and often a most legitimate and valuable one, but it requires justification equally with belief and disbe- 256 LIMITS OP DOUBT lief,' and it can only be justified by showing that the reasons both for belief and disbelief, for affirmation and negation, are insufficient — that they counterbal- ance and counteract one another. This implies, how- ever, that the mind is competent to estimate the rea- sons both for belief and disbelief, for affirmation and negation, and to weigh the one set of reasons against the other set. It supposes that belief, disbelief, and doubt should correspond to evidence, and that evi- dence may be so apprehended and appreciated as to explain and effectuate the correspondence. A mind altogether incapable of knowing itself entitled to be- lieve and disbelieve must be as incapable of knowing itself entitled to doubt, and, consequently, must be as much bound to suspend its doubt as its belief or its disbelief; or, in other words, must not reason, judge, or think at all. Everything short of, or different from, the entire ejection of intelligence, the absolute suppression of rational activity, must be irrational in a mind so constituted ; and, in fact, such a mind would be a reason of which every movement would be neces- sarily unreasonable. The mind of man has not been so constituted, and is not thus under the appearance of rationality realised absurdity ; and hence its doubt- ing, not less than its believing and disbelieving, prop- erly exercised, is a perfectly legitimate mode of existence and activity — one dependent on and accord- ant with reason. Doubt, pushed to its utmost extent, is only intel- ligible on the supposition that the mind can appreciate evidence, and distinguish between truth and eiTor. 257 COMPLETE OE ABSOLUTE AGNOSTICISM It presupposes, in other words, the very truth and cer- tainty which the agnostic would persuade us it sets aside. Thus we have only to compare the latent as- sumptions from which the agnostic reasons with the conclusion at which he arrives, to iind that his alleged demonstration of absolute doubt is also a reductio ad absurdum of such doubt, the doubt itself being essen- tially inconsistent. The agnosticism of absolute doubt is self -condemned by its self-contradiction. The agnostic, it is often said, can have no creed. The saying shows lack of reflection : no rational being can be creedless. The agnostic can no more dispense with a creed than his neighbours, although it may be peculiarly difiicult or inconvenient for him. to profess that he has one. He needs a creed even in order to prove that there should be none. The advocate of universal doubt cannot take a single step towards the vindication of his doubt iinless he believes, and be- lieves himself to know what certainty, knowledge, evi- dence, and truth are. He may, and, indeed, as a matter of course, will prefer such views of certainty, knowledge, evidence, and truth as seem to him most likely to subserve his purpose, but the very choice and use of these or any views regarding them implies a belief in the very things — certainty, knowledge, evi- dence, and truth — of which universal doubt is. the negation. " The very logic," says Edward Caird, " by which the sceptic overthrows the dogmas of phi- losophy, implies that the mind possesses in itself the form and idea of truth. His deepest doubt reveals a certitude that transcends and embraces it." 358 EELATION OP BELIEF TO KNOWLEDGE There is a haziness of conception in the minds of many persons as to the real relationship of doubt to belief and disbelief which cannot but prevent full rec- ognition of the force of the preceding remarks. Per- haps a few words may somewhat help to clear it away. Belief is the assent of the mind to what it regards as true — to what it thinks it knows. It is easily dis- tinguishable from such mental states as imagination, feeling, desire, and volition, but inseparable from all rational intellection both intuitive and discursive, and coextensive with true and erroneous judgment, real and imagined knowledge. No man can believe anything which he does not suppose that he knows to be true. What the mind in belief regards as true may not be true, but it cannot believe what it does not apprehend as true ; what the mind believes it knows it may not know, but without believing that it knows it cannot believe at all. There is thus in the very nature of belief a direct reference to knowledge and truth. Those who would base all knowledge on mere belief or reduce all knowledge to mere belief overlook that there is no such thing as mere belief, as entirely self-contained belief ; that there is only belief which includes a reference and appeal to knowledge and truth. Those who talk of a belief which is its own guarantee directly contradict the testimony which belief bears regarding itself. The voice of all belief is : I speak not on my own authority ; I have no right to acceptance or existence except what I re- ceive from knowledge and truth. 259 COMPLETE OE ABSOLUTE AGNOSTICISM Belief adheres indissolubly to all knowledge. Whatever we know, we believe. There is no differ- ence in this respect between immediate or intuitive and mediate or discursive knowledge; between pre- sentative and representative knowledge; between knowledge of the past, present, or future; between the knowledge which comes to us through sense, or through the understanding, or through the reason. It has often been attempted by the perverse use of terms to separate belief from knowledge, and to oppose the one to the other, but every attempt of the kind is so- phistical and irrational. The opposition of belief to knowledge has no proper meaning or justification. Wherever there is knowledge, there is belief founded on the knowledge ; and wherever there is belief not foimded on knowledge, it is illegitimate and self -con- tradictory belief. Of course, there is an immense amount of belief of the latter kind — of belief in a knowledge which is not real but imaginary, of assent to error under the impression that it is truth. Be- lief, while co-extensive with real knowledge, is far more extensive; it is as inseparable from false as from true judgment, from the abnormal as from the normal workings of the mind in the exercise of its cognitive faculties. A world where belief was precisely co-extensive with knowledge, precisely in accordance with evi- dence, would be a world where there were no erro- neous beliefs. Our world is still very far, indeed, from being such a world. It is, however, the goal which a rational world should strive to reach. It is 260 BELIEF AND DISBELIEF the ideal at which a rational man should aim. We cannot believe what we do not know, or think we know ; but we have no right to believe more than we know, or to be content with merely thinking we know instead of trying our best really and truly to know. Evidence should be the measure of assent. All real evidence we are bound to receive, and to estimate ac- cording to its actual weight and value. What is true of belief is equally true of disbelief, and for the simple reason that disbelief is belief. But slight reflection is needed to dispel the common notion that disbelief is the opposite of belief. The man who disbelieves in Irish Home Kule believes just as much as the man who believes in it, only he believes that it would be bad, whereas the other believes that it would be good. Disbelief is not the opposite of belief, but belief of the opposite — belief that a particular propo- sition is not true. The believer and the disbeliever differ only in that their beliefs differ and conflict. Both have beliefs, and they are alike responsible for the character and correctness of their beliefs. !N"or is doubt the opposite of belief. To doubt is to believe that there is not warrant for a flrm decision, — that there is insufiicient evidence for a resolved and settled belief. It implies a commingling of belief and disbelief; or, as it may be also expressed — since disbelief is itself belief — a combination of positive and negative is belief. When the evidence in favour of a proposition seems to a man full, he believes and does not doubt ; when the evidence against it seems to him full, he disbelieves and does not doubt ; when the 361 COMPLETE OR ABSOLUTE AGNOSTICISM evidence in regard to it seems to liini inadequate in amount or ambiguous in character, partly in favour of and partly against its affirmation, he partly be- lieves and partly disbelieves — believes because there is evidence, and disbelieves because it is not of such quantity and quality as to show whether the proposi- tion be true or not — and only in this ease does he doubt. Doubt is thus of a double nature : a mixture of belief and unbelief ; the opposite of neither belief nor disbelief, but only of assured belief or assured disbelief. It is so far from being a state of mind in- dependent of, or distinct from, belief or disbelief, that it may approximate closely to both and be diffi- cult to distinguish from either. We may think that we believe when there is much unbelief in our belief, and that we doubt when there is much faith in our doubt. " More faith," the poet truly tells us, " may live in honest doubt than in half the creeds." The conflict of judgments and the counteraction of belief and disbelief in doubt are what is characteristic of it, and what the very terms for doubt in the various lan- guages of the world show that men have everywhere recognised to be its characteristic. Belief and disbelief, then, are two species of belief, and doubt contains both and arises from their coun- teraction. Wherever there is perceptive or intellect- ive judgment, intuition, or inference of any kind, there also is belief in its positive, negative, or dubita- tive form ; and in whatever form it appears, it should correspond to the relevant attainable evidence. When .Dr. Bain and other psychologists tell us that doubt 262 NATUEE OF DOUBT is the opposite of belief, they are obviously mistaken. It is belief, and belief of a particular kind — belief that the reasons for and the reasons against some opinion or proposition tend more or less to counteract and cancel one another, and so warrant neither a de- cidedly afiSrmative nor a decidedly negative belief. It necessarily supposes in every case some degree of belief, some perception of evidence, and a certain power of estimating the weight and worth of evi- dence. The only opposite to belief is the absence of belief, and there can only be the entire absence of belief in a mind devoid of all judgment as to truth and error and of all apprehension of evidence. Entire ignorance is the only complete security against doubt. " Men that know nothing in sciences," says Arch- bishop Leighton, " have no doubts." If doubt be of the nature now described, the essen- tial inconsistency of the agnosticism of absolute doubt is apparent. Doubt in every ease requires to justify itself no less than belief or disbelief. It ought equally to be in accordance with evidence, and it has specially to judge the evidence both for and against what is doubted. It should give heed even to the least evidence, and to all the evidence pro and con. It is the most complex form of belief, the latest to make its appearance in consciousness and history, and the most difficult correctly to regulate or appreciate. The child, the savage, and the common man believe and disbelieve more readily than they doubt. Doubt is a peculiarly imstable state of mind. Dubious questioning is to men in general unpleasant, and to 263 COMPLETE OE ABSOLUTE AGNOSTICISM many men intolerable. Dull believing or vehement disbelieving is easier to them, and more in favour with them even when much less commendable. What, then, would justify such a state of mind as the scepticism of absolute or universal doubt ? Only a completely self -contradictory world; one in which the evidence for all opinions was equal to the evidence against them ; one in which reason would be con- demned to perpetual self -stultification ; one in which all search for truth and weighing of evidence would necessarily lead only to learned ignorance strictly and literally understood — an ignorance absolute and complete, and yet one only capable of being estab- lished by an absolute and complete knowledge. In a world so strangely constituted self-contradiction would be the one great law, and the pure Pyrrhonist the only wise man, if even he were wise. The inhab- itants of it would need no other excuse for their indi- vidual contradictions and inconsistencies than the words of the poet : — " Die Welt ist voller Widerspruch, Und sollte sich's nioht widersprechen ? " The uniformly self-contradictory person in a com- pletely self -contradictory world would, if I may say so, be the only self-consistent character. The existence of a self-contradictory world, how- ever, has never yet been proved, and must be pecul- iarly difficult to prove by those who think nothing can be proved. So far as I can judge, it has never been shown that there are any other contradictions 264 A SELP-OONTKADIOTORY WORLD ASSUMED in the world than those for which such beings as our- selves — beings who too frequently judge and act irra- tionally — are responsible. IV. INCONSISTENCY OF SYSTEMATIC AND UNIVERSAL DISBELIEF Agnosticism, I proceed to maintain, cannot be self- consistent in the form of systematic and universal disbelief. In the very act of maintaining that truth cannot be reached, it implies that it has been reached. It is a denial that truth can be attained, but an af- firmation of the untrustworthiness of the mind. It rejects all that the mind ordinarily regards as true, but on the ground that the mind is incompetent to ascertain what is true. Is, then, we are bound to ask, this allegation of the mind's incompetency to ascertain truth itself true ? It obviously must be held to be so by those who make it, and who reject all other affirmations on the strength of it. Unless it be a truth, and a truth better established than all other statements asserted to be truths, agnosticism as universal disbelief, as denial of the existence and possibility of knowledge, can have no rational war- rant. If, on the other hand, it be a truth, what is to be made of the doctrine that truth is unattainable ? Why, in this case truth has been attained. One truth so comprehensive as to be a whole philosophy in itself — a truth which enables us to decide on the worth of every proposition which the human mind can enter- tain — has been actually and adequately established. 365 COMPLETE OE ABSOLUTE AGNOSTICISM If the mind, however, can acquire even one truth, and especially if it can make itself master of so ab- struse and significant a truth as is alleged, it cannot consistently be held to be so untrustworthy as the ag- nostic represents it to be. If the mind be justified in one instance in saying No, it may be warranted in other instances in saying Yes. The mind which- can prove its own incompetence can hardly be so incom- petent after all. It thereby shows itself capable of accomplishing an especially arduous task, the ascer- tainment of its own utmost reach of capacity and fac- ulty, of what it absolutely can and cannot do. This must require a most difficult and elaborate investiga- tion into the nature and limits of intelligence, and the reason which can successfully prosecute it cannot be so weak as is asserted. There is no kind of re- search in which failure is more probable. There is no question as to which the mind is less likely to suc- ceed in finding an answer than that as to the limits of its own capacities. Hence the agnostic negation is a denial that truth can be reached in cases where its attainment should be comparatively easy, based on the presupposition that it has been reached in a case where its attainment must be peculiarly difiicult. That it is a negation — a denial of the right of the intellect to accept anything as true — clearly does not affect the argument. It has no relevancy as an an- swer to it. A negative conclusion should be as much a result of investigation as a positive one. A nega- tive judgment, if really warranted, is as much a truth as an afiirmative one. Disbelief, as already shown, 366 WHAT UNIVERSAL DISBELIEF IMPLIES is not the opposite of belief but belief of the opposite, and as much dependent on truth and evidence as the opposite belief. Nor is disbelief — negative belief- easier to prove than belief — positive belief. ISTay, a negative is often specially difficult to prove. And the difficulty of proving the vast and daring negative distinctive of complete agnostic disbelief must be enormous. In fact, it would require omniscience to accomplish such a task. To affirm rationally what cannot be known one must have a comprehensive ac- quaintance with whatever is or may be; in other words, to know that nothing is knowable one would require to have a thorough knowledge of everything. But an agnosticism thus absolute would be identical with a complete gnosticism. The sceptical " there is no attainable knowledge of truth " is uttermost scep- ticism ; but it implies an " I know that there is no attainable knowledge of truth," which is an expres- sion of the uttermost conceivable dogmatism, and all the more dogmatic owing to its self-contradictoriness. Suppose disbelief pushed to the uttermost point conceivable. Suppose a man to maintain that we have no warrant to believe anything and should dis- believe everything. Does he thereby get rid of belief or its obligations? By no means. He is left with an enormous amount of belief for which he ought to have good reasons. His disbelief includes belief that every affirmative proposition which the human mind can entertain is false, and implies belief that the evi- dence seemingly for every such proposition is unsat- isfactory while the evidence against it is conclusive. 267 COMPLETE OR ABSOLUTE AGNOSTICISM ISTow a man with so much belief as that has surely more instead of less of it than his neighbours. And although he may, in one sense, rightly call himself an unbeliever or sceptic, he may in another and as legitimate a sense be justly maintained to be a greater dogmatist than any scholastic metaphysician or infal- libilist theologian known to history. Such a sceptic has much faith of a kind, and of a kind greatly in need of strong proof. It is a faith which presupposes a demonstration that the world is one which warrants the inference only of negative propositions. What sort of world would that be? One entirely disappointing to intelligence. One of which the very existence is inconceivable, and which were it real would be at every moment and point of its existence an offence and torture to thought. Not even the most self-confident, perhaps, of transcen- dentalist metaphysicians will dare to grapple with the dread idea of a world of which nothing except negations are true ; and certainly no one else will be so audacious. Fortunately the world of experience neither demands from us the superhuman intellectual exertions nor inflicts on us the continuovis and intol- erable intellectual disappointment which the world of the absolute agnosticism of disbelief must do. The actual world often yields, indeed, to our investigations merely so-called " negative results " ; but they are " negative " only in the sense that they negate our misconceptions of its realities; not in a sense which would put the world itself and reason itself to shame. 268 AGNOSTICISM AND FIKST PKINCIPLES V. ABSOLUTE AGNOSTICISM AND FIEST PRINCIPLES Absolute agnosticism we have argued to be inher- ently inconsistent. Let us now consider it in relation to the primary grounds of belief, the ultimate princi- ples of knowledge. The reality and validity of such primary grounds or ultimate principles are implied in all knowledge and reasoning. The most radical and resolute scep- ticism cannot dispense with the use of them even when attempting to displace and discredit them. It must assume and proceed on them even in order to vindi- cate its rejection of them. However complete, it cannot free itself from the obligation of trying to prove its assertions and endeavouring to convince others of their truth. Committed although it be to deny or question the reality or attainability of truth, it must claim to be itself true and truly established, and so far imply that there is truth, and that truth can be distinguished from error. While unable to admit that there is knowledge, it is not entitled to believe or assert that there is none unless it knows its belief or assertion to be well founded, which of itself would prove that there is knowledge, and that knowl- edge is distinguishable from ignorance and illusion. The very doubt or disbelief distinctive of the agnostic supposes, in fact, a faith which implies a creed, a whole system of judgments, which, notwithstanding the agnostic denial of knowledge, only knowledge can justify. Further, agnosticism professes to be a kind of philosophy, and undertakes to support and defend 269 COMPLETE OE ABSOLUTE AGNOSTICISM itself and to assail and overthrow other systems by means of. reason and reasoning. And this implies that there are laws of rational procedure, and some criterion or criteria by which it may be determined when these laws are observed and when violated. It follows that the question as to the relationship of absolute agnosticism to primary principles of knowledge must be one which vitally concerns it. What, then, is that relationship ? Well, in the first place, such an agnosticism, if an agnosticism of doubt, must obviously doubt all first principles, and if an agnosticism of disbelief, must disbelieve them. What it clearly cannot do is to be- lieve them. It must reject them ; cannot without self- destruction accept them. Its attitude towards self- evidence is necessarily that of distrust or denial, not that of trust. In a word, it must assume that there are no primary grounds of belief, no first principles of knowledge. If there be any such grounds or prin- ciples knowledge exists ; its foundations are laid, and a complete agnosticism is manifestly extravagant. In the second place, the agnosticism in question is not only logically bound to ma"ke the assumption that there are no first principles, but vitally interested in adhering to it. To have to admit that the assump- tion is unwarranted is for it equivalent to having to acknowledge itself throughout essentially untenable. If the foundations of knowledge be solid, if the laws which regulate intellectual activity be trustworthy, the theory that the mind of man can build up only false and illusory structures must be extravagant. 270 AGNOSTICISM REJECTS FIRST PRINCIPLES Absolute agnosticism, then, is incapable of either taking up or maintaining an impartial attitude tow- ards first principles. It may profess to be fairly and reasonably critical of them, and neither more nor less; but it cannot really afford to be so. Its rela- tionship towards them is of necessity as faulty as that of the most thorough dogmatism. A right relation- ship to them is one which does not exclude criticism of them, but which does exclude alike arbitrary rejec- tion of them and predetermination to prove them un- trustworthy. It does not exclude, I say, criticism of them — any criticism of them which is just and rational. On the contrary, it is' a manifestly incumbent and important part of the work of philosophy to criticise and test all principles alleged to be primary either as constitutive of knowledge or regulative of its growth. Ordinary thought, of course, does not do so. It accepts them without question ; apprehends, believes, and acts on them unreflectively as self-evident. And this is quite natural. It is all that the ordinary man can do, and all that he feels the slightest need of doing. But there is an obvious disadvantage attached to his mode of procedure. The ordinary man very often accepts as Self-evident what is extremely questionable or en- tirely erroneous. What he deems primary certainties may be merely inherited or current prejudices. What he trusts as natural reason or common-sense may be unnatural or nonsensical. A genuine philosopher cannot take the ordinary man as his guide or example. Nor can he take as such the ordinary scientist. 371 COMPLETE OK ABSOLUTE AGNOSTICISM The scientific specialist is, of course, much more care- ful in his dealing with first principles than the ordi- nary man ; but his attitude towards them is not essen- tially different. He does not any more than the ordinary man make them a subject of special investi- gation. He does not discriminate them from all else that is to be found in thought, and examine them in themselves, in their inter-relations, and their bear- ings on knowledge as a whole. He simply selects those of which as a specialist he has need, and of the peculiar worth of which he is aware. Scientific thought is thus, like ordinary thought, uncritical in its attitude towards knowledge and the first princi- ples thereof. Kone of the special sciences start with a criticism and theory of knowledge. And in so do- ing they act wisely, for otherwise they would find it difficult to start at all. Philosophy even may so far proceed in the same way. Its province is, not like that of the sciences, mere sections of knowledge, but knowledge as a whole. It may, however, simply accept knowledge as given to it through the sciences ; or, in other words, may make the sciences the object of its study, trace their rela- tions, exhibit them as an organic whole, co-ordinate and combine their conclusions, and present to the mind as correct a picture as it can of the whole intel- ligible world from the results thus obtained. Philos- ophy at this stage — what may be called positive or scientific philosophy — differs from ordinary thought and special scientific thought simply in virtiie of its generality or comprehensiveness. It is not self-criti- 272 PHILOSOPHY AND FIKST PEINCIPLES cising thought; although reasoned it is unreflective; it builds up what is admitted to be knowledge into a systematic or structural unity, but it does not inquire what so-called knowledge is or is essentially worth ; it is merely an advance on special science, as special science itself is on ordinary knowledge, and ordinary knowledge on crude sensation. Along the whole line the mind never changes its attitude towards its ob- jects ; at the end this is- just what it was at the begin- ning ; it is assumptive and dogmatic throughout. Philosophy, however, may assume, and is bound to assume, another attitude ; may pass, and ought to pass, from a dogmatic to a critical stage. It is called on to undertake a task which neither ordinary thought nor special science can perform, and yet which is a much- needed supplement to the work of both — namely, a methodical and impartial examination of the condi- tions and guarantees of knowledge as such, and in whatever form it may appear. And in the fulfilment of this duty it must be largely a criticism of the so- called primary or ultimate principles of knowledge. The criticism may conceivably lead to a completely sceptical result; that is to say, it may show all so- called knowledge to be credulity and all so-called science to be illusory. It may conceivably convict reason itself of being responsible for the inconsist- encies in agnostic argumentation, and make so mani- fest the constitutional invalidity and vanity of thought as, in a sort of way, to justify the claim of absolute agnosticism to be the best philosophy attain- able. The conceivability of the criticism having so 273 COMPLETE OE ABSOLUTE AGNOSTICISM tremendous an issue, however, is not a sufficient ground for refusing to undertake it. Eather is it a reason for undertaking it, and for conducting it in as earnest and thorough a manner as possible. The right, then, of the sceptic to institute a criti- cism of the conditions of knowledge is not here called in question. On the contrary, to institate it is fully admitted to be a philosophical duty. ISTor is it over- severity or even over-subtility which is held to be the fault of the sceptical criticism of principles. What it is charged with is unfairness, unreasonableness. The most thorough sceptic can no more refuse to proceed from and make use of first principles than the most absolute dogmatist. Let him analyse the men- tal processes and verbal argumentations through which he reaches and justifies his sceptical views and conclusions, and he must inevitably find those princi- ples to have been his own primary assumptions. Hence he is as much bound as other thinkers to be- ware of taking for first principles what are not such. He ought carefully to distinguish them from all that is of an a posteriori, particular, contingent, or infer- ential character in intellection and belief. He should criticise every apparent first principle with a view to determine whether it is not merely apparent, or secondary, or false. He should only accept it as what it seems or is said to be after he has satisfied himself that it really is primary or a priori; that it is self- evident and necessary — not only immediately seen to be true, but what must be true ; and that it is natural and universal — true always and everywhere, true for 274 PEOOF OF FIRST PEINCIPLES all persons and in all cases. So far as he merely does that he is plainly within his right, and only acts as he ought to do. But as plainly he has no right to resist real self- evidence or to reject what are truly first principles. That, however, is precisely what an absolute agnostic never fails to do, and indeed, must do. His whole hypothesis compels him to take up a distinctly an- tagonistic attitude towards first principles. He can- not afford to assent even to self-evidence. Were he to do so he would have no case. He must refuse to acknowledge the reality and validity even of first principles. And that is an obviously wrong at- titude to assume towards them. Primary and self- evident truths, necessary conditions of knowledge, are entitled to be trusted. Mental sanity requires their acceptance. Whoever rejects them, whoever begins with doubt or disbelief of them, starts as an agnostic in order that he may end as one, and so be consistent in absurdity throughout. The absolute agnostic must act thus. His demand for proof of what are truly primitive judgments or first principles is, of course, one which cannot be met, but it is also one which it is irrational to make. They cannot be conclusions of any process of proof seeing that they are the conditions of all proofs and conclusions. " Did not reasoning," said Eoyer-CoUard, " rest upon principles anterior to it- self, analysis would be without end, and synthesis without commencement." As all reasoning supposes knowledge, all knowledge cannot be gained by reason- 275 COMPLETE OE ABSOLUTE AGNOSTICISM ing. In refusing to accept first principles without proof an agnostic acts as foolishly as a man who should insist on being provided with a medium wherewith he may see light, although light is itself the only medium by which anything can be seen. The right attitude of mind, then, towards first principles is that of belief, because of their self-evi- dence. Doubt or disbelief of their truth and validity is a wilful rejection of the light of self -evidence, and begs the question in favour of scepticism. This is all the more manifest inasmuch as agnos- ticism itself has to assume and make use of them in order to vindicate its rejection of them. Wo other- wise can it justify its doubt or disbelief of truth or knowledge. But it thus places itself in a most equiv- ocal and inconsistent position relatively alike to truth, knowledge, and the laws of reason. It need not deny that truth exists. It may or may not admit that there is truth. But it must deny or question that truth can be found, and yet must also claim to be itself true, and truly established. It cannot admit that there is knowledge; for knowl- edge even of a phenomenon is not itself phenomenal, and so-called subjective certainty is mere feeling. Wherever there is knowledge mere feeling and sub- jectivity are transcended. Knowledge implies judg- ment, but the judgment that everything is or is not, or that it is doubtful whether it is or is not, a phe- nomenon as contradistinguished from a reality, is not itself given as a phenomenon. What agnosticism really asks us to accept, therefore, is not the simply 276 REJECTIOlSr OF LAWS OF THOUGHT phenomenal, but a system of judgments regarding phenomena. But that it can only do on the ground of knowledge, notwithstanding its denial of the pos- sibility of knowledge. It must, further, refuse to accept even the necessary laivs of thought as true, or to admit that anything really is what it necessarily appears in thought to be ; for not to do so would be the retractation of all that is distinctive of it. And yet it is only by availing it- self of those laws that it can give any plausibility to its own reasonings. The reasoning of the agnostic is as dependent as the reasoning of other men on the existence and validity of the necessary principles of thought. In setting those principles aside, therefore, he as thoroughly refutes his own conclusions as those of his opponents, — or rather more so, for his oppo- nents do not admit that he is entitled to discard first principles. If he cannot show that he is warranted to do that, his explicit refutation of others is the part of his procedure in which he fails, and his implicit refutation of himself the part of it in which he suc- ceeds. He does not accomplish what he wishes, and does accomplish what he does not wish. It has always been the boast of the absolute philo- sophical sceptic that no opponent can refute him. It is so far true. There can be no direct demonstrative contradiction of a scepticism wbich is content to jus- tify universal doubt simply by the possibility of such doubt. Whatever answers be given to it, whatever reasons be urged against it, must fall under what it questions, seeing that it refuses to acknowledge the 377 COMPLETE OR ABSOLUTE AGNOSTICISM truth of the conditions on which all intelligence and inference depend. All thought must rest on first principles — on truths which have their evidence in themselves, and which, in order to be believed, re- quire only to be apprehended. If a man deny them, you cannot deductively prove them to him, nor can you prove anything to him, for they are the conditions of all rational and sane thinking. If, when you ap- peal to one of those truths, a man, without endeavour- ing to show that it is intrinsically untrue or doubtful, simply says, " I do not choose to admit it," " I find it possible to reject it, and therefore I reject it," there is no further argument possible between you and him in the direct line. But can the agnostic fairly claim this as a triumph ? Assuredly not. It merely means that rather than be considered a bad reasoner, he is willing to accept an absurd premiss ; that, in order to justify an. argument which implies the falsity of a self-evident principle, he will not hesitate to adopt the falsity as a truth. But every alleged logical vic- tory of this kind must be deemed a real rational defeat by every truly reasonable mind. It is the triumph of will over reason, the substitution of will for reason. Assent to first principles is not, as the agnostic would have us suppose, mere belief or blind trust. It is an acceptance of self-evidence, just and rational in itself, and capable of being corroborated by legitimate and adequate criteria. In withholding it, the abso- lute agnostic, the genuine and thorough sceptic, de- mands to be directly refuted, which is absurd, but 278 EELATION TO PEACTICAL LIFE makes no attempt directly to justify himself, although that is greatly needed. What he opposes to self -evi- dence is self-will. What he opposes to intuitive rational insight is intellectual caprice. He decides against reason ab initio without reason. In a vyord, his rejection of the laws of thought is an essentially arbitrary, irrational act. VI. ABSOLUTE AGNOSTICISM AND PEACTICAL LIFE " By their fruits ye shall know them " is an axiom which holds good of propositions and theories as well as of things and persons. All truths tend to good, and all errors to evil. A theory or system which cannot be acted on is one which is greatly to be dis- trusted. How stands it in this respect with agnosti- cism ? Can it be made to harmonise with the require- ments of practical life, or with the nature of man as a being formed for action ? The answer must be in the negative. Agnosticism is not a system which will work. Its relation to practice is unnatural and un- satisfactory, and it is inconsistent with any accepta- ble theory of duty and conduct. Both our physical and moral life have imperative practical requirements with which every consistently and completely agnostic theory, either of doubt or disbelief, must inevitably come into conflict. Man is born to act, and must act on pain of death. In act- ing he comes under obligations which he must fulfil, otherwise conscience will pronounce him deserving of contempt and punishment. With this state of 279 COMPLETE OE ABSOLUTE AGNOSTICISM things, neither absolute disbelief nor absolute doubt can be got to accord. If there be no truth, there can be no moral truth. If reason be untrustworthy, its ethical decisions can have no claim to be trusted. If we have no right to believe, we can have no ground to act. If total suspension of judgment be the proper rule of intelligence, total cessation from action must be the proper rule of will. Here agnosticism seems in presence of an insuperable difficulty, and certainly of one which has never been surmounted. Some have evaded it by saying that man was so self -contradictory a being that this additional contra- diction need not be taken into account, or should be credited to human nature instead of charged against the agnostic representation of that nature. This may pass as a joke, but it cannot be allowed as an argu- ment. Any view of the human intellect which exhib- its it as essentially self-contradictory is already, ipso facto, highly suspicious ; but all suspicions against it receive strong confirmation when it is seen to be in opposition also to the implications of instinct and appetite, of affection and duty, and to be, in fact, such as would paralyse the entire emotive and active nat- ure, from its lowest physical prompting to its highest spiritual aspiration. There are others who have, in substance, said: Adhere to agnosticism as a theory, but do as others do in practice. Conform to common thought and the ordinary modes of life as regards conduct, follow the promptings of nature, listen to what sounds as the voice of duty, while sceptical as to the grounds and 380 NOT EEASONABLE IN PEACTICE worth of human judgments. Doubt or disbelieve all that is received by human beings as true and certain, yet be as prompt as others to decide and as energetic to execute when action is required. That means,- however, that the best thing which an agnostic can do is to act as if agnosticism were not true. And, in fact, the shrewdest and most ingenuous of agnostics have confessed that they did so, and could not help doing so, in regard to the affairs of common life. But why, then, suppose that their theory can be acted on at all ? If they cannot act on it as regards ordinary things, how can they assume that it may be acted on as regards high'er things ? If a theory which pre- tends to be universal will plainly not apply — not work — in one sphere, is it not likely to be equally at fault in others ? Is not the proper inference that it will work nowhere ; that as regards action or conduct it completely breaks down; that it is to be trusted neither as to our lower nor higher life — neither as to this world nor any other ? Yet is it credible that thought should be so related (or unrelated) to action, truth to life ? There are agnostics who have dealt with the diffi- culty in question in still another way. They have entirely separated theory and conduct, so divided rea- son as to destroy its unity, and formed, instead of one homogeneous and harmonious philosophical doctrine, two heterogeneous and discordant ones — the one spec- ulative and sceptical, the other practical and dog- matic. Could this procedure be justified, no further proof would be needed of the constitutional self-con- 281 COMPLETE OE ABSOLUTE AGNOSTICISM tradictoriness of the human intellect. But no evi- dence is to be found for such dualism as is alleged, or warrant for such " double book-keeping " as is adopt- ed. The agnostics referred to have seen that they must sacrifice either their agnosticism to morality, or morality to their agnosticism ; and their reverence for morality has been sufficiently strong to induce them to choose the former course as the lesser evil. They have thus made in favour of morality the greatest sac- rifice vsrhich, as philosophers, they could make — the sacrifice of their philosophical principles and consist- ency. The moralist may commend them in conse- quence, but the approval of the logician cannot be expected. Absolute agnosticism, then, owing to its intrinsic self-contradictoriness, has among other defects that of logically necessitating either a tremendous intellect- ual or a tremendous ethical sacrifice, or both. It must be inconsistent either with reason or duty, or both. The nearer it approaches to absoluteness, or essential universality and completeness, the more cer- tainly will it show itself incompatible with either true science or right practice. As I am at present dealing merely with absolute agnosticism, to have indicated in this general way that it affects the latter as well as the former may be, perhaps, all that is here strictly required. Yet it can hardly be irrelevant also to refer in a few sentences to the agnosticism which is specially direct- ed against knowledge and certitude in morals. There is such an agnosticism. Morality has never 382 AGNOSTICISM AND DUTY had any exceptional immunity from the assaults of sceptical criticism ; nor is it likely ever to have it. The highest truth accessible to the human intellect is just the truth most in danger of being suspected and rejected by it. The impressions of sense find, as a rule, a readier assent than the dictates of conscience. When Kant assumed the moral imperative to be a limit which even a criticism that disregarded every other might be expected to recognise, he made the mistake of judging of others by what he was himself. He credited mankind, that is to say, with such a sense of the sacredness of duty as is possessed only by a few. And he forgot the teaching of experience transmitted to us by history; overlooked the historical fact that the agnosticism which questions the reality of moral distinctions is as old, and has been as prevalent, as that which throws doubt on the existence of external things, or any other form of scepticism. Long before the Christian era there were agnostics who traced all moral beliefs to non-rational causes. The sophists and sceptics of ancient Greece — a Gor- gias and Protagoras, Arcesilaos and Carneades, iEne- sidemus and Agrippa, for example — ^were wont to expatiate on the diversity, conflict, and arbitrariness of those beliefs, and of the customs, laws, and institu- tions to which they had given rise, and on the impossi- bility of finding for them any fixed standard or sure criterion. The same must be said of the succession of sceptical thinkers from Montaigne to Hume. And agnostic attacks on the cognoscibility of aught real and regulative in morality have, perhaps, never been 283 COMPLETE OE ABSOLUTE AGNOSTICISM more numerous and varied than in recent times. In- dividualism, positivism, naturalism, sensationalism, pantheism, pessimism, and anarchism have all been prevalent during the latter half of the nineteenth century, have all shown agnostic tendencies, and have all supplied agnosticism with keenly sceptical assail- ants of the very bases of a real or credible ethics. The history of philosophy leaves us in no doubt at all that agnosticism as to morality is not only possible but may assume many and plausible forms. The idea of duty on which all morality rests, is as capable of being impugned as the idea of God, on which all religion rests. Wherever a real agnosticism finds entrance ethical agnosticism may be expected to follow. A sincere agnosticism must tend towards completeness. Hence it will naturally and necessarily invade and seek to make its own the sphere of morality. And it will be especially difficult to prevent its succeeding if it be to any considerable extent of an anti-religious character, seeing that the connection between religious and mor- al faith, religious and moral character and conduct, is especially close and strong. Agnosticism as to the bases of either religion or morals cannot fail to spread and intensify agnosticism as to those of the other. There is, of course, no species of agnosticism so harm- ful as that which undermines moral principles and weakens and vitiates moral practice. But all agnos- ticism contributes, and anti-religious agnosticism es- pecially, to feed and foster that form of it. Hence all agnosticism, and especially anti-religious agnosticism, 384 ETHICALLY UNSATISFACTORY may fairly be held to tend to the demoralisation of individuals and of societies. For a man like Kant or Fiehte, in whom the voice of conscience sounds clearly as the very voice of God, the moral law may not unnaturally seem as the strong- est and surest, or even as the sole yet sufficient, barrier to sceptical doubt or denial of objective existence. It is not so inexplicable as is commonly supposed that Kant, after he had laboriously sought to show that the speculative use of reason only leads us stage after stage, through its forms, categories, and ideas, deeper and deeper into subjectivity and illusion, could yet fancy that the practical reason was sufficient to secure us a foothold on eternal reality. For by men like Kant the moral law is vividly realised as standing in far closer relationship to their very selves than the outward world does, and at the same time in a far more independent relationship. The world of the senses is to so large an extent what it is owing to the constitution of the senses that it is comparatively easy for them to regard it as wholly a creation of the mind. The moral law, on the other hand, presents itself to them as beneath and beyond sense, indepen- dent of and above them, universal and eternal, immut- able and divine. But will the generality of men, or even of philos- ophers, be so impressed? Experience and history clearly teach us that they will not. Convince them that their faculties are deceptive, and that the objects of sense and the contents of the positive sciences are only subjective appearances and their ideal connec- 285 COMPLETE OR ABSOLUTE AGNOSTICISM tions, and hopeless must it be to try to persuade them that the categorical imperative is an absolute reality and a law binding on all intelligences. Ordinary humanity will only regard moral judgments and be- liefs as on a level, so far as truth and certainty are concerned, with other kinds of judgments and beliefs. Bring men to think that there is no objective truth outside of the region of morals, and, as a rule, what they will conclude is not that there is such truth there, but that there is such truth nowhere ; that so- called moral knowledge must be as deceptive as all else that is called knowledge, and morality itself of no exceptional validity. Agnosticism, then, is ethically as well as intellect- ually unsatisfactory. It cannot be reasonably ex- pected to yield good fruits ; to enlighten and guide practice ; to invigorate, purify, or ennoble life. On the contrary, it tends to weaken and destroy all trust, even in the foundations of virtue and duty, and to produce and diffuse that sort of doubt and disbelief of which the inevitable issues are despair and desola- tion. A soul from which all moral faith has gone is, indeed, a soul that has lost all true good, and is itself a lost soul. " As music and splendour Survive not the lamp and the lute, The heart's echoes render No song when the spirit is mute : — No song but sad dirges Like the wind in a ruined cell, Or the mournful surges That ring the dead seaman's knell." —Shelley's " Adonais." ' 386 CHAPTEE VI ON MITIGATED AND PARTIAL AGNOSTICISM AND THEIR FORMS Absolute agnosticism may be verbally professed, but is not really credible, and cannot be consistently pre- sented or logically defended. A universal suspension of judgment or entire negation of knowledge is not only a false but an unattainable ideal. Its realisation would be the extinction of intelligence. Some de- gree of faith in and knowledge of truth is as necessary to the mind as some measure of breath and air to the body. Reason can no more be sustained and exer- cised in a vacuum than can any of the other powers of life. Hence agnosticism has never been able to present itself in a pure and full form. Absolute ag- nosticism has not attained to actual existence. His- tory shows us only more or less close approximations and more or less ingenious counterfeits of it. All known types or schemes of agnosticism have been either incomplete as to nature or extension or as to both nature and extension ; in other words, all agnos- ticism has been either of what may be called a miti- gated or a partial kind or both mitigated and partial. That fact, however, raises the very important questions. Can agnosticism be either mitigated or lim- ited in a legitimate and satisfactory manner ? Has 387 MITIGATED AND PAKTIAL AGNOSTICISM it ever been so mitigated and limited ? They cannot be here quite passed over. That agnostics themselves so frequently ignore them makes it only the more necessary that non-agnostics should not, especially as any critical survey of the historic forms of agnosti- cism soon shows that both the mitigation and limita- tion have always been fruitless so far as concerned their main object, and that it is vain to endeavour to rationalise the irrational. I. MITIGATED AGNOSTICISM. ITS TJETDEELTIITG ASSUMPTIONS Mitigated agnosticism is invariably scepticism modified by a dogmatism in which agnostics are of all men the least entitled to indulge. Only through a surreptitious commingling of scepticism with dogma- tism can any form of mitigated agnosticism be made to assume an appearance of plausibility. Continuous self-contradiction is accordingly its inevitable and predominant characteristic. That characteristic, in- deed, is what distinguishes it from the consciousness, however vivid, of the necessary imperfection of hu- man knowledge. The latter, a due sense of one's ig- norance, is not only a quite legitimate but a habitual- ly appropriate frame of mind for all mankind. No man knows anything completely — ^knows anything in its whole nature and in all its relations. A perfect knowledge of any object, however simple and small, is only possible on the presupposition of a perfect knowledge of the omne scibile, — of all truth and of 388 ASSUMPTIONS OF MITIGATED AGNOSTICISM all reality, — of God, the universe, and man, — such knowledge as can belong to God alone. The wiser a man is the more likely will he be to feel that he knows so little, and that little so superficially, — that any knowledge he may be credited with is not only noth- ing to boast of but hardly worthy of the name of knowledge. The words of Socrates, " I know noth- ing, except that I know nothing," and those of St. Paul, " If any man think that he knoweth anything, he knoweth nothing yet as he ought to know," bore in them no agnostic or sceptical meaning, but were sim- ply somewhat paradoxical, yet apt and effective ex- pressions, of what Socrates and St. Paul felt as to the littleness and defects of their own knowledge and of all creaturely intelligence. If treated as strictly and speculatively true they are thoroughly sceptical for- mulae, and also thoroughly dogmatic formula. To affirm one's entire nescience, to declare that one knows that one knows nothing, is to attribute to one's self a very marvellous knowledge of a very marvellous ignorance — a kind of omniscient nescience of all one's objects of sense, data of consciousness, beliefs, intui- tions, and inferences. It is to propound in a single sentence an incredible dogmatism and an equally incredible scepticism. An analysis and critical examination of all the forms of mitigated agnosticism which have appeared in the course of the history of philosophy would be required in order completely to prove its mitigation to have been always effected through the illegitimate combination of dogmatism with scepticism, — through 389 MITIGATED AND PAETIAL AGNOSTICISM implicit assumptions of the attainment of knowledge in order to justify explicit doubts or denials of its attainability. But manifestly I must be content with much less than full proof. I shall merely try to in- dicate how my readers may obtain such proof for themselves. Both the merits and defects of agnosticism, of course, show themselves in its history. But it is only with its defects that we are now concerned, and in- deed only with such defects as are so inherent in it and characteristic of it as to appear in every stage of its history. Although those defects, however, may be found wherever agnosticism is to be found, it seems desirable to seek and take note of them as near to the rise of agnosticism as possible. But it was in Greece that agnosticism, under the name of scepticism, first appeared in distinct forms. In the oriental world it was only enveloped and involved in ontological and theological creeds. Let us turn our eyes therefore to ancient Greek scepticism. In its oldest forms we may easily trace all the root-errors of the most mod- ern English, French, and German agnosticism. The doubts and questionings of the Pyrrhonians, as the earliest Greek sectarians of scepticism were called, seem to have been bold, radical, and wide- reaching. Yet their teaching was largely modified by manifestly dogmatic assumptions, and largely de- pendent on them for what plausibility it possessed. This can be easily shown by a brief and summary statement of what they were. 1. The Pyfthonians, then, did not doubt or dis- 290 PYERHOmSM believe that human life had a chief end; that that end could be known; that they themselves knew what it was ; and that they also knew how it was to be attained. On the contrary, they thought and acted as if those four closely connected yet distinguishable assumptions were positive and reliable facts. It was owing to their faith in them that they advised their contemporaries not to trouble themselves in the vain search of what they held to be unattainable, — knowl- edge and truth. Pyrrhonism was professedly a prac- tical philosophy, — one which undertook to guide men to the possession of the chief good, the highest satis- faction of their nature. Yet it was also a reasoned refusal to allow that knowledge was attainable. The self-contradiction is obvious. The assumptive and positive portion of Pyrrhonian teaching was clearly inconsistent with the sceptical and negative portion of it, and with the maintenance of a philosophy of doubt or nescience. How could, how did, such self- contradiction originate? Largely at least from a crude and erroneous belief that knowing and doing, true thought and right practice, are separable in a way and to an extent altogether incompatible with the spiritual unity of the human mind and of human life. The mind is indivisible into contrasted or un- connected departments, and its life is a process in which all its energies and activities are combined with a view to co-operation. Knowing is itself a kind of doing. The doing which is without knowing is auto- matic, mechanical, or instinctive action, not properly human action. Intellectual activity is sustained by 291 MITIGATED AND PAKTIAL AGNOSTICISM volitional energy, and volitional energy is guided by intellectual illumination. Knowledge, as Bacon says, is power. ISTeither physical nor moral ends can be attained when causes, conditions, and laws are ig- nored. What is man's chief good is itself a question for enlightened reason to answer, and even a difHcult question which admits of and has received many and conflicting answers. If we do not know what is true how can we know what is good, and still more what is best ? If all else be doubtful, what right can we have to assume anything ethical to be certain ? The same kind of assumptiveness and self-contra- dictoriness which has thus been referred to as charac- teristic of Pyrrhonism has constantly reappeared in the subsequent history of scepticism. The scepti- cism of the Academics and Empiricists of ancien"fc Greece is marked by a similarly unnatural severance of knowledge and practice as that of the Pyrrhonians, although the Academics introduced probability and the Empiricists experience with a view to bridge over the chasm and to present some appearance of rational basis for conduct. The ISTeo-Sceptics of Greece, on the other hand, preferred to build on the original Pyrrhonian basis. The majority of the avowed scep- tics of modern times to whom I have referred in chap, iii. were generally called Pyrrhonians, and did not regard themselves as wronged by being so called. Ritschlian divines separate religion from knowledge in much the same way as Pyrrhonian sceptics sepa- rated the conduct of life from knowledge. Their rep- resentation of religion as dependent only on judg- 393 INCONSISTENCY OF PYREHONISM ments of value, and independent of any knowledge of objective reality or of relationships which can be ex- pressed in existential or theoretical judgments, is assuredly Pyrrhonianism in theology. 2. There were other assumptions involved in the Pyrrhonian demand for suspension of judgment. Por instance, Pyrrhonians did not doubt of knowing phe- nomena, but held that they knew only phenomena. Nor did they doubt of knowing realities, but denied that they knew them. Neither as to things in them- selves nor as to appearances of things was their atti- tude of mind one of mere suspension of judgment or of pure doubt. On the contrary, as to the former, it was one of negation, denial of the knowledge or knowability of things themselves ; and as to the latter, one of affirmation, of belief that appearances only are known. Thus the Pyrrhonian doubt had reference merely to the existence and nature of things in them- selves, of realities which do not appear. But on what did such doubt itself rest ? Was it on either a scepti- cal or a rational judgment ? Manifestly not, but on the dogmatic and absurd assumption that realities and phenomena, things and appearances of things, were entirely distinct, absolutely separate, and known by Pyrrhonians themselves to be so. If things in them- selves are things which appear, there can be no more reason for doubting of things in themselves than for doubting of things which appear. And if there be no things in themselves, none which do not or may not appear, doubt as to so-called " things in themselves " must be doubt about nothing at all— objectless, mo- 393 MITIGATED AND PAETIAL AGKOSTICISM tiveless, reasonless doubt. To doubt of realities while believing in phenomena assumes a distinction between them, and enough of knowledge to draw the distinc- tion. There cannot be intelligent or even intelli- gible doubt about things altogether unintelligible, such as the Pyrrhonians pronounced things in them- selves to be. It was not Pyrrhonians, or sceptics of any kind, who first represented the distinction between reality and appearance, being and becoming, the noumenal and phenomenal, as an absolute one. Like all that is distinctive of scepticism, it sprang from the exclu- siveness and exaggerations of dogmatism. In Greece it was the conflict between the Eleatics and Heracli- teans which brought it into prominence. Plato gives it a large place in his teaching, and threw such a glory and a charm over it as to secure for it a remarkable history even far beyond the confines of scepticism. The sceptics have had only to adopt and apply it in a special way. They have done so with the most in- structive unanimity. There is, perhaps, no form of developed scepticism which does not depend on the distinction in question as one of its chief supports. It is one of the main pillars of Kantianism, and of all post-Kantian agnostic, theories. Even agnostics, in- deed, seem now too ashamed of it to venture to empha- sise or formulate it ; but they have not had the courage to discard it, or been able to show that they can dispense with it. Their sceptical doubts and de- nials still depend on it, and presuppose its intelligi- bility and accuracy. Mr. Alfred Sidgwick, the most 394 KNOWLEDGE AND OPINION philosophical representative of scepticism in England, holds in the present day that reality cannot be known, just owing to his distinguishing reality from appear- ance in the preposterous way which Pyrrho did in the age of Alexander the Great. And such a distinc- tion ! The distinction between Eeality which does not and cannot appear, and Appearance in which nothing really appears. How can any one reasonably believe either in such Reality or in such Appearance ? It would seem as if agnostics must believe in both. 3. The separation and contrast of reality and ap- pearance naturally implied another separation and contrast. Granted that reality and appearance were so apart and unlike in themselves, they could not be united in or alike to the mind — could not, for instance, be equally objects of knowledge. If realities only be known, appearances must be below knowledge ; and if appearances only be known, realities must be above knowledge. Accordingly, the Greek dogmatists, who dissevered and contrasted reality and appearance, noumenon and phenomenon, gave to the words knowl- edge and belief (eVtcTTJj/M/ and Bo^d) the significations required to express the correspondent mental states. In other words, they termed knowledge only what they held to be apprehended by pure reason and demonstratively certain, and called opinion all that presented itself to sense, and was consequently viewed by them as in contact merely with semblance or illu- sion. The sciences which are now called positive, and which are so often spoken of as the only sciences, Plato and the speculative philosophers of antiquity 395 MITIGATED AND PARTIAL AGNOSTICISM did not regard as worthy of the name of science. They held all sense-perceptions and ordinary judg- ments to be essentially different from true cognition, and relegated them to the limbo of mere opinion. The sceptics accepted the same distinction between knowledge and opinion, but they made another appli- cation of it, and drew from it an opposite inference. They concluded that knowledge was unattainable; that truth, if there be such a thing, must be beyond the reach of the human mind; and that men should be content to do without them, making the most of such substitutes for them as appearances, probabili- ties, and experiences, and seeking only to gain prac- tical ends. The Greek sceptics, however, who, from Pyrrho to Sextus Empiricus, represented knowledge as beyond human reach, either did not define knowledge in any reasonable way or assumed that there was no knowl- edge short of absolute knowledge, and no valid proof of any kind unless there was some one perfectly clear and unquestionable criterion of truth. Modern scep- tics have proceeded in the same way, but it is a mis- leading one. Men may have true knowledge without being infallible. It is easy to show that our senses are often at fault. Their illusions and the fallacies of inference associated with them are innumerable. Hence one of the arguments on which sceptics have placed the greatest reliance. Yet all the errors and contradictions which can be fairly charged on the judgments of sense are very far from disproving that all our senses yield us a large amount of real knowl- 396 PYEEHONIAN USE OF REASON edge. The inference to the contrary drawn from their defects and errors is excessive and fallacious. The illusions and contradictions adduced are excep- tional; and, further, they are explicable, and so ex- plicable as to cease to have any argumentative value against the existence of truth and the reality of knowledge whenever they are naturally accounted for. If we can discover the causes of either our erroneous perceptions or inferences, the scepticism which has based itself on those perceptions or inferences is left without foundation and must fall. Their causes al- ways can be discovered. All that the sceptical argu- mentation referred to really proves is that the search for truth is a serious affair, one which requires exer- tion, circumspection, and method. 4. Pyrrhonism also assumed that there was in man a reason capable of weighing reasonings regarding things, and of determining what weight ought to be assigned to them. Pyrrho himself, in order that he might overtly deny that man had such a reason, re- quired to reason as if he had it, and thus also to miti- gate his open scepticism with secret dogmatism. The assumption was manifestly implied in his argument that he could neither legitimately affirm nor deny the reality of motion because the reasoning of Parmenides that there is no motion, and the reasoning of Heracli- tus that all is motion, being of equal but contrary weight, balance and annul each other. In order to be entitled so to infer that the two opposite views were supported with reasons of equal weight and worth he must have had a power competent to weigh and appre- 397 MITIGATED AND PAETIAL AGNOSTICISM ciate reasons aright. The assertion that reasoning yields contradictory conclusions which are supported by proofs of the same cogency in reality presupposes its veracity and validity, although meant to discredit it. Further, it is an assertion which ought not to be dogmatically affirmed, but which requires to be justified in each and every instance. A universal conclusion cannot be rationally inferred from a par- ticular case. And there is obviously a special and tremendous improbability in supposing that reason, the general validity of which is implied in all rea- soning, will uniformly proceed to contradict and stulify itself in particulars. That reason thus contradicts and stultifies itself the agnostic has often asserted but never proved. Pyrrho obviously did no more than give the assertion a kind of plausibility by confounding the contradic- tions of one-sided and reckless reasoners with the con- tradictions of reason itself. He had no right to infer because Parmenides had argued that there was no motion and Heraclitus that all was motion, and the arguments of the one seemed to him to be just as good or just as bad as the arguments of the other, that reason necessarily falls into self-contradiction when applied to investigate the nature of motion. Grant that the opposing arguments of Parmenides and Her- aclitus are equal, and all that can be fairly deduced is that Parmenides and Heraclitus contradict each other. To conclude that therefore reason contradicts itself is a leap of logic quite unwarranted. The more natural view is that both Parmenides and Heraclitus 398 BOTH SCEPTICISM AND DOGMATISM have erred ; that they have proceeded from inadequate or false conceptions of motion; that their respective findings, " there is no motion " and " all is motion," are alike extravagant; that we should be content to affirm that " there is some motion," so that the percep- tion of motion is not a mere perception without ob- ject, but, under normal conditions, a real perception of an object — i.e., the perception of a real object. If this view be correct, reason must be held to be consist- ent both with itself and with experience, where the sceptic most confidently ascribes to it self-contradic- tion and unconformity with experience. Arcesilaos and Carneades, I must add, reasoned in the same way and with the same intent as Pyrrho. ^nesidemus and Agrippa placed the argument from the contrariety of judgments among the so-called scep- tical tropes. Montaigne, Le Vayer, and Bayle made constant use of it. It reappeared in Kant's doctrine of antinomies ; and it is very conspicuous in the agnos- ticism of Hamilton, Mansel, and in various other nineteenth-century forms of scepticism. 5. I shall mention yet another dogmatic assump- tion in Pyrrhonism — namely, the assumption that such doubt as it inculcated would free men from the cares and fears of life, and secure them mental tran- quillity. What warrant was there for that assump- tion ? None, so far as either reason or experience shows. The great mass of our cares and fears, our pains and sorrows, have their sources not in things in themselves, but in what things are or may be to us; not in so-called realities, unknovni and unknowable 399 MITIGATED AND PAETIAL AGNOSTICISM through experience, but in such as do or may appear in the actual or possible phenomena of experience. "Who troubles himself about fire and water in them- selves ? Yet how troublesome may be the fire which bums and the water which drowns ? The preceding observations on Pyrrhonism may suffice to show that, so far from being pure, com- plete, absolute scepticism, it was very largely indeed a scepticism dependent on and made up of dogma- tism; a system mitigated or modified through the mixture of sceptical with dogmatical elements, and consequently one composed of incongruous and dis- cordant elements. An analysis of most other forms of scepticism would show them to be of the same char- acter; not less full of dogmatic assumptions, nor less self-contradictory and untrue. I must leave, however, my readers to institute for themselves any further analysis of the kind which they may deem necessary. It will now, I hope, be enough for me at this point to consider how Hume has treated the question of the relation of mitigated to absolute agnosticism. II. HUME OI^ MITIGATED AND ABSOLUTE SCEPTICISM Hume, in the essay entitled " Of the Academical or Sceptical Philosophy," has clearly defined his atti- tude both to absolute and mitigated agnosticism by professing himself to be not a Pyrrhonian but an Academic sceptic. Pyrrhonism was the term he em- ployed to denote absolute scepticism. It had often been so used before, and has not infrequently been so 300 HUME ON PYREHONISM used since. For such use of it there is, however, no proper historical warrant. Pyrrhonism, as I have already shown in this chapter, was not absolute scep- ticism. There is no evidence even of its having been a nearer approximation to such scepticism than Aca- demic scepticism was. On the contrary, the docu- mentary testimony seems to prove that the scepticism of the Pyrrhonists was much less radical and com- plete than that of the Academics. M. Brochard has very plausibly, and perhaps justly, maintained that the so-called " Pyrrhonian suspension of judgment " was not taught by Pyrrho, but appropriated by those who called themselves his disciples, from Arcesilaos or Carneades, who undoubtedly inculcated such sus- pension of judgment as to knowledge. The great concern of Pyrrho was that men should live conform- ably to the chief end of life, and his scepticism seems to have had its source mainly in his aversion to spec- ulation and sophistry as incompatible with such a life. In the essay " Of the Academical or Sceptical Phi- losophy," Hume indicates the grounds on which the absolute sceptic challenges the worth of belief in the existence of an external world, in the certainty of mathematical demonstration, and in moral evidence, and implies throughout that no rational refutation of them is to be found. At the same time he admits that although absolute scepticism cannot be refuted, it will not, and should not, be accepted. " Its prin- ciples may flourish and triumph in the schools, but they' must vanish like smoke in real life." " A Pyrrhonian," he says, " cannot expect that his 301 MITIGATED AND PARTIAL AGNOSTICISM philosophy will have any constant influence on the mind ; or, if it had, that its influence would be bene- ficial to society. On the contrary, he must acknowl- edge, if he will acknowledge anything, that all human life must perish, were his principles universally and steadily to prevail. All discourse, all action, would immediately cease, and men remain in a total leth- argy till the necessities of nature, unsatisfied, put an end to their miserable existence. It is true so fatal an event is very little to be dreaded. Xature is always too strong for principle. And though a P;5Trhonian may throw himself or others into a mo- mentary amazement and confusion by his profound reasoning, the first and most trivial event in life will put to flight all his doubts and scruples, and leave him the same in every point of action and speculation with the philosophers of every other sect, or with those who never concerned themselves in any philo- sophical researches. When he wakens from his dream he will be the first to join in the laugh against him- self, and to confess that all his objections are mere amusement, and can have no other tendency than to show the whimsical condition of mankind, who must act and reason and believe ; though they are not able, by their most diligent inquiry, to satisfy themselves concerning the foundation of these operations, or to remove the objections which may be raised against them." Absolute scepticism, then, according to Hiime, is excessive, and can be in itself neither durable nor useful. It may, however, he thinks, in part give rise 302 HUME ON MITIGATED SCEPTICISM to two very desirable species of mitigated scepticism — the first being a degree of doubt and caution and modesty in all kinds of scrutiny and decision, and the second being the limitation of our inquiries in such respects as are best adapted to the natural capacity of the human understanding. A tincture of universal scepticism — a certain sense of the universal perplex- ity and confusion inherent in human nature — may, he considers, be serviceable in abating the pride and obstinacy and self-confidence of dogmatists, and in inducing men to avoid all distant and high inquiries, and to confine their judgments to common life, and to such objects as fall under daily practice and expe- rience. " Those who have once," Hume says, " been thor- oughly convinced of the force of the Pyrrhonian doubt, and of the impossibility that anything but the strong power of natural instinct could free us from it, will never be tempted to go beyond common life, so long as they consider the imperfection of those facul- ties which they employ, their narrow reach, and their inaccurate operations." Again, he asks, " While we cannot give a satisfactory reason why we believe, after a thousand experiments, that a stone will fall or fire burn, can we ever satisfy ourselves concerning any determination which we may form with regard to the origin of worlds, and the structure of nature, from and to eternity ? " ISTo reasoning, he then argues, except abstract rea- soning, concerning quantity and number, and experi- mental reasoning, concerning matter of fact and 303 MITIGATED AND PARTIAL AGNOSTICISM existence, cau contain anything but sophistry and illusion. Hume's own scepticism, then, is professedly a mit- igated scepticism, but one which is so far founded on the absolute scepticism which he rejects as excessive. His rejection of absolute scepticism is not rested on reason, but on instinct, common-sense, practical in- credibility. The absolute sceptic is held by him to have reason, so far as can be made out, on his side. Hume had, in other words, according to his owu ex- plicit confession and declaration, nothing to urge against what he calls excessive scepticism but an in- stinct which he alleges can be proved to be irrational, and the evil consequences which would flow from ad- mitting as true what he holds cannot be shown to be false. If Dr. Thomas Eeid — ^his most effective Scottish opponent — ^had merely appealed in refutation of such scepticism to blind instinct or to common-sense in its vulgarest and not in its philosophical acceptation, he would have met it in the only way in which Hume met it, or professed to think it could be met. Of course, Eeid was not content so to meet it. He did not believe that any of the original instincts or origi- nal principles of human nature could be shown to be contrary to reason. He held, and tried to prove, that it was only by false reasonings that reason could be represented as contradicting either itself or instinct. He may or may not have successfully maintained his position ; but surely the position itself is incompara- bly superior to that of Hume, who holds absolute ag- 304 HUME'S OWN SCEPTICISM nosticism alone reasonable, and yet quite incredible — ^who acknowledges that were he faithfully to follow reason he must be an absolute agnostic, yet that, in order not to be ludicrous, he must yield to a blind in- stinct, or, in other words, prefer unreason to reason. "When a philosopher tells us that the state of man is a condition thus " whimsical," we ought not readily to admit that he is entitled to speak for any one except himself. He, owing to his agnosticism, may be in that condition, but the " whimsicality " of his situa- tion may be entirely due to the irrationality of his ag- nosticism. That agnosticism may be a dream, and he may only have to awaken from it to find himself in a world of light and order, where sound reason is never at variance with healthy instinct. It is further to be observed that absolute scepticism, according to Hume, would, if accepted, put an end to all discourse and to all action. In his opinion, if the sceptic were to follow his reasonings to their legit- imate conclusion, and then seriously to adopt that conclusion, he would soon perish. In other words, he held a view directly opposed to that of those who maintain that even if scepticism were to justify its doubts and negations, and to get the validity of its arguments acknowledged, ordinary life would be quite unaffected. He did not think that a merely phenom- enal world would have the same influence as a real world on any one who believed it to be merely phe- nomenal; he thought it could only have the same influence on those who were not thoroughly awakened out of their dream that it was real. Those who sup- pose that they hold his doctrine, and yet censure in 305 MITIGATED AND PARTIAL AGNOSTICISM his opponents an opinion which he so explicitly held, should find in this matter for reflection. Are they not meaning by phenomena realities ? Is their phenomenal sun, for example, simply a mental im- pression, or a group of such impressions, or an idea derived therefrom, or is it not a real body some ninety- four millions of miles away from them, and from every impression which it is possible for them to have ? Hume was not so unwise as to fancy that in the view of a consistent agnosticism the mind can get a hairsbreadth beyond itself. He knew that it must deny the objectivity of space, the validity of causality, the reality of substances, and that these external phe- nomena could not be reasonably held to be the equiva- lents of realities, but only of illusions. Hume represents absolute scepticism as logically and legitimately leading to the mitigated scepticism which he recommends. But in that he obviously errs. Absolute scepticism neither acknowledges nor con- tains nor yields any measure. Mitigation and limi- tation are contrary to its nature ; it can only be mitigated and limited by being so far successfully refuted. How can a sense of the universal perplexity and confusion inherent in human nature produce merely care and caution and modesty in reasoning? Why, if the conviction involved in that sense be cor- rect, no care, caution, or modesty in reasoning can in the least secure that reasoning will reach truth. Rea- son, according to the absolute sceptic, must necessarily fail to attain knowledge; and, according to Hume himself, must, even when exercised faultlessly and to 306 HUME'S AGNOSTICISM ABSOLUTE the full, lead to conclusions which can neither be be- lieved nor acted on. With true criticism, modesty and moderation, caution and carefulness, must ever be closely allied ; but they have no natural connection with the scepticism which teaches that reason is essen- tially unreasonable, and that the whole constitution and condition of mankind are essentially absurd. ISTor can absolute scepticism logically warrant the limitation of reason to any particular sphere. Indeed it cannot, perhaps, warrant any conclusion, as it im- plies the worthlessness of logic ; but if any conclusion may be inferred from it, it must be not the propriety of limiting but of wholly suppressing reasoning and research. If our rational faculty is essentially inca- pable of attaining truth, it will not do to say that we must not in the exercise of it go beyond common life. What we must say is, let us not employ it at all. The assumption that reason is valid in any sphere implies that it is not essentially incapable of attaining truth, and logically forbids our excluding it from any sphere, until we have proved it powerless within that sphere. And such proof must be furnished by reason itself acting in accordance with its own constitutional laws. We have no right, so far as reason and philosophy are concerned, to discourage curiosity and research in any direction; they must be free to turn to any question. We have a right only to insist on thor- oughly testing their reports. Things remote from us are often more easily answered than those which are close to us. It is often only in things very far away 307 MITIGATED AND PAKTIAL AGNOSTICISM that we find the explanation of things near at hand. We know that a stone falls and at what rate it falls, but not why it falls, — we know, that is, the fact of gravitation and its law, but not its cause; and long before we know the why or cause of gravitation, fa- miliar although it be as a fact and certain as We are of its law, we may have a scientific proof that the present physical constitution of things had an origin at an approximately assignable date. Indeed, some of the most eminent scientists of Europe hold that they have already found in Fourier's theory of heat a basis for a strictly scientific inference as to the origin of worlds, the very question which Hume thought it especially hopeless to discuss. Absolute agnosticism, then, does not lead to a miti- gated agnosticism such as Hume professed and recom- mended to others. His agnosticism, however, logi- cally emerges and issues at all points into absolute agnosticism. To admit that reason is on the side of absolute agnosticism is to admit that so long as you follow reason only — that whenever you allow yourself to yield to the guidance of reason without bias or caprice — you are bound to be an absolute agnostic. It is to grant that whenever you have the sincerity and courage to philosophise with freedom and thor- oughness, you will not mitigate, modify, or limit your agnosticism. And it must be said, I think, of Hume, that in his philosophising on fundamental questions he was thus true to himself by being thoroughly ag- nostic. The agnosticism at which he arrived implies (as I have endeavoured to show in chap, iii.) that all 308 VAEIETIES OF AGNOSTICISM that seems knowledge of existence is not really so; that belief is not essentially distinct from imagina- tion; that substances are reducible to collections of ideas, time and space to subjective conceptions, the causal connection to habitual association ; reason to custom ; that science has no principles, and religion no satisfactory grounds. The agnosticism which goes thus far ought to go farther. Any mitigating ele- ment which may be claimed to be in it has obviously no right to be there, and will but slightly alter its general character or affect its general influence. Virtually and implicitly such agnosticism is absolute. III. PARTIAL OE LIMITED SCEPTICISM: ITS POEMS AND THEIE INTER-EELATIOITS Partial or limited agnosticism — agnosticism in- complete as regards extension — is more prevalent than either absolute or mitigated agnosticism. Like mitigated agnosticism it always shows itself incapable of justifying its own incompleteness. The arguments which it employs against the species of knowledge and certitude that it rejects are as applicable to the spe- cies that it accepts. All its weapons may be turned against itself. It never clears itself of self-contra- diction. There are various forms of partial or limited agnos- ticism, and they may be distributed or classified in more ways than one. I must distribute them with a view to the work I have in hand, a treatment of agnos- ticism in relation to religion. It is with the agnosti- 309 MITIGATED AND PARTIAL AGFOSTICIS:\I cism which directly refers to religion that I have mainly to do; it is it which I must throughout keep in view. And yet it is impossible, and were it possi- ble it would be unwise, to deal with it exclusively, seeing that the agnosticism which has no special refer- ence to religion has in all its forms and varieties a general and indirect reference to the agnosticism di- rectly occupied with religion. Hence I distribute partial or limited agnosticism into: (A) agnosticism which has a special reference to religion, and (B) agnosticism which has not such a reference. A. The agnosticism which has a special reference to religion is of two kinds. It is either — 1. An agnosticism which opposes religion and seeks to discredit and destroy it, — anti-relig- ious (anti-theological) agnosticism; or 2. An agnosticism which aims at the support and defence of religion, — religious (theologi- cal) agnosticism. B. The agnosticism which has no special reference to religion may be subdivided thus : — 1. The agnosticism which originates in over- hasty and ambitious theorising, and is insep- arable from the systems of speculation to which such theorising gives rise. 2. The agnosticism which displays itself in given departments of knowledge or regions of inquiry. 3. Agnosticism as to particular powers of mind or principles and conditions of thought. 310 CLASSIFICATION OF AGNOSTICISM And 4. Agnosticism as to the ultimate objects of knowledge. A. Agnosticism must not be supposed to have nec- essarily any special reference to religion. It may have no more a special reference to religion than to various other things, and may have a special reference to other things when it has none to religion. It may be neither religious nor anti-religious, theological nor anti-theological. Still less is it to be assumed that the agnosticism which is specially related to religion can only be antagonistically related to it. To iden- tify agnostics with atheists or anti-theists, or to repre- sent them as irreligious and impious, is to misrepre- sent and calumniate a large section of them. Many persons who may justly be called agnostics have a right to be regarded as sincere believers in God, and even as convinced and earnest Christians. Agnosti- cism has been often employed honestly and zealously for the defence of theistic and Christian faith. It has been so employed both by philosophers and theo- logians, both by Catholics and Protestants. !No species of agnosticism, however, is unrelated to its genus. No agnosticism with a special reference or limited sphere is without reference to the agnostic idea, spirit, and aim. On the contrary, every kind of agnosticism tends towards agnostic completeness. Agnosticism in any form is of the nature of agnos- ticism in every form, and whether in peace or at war with other forms is certain of contributing to the diffusion of the agnostic spirit and the strength of the agnostic movement. Agnosticism cannot be got 311 MITIGATED AND PAETIAL AGNOSTICISM rid of by the help of agnosticism. Science, philoso- phy, and religion are all sure to suffer when they enter into alliance with agnosticism of any kind. There is a religious and an anti-religious agnosti- cism, but both are hurtful to religion: the former not less so than the latter. Religious agnosticism has had among its advocates men of ardent piety, of per- suasive eloquence, and of remarkable dialectical subtlety, who have thought that they could make scepticism the shield and sword of religion. Accord- ingly, they have striven zealously to discredit human reason and secular knowledge, and represented those who could not recognise the wisdom of so doing as rationalists and irreligious. But their labour has been in vain ; they.have never succeeded in justifying their procedure at the bar of reason; and experience and history certify that although attempts of the kind referred to may have a brief notoriety, their failure is sure soon to become evident. The alliance of scep- ticism as to reason and science with dogmatism as to faith and religion, is thoroughly unnatural and irra- tional ; and it is not religion, or even religious scepti- cism, but anti-religious scepticism, or scepticism pure and simple, which always is and must be the chief gainer by it. Ardent religious agnostics have not infrequently become ardent anti-religious agnostics. They have made many more unbelievers than believ- ers in religion. Religious agnostics try to further the cause of re- ligion by labouring to discredit reason with reason- ings which can have no validity unless reason is trust- 313 EELIGIOUS, ANTI-EELIGIOUS AGNOSTICISM worthy. Anti-religious agnostics are, perhaps, less manifestly inconsistent, but they can only give any semblance of plausibility to their scepticism as to the attainability of religious truth by the employment of arguments which do not tell against religious truth alone — arguments of which the conclusions cannot be reasonably confined within the sphere of religion. Their reasonings are mostly as applicable against what anti-religious agnostics themselves accept as gen- uine knowledge and strict science as against the relig- ious knowledge and theological science which they declare to be delusion and pseudo-science. They are mostly, in fact, substantially the same reasonings which have been employed by agnostics for more than two thousand years against every, or almost every, species of knowledge. They are arguments, that is to say, of which the conclusion, were they applied with- out prejudice or partiality, would be absolute agnosti- cism. B. There are, however, as I have indicated, many . forms of agnosticism which have no special reference 'to religion ; which are neither directly favourable nor directly hostile to it ; neither specifically religious nor anti-religious, theological nor anti-theological. But although they have not a special reference to religion they have a general one, although not a direct an in- direct one. And they have all the same sort of gen- eral and indirect reference to it ; all affect it in the main in the same way ; are all on the whole unfavour- able and injurious to it. Religion should be wholly true, and can only be profited by what is true, whereas 313 MITIGATED AND PAETIAL AGNOSTICISM agnosticism as such, however much truth may be con- joined with it in particular minds or systems, is an " ism " which is not true, and cannot benefit religion. That is a fact which cannot be too thoroughly realised. 1. One class of the forms of agnosticism which have no special reference to religion originates in faulty philosophising. The varieties of agnosticism within it are corol- laries or complements of all the narrow, extreme, over-ambitious speculative theories which pretend to explain the universe of being and becoming with inadequate means and in inappropriate ways. Such theories naturally lead to agnostic conclusions as to the grounds of religion. When philosophy and relig- ion are both of a comprehensive, reasonable, and self- consistent character, there can be no conflict, there can only be harmony and mutual helpfulness between them. But when a philosophy has none of these qualities, it is most likely to come into collision with religion, and to take up an antagonistic attitude to it. A philosophy which maintains that knowledge is only of things we see, and that matter is the one sole ultimate reality, cannot logically concede that there is religious knowledge or spiritual reality properly so called. To a consistent materialist religion can- not fail to seem an illusion, and theology merely a kind of agnosticism. His philosophical theory must have an anti-theological agnostic supplement. I do not infer from this that every materialist must be an atheist (an anti-theistic agnostic). I am quite aware that there have been theistic and even Christian mate- ■ 314 AGNOSTIC PHILOSOPHY ANTI-EELIGIOUS rialists, of the sincerity of whose religious faith and the genuineness of whose piety fair-minded men could have no doubt. I am quite willing to grant that Dr. Priestley was a better Christian than Bishop Horsley, who enlisted against him " the bad passions of men, and the cruel prejudices of party." To ques- tion, however, the consistency of a man's thinking is one thing, and to deny his sincerity or piety is an- other. Philosophers like other men, and materialists like other philosophers, may lapse into what are called " happy inconsistencies." But " happy inconsisten- cies " are always exceptional cases. As a rule, a materialistic philosophy will not arrive at spiritual- istic or religious conclusions ; on the contrary, it will almost always be found associated with an anti-theo- logical agnosticism. It was so in ancient times, and is so now. During the last half century we have seen materialism and agnosticism closely conjoined in ac- tive hostility to religion in every European country. The philosophical theories known as sensism, em- piricism, phenomenalism, and positivism are akin to materialism, although distinguishable from it, and like materialism they all lead to varieties of scepti- cism of a nature conformed to their own. There is also subjectivism, a subjective or idealistic scepticism, just the opposite of materialistic scepticism, but not less antagonistically agnostic towards religious truth. Its full logical outcome is the form of scepticism known as solipsism. All these theories are agnostic, and also anti-religious and anti-theological in ten- dency. 315 MITIGATED AND PAKTIAL AGNOSTICISM 2. A second class of the forms of agnosticism which have not a special or direct but only a general and indirect reference to religion contains those which are associated with particular departments of inquiry. There is no science which may not be, or which even has not been, subjected to sceptical criticism and declared unworthy of the name of knowledge. (a) The opinion that mathematics at least has been unchallenged is a vulgar error. The logical perplex- ities involved in its fundamental conceptions had oc- cupied the thoughts of some of the Greek philosophers even before the days of Pyrrho ; and there is no rea- son, so far as I am aware, for supposing that any of the Greek sceptics considered it entitled to immunity from their attacks. Sextus Empiricus was probably a generally accurate representative of their views when, in his Pyrrhonic Institutes, he questioned the very possibility of demonstration (bk. ii. c. 13), and dwelt at length on the difficulties implied in the very ideas of motion, magnitude, addition and subtraction, whole and part, continuance, change, place, time, and number (bk. iii. c. 7-18) ; as also when, in his treatise Against Mathematicians , he disputed the certainty of geometry (bk. iii.), of arithmetic (bk. iv. ), and of astronomy (bk. v.). Doubts and difficulties of a kind similar to those urged by Sextus — epistemological and metaphysical doubts and difficulties — are not even now all solved or eliminated ; nor are they likely soon to be. Metageometry has quite recently been bringing mathematicians face to face with previously unsuspected doubts and mysteries which suggest that 316 MATHEMATICS AND AGNOSTICISM the claims even of their science may be assailed not merely from its under or empirical side but also from its upper or speculative side. In a word, absolute knowledge, absolute certainty, in a strictly absolute sense of the terms, may be argued without absurdity to be even in mathematics beyond human attainment, and the mathematical sciences themselves to be sur- rounded with nescience and dependent on supposi- tions which involve metaphysical propositions. That conclusion, however, will not warrant math- ematical scepticism. It means no more than that a finite intelligence cannot be an infinite intelligence, and that only to the latter can there be "no dark- ness at all." The mathematician may safely rest content with logical demonstration from propositions self-evident to him, and with such certainty as such demonstration gives, until he is shown that in math- ematical processes the axioms are not self-evident or the inferences logical, or that there are counter-axioms as evident or counter-inferences as valid as those which he has accepted, or that there are at least clear and weighty positive ab extra reasons for suspecting the rationality and certainty of his science. Were mathematical agnosticism, however, thus vindicated, reason itself would be so discredited that it could not be trusted in the religious or in any other sphere. Reason would be proved to be rooted in unreason. But mathematical agnosticism cannot be thus estab- lished without anti-theological agnosticism also being established. The trustworthiness of reason is im- plied in the knowledge of God, and so as much a pre- 317 MITIGATED AND PARTIAL AGNOSTICISM supposition of theology as of mathematics, while the trustworthiness of God is a guarantee of the trust- worthiness of reason in all its normal processes and legitimate acquisitions. (b) Scepticism as to the possibility of physical sci- ence was prevalent in the classical and in the medie- val world. Grecian sages generally looked with contempt on the facts with which such science deals, and saw in them no signs of law or order. Plato relegated the whole world of sense to the limbo of mere opinion, and denied that there was any science of phenomena, or that science could be reached through the study of phenomena. The scholastic divines were no wiser. Theology was so dominant in the Middle Ages that, while unlimited trust was given to all Biblical refer- ences to physical things, little interest or confidence was felt in their direct study. Hence the views of the men of those times on the subjects of which the positive sciences treat were very stranga and erro- neous, being largely due both to an irrational credulity and an irrational incredulity, or, in other words, to a combination of dogmatism and scepticism — of dog- matism as regarded the words of a book, and of scep- ticism as regarded the facts of nature. Only slowly and with dijfficulty, — only through protracted and painful conflicts, — have the studies occupied with natural objects become genuine sciences ; and there are some who have represented the history of their progress as an exemplification of the triumph of scep- ticism and science over religion and theology. 318 SCEPTICISM AS TO PHYSICAL SCIENCE But it is assuredly nothing of the kind. Religion has gained as much from what has taken place as science. True theology finds strong support and rich nutriment in those emancipated sciences which are now so zealously and successfully reading and ex- plaining the book of nature. That book is the pri- mary, universal, and inexhaustible text-book of divine revelation, and although inadequate to satisfy all the wants of sinful man, it is, and will always be, neces- sary to him, not only as a physical but a spiritual be- ing. It is the oldest and most comprehensive of the media of divine revelation, and the correct interpreta- tion of it is only possible through the aid and instru- mentality of the appropriate sciences. Hence every enlightened theologian of to-day sees in the dogma- tism which would obstruct or enslave those sciences an ally of the scepticism which is an enemy both of pure religion and true theology. The more accurate- ly and fully physical nature is investigated and ex- plained by the sciences of nature, the more must the human mind recognise it to be pervaded by thought akin to its own ; the more must the human spirit find itself " at home " therein. (c) Historical scepticism, otherwise hnown as " erudite scepticism " and " historical Pyrrhonism," belongs to the same group. What is distinctive of it is the extent to which it challenges the credibility of historical narrative and questions the possibility of historical science ; and the conclusion at which it arrives may almost be formu- lated in the terms of the hon-mot attributed to Fon- 319 MITIGATED AND PAETIAL AGNOSTICISM tenelle, I'histoire n'est qu'une fable convenue. It contends that history, as an account of events, is very little to be trusted, and that a science of history can- not reasonably be looked for. Much may be said for that contention. History is rarely the record of deeds witnessed, or of words heard by historians themselves ; it is to a small extent founded on direct observation. Its data are, of necessity, largely reached by reasonings and guessings far from indis- putable. The history of man is known to us during only a very short portion of the time that he has been on earth. So-called ancient history is largely fabu- lous. Most of the classical historians were very un- critical. Medieval historians were exceedingly cred- ulous, and often relied on forged documents. It is impossible for even the most honest, learned, and la- borious historian to give a detailed account of any lengthened period, or comprehensive view of any com- plex portion of history, without falling into many errors. And there are few historians who are not biassed by self-interest, by prejudice, by party spirit, by the desire to be vivid, picturesque, and popular, and, in a word, by a multitude of perverting influ- ences. It is certainly not on the side of scepticism that ordinary readers of history err. Many are ready to accept in blind faith whatever is presented to them. There have been those, however, who may fairly be designated " historical Pyrrhonists." As typical ex- amples of erudite scepticism may be mentioned Bayle, a main purpose of whose famous Dictionary was the 320 HISTORICAL SCEPTICISM suggesting of historical doubts; Father Hardouin, who maintained that the works attributed to Thucyd- ides, Livy, and most of the so-called classical writings, as well as the chronicles and documents relating to the Franks, were forgeries; Schopenhauer, who has as- sailed the historical muse Clio in terms the most con- temptuous and even indecent; and M. Louis Bour- deau, who, in his L'Histoire et les Historiens, 1888, has learnedly argued that of true history there is as yet almost none, and that the historical method should be abandoned for the statistical. Almost the only " historical sceptics " of the pres- ent day, however, are not those who deny the possibil- ity of discovering historical truth and presenting it in an accurate and appropriate narrative form, but those who maintain that there can be no science or scientific study of history. Their attitude towards history is very much the same as the attitude towards nature of the ancient philosophers and medieval doctors, who thought a direct study of the material world would not yield physical science. The unwisdom of it will doubtless be made evident in the same way. In fact, a science of history is manifestly in course of forma- tion, and was never so eagerly cultivated as at present. All sociological studies are of the nature of contribu- tions to historical science, and the last quarter of a century has probably produced more such studies than all previous centuries together. Historical sci- ence may likely enough never attain the exactness of physical science, and yet reach greater depth and ful- ness of knowledge. Man, just because man, is eapa- 331 MITIGATED AND PAETIAL AGNOSTICISM ble of knowing more that truly deserves the name of knowledge about human nature and human history than about what is merely material or animal. He can enter more deeply into his own heart than into the nature of a stone, into the thoughts of Buddha or Plato than into the mind of an ox or sheep. He has to interpret nature by himself, not himself by nature. The human mind and its history are in themselves more intelligible than the physical world and its evo- lution, and may be expected when scientifically stud- ied and philosophically interpreted to contribute more to knowledge in general and to religious knowledge in particular. Matter is the stage prepared for the drama of the spirit. There is, we may be sure, more significance in the drama than in the stage, and what that significance is will be gradually brought more fully to light. The refutation of historical scepticism may safely be left to the future. The future will not fail to undertake the task, and will accomplish it by simply marching onwards. Solvitur ambulando. (d) Another variety of the same kind of agnostic cism is " ethical " agnosticism. It also has had a lengthened history, and has at times had considerable popularity. The diversity and contradictions of the moral judgments of man- kind has always been its favourite argument. Yet it is a very inconclusive one, as it owes whatever ap- pearance of validity it possesses to a manifest over- sight, the overlooking of the comprehensive unity of principles underlying the easily explicable differ- ences of applications and inferences. An impartial 323 ETHICAL SCEPTICISM study of the relevant facts cannot fail to show that man is always and everywhere a moral being, and that the more truly man he becomes the more does his mo- rality commend itself to the common conscience of mankind. Far from there being any incompatibility between a continuous moral progress and the immuta- bility of moral truths, there is a complete harmony. Further, all ethical scepticism is compelled to as- sume the ethical ideas and distinctions which it repu- diates. The moral law in its essentials is not only confirmed by the common consent of mankind, but practically recognised where it is not explicitly af- firmed, and spontaneously obeyed by those who logi- cally should disobey it. How should it be otherwise ? Only where there are order and reason of some kind can there be any truth ; and wherever there are order and reason there must be truth, and essentially the same truth, for truth is just conformity to the order of things and the requirements of reason. All the heavenly bodies may at some time or other be inhab- ited by moral agents. But there can be no moral agents except in intelligible and orderly worlds, and in all such worlds the ethics of rational agents must be, like their logic and mathematics, as Dr. Paul Ca- ms has justly argued, " in fundamentals the same." ^ Obviously to the extent that ethical scepticism is a partial and exclusive scepticism, tacitly or openly claiming to be the only scepticism, it is illogical and self-contradictory. There is no good reason for con- fining scepticism to the sphere of morals. Nay, if ' Fumda/mtntal ProUems, 46-52. 333 MITIGATED AND PAETIAL AGNOSTICISM consciousness, in the form of conscience, cannot be relied on, how can it be relied on in any other form ? If ethical agnosticism must be accepted, how can re- ligious agnosticism be rejected, or the claims of re- ligious science vindicated? If scepticism as to the knowledge of ethical truth be warranted, so must scepticism as to the knowledge of religious truth. The shortest way to complete religious agnosticism is to dispute the possibility of a knowledge of God, and God must be admitted to be unknowable if ethical truth be unknowable, — if reason be unable to appre- hend goodness, righteousness, and other ethical excel- lences. The very thought of a non-moral or immoral God is one in which no sane mind can find rest or satisfaction. It is a self -contradictory and monstrous thought. Were it a necessary or legitimate conclu- sion of reason, reason would be self-stultified, and neither science nor religion could be shovsm to be valid. (e) Another form of the same hind of scepticism as ethical agnosticism is metaphysical agnosticism. While undoubtedly prevalent, it is apt to seem even more so than it really is. The chief reason of that is that many who profess to be metaphysical agnostics do not know what metaphysics means. Obviously before a man declares metaphysics to be a pseudo- science or fancied knowledge, and that he has no faith in it, — or, in other words, before he poses as a meta- physical sceptic, — ^he shovild know what thoughtful writers on metaphysics mean by it, and should have studied the history, the chief systems, the main prob- 334 MEANING OF METAPHYSICS lems, the methods and the claims of metaphysics. But that is what comparatively few men have done. The ordinary man does not even ask what metaphysics is. The generality even of scientists are innocent of . metaphysical curiosity. The majority of self-styled metaphysical sceptics have never heen earnest meta- physical students. Many of them show, as I have said, that they do not even know in what sense the term metaphysics is, or ought to be, employed. No one who does attach a reasonable meaning to the term " metaphysics " will be inclined to entertain or advocate " metaphysical agnosticism " with a light heart. Whoever understands aright what metaphys- ics is, and consequently what metaphysical scepticism properly signifies, must recognise such scepticism to be a most radical and far-reaching agnosticism, a form thereof assent to which must involve grave and tre- mendous issues. What, then, is metaphysics ? It has been suggest- ed that no one knows what it is, and that there are as many different conceptions of it as there are inde- pendent metaphysical thinkers, or at least as there are distinct metaphysical schools. Nor need it be denied that there is some slight, although only very slight, appearance of foundation for the opinion. Meta- physicians often arrive at very different and conflict- ing results, and still oftener perhaps fail to arrive at any definite or positive results. There is much truth and wisdom as well as wit in De Morgan's humorous definition of metaphysics : " The science to which ignorance goes to learn its knowledge, and knowledge 325 MITIGATED AFD PAETIAL AGNOSTICISM to learn its ignorance. On which all men agree that it is the key, but no two upon how it is to be put into the lock." In the course of its history the word meta- physics has been employed in very different ways; and even at the present day all who expressly treat of metaphysics do not mean by the term precisely the same thing. But, certainly, so very general an agree- ment as to how it should be understood has at length been arrived at that there can be no reasonable doubt as to what in the main it ought to mean. Almost with- out exception metaphysicians now avoid confounding metaphysical with either physical or psychical sci- ence in general, or with any of the physical or psy- chical sciences, and treat of it as the science or theory which concerns itself with what both underlies and overlies all the special sciences, mathematical, nat- ural, mental, and theological; or, to express myself more precisely, which deals alike with the first prin- ciples and the last results of rational inquiry — alike with the fundamental conditions, categories, and lim- its of knowledge, and with the ultimate nature, rela- tions, and laws of reality. Thus understood, it is the theory of knowing and being, or of the universal and essential in truth and existence, and includes episte- mology (which should be carefully distinguished from logic and methodology) and ontology. Some metaphysicians indeed would identify it with the lat- ter, to the exclusion of the former; but the larger view, comprehensive of both, is, I think, much to be preferred. In fact, it is practically impossible to adhere to the narrower view ; impossible to act on it 326 METAPHYSICAL SCEPTICISM consistently for a single instant of time. Truth and reality are inseparable. There is no knowing with- out being, or being unrelated to knowing. Truth and reality, knowing and being, are throughout correlative and coincident. Epistemology and ontology are in intimate connection at every point. If metaphysics be what has now been indicated, and what is now almost universally regarded as the only reasonable conception of it, the nature and sig- nificance of metaphysical scepticism must be at once apparent. Metaphysical scepticism is scepticism as to what is primary in rationality and knowledge ; or as to what is ultimate in being and appearance ; or, and this is the more consistent as well as more compre- hensive view, as to what is universal and essential both in thought and existence. If understood in the first sense or reference, however, it means that there are no real or rational bases for any kind of knowl- edge or science ; if in the second, that there are no known grounds of reality and that all appearance is illusory and inexplicable ; and if in the third, that all epistemology and ontology are worthless, knowledge wholly unattainable, and existence altogether vanity. There can consequently be no deeper depth of scepti- cism than metaphysical scepticism. It leaves the mind with nothing to rest on or hold by. The contra- dictions of the senses, the contradictions of reason and reasoning, the contradictions between experience and theory, are what it appeals to in its own behalf, and these can warrant no trust in any positive truth. Thus living, moving, and having its being in self-contradic- 327 MITIGATED AND PARTIAL AGNOSTICISM tion, it can itself be a support to nothing, while it strives to undermine all the real foundations of sci- ence and philosophy. The only conclusion to which it naturally leads is the unattainability of knowledge, the incognoscibility of existence. It signifies as re- gards philosophy that all its problems are insoluble, and as regards the sciences that all their findings are dependent on unwarranted assumptions. Its bearing on theology is obvious. Theology is professedly not sceptical inquiry but positive science. It rests on faith in truth, and in truth of a metaphys- ical nature. It seeks, believes that it finds, exhibits, and defends such truth. Metaphysical scepticism implies and includes theological scepticism, and hence necessarily combats theology and denies its right to existence. If it be true, theology is false ; if theology be true, it is false. Theology is primarily and main- ly knowledge of God — a knowledge which has to be attained through reason and experience, through nat- ure and history, and, in a word, through all the ways and forms in which God has made Himself known. Metaphysical scepticism questions and denies our right to regard anything as a medium of knowledge of God. But, — " Of Grod above or man below, What can we reason but from what we know ? " (/) Tlie forms of agnosticism rrmy likewise he grouped with reference to the mental powers or prin- ciples of which the validity and veracity are disputed. Every power and principle of mind may be scepti- 338 SCEPTICISM AND THE SENSES cally treated, and, in fact, there is not one of them which agnosticism has not at some time and in some form assailed. To the simple and rude mind the clearest and most satisfactory of all testimony appears to be that of the senses. To the critical and reflective mind doubts respecting its reliability and worth necessarily sug- gest themselves. The seeming anomalies, the monot- onous and ceaseless changes, the apparent purposeless- ness, the labour and sorrow, which perplexed the soul of the author of Ecclesiastes when he contemplat- ed the world of the senses, also left many traces of doubt and sadness in the lines of the poets and the re- flections of the sages of ancient Greece. Both Brah- manists and Buddhists regard the world of the senses as a world of illusion, — a world of which " illusion " is " the material cause." All the chief sceptics of the Western world have disputed the credibility of the senses as witnesses to objective reality. The so-called errors and contradictions of the senses have, of course, afforded the materials for one of the main arguments in support of distrust of their testi- mony. It is easy to adduce numerous instances of various kinds of phenomena which may be so called and so represented. Attentively regarded, however, all phenomena of the kind will be found to be the results either of hasty and inconsiderate inferences or of abnormal conditions of the organs of sense, and not the deliverances of sound senses properly exercised. They are self-deceptions for which not the constitu- tion or action of men's senses are to blame but men 329 MITIGATED AND PAETIAL AGNOSTICISM themselves. Opinionis mendacium est non oculorum. The subjectivity of the senses has been not less relied on as an argument to justify scepticism as to their testimony. It is represented as implying the inabil- ity of the mind to apprehend really external objects. Descartes, Malebranche, JSTorris, Berkeley, and oth- ers, made use of it before Hume gave full and explicit expression to its implicit scepticism, — a scepticism which centred in the assumption that not things, not realities, but merely ideas or images are consciously apprehended. Those who have succumbed to such scepticism, while right in regarding sensations as in- dispensable to the knowledge of external objects and yet in themselves incapable of attaining or constitut- ing it, have erred entirely in conceiving of them as existing apart from perceptive and rational concomi- tants, and in disbelieving that what they could not do alone they could not do when not alone. Mere sensa- tions are mere abstractions which have no existence in any individual mind or actual experience. All real sensation is conjoined with perceptive and appetitive power, and in man at least with conception and rea- son. It is only an element, although an important element, of the psychical process implied in the cog- nition of external things. Scepticism as to memory is as possible, and may be advocated as plausibly, as scepticism regarding per- ception. Eemembrance is an act no less mysterious than vision. Of the many attempts which have been made to explain it not one has found much accept- ance. There is further between its testimony and 330 SCEPTICISM AS TO MEMOEY that of the senses a radical difference by no means in its favour. Acts of perception may be reasonably regarded as immediate and direct apprehensions of facts, and are generally so regarded. Not so acts of memory. Memory is dependent on immediate and intuitive knowledge, but cannot possess or supply it, — cannot know the past as present, the non-existent as existent. Probably no psychologist now holds Dr. Reid's view to the contrary. Recollections are never so vivid and exact as the perceptions and experiences recalled, and are generally very vague and blurred, very effaced and fragmentary, in comparison. Of all our cognitive powers, memory is the most closely con- joined with imagination, and has even been deiined as " reproductive imagination." But imagination, as every one knows, changes the appearances of all that it acts on, and shows little preference for truth over error. Memory is also largely affected by the disturbing influences of external surroundings, cor- poreal conditions, emotions, passions, habits, &c. All experience teaches that it is exceedingly apt to play us false. Its illusions are innumerable, and even its hallucinations are of many kinds. It is habitually inaccurate in the performance even of what may be regarded as its most special function — the measure- ment of time. A very poor chronometer can tell more exactly the duration of a second, a minute, or an hour, than the best memory. There are certain situations in which minutes seem to us intolerably long, and oth- ers in which we hardly notice the flight of hours. In early youth years, as recalled, seem long ; in old age, 331 MITIGATED AND PAETIAL AGNOSTICISM short. The entire mnemonic process — how anything whatever enters into memory and can be retained or recalled — is as yet an altogether unexplained mys- tery. The facts just referred to may sufEce to indicate to my readers how easy it may be, by simply dwelling on the defects and errors of memory, to get up a plausible plea for scepticism as to its trustworthiness. It is largely by such " exclusive dealing " that scepti- cism in all its forms is, and always has been, support- ed. Obviously, the method is as applicable against the credibility of any one faculty as of any other. As obviously, however, it is a fallacious method, and one by which in no case can agnosticism as to any of the faculties of mind be established. The agnos- ticism which bases itself exclusively on the errors and defects of any of our faculties must be, as regards even that faculty, a failure. What is overlooked or concealed in it is not destroyed, or lessened, or in any way got rid of by merely being ignored. Hence, when all that can be said to depreciate and discredit mem- ory has been fully said, its essential veracity and in- estimable value will remain intact and undiminished. Whoever has read the " Confessions " of St. Au- gustine will not be likely to forget the eulogy on mem- ory in Book x. Augustine there descants with mar- vellous eloquence and clear introspective vision on the spacious regions and palaces of memory ; on the treas- ures of innumerable images of things of all sorts con- tained in them ; of the media in and through which they have been acquired ; of how they are preserved 332 AUGUSTINE ON MEMORY from loss and brought up for use ; and of how so much of heaven, earth, and sea, of the histories of men and nations, of learning, art, and science, as well as of one's own self, feelings, deeds, and experiences belong to them. Realising how divinely wonderful the gift of memory is, he sought in chapter after chapter of the book to which I refer to make others appreciate it as he felt himself constrained to do. JSTow let us mark this fact. In all that he has written in Ms elaborate eulogium of memory there is probably not a sentence which has been disproved or discredited by agnostic or any other criticism. Between the criti- cism and the conclusion of agnosticism as to memory there is an enormous and irrational interval. The criticism, wherever true, only indicates defects quite compatible with the essential truthfulness of memory, and with its being all that Augustine has described it to be. Even a very ordinary memory can retain, with an extraordinary degree of evidence and accuracy, a wonderful wealth of experiences and acquisitions. And in the great majority of cases in which memory fails us, it does not even so fail us as not to leave us conscious that it fails us. We remember that we have forgotten, and, as Augustine says, " we have not yet forgotten that which we remember ourselves to have forgotten." Where there is no remembrance there is utter effacement of memory, but no error or deception of memory. Scepticism as to reason is another form of the same species of agnosticism. Here I would only remark that as in every other form of scepticism as to partic- 333 MITIGATED AND PARTIAL AGNOSTICISM ular powers or principles of cognition there will be found associated with the excess of distrust, distinc- tive of it, a correlative excess of confidence in another power or principle, so is it in the case of scepticism as to reason. The chief cause of such scepticism is an exaggerated estimate of the place and function of sense in cognition. Knowledge is attempted to be traced exclusively or mainly to sensation. That, however, can only be done, or even seem to be done, by an unnatural abstraction of sense from reason, which makes sense itself impotent and untrustworthy. Mere sensation has not been shown ever to exist alone, or even to be conceivable as existing alone. And, fur- ther, mere sensation, even if it existed, would be ex- clusively individual and subjective; but where there is no universality or objectivity there can be no knowl- edge or intelligibility. To accept sensation alone as the foundation of knowledge, or knowledge to be merely transformed or associated sensations, is en- tirely to betray the cause of knowledge. Having already, however, had to some extent to deal with agnosticism as to reason, and as I must nec- essarily have it further under consideration in the chapters which follow, I shall not dwell on it here. In the next chapter I shall enter on the considera- tion of that group of forms of agnosticism which directly refer to {he objects of hnowledge. 334 CHAPTEK VII PARTIAL OR LIMITED AGNOSTICISM AS TO ULTIMATE OBJECTS OF KNOWLEDGE There are forms of agnosticism distinguished from one another by the objects to which they refer. In so far as distinct from and exclusive of one another they are necessarily of a partial and limited nature, and as such they must, like the forms of mitigated or modified agnosticism, not only fail to realise but must contravene and contradict the ideal of agnosti- cism, which, as we have seen, must be arbitrary and inconsistent unless unlimited and universal. What Sextus Empiricus said of scepticism holds true of agnosticism: it should not be the belief of a school or sect, or a definite doctrine as to anything, but is " a certain line of reasoning," an arymjri or move- ment, the Svvafii^ of so opposing in every way the appearances of sense and arguments of intelligence to each other as to produce by their equilibrium sus- pension of judgment. Agnosticism thus understood is a potentiality of which all the actual manifesta- tions must be self -contradictory. ' That that is not an inaccurate view of it has, I trust, already been sufiiciently shown. It may be thought that as the forms of agnosticism now to be considered derive what is distinctive of them from their relationship to the objects of kriowl- 335 PARTIAL OR LIMITED AGNOSTICISM edge, our first question regarding them should be, What are the objects to which all real or supposed knowledge may be reduced, and with reference to which all partial forms of agnosticism may be dis- tributed? There is a question, however, prior even to that. It is the question, Under what conditions, or by what right, do those who advocate a partial agnosticism, a scepticism incomplete as regards ex- tension, draw limitations as to the sphere of knowl- edge? This is a question, it seems to me, which agnostics have generally neglected altogether or answered only in an arbitrary and dogmatic manner. Yet all agnosticism seems to depend on it. What is its theory of knowledge? How has it got it? How has it attained a theory of knowledge which can war- rant it to assign limits to knowledge, and say thus far but no farther? here man may search but there he ought not? matter is knowable but not mind? the ways of man may be traced but not those of God? At present I require only to consider the preliminary question of method. How may such a theory of knowledge be obtained as will warrant any limita- tion of the sphere of knowledge? I. ON ASSIGNING LIMITS TO KNOWLEDGE The limits of knowledge cannot be laid down in an a priori manner. There is no useful theorising on knowledge possible apart from knowledge. It is pre- sumptuous to warn intelligence off from investiga- tion in any direction until we have informed our- selves that in that direction there is nothing for it to investigate. We have no right to affirm that any 336 NOT TO BE ASCEETAINED A PRIORI question which the mind can seriously ask, with- out manifest self-contradiction, may not be found answerable, until we have learned that all rational ways of answering it have been exhausted. "We do not require to deny — we are not entitled to deny — that there are a priori limits or conditions of knowl- edge; but we are bound to deny the legitimacy of theorising on knowledge without knowledge, and without study of the various kinds of knowledge and of methods of investigation. Further, we cannot hope to ascertain the limits of knowledge by a mere critique of the powers of knowledge. We cannot measure the range of our mental tether by simply taking an introspective view of it. No mere psychological analysis of the consti- tution of the intellect will enable us to trace the bounds of its competency. When, indeed, the laws of the intellect are .violated truth cannot be attained, but to know that, and to know and conform to the laws of the intellect, is not to know how much truth may he attained, how far intellect may advance in its quest after knowledge. Everything at least that man does know he can know. Any estimate of man's power to know which leaves out of account what he actually knows must be an erroneous one. Hence the ques- tion. What do we know ? should precede the question. What can we know ? The positive grounds adduced in proof of knowledge ought never to be set aside or left unexamined because some general theory of knowledge has ignored them. "So theory of knowl- edge is universally valid which does not apply to every instance or fact of knowledge ; and consequent- 337 PAETIAL OE LIMITED AGNOSTICISM ly it is vain to appeal to any theory of knowledge against positive evidence for knowledge. Epistemology is the department of philosophy which undertakes to provide us with a theory of knowledge. It is concerned, therefore, not merely with some but with all knowledge, with knowledge as such in its entirety and universality, seeing that to attain a true and complete theory of anything an accurate and full knowledge of that thing is the indispensable condition. Epistemology as the theory of knowledge should be essentially vorjaK voi](7e Knovmig ound. Being, p. 3. 347 PAETIAL OE LIMITED AGNOSTICISM say, need to do. Physicists, as a rule, do not trouble themselves about theories of knowledge. They take it for granted, and can afford to do so, that no one will dare, in the name of an epistemology, to set at nought and refuse to look at the evidence which they adduce for their findings as to physical things. Theo- logians cannot act so because they are not equally sure of fair treatment. They have to consider that there are professed philosophers with an anti-theological bias who seek to arrest their inquiries and reject their findings with epistemological hypotheses. In those circumstances theologians little versed in philosophy may rightfully insist on pursuing their own labours and holding to their own conclusions so long as they are not met on their own ground and their own reasons are not weighed; and those of them who are competently conversant with it may further venture to criticise any epistemology to which their adver- saries appeal. There are some truths regarding the limitation of knowledge which must not here be left wholly unin- dicated. 1. JSTo object of belief or thought, not evidently self-contradictory, should be assumed to be unknow- able. It may just as rationally be assumed to be knowable. It is no less incumbent to give reasons for holding any conceivable object or proposition unknowable than for holding it knowable. There is as much demand for evidence for the denial as for the assertion of cognoscibility. A man who says that God is unknowable is under as much obligation to justify the statement as the man is who says God 348 LIMITATIONS OF KNOWLEDGE may be known. The only difference between them is that the man who says God is unknowable has much the more difficult proposition to prove. A negative proposition, unless it involve a manifest self-contradiction, is always more difficult to prove than an affirmative one. There are persons, how- ever, and agnostics are very frequently of the num- ber, wha seem to think that only knowableness re- quires to be proved, and that unknowableness may be assumed without evidence. 2. All that we have reason to believe real we have also reason to believe knowable. Much that is real may be unknowable to us; yet so far from being unknowable to us because it is real, in so far as we have any good reason to believe it real we have also reason to believe it knowable. It is the unreal which is necessarily unknowable, for it is no object of thought at all. One cannot prove anything about nothing; one cannot prove to exist what does not exist. The unreal is the negative at once of the real and the knowable. Existence and knowableness — reality, truth, and proveability- — are coincident and inseparable. In the Absolute Reality, with which philosophy and theology are alike concerned al- though in different ways, there can be no darkness, no unintelligibility, at all. Itself must fully know itself. To say that the world, the soul, or God is, yet cannot he Jcnown, is a statement both presumptu- ous and nonsensical. So far as anything really is it is knowable through the manifestation of what it really is. 3. It seems erroneous to suppose that we can draw 349 PAETIAL OK LIMITED AGNOSTICISM definite objective lines of demarcation between the knowable and the unknowable. We may draw lines between the known and the unknown, and it is highly desirable to draw such lines when we can, and as distinctly as possible. It is the characteristic of an accurate and careful thinker to distinguish as pre- cisely as he can between what he does and does not know; and to do so is always a forward -step in a man's pursuit of knowledge. But it is at once a mark of mental confusion and a perverse exercise of ingenuity to attempt to trace the external or objec- tive boundaries of rational research, — to draw lines in the outward universe beyond which all must be a terra incognita and within which all is explicable. Dr. Bithell — an agnostic writer to whom I have al- ready had occasion to refer — declares his inability to understand this objection to his agnosticism, and ventures to affirm that " the line of demarcation be- tween the knowable and the unknowable is at least as sharp and clear as the mathematical line which separates two plane surfaces." ^ Indeed ! There is no difficulty in drawing — in mentally realising — a clear and sharp line of demarcation between the hnown and the known, especially when both knowns are of the same nature. But is it as easy to draw such a line between the known and the unknown? Cer- tainly not. For the ordinary human intellect there is no clear and sharp line of distinction like a mathe- matical one between those two. Their boundaries are continually changing and commonly very indistinct. But what Dr. Bithell, with " a light heart," ventures ' Agnostic Problems, p. 3. 350 LIMITATIONS OP KNOWLEDGE to undertake is a far more difficult task than to fix the boundaries between the known and unknown ; it is to draw a line as clear and sharp as if it were a mathe- matical one betiveen two unknowns, and one of which is not only unknown but unknowable. That I vent- ure to think must be a problem which no finite being can solve. Dr. Bithell has certainly not solved it by telling us that " the line of demarcation between the Knowable and the Unknowable may be defined as that which separates those phenomena that come within the range of consciousness from those facts or truths which lie beyond the reach of consciousness." ^ That is only equivalent to saying that we know what we know and cannot know what we cannot know, — a truism which defines and distinguishes nothing, and is of no value whatever. Dr. Bithell represents Kant and Ferrier as having taught to the same effect as himself, but the two quotations (p. 5) adduced in proof are to quite a different effect. There is no war- rant in the history of philosophy for his statement that " philosophers, generally, are pretty well agreed in making consciousness the line of demarcation be- tween the Knowable and the Unknowable." I am not aware of any philosopher of eminence having come to such a pass as that. Philosophers, generally, are pretty well agreed, I think, that to draw a line of demarcation between the Knowable and the Unknow- able is impossible ; that there is absurdity — self-con- tradiction — in the very attempt ; that to draw such a line we must have already done what we affirm to be impossible — ^known the unknowable; that we cannot ' Agnostic Prohlems, p. 4, 351 PARTIAL OE LIMITED AGNOSTICISM draw a boundary unless we see over it, or, as Hegel says, — " No one is aware that anything is a limit or defect until at the same time he is above and beyond it." 4. The only ascertainable limitations of the mind manifesting itseK as reason — i.e., in the appropria- tion of knowledge and truth — are those which are inherent in its own constitution. They are subjective not objective limitations. They are inherent in and constitutive of intelligence. Reason — the mind as cognitive or rational — has its limits in its own laws. To discover, state, and expound those laws is the business of Epistemology or Theory of Knowledge, which is intimately connected with Psychology, Formal Logic, and Methodology. The laws of rea- son — laws of intuition, evidence, and in,ference — are manifestly not external boundaries, but they are the only discoverable expressions of the Divine " Thus far." So long as reason conforms to its own laws it cannot go too far. When it does not conform to them it ceases to be reason and becomes unreason. Reason is entitled to examine any and every thing which comes under its notice, and cannot push examination too far so long as it remains reason. Only when it violates some law or laws of its own has it gone too far, — has it erred and strayed, — and then simply because it has ceased to be rational. Does the agnostic say that that may be true of reason and its sphere the Knowable, but that beyond them there are faith and its sphere the Unknowable, and that " he is prepared to work on both sides of the line of demarcation? On the side of the Know- 352 LIMITATION'S OF KNOWLEDGE able he founds and cultivates his Science; on the side of the Unknowable he finds an illimitable arena for the exercise of Belief and Faith." ^ Eeason and Belief or Faith, however, cannot be so separated. Where there is no reason or knowledge there should be no belief or faith. In the Unknowable there is no arena for the exercise of a reasonable belief or an honest faith. All that the mind can do on the side of the Unknowable is to play at make-belief, to feign faith, to worship nothingness. Such exercise must be both intellectually and morally a very dan- gerous sort of exertion. Madness that way lies. 5. Knowledge is limited by evidence. We lack knowledge of what we have not sufficient evidence for. !Rothing, however, sufficiently proved by evi- dence of any kind is to be rejected because it cannot be proved by evidence of another kind. Demonstra- tion is the proof appropriate in mathematics, but it is a kind of proof which one has no right to demand in psychology, ethics, or history, or even in the phys- ical sciences. Proof, and thoroughly satisfactory proof too, has many forms. Hence the words prove and proof have necessarily many variations of signi- fication. Agnostics often make an abusive applica- tion of that fact. Their favourite quotation is drawn from Tennyson's Ancient Sage: — " Thou canst not prove the Nameless, O my son, Nor canst thou prove the world thou movest in. Thou canst not prove that thou art body alone, Nor canst thou prove that thou art spirit alone. Nor canst thou prove that thou art both in one : Thou canst not prove thou art immortal, no 1 Bithell, Agnostic Problems, p. 17. 353 PAETIAL OE LIMITED AGNOSTICISM Nor yet that thou art mortal — nay, my son, Thou canst not prove that I, who speak with thee, Am not thyself in converse with thyself. For nothing worthy proving can be proven, Nor yet disproven : wherefore thou be wise. Cleave ever to the sunnier side of doubt, And cling to Faith beyond the forms of Faith ! " These lines are very beautiful, and perhaps precise enough for the poet's purpose, but they have no claim to be regarded as a correct expression of a true philosophical creed. The " Nameless " is the Being who has been named more or less aptly in all the languages of the earth, and who has been almost universally recognised by mankind as the most self-revealing of Beings. In the ordinary sig- nification of the word " prove," all sane men accept as adequately proved the existence of the world, of them- selves, of their bodies and spirits, and that in each of them body and spirit are united ; and if many of them are in doubt as to whether they are mortal or immor- tal it is because of a conflict of reasons which mahes them dubious as to whether there is proof or on which side it is. The evidence for the distinction between " I " and " Thou " excludes all rational doubt. It is "proof" as strong as the self -evidence of a mathemat- ical axiom. " Nothing worthy proving " has been left without the power of proving itself. " Cleave ever to the sunnier side of doubt " is very questionable ad- vice — an encouragement to selfishness and indulgence — unless it mean, Cleave to the side on which the light of reason, the sun of truth, shines clearest. I do not in the least blame the poet for his use of the word prove; demonstrate would have taken all the 354 ULTIMATE OBJECTS OP KNOWLEDGE music out of his lines. What I object to is that ag- nostics should expect us to accept his words as liter- ally, or even substantially, true. 6. The existence of obscurity, mystery, and diffi- culties in connection with the objects of knowledge does not disprove knowledge of them. Propositions may be perfectly true, and conclusively proved to be true, although they involve incomprehensible con- ceptions, and are associated with unanswerable diffi- culties. The ultimate truths even of mathematics have all a side which is lost in difficulty and darkness. The conclusions of the infinitesimal calculus when- properly worked out have to be accepted in spite of air the perplexities which may be suggested by thinking of infinites and infinitesimals of different orders. It shows a lack of clearness of thought to reject truths because of merely connected difficulties. Whatever reason assures us to be real and certain is to be accepted, however much there may be as- sociated with it which is dubious and perplexing. The mysteriousness inseparable from the immensity, infinity, eternity of God, and Space, and Time does not make their existence in the least degree doubtful. As our knowledge that the grass grows is not in the least subverted by our ignorance of how it grows, so our knowledge of the existence of an Infinite and Absolute Being is quite compatible with our inability to form clear and adequate conceptions of Infinity, Absoluteness, and Being. II. THE ULTIMATE OBJECTS OF KNOWLEDGE The most generally adopted distribution of the ultimate objects of knowledge is the threefold one — 355 PARTIAL OR LIMITED AGNOSTICISM Self, the World, and God. Those three objects Bishop Westcott, for instance, designates " the three final existences which sum up for us all being," and treats of them as such in a very instructive and sug- gestive way in his Gospel of Life (eh. i., pp. 2-42). So Professor Fraser in his Philosophy of Theism (Lect. II.) describes them as "the three primary data differently conceived by different minds " — " the ultimate threefold articulation of the universe of existence " — and admirably emphasises the im- portance of the right correlation of them in human thought and life. That mode of distribution, besides being the most familiar one, is also the most con- venient as regards all that I have at present in view, a consideration of agnosticism as to Self and the World in so far, and only in so far, as it bears on agnosticism as to God. Therefore I avail myself of it. That it is a faultless distribution of the ultimate objects of kaowledge, or one which can be safely accepted as the principle of a classification of the sciences or of the organisation of a philosophy, I am far from affirming. On the contrary, I admit it to have various defects. Two of them it seems neces- sary to indicate. The first is that the terms " Self," " World," and " God " are not unambiguous terms. " Self " and " world " are apt to seem quite clear and definite. They are really very much the reverse. " Self " ! What " self " ? Is it merely the individual self, the self of self -consciousness, the subject of a mind as cognisant of itself in feeling, desiring, believing, 356 ULTIMATE OBJECTS OF KNOWLEDGE knowing, willing, &c. If so, every " self " except the individual's own must be included not in " self " but in the " world," or at least in a distinct category of selves — say " other selves." The difference be- tween the " self " of self -consciousness and the " selves " which are to that " self " merely objects of knowledge is in some respects even greater than that between the subjective self and physical objects. If by " self " be meant both the subjective self and objective selves, humanity or the human mind or human nature would seem to be what is denoted by it. But is even that all that should be meant? Can we stop even there ? Should not all that feels, every sentient creature, be regarded as a "self"? If so, by " self " must be understood not the individual self, but the whole finite animate or conscious world, or even all spiritual being and life, the Divine in- cluded. But in that case by the " world " would have to be meant the merely material, the exclu- sively physical, world. ISTo one, however, so restricts the signification of the term in ordinary speech. The external world is not merely composed of dead and physical things but to a large extent of living and conscious things. " ^N^ature " aid " universe " we often vaguely call it. And under those names monis- tic physicists are in the habit of identifying God with it or including God in it, while thorough and consistent pantheists represent it not as an object of knowledge but as eseentially an illusion, a decep- tive appearance of reality. The term " God " as used in agnostic controversy ig — notwithstanding all the different conceptions 357 PAETIAL OE LIMITED AGNOSTICISM which men have formed of God, and notwithstand- ing also that agnostics deny God to be an object of knowledge, or affirm that He is only an object of belief — less ambiguous than either " self " or " world." And the reason is obvious. Non-agnos- tics have to state clearly what they mean by the term " God," and agnostics are bound to show that in the sense affirmed there is no known or even know- able God. When, therefore, the non-agnostic de- clares that he has, and that others may have, good and sufficient reasons for holding that there is a self- existent, inffiiite, eternal, morally perfect spirit or mind, the source, sustainer, and controller of all finite minds and existences, the agnostic may, or rather must, deny his statement, but he cannot deny that he knows what the term " God " as employed by his opponent means, and what both the affirmation and the denial of God in that sense mean. In contro- versy between an agnostic and a non-agnostic there need be no ambiguity as to what is meant by " God," and there seldom is any. One cannot say the same of the terms " self " and " world." It is miich easier, however, to indicate than to remove the defects of the ordinary thre'efold distribution of objects of knowledge ; much easier to criticise it than to replace it by a better. There is happily no reason why I should undertake the latter task. None of the ambiguities in the terms of the afore- said distribution of ultimate objects of knowledge can affect anything which I have to say in the pres- ent chapter regarding agnosticism as to either self- knowledge or world-knowledge. But this question 358 ULTIMATE OBJECTS OF KNOWLEDGE does so to some extent, — Are we entitled in any dis- tribution of such objects of knowledge to ignore Space and Time, which, while identical neither with " self " nor the " world," mind nor matter, are yet not mere imaginations or nonentities, but necessities of thought, conditions of existence, and the very foundations of mathematical science, on which so much other science is dependent? True, the mathe- matical sciences are not dependent only on the quantitative relations with which they are occupied. They depend also on the formal laws of thotight ■ which it is the business of Logic to expound. Sub- jective laws, however, would not lead to objective truths had they not real objects to deal with. It has never been shown to be even conceivable that with- out apprehensions of space and time we could have any valid or consistent conception whatever of objec- tivity or externality. On those apprehensions the mathematical sciences rest, and of all sciences they are the most certain and exact. They can dispense with observation and are independent of experimen- tation. They need no external verification. They prove by their very existence that there is a knowl- edge perfect of its kind which has not its exclusive or even its main source in sense, and the limits of which are not those of sense. They are in themselves an irrefragable refutation of the hypotheses as to the nature and limits of knowledge propounded by empiricists and positivists. l^either their principles nor their conclusions are generalisations of the data of sense. And yet they are — regarded merely as knowledge — knowledge at its best. Plato and ISTew- 359 PAETIAL OR LIMITED AGNOSTICISM ton have spoken of God as " thinking mathemati- cally." JSTovalis has enthusiastically declared — " Pure Mathematics is religion, the life of the gods is mathematics, the mathematicians are the only happy men." John Bright has been credited (I forget by whom) with having said — " Teach a boy arithmetic thoroughly and you will make a man of him." If he said so he must have felt that there was something more in arithmetic than most people imagine, something ethical and divine, in virtue of which, if " thoroughly taught," it would not merely exercise "the arithmetical understanding" and make ' quick and accurate calculations, but also so influence the whole character and life as to make " men." Space and Time are not mere subjective concep- tions. They are not arbitrary creations of thought. It is not in any man's power to accept or reject them at will, or to apprehend them otherwise than as all men apprehend them. They are objects of in- tuition and forms of thought, but not merely or exclusively so; on the contrary, they are intimately and inseparably connected with all the facts of ex- perience and all the objects of nature. Idealism and empiricism are alike incompetent adequately to ac- count for or even accurately to describe them. Kant in his Critique of Pure Reason has attempted to do so from the standpoint of the former, and Dr. Shadworth Hodgson in his Metaphysic of Experience from the standpoint of the latter. Both have failed. Both have had to assume as regards alike Space and Time what they professed to prove. They have cer- tainly not failed from lack of either ingenuity or 360 ULTIMATE OBJECTS OF KNOWLEDGE industry, of either ability or zeal. To all appear- ance they have failed because they attempted the impossible, and there is little hope of any one else succeeding who confines himself to the course or method of either. Eeason cannot refuse to recognise that Space and Time are iniinite and eternal. Only so can it think of them. To affirm them to be iinite as regards either extension or duration would be for reason a suicidal act. The familiar words everywhere and always imply all that is explicitly expressed by in- finite and eternal. It is distinctive of man as a rational being to have an implicit knowledge of in- finity and eternity. As soon as he is capable of reflection he finds himself cognisant of those two transcendental realities. He can confidently affirm space to be infinite in every direction, as it is a self- contradiction, manifestly irrational, to regard it as finite in any direction. The finite is the limited. But by what is space limited? It must be either by a vacuum or a plenum, and yet it is absurd to regard it as limited by either. There is space where there is neither the one nor the other. Were there no matter in existence, — were not merely the gross matter which we apprehend through all our senses but also the subtle and mysterious ether which is the subject of so much speculation and the object of so little positive Imowledge annihilated, — there would not be an inch of space either more or less in the universe. Time is infinite in two directions. It has no limit either on the side of the past or of the future. To say of anything that it happened 361 PAETIAL OE LIMITED AGNOSTICISM or will happen at no time is equivalent to saying that it has never happened or will never happen. Estimates of time may vary indefinitely. Short- lived creatures may, perhaps, in some species be so organised as to feel life as long as those that are really long-lived. A drowning man may in a few minutes feel as if he were passing through the whole course of his past life. A dreaming man may in as short a time imagine himself passing through hours of exertion, danger, or sorrow. A thousand years may be as a day and a day as a thousand years according to the differences in the rapidity, vivacity, intensity, &c., of subjective states experienced. Yet Time itself does not vary — does not flow faster or slower, but continuously and equably through innumerable imperceptible instants. There is not a minute more in an hour felt to be long than in one felt to be brief. That Space and Time are we know and cannot fail to know. Mathematics shows that a vast amount may be known with certainty about them and in dependence on them. Yet how mysterious they are ! How difficult, and indeed even impossible, it is to find or invent fitting words to express what they are ! To say that space is extension and time duration is just to say that space is space and time time, or that space and time are what they are, which is no doubt true, but no doubt also does not add in the least to our information as to their nature. Are they "things"? Certainly not, if all things are either material or spiritual. They are neither material nor spiritual, although there is nothing finite, whether 3C2 ULTIMATE OBJECTS OF KNOWLEDGE material or spiritual, conceivable by us otherwise than as within them. Are they properties or quali- ties? Perhaps they are. But before we are entitled positively to affirm that they are, we should require to know what qualities or properties are, and of what substance or substances, being or beings, space and time are the qualities or properties. As they are infinite and eternal they cannot be confined or spe- cially belong to any finite being or substance. The infinite and eternal can be coexistent and coexten- sive with itself alone. To characterise tinie and space as merely forms of thought has the serious defect of not describing them as they are actually thought of, namely, as necessary objects of thought and necessary conditions of objective existence. Our apprehensions of space and time are insep- arable from thoughts and convictions of immensity and eternity, and consequently bring with them the same thoughts and convictions as our apprehensions of God. They are in the same way mysterious, and in the same way so far from self-contradictory that they cannot fail to command assent. While matter is unthinkable as either infinite or eternal, — while it can only be conceived of as within time and space, as having begun at a given time and reached a definite date, and as being of some particular mag- nitude and form, — space and time are like God in that they are only truly thinkable as infinite and eternal. Hence our thoxights of them bring with them some of the same difficulties as our thoughts of God. They bring with them the same great mysteries of self-existence, eternity, and infinity. 363 PAETIAL OE LIMITED AGXOSTICISM Yet they are not therefore incompatible with knowl- edge and rational belief. On the contrary, they are vehicles of a real knowledge of time as eternal, of space as infinite, and of both as necessarily existent; snch a knowledge as should at least suffice to prevent us from venturing to deny that God can be known as eternal, infinite, and necessarily existent. Xo otherwise can God be consistently thought of than as possessed of those attributes. To think of God as having begun to be in time or as confined to a limited space is to think of Him as a god who must have been created by another God; in other words, must be to regard Him as a false god, as not truly God. Xo thoughtful atheist even can think of any being not eternal and infinite as truly God. Such an atheist, if the most thorough-going materialist, must feel bound to invest matter itself with the at- tributes of infinity and eternity. He cannot ration- ally maintain that it has created itself or assigned conditions or limits to itself. Xor can he reasonably maintain that it has been created either by eternal time or infinite space, for neither the one nor the other causes or creates anvthing. They are con- ditions of existence but not efiicient agents, not endowed with any kind of power. They in no way account for the existence, organisation, peculiarities, or activities of anything. Infinite space and eternal time can originate and explain nothing unless con- joined with Absolute Being, — self-existent, self-ac- tive, and spiritual Being, — the Being on which all finite and dependent beings, all animate creatures, all selves, all societies, live, move and have their 304 ULTIMATE OBJECTS OP KNOWLEDGE being. The infinity of space and eternity of time, instead of entitling us to dispense with faith in an infinite and eternal God, seem rather to demand such faith. The self-consistency of thought requires it. Keason insists that the empty infinities of space and time be filled with the powers and perfections of reason in order to be the Absolute Infinity which can alone satisfy rational minds and explain a rationally organised universe. And the most resolute material- ists have had practically to acknowledge the justice of the claim. They have been compelled to exercise their imaginations at the cost of their reason in the deification of matter. Holbach, for example, in his Systeme de la Nature ascribed to matter much which he denied to God, but which cannot be sanely con- ceived to belong to matter, and which contradicts the teachings of genuine science. Haeckel, whose so-called Monism is the present-day counterpart of Holbach's Naturalism, attributes infinity and eter- nity to a world-substance, that in the form o± " Ether " pervades, fills, and animates all space and time, and is, in his opinion, the only satisfactory basis of either religion or morality. Nature or the World he divides into " Ether " (= spirit), mobile or active substance, with vibration as its property, electricity, magnetism, light, and heat as its func- tions, and a dynamical, continuous, elastic, and prob- ably non-atomic structure ; and " Mass " (= body) inert or passive substance, with inertia as its prop- erty, gravity and chemical affinity as its functions, and a discontinuous, inelastic, and probably atomic structure. Ether, we are told, is, theosophically 365 PAETIAL OE LIMITED AGNOSTICISM speaking, God the Creator, and Mass or Body the created world (Monism, p. 106). Such a doctrine is surely no improvement on the materialistic systems of earlier times, very unlikely indeed to solve any Weltrdthsel, and worthless as a basis for either re- ligion or morality. The word God has a definite meaning, and no man has any more right to identify it with the Ether than with a stock or a stone. That eternity and infinity are not in themselves distinctively religious ideas I fully grant. To worship pure space or mere time is impossible. 'No human being has ever done anything so foolish. I must, however, entirely dissent from the opinion of Dr. Paul Carus '- that religion would not suffer if the ideas of eternity and infinity were abstracted or dis- sociated from our thoughts of God. It seems to me that it would suffer dreadfully; that the abstraction referred to would leave little room for rational faith in God or enlightened piety towards God. To deny the infinity and eternity of God appears equivalent to afiirming that there are places where He is not or even cannot be, and to imply that He is such a god as Elijah described Baal to be, one to whom it might be necessary to cry very loud as he might be wholly engrossed with his own thoughts, or on a journey, or peradventure asleep and must be awaked. How has so earnest and able a thinker as Dr. Carus taken up such a position? Apparently in consequence of meaning by the term " God " " cosmical law," and by the term " religion " " morality " or " ethical conduct." Can any one, however, have a right so to ^ Homilies of Science, pp. 108-112. 366 ULTIMATE OBJECTS OF KNOWLEDGE employ those terms in controversy with others who are using them in their ordinary sense ? Surely not. Such an employment of them manifestly tends to efface the distinction between theism and atheism, and to make rational discussion between the theist and atheist, the religious and non-religious man, im- possible. Our cognitions even of Time and Space imply some knowledge of the Absolute and Infinite. Much more must such knowledge be involved in our appre- hensions of God, the true Absolute, the positive In- finite, the self -existent and all-perfect Being. It does not follow, however, that we have any absolute or infinite knowledge. All human knowledge is rela- tive and finite. Even the mathematician has only a relative knowledge of the absolutes on which all his science rests. One of them is " time." The mathematician does not know what time is, but must assume that it is, and must reason and calculate on that assumption. He thinks of it as an absolute — as that which always was, which necessarily is, which must for ever be, and yet which is constantly disappearing and reappear- ing, ceasing to be and coming to be, at every instant, — a continuum which is ever the same and yet never the same, and which fiows ever onwards at an unvary- ing rate. He cannot do otherwise than so think of it, although when so thought of it is profoundly myste- rious and accompanied by apparent self-contradic- tions which no one seems to have satisfactorily ex- plained away. He would seriously err were he to conceive of it as what may once not have been or as 367 PARTIAL OE LIMITED AGNOSTICISM what may not always be, as capable of going not only forwards but backwards, as flowing at one time fast and at another slow or occasionally ceasing to move at all. Yet while he would' seriously err were he to reject the idea of " mathematical, true, and absolute time," he may well be profoundly thankful that in the prosecution of his science or in the application of it what he has to deal with is " relative, apparent, and common time." Not otherwise is it as regards " space." Space is necessarily thought of as absolute and infinite, as what cannot not be or be otherwise than as it is thought to be — i.e., as immovable and irremovable, immuta- ble and indivisible, timeless and changeless. Yet no human being has an absolute apprehension of space. The mathematician and the metaphysician alike must be content with a relative, vague, and imperfect ap- prehension of it. The more they know of it the more conscious will they be how relative and imperfect their conceptions of it are. Until modern times men thought that they knew with entire certainty where the centre of the world was — that it was at Jerusalem — and that there could not possibly be men on the side of it opposite to their own. Will any one now venture to affirm that there is an absolute centre or absolute up or down in space ? Would doing so not be to follow the bad example of the Christian fathers and medieval schoolmen who pronounced it senseless and profane to believe that there could be human be- ings on the side of the earth opposite to their own ? It was for many ages firmly held that the sun and planets turned round the earth. Copernicus and Gal- 368 AGNOSTICISM AND THE SELF ileo proved that the geocentric theory should give place to a heliocentric one, and, notwithstanding the long and bitter opposition alike of Catholic and of Protestant divines, the latter theory is now universally accepted. Its superiority over the geocentric theory as regards tracing the movements of the universe is obvious and immense. It is vastly more convenient to take the sun as the standpoint of observation than the earth. But the geocentric theory was not wholly erroneous ; on the contrary, the predecessors of Gali- leo observed, from their point of view, as correctly as he did from his. Nor is the heliocentric theory ideal- ly perfect. The sun no more affords an absolute posi- tion than the earth does. In some of the far-off worlds of God's great universe there may quite con- ceivably be astronomers who have enormous advan- tages in the prosecution of their studies over their terrestrial brethren. Although then a vague appre- hension of absolute space seems to underlie and to be implied in all our definite and relative conceptions of space, it would seem as if we can only deal in a prac- tical way with relative space. " Any one," Clerk Maxwell wisely and wittily said, " who will try to imagine the state of a mind conscious of knowing the absolute position of a point will ever after be content with our relative knowledge." * III. AG]