r. 
 
 L 
 
 LIBRARY 
 
 Owivlk'hy of 
 
 california 
 
 SAN DITOO 
 
 ^ 
 
 J
 
 ■1 
 
 3""i822""01147 9193 
 
 M ■>.!■ 
 
 Central University Library 
 
 University of California, San Diego 
 Note: This item is subject to recall after two weeks. 
 
 Date Due 
 
 NOV ^ 2 m? 
 
 
 
 
 
 
 HAY 2 9 1994 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 0139(1/91) UCSDLib.
 
 A STUDY IN REALISM
 
 CAMBRIDGE UNIVERSITY PRESS 
 
 C. F. CLAY, Manager 
 
 LONDON : FETTER LANE, E.G. 4 
 
 NEW YORK : THE MACMILLAN CO. 
 
 BOMBAY 
 
 CALCUTTA ■ MACMILLAN AND CO., Ltd. 
 
 MADRAS 
 
 TORONTO. : THE MACMILLAN CO. OF 
 
 CANADA, Ltd. 
 TOKYO: MARUZEN-KABUSHIKI-KAISHA 
 
 ALL RIGHTS RESERVED
 
 A STUDY IN REALISM 
 
 BY 
 JOHN LAIRD, M.A., 
 
 PROFESSOR OF LOGIC AND METAPHYSICS IN THE QUEEn's 
 UNIVERSITY OF BELFAST 
 
 CAMBRIDGE 
 
 AT THE UNIVERSITY PRESS 
 
 1920
 
 To thee, O man, the sua his truth hath given, 
 The moon hath whisper'd in love her silvery dreams ; 
 Night hath unlockt the starry heaven, 
 The sea the trust of his streams : 
 And the rapture of woodland spring 
 
 Is stay'd in its flying; 
 
 And Death cannot sting 
 
 Its beauty undying. 
 
 Robert Bridges.
 
 TO THE MEMORY OF 
 MY BROTHER 
 
 HUGH BLACKHALL LAIRD 
 SEOOND-LIEUTENANT, YORKSHIRE REGIMENT 
 
 WHO FELL IN ACTION 
 
 AT TRONES WOOD 
 
 DURING THE BATTLE OF THE SOMME 
 
 8th JULY, 1916
 
 PREFACE 
 
 WHEN I set out to write this book, I meant to give the 
 reader a methodical, brief survey of some of the chief 
 problems in philosophical realism (as I understood it), and to 
 spend most of my labour on certain points in the theory which 
 I found especially perplexing. I thought, indeed, that many 
 of the cardinal features of realism had been investigated so 
 minutely within recent years ^ that I could afford to omit some 
 of them from my discussion and to be very brief in my treat- 
 ment of certain of the others. Per contra, I considered that 
 realists had commonly paid too little attention to certain 
 varieties of knowledge in which, at the first look, other theories 
 seemed better suited to the facts. The creative imagination of 
 the artist, for example, the constructions of science, and even 
 the meanings of perception, might seem to belong to a wholly 
 different order from the simpler ways of apprehending which 
 the realists dissect (I except Mr Alexander, whose Space, Time, 
 and Deity had not appeared when I wrote) and I wished to 
 examine whether contrasts of this striking kind were securely 
 established in fact. In other words, I wished to search those 
 other theories on the very ground they had chosen for them- 
 selves, feeling convinced that realism was strong enough to 
 occupy it, and knowing, as a thing of course, that if realism 
 were to fail in this enterprise, it could only be a provisional, 
 departmental theory, and not (as it claims to be and as I 
 believed) a final, catholic one. 
 
 1 The theory of relation is an instance.
 
 viii PREFACE 
 
 I never expected, I hoj)e, to do more than an underling's 
 work in so slight an essay upon so large a theme, but, even 
 so, I wish that my confidence had not oozed so persistently as 
 I wrote, and that I could have felt less like a child in chase 
 of a rainbow. And now that the printer has put an end to 
 my struggles with logic and my nibblings of the pen, I am 
 better able to appreciate the yawning chasm between antici- 
 pation and achievement. I am doubtful, now, whether I should 
 not have included much that I passed over of set purpose ; and 
 I sigh for the equipment which would have enabled me to deal 
 more adequately with many of the problems that have sought 
 me out. On the other hand, I think I may claim that I have 
 faced any of the difficulties I was able to understand, openly 
 enough and squarely; and that I have honestly endeavoured 
 to keep objections in the foreground instead of attempting to 
 gloss them over. 
 
 Be that as it may, I make no apologv for the spirit of this 
 adventure; and I should not wish to do so, even if such an 
 apology could ever condone the offence. There can be no 
 health in philosophy, I am sure, without continual discussion ; 
 and I still believe most firmly that realism is a truly philo- 
 sophical theory of knowledge, by which I mean that the realists' 
 point of view, literally interpreted and resolutely argued, 
 may be sustained, consistently and without special pleading, 
 throughout the whole wide territory of the theory of know- 
 ledge. 
 
 No part of this book has been published before; but I 
 have contributed (copiously, I am afraid) to the philosophical 
 journals during the past year or two, and these pages show 
 traces of portions of this published matter and of certain com- 
 ments upon it. To be more precise, two articles in Mind are 
 connected with the subjects of the third and of the eighth
 
 PREFACE ix 
 
 chapters in this book; an article in the British Journal of 
 Psychology dealt in advance with some of the problems of the 
 second chapter; the concluding pages of the sixth chapter are 
 tinged with the remembrance of an article I wrote for The 
 Mo7iist; and the general argument of the book has a certain 
 affinity with the contentions I put forward in a paper pub- 
 lished in the Proceedings of the Aristotelian Society. In 
 writing the eighth chapter I had to deal, in part, with the 
 subject-matter of a former book; and I hope I have learned 
 from my critics. 
 
 I am most grateful to my father for the pains he has taken 
 in reading the book in proof, and for advising me of many of 
 my mistakes. And, like so many others, I have the pleasant 
 privilege of thanking the Syndics of the Cambridge University 
 Press for the honour they have done me in publishing the 
 book, and of expressing my gratitude to all who may be con- 
 cerned for the skilful care which has been given to the printing 
 of it. So much in the book is due to what I learned at Cam- 
 bridge that I may be pardoned, I hope, for finding a peculiar 
 delight in this privilege. 
 
 J. L. 
 
 Juhj V2, 1920
 
 CONTENTS 
 
 CHAPTER 1. INTRODUCTION 
 
 Realism ...... 
 
 A short retrospect: Arnauld and Raid 
 Assumptions ..... 
 
 PAGE 
 
 1 
 
 2 
 8 
 
 CHAPTER II. THE THINGS WE PERCEIVE 
 
 Perception, judgments of pei'ception, and sensation 
 Direct perception ........ 
 
 The theorj' of sensory atomism ...... 
 
 Dr \\'ard's theory ........ 
 
 Sign-facts and continuants ....... 
 
 Perception and the discovery of an independent world 
 The perception of 'matter' ...... 
 
 CHAPTER III. THINGS REMEMBERED AND 
 THINGS EXPECTED 
 
 The 'specious present' ...... 
 
 Immediate knowledge of the past and of the future 
 Analysis of memory : reproduction and recollection 
 The place of memory-images ..... 
 
 CHAPTER IV. THE STUFF OF FANCY 
 
 Fancy and imagination .... 
 
 Images ....... 
 
 Dream and waking ..... 
 
 The elements of fancy : and their meaning . 
 Constructive imaging : and psycho-analysis 
 Theory of fancy ...... 
 
 CHAPTER V. THE WORLD OF COMMON BELIEF 
 
 Perception and judgment . 
 
 Propositions about the world 
 
 The space and time of common belief 
 
 Causal interpretation . 
 
 Error. ..... 
 
 15 
 17 
 18 
 25 
 
 27 
 30 
 30 
 
 45 
 49 
 51 
 55 
 
 60 
 61 
 65 
 67 
 75 
 81 
 
 83 
 86 
 93 
 95 
 103
 
 CONTENTS 
 
 CHAPTER VI. PRINCIPLES 
 
 General facts : ami first principles 
 
 The order of tlieir discovery 
 
 Tlie being of categories and of universals 
 
 Univeraulia in re . . . . 
 
 Induction and pure logic 
 
 The law of parsimony .... 
 
 CHAPTER VII. 
 The value of things . . . . 
 Whether beauty is only delight . 
 Whether morals are matter of feeling 
 Value and existence . . . . 
 
 VALUES 
 
 CHAPTER VIII. THE MIND 
 
 The realistic theory of knowledge 
 The place of psychology 
 Consciousness 
 
 Consciousness and behaviour 
 The nature of consciousness 
 Introspection and its difficulties 
 The spiritual substance 
 Objections .... 
 Realism and the self . 
 
 CHAPTER IX. THE LARGER OUTLOOK 
 The sufficiency of realism ..... 
 Realism and the theory of probability 
 Realism in hypotheses ..... 
 
 Tlie methods of biology ..... 
 Realism in the human sciences : and, first, economics 
 Historical knowledge and the philosophy of history 
 Finding and making in knowledge 
 Imagination and the world of art 
 Representative knowledge ..... 
 Religion and mysticism ..... 
 
 EPILOGUE 
 
 The vision of Thales 
 INDEX . 
 
 PAGE 
 105 
 
 107 
 108 
 115 
 119 
 123 
 
 125 
 126 
 135 
 144 
 
 149 
 150 
 1.52 
 155 
 160 
 162 
 172 
 173 
 179 
 
 180 
 182 
 183 
 186 
 189 
 193 
 201 
 204 
 209 
 211 
 
 218 
 223
 
 CHAPTER I 
 
 INTRODUCTION 
 
 Comme done il est clair que je pense, il est clair aussi que je pense 
 a quelque chose, c'est-a-dire, que je connais, et que j'aper^ois quelque 
 chose. Car la pensee est essentiellement cela. Et aiiisi, ne pouvant y 
 avoir de pensee ou de connaissance sans objet conuu, je ue puis non 
 plus me demander a moi-meme la raisou pourquoi je pense a quelque 
 chose, que pourquoi je pense. 
 
 Arnauld, Des vrayes et desfausses Idees. 
 
 Ihere is no best way of beginning a book, but journeys have 
 to start somehow, and intending travellers expect to be ap- 
 prised of certain matters before they set out. If you would 
 go with us, gentle reader, you have the right to ask what we 
 intend to discuss, and what our chief assumptions are. You 
 will not ask more than this from an introductory chapter; for 
 you are discerning and experienced, dear sir or madam, and we 
 would not address you if you were not. But you cannot ask 
 less, and we cannot do less than comply. 
 
 No philosopher wants to talk about words more than he 
 can help doing in the ordinary way of business, and the retort 
 that philosophy is a wordy business at the best is far too 
 cheap to be worth a glance. There would be some excuse, 
 it is true, and perhaps some little interest, in discussing the 
 various senses in which critics and philosophers have used the 
 word realism. It is a hard-used drudge of a word in art and 
 philosophy (it would turn if a word could), and that is not 
 surprising, for reality is a difficult thing to get away from. 
 Those who try to turn their backs upon it set their faces 
 towards another reality, and those who desert the actual for 
 the ideal soon bestir themselves to prove that this ideal is the 
 only genuine fact. Realists by profession, therefore, are very 
 apt to assume a virtue to which others are equally entitled, 
 and the end of this thing is confusion. If everyone is a realist
 
 2 INTRODUCTION lch. 
 
 after his own fashion, and if the fashions differ, how can the 
 word realism always mean the same thing? 
 
 Plainly it has not always meant the same thing in the 
 mouths of philosophers. In mediaeval times, as we all know, 
 realists disputed with conceptualists and nominalists concerning 
 the logical preeminence and the dynamic potency of Universal 
 Forms. There is so little affinity, however, between the 
 mediaeval and the modern usage of the term realism that 
 even the ghost of this ambiguity has ceased to haunt the word. 
 On the other hand the modern usage is amazingly and un- 
 comfortably protean. If the shade of Reid could visit these 
 regions to-day it would greet Mr Prichard of Oxford, but it 
 would be startled by Mr Alexander, bewildered by Mr Russell 
 and distressed by Mr Holt. Indeed one is tempted to think 
 that any realism defined to the quick becomes nothing but 
 the definer's private philosophy, and that the term itself cannot 
 signify more than an attitude and a tendency. 
 
 Realism in modern philosophy is born in controversy, and 
 its foe is idealism in some form. History repeats itself in this 
 matter, and there is a very clear similarity between Arnauld's 
 reply to Malebranche, Reid's reply to Berkeley and Hume, 
 and Mr Moore's criticisms of Mr Bradley. On the other 
 hand, the three idealisms thus attacked were, after all, very 
 different philosophies, and the Greek rule that a thing is best 
 known by contrast with its opposite has a very precarious 
 value when the ' opposite ' does not remain the same. The 
 choir of heaven and furniture of earth, as Berkeley saw them, 
 look like a cockle-boat on the ocean of the Absolute, and 
 Reid's cudgels use a ruder science than Mr Moore's rapier. 
 
 If anyone were to write a history of realism (and there is 
 room for this enterprise) he would have to take Arnauld very 
 seriously. The ' great doctor,' ' le plus savant mortel qui j amais 
 ait ecrit ' as Boileau's stately epitaph puts it, had too little 
 leisure in his tempestuous career to become a great philosopher. 
 Still, he was eighty-two when he died, and he never understood 
 how anyone could need repose ' when he had all eternity to 
 rest in' ; so he found time to take the lion's share in the Port 
 Royal Logk^ to write the best set of objections to the Medita-
 
 i] INTRODUCTION 3 
 
 tio7is of Descartes and to correspond doughtily and lengthily 
 with Leibniz. His greatest achievement in philosophy, how- 
 ever, was his criticism of Malebranche in a book which he 
 described (perhaps sincerely) as a 'bagatelle,' and entitled Des 
 vraies et desfaiisses Idees. Even those who, like Sainte Beuve^, 
 maintain that Arnauld was no philosopher because they 
 detest his terre-a-terre methods and love the beauty and 
 polish of Malebranche, have to admit that the rigour, strength, 
 and sureness of Arnauld's logic made him an easy victor. His 
 relentless pursuit of Malebranche's doctrine of representative 
 knowledge is still the classic exposure of that theory and would 
 have killed it if philosophers had learned to avoid the mistakes 
 of their ancestors. What is more, Amauld laid the foundations 
 of a comprehensive theory of knowledge, all the more interesting 
 on account of its Cartesian assumptions, and on account of the 
 formal precision of its statement. 
 
 We must hurry on, however, and avoid history except when 
 we need it. But we shall be the better of a little history, and 
 we may approach our subject by a short consideration of Reid's 
 philosophy. 
 
 Reid''s earliest and most interesting book was his Inquiry 
 into the Human Mind on the Principles of' Common Sense. It 
 was a treatise on the problem of perception, and Reid claimed 
 that all previous philosophers had espoused a most vicious 
 fallacy. They all supposed that we perceive, not things them- 
 selves, but their representatives; and Reid tried to show that 
 the rerum simidacra tenuia of Lucretius, the species of the 
 Greeks and the schoolmen, and the 'ideal theory"* of Descartes 
 and Locke, Malebranche, Berkeley and Hume were only variants 
 of this radical misconception. According to Reid, every one 
 of these philosophers believed that perception is a kind of 
 contact between mind and thing, so that anything directly 
 perceived must touch the mind in space, and be present with 
 it at the same moment of time. If so, it is clear that what 
 we call the external world cannot be directly perceived. The 
 
 ^ Cf. his Port Royal, vol. v. p. 449, "Allons! on peut faire d'Arnauld un 
 grand logicien, on en peut faire un cart^sien disciple, et le premier entre les 
 disciples; on n'en fera jamais un philosophe." 
 
 1—2
 
 4 INTRODUCTION [ch. 
 
 sun affects our bodies (and perhaps our minds) when rays from 
 it reach us, but the sun itself does not wander into the optic 
 nerve, and there cannot be any instantaneous compresence 
 between the mind and the sun since the rays take time to 
 travel. It follows that the plain man is mistaken when he 
 supposes that he can see the stars and the hills, or feel the 
 support of good shoe leather ; and the ancients, Reid argued, 
 failed to notice their total disagreement with common sense 
 simply because they con-ected one bad hypothesis with a worse 
 one. They supposed that the circular yellow patch which we 
 see when we look at the sun is the copy or representative of 
 that orb. But the moderns, and especially Berkeley, easily 
 proved that this correcting hypothesis was utterly baseless, and 
 then they were left without any world at all. 
 
 These reflections of Reid's go to the heart of the question, 
 and they might well have proved more disturbing to common 
 sense than Reid supposed. The plain man believes, it is true, 
 that he perceives the sun and the earth, but he also believes 
 that the cause of his perceiving is the fact that the sun and 
 the earth affect his eye and his hand. If he believes further, 
 as in fact nine men do out of ten, that all causal action is by 
 contact, he has a very pretty problem on his hands, quite hard 
 enough to gravel most philosophers. 
 
 The problem was certainly too hard for many members of 
 the Scottish school which Reid founded. So many Scottish 
 clergymen knew that Hume was wrong, so many Englishmen 
 of Dr Johnson''s type found Berkeley''s immaterialism absurd, 
 and so very few of them were able to support their convictions 
 by argument that any attempt at a reasoned defence of common 
 sense fell on very quick ears. There is no other explanation 
 for the immediate success of Oswald's ponderous invective or 
 of Beattie's shallow elegance in his Essay on Truth. On the 
 other hand, Reid himself was neither a furious zealot nor a 
 plain man in enormous blinkers, and Priestley showed little 
 penetration when he arraigned the whole ' triumvirate ' com- 
 posed of the Glasgow professor, the author of The Minstrel^ 
 and the minister of Methven. One can sympathise, indeed, 
 with Priestley's annoyance at ' this sudden torrent of nonsense
 
 i] INTRODUCTION 5 
 
 and abuse that is pouring down upon us from the north' 
 threatening to overturn the sciences and to lead to a state of 
 affairs in which ' the whole business of thinking will be in a 
 manner over, and we shall have nothing to do but to see and 
 believe^,' for most of the partisans of the new philosophy 
 understood it no better than Burns did when he wrote : 
 
 Philosophers have fought and wrangled 
 And mickle Greek and Latin mangled. 
 Till, wi' their logic jargon tired 
 And in the depth of science mired. 
 To common sense they now appeal, 
 That wives and wabsters see and feel 
 
 and that interpretation is unfair, even to Beattie. Indeed, a 
 grand jury of women and weavers would have been too sophis- 
 ticated for some of the arguments given in the name of common 
 sense; and some of Reid's appeals to the constitution of human 
 nature are liable, in principle, to the same condemnation. In 
 their essence, however, Reid's investigations were of a wholly 
 different order from this crude acceptance of everyday beliefs, 
 and there is really no excuse for identifying his philosophy, 
 or any realism, with a blind belief in the existence of matter. 
 The theme of his Inquiry was restricted, it is true, but the 
 Inquiry itself, as Hume said in a letter to Reid, was ' deeply 
 philosophical-,'' and Reid's survey of the mind and the world 
 in his Intellectual Powers was both penetrating and compre- 
 hensive despite its limitations and its occasional inconsistencies 
 on points of detail. 
 
 It is unlikely, indeed, that Reid's influence would have 
 endured so long had there been no salt of philosophy in it. 
 The Inquiry was published in 1764, and as late as 1857 Cousin 
 distinctly stated that any radical departure from Reid's 
 philosophy in Aberdeen, Glasgow or Edinburgh would be a 
 European calamity*. The great influence of Reid's ideas in 
 France during the first half of the nineteenth century began 
 
 1 An Examination of Dr Reid's Inquiry, etc. p. 200 and p. 202. 
 
 2 Hill Burton's Life and Correspondence of David Hume, vol. ii. pp. 153 — 
 154. 
 
 3 In the preface to the third edition of his Philosophic ICcossaise. Cousin 
 omitted St Andrews from the list because Ferrier was there !
 
 6 INTRODUCTION [ch. 
 
 with the chance which led M. Royer-Collard to purchase a 
 copy of the Inquiry at a book -stall near the Seine in 1811, 
 and was stimulated both by Hamilton's ostensible discipleship 
 of Reid, and by the desire of the French people to avoid 
 'quelque importation de la mauvaise metaphysique de 
 rAllemaffne deaeneree^"' This Franco-Scottish alliance, how- 
 ever, could not have been built wholly upon sand and prejudice, 
 and since Cousin's Philosophie Ecossaise is still the best com- 
 mentary on the movement, there is good reason for considering 
 what Cousin said of it. 
 
 Cousin claimed that Reid's discoveries in metaphysics were 
 of the same fundamental importance as Adam Smith's in 
 political economy^, and he found the essence of Reid's discovery 
 and method in a passage at the end of the Inquiry. In this 
 passage Reid contrasted the 'way of reflection' with the 'way 
 of analogy.' All previous philosophers, he maintained, chose 
 the ' way of analogy.' They tried to interpret the mind in 
 the light of inappropriate analogies ultimately derived from 
 the contact of bodies in space, and so they went to their 
 destruction. The ' way of reflection,' Reid continued, avoids 
 this initial fallacy. Its beginnings are set in ' reflection,' and 
 that, in its turn, is just accurate attention to the mind itself. 
 When our mental processes are carefully discriminated without 
 prepossessions, and particularly without the prejudice that 
 results from supposing that explanations of the mind must 
 conform to causal and spatial canons which in fact are wholly 
 inapplicable, the chief problems of knowledge solve themselves 
 
 1 Cousin's phrase, ibid. There is a curious irony in reading these state- 
 ments nowadays, and the reader may be interested in the similar attitude of 
 Scottish theologians in those times. "For those who are not inclined to 
 study German philosophy" Dr Chalmers said a few years earlier "I do not 
 recommend that they should suspend for it their ordinary readings. Their 
 very ignorance of the German idealism, the very confinement of their 
 mental philosophy to the doctrine and metaphysics of the Scottish school, 
 are guarantees in themselves against the deleterious influence of these out- 
 landish speculations" (Fraser, Biographia Philosophica, p. 74). Chalmers, 
 for his part, preferred ' plain Scottish boluses ' ; he was convinced that 
 ' the unintelligible does not always imply the solid or even the profound ' ; 
 and of much more to the same effect. He preferred Kale to Sauerkraut. 
 
 2 Philosophie Ecossaise, Avertissement, p. ii.
 
 i] INTRODUCTION 7 
 
 in the sense that accurate observations followed by careful 
 reasoning give an answer that can neither be impugned nor 
 rejected. "When the operations of the mind are exerted, we 
 are conscious of them, and it is in our power to attend to them, 
 and to reflect upon them, until they become familiar objects 
 of thought. This is the only way in which we can form just 
 and accurate notions of those operations^"" 
 
 In itself, this account of the spirit of Reid's enterprise, 
 does not differ importantly from the programme of Locke's 
 Essay -^ and Reid's hint, later on in the same chapter, that this 
 method of direct deduction from the phenomena without 
 analogy or hypothesis had been attended by great success in 
 the domain of physics is thoroughly characteristic of the 
 eighteenth century. The glamour of Newton's achievements 
 led all the philosophers of that age to have great hopes of 
 experimental inquiries into human nature. "When Reid was a 
 student in Aberdeen he learnt as much as that from his master 
 Turnbull^, and the subtitle of Hume's Treatise declares it 
 in so many words*. There is nothing peculiarly distinctive, 
 therefore, in Reid's conception of his task. His merit lies in 
 the tenacity with which he clung to the phenomena he found, 
 and in his refusal to be fobbed off with anything else. 
 
 What, then, are these phenomena.? It would seem from 
 the above quotation that Reid took them to be the operations 
 of the mind, or, rather, those mental operations which are 
 specifically concerned with the business of knowing. If so, 
 he deserved great credit for his thorough and searching survey 
 of these complex and varied operations in his Intellectual 
 Powers, and for his courage in insisting, to the point of tedium, 
 on the fundamental doctrine that these operations should be 
 studied for themselves alone and should not be supposed to 
 have the characteristics of other things unless and until they 
 
 ^ Inquiry, Hamilton's edition, p. 201. 
 
 2 George TurnbuU (1698—1748) Wcas a regent of Marischal College, Aber- 
 deen, from 1721 till 1727. Eeid's name was on his roll in 1726. Turnbull 
 wrote many books, and his Anticnt Paintings is one of the unfortunate tomes 
 ■which the porter found too heavy in Hogarth's picture. 
 
 » "A Treatise of Human Nature: Being an Attempt to introduce the 
 experimental Method of Reasoning into Moral Subjects."
 
 8 INTRODUCTION [ch. 
 
 have been proved to have them. On the other hand, the bays 
 which he rightly earned on this account would make Ijut half 
 a diadem. The operations of the mind that is bent on knowing 
 are only a part of the relevant phenomena. While there is 
 remembering, supposing and believing of the one part, there 
 are the things remembered, supposed or believed of the other 
 part. Anyone, that is to say, who sets himself to reflect upon 
 the operations of the mind in knowledge, has also to reflect 
 upon the objects before the mind, and anyone who distrusts 
 specious analogies concerning the process of knowing should 
 also distrust elusive and figurative descriptions of the objects 
 which in fact are known. He must examine and consider most 
 scrupulously what it is that we apprehend in any given instance, 
 instead of arguing that we must apprehend this or the other 
 kind of thing because our theories of the universe, untested by 
 observation, have it so. 
 
 Reid's detailed investigations (and the concluding chapter 
 of the LTquiry, for that matter) show that he had grasped this 
 double aspect of his problem very firmly indeed, even if some 
 of his definitions incautiously omit it. If this be allowed, Reid's 
 work as a whole is a sane and resolute application of the funda- 
 mental principle of any realism. For realism is a theory of 
 knowledge whose essence is to supply a complete phenomeno- 
 logy of knowing and of things known, or, in other words, to 
 make an accurate and thorough survey both of the processes 
 of knowing and of the objects directly known through these 
 processes. 
 
 The trouble is, of course, that so many philosophies make 
 precisely the same claim. Did not Hegel write his Plieno- 
 menology, and do not James or Bergson or Avenarius give us 
 a philosophy of pure experience, each in his several way ? All 
 these philosophers, it would seem, want the same thing and 
 they attain something very different. There must therefore 
 be something peculiar and distinctive in realism to explain its 
 difference from these other philosophies. 
 
 This distinctive thing, I suppose, is an affair of assumptions, 
 and, perhaps, of hopes and expectations. The main assumption 
 of realism is that things can be known as they really are. The
 
 I] INTRODUCTION 9 
 
 secondary, but scarcely less important assumption, is that 
 anything is precisely what it appears to be when sufficient 
 precautions have been taken to avoid confusion between the 
 actual genuine appearance and spurious though very plausible 
 glosses upon it. It follows, of course, that these genuine 
 appearances cannot contradict one another; for things cannot 
 contradict one another. It also follows that true knowledge 
 of this or the other thing need not logically imply the knowledge 
 of all its conditions. To say that things can be known means, 
 of course, that they can be known by us. We, however, are 
 finite beings, and so we cannot hope to know more than a very 
 small part of the infinitude of existence. Peo' contra^ we have 
 no right to deny the usual, and, in all probability, the very 
 just belief that everything in the universe has strictly infinite 
 ramifications, so that, if we were sagacious enough, we might 
 pass from cats to clover and from clover to the stars. Similarly 
 we have no right to deny the orthodox assumption of psychology 
 that any piece of thinking is a subtle web whose pattern, 
 perhaps, was woven long before the days of our eolithic 
 ancestors, and whose yarn, even now, is three parts spun in a 
 blind loom of miles of branching nerves. Thus if we know 
 anything as it really is we must be able to know it despite the 
 fact that we do not know much that pertains to it in the way 
 of conditions and connections. 
 
 These assumptions distinguish realism very sharply from 
 the Anglo-Hegelian idealism which was lately dominant and 
 still is fashionable in these islands. Even the Oxford idealists, 
 however, might find a meaning for them which they would 
 consider tolerably innocuous and moderately true ; and the 
 pragmatists or M. Bergson might contrive to accept them 
 totidem verbis. Some further explanation is needed, therefore. 
 The statement that things can be known as they really are is 
 simple in appearance only. We need not stay, it is true, to 
 consider what is meant by a 'thing,*' for 'things' in this general 
 statement must clearly be understood in the most general sense 
 possible. Any entity whatsoever that can be apprehended by 
 the mind is a ' thing ' in this sense, so that rainbows, dream 
 castles, a yearning for Nirvana, and the null-class are included
 
 10 INTRODUCTION [ch. 
 
 in the statement as well as the ships and the rifles which take 
 part in the executive order of the physical world. On the 
 other hand, two points at least require special discussion. 
 
 To say that things may be known does not tell us what 
 their reality is. That is a problem for investigation, not 
 something that can be defined in advance, and, of course, there 
 is no implication that knowledge can be satisfied in all its 
 enterprises. There may be many things which we cannot 
 begin to apprehend. A being like Voltaire's Micromegas, for 
 example, with his thousand senses, would be acquainted with 
 more than nine hundred and ninety kinds of sensible qualities 
 from which we are cut off. Again, there are many things of 
 which we know only that they are, not xohat they are. The 
 meaning of the statement, then, is only that there is nothing 
 in the relation between the mind and things which, of itself, 
 makes anything inaccessible to knowledge. To put it other- 
 wise, the reason for ignorance never lies in the ineptitude of 
 knowledge. It is due, when it occurs, simply to the empirical 
 iact that the mind either does not apprehend these things, or, 
 for some reason of fact, is not in a position to apprehend them. 
 A blind man should blame his eyes and not his mind when he 
 cannot see the sunset. 
 
 What, then, is this knowledge for which so much is claimed.'' 
 According to M. Bergson, true knowledge is intuition \ and 
 that, in its turn, is a process of union and becoming. The 
 man who grasps anything by intuition worms his way into the 
 very being of that thing until it is incorporated into him and 
 he into it. We know a thing by becoming it, and it is known 
 by becoming us. Others, again, maintain that knowledge of 
 a thing is the possession of an image or representative of it, so 
 that we know anything when we possess certain pictures or 
 tokens, and not otherwise. The pragmatists, for their part, 
 are shy of such theories because they do not take knowledge 
 very seriously. They consider it a temporary adjustment 
 between ourselves and our environment, a useful compromise 
 which enables us to get along ; and from that point of view 
 
 ^ See his Introduction to Metaphysics, passim.
 
 i] INTRODUCTION 11 
 
 it is only idle fancy to believe that anything could be finally 
 and utterly what we take it to be. 
 
 Realists, however, deny all these theories, though they ad- 
 mit a subsidiary and consequential truth to some of them. The 
 first they deny altogether. Knowledge, they think, is never a 
 kind of identity, and they are apt to choose very common- 
 place illustrations to support their contention. We do not 
 become Niagara by looking at it ; we do not become the past 
 by remembering the Great War ; we do not become a set of 
 figures by contemplating the multiplication table ; and so on. 
 On the contrary, if we became these things we could not know 
 them at all. According to realists, the process of knowledge 
 always implies that the mind is confronted with an object, and 
 always implies that we are never under any conceivable circum- 
 stances identical with that object. Even when we apprehend 
 our own experiences, the process of apprehension cannot be 
 identical with the experience which is apprehended. 
 
 Realists therefore deny the reality of intuition in M. 
 Bergson''s sense but they need not be quite so intransigeant in 
 respect of the other theories. They need not deny that much 
 of our knowledge is merely representative. What they deny 
 is, firstly, that knowledge means representation, and secondly 
 that representative knowledge could occur without a direct, 
 non -representative basis. If knowledge meant representation, 
 statues of dead men would know the dead men, and the still 
 pools would know the clouds and the trees which they reflect. 
 Indirect or representative knowledge, again, implies direct 
 acquaintance at some point. The collector who finds an ancient 
 coin, for example, has only an indirect acquaintance with the 
 potentate whose image is stamped upon it, but he is directly 
 acquainted with the coin, and he could not know that the 
 impression stamped upon it really is an image unless he were 
 able to compare some portraits with some originals from direct 
 acquaintance with both. And realists are in earnest concern- 
 ing truth and knowledge while pragmatists are not. They 
 need not deny, indeed, that knowledge is useful precisely 
 in proportion as it affords guidance here and now, or that a 
 lucky guess or a vague approximation may often work as well
 
 12 INTRODUCTION [ch. 
 
 as, or even better than, well-grounded knowledge. What they 
 are bound to deny is that the mere fact of being guided by 
 ideas,expectations,preiudicesandthelikeis</«^?-^/b7T knowledge. 
 We work as we can, and how we can ; but that does not tell 
 us how or what we know. 
 
 To take knowledge seriously is to believe that the ap- 
 pearances presented to the mind can be saved from contradiction 
 if they are attentively and judicially discerned. The pheno- 
 menology of knowledge, in other words, in its twofold division 
 into apprehension and thing apprehended, may very well be 
 final truth ; and, in particular, it need not be a compromise 
 or only a provisional, approximate makeshift. Indeed, the 
 reasonable hope and expectation of most realists is that certain 
 at least of the tried and tested results of scientific investigation, 
 and some of the most stubborn beliefs of workaday life, will 
 prove to be as true ' in the end ' as at any other stage, that is 
 to say they will prove to be quite true. The truth of logic is 
 the crucial test, for if logic chokes you, what will you drink ? 
 According to the absolutists and the pragmatists any logical 
 argument and any part of one cannot be quite true because it 
 and its parts are always subject to qualification in a wider 
 context. For realists, on the contrary, each link in a chain 
 of logical argument m.ay be true in its own right and the whole 
 chain, consequently, need not be a thing of cobwebs. The 
 validity of pure mathematics, again, stands or falls with the 
 validity of logic, and a philosophy which is bound to impugn 
 mathematics has little chance of certainty on its own account. 
 It has to substitute something much less certain in the place 
 of mathematical certainty ; and the goddess of Reason will 
 see to its undoing. Other beliefs, perhaps, are not so likely to 
 remain unshaken. The belief in matter and the belief in the 
 soul have commonly chaff as well as grain in them, and common 
 sense may not relish the process of winnowing. Reid's sieve, 
 for example, was probably far too wide, and he was more 
 scrupulous than most. But if the world in which we think we 
 live sometimes seems to take wings to itself when even a 
 realistic philosophy startles it, there is a fair presumption that 
 such flights may be short.
 
 i] INTRODUCTION 13 
 
 Horace was not the first to see that nature has a way of 
 returning after being driven out with a pitchfork, and this 
 furtive habit of nature's partially explains the periodic and 
 independent return of realistic philosophies in modem times. 
 As we have seen, the first of the major realisms was Arnauld's 
 attack on Malebranche, the second was Reid's and the third 
 was the contemporary movement. All these were independent 
 of one another, all attempted to refute a certain variety of 
 idealism, and all, in a sense, strove to rehabilitate familiar 
 ways of thinking at the expense of what they considered to be 
 ungrounded and overweeningpretensionson thepart of accepted 
 philosophy. The manner of Arnauld's criticism of the theory 
 that 'we see all things in God' may have been somewhat grim 
 and pedestrian, but his thrusts were shrewd, and the main 
 principles of his strategy were very similar to Reid's. Reid, 
 for his part, recoiled from the scepticism in morals, religion 
 and science which he found implicit in the ' way of ideas ' 
 despite Berkeley's piety and Locke's cautious tolerance. The 
 contemporary movement, in its turn (including Mr Bertrand 
 Russell's earlier writings though not his later), rejected the 
 conclusions of Absolutism. This modem movement, it is true, 
 did not set out in the first instance to rejuvenate the body of 
 common sense beliefs, although Mr Moore, in his early papers, 
 maintained that ' matter ' could be directly perceived, and 
 although Mr Russell sometimes ironically called himself naive. 
 On the contrary, the main interests of the new realism were 
 sternly logical (except in Oxford where Reid's tradition had 
 remained active), and Mr Russell came to follow Mr Moore 
 because he became convinced that the main assumptions of 
 this realism were ' quite indispensable to any even tolerably 
 satisfactory philosophy of mathematics \' 
 
 It is plain that any realism of this kind, even if it defends 
 common sense, defends a common sense which is very sophisti- 
 cated indeed. A philosopher's attitude, in a word, must always 
 differ from the plain man's. The plain man accepts or rejects 
 conclusions en bloc. He may be right in this just as he may 
 be right in believing that any form of idealism is only a fable 
 ^ The Principles of Mathematics, Preface, p. viii.
 
 14 INTRODUCTION [ch. i 
 
 convemie, but these wholesale methods are not philosophy. 
 Philosophers look for reasons and (perhaps unfortunately) need 
 the dialectical skill which technical palaestrics in philosophy 
 demand. Judged by this criterion, Arnauld, Reid and many 
 contemporary realists are quite as good philosophers as their 
 opponents. 
 
 The fundamental point on which Arnauld, Reid and 
 Mr Moore agree is that the object of true knowledge is in a 
 certain sense independent of our knowing of it. This inde- 
 pendence does not mean unrelatedness, for everything has some 
 relations to everything else. It means that the fact of being 
 known does not imply any effect upon the character or existence 
 of the thing which is known. If it did, nothing could be known 
 as it is in itself ; for everything in that case would be changed 
 simply because it had become known. All idealists, in spite 
 of their differences, dispute this independence of the objects 
 of knowledge, even when they give a qualified and hesitating 
 blessing to what their opponents call ultimate fact, or when 
 (by a most insidious device) they try to soothe their critics by 
 apparent acquiescence and then divert the argument to another 
 plane. According to realism, the plane of observation and logic 
 is the only possible plane of truth. 
 
 I intend to make this assumption in the sequel, and to con- 
 sider, in some detail, whether it is consistent with the facts of 
 perception, memory, imagination, and similar processes. And 
 I shall try to describe what it is that we perceive or imagine 
 or believe. The general thesis of realism is that knowledge is 
 a kind of discovery in which things are directly revealed or 
 given to the mind. This statement, to be sure, is not very 
 precise, but perhaps it may achieve greater precision as we 
 proceed.
 
 CHAPTER II 
 
 THE THINGS WE PERCEIVE 
 
 In my view, a thing is what it looks, and looks what it is. 
 
 John Grote, Exploratio Philosophica. 
 
 JliVERYBODY knows that philosophers debate interminably 
 whether or not the external world can be directly perceived. 
 Everybody expects a philosopher to talk about this question, 
 just as everybody expects linnets to sing ; and most people 
 would add that the orhis terrarum need not heed the dispute. 
 The truth is that philosophers have no choice. They are bound 
 to investigate this problem whether they will or no. 
 
 The question itself seems very straightforward, and the way 
 of setting about to answer it reasonably clear. We have to 
 ask what perception is and whether it is direct. If it is not 
 direct the question falls ; but if it is, the next step in the 
 argument is to enquire what precisely is directly perceived ; 
 and when we have satisfied ourselves on this head it only re- 
 mains to consider whether or not this direct perception is 
 literally the discovery of an external world. 
 
 There is little difficulty in recognising what perception is. 
 A man can scarcely miss it, for he perceives every time that 
 he glances at the books on his shelves, or hears a dog barking, 
 or smells the new-mown hay. 
 
 Still, some explanations are required. In the first place, it 
 is necessary to distinguish perception from judgments of 
 perception. When a man says 'Thafs my Apuleius,'' or 'That's 
 the scent of clover' he not only perceives, but makes a judg- 
 ment based on perception. Perception is the apprehension of 
 the fact on which such judgments rest, and that is the 
 simplest way of describing it. These judgments are based 
 upon apprehended fact which is present and sensory. Per- 
 ception is not memory or expectation, and it always implies 
 the use of one or more of the senses. 
 
 In the second place, it is usual to distinguish perception
 
 16 THE THINGS WE PERCEIVE [cii. 
 
 from sensation. This distinction is not so clear as the former^ 
 and depends upon two different ways of looking at the matter. 
 According to one of these ways of looking at it, anything 
 perceived is a complicated affair in which several distinct 
 elements can be discerned by analysis. Thus the perceived 
 thing corresponding to the judgment 'That's my dog Argus'" 
 certainly contains a structure of space and time, and a mani- 
 fold of sensory qualities. It would do so even if the Argus 
 perceived at the moment were little more than a series of 
 barks, for these barks would have an Argus-meaning for the 
 dog's master. Sensory elements of this kind, therefore, must 
 be distinguished from the object of integi-ated perception. 
 Such elements probably do not exist in isolation, but they 
 can be discerned analytically. 
 
 The other way of distinguishing sensation from perception 
 does not coincide with this one, and often seems iiTelevant, 
 It results from regarding the facts in the light of their de- 
 velopment. Adult perception, it is plain, has a history. The 
 infant, looking round him from his cot, does not perceive 
 quite the same things as he will three years later. What a 
 patient perceives when he is recovering from an anaesthetic 
 is a shadowy booming mass, and not the crisp outlines or the 
 clear tones he will notice in half an hour or so. And so on. 
 It would be absurd to suppose, however, that these primitive 
 or confused sensory appearances are identical with the sensory 
 elements of the former argument. On the contrary, the ana- 
 lytic discrimination of these sensory elements only occurs at 
 a highly developed level ; and, for the rest, it is hard to see 
 why the history of perception should be allowed to correct 
 or to supersede the description of it. Perception has de- 
 veloped from a more primitive form of apprehension which 
 may, perhaps, be fitly called sensation ; but this circumstance 
 seems irrelevant to the description of what perception is. 
 There are no Melchizedeks among us nowadays, without 
 father, without mother, and without descent ; but surely we 
 can see for ourselves whether a gardener has done his job 
 without reflecting that he could not have done it if he had 
 not had a grandmother.
 
 ii] THE THINGS WE PERCEIVE 17 
 
 Assuming, then, that we know what perception is, we have 
 to ask, in the next place, whether it may be direct ; and the 
 answer to this question is not doubtful. Apprehension is 
 always direct unless there are intermediaries between the 
 thing apprehended and the apprehending of it. Now it is plain 
 thatwe are always confronted with somethingin perception, and 
 that there need not be anything between our minds and it. The 
 thing which confronts us, to be sure, may also be the cue to 
 association and inference, but that is not to the point. When 
 I perceive, let us say, a coloured patch, I am directly and 
 immediately acquainted with this patch, and no process of 
 argument can overthrow this palpable certainty. Direct per- 
 ception occurs, therefore. What, then, is directly perceived "^ 
 
 There are many answers to this question, and it would be 
 useless to try to enumerate them all. It seems plain, however, 
 that some of these answers either do not describe perceptible 
 facts at all, or else profess to give information which is wholly 
 additional to anything perceptible. Anyone who maintains, 
 for example, that we perceive brain changes or nervous shocks 
 on the one hand, or ' primary ' qualities without ' secondary ' 
 on the other, simply does not describe the facts perceived. 
 We do not perceive the pulpy hemispheres within our skulls 
 or any tremors at the synapses between the neurons in the 
 cortex, and we do perceive some part at least of the things 
 we call chairs and tables, stars and medallions. Similarly we 
 do not perceive colourless shapes or soundless vibrations ; and 
 those who maintain that sweetness and colour are figments of 
 the mind while figure and motion are immitigable realities 
 have perforce to admit that we cannot perceive the figments 
 without the realities or the realities without the figments. 
 Those again who tell us that we perceive states of our own mind, 
 or of God's mind, or of a cosmic Super-Experience plainly do 
 not describe a fact of direct perception, like the colour saffron 
 or the flavour of pine-apple ; and those who say that we per- 
 ceive matter, and that matter is composed of electrons, do not 
 describe what we perceive but give us an argument in its 
 place. The imperceptible, ultra-microscopic constitution of 
 perceived things may be as they say, but the things we 
 L. 2
 
 18 THE THINGS WE PERCEIVE [ch. 
 
 perceive are macroscopic, and so they must be described in 
 macroscopic terms. 
 
 When views of this kind are omitted, the choice of alter- 
 natives becomes much more Hmited, and some of the more 
 important of them may be considered without further parley. 
 I propose to begin by considering a very fashionable theory 
 which I shall call the theory of sensory atomism. According to 
 this theory we perceive sense data and we perceive nothing else. 
 
 We arrive at the notion of sense data by paring away some 
 of the imperceptibles which are commonly supposed to belong 
 to perceiving. When the plain man says that he sees a golf 
 ball he might describe his meaning by saying that he perceives 
 a spherical white object which is capable of responding to his 
 attentions in a way that everybody knows well enough. It is 
 clear, however, that when he looks at the golf ball, he cannot 
 perceive, at that moment, what it would do if he hit it. Golf 
 would be a very different game if he could. It is also clear 
 that he cannot look at every part of the surface of a sphere 
 all at once ; and, granting that he could touch the whole sur- 
 face at once if he grasped the ball, even then he could not 
 touch the interior of it. 
 
 The interior, then, and many other parts or qualities which 
 we commonly ascribe to golf balls (not to speak of their uses) 
 are not directly perceived at any one instant of perception. 
 Again, what is perceived by one sense differs from what is 
 perceived by another. The colour of the ball, for instance, is 
 not touched, and it is a commonplace of psychology that sight- 
 space and touch-space are not identical, despite the general 
 similarity between the two. It is impossible to join a visual 
 inch and a tactual one : the objects of touch-space do not grow 
 smaller as they recede : and so on. Hence it would seem that 
 the plain man's idea of a golf ball should really be resolved into 
 a series of visual, tactual, and other separable data. The golf 
 ball of common sense is a collection of perceptible entities, 
 including all the coloured patches which we perceive when we 
 look at it from different angles and positions, the touch-sphere 
 that may be grasped, and the like. 
 
 The members of this collection are sense data, and these,
 
 ii] THE THINGS WE PERCEIVE 19 
 
 it is claimed, are ultimate facts which cannot be resolved by 
 any alchemy. We are aware of them in perception, and we 
 may discern them readily enough if we take the trouble. Sense 
 data, then — a fleeting yellow patch, a glabrous evanescent 
 contour, and so forth — are what we actually perceive, and, 
 in the last analysis, we perceive nothing else. 
 
 It is plain that this theory is amply justified in what it 
 affirms. Sense data are perceived, and no theory which neg- 
 lects their variations and their differences is worth considering. 
 On the other hand there may be a snare in it concerning what 
 it denies. Granting that we perceive sense data, is it true 
 that we perceive nothing else ? Is the theory of sensory 
 atomism a complete description of the facts perceived '^ 
 
 The examples previously given are typical of those which 
 are usually selected by the sensory atomists. They ask them- 
 selves what we perceive, and they are not content with the 
 answer that we perceive inkstands and pennies, buttercups and 
 pebbles. For they proceed, in their questioning mood, to 
 draw further distinctions. They ask. What is the seen inkstand, 
 the J'elt penny, the smelt buttercup.? And they conclude in 
 the end that the seen inkstand is a shining silvery patch, the 
 felt penny a rounded cool disc, the smelt buttercup a mere 
 aroma, and so of the rest. These patches and odours, there- 
 fore, are the ultimate units of their theory. 
 
 At this point objections begin to be heard. Ordinary 
 psychologists (sophisticated, perhaps, and pedantic, but honest 
 observers according to their lights) would state in reply that 
 the shining silvery patch is not an ultimate unit but only a 
 part of a perceived whole, that the rounded cool disc is felt 
 along with its surroundings, and so on. 
 
 The sensory atomists shrug their shoulders at this objection. 
 Their point is that they certainly do perceive the silvery patch, 
 whatever else they may see at the same time, and they submit 
 that the rest of what they see at any moment is just another 
 set of coloured patches. When they have to meet the objection 
 that the colour of the suri'oundings certainly modifies the 
 colour of any perceived patch, they reply that this does not 
 matter. They know very well, they say, that a disc which is 
 
 2—2
 
 20 THE THINGS WE PERCEIVE [cii. 
 
 grey within a white border may be reddish within a green 
 border; but that is a statement in popular language which 
 really means that the same stinndi may give rise to different 
 sense data when the surrounding stimuli are altered. It is not 
 true that one and the same sense datum changes in these 
 cases. On the contrary the sense data become different because 
 of the experimental change in the stimuli. In this experiment 
 a grey patch is seen at one time and a reddish patch at another 
 time. And that, precisely, is the sensory atomist's contention. 
 
 Serious difficulties remain, however. To begin with, there 
 are the facts of continuous sensory adjustment and discrimi- 
 nation. The examples chosen (principally in the field of vision) 
 are snap-shot glimpses, focussed and discriminated. Careful 
 psychological investigation shows, however, that these snap- 
 shots fluctuate so much that recognisable changes occur within 
 a fraction of a second. Anyone who looks steadily at a brass 
 candlestick, for example, quickly discovers that the sense data 
 are not at all steady. The solid candlestick is a quivering 
 slough in terms of visual sense data. There are perpetual 
 slight oscillations in it, perpetual shifting of discriminated 
 brightness, and so forth. 
 
 Now whatever we mean when we say that we perceive a 
 candlestick we certainly do not mean that we perceive a 
 fluctuating thing, and there must be some legitimate meaning 
 in our assertion that we perceive a candlestick which is steady. 
 In view of this objection the sensory atomists tell us either that 
 we infer the steadiness or neglect the fluctuations. According 
 to the first of these alternatives, although the sense data per- 
 ceived are fluctuating, we infer, more or less illegitimately, that 
 the candlestick is steady. According to the other alternative 
 we neglect the fluctuations, so that the candlestick seems to 
 look steady although it does not really look so. And neither 
 alternative is at all satisfactory in view of the original theory. 
 
 The inferential theory is not a description of the facts, and 
 it forces the sensory atomists to run with the hare and to 
 hunt with the hounds. There is certainly no conscious in- 
 ference in the case, and the judgment that we perceive a 
 steady candlestick seems to be a direct transcript of the facts
 
 ii] THE THINGS WE PERCEIVE 21 
 
 perceived just as much as the judgment that we perceive a 
 yellow shape. Anyone who argues, therefore, that the yellow 
 shape is given fact and the steadiness spontaneous inference, 
 substitutes a theory for a description, and yet the sensory 
 atomists claim to dispense with everything except incorrigibly 
 hard sense data. They have to shift their ground, therefore, 
 in order to meet the facts, and if they try to argue that the 
 steadiness of the candlestick is an acquired meaning whereas 
 the yellowness is ultimate bare fact, the retort is obvious and 
 crushing. What right have sensory atomists, of all people, to 
 draw this distinction ? In all probability the power of dis- 
 criminating yellow is also acquired; for all the senses, if our 
 evidence can be trusted, spring from a blurred matrix, and 
 it is quite impossible to tell in these cases what precisely is 
 acquired and Avhat is not. To make a beginning with their 
 argument, sensory atomists have to leave hypotheses concerning 
 development on one side, and accept the data which have become 
 hard. The hard fact which is found, expressed as directly as 
 possible, is nothing short of a yellow, steady, solid candlestick. 
 The alternative of maintaining that these fluctuations are 
 perceived but neglected has the merit of calling attention to 
 facts which are true and important, but it leads, in principle, 
 far beyond the point where sensory atomism conmionly stops 
 short. Sensory atomists usually accept Stumpfs argument in 
 favour of undiscriminated differences in sense data. Briefly 
 stated, this argument runs as follows. A weight of four lbs., 
 let us say, may be discriminably different from a weight of four 
 lbs. two oz. but a weight of four lbs. one oz. is not discriminably 
 different (normally) from either of these extremes. Hence the 
 intermediate weight appears to be equal to two things which 
 do not appear to be equal to one another, and the natural if 
 not the inevitable conclusion is that the three sets of sense 
 data are all really different though only two of them ap})ear 
 to be different. It is clear, however, that the principle of this 
 phenomenon reaches very far indeed. As Ruskin tells us, the 
 painter has first to recover the ' natural innocence of the eye "* 
 before he can leaoi to paint. It takes training to see what 
 things really look like, just as it takes a special effort to notice
 
 22 THE THINGS WE PERCEIVE [ch. 
 
 that the candlestick always fluctuates a little when we look at 
 it steadily. Some psychologists even argue, like Mr Pillsbury^ 
 that we usually perceive types and not individuals. 
 
 To be sure, it is possible to maintain that the painter comes 
 to have different sense data as the result of his ambition to 
 attain a state of innocence, that the candlestick does not 
 fluctuate until we make it do so by bringing the expectations to 
 bear upon it which meddling psychologists have taught us, 
 and that untrained persons really do perceive types and not 
 individuals. If that be true, however, we are far away from 
 the chosen ground of sensory atomism. The sense data which 
 sensory atomists take to be the last results of ruthless attention 
 to fact either disappear in large measure or become exceedingly 
 doubtful from this new standpoint. They are either ' neglected ' 
 or non-existent, and many of them are discerned only in the 
 special case in which sensory atomists play tricks with their 
 own perceptual apparatus. 
 
 Very similar conclusions follow from considering the margin 
 of perception. Since the sensory atomists have no right to 
 base their theories on anything less than the whole of what 
 we perceive at any moment, they have to reckon with this 
 margin. There are very marked differences, however, between 
 the focussed centre and the penumbral fringe of the visual 
 field at any moment, and their theory seems to have little 
 room, if any, for these differences. Sensory atomists tell us 
 that we perceive a yellow shape, diversely illuminated, w^hen- 
 ever we look at what we call a candlestick, but, in fact, we 
 never perceive the candlestick quite alone without surroundings, 
 and the character of these perceived surroundings is very in- 
 teresting and highly important. Imprimis this border or 
 fringe is very little discriminated and, except by a special 
 effort, is scarcely discriminated at all towards its periphery. 
 Amplius, it has no crisp boundary although it does not stretch 
 out to infinity. Praeterea, when there is a change in attention 
 (or in sensory adjustment, or otherwise), focus and fringe 
 change places, and yet this difference is commonly neglected 
 
 * The Ptychology of Reasoning , especially pp. 73 sqq. referring to Lehmann's 
 experiments, and to Dr Hayden's and Schumann's.
 
 ii] • THE THINGS WE PERCEIVE 23 
 
 and is never taken to mean that a new thing has come into 
 being. AdJwc, the margin, in critical respects, contributes to 
 the meaning of the focussed, central area. 
 
 Let us suppose, for instance, that the candlestick is in the 
 centre of our visual field. In that case the candlestick is local- 
 ised as well as coloured and shapely. This localisation is 
 relative to the rest of the perceived field and to the body of 
 the percipient (in so far as he is aware of his body), and we 
 shall consider some of the implications of this circumstance 
 before long. For the moment the important point to notice is 
 that the localisation is perceived as directly as the colour or 
 the shape. Because the candlestick is localised the margin 
 counts as well as the centre, and yet the sensory atomists"' 
 description of sense data takes account of nothing except the 
 yellow shape in the centre. 
 
 The margin, to be sure, is usually neglected in perception, 
 but it is perceived none the less, and it is quite absurd to 
 describe the field of vision as if it were a mosaic of sense data 
 M hich appear the same whether they are focussed or not. We 
 do not, indeed, perceive a world all at once, but we certainly 
 do perceive an expansive pattern, clear in the centre and 
 nebulous towards its misty border. Moreover, when we alter 
 our attention within this perceived expanse, we perceive a 
 difference of outline and emphasis within the same field and 
 not two different sense data (as the sensory atomists aver). 
 Again, the very nebulousness of the margin is rich in con- 
 sequences. If, for example, we turn our eyes from the candle- 
 stick on the mantelpiece to the tiles of the fire-place, something 
 more has happened than a mere change of emphasis within 
 the same field ; for the passage of our eyes continuously annexes 
 a new field only partially coincident with the old one. What 
 was formerly central is now marginal, what was formerly 
 part of the margin is now the centre, and the new mai'gin 
 has come to include something that was previously not per- 
 ceived at all. Yet because the earlier field had no crisply 
 defined boundaries we always suppose that we are exploring 
 more of the same world, and not that we have come to per- 
 ceive a fresh sense datum part of which is similar to a former 
 one; and this reference to a world beyond the expanse directly
 
 24 THE THINGS WE PERCEIVE [ch. 
 
 perceived at any moment is always part of the meaning of 
 anything whatever that we perceive. 
 
 Furthermore, the sensory atomists, in distinguishing so 
 trenchantly between the sense data of sight, touch, sound and 
 the rest, omit the crucial factor which psychologists call 
 ' complication,"' and the reader, in his charity, must pardon 
 me for introducing a little elementary psychology at this 
 point. When a critic talks of the ' tactile values ' of a painting, 
 or a schoolboy says that ice looks cold, some cantankerous 
 purists might point out that neither coldness nor tactile 
 values can, strictly speaking, be seen, and consequently that 
 the correct way of putting it is to say that the coldness or 
 the tactile values are inferred from what is seen or associated 
 with it. In point of fact, however, these purists would be 
 wrong. The coldness of the snow is perceived at a distance 
 just as directly and immediately as its whiteness or its lustre, 
 for it is part of the meaning of the wintry landscape which is 
 perceived. The visual sense datum is as much a sign as Sifact 
 and it is always apprehended so. The meaning has been 
 acquired, to be sure, but we perceive it when it is acquired. 
 I do not mean, of course, that we perceive what is meant 
 precisely in the way in which we perceive meaning. A sign 
 is never what it signifies. I mean that we perceive significance ; 
 and, indeed, that we always perceive sign^/acts, not sense 
 data devoid of meaning. The snow has a cold look, and this 
 cold look belongs to it in precisely the same sense as its 
 surface or shade, its brightness or its colour. 
 
 This example has been chosen from the sense of sight, and 
 the other senses make the argument still more forcible. The 
 blending of taste and smell is so notorious that it is manifestly 
 absurd to analyse the taste of roast beef into so many data 
 of sweetness and sourness plus the characteristic flavour of 
 that succulent substance. Again, this blending of sensory 
 qualities is so hopelessly embrangled with tactual and even 
 with visual meaning that it is impossible to say whether 
 tastes and smells are literally extended or merely correlated 
 with extended things. Indeed, it is impossible to say whether 
 these qualities appear to be ' external ' or not. No one is likely 
 to be in two minds on the question whether the yellowish
 
 II] THE THINGS WE PERCEIVE 25 
 
 liquid that he calls whisky is inside his body or outside it, 
 but anyone may wonder interminably whether the smell of 
 the whisky is inside the glass or inside his nostrils, and any- 
 one may argue without end whether the shrewd wind is itself 
 chill or only makes human beings feel so. The significance 
 of these sign-facts is so thoroughly wrapped up in them that it 
 is useless to try to uncover their signlessness. 
 
 Our conclusion, then, is that we perceive more than mere 
 sense data as these are commonly described by the sensory 
 atomists. We always perceive meaningful data which are as 
 much signs as facts. It w ould be a mistake, however, to treat 
 the objects of perception as mere signs, although most of the 
 scholastics and many of the moderns have done so. Particular 
 shapes and colours (to mention no others) are directly per- 
 ceived, and, when perceived, are unquestionable facts. On the 
 other hand they also signify a w ider context in time, space and 
 material. Thev are signs, although they are not mere signs ; 
 and their significance is not extraneous or adventitious to 
 their perceptible being. 
 
 Dr Ward's theory^ that we perceive segregated portions 
 of a presentational continuum stands at the opposite pole 
 from sensory atomism. According to this view, the mind 
 (racial or individual, human or non-human) is confronted 
 initially with a sensory totality in which it does not dis- 
 criminate extension or duration, colour or sound. Speedily, 
 however, it comes to discriminate within this totality ; patterns 
 begin to show themselves; and, finally, this primitive sensory 
 vat is transformed into a recognisable and even into a solid 
 world. This development takes place through the segregation 
 of what was originally continuous, not through the aggre- 
 gation of ultimately discrete particles. Even the appearance 
 of discreteness at the end is only partial because the fringe of 
 any perceived thing melts into the continuum and mingles 
 with it. 
 
 This theory gives a very good account of perceptual develop- 
 ment, and also keeps close to the facts of developed perception. 
 Some writers, it is true, speak as if no one except a mathe- 
 1 PaycJiological Principles, especially chapter iv.
 
 26 THE THINGS WE PERCEIVE [ch. 
 
 matician had the right to speak of continuity at all. Con- 
 tinuity, according to Mr Russell, is just a kind of fog in which 
 those who are not mathematicians love to dwells It is diffi- 
 cult to see, however, how a better word could be chosen to 
 describe the facts which sensory atomists (as we have seen) 
 neglect, and those who use it as a descriptive term need not 
 quarrel with the mathematicians. Indeed they need not 
 enquire very closely whether their 'continuum' is ' continuous"* 
 or only 'compact' in the strict mathematical sense. In a 
 word, there is continuity in what we perceive, and there is 
 continuity between what we perceive at any moment and 
 its meaning in terms of a wider context. What is more, our 
 usual notions of space and time are in all probability too 
 discrete for the facts of perception. Recent enquiries, and 
 recent theories of natural knowledge point to the conclusion 
 that space and time, as currently conceived, are cumbrous and 
 inexact descriptions of the perceived phenomenal We perceive 
 spatio-temporal events whose space-time is a single fact more 
 intimately joined and also more fluid than either common 
 sense or traditional physics is accustomed to admit. 
 
 However that may be, it is plain that Dr Ward's theory 
 would be misleading without a certain reservation. The world 
 as a whole is never perceived either at the beginning or at the 
 end of psychological development. What we perceive is always 
 a shifting fragment with tenuous edges. These fragments, it is 
 true, are apprehended by creatures who are always immersed 
 in the sea of being, and are therefore felt to be only fragments. 
 Moreover, this initial and continuing fragmentariness of our 
 percepts may be the basis, in the end, of our conviction of 
 the existence of a single world. This world, however, cannot 
 be perceived in even a vestige of its entirety. We perceive 
 a minute patch of it at the best, and the world as a whole, so 
 far as perception goes, is only a signified thing clinging with 
 a tag of meaning to the fragments we perceive. 
 
 Again we must beware of neglecting the facts of perception 
 in our zeal to pave the way for a theory of development. Even 
 
 1 Introduction to Mathematical Philosophy, p. 105. 
 
 ^ See e.g. Whitehead's Principles of Natural Knowledge, passim.
 
 II] THE THINGS WE PERCEIVE 27 
 
 if the beginnings of the presentational continuum were as 
 vague as the sensations of a half-stunned crayfish, there is 
 no such vagueness in the normal perception of men and women, 
 or of the higher animals. The eagle and the gazelle, it is true, 
 cannot tell us what they perceive, or do not care to do so. 
 But men and women do tell us, and they say that they per- 
 ceive such things as gargoyles and cathedrals, twinkling stars 
 and a gibbous moon. These statements are judgments based 
 on perception, and it is worth while enquiring, once again, 
 whether judgments of that kind are faithful accounts of the 
 facts directly perceived. 
 
 As we have seen, these statements seem to indicate that 
 physical things interpreted after the fashion of common sense 
 are directly perceived. It is not difficult to decipher the main 
 outlines of this common sense interpretation. The physical 
 things of common sense differ from scientific objects on the 
 one hand, and from a collection of sense data on the other. 
 Unlike electrons and ions, physical things are supposed to be 
 perceptible and to be characterised by all the sensible qualities. 
 One and the same bar of common soap, on this interpretation, 
 can be smelt and touched and seen, whereas ions and electrons 
 are imperceptible, and are not credited with any properties 
 except the spatial and temporal qualities required for the 
 equations of physics. For a different reason the physical things 
 of common sense are not merely sense data or collections of 
 them. They are continuants in time, and sense data are fleeting 
 intermittent things. They preserve the same recognisable 
 spatial contour for a considerable time, whereas the sensory 
 atomists give a mere parody of this statement when they say, 
 instead, that a given set of intermittent sense data are broadly 
 similar in respect of outline. Physical things, again, are 
 supposed to have causal relations with other physical things. 
 Soap dissolves in water, and it will stop a leak in a gas pipe. 
 This language could scarcely be used of mere sense data. 
 
 Our previous argument has shown that the account of the 
 sensory atomists is true in most particulars but inconiplete 
 because it does not mention meaning, and because meaning is 
 directly perceptible just like colour and sound. It follows that
 
 28 THE THINGS WE PERCEIVE [ch. 
 
 common sense descriptions are much too hasty. According 
 to common sense, a penny is a circular object; but what we 
 directly perceive in the penny when we look at it from different 
 positions is a series of ellipses of varying eccentricity, and it 
 is impossible to deny this and also to accept the facts of 
 direct perception. Again, it is true that our sense data 
 fluctuate when we perceive, as we say, a steady candlestick, 
 and it can be proved to demonstration that sight-space and 
 touch-space are not literally identical. It is useless to suppose, 
 therefore, that these differences and these fluctuations can be 
 either ignored or repudiated in any faithful account of the 
 facts of direct perception. Moreover, the case against this 
 hasty common sense interpretation seems to become inde- 
 finitely stronger when we consider the variations which are 
 possible in the perceptions of one and the same person, or, again, 
 the variations between different human beings and between 
 mankind and the animals. The field that is green to Hodge in 
 his normal condition is yellow to him when he is jaundiced, and 
 has many colours when he is intoxicated. No one can reasonably 
 suppose that a man and a dog are aware of precisely the same 
 odours when they sniff, as we say, the same piece of meat. Bacon 
 and eggs taste very differently to those who have a cold in the 
 head and to those who have not. A colour-blind man and a 
 normal woman do not perceive precisely the same scarlet patch 
 when they look at geraniums in flower. These examples could 
 be multiplied indefinitely, and no one who has read Berkeley's 
 Dialogues between Hylas and Philonous has any shred of 
 excuse for neglecting them. ' You may draw as many absurd 
 consequences as you please,' poor Hylas complained, ' and 
 endeavour to perplex the plainest things, but you shall never 
 persuade me out of my senses. I clearly understand my own 
 meaning.' ' I wish you would make me understand it too,' 
 said Philonous^. 
 
 The truth is that these variable, fluctuating, intermittent 
 
 sense data are not the whole of what we perceive, for they are 
 
 always suffused with meaning when we perceive them ; and 
 
 that, in principle, solves the problem. Our glimpses of the 
 
 1 Dialogues, Fraser's edition, vol. i. p. iJ93.
 
 ii] THE THINGS WE PERCEIVE 29 
 
 candlestick are not steady like the candlestick, but they have 
 a meaning which we interpret in terms of steadiness. Ac- 
 cording to one way of it, an emmetropic and an astigmatic pair 
 of eyes cannot perceive the same thing. According to another 
 way of it, they can, because there is a common meaning and 
 an identical reference in the things they perceive. To say 
 that we cannot perceive continuants intermittently is true on 
 the surface, but on the surface only ; for continuants are 
 conveyed to us through the intrinsic meanings of what we 
 perceive intermittently. What is more, anything that we 
 perceive is part of a world, and this meaning is found 
 in it unalterably. Sense data are sign -facts and they signify 
 continuants in a world. 
 
 What must be avoided is the confusion between the meanings 
 which we perceive and our reflective judgments concerning 
 them. To borrow a phrase from William James, our perceived 
 meanings are raw and unverbalised, and even the simplest of 
 the plain man's judgments of perception implies a system 
 of meanings which need not be identical with the unverbalised 
 meanings of perception. Common sense is unreflecting on the 
 whole, and therefore its reflective theories are not very likely 
 to be good ones. Clearly, therefore, there is nothing to prevent 
 science or philosophy from being better interpreters of the 
 plain man's meaning in perception than the plain man himself. 
 More especially, this recognition of perceptual meanings does 
 not imply any particular theory of space or time or the world. 
 To speak of the spatial meaning which clings to a patch of 
 colour, for example, does not imply any one interpretation of 
 space. The phrase simply describes the kind of meaning, what- 
 ever it may be, which a man sees when he prods a hole with his 
 dibble to plant a cabbage, or looks before he leaps. To say 
 that we perceive continuants is only the beginning of a theory 
 concerning what these continuants are, and to say that these 
 continuants belong to a world is not a cosmology, or even an 
 articulate theory of the world itself. The point is that the 
 meanings perceived in any instance of perception are of the 
 type indicated by these expressions. In other words, the work 
 of reflection in these matters is not primarily an addition to
 
 30 THE THINGS WE PERCEIVE [ch. 
 
 the facts of perception or a wholly novel species of interpreta- 
 tion. It is an attempt to clarify and to continue those un- 
 verbalised perceived meanings which are always given and dare 
 not be overlooked. 
 
 Assuming, then, that the things we directly perceive are 
 the sign-facts described above, our next and most important 
 question is the bearing of this conclusion on the theory of 
 realism. Realists maintain that perception is the discovery 
 of a world independent of the perceiver. Now, whatever is 
 discovered is given or found. We may ask, therefore, whether 
 the sign-facts which we perceive are either given or found. 
 
 According to many writers it is absurd to say that perceived 
 things are given or found, since all perceived things are products 
 of the nervous system. This argument, however, plainly 
 contains a fallacy. The nervous system itself is also perceived, 
 and we have to trust our own eyes before we have the right to 
 trust anyone''s description of our ocular apparatus. That would 
 be true even if we were able to observe our own retinae, or 
 optic nerves, or brain ; and we cannot, in fact, do this. Our 
 evidence for the existence or for the character of the nervous 
 system depends upon our perception of other people's nerves 
 and their arborescent entanglements. We must trust percep- 
 tion, therefore, to make a beginning with this theory, and if 
 we have the right to trust our observations of the nerves upon 
 the dissecting table, no possible theory can consistently deny 
 that we have an equal right to trust our observations of the 
 dissecting table itself. We cannot see the nerves better than 
 we see the knife, and therefore if we know that there are 
 nerves we also know that there is a knife. It is also true, to 
 be sure, that when we assume that we can perceive things, we 
 find that we can perceive the nerves on the dissecting table as 
 well, and we are then in a position to infer, by exceedingly 
 probable reasoning, that the nervous system must be stimulated 
 whenever perception occurs. The most that can be extracted 
 from this circumstance, however, is that the nervous system is 
 an indispensable instrument for perceiving; and this harmless 
 truism is consistent with any theory. 
 
 According to a much more important set of arguments,
 
 ii] THE THINGS WE PERCEIVE 31 
 
 perception can never be the literal discovery of an independent 
 world because the process of perceiving always permeates and 
 transfigures the perceived thing. Perceived things cannot be 
 given to the mind because they always have a meaning, and 
 this meaning must be contributed by the mind. 
 
 Some writers maintain that what is given to the mind is a 
 bare stimulus, a naked shock or tremor of the nerves, without 
 any clothing of space or time or sensory quality. This theory, 
 however, is liable, in principle, to the same objections as the 
 former. We have to trust our developed perceptions before 
 we can observe or infer the existence of this naked shock or 
 tremor, and this theory, consequently, may be left by the 
 wayside along with the other one to comfort it with the fond 
 recollections of a vicious circle. 
 
 It might still be argued, however, that even the rudiments 
 of perception suffice to vanquish this drculus iii probanda, and 
 that if the human mind were given something as rudimentary 
 as the crustacean perceptions of a lobster it might be capable 
 of weaving our present system of acquired meaning into these 
 perceptions. And then the fight begins. When anything 
 is given to the mind, we are told, the mind must be wholly 
 receptive. We have lost this innocence long ago, and the mind 
 clothes everything it receives with sophisticated and elaborate 
 raiment. Whatever is perceived is suffused with acquired 
 meaning, and what is given cannot be also acquired. 
 
 There are two answers to this argument, and these, together, 
 seem conclusive. In the first place there is no reason why a 
 thing which is given, found, or discovered must therefore be 
 supposed to be passively received. Many discoveries require 
 painful effort, and the discoveries of perception seem to follow 
 the same law although we may have forgotten our infantile 
 labours in this direction. Attending, discriminating, inhibiting 
 the irrelevant and organising our mental resources are pro- 
 cesses which must be performed with the sweat of our minds, 
 and the theory of realism does not contain the faintest impli- 
 cation to the contrary. In the second place, if perception is 
 admitted to be discovery at any stage, there seems to be no 
 good reason why it should be less of a discovery at an advanced
 
 32 THE THINGS WE PERCEIVE [ch. 
 
 stage of development than at a more primitive level. The 
 development may be progressive discovery, and the acquired 
 meaning ascertained, not imposed. 
 
 Indeed, there seems to be no good reason for maintaining 
 that acquired meaning is contributed by the mind, unless it 
 is argued that all meaning is so contributed. That, however, 
 is precisely the contention of many philosophers, and these 
 philosophers conclude in consequence that since a meaningless 
 thing is nothing to the mind there is no possibility of discerning 
 the given as it is in itself. This argument is enormously 
 important if it is true, and so it must be considered very 
 carefully. 
 
 The principal reasons in its favour are three in number, 
 the dogmatic, the intuitive and the critical ; and I shall deal 
 with them in turn. 
 
 The dogmatic argument consists of the assertion that all 
 meaning is synthesis, and that synthesis cannot be given. All 
 synthesis, it is argued, is spontaneous. It can only be thought, 
 not perceived, and it is the work of the individual mind, or of 
 Cosmic Reason, or of Absolute Spirit. 
 
 This dogmatic assumption permeates Kant's position and 
 that of his successors. It has a much older history, however, 
 and Kant himself seems to have inherited his theory of a 
 spontaneous, non-sensuous, active Reason from the scholastics 
 who themselves traced it back to Aristotle. In modern times 
 Green and Lotze do fealty to the dogma in the special form 
 which maintains that meaning is synthesis, that synthesis is 
 relating, and that all relating is the work which the mind does 
 in making a finished product out of raw sensory material 
 which, in its turn, is either given to the mind in feeling, or 
 generated from the mind itself. It is unfortunate from one 
 point of view, and instructive from another, that Kant's critical 
 philosophy was not critical enough to examine this fetish. 
 Meaning, it is clear, implies togetherness, but things may be 
 found together as well as put together. A cow's tail is together 
 with the cow, but if anyone was responsible for this conjunction 
 it was God. It is certainly true that there cannot be meaning 
 without connection. Utter isolation kills meaning. But the
 
 ii] THE THINGS WE PERCEIVE 33 
 
 assumption that connection is only the mind's carpentry, or 
 that it cannot be discovered in things, is a mere dogma which 
 should never have lived. We should pray for peace to its ashes, 
 for its ghost, even now, is horribly malignant. 
 
 According to the intuitive argument, it is self-evident that 
 meaning is wholly an affair of the mind. Things exist but 
 they do not mean ; and that is an end of the matter. 
 
 The plausibility of this argument depends on its ambiguity. 
 ' Meaning ' is sometimes synonymous with intention. My 
 meaning is what I intend to convey to myself or to some other 
 person. Now my intentions, of course, are my own, but the 
 facts I intend to indicate need not be mental at all. And when 
 we say that such and such a perceived thing means so and so 
 our intention is to express a certain state of the perceived facts, 
 i.e. the way in which some fact or aspect of fact indicates 
 something else in the facts. To be sure, it takes a mind to 
 apprehend this indication just as it takes a mind to apprehend 
 at all; but there is no reason in the nature of things why this 
 indication should be less radically objective than anything else 
 that can be apprehended. 
 
 The critical argument puts these general issues to one side 
 and goes straight to the facts of knowing. These facts, it is 
 claimed, show to demonstration that all meaning is mental. 
 Our meanings, indeed, are wholly personal. Things, it is true, 
 are connected with one another, but it is a mere accident if 
 any of these connections coincide with what we call meaning. 
 It might be otherwise if signs had anything in common with 
 the things they signify or if these twain necessarily obeyed 
 the same laws or had any recognisable affinity ; but, as it is, 
 none of these conditions holds. 
 
 The written letters ' fireship ' signify a certain noise ; and 
 this noise signifies a thing which is highly combustible and 
 very dangerous. The letters, however, are utterly different 
 from the sound, and the sound itself is utterly different from 
 the thing it signifies. It would be inconceivable otherwise how 
 such different sounds as 'fireship' in English and 'brulof in 
 French should signify the same thing. Thus the nature of a 
 sign, its laws and connection, may differ completely from those 
 L. 3
 
 34 THE THINGS WE PERCEIVE [ch. 
 
 of the thing signified. The laws of v/ords are the laws of sound 
 and of the human pharynx. The laws of things are what you 
 will. And the connection between these signs and things is 
 wholly arbitrary. Notliing but a mind can give them meaning. 
 The theory of symbolism is a complicated business, and 
 we might simplify our thesis by maintaining that perceptual 
 significance is an affair of 'natural' signs and not of arbitrary 
 ones (as Berkeley supposed when he described perception as a 
 divine visual language). It is better, however, to take the case 
 at its strongest, for if even the connection between language 
 and things is not peculiarly mental in its essence, there is not 
 much likelihood that any other sort of meaning must be mental. 
 A sign is something which is capable of doing duty for the 
 thing it signifies, and it is interesting to consider the psycho- 
 logical conditions of the mind's use of signs. Nothing can 
 become a sign unless it has been experienced along with some 
 other thing. When it has been experienced in this way, it may 
 stand for, or be a substitute for, that other thing or for the 
 whole composed of itself and the other thing. The most useful 
 kind of signs are those which permit of the same logical de- 
 ductions as the wholes for which they stand. Another kind 
 of sign, also very useful at times, is that which, like the 
 proverbial knot in a handkerchief, leads the mind to think, 
 by association, of the event which was thought of when the 
 knot was tied. 
 
 The usefulness of a sign lies in the ease and celerity with 
 which it can be used as a substitute. The more time gained, 
 the better the sign. On the other hand, these substitutes do 
 the same kind of work as the things they signify : they are 
 quicker at it ; that is all. If we use a sign because it has the 
 same logical properties as the things it signifies, the important 
 point to consider is what these logical properties are. If they 
 are mental, the sign must needs be mental ; and if not, not. 
 If, again, this connection is illogical, like mere association, we 
 have to ask whether the things we associate are necessarily 
 mental. If they are, this kind of meaning must be mental ; 
 and if not, not. 
 
 Now it is clear that logical relations are not peculiarly
 
 ii] THE THINGS WE PERCEIVE 35 
 
 mental. The idea that they are depends upon the antiquated 
 prejudice that everything is mental which cannot be seen 
 or touched. Again it is plain that we associate things. We 
 associate pony -phaetons with old ladies because we have often 
 seen old ladies driving in them. If pony-phaetons are not 
 mental they do not become mental by standing for old ladies. 
 And the old ladies need not become mental either. 
 
 The only serious difficulty in the case is that mere con- 
 junction casually experienced may serve as well as a settled, 
 intrinsic conjunction in things. Sunday may stand for gloom 
 and overfeeding to me, because gloom and overfeeding used to 
 be my lot on that day of the week. That was a settled con- 
 junction in my life. On the other hand the nayne Sunday has 
 no intrinsic connection with the first day of the week. The 
 prehistoric inventor of the name must simply have thought of 
 name and day together, and other people must have accepted 
 the conjunction because they thought that name as good as 
 any other. Even in that case, however, the name and the day 
 are things which are experienced together, and neither of them 
 is part of the mental process of experiencing. The mere 
 conjunction of the two in time makes it possible for one of 
 them to be the sign of the other — to mean that other — but in 
 the original conjunction both were things which were appre- 
 hended; and the sign, when it is used, is an apprehended thing 
 too. The mind takes a cue from it and comes to think of 
 something else, but the fact of meaning does not in the least 
 annul the distinction between the process of thinking and the 
 things of which we think. 
 
 We may conclude, then, that even in the most arbitrary 
 cases, meaning or significance is something thought of, not 
 part of the mental process of thinking. Meaning may need a 
 mind but it is not mental. And the opposite view is untenable. 
 Even if meaning were a mental product, the mind would have 
 to apprehend this product, and the product given to the mind 
 in apprehension would be an apprehended thing just as surely 
 as if the mind had nothing to do with the production of it. 
 Knowing is never making. It is just knowing. 
 
 For clearness'' sake we may now review the foiegoing 
 
 3—2
 
 36 THE THINGS WE PERCEIVE [ch. 
 
 argument. Perception is the awareness of a sensory complex 
 which confronts the mind directly. This complex is a sign-fact 
 which has meaning or significance as well as extension, duration, 
 colour, or tone, and this meaning or significance belongs to it 
 quite as indefeasibly as hardness, extension and the rest. In 
 most cases (if not in all) the sign-fact perceived has, broadly 
 speaking, the kind of meaning which is elaborated into the 
 common sense notion of a physical thing. At any rate a con- 
 tinuant is signified, and the continuants perceived indicate 
 other continuants and furnish an analogue, in perception, to 
 what common sense calls the connection of physical things in 
 a physical world. That is what we perceive, and the arguments 
 which attempt to show that these perceived things cannot 
 really be discovered because they are not given or because 
 their meaning is somehow a mental thing are one and all 
 mistaken. 
 
 This explanation is certain to give rise to the impatient 
 question whether or not ' matter ** is directly perceived. Even 
 the most impatient critic, however, must surely admit that this 
 question is not so simple as it looks. What are we to make, 
 for instance, of the palpable differences between what the 
 plain man calls matter, and the elaborately different world 
 described in scientific theories of the constitution of matter ? 
 The rocks and trees of common sense are relatively constant 
 as the ages advance, but the scientific theories of them are 
 constantly changing, and are very frequently in solution. This 
 relative finality of perception, however, may perhaps be unim- 
 portant ; for science, very likely, becomes progressively more 
 penetrating while vision and hearing do not ; and there may, 
 after all, be sufficient affinity between the common sense point 
 of view and the point of view of science to justify the tra- 
 ditional attitude in which common sense boggles at philosophy 
 and meekly accepts whatever science has to say. On the other 
 hand, we have seen already that there is at least one difference 
 between common sense and the prevailing scientific theories 
 which cannot be passed over lightly. Scientific tradition 
 purges away the secondary qualities of physical things in a 
 ruthless electuary, and crams their interstices with imper-
 
 ii] THE THINGS WE PERCEIVE 37 
 
 ceptible substances. As a general rule we are willing enough 
 to accept these additions, for we take little account of 'Occam's 
 razor,' and we know, like a friend of our youth, that ' our 
 wision's limited"'; but we may reasonably complain of being 
 forced to accept less than nature offers us, and that is why the 
 plain man himself often makes a wry face at the Democritus' 
 purge which banishes colour, sound and odour (along with the 
 rest of the secondary qualities) from his world. 
 
 Indeed, it is manifest that the .secondary qualities are 
 directly perceived in the same sense as the primary, and it is 
 abundantly clear that any arguments which discredit the 
 secondary qualities tell equally against the primary. Science 
 therefore must accept one or other of two alternatives. Either 
 the ' matter "* of scientific theory is wholly imperceptible, or it 
 has secondary qualities as well as primary. In the former case 
 there is complete disagreement between science and common 
 sense, and science itself is in a quandary since the only evidence 
 for the existence of 'matter' is the evidence of perception. On 
 the latter alternative, scientists are at liberty to argue that 
 secondary qualities are secondary only in the sense that they 
 are not needed in scientific descriptions and formulae. The 
 explanation of colour may be given in terms of colourless 
 vibrations ; and so on. In that case, science and common 
 sense, with a little good will, may contrive to adjust their 
 differences. 
 
 The philosopher's problem concerns this world of per- 
 ceptible matter, spatial, temporal, coloured, resonant and 
 tactile. Is there such a world .^ Can it, literally and directly, 
 be perceived as it is in itself.'' Could it exist before the race 
 of sentient beings began, and after they have ceased to be 't 
 Is velvet really black, and smooth, and glossy 't Could a rose 
 be fragrant still in an unpeopled waste .'' And common sense 
 expects a plain answer to its plain question. 
 
 A plain question is a euphemism for the fallacy of many 
 questions, and a simple 'yes' or 'no' in these matters is worth 
 nothing at all. Any theory worth a peppercorn must take 
 account of the difficulties, and it is but trifling with the matter 
 to speak of the direct perception of an independent world
 
 38 THE THINGS WE PERCEIVE [cii. 
 
 without considering the differences between one man's percep- 
 tions at different times or between different percipients. Do 
 I and my neighbour and my dog perceive the same rainbow ? 
 The final answer to this seemingly simple question is elusive 
 enough to justify the northern legend that truth lies buried 
 at the end of a rainbow. 
 
 There seem to be two principal ways of meeting these 
 difficulties. According to the first of them, the difficulties 
 arise from a false conception of thinghood. We say that a 
 penny is round but that it often appears to be elliptical, that 
 the trail which the bloodhound pursues only appears to be 
 odourless to his human followers, that one of Dalton's celebrated 
 stockings was red and the other green although both seemed 
 to have the same colour to him because he was colour blind. 
 Indeed, we might even say that because these things are so and 
 so, they must appear otherwise under certain conditions. We 
 assume, that is to say, that the penny has just one shape and 
 each of the stockings just one colour, but we should reverse 
 these assumptions according to the theory we are now con- 
 sidering. If we do not, it is argued, our perplexities become 
 hopeless. Many shapes of the penny, several colours of the 
 stockings, are perceived in fact. Why then should we conclude 
 that all save one are illusory appearances .f* And when we have 
 come to see that there is really no justification for assigning 
 a privileged position to one shape or to one colour it seems 
 difficult to stop short of the conclusion that all percepts are 
 but appearances, and that physical things are never perceived 
 as they are in themselves. 
 
 When the ordinary conception of thinghood is reversed, 
 another possibility appears. Instead of supposing that the 
 penny has just one shape, let us suppose that it is always the 
 whole collection of shapes which would appear from different 
 points of view ; and so in the other cases. On this hypothesis, 
 what we perceive at any moment is just one selection from the 
 countless host of appearances which together are the thing; 
 and therefore it is easy to explain the different appearances 
 which things present to different people or to the same person 
 at different times. For each of these selections is a different
 
 ii] THE THINGS WE PERCEIVE 39 
 
 selection, and the thing itself is rich enough to contain all 
 possible selections. This theory holds good, even in the cases 
 in which the different selections seem to contradict one another. 
 To most of us the elliptical shape of the penny seems to con- 
 tradict the roundness which we also attribute to it, for we are 
 accustomed to suppose that round things and elliptical ones 
 cannot occupy the same space ; and similarly the brown tint 
 which Dalton saw in both of his stockings seems to distort the 
 red of the one and the green of the other. This theory, however, 
 is ready with its answer; for, according to it, what we call one 
 space is really a collection of spaces. The roundness of the 
 penny,it maintains, is an inaccurate expression which is justified 
 by its convenience only. The phrase ought to mean (and really 
 does mean) that series of perceptible shapes which approximates 
 in a certain determinate fashion to a limit of circularity. There 
 is no difficulty in defining this series in such a way as to dis- 
 tinguish it clearly from other series which approximate towards 
 an oval or an oblong limit, and so the shape of every penny 
 differs from the shape of every egg or every envelope. The 
 supposed distortion, therefore, is due to a false conception of 
 what one shape, one colour, and the like, really mean. 
 
 In some ways common sense might welcome this theory, and 
 frankly prefer it to the usual ways of speaking. The plain man 
 believes, for example, that he can perceive his house just as it 
 really is. But although his house exists as a whole he never 
 perceives it as a whole. The most he can hope to do is to catch 
 a series of glimpses as he walks round it. Now if each of these 
 glimpses is really a part of the house, logic compels the plain 
 man to admit that the house at any moment contains all the 
 glimpses which he might successively observe during his peram- 
 bulations. The house, to be sure, is more than the sum of these 
 aspects, but it cannot be less than the sum of them ; and that 
 is the substance of the contention before us. 
 
 At the same time some further implications of the theory 
 might be welcomed less cordially. If our hypothetical friend 
 went up in an aeroplane and looked down at his house, this 
 odd aspect would also be part of what the house always is. If, 
 descending from the areoplane, he felt giddy and saw the house
 
 40 THE THINGS WE PERCEIVE [ch. 
 
 reeling before him, this reehng aspect would also be part of 
 what the house always is, and, by the same argument, the house 
 always is precisely what dogs and rats and mice perceive in it 
 whenever they happen to use their senses. These consequences 
 may not lead to contradiction, but they certainly seem very 
 strange. 
 
 Again, there is a difficulty about meaning. The theory we 
 are considering is usually expressed in the language of sensory 
 atomism, but we have already seen that any sense datum is 
 tinged with a meaning beyond itself. If this meaning were 
 drawn out explicitly it might, indeed, result in the apprehension 
 of other determinate sense data, but, in the general case, that 
 is not so. The cold look of the ice, for example, does not 
 usually lead to any reinstatement of the feelings of cold. It 
 is just a tinge in the look of the ice. These tinges of meaning 
 belong to anything perceived ; but it is strange, perhaps, to 
 conclude that they are literally part of the thing's independent 
 reality. To ignore them, on the other hand, robs perception 
 of three parts of its powers. 
 
 And here another difficulty emerges, far more serious. Un- 
 doubtedly we consider some of our perceptions better or more 
 adequate than others. Attentive scrutiny of a flower is better 
 than a casual glance; and when the steel worker, to borrow an 
 illustration of Dr Ward's, perceives many tints in the flame 
 which is only a uniform glow to others, we should commonly 
 hold that his perception is therefore more adequate. We believe 
 that a dog smells better than we do and that we see better 
 than a dog. Anyone, indeed, who admits that the senses are 
 educable is plainly of this opinion. 
 
 The theory before us might be able to admit the truth of 
 this belief provided that it were interpreted in some tortured 
 or Pickwickian sense. ' More adequate,' for example, might 
 be taken to mean 'more effective for practical purposes.' This 
 theory however must deny in toto that anything directly per- 
 ceived is, in itself, better or more adequate than any other; 
 and we all believe precisely the contrary in fact. According 
 to the theory before us, the margin of perception must, in its 
 mere marginal outlines, be immitigable objective fact in the
 
 n] THE THINGS WE PERCEIVE 41 
 
 same sense as the centre. For it is perceived; and perception, 
 ex hypothesis is only a selection from what always is. We all 
 believe, on the contrary, that we perceive a thing better when 
 we focus it. If this belief has to be relinquished there seems 
 to be no good reason for clinging to any other belief about 
 perception, and we might as well surrender perception to the 
 sceptics as defend it at such a cost. 
 
 Some perceptions, therefore, are intrinsically better than 
 others. Improvement in perception, however, is not merely 
 a process of coming to discriminate more qualities, it often 
 involves the transformation of the whole texture of the per- 
 ceived thing. Consequently it is impossible to maintain that 
 perception is always a faithful witness, or that anything per- 
 ceived is only a selection of something which, precisely as it 
 appears to perception, always belongs to an independent world. 
 
 This conclusion points towards the alternative method of 
 attempting to reconcile these apparent contradictions. This 
 alternative is to maintain that the material world is, broadly 
 speaking, what it is perceived to be, although there is much 
 error, and enormous risk of error, in our perception. According 
 to this view, the world is really independent of the percipient, 
 and it is really spatial, temporal, coloured, and resonant. We 
 see colours and we hear sounds, for the most part, just as they 
 really are; and we can generally overcome our errors by closer 
 attention, or allow for them by careful reflection. 
 
 This alternative seems provisional and unsatisfactory to some 
 philosophers, and impossible to others. We may consider the 
 latter first. 
 
 According to these critics this theory makes the radically 
 untenable assumption that perception may or may not be 
 veridical, or, in plain English, that it may or may not be true. 
 But nothing except judgment can be either true or false; and 
 perception, therefore, cannot be either. 
 
 This objection is only a dogma. There is a risk of error in 
 every species of apprehending, and not merely in judgment. 
 That which confronts the mind may or may not be as it seems. 
 An illusory percept, to be sure, claims to be as it seems, and 
 it is verily a determinate appearance ; but that is not to the
 
 42 THE THINGS WE PEKCEIVE [ch. 
 
 point since precisely the same thing occurs in a false judgment. 
 Anyone who judges that Caesar died in his bed has a thinkable 
 complex before his mind which is something, appears to be true, 
 and is false in fact. If it is a misuse of language to say that 
 perception may be false, let us say that it may be mistaken. 
 The name does not matter. 
 
 It might be otherwise if judgments based on perception 
 referred to the percepts on which they are based, but that is 
 certainly not the case; and it is essential to notice that it is 
 not. Psychologists, it is true, make judgments about their per- 
 cepts, but that is quite a special case. When I judge "That 
 is a goblet" my judgment is based on my percept, but it does 
 not refer to this percept. It refers to the goblet ; and pre- 
 cisely and numerically the same thing is signified both in the 
 perception and in the judgment. The percept is a sign-fact 
 having certain sensible qualities and a meaning which may be 
 expressed in terms of continuance or other properties not 
 directly perceptible at any one moment. The goblet is the 
 continuant itself. This continuant has the features discerned 
 by any act of perception, and it also has the qualities signified 
 in the perceived sign-fact. And the continuant itself is judged 
 as well as perceived. 
 
 This alternative theory, therefore (if anything so common- 
 place can be called a theory), is not possible, but it may well 
 seem unsatisfactory. When we once admit that error may 
 creep into perception, we find the spectre of ineradicable error 
 in all perception looming before us. For example, there seems 
 to be a personal equation in most perceiving. Everyone who 
 hears a clock ticking hears a rhythmical tick-tack, but there 
 is no physical difference of emphasis in the beats themselves. 
 Very likely there is an impersonal equation too ; for what men 
 see or hear is not what dogs hear or see, and yet the world 
 does not depend on the men or on the dogs. If, then, there 
 is no absolute guarantee for the truth of any perception, and 
 if any perceived thing can be perceived more and more ade- 
 quately in proportion as we attend to it more carefully, what 
 reason have we for serious confidence in perception ? The best 
 attention we can give must always be inferior to that which a
 
 i] THE THINGS WE PERCEIVE 43 
 
 being with better equipment than ours could bestow. And who 
 can measure the degree of our infirmity ? 
 
 The point at issue is not the barren topic of scepticism. 
 Sceptics deny logic ; and if logic goes, all is gone, including 
 scepticism itself; for the sceptic is a would-be Samson who 
 must himself perish in the ruins he has made. Distrust of the 
 senses however (or rather of the tidings they are taken to 
 convey) need not carry distrust of logic along with it, although 
 it is bound to perplex our beliefs in the whole realm of 
 existence ; for the senses are our only evidence for the exist- 
 ence of a physical world. So much must be admitted. On 
 the other hand, the 07ius prohandi certainly rests with the 
 doubter of sensory testimony. Credulity is more ancient than 
 doubt, and although credulity is the philosopher's bane, dis- 
 belief or suspension of judgment is a logical attitude when, 
 and only when, there are positive reasons for denying that 
 something is as it appears to be. We have seen that there are 
 such reasons ; but we have to grant in the first place that 
 something is perceived, and in the second place that the best 
 of all reasons why anything should appear so and so is that it 
 really is so and so. The reasons for doubting depend, not on 
 the imbecility of apprehension but upon the character of the 
 perceived facts themselves. It is surely rash to conclude that 
 there are no colours in a sunset, even granting that the angels 
 might see the sunset differently from us, and that sheep and 
 oxen almost certainly do. Some have thought, even, that the 
 angels can see no colours because of the perfection of their 
 vision, and that God perceives nothing by sense. It is sad that 
 men should have thought so. A Providence who cares for the 
 sparrow may delight in the lark's song, and if the spheres ring 
 with sweet music the angels may surely be privileged to hear. 
 There is nothing derogatory in the perception of physical 
 realities. 
 
 To come down to earth, we have a very good right to believe 
 that colour, sound and figure are the very stuff of a continuing 
 world, apprehended in part, and not inadequately, by men and 
 other sentient creatures. Those philosophers who treat this 
 belief as a sort of useful fairy tale told by the Absolute to
 
 44 THE THINGS WE PERCEIVE [en. ii 
 
 that immature part of itself which is incapable of deeper ap- 
 prehension have first to prove that this physical world is 
 incapable of existence. They cannot prove their contention 
 by showing that the physical world is really a mental thing, 
 for this mentalism remains unproved. They cannot prove it 
 by denying that anything can be apprehended truly unless 
 everything is apprehended truly, for that argument makes an 
 end of logic and of their own theories too. They cannot prove 
 it a prioi'i by contending that facts must be sublimated by 
 reason ; for reason is capable of dealing faithfully with what 
 is given. To say that sense experience is partial and provisional 
 is true enough, but need not lead to wholesale distrust in the 
 testimony of the senses. According to some writers, all that 
 can be said of careful discriminating perception is that it is 
 the best we can do and perhaps good enough for us. It is 
 another and a very different thing to say that perception is 
 good enough for any knower because it may reveal, and com- 
 monly does reveal, physical things as they really are. And 
 that, with some presumption and with tedious emphasis and 
 explanations, has been the thesis of this chapter.
 
 CHAPTER III 
 
 THINGS REMEMBERED AND THINGS EXPECTED 
 
 The mind is not where the body is, when it perceives what is distant 
 from the body, either in time or place, because nothing can act but 
 when and where it is. Now the mind acts when it perceives. The mind, 
 therefore, of every animal who has memory or imagination, acts, and 
 by consequence, exists, when and where the body is not ; for it perceives 
 objects distant from the body, both in time and place. 
 
 Lord Monboddo, Antient Metaphysics. 
 
 The things we perceive are taken to exist at the time when 
 we perceive them, or, to be still more accurate, at the time 
 when we feel the bodily accompaniments of perception. We 
 think we see the shot fired at the time when we see the sports- 
 man pressing the trigger, but, in point of fact, our vision occurs 
 a fraction of a second after this event, because light takes time 
 to travel, because the rate of transmission through the optic 
 nerve is only some thirty miles an hour, and because there may 
 be still further delay in the cortex and cerebral hemispheres. 
 We all know that the report of the shot reaches us rather late, 
 and that Orion, as we see it, is centuries old. Common sense, 
 however, finds no difficulty in accepting these facts. The plain 
 man is reassured when he is told that the interval of time is 
 insignificant for objects near at hand, and he is willing to correct 
 his hasty assumptions in the light of the information which 
 science gives him. 
 
 On reflection, however, many philosophers have become con- 
 vinced that what we call perception contains memory and 
 expectation as well. This fact, if fact it be, is usually said to 
 be an implication of the ' specious present^ '; and, since some 
 of the arguments on this topic are very confused, it is necessary 
 to dwell upon it for a little. According to these arguments, 
 
 1 James borrows the name from Mr E. R. Clay whom he quotes {Principles 
 of Psychology, vol. i. p. 609).
 
 46 THINGS REMEMBERED [ch. 
 
 the present moment is only a mathematical instant having 
 position in time but no duration. Any act of attention, on 
 the other hand, is an enduring process, and any event or slab 
 of existence which is apprehended also endures. Hence, we are 
 told, the apparent, or specious, present is always past and 
 future as well as present. 
 
 The whole question, to be sure, is very complicated, so com- 
 plicated, indeed, that there is an excuse, at least, for those who 
 maintain that no pertinacity in thinking can straighten it out. 
 In the first place, present events must be distinguished from 
 earlier and from later ones, and yet any one of these earlier 
 events was formerly in the present, and any one of the later 
 events will be present when the time comes. That is perplexing 
 enough, but our perplexities are doubled when we notice that 
 there is always a twofold series in the case. The process of 
 apprehending takes time, and the event apprehended endures. 
 If I look at a moving train, the train is passing and so is my 
 mind, and neither the passing of the train nor the transition 
 in my mind is a durationless, or mathematical, instant. More- 
 over, although we commonly attend to something other than 
 ourselves and not to our own attending, still we usually call 
 an event present when we take it to exist simultaneously with 
 our apprehension of it. It is small wonder, then, that there 
 are perplexities. 
 
 At the same time the usual accounts of the specious present 
 are needlessly self-contradictory. They aver, in effect, that what 
 appears to be present also appears to be past and future, and 
 go on to speak as if the specious present were both earlier and 
 later than itself. That is nonsense without excuse. The fact 
 of the matter is that whenever we perceive anything we always 
 perceive an enduring slab of time or a stretch of change. A 
 stretch of change, we know, has an irreversible order of earlier 
 and later in it, and we can infer that the same order holds for 
 a slab of time that endures without perceptibly changing. The 
 rest is an affair of definition. If the present existence of the 
 things we perceive is defined to be the existence of these things 
 which is contemporaneous with any present act of attention, 
 then the present, in this sense, endures because the act of
 
 iiij AND THINGS EXPECTED 47 
 
 attention takes time, and it should not be impossible to Hnd 
 out experimentally how long it endures. If, on the other hand, 
 the present is defined to be only a mathematical instant, then 
 the truth is that events are never perceived in or at a mathe- 
 matical instant. What is perceived is a stretch of duration, 
 and any stretch of duration contains an infinite number of 
 mathematical instants in the connnon order of time. 
 
 From this standpoint, the facts are complicated but not 
 contradictory. The supposed contradictions arise from first 
 assuming that perception tries to skewer the perceived event 
 on to a durationless instant, and then discovers that the event, 
 trapped and insulted, cries out that it endures all the same. 
 That is a complete mistake. We do not stop a changing event 
 by observing it. On the contrary, we observe it changing, and 
 neither let nor hinder the change. 
 
 When we observe a stretch of transience (as we constantly 
 do) the stretch, just because it is passing, has certain character- 
 istic features which belong to the essence of what we mean by 
 past, present and future. The stretch, being transient, contains 
 the order of earlier and later, but it contains more than that. 
 If it did not, past, present and future would lose their meaning, 
 for the order of earlier and later holds of past events them- 
 selves without reference to the present. Quatre-bras, for instance, 
 came before Waterloo. This additional element is felt by all 
 of us. Just because the perceived event is transient it is felt 
 to be slipping away from something and to be approaching 
 something; and these peculiar feelings of ' something slipping 
 away ** and of ' something in the offing "' are pastness, present- 
 ness and futurity Kar e^oxw- They are found in every piece 
 of apprehending, and they are the principal source of the im- 
 portance we attach to the order of earlier and later. The 
 instants of time have many orders, but we select this irreversible 
 order because of its distinctive connection with these universal 
 feelings. 
 
 When perception is defined to be the apprehension of any 
 physical thing which appears to be simultaneous with the act 
 of perceiving (and that is the usual definition), the thing per- 
 ceived must be supposed to exist precisely at the time occupied
 
 48 THINGS REMEMBERED [ce. 
 
 by the apprehension of it, no earliei- and no later. This defini- 
 tion is reasonable enough, and perhaps necessary, but it would 
 probably be misleading without some further explanations. 
 
 In the first place there is a complication due to the manner 
 of perceptual attention itself. While it is legitimate to speak 
 of an act of attention being succeeded by another act, it would 
 be false to suppose that these acts succeed one another dis- 
 continuously. Attention moves with a wormlike motion. It 
 has a forward impetus in every coil and it is never at rest. It 
 is not in two times at the same time any more than a worm 
 is in two places at the same time, but each successive pulse of 
 attention has itself a waxing part and a waning part, and this 
 adds immensely to the complexity of the problem. 
 
 In the second place there is the fact of meaning. As we have 
 seen already, what we perceive is a sign-fact. It is a fact itself 
 and part of the factual order of existence, but it is also a fact 
 which in its very texture is significant of a wider range of 
 existence. This wider range, to be sure, is probably of the 
 same stuff as the data of sense, but that is a problem for 
 reflective interpretation, not something given directly in per- 
 ception. Now these ' slipping away ' and ' something in the 
 offing ' indications in a perceived fact are, in the first place, 
 enisled in the very being of that fact, and, in the second place, 
 are signs of facts which are definitely earlier and definitely later 
 than what is perceived. The earlier events and the later ones 
 are not themselves perceived, but they are signified by the fact 
 perceived. That is part of what is meant by 'primary memory,' 
 or the indications of the past in the present. By the same 
 logic ' primary expectation ■" should also be included. 
 
 In the third place it must be remembered that every event 
 attentively discriminated has a marginal setting. Most accounts 
 of the specious present, it is true, lay gi-eat and perhaps undue 
 emphasis upon this circumstance. A thunderstorm, for example, 
 is a complex piece of atmospheric artillery in which the thunder 
 booms, lightning flashes, and great drops fall. At the moment 
 when I feel the drops, it may happen very easily that the crack 
 of the thunder has just passed its premonitory stage and that 
 the lightning has not quite ceased to stir me. In this case.
 
 Ill] AND THINGS EXPECTED 49 
 
 therefore, there is overlapping on the margin of attention, and 
 this marginal overlapping may partially conceal the difference 
 in the times at which these interlacing events begin and end. It 
 is a mistake, however, to say as many writers do, that the flash 
 as a whole and the whole reverberation of the thunder are per- 
 ceived simultaneously with the drops of rain just on account 
 of this partial overlapping. None of these events occurs twice 
 over. The perceived flash and the perceived boom are not 
 telescoped into the perceptible present, and also made to stretch 
 out beyond it ; and if primary memory and primary expecta- 
 tion really meant (as they are often said to mean) an overlapping 
 of this impossible kind, then primary memory and primary 
 expectation would be only illogical fictions. On the other hand, 
 it is most certainly true that any event perceived has a present 
 setting and fringe, part of which wanes while part waxes. What 
 is more, these mai-ginal waxings and wanings are often irre- 
 sistible incentives to memory proper and to explicitexpectation. 
 Frequently we cannot help asking ourselves what it was that 
 is just passing away, and what it will be that is just beginning; 
 and the answers to these questions are given by articulate 
 memory and by developed expectation. 
 
 With these explanations we may now proceed to our problems. 
 It is evident that the things we remember must have an earlier 
 date than our apprehension of them, and that what we expect 
 occurs later. Memory, in other words, looks back to the past, 
 and expectation looks forward to the future. The problem for 
 realists is the precise sense in which these statements are true. 
 Can memory be literal discovery of the past, or expectation 
 literal discovery of the future.'' Are we directly aware of former 
 things in memory, and of things to come in expectation ? 
 
 " It is by memory," Reid said, " that we have an immediate 
 knowledge of things past^*" This statement was more than 
 his officious editor could bear, and Hamilton therefore added 
 a footnote which runs as follows. " An immediate knowledge 
 of a pa,st thing is a contradiction. For we can only know a 
 thing immediately, if we know it in itself, or as existing; but 
 what is past cannot be known in itself; for it is non-existent-." 
 1 Works, Hamilton's edition, p. 339. ^ Ibid. 
 
 L. 4
 
 50 THINGS REMEMBERED [ch. 
 
 This statement is brimming with confusion. In the first 
 place, we can immediately apprehend things that do not exist 
 at all. If a man says that 'good differs from bad as heaven 
 from heir he may intend, indeed, to confine his assertion to 
 existing good things and existing bad things, but he may 
 equally well intend to assert the radical incongruence between 
 goodness and badness themselves, these very universals; and 
 universals do not exist, they only subsist. In the second place 
 it is simply false to maintain that any assertion of existence 
 is confined to the present tense. One might wish that European 
 languages, like some oriental ones, had a tenseless mood, but 
 even a European can see the point. It is true that the past 
 does not exist now, just as it is true that the present did 
 not exist formerly; but existence itself means the whole of 
 existence, not merely present existence; and past events, like 
 present ones, have their determinate place in the determinate 
 series of existence. 
 
 So we must refuse to be deluded by the fable that the past 
 cannot be directly discovered because it is dead, or that there 
 is any contradiction in holding that direct acquaintance 
 extends beyond the present. On the other hand, it is possible 
 that this restriction holds good in fact, and most philosophers 
 allege that it does. We have only a representative acquaint- 
 ance with the past and the future, they tell us. Memory is 
 always a present fact representing the past, just as expectation 
 is always a present fact representing the future. 
 
 This account of expectation seems to be just. Clairvoyants, 
 it is true, claim to apprehend the future directly, but we need 
 not take their pretensions very seriously until they give us 
 better evidence and less fraud along with it. It is impossible, 
 no doubt, to prove that clairvoyance 7nust be a miracle and an 
 absurdity, although the indeterminists say so, very stubbornly. 
 Even in the most chaotic world there could not be more than 
 one set of events at any one time, and this very set might be 
 revealed to the clairvoyants. With or without free will or 
 utter contingency, the past is just the past; and, by the same 
 logic, the future is just the future. In point of fact, however, 
 we expect or anticipate in the present, and the things which
 
 Ill] AND THINGS EXPECTED 51 
 
 we expect to happen in the future are not really observed 
 intuitively before they happen but are either 'primary ex- 
 pectations' in the sense already explained or else forecasts of 
 what we think the future will be like. Such forecasts of the 
 future are certainly not the future itself; and primary expec- 
 tation is plainly a present event. It is the promise and 
 prospect contained in the present. 
 
 The main problem before us is whether the analysis of 
 memory is fundamentally similar. Is memory an affair of the 
 present signs of pastness and of inference from these, or is it, 
 perhaps, direct apprehension of past events themselves.'^ 
 
 The usual account of it seems to be that memory is a species 
 of reproduction of the past in the present. It is the existence 
 of a memory-image which resembles, or at least represents, the 
 past. The past disappears but leaves traces ; and these, being 
 revived, are memories. 
 
 Undoubtedly, we are capable of reproducing what we have 
 done in the past, and we often call this reproduction memory. 
 When a schoolboy says that he remembers Horace's Nunc est 
 bibendum he means that he can repeat the lines at will from 
 *nunc pede libero' to the end. This repetition is manifestly 
 a reproduction in the present of his former conning of the 
 lines. The lines were 'committed to memory,' and we usually 
 think of memory in this sense when we speak of 'phenomenal 
 memories' like Dr I^ey den's ability to reproduce Acts of 
 Parliament vej^batim after a single reading, or the magic which 
 enables half a score of Moslems to repeat the whole Koran, or 
 Scott's marvellous knack of repeating pages of Spenser or 
 Ossian or Border ballads. Memory of this sort belongs to the 
 same tribe as habit and instinct. It is a kind of 'perseveration,' 
 re-doing what we or our ancestors have done before. The boy 
 who keeps a straight bat because he has been taught to do so, 
 and Fabre's little beetle trundling a ball of provender which 
 he has patted into a perfect sphere, are reproducing what they 
 or their fathers have learnt. And that is what happens in 
 memory of this kind. 
 
 On the other hand memory often means recollection, and 
 that is much more important. It is very doubtful, indeed, 
 
 4—2
 
 52 THINGS REMEMBERED [ch. 
 
 whether memory without recollection is memory at all, for 
 mere repetition or mere reproduction does not involve any 
 reference to the past. A revolving piston repeats itself but 
 does not remember, and we often feel that it is an abuse of 
 language to say that the beetle remembers how to pat his ball. 
 If the beetle merely repeats the performances of his ancestors, 
 he does not, strictly speaking, remember these performances 
 even if he is a reincarnated beetle, and we should commonly 
 hold that no repetition is memory unless it is accompanied by 
 recollection in some slight degree. There must be a feeling of 
 familiarity at least, and that feeling must have been acquired 
 in individual experience. Recollection, then, is more funda- 
 mental for the theory of memory than repetition, and we may 
 confine ourselves to it. 
 
 Recollection appears to be the direct apprehension of the 
 past, and an illustration will show its difference from memory in 
 the sense of reproduction. When Jones asks himself what he 
 recollects of his childhood he does not mean to ask what he 
 learnt in his childhood. He learned to walk and to speak, he 
 learned table manners and toilet conventions, but the odds 
 are that he has forgotten how and when he acquired these 
 accomplishments. On the other hand, he can recollect specific 
 events. He recollects the arrival of a baby brother, let us say, 
 or some juvenile delinquencies Avhich brought an impressive 
 retribution, together with a few trivial incidents which seem 
 to have clung to him for no reason that he can assign. These 
 events are not habits or repetitions. On the contrary, recollec- 
 tion seems to be the mind's power of returning, again and 
 again, to precisely the same event in the past. These peaks 
 in the past stand out in relief where everything else is a blank 
 or a crapulous haze, and Jones's repeated attention to the same 
 thing is quite different from any habit he may have learnt. 
 Reproduction is a fresh performance similar to a past one. 
 Recollection seems to be direct acquaintance with the past 
 itself. 
 
 It might seem, indeed, that this analysis of recollection 
 could be proved to demonstration. The alternative hypothesis 
 is that Jones's recollections are recurrent memory -images which
 
 Ill] AND THINGS EXPECTED 53 
 
 indicate the past. Now, as we have seen, the mere recurrence 
 of similar events is not memory at all, much less recollection. 
 It would be otherwise if these recurring events were hnoron 
 to be similar to past ones, but such knowledge surely implies 
 a comparison between present repetition and the past events 
 themselves. How can Jones Jcnoza that any memory-image 
 represents the past unless he can compare the memory -image 
 with the past itself.? And does not the possibility of this 
 comparison prove that Jones is directly acquainted with the 
 past as well as with the present ? 
 
 This argument, however, falls short of demonstration, as 
 a reference to the kindred facts of expectation shows. Our 
 anticipations represent the future; and yet we can never be 
 directly acquainted with the future. At first sight, indeed, 
 this expectation of the future seems itself to depend on 
 recollection. Our knowledge of the future depends on our 
 knowledge of the past, and our confidence in our expectations 
 depends upon our recollection of our former premonitions 
 together with our recollection that the event confirmed them. 
 These facts, however, do not prove the case. The future is 
 never observed, and all that we can observe in the past is a 
 certain consecution from earlier to later. Now the order of 
 earlier and later does not need recollection, for it is directly 
 observed in the specious present. Transience is directly per- 
 ceived, and transience proceeds from earlier to later. This 
 perception of earlier and later is a sufficient basis for reference 
 forwards or reference backwards, and so the argument falls. 
 For it asserted that this temporal reference before and after 
 was impossible without direct recollection of the past. 
 
 On the other hand, even if there is no demonstration, there 
 is at least a very strong probability that recollection is direct 
 acquaintance with the past. That seems, indeed, to be the 
 reason why memory and expectation are so palpably different 
 in our experience. Expectation is only the present sign of 
 a hidden future, and so it is plausible conjecture at the best. 
 Recollection, on the other hand, is often supposed to be certain, 
 precisely because we believe that the past itself is accessible to 
 our observation when we recollect. Even those who hold that
 
 54 THINGS REMEMBERED [cii. 
 
 memory is always a reconstruction of the past must admit that 
 it is a very different sort of reconstruction from expectation, 
 and they would be very hard pressed to explain this most 
 manifest difference. The simplest hypothesis is probably the 
 best, and we should accept the facts as they seem to be, unless 
 something in the character of apprehension shows that our 
 acquaintance with the past must be indirect. 
 
 As we have seen, there is no obstacle of this kind, but so 
 many philosophers find one gratuitously that it is necessary 
 to pursue the point a little further. If we go the way of the 
 mystics and suppose that we have to become a thing in order 
 to know it, then, of course, we cannot know the past; but, in 
 that case, our own perverse assumptions have worked our 
 undoing. Mysticism as a way of being should command our 
 respect, if not our aspirations, but when the mystic, becoming 
 arrogant, waves his wand before the multitude and whispers 
 darkly that knowledge is not what it is, it is time to be on our 
 guard against this wizardry. If, again, with a perverted matter- 
 of-factness, we declare that apprehension is an immediate 
 contact of present mind with present thing, and infer that no 
 mind can encounter the past, we have dug the same pit for 
 ourselves with a spade instead of a spell. Change the language 
 and the fallacy remains. Knowledge is not communion with 
 the thing nor contact with it. It is just knowledge; and we 
 may inspect the past as well as the present. 
 
 To those who are not hampered by these prepossessions the 
 problem is simply one of analysis, but the facts are intricate 
 enough to make the analysis very difficult. Psychologists seem 
 to be agreed that the first requisite of memory is the revival of 
 an image which represents some past event. A bad psychologist 
 stops there. A better one remembers that this resuscitated 
 copy is, after all, 'snug in its own skin' (as James puts it^), 
 and must be expressly referred to the past if it is memory at all. 
 The main psychological problem, therefore, is the character of 
 this reference of a memory-image to the past, and we are told 
 that the past must be our past, and that a clear recollection 
 is one that can be dated in our former personal experience. 
 1 Principles of Psychology, vol. i. p. 650.
 
 Ill AND THINGS EXPECTED 55 
 
 This dating in personal experience, however, does not affect 
 the logic of the question. A dated memory is something that 
 we remember in its context. It is an event remembered in 
 connection with other remembered events. Smith dates his 
 recollection of the Matterhorn by remembering the details of 
 his ascent, and of that memorable visit to Zermatt. He re- 
 members the ascent in connection with Robinson^s aspersions 
 on his climbing powers before he went up, and with Jones's 
 envious belittling of his exploit after he had returned to the 
 hotel. If an isolated recollection, therefore, is held to be the 
 existence of a memory-image, a dated recollection can be 
 nothing but the existence of a group of connected memory- 
 images. 
 
 This explanation in terms of memory-images is plainly 
 opposed to our earlier argument which maintained that recol- 
 lection is direct acquaintance with the past. According to the 
 usual theory, Smith's recollections of his ascent of the Matter- 
 horn are a series of representative images in his specious 
 present. These images are what is before his mind when he 
 relates his adventures at his own fireside, and in that case there 
 is no room for direct recollection of the ascent itself. Smith's 
 memory is not split into two. He does not see these images 
 and also the Matterhorn itself, as if he stood on the bridge 
 at Zermatt and compared the mountain with a picture-post- 
 card. There is only one thing before his mind as he tells his 
 modest story, and our problem is what that thing is. 
 
 He remembers, I think, the very thing that he perceived. 
 What he saw as he toiled up the slopes of the mountain was 
 a series of sign-facts, each signifying the same continuant, and 
 each belonging to that continuant itself. What he sees with 
 the eye of reminiscence is also a series of sign-facts, only there 
 is rather less of fact in them though quite as much of sign. 
 Smith remembers the same thing as he formerly observed, for 
 in both cases he is aware of the Matterhorn. The perceived 
 Matterhorn, however, differs in ma^ny ways from the remem- 
 bered Matterhorn, because a great deal of the perceived detail 
 of it escapes the mind. This loss is the penalty of obliviscence. 
 Memory is a poor affair compared with perception, and that
 
 56 THINGS REMEMBERED [ch. 
 
 is why Smith intends to go back to Zermatt as soon as his 
 funds permit. But although the remembered colour is fainter 
 and more schematic than the colour formerly perceived, it is 
 still the very colour which shone on the Matterhorn when 
 Smith climbed it. 
 
 This theory would be nonsense if it implied that Smith 
 remembered his former percepts, for these could not change 
 through obliviscence and also be literally the same. It would 
 be a metaphor, therefore, and a bad one, if we said that 
 Smith's (so-called) memory-images are just his former percepts 
 glimmering faintly through a misty veil of time. But Smith 
 did not perceive a percept. He perceived the mountain, and 
 there is no contradiction in maintaining that he remembers 
 the very mountain which he formerly perceived. His percept 
 was that in the mountain which he formerly perceived at a given 
 point in time. It was the Matterhorn itself cabined within the 
 limitations of Smith's vision. Similarly his so-called memory- 
 image is that in the mountain which Smith is able to remember. 
 Memory has its own restrictions. Smith's memory is limited 
 to the past Matterhorn just as his perception was limited to 
 the Matterhorn at the time he perceived it (except for the 
 signs which indicated continuance). Smith therefore remembers 
 the mountain in the state in which he formerly perceived it. 
 Again, remembered things are not apprehended so fully as 
 things perceived, but, on the other hand, the scope of memory 
 is wider than the scope of perception. The mind's eye can take 
 in more at once than the perceiving eye, and some persons 
 even claim that they can see all the sides of a cube at once in 
 memory. It is needless to pursue these details, however. Our 
 contention is that the things we perceive are also the things 
 we remember, and that memory-images are commonly mis- 
 named. Memory does not mean the existence of present 
 representatives of past things. It is the mind's awareness of 
 past things themselves. 
 
 Is there no such thing as reproduction in memory, then, and 
 is the whole theory of memory-images simply mistaken .'' That 
 would be going too far. Although Smith's memory is not 
 split into two, it is probable that some reconstructions in the
 
 Ill] AND THINGS EXPECTED 57 
 
 present accompany his recollection and blend with it. When 
 he remembers his ascent of the Matterhorn, he remembers not 
 merely the steep places and the insecure footholds, but also 
 his anxiety and his laboured breathing. For the most part, 
 these memories can be explained by the hypothesis of direct 
 acquaintance with the past without further ado. Still, there may 
 also be some reconstruction in the present. If it comes to that, 
 Smith's body does a little reconstructing for him. His heart 
 may begin to beat appreciably faster and his breath come 
 quicker as he remembers those painful panting moments before 
 his second wind came to him. Indeed, there is always a slight 
 bodily effect of this kind though it seldom makes any appreci- 
 able difference to anyone's alveolar percentage of CO,. 
 
 What Smith's body does in these cases, his mind may do 
 also. While he may recollect a past emotion without any 
 present excitement, the chances are that his recollection of an 
 exciting event will excite him again, and it is common enough 
 for excitement of this kind to be attributed mistakenly to the 
 past. It is easy to feel a hero when the danger is over and not 
 very difficult to be duped into the recollection of having felt 
 like a hero. Again, although memory has not quite the same 
 limitations as perception (and therefore should not be dis- 
 counted merely because we remember more at a given moment 
 than we ever perceived all at once), still there are limits to its 
 powers in this respect. Smith during his genial reminiscences, 
 throws himself back, as it were, into this ascent of the Matter- 
 horn, and very likely sees himself toiling up it, much as he 
 might see himself in a dream. It is impossible to suppose that 
 he ever saw the sweat pouring from his brow, but he can see 
 it well enough in memory, or, rather, in what passes for 
 memory. This experience is due to imagery eking out the 
 memory, and although it is generally easy enough to distin- 
 guish between memory and this embroidery of fancy, still it 
 is not always easy and sometimes it is not even possible. AVhen 
 George IV remembered how he had fought at Waterloo, his 
 memory was all embroidery. To be sure he had been em- 
 broidering so long that, in a sense, he really did remember; 
 but the date which he remembered was later than 1815.
 
 58 THINGS REMEMBERED [ch. 
 
 Illusions of memory are not usually so egregious as this one, 
 but they are notoriously frequent and they seem to be due 
 either to mistaking a present reproduction for recollection of 
 the past, or to associating recollections wi'ongly and so mis- 
 dating them. George IV's illusion concerning Waterloo 
 illustrates the first case. The child who thinks he remembers 
 an episode of his childhood when the truth is that others 
 have mentioned it so often in his presence that he comes to 
 appropriate the incident to himself, probably illustrates both 
 cases; for he mistakes his present reproduction for past fact 
 and associates the episode with his early history and not with 
 the later time at which it was told him. The case of misdated 
 recollections is an everyday affair of which examples leap to 
 the mind. Our proneness to this error is connected with the 
 fact that we voluntarily give rein to association when we try 
 to remember. These associations, when we dwell on them, tend 
 to take the form of images, and the cluster of images so arising 
 is mistaken for a determinate event in the past. 
 
 Facts of this kind show how untrustworthy memory is. He 
 would be a bold man indeed who trusted his recollection 
 implicitly except for a few striking events, and these, for the 
 most part, very near. On the other hand, there is no sufficient 
 reason for ultimate scepticism in these affairs. The separation 
 of genuine recollections fi-om fictitious ones need not be mere 
 guess-work, and with a little care, the imagery in the present 
 can usually be distinguished from the apprehension of the 
 past. It would be a pleasant world if memory were more 
 reliable than it is; but we cannot change the world, and we 
 must be content with keeping our theories of it as accurate as 
 may be. 
 
 Indeed, we should not be too sceptical, even concerning ex- 
 pectation. After all, we are passably prescient about many 
 things, even granting that the medical text-books are bound 
 to warn us that ' prognosis should be guarded,' Fuller con- 
 sideration of this topic, however, must be deferred until a later 
 chapter. 
 
 Before passing to the problems of imagery we may perhaps 
 consider an objection which is frequently urged. 'Your
 
 hi] and things EXPECTED 59 
 
 theories,'' the objector says, 'may be well enough, but how do 
 you explain memory ? You say nothing about the brain, or 
 channels in the nerves, or psycho-physical dispositions, and 
 until you do that you have explained nothing/ 
 
 This objection need not detain us long. It is plainly im- 
 possible to explain the fact of memory itself. Memory is 
 possible, and that is all we need to know. All we can do, and 
 all we should want to do, is to explain why we remember one 
 thing at one time and another thing at another time, and ex- 
 planations of that kind do not seem to be beyond the reach 
 of very moderate resources without any reference to physiology. 
 We remember this or that when something in the present acts 
 as a cue, and when the mind, in apprehending it, is led to 
 think of something in the past connected with the cue. There 
 is nothing strange in the circumstance that these cues should 
 lead us to think of the past; for the past, speaking generally, 
 is always with us. We do not lose hold of it as the hours 
 hurry forward, and we only attend to it more carefully when 
 we follow up our cues. The past, then, is neither trackless 
 nor unknown. The recency, vividness, or singularity of particu- 
 lar events in it make some recollections easier than others, and 
 we have usually a sufficient motive for setting ourselves to 
 remember. Those who believe that history repeats itself ought 
 to believe in the utility of history, and our lives would be 
 robbed of half their interest if we had no curiosity about 
 former things. 
 
 The mere fact that the brain is stimulated does not explain 
 perception, and the mere fact that the brain endures and 
 retains traces of former stimulation does not explain memory. 
 We have to trust perception and memory in order to obtain 
 this physiological information, and we have no right to disown 
 these witnesses at a later stage of the argument. The ultimate 
 fact in the case is the fact that apprehension itself occurs, and 
 it is quite unreasonable to be dissatisfied with the analysis and 
 description of apprehension as we find it. We can perceive 
 the present and recollect the past; and we are not required to 
 explain the inexplicable.
 
 CHAPTER IV 
 
 THE STUFF OF FANCY 
 
 Whence come ye, so wild and so fleet. 
 For sandals of lightning are on your feet, 
 And your wings are soft and swift as thought, 
 And your eyes are as love which is veiled not? 
 
 Shelley, Prometheus Unbound. 
 
 1 HE subject of this chapter is imaging, not imagination. This 
 distinction between images (or fancy) and imagination is at 
 least as old as Coleridge's Biographia Literaria, and Coleridge's 
 account of it is well worth quoting. Milton, he says, ' had a 
 highly imaginative^ Cowley a very fanciful mind,' and ' the 
 division is no less grounded in nature than that of Otway's 
 
 Lutes, laurels, seas of milk, and ships of amber, 
 from Shakespeare's 
 
 What ! Have his daughters brought him to this pass ? ' 
 The imagination, he tells us in another passage, 'dissolves, 
 diffuses, dissipates in order to recreate : or where this process 
 is rendered impossible, yet still at all events it struggles to 
 idealise and unify.... Fancy, on the contrary, has no other 
 counters to play with, but fixities and definites. The fancy 
 is indeed no other than a mode of memory emancipated from 
 the order of time and space".' 
 
 We should accept this distinction, although we need not 
 tie ourselves to the letter of Coleridge's transcendentalism. 
 Imagination (to quote Mr Conrad who ought to know) is ' a 
 creative effort in which mind and will and conscience are 
 engaged to the full, hour after hour, day after day^' It is the 
 faculty with which Mr Conrad, when he wrote Nostromo, 
 ' wrestled with the Lord for his creation, for the headlands 
 
 1 Biographia Literaria, chap. iy. - Op. cit. chap. xni. 
 
 ^ A Personal Record, chap. v.
 
 CH. iv] THE STUFF OF FANCY 61 
 
 of the coast, for the darkness of the Placid Gulf, the light on 
 the snows, the clouds on the sky, and for the breath of life 
 that had to be blown into the shapes of men and women, of 
 Latin and Saxon, of Jew and Gentile\' This high imagination, 
 to be sure, often uses imagery. The romance of green mansions 
 or a purple land is clothed in imaged raiment. The light that 
 never was is often the best part of poetry. None the less, 
 imaging in itself is usually a pedestrian affair. Even children, 
 though their fancy is sometimes quaint and nimble, are earth- 
 bound in most of their imaging, granting that a Daisy Ashford 
 arises here and there to wrestle with some bright spirit and a 
 stubby pencil for The Young Visiters. Most images, in a 
 word, are mere occurrences. Dream pictures, and the chaos of 
 an Ideenflucht are so many facts presented to the mind ; and 
 it is part of our business to consider what these facts are. 
 
 Images tend to arrogate a quite disproportionate importance 
 in many psychologies and theories of knowledge. Some writers 
 even speak as if the mere existence of symbolic images were 
 the same thing as judging, and others have never overcome 
 the ancient fallacy of confusing universals with attenuated 
 images. Half the talk about ' concepts ' is sunk in this con- 
 fusion, and even those who avoid this pitfall are very apt to 
 neglect thinking in their eagerness to explore the imaged acces- 
 sories of thinking. Even if there is no such thing as image- 
 less apprehension, images are usually only illustrations of what 
 we think about. A man who thinks of the foreign exchange 
 may always have an image of the ' Lusitania' and its bullion, 
 or of the Bourse at Brussels when he does so, but it is surely 
 manifest that he does not identify the foreign exchange with 
 the Bourse or the ship. 
 
 It is doubtful, indeed, whether the elaborate investigations 
 into types of imagery which are so prominent in many psy- 
 chologies have a tithe of the value which they are commonly 
 supposed to have. It is very interesting, no doubt, to know 
 that some people can visualise their breakfast tables in 
 minute detail, that others can visualise shapes and not colours, 
 that a few cannot visualise at all, that there is no connection 
 ^ A Personal Record, chap. v.
 
 62 THE STUFF OF FANCY [ch. 
 
 between the power of visualising and keenness of sight, that 
 those who cannot visualise can see in their dreams as well as 
 those who can, that some who have become blind can visualise, 
 that there is audition coloree, that women have clearer taste- 
 images than men, and that a few privileged persons can smell 
 the imaged perfume of the rathe primrose that forsaken lies, 
 when they are asked to perform this feat in a stuffy classroom. 
 These interesting facts, however, do not seem to have any 
 great importance beyond themselves. It looks promising, to 
 be sure, to classify pupils into audiles, visiles and motiles ac- 
 cording as auditory, visual or motor images predominate in 
 their thinking, for it might seem that the audiles would be 
 tauo-ht most easily through illustrations borrowed from sound; 
 and the others similarly. This idea, however, is wholly mis- 
 taken. The present writer, for example, cannot visualise at all, 
 but he can understand a description of St Sofia or of the Golden 
 Horn as well as other people. Anyone, indeed, can appreciate 
 illustrations borrowed from sight, sound and the rest, provided 
 that he has the normal equipment of senses. A poet blind 
 from birth, it is true, cannot describe the look of things ; but 
 that is quite another story. 
 
 Images exist, however, even if they are overrated, and the 
 elaborate investigations of Galton, Ribot and the others 
 should tell us at least what they are. But we know that well 
 enouo-h. Images are the mimics of percepts. A visual image 
 has shape and colour ; Bach could hear his Fugues in his 
 mind's ear; and so on. Compared with things perceived, it 
 is true, images are usually fleeting, ghostly and unstable, but 
 that, after all, is a relative matter ; for some percepts are 
 fainter than some images, and some imagery steadier than some 
 perception. Indeed, imaged things and perceived things are 
 so manifestly similar that it is difficult to explain why we 
 mistake the one for the other so seldom. The chief reason 
 seems to be bodily. When we perceive anything we are aware 
 of our bodies as well, and the bodily adjustments of imaging 
 are different from those of perceiving. We close our eyes to 
 see images and open them to perceive things; and this difference 
 in bodily consentience enables us to distinguish the two.
 
 iv] THE STUFF OF FANCY 63 
 
 This intrinsic similarity between imaging and perception 
 makes the status of images a very difficult problem. Common 
 sense, it is true, has its simple solution pat and ready, but 
 this solution, by an unhappy chance, is quite untenable. Ac- 
 cording to common sense, images are unreal and merely mental 
 things. They are the stuff that dreams are made on, that is 
 to say, nothing but the dreamer's private mind-stufF. Berkeley''s 
 critics, for example, supposing most absurdly that it was im- 
 possible to distinguish between dream and waking on the 
 good bishop's principles, proceeded to waste much ink in trying 
 to show that a world composed of images would be only a 
 mental fiction. So Beattie told Hume that the idea (or image) 
 of a roaring lion is not a roaring idea, and that the image of 
 an ass is not a long-eared sluggish idea^ ; and he put certain 
 * clownish questions ' to Berkeley in the same spirit. " Where," 
 he asked, " is the harm of my believing that if I were to fall 
 down yonder precipice and break my neck, I should be no 
 more a man of this world ^ My neck, sir, may be an idea to 
 you, but to me it is a reality and an important one too. 
 Where is the harm of my believing that if, in this severe 
 weather, I were to neglect to throw (what you call) the idea 
 of a coat over the ideas of my shoulders, the idea of cold would 
 produce the idea of such pain and disorder as might possibly 
 terminate in my real death .^^^ 
 
 These heavy witticisms, of course, did not touch Berkeley's 
 case. He could, and did, distinguish between an ' idea of 
 sense ' and a mere image, and his theory implied no confusion 
 between a square meal and a Barmecide feast, even granting 
 that the language of it compelled him to admit that we are 
 fed and clothed with ideas. What he did maintain was that 
 percepts (or ideas of sense) have the same status in relation 
 to the mind as images have. And he held that both were 
 mental. His critics replied, in effect, that images are mental, 
 and perceived things independent of mind. 
 
 This path, however, was not open to them. These mimics 
 of sense which we call images must have the same status as 
 percepts. If the latter are objects the former are too. If one 
 1 Essaxj on Truth (vol. i. 1776), p. 217. - Op. cit. p. 244.
 
 64 THE STUFF OF FANCY [ch. 
 
 is a mental event so is the other. The imaged St Sofia is 
 domed and minaretted and shapely just as the perceived 
 St Sofia is, so that if the perceived St Sofia cannot be mental 
 (on the ground, say, that the mind itself cannot be coloured 
 or extended) the imaged St Sofia cannot be mental either (for 
 precisely the same reason). Per contra^ if the perceived St Sofia 
 is not mental the imaged St Sofia cannot be mental either; 
 and even common sense must accept the logic of this situation 
 if it hopes for consistency. It is not at liberty to make a hand- 
 some present of images to the idealists, like the parson who 
 did not grudge his sermons to the burglars : 
 
 They came and prigged my silver^ my linen, and my store ; 
 But they couldn't prig my sermons ; they had all been prigged 
 before. 
 
 We have argued in the two preceding chapters that things 
 perceived and remembered are independent of the mind and 
 directly apprehended by it. Our grounds for this conclusion 
 were, briefly, that they seem to be so, that the best possible 
 reason for their seeming so is that they really are so, and that 
 the arguments which purport to prove that they are not so 
 are inconclusive. And now we have discovered that images 
 have the same status, in this respect, as perceived or remem- 
 bered things. They are apprehended things confronting the 
 mind, and not varieties of mental operations. They are given 
 to the mind, like anything else that it discovers. Coleridge 
 said that when he dreamed of Kubla Khan's palace ' all the 
 images rose up before him as things.' It may well have been 
 so, for that is what the images of our dreams commonly do. 
 They stand before us like the Brocken spectre or the Fata 
 Mo7'gana, like Tasso's familiar spirit or the angel that touched 
 Elijah in the wilderness. Images, in a word, always appear as 
 ' things ' unless perceived things conflict with them, and some- 
 times even then. When they conflict with perceived things 
 they are accounted shadows and so contrasted with the sub- 
 stance of perception. Yet shadows, after all, are things in their 
 own way, and Coleridge admitted as much when he went on 
 to say that his vision passed away after his interview with that 
 meddlesome ' person on business from Porlock,' ' like the
 
 iv] THE STUFF OF FANCY 65 
 
 images on the surface of a stream into which a stone has been 
 cast^' 
 
 If realists, however, are bound to hold that images have 
 the same status as perceived things in many ways, they are not 
 bound to identify the two. On the contrary realists are bound 
 to contrast them, despite the fact that they have colour and 
 shape and sound in common. 
 
 The contrast between dreams and waking life is usually 
 unfair to the dream. The dreamer is not usually in a position 
 to test his dream-world, and if he experiments with Reid's 
 remedy for curing nightmare and tries to throw his dream- 
 body over the dream-precipice, the dream promptly vanishes. 
 The memory of a dream, on the other hand, is such a very 
 feeble affair that it is seldom a serious competitor with waking 
 life, and yet the dream may have been very lusty indeed when 
 it had no competitors. As Hobbes remarked long ago, ' the 
 light of the Sun obscureth the light of the Starres^.' The 
 only fair comparison would seem to be between the memory 
 of a dream and the memory of something perceived, and, even 
 then, the luck is against the dream, for the memories of waking 
 life are so closely knit with one another and with the per- 
 ceptions of the moment that they acquire a kind of solidity 
 which the dream lacks. If we leave dreams out of account, 
 however, and try to compare present imagery with present 
 perception, the images again are unfairly placed. We are 
 practical men (we boast of it) and we cultivate our images as 
 little as may be. Why, then, should we be surprised if these 
 neglected things should appear to be only pale, starved, shiver- 
 ing, gibbering ghosts when they receive their tardy summons ? 
 
 It is clear, to be sure, that images do not fit into the con- 
 tinuous context of waking life. Anyone who dines out on the 
 strength of a dream invitation may be welcome but was not 
 expected. Our beds are not stained by the wounds of dream 
 scimitars. And so on. The explanation of all this, however, 
 is not so easy. If we have a right to maintain that Buckingham 
 Palace is in space and time because we perceive it to be 
 
 ^ ' Of the Fragment of Kubla Khan.' Note preceding the poem (1816). 
 ^ Leviathan, bk i. chap, ii., ' Of Imagination.' 
 
 L. 5
 
 66 THE STUFF OF FANCY [ch. 
 
 extended and enduring we seem to have the same evidence about 
 dream -palaces. We certainly apprehend these nocturnal in- 
 truders, and we find it very difficult to discover a place for them. 
 
 It has been suggested that dream-palaces inhabit our bodies 
 since they are only disturbances of the nervous system ; or 
 again, that they occupy their own spaces and times although 
 they are excluded from the privileged order of space and time 
 which science and common sense select ; or again, that they 
 belong to the physical world but that the apprehended 
 meaning of them is false. 
 
 The first suggestion maybe dismissed. The nervous system, it 
 is true, plays its part in dreams and imaging; and the stomach 
 does so too. Dryden, we are told, used to eat raw meat in the 
 hope of obtaining splendid images. Mrs Radcliffe used to pre- 
 pare for fantastic dreams by preparing fantastically indigestible 
 suppers, and many anchorites have tried to summon visions by 
 fasting and scourging and other calculated austerities. But the 
 images which we perceive do not inhabit either the stomach or 
 the nervous system, and they do not nestle within the dark coil 
 of our nerves. They require the nervous system in the same 
 way as perception does, and not in a totally different fashion. 
 
 The second suggestion looks more promising, but seems to 
 involve the absurd consequence that there really are dream 
 spaces in the same sense as there really are perceived spaces. 
 Hashish, on this assumption, would be a means of enabling 
 us to behold a different time order from the usual one, and 
 flagellants would scourge themselves into meeting real angels 
 in a real celestial world. On the other hand this suggestion 
 is not absurd unless it is misunderstood. The celestial worlds 
 of this hypothesis would be different celestial worlds for every 
 anchorite. One dream does not fit into another dream any 
 more than it fits into waking life, and dreamers, in the general 
 case, visit a whole series of dream worlds in a single night. 
 Per contra^ although there is no contradiction in this infinity 
 of discoverable worlds there are some grounds for suspecting 
 that worlds of that kind are as good as no worlds at all. 
 
 Still, as we have seen in an earlier chapter, we never perceive 
 the physical world as a whole, and, as we have also seen, the
 
 iv] THE STUFF OF FANCY 67 
 
 things which Smith or Jones perceive directly, with all their 
 variations and their perplexing differences at different times, 
 may not fit into a single physical world quite so tidily as we 
 naively suppose. Images, according to the most usual theory 
 of those who accept their objectivity, are quite as objective as 
 perceived things. They are only selections from a manifold of 
 objective being, and they are discovered precisely as they are 
 in themselves. The physical world, according to this view, is 
 the selection from this manifold of being which is peculiarly 
 useful for everyday life. Unless we are psycho-analysts it does 
 not pay us to take account of dreams, and the apparent un- 
 reality of dreams is neither more nor less than the uselessness 
 of our drowsy selection. It stands to reason that images are 
 somehow real, simply because they are plainly apprehensible 
 things. According to this theory, they are real in every sense 
 in which the objects of perception are real, except that they 
 do not take part in the work of the physical world. 
 
 If the physical world were really a collection of percepts 
 and unperceived sensibles (as the sensory atomists hold) this 
 theory would be tenable. It requires to be modified, however, 
 if, as we have maintained, we perceive physical things and 
 not percepts, and the kind of modification which is required 
 should be tolerably evident to the reader. Images plainly are 
 coloured, extended, and so forth, just as percepts are. Why 
 should we not suppose that they are physical facts forming 
 part of the world of physical things, but that the meaning 
 of imaged things is different from the meaning of perceived 
 ones ? If that were so, errors would be bound to arise (as they 
 do in cases of illusion or hallucination) if the meaning of 
 imagery were confused with the meaning of perception. And 
 that is our third suggestion. 
 
 Images are relatively detached in comparison with the 
 objects of perception, and this detachment appears in three 
 principal ways. In the first place images are not usually 
 suffused with a cluster of meanings derived from several senses. 
 A visual image is almost wholly visual. It shows very few traces 
 of tactile or other values. In the second place, imaging has a 
 different bodily margin from perceiving, and the bodily fringe 
 
 5—2
 
 68 THE STUFF OF FANCY [ch. 
 
 of all perception helps us very much in localising the things 
 we perceive, and in grasping their steadiness and solidity. 
 Visual images, for example, are not usually localised relatively 
 to our bodies, for although they float before us vaguely they 
 do not usually appear to be at any determinate distance from 
 us; and when they do (as in the cases in which we see double 
 by pressing our eyeballs) they are localised relatively to per- 
 ceived things. Again, and this is the third point, images are 
 seldom localised relatively to other images, because images in 
 the vast majority of instances reveal themselves as detached 
 beings each fenced in its own pen. 
 
 Illusions and hallucinations occur when imaged things and 
 their meaning are interpreted as if they had a perceived 
 meaning. When Blake asked the lady, " Madam, did you ever 
 see a fairy funeral ? " and went on to say that he himself had 
 often seen one on a dewdrop, he purposely made this very 
 confusion ; for these elfin obsequies did not take place on the 
 surface of the dewdrop. The man who slew the phantom with 
 his hatchet really saw a phantom form ; but he made a fearful 
 error when he mistook the phantom for a perceived things 
 The drunkard's loathsome rats and crawling serpents do not 
 deceive him up to a point. As I have heard it put, they are 
 very good serpents as far as they go ; but they do not strangle 
 him as perceived snakes might. 
 
 These errors would not be likely to occur either if images had 
 no meaning or if their meaning had nothing whatsoever in 
 common with perceived meanings. Although they are relatively 
 detached, imaged things are apprehended with an imaged 
 meaning which indicates a wider sphere of being. On the other 
 hand, the definite detailed meaning of imaged things is found 
 to be different from the meaning of perceived things when it is 
 carefully examined. We do not usually confuse the two systems 
 of meaning in waking life, and we can usually discriminate 
 passing well between them in daydreams and in illusions hypna- 
 fi^ogiques. There is no occasion to discriminate in dreaming 
 proper, because the sense of perceived things is quite shut out. 
 
 ^ Trial of Bernard Schidmaizig in Silesia. Quoted by MacNish, The Philo- 
 sophy of Sleep, pp. 75-77, from Hoffbauer's Treatise on Legal Medicine.
 
 iv] THE STUFF OF FANCY 69 
 
 If we were to draw a bow at a venture, we might hazard the 
 guess that images are really physical facts, partly identical 
 with perceived or remembered things, and differing from these 
 latter precisely in the respects in which their image-meaning 
 claims to be different. I wish to defend this thesis, though I 
 hope I am sensible of its difficulties. I am confident, at least, 
 that the elements of imagery should be explained in this way. 
 The imaged order of these elements is certainly harder to 
 account for, but even this difficulty may not prove insuperable. 
 
 It is a commonplace that imagery is borrowed from per- 
 ception. Those who are born blind see no white vistas when 
 they think of the frozen Caucasus. Those who are born deaf 
 cannot hear the nightingale in their imagings. It is precarious 
 reasoning, no doubt, to infer from this circumstance that 
 every note of elfhorns in our dreams has actually been heard 
 at some time or other, or that every tint that the fancy can 
 summon has once been seen with our eyes. Hume raised the 
 doubt very pertinently when he asked whether a mind that 
 had perceived all the shades of blue, save one, could not supply 
 the missing tint from its own resources ^ On the other hand, 
 Hume's experiment could never be put to the proof, for no 
 one knows how many shades of blue he has perceived during 
 his lifetime, and most of the possible shades of blue could be 
 discerned during any summer afternoon by the sea. What is 
 more, the very modesty of Hume's doubt is excellent evidence 
 of the solidity of the usual theory. We may agree, then, that 
 the elements of any image were once perceived; and in that 
 case there is nothing to hinder us from supposing that the 
 elements imaged at any time are literally the same elements 
 as those formerly perceived. These imaged elements, to be sure, 
 have lost some of their former fulness of detail, but this 
 partial loss does not annul the identity of what remains. 
 
 Proceeding with our hypothesis, then, we find, in the next 
 place, that it agrees very well with our earlier account of 
 memory. We found, in the previous chapter, that although 
 some reproduction in the present may blend with our recol- 
 lections, still the objects of recollection are not really memory- 
 1 Treatise, bk i. pt i. sect. i.
 
 70 THE STUFF OF FANCY [ch. 
 
 images existing in the present but the very scenes and events 
 in the past which were formerly perceived. It is doubtful, 
 indeed, whether anyone would dispute this account of the 
 matter unless he supposed that it implied insuperable diffi- 
 culties with regard to space and time. When Smith, to return 
 to our illustration, recounts his ascent of the Matterhorn, his 
 memory, we are told, is certainly a present phenomenon, and 
 the ice and the snow which he sees in recollection cannot 
 really be part of the Matterhorn, since Smith is in England 
 and the Matterhorn is in Switzerland. If the elements of all 
 imagery, however, are really identical with something formerly 
 perceived, and if the temporal and spatial meaning of imaged 
 things should never be identified with the spatial and temporal 
 meanings of present perception, it is possible to maintain that 
 the ' memory-images ' which Smith is said to recollect during 
 his narrative are ' images ' whose date is in the past and whose 
 place is in Switzerland, while images which are not memory- 
 images are fragments of the same order but emancipated from 
 the perceived or remembered world of space and time. And 
 that, in a word, is our theory. 
 
 These difficulties concerning the place and the time of 
 recollected things are dispelled by closer examination. Smith 
 recollects them now, but he need not suppose that they exist 
 now. Again, he is in his room when he recollects them, but he 
 need not suppose them to be in his room, and he would not 
 suppose them to be there (unless he were hopelessly confused), 
 even if they were mere images, fully emancipated. Smith, to be 
 sure, might refuse to accept the suggestion that these recol- 
 lected scenes are really in Switzerland and in the past, because 
 he would suppose that the apprehension of anything in its 
 place and in its time requires the whole system of meanings 
 involved in perceiving a thing in its place ; and it is the room 
 and not Switzerland which has these perceived meanings for 
 him at the time of his recollecting. If he could be induced to 
 discard the distinctive meanings of perception, however, he 
 would have no good reason for denying that the place of his 
 remembered ascent of the Matterhorn is really Switzerland, 
 and the time of it the past.
 
 iv] THE STUFF OF FANCY 71 
 
 As we have seen, an explicit recollection is dated, that is 
 to say, it is recollected in a context of other recollected events. 
 As soon as we direct our minds to any landmark in the past 
 we begin to recollect the surroundings of the landmark, and 
 this process calls for no special comment so long as it goes 
 on spontaneously and readily. There is a great deal to be 
 learned from it, however, when it is baulked and forced to 
 hesitate. 
 
 The process of gripping an elusive or uncertain recollection 
 calls for a sort of tentative palpation in the dim borderland 
 of memory. We have a hazy idea of what the facts were, and 
 then we ask ourselves whether they were thus and thus. Ac- 
 cording to the usual theories, these tentative palpations are 
 so many halhns d'essai in the way of present reconstruction 
 which we try to include in the representative fabric which we 
 are weaving at the moment. A much simpler hypothesis is 
 open to us. Let us suppose that these hallons d'essai are really 
 dateless and placeless recollections which appear to float before 
 us just because we cannot apprehend their determinate position 
 at some former place and time. In point of fact, recollections 
 must float so long as they are dateless and placeless, and the 
 date or place of anything can scarcely be an intrinsic property 
 of it since it is always a relative matter which the context 
 decides. Even percepts appear to float in this way during 
 vertigo or in a summer haze, simply because their relation to 
 our usual points de repere is suddenly altered. These tentative 
 palpations towards explicit recollection, therefore, need not 
 be present candidates for an obsolete constituency. They may 
 very well be old things whose specific context in the past is 
 not remembered at the time of this tentative palpation, and 
 then we search for the context. " I remember a round tower 
 such as you describe. Where was it again ? Oh, yes ! It was a 
 signal tower in Donegal. I saw it from a boat, I remember. 
 Yes, the headland ran so and so and the tower was — No, that 
 was a Martello tower, Napoleonic, I think, or perhaps Eliza- 
 bethan. I am vague about these things, and the boatman 
 told me one or the other. Let me think. Oh, yes ! I mean the 
 Tower at Glendalough, among the Seven Churches. It is at
 
 72 THE STUFF OF FANCY [ch. 
 
 the head of one of the h)veHest valleys in Wicklow..." That 
 is the kind of process that occurs. 
 
 The context of a recollected event is usually said to be 
 revived through association, and this description of the facts 
 may be accepted without challenge, since association, in 
 modern psychology at all events, is interpreted very loosely. 
 Jung, for example, classifies associations into numerous groups, 
 some of them logical like ' simple predicates ' or ' relations of 
 the verb to the subject,'' some linguistic like ' clang associa- 
 tions ' and ' completion of words,'' some factual like ' co- 
 existence,' some classificatory like ' co-ordination '' and ' sub- 
 or supra-ordination,'' and so on\ The truth seems to be that 
 any conjunction once apprehended may be subsequently 
 associated, and the examples range from mere togetherness 
 in time to logical implications. Association is just the con- 
 servatism of the mind apprehending things in their old setting, 
 so that anything that recurs in our experience tends to be 
 apprehended along with its former accompaniments. It is not 
 difficult to find examples, but perhaps I may be allowed to 
 quote a passage from an entertaining writer by way of 
 illustration : 
 
 Nor a clergyman. I would not encourage that — not because I do 
 not admire and esteem the clergy, but I don't like the idea of a young 
 man leaving the sea for the Church. Not the right preparation, I'm 
 sure. Sailors, my dear. So profane. Parrots you know ; one hears such 
 dreadful things. Tlie Church should be a call, I think, not a deliberate 
 choice like that. And then again the expense of the preliminaries ; the 
 examinations, so trying ; Hebrew, 1 believe : Greek : the waiting for a 
 curacy. And then the curacy itself. Vicars' wives, I am told, can be 
 so vexatious. I remember poor Mrs Rackshaw very vividly. Such a 
 temper, my dear. But, of course, if he was successful it would be very 
 nice for you, although bishops' wives have no title, nothing. Very 
 unfair, I think — the Lord Bishop and plain Mrs. Not just ^. 
 
 If the good lady thus reported had developed her associations 
 into full recollections she might have been excellent company. 
 There is hope in the parrot and the Lord Bishop and poor 
 Mrs Rackshaw. If, again, she had resolutely controlled her 
 
 ^ Studies in Word Association, chap. ii. 
 * E. V. Lucas, Mr Ingleside, pp. 275-276.
 
 iv] THE STUFF OF FANCY 73 
 
 associations in the interest of her subject, her advice might 
 have been ahitiost as valuable as she thought it was. Associa- 
 tion, for the most part, falls between these two stools, and. 
 so is vaguely reminiscent without recollection, vaguely pertinent 
 without logic, vaguely old without the beauties of antiquity, 
 and vaguely new with all the insipidity of staleness. 
 
 Association, indeed, is a species of imperfect recollection. 
 It is half-recollection. To apprehend a connection is, in 
 principle, to retain it, and this fact explains the status and 
 the character of all associations. The ' association of ideas ' 
 was fondly cherished by a former generation of psychologists 
 because it was interpreted primarily as the association of 
 images, and these, in their turn, were half-recollections of some 
 connection formerly perceived. The fact that association is 
 half-recollection distinguishes mereassociates from recollections 
 proper ; for mere associates lose something in comparison with 
 memories. Their edges, to speak metaphorically, are rubbed 
 off when their context is forgotten. They are even more 
 corroded and decayed than memories, and this corrosion makes 
 them, to all appearance, intrinsically dateless and placeless. 
 These associated images have lost too much detail to have any 
 assignable position, and so they have to Hoat. Sometimes, it 
 is true, they float only temporarily because of a momentary 
 lapse of memory. Psycho-analytic literature, for example, is 
 crowded with unidentified images which finally receive their 
 place in some definite recollection. But, for the most part, 
 associated images are not of this order. They result from 
 repeated perceptions of the same kind of thing. If Nimrod 
 has seen an ibex's head but once, he may always be able to 
 remember a place for it, but if he has seen a moose's head 
 scores of times, the odds are that he retains only the common 
 residuum from those perceptions, and not a specific, individual 
 recollection. 
 
 Because some images are so corroded that they are, so to 
 speak, one thing as much as another, some authors have con- 
 cluded that images are inherently general in their nature, and 
 hence have inferred either that the existence of images is 
 general knowledge, or that they are fundamental in all general
 
 74 THE STUFF OF FANCY [ch 
 
 knowledge. It is not so in fact. A perceived thing can 
 symbolise a general idea quite as well as an image can. Peaches 
 may symbolise opulence as well as the apples of the Hesperides, 
 and the poppies in Flanders are as suggestive of death as 
 Charon and his fabled ferry. Images, to be sure, are often 
 chosen as illustrative symbols. For this purpose we often prefer 
 half-recollections to full recollections because it is too much 
 trouble to recollect properly. A great many people tell us that 
 ' doctors say so and so ' to save themselves from considering 
 what doctor ever said anything of the kind. 
 
 Images, then, have not in themselves a more general meaning 
 than perceived or remembered things. On the other hand, 
 the meaning of imaged things is emancipated from any parti- 
 cular context in space and time, and so it may simulate logical 
 generality because of its omission of manifestly particular 
 circumstance. Images, however, are thoroughly particular 
 although they are placeless and dateless. An image has 
 extension, figure and duration just like anything perceived 
 or remembered. The only difference is that it is cut loose 
 from the order of perceived and remembered fact. In a word, 
 images are precisely what they appear to be, spatial, temporal 
 and physical, yet without a home in the perceived order of 
 time and space. The fact thafgeneraP thinking is so frequently 
 accompanied by imagery is only a proof of the hold which 
 the physical world has upon our spirits. We feel that nothing 
 really counts unless it has a place in perceived fact, and when 
 we cannot perceive or remember an illustration we do the 
 next best thing by imaging one. 
 
 Summing up, then, we may fairly claim that this accoinit 
 of images has been shown to hold of all the elements of 
 imagery, of associated images, and of illustrative ones. On 
 the other hand, the problem of creative imagery has not yet 
 been touched, and the reader may reasonably suspect that there 
 are peculiar difficulties concerning it. Our theory, he might 
 say, would supply the stones for Kubla Khan's dream palace, 
 but it could not build that stately edifice. Who ever saw a 
 dragon, or a wyvern, or a chimera, and who had ever the 
 slightest difficulty in imaging one ? A cruel fate has con-
 
 iv] THE STUFF OF FANCY 75 
 
 demned thousands of little waifs to loiter and starve in dingy 
 streets, but sleep may bring them its recompense, and their 
 dreams may unfold the limitless spaces of enraptured lands. 
 What are we to say, then, of these cases ? 
 
 The most striking examples of constructive imagery (as 
 opposed to constructive imagination which is not our present 
 topic) are probably to be found in dreams. If we can show, 
 then, that the by-ways of dreamland are not radically different 
 from the thoroughfares of waking life, there will be a strong 
 presumption that our theory holds for the whole range of 
 imaging. 
 
 The labours of the psycho-analysts have shown this to be 
 exceedingly probable. Psycho-analysts have proved that 
 dreams are not the utterly sporadic anomalies that credulous 
 ignorance or misplaced romanticism have supposed, and they 
 have also shown that there are principles in dreaming and not 
 mere caprice. The detail of these principles, to be sure, is still 
 matter of dispute. Indeed, it will continue to be disputed so 
 long as merely figurative beings like the so-called ' censor ' 
 are supposed to work in earnest, so long as the crude psy- 
 chology of ' outer ' and ' inner ^ (sparingly flavoured a la 
 James or a la Bergson) is taken to be the basis of extroversion 
 and introversion, and so long as perverted ingenuity is wasted 
 in showing that any straight stick is ithyphallic or any want 
 sexual on the gi'ound that any stirring of the body is remotely 
 connected with sex. It is time to call a halt when the 'Electra- 
 complex ' or the ' father-imago ' is taken to be the root of the 
 soul, when avarice is solemnly alleged to be a symbolic develop- 
 ment from a child's curiosity in his excrement, when Jung 
 takes fifty pages to show that Hiawatha is an incest-phantasy', 
 or when Freud tracks every associate back till he can point 
 to an infantile fixation and then triumphantly assumes that 
 every infant has a midwife's knowledge. Still, when all has 
 been said, the psycho-analysts have certainly discovered some 
 strands of the Ariadne thread which is needed for these mazes. 
 Let us glance at the phenomena and the method of inter- 
 preting them before considering the theory. 
 
 1 Psychology of the Unconscious, chap. vii.
 
 76 THE STUFF OF FANCY [ch. 
 
 The phenomena of dreams seem to show a Bacchantic caprice 
 of images made out of nothing, leading nowhither, and con- 
 taining in themselves neither rhyme nor reason. Everyone, 
 however, by looking a little closer, can find that these bricks 
 are not made without clay even if they are made without 
 straw. There is plenty of experience to show that incubus or 
 nightmare is partially explained by difficulties in breathing 
 and that it is much more pronounced when the sleeper is lying 
 on his back; or that the flapping of a window-blind often 
 synchronises with the dream of a tempest. Again, there is 
 plenty of experience to show that many dreams have a point 
 de repere in the dreamer's recollection of the day before. 
 Sleeping dogs follow the chase when the hunt is over, and a 
 sharp word spoken in haste often finds its nemesis in the 
 regi'ets and disasters of dreamland. It is clear, however, that 
 this type of explanation is very partial indeed. These scanty 
 fragments of yesterday"'s dream are related to the dream as a 
 whole much as half a dozen pebbles are related to a concrete 
 pillar. Granting that the flapping of the window-blind gives 
 rise to an unquiet sort of dream, it will not usually explain 
 what in detail is dreamt. Even the general atmosphere of 
 the dreamer's yesterday will not explain that. Nightmare, 
 it is true, is rife when the plague is near, and war which 
 oppresses the waking spirit continues to haunt it in dreams ; 
 but these facts, it is held, do not explain the distinctive details 
 of the dream, though they may explain why the general trend 
 of the dream is menacing. Accordingly, if the hizarrer'ie of 
 dreamland is subject to law, the interpreter must probe deeper 
 than these surface resemblances. 
 
 The method of psycho-analysis is perfectly straightforward 
 in its essence. The analysis tracks the elements out, and 
 the method consists in giving association free play. It is 
 assumed that no detail in the dream is too insignificant to be 
 neglected, and that every element can be tracked by association 
 to its source in former experience. It is also assumed that 
 every disturbance in the normal time of associative response 
 indicates emotional perturbation. This is not the place to 
 consider the technique of psycho-analysis, or to multiply
 
 iv] THE STUFF OF FANCY 77 
 
 examples. A single example must suffice, and this one from 
 Jung will serve : 
 
 A young patient dreams as follows : / am standing in a strange 
 garden, and pluck an apple from a tree. I look about cautiously, to make 
 sure no one sees me. 
 
 The associated dream material is a memory of having once, when a 
 boy, plucked a couple of pears surreptitiously from another person's 
 garden. 
 
 The feeling of having a bad conscience, which is a prominent feature 
 in the dream, reminds him of a situation he experienced on the previous 
 day. He met a young lady in the street — a casual acquaintance — and 
 exchanged a few words with her. At the moment a gentleman passed 
 whom he knew, whereupon our patient was suddenly seized with a 
 curious feeling of embarrassment, as if he had done something wrong. 
 He associated the apple with the scene in Paradise, together with the 
 fact that he had never really understood M'hy the eating of the forbidden 
 fruit should have been fraught with such dire consequences for our first 
 parents... 
 
 Another association was, that sometimes his father had punished him 
 for certain things in a way that seemed to him incomprehensible. The 
 worst punishment had been bestowed after he had secretly watched girls 
 bathing. 
 
 This led up to the confession that he had recently begun a love affair 
 with a housemaid, but had not yet carried it through to a conclusion. 
 On the day before the dream he had had a rendezvous with her^ 
 
 Stolen fruit and a bad conscience often go together. So do 
 guilt and anything else that deserved or received a whipping. If 
 the boy had dreamed of stealing jam from a cupboard we may be 
 sure that our psycho-analyst would eventually have discovered 
 the affair with the housemaid via the interview and the bathing 
 girls. We may trust his Jlair for an Actaeon-complex. Even if 
 psycho-analysts could explain (as they pretend) why the boy 
 did not dream directly of the housemaid, they could not explain 
 why he dreamed of the garden and not of the cupboard. That, 
 however, should not be expected of them. It is enough if they 
 can trace the association apres coup ; and that is what Jung 
 claims to have done in this example. 
 
 In broad outline, the theory of the psycho-analysts is that 
 a dream is a sort of translation of a ' latent content "* into a 
 ' manifest content.' This transformation is the ' dream-work,' 
 1 Analytic Psychology, pp. 303-304.
 
 78 THE STUFF OF FANCY [ch. 
 
 and the dramatic portrayal which it adopts usually shows 
 transference and condensation. This transference, however, is 
 not peculiar to dreams since the transference of desires from 
 their original objects to the physician is part of the standard 
 psycho-analytic cure. The condensation or telescoping of ima- 
 gery, again, is not peculiar to dreams, for it is characteristic 
 of all imagery, as the crowded history of a drowning man's 
 imaging proves. The 'latent content' itself is usually de- 
 scribed as a wish or an impulse or a complex of these, and the 
 theory is that the latent content is sometimes dramatised 
 directly and shown in the manifest content. Thus the Arch- 
 duke Charles of Austria, wishing to waive his claim to the 
 crown in favour of his son, yet doubting his right to do so, 
 had his scruples removed when he saw his father, in a vision, 
 laying his hand on the young child's head^ More usually, 
 however, the translation is not straightforward, and in that 
 case it is held that certain ' wishes ' are suppressed in waking 
 life either because of social conventions (as in the case of sex) 
 or because (like the instincts of childhood) they are not adapted 
 to the kind of reality which adults encounter. It is supposed, 
 however, that these suppressed wishes find a partial and in- 
 direct fulfilment in dreams, where they are bodied forth in a 
 fantasy which has its roots in them, however meaningless it 
 seems. 
 
 The further subtleties of this investigation need not concern 
 us here. What does concern us is the general conclusion that 
 dreams are dramatic presentations which can be traced to an 
 emotional or impulsive source. What relation has this con- 
 clusion to our general theory of imaging ? 
 
 The imagery in a dream is, of course, its manifest content, 
 but this manifest content, according to the psycho-analysts, 
 is not really creative. It is one of the termini in a chain of 
 association whose other terminus can be recalled to memory 
 when sufficient pains are taken. The manifest content, in a 
 word, consists of half-recollections, as in any other case of 
 association. This new theory, therefore, supports our general 
 analysis in this important circumstance, and if it were possible 
 1 Example quoted by Morton Prince, The Unconscious, p. 223.
 
 iv] THE STUFF OF FANCY 79 
 
 to explain why there should be any such thing as dramatic 
 portrayal in dreams, and how the manifest content of a dream 
 symbolises or expresses a suppressed wish, or desire, or libido, 
 this support would be very strong indeed. 
 
 Dramatic portrayal in dreams is the natural result of the 
 special circumstances of dreaming. During sleep our minds 
 are shut off from perceptual acquaintance with our surround- 
 ings, and from the usual responsive movements of our bodies. 
 For that reason the sleeper cannot have any direct acquaint- 
 ance with sensory fact except in the way of imaging or memory, 
 and the images and memories of dreams, like the stars in 
 Hobbes's metaphor, reveal themselves very clearly just because 
 they have no competitors in perception. In the second place, 
 the dreamer's mind is dissociated; indeed, dreaming is an even 
 more extreme case of dissociation than the hypnotic trance, 
 or the trance-like condition which frequently accompanies 
 extreme physical weariness. Soldiers after a long forced march 
 often wonder whether they are going mad because, when they 
 pull their minds together, they find that they have spent half 
 an hour without having been aware of anything at all except 
 a few waifs and strays of imagery rambling and tumbling and 
 lolloping after one another. Dreaming is the same kind of 
 thing ; and we all know that we can ask ourselves in a dream 
 whether we are dreaming or not, and that we may have too little 
 command over our thoughts to be able to answer the question. 
 
 Mental dissociation, in its turn, means the mind's failure 
 to achieve comprehensiveness or coherence in thinking, its 
 inability to recollect when it tries, and its powerlessness to 
 follow up the meaning of anything beyond its most super- 
 ficial aspects. It is not surprising, therefore, that the half- 
 recollections of dreams should remain half-recollections, or that 
 dream images should be accepted in childlike trustfulness with- 
 out the smallest critical effort on the dreamer's part. There 
 are exceptions, to be sure ; indeed, we sometimes suppose in 
 our dreams that we have reached the most profound conclu- 
 sions by the most masterly reasoning. Our satisfaction, in 
 these cases, however, is due to our own critical ineptitude, and 
 that is not really an occult Mant7-a exorcising the enigmas of
 
 80 THE STUFF OF FANCY [ch. 
 
 existence. In any case, there is no reasonable doubt that this 
 description holds of every twelve dreams in a baker's dozen, 
 and imagery, therefore, has free scope in dreamland. It is 
 taken at its face value, and the mind is powerless to begin to 
 scrutinise its meaning. What is more, the condensation of 
 imagery, as we have seen, is precisely what we ought to expect, 
 for images always tend to become condensed like vapour on 
 a spoon. If thought is quick, images are meteors; and this 
 huddled throng of images crowding together like a rain of 
 meteors in the sky is just the dramatic portrayal which we 
 find in dreams. 
 
 The statement that the manifest content of a dream ex- 
 presses a desire or libido is very misleading. In strictness of 
 logic a dream is only its manifest content, and its meaning is 
 just what the manifest content is taken to be, i.e. so much 
 presented fact. It is the psycho-analysts who find the expres- 
 siveness, and they find it when they succeed in tracing a con- 
 nection between the dream and the emotional source towards 
 which the associations of the dream converge. There is every 
 reason to suppose, indeed, that the emotion or the libido has 
 the same relation to dreaming as emotion has to thinking or 
 perceiving or imaging in waking life. Emotion often accounts 
 for the selection of the objects which we notice. Suspicious 
 people always find suspicious circumstances, because they have 
 no eyes for anything else. Hopeful people notice cheerful 
 things, and they see everything coulcur de rose, because they 
 do not notice the sombre tints. The emotions make moun- 
 tains out of molehills, because they keep our attention away 
 from everything except the molehills. It is the same with 
 dreams and their latent content. The dreamer selects those 
 images which are consonant with his desires or libido. The 
 images he selects do not mean the libido, but the libido is the 
 cause, or part of the cause, of his selection. Dream meanings, 
 therefore, are not peculiarly occult; and the facts of dreaming 
 support our earlier analysis. 
 
 Many will object still, however, and think that the crucial 
 difficulty of creative imaging remains unanswered. There are 
 no gorgons, and yet it is possible to have gorgon-images readily
 
 iv] THE STUFF OF FANCY 81 
 
 enough. Snakes we know and women's heads we know, but 
 not women's heads with snakes for hair. The elements of 
 dreams and of all imagery, it is urged, may be half-recollec- 
 tions ; their order and arrangement, broadly speaking, may be 
 similar to that of perception or reminiscence ; but the image 
 itself, the gorgon or the dream palace or whatever it may be, 
 is a wholly new combination of these elements, and so a new 
 creation. This argument, to be sure, does not prove that 
 images are mental. A house is not more mental than a glacier, 
 although a glacier is constructed by nobody and a house is a 
 combination of material which human beings have brought 
 into existence ; and it is hard to see how anything must be 
 mental if its elements need not be. On the other hand, the 
 principal problem of this essay is to consider whether things 
 are literally discovered by the mind, and these arguments, it 
 may be said, are amply sufficient to prove that creative imagery 
 is not in any serious sense discovered fact. 
 
 As it seems to me, the natural inference from this line of 
 argument is that the imaged gorgon is a combination of ele- 
 ments which the mind has put together. If so, the apprehen- 
 sion of this result of the mind's workmanship would be just 
 the discovery of that special kind of product. The man who 
 has built a hut perceives his handiwork in the same way as he 
 perceives the hills or the sky, and if knowing and making are 
 distinct in this instance they should also be distinct in the case 
 of images. To combine half-recollections into a single whole is 
 one operation; to apprehend this product is another operation. 
 The second only is knowing, and might very well be discovery. 
 
 There is another side to the story, however. If it is a mis- 
 take to confuse between the elements of an image and the 
 combination of these elements, it is also a mistake to suppose 
 that the whole question turns upon the novelty of imaged com- 
 binations. The things of the fancy may fill us with wonder and 
 delitiht because of their freshness, as when we read of 
 
 Silver hammers falling 
 On silver anvils, and the splash and stir 
 Of fountains spouted up and showering down 
 In meshes of the jasmine and the rose.
 
 82 THE STUFF OF FANCY [cii. iv 
 
 On the other hand, the freshness of these pictures only proves 
 that we have never perceived this fancied arrangement of 
 physical things ; and it is not at all absurd to suppose that 
 the poefs fancy was literally finding. The fancy does not dis- 
 cover perceived fact, but no one ever supposed that it did. 
 When we speak of physical things we mean to include all that 
 we can perceive in them and much more than that ; but we 
 perceive more in things than we can image, and we cannot 
 ascribe the perceptual order of meanings to imaged things 
 without becoming the victims of hallucination. Fanciful 
 meanings are image-meaning's, fancy's laws are the laws of 
 imaged things, and its reality the reality of the world as 
 imaged. Images, in a word, are parts of the physical world 
 imaged, and that is what we discover through the fancy. 
 
 The trouble is that images are sometimes so similar to per- 
 ceived things that we ascribe the same properties to both, and 
 then we fall into confusion. There is some excuse for this 
 since the most impressive images contain an illusion of per- 
 ceptibility, and even the painter tricks us into the half-belief 
 that we perceive the things of his fancy. It is seldom, how- 
 ever, that we are wholly deceived like the birds that pecked 
 at the grapes of Apelles ; and dreamers are not really tricked 
 so long as they continue to dream. They would only be 
 tricked if they slept, like a hai-e, with their eyes open.
 
 CHAPTER V 
 
 THE WORLD OF COMMON BELIEF 
 
 These judgments may, in the strictest sense, be called judgments of 
 
 nature I acknowledge that, if we were to rest in these judgments of 
 
 Nature of which we now speak, without building others upon them, 
 they would not entitle us to the denomination of reasonable beings. 
 But yet they ought not to be despised, for they are the foundation upon 
 which the grand superstructure of human knowledge must be raised. 
 
 Reid, Intellectual Powers. 
 
 Ix this chapter I intend to deal with a set of questions arising 
 directly out of the position reached in the three preceding 
 ones. This conclusion may be summed up by saying that we 
 perceive sign-facts which are parts of the physical world, and 
 indicate other parts of it; and that we encounter the same 
 world in recollection and imaging through our acquaintance 
 with sign-facts partially identical with perceived ones, although 
 corroded in feature and altered in significance. 
 
 This physical world, however, is a believed thing rather 
 than a perceived or remembered one. Perception and memory 
 are only plummets giving us a sounding here and there, and 
 I wish to consider the systems of meanings which, taken to- 
 gether, are the believed thing which common sense calls the 
 
 \ 1 physical world. Belief, of course, is not confined to this world, 
 for we may believe in pure mathematics or in abstract ethics; 
 and although Newman has told us that ' no one will die for 
 his own calculations ' he has also told us that ' many a man 
 will live and die upon a dogma ^^ On the other hand, we do 
 
 >">- believe in this physical world of common sense, and our judg- 
 ment (as Hume said) ' peoples it-."' We can perceive, remem- 
 ber and judge the same things, but our judgment apprehends 
 
 1 The Grammar of Assent, 3rd ed. p. 90. 
 
 2 Treatise, bk i. pt in. sect. ix. : " 'Tis this latter principle which peoples 
 the world, and brings us acquainted with such existences, as by their removal 
 in time and place, lie beyond the reach of the senses and memory." 
 
 6—2
 
 i 
 
 84 THE WORLD OF COMMON BELIEF [ch. 
 
 these things at its own level, and interprets the rudimentary 
 meanings of direct memory and perception in a much more 
 explicit fashion. And we have seen already that perception 
 and judgments of perception are not the same. To choose 
 Meinong"'s example, there is a difference between the red cross 
 which is perceived and the belief that the cross is red^ True, 
 the only possible evidence for the belief that the cross is red 
 is the perception of the red cross, and this evidence, we are 
 sure, is amply sufficient. None the less, a thing, in so far as 
 it is perceived, is not a proposition, and propositions, in 
 strictness, are what we believe. There is the same kind of dis- 
 tinction between perceived meanings and the set of propositions 
 which interpret these meanings in terms of judgment. 
 
 It may seem very absurd, prima facie ^ to pay any attention 
 to this believed thing which common sense calls a world, 
 because science and philosophy pride themselves upon being 
 better interpreters of fact than the plain man. What the 
 world is in detail, what space and time are, what the consti- 
 tution of matter is, must be inferred from the detail of what 
 we perceive or remember, and require the most intimate 
 partnership between rigorous deduction and unbiassed obser- 
 vation, between parsimony of principles and genius for experi- 
 ment. On the other hand, the data of science and philosophy 
 include judgments of perception as well as perception itself, 
 so that even the most philosophical improver of common sense 
 cannot avoid the problem altogether, and the reflective inter- 
 pretations of common sense, however hasty and crude and 
 dogmatic they may be, must have gold in them as well as 
 dross. At the worst they are a muddy deposit which may be 
 clarified by those who make a business of clarifying, and, at 
 the best, common sense may be right in its main conclusions 
 although impatient of detail, confused in its expressions, and 
 halting in its proofs. 
 
 We believe (I think justly) that there is a physical world. 
 
 This world contains the things we perceive and remember and 
 
 expect, and it also contains a vast apparatus of things which 
 
 connect these. This connective tissue permeates and encom- 
 
 1 Veber Annahmen, Kap. 6, 1st ed. p. 110 and passim.
 
 v] THE WORLD OF COMMON BELIEF 85 
 
 passes these perceived things, and it is beheved to exist just as 
 certainly as they and much in the same manner. If a letter 
 reaches me from America I can obtain some evidence about 
 its journey from my own perception and from that of other 
 people. The letter was perceived by the writer and by some 
 postmen and sorters. Someone saw it when it was dropped 
 into a bag, and someone saw the bag put on board a ship. 
 The pilot who steered the ship out of the harbour used his 
 ejes; the captain or the mates perceived the ship and the 
 sea during the whole voyage ; and so on. Still there was far 
 more in the ship than the ship's company perceived at any 
 moment, and more in the ocean than the watchers saw from 
 the bridge. The ship, and the ocean, and the world itself are 
 believed things inferred from these partial and intermittent 
 soundings of perception and memory, even granting that these 
 soundings have always a meaning beyond themselves. The 
 world is a continuing thing, spread out and enduring; and its 
 features, except tentatively and in shreds and patches, are not 
 perceived but judged and inferred. The things which appear 
 to perception also appear to belief, and philosophers should 
 scrutinise these beliefs in the same spirit as they scrutinise per- 
 ception. We perceive sign-facts, and we believe — what .'' 
 
 When Hume said that judgment 'peoples the world"* he 
 assumed without any question that belief attaches itself to the 
 impressions of the senses and of memory^ We are tied down 
 to these beliefs and constrained by them just as if our ideas 
 were things (to borrow a phrase from Tolstoi). Hume did 
 not suppose, however, that this sense of constraint, this firm- 
 ness and steadiness of conception, was confined to perception 
 and memory. He knew that it also belonged to anything 
 
 1 Treatise, bk i. pt in. sect. ix. : "Of these impressions or ideas of the 
 memory we form a kind of system, comprehending whatever we remember to 
 have been present, either to our internal perception or senses; and every par- 
 ticular of that system, join'd to the present impressions, we are pleased to call 
 a reality. But the mind stops not here. For finding that with this system of 
 perceptions there is another connected by custom, or, if you will, by the rela- 
 tion of cause and effect... it forms (these ideas) into a new system, which it 
 likewise dignifies with the title of realities. The first of these systems is the 
 object of the memory and senses; the second of the judgment."
 
 86 THE WORLD OF COMMON BELIEF [ch. 
 
 which we take to be causally connected with perceived or re- 
 membered things. Indeed, Hume meant by 'judgment' pre- 
 cisely this extension of perception and memory by causal 
 interpretation, and he was concerned to show that all causal 
 interpretations are extensions of the memory and senses. It 
 is plain that he was right in this. Causal interpretation, to 
 be sure, supplements observation and interpolates numberless 
 unobserved links, but the world of common belief rests on per- 
 ception in the end. If we can observe or remember that 'the 
 stick began to beat the dog ' we may be able to infer the chain 
 of events which led to the old woman's difficulty at the stile 
 and the rest of it, but without this observation or recollection 
 the whole chain of inference would dangle without a support, 
 and our ideas about the world would be as impotent as Baron 
 Munchausen's method of descending from the moon, when he 
 slid halfway down the rope and then made use of the upper 
 portion that had become useless. 
 
 Assuming, then, that the physical world is a realm of 
 existence in which fragments are perceived or remembered and 
 the rest ' peopled ' by things which are required to satisfy our 
 causal interpretations of these fragments, we have to ask how 
 far, in broad outline, this world may be said to be discovered. 
 To speak more accurately, the physical world is that system 
 of things which, if it exists, is the foundation of the truth of a 
 certain set of beliefs, i.e. the beliefs based directly on perception 
 and memory and the beliefs derived from these by causal in- 
 terpretation. To believe in the existence of this world is to 
 believe in the truth of these propositions, and plainly the most 
 important question to ask is what these propositions are. 
 
 Nearly anyone would admit that these propositions, if they 
 are true, are about the world or parts of it, and that the world 
 guarantees and controls them. That admission, however, does 
 not tell us what these propositions are, how they are about 
 the world or how the world controls or guarantees them ; and 
 so we must ask these questions for ourselves. 
 
 Let us ask, then, in the first place, what these propositions are. 
 What precisely is before the mind when anyone believes proposi- 
 tions like ' The cross is red ' or ' lialage has torn her pinafore ' ?
 
 v] THE WORLD OF COMMON BELIEF 87 
 
 Adapting a term of Meinong's^, we may say that proposi- 
 tions are asserted 'objectives.'' An example will show what is 
 meant. When we perceive a red cross we are justified in be- 
 lieving that the cross is red. ' That-the-cross-is-red ** (or ' the- 
 redness-of-the-cross,*' if the reader prefers) is the objective of 
 the judgment. The proposition ' The cross is red ' asserts this 
 objective, but anyone who questioned this judgment or began 
 to consider ' Well ! what about this redness of the cross .''' 
 would merely contemplate the objective without also asserting 
 it for true. The indicative mood expresses assertion, while the 
 optative or subjunctive moods do not ; and yet precisely the 
 same objective — this 'redness-of-the-cross' — is before the mind 
 in all these moods. 
 
 Propositions, therefore, are asserted objectives; and these 
 objectives in their turn are manifestly complex. The next 
 step, therefore, is to find the elements of the objectives. An 
 objective is the kind of fact which we express by a verbal noun, 
 and a verb usually expresses a relationship between at least 
 two things. That is clearly true in the case of Lalage and her 
 pinafore, so perhaps we may turn to the more fundamental 
 example of the red cross. The objective in this case contains 
 a relation between the cross and redness, i.e. a relation between 
 a logical subject or individual and a universal quality belong- 
 ing to it. And that raises difficulties. 
 
 It is clear that the red cross is an existing thing and that 
 redness is not. ' Redness ' does not do any work. It does 
 not even keep its place through inertia, for it has no place and 
 no inertia. It is only a universal. So much is clear, but the 
 very obviousness of this analysis is beset with difficulties. The 
 objective, we have said, is complex and contains a universal 
 related to its logical subject. But what is this subject .'' If it is 
 the red cross, the subject, so to speak, is red already, and there 
 is therefore no need for, and no value in, the universal. If, on 
 the other hand, the logical subject is the cross unqualified by 
 redness, the trouble is still greater, if that be possible. For the 
 cross is red, and it is never unqualified in fact. The supposed 
 subject would, therefore, be a nonentity. Again, the red cross 
 ^ Ueher Annahmen , passim.
 
 88 THE WORLD OF COMMON BELIEF [ch. 
 
 is perceived. It is just one thing, and to analyse it into two 
 things is the merest makeshift. 
 
 Perhaps we may circumvent these difficulties by considering 
 another aspect of the affair. We perceive the red cross and 
 we believe that the cross is red. Our belief, we say, is about 
 the red cross, and the perception of the cross is the evidence 
 for the belief. What, then, is the relation of the belief to the 
 percept, and in what sense is it about the thing perceived ? 
 
 As we have seen, we perceive physical things, and we do not, 
 strictly speaking, perceive mere percepts any more than we 
 expect expectations or promise promises or intend intentions. 
 Perception, it is true, like any other mode of apprehension, im- 
 plies certain limitations in our acquaintance with the thing, 
 but although we apprehend things in this limited fashion we 
 never attribute these limitations to the things themselves. 
 Thus although the red cross is perceived, it is not a mere per- 
 cept, but a thing. The percept, on the contrary, is only a 
 sign-fact, in part literally identical with the thing, in part 
 significant of certain further aspects of the thing's thinghood. 
 
 Now this same thing that we perceive is also the subject 
 about which we judge, and the universal 'redness' is about its 
 subject simply in the sense that the thing is characterised by 
 this universal quality just as it is characterised by many other 
 universal qualities. We refer to things in judgment, not to 
 objectives, precisely as we perceive things and not percepts ; 
 and an objective of this kind means a thing in so far as it is 
 restricted in the special mode of judgment, while a percept is 
 the same thing in so far as it is restricted in the special mode of 
 perception. Judgment always selects some one of the many 
 universal characteristics which things have. Perception does 
 not select in this fashion although it selects in another fashion. 
 None of the restrictions implied in either judgment or percep- 
 tion is attributed to the thing itself, and, at the same time 
 there is no need for supposing that things themselves are not 
 literally all that they are perceived to be and all that they are 
 judged to be, though they are not limited to the features or pro- 
 perties which perception or judgment is capable of selecting. 
 
 Let us consider some of the objections to this position.
 
 v] THE WORLD OF COMMON BELIEF 89 
 
 Those who beheve that we perceive percepts (or sense data) 
 usually regard this percept as an ultimate of ultimates, so 
 much pure fact wholly self-enclosed and incorrigibly real. 
 This view, I have tried to show, is mistaken, but it would 
 vitally affect our theories of the world if it were true. As we 
 have seen so very often, it is plain that a thing endures al- 
 though it is perceived intermittently, while it is equally plain 
 that a percept cannot endure, simply because it is confined to 
 intermittent apprehension. Moreover, it is plain that a per- 
 cept cannot be identical with an objective since the objective 
 is judged and not perceived, since the percept is particular 
 while the objective contains a universal, and since the percept 
 has not the kind of complexity which the objective has. For 
 these and similar reasons writers of this school give a totally 
 different analysis of the facts from the one we have chosen. A 
 judgment of perception, they say, is the apprehension of an 
 objective which corresponds to a percept. The objective, as a 
 whole, is about the percept, and the percept is never a con- 
 stituent of it. 
 
 As it seems to me, anyone who holds this theory is forced 
 to relinquish the most certain thing in this puzzling matter. 
 What seems to me most certain is that I discover the character 
 of the thing I perceive in all true judgments of perception. 
 * This book which I see is red.' If I really mean what I say 
 when I make this assertion, my belief must refer to the book 
 itself, and I must also perceive that very book. The theory 
 before us ignores the book altogether, and cheats us with ob- 
 jectives and sense data instead; and this omission seems to me 
 a sufficient refutation of it. 
 
 A second objection runs somewhat as follows. When we 
 assert that the cross is red, the 'is' in our assertion is unintel- 
 ligible unless it asserts identity; and that is inconsistent with 
 our analysis. For 'redness' is not a part of the book, like one 
 of its pages, still less identical with the book; and the book 
 is a thing and not a collection of universals like redness, 
 heaviness, and the like. 
 
 This argument raises many problems concerning the status 
 of universals and some of these must be postponed until the
 
 90 THE WORLD OF COMMON BELIEF [ch. 
 
 next chapter. Its main purport, however, can be examined 
 independently of these investigations. Redness, I take it, is a 
 characteristic of many red things — of sunsets and uniforms 
 and sealing wax and new-born infants, for example. When we 
 say that any of these things is red, we mean that it is red and 
 we do not mean that it is identical with redness. The 'is' of 
 predication is not the 'is' of identity. What is it, then ? 
 Why, the 'is' of predication. 
 
 According to a third objection 'redness' is only a concept, 
 and consequently only a mental gloss upon things, not a dis- 
 coverable property of them. We perceive reds, not redness, 
 and redness itself is only a product of mental comparison. It 
 is the result of mental experiments which n)ay be utterly 
 capricious, for, if we choose, we may compare anything we like 
 from carburetters to chalcedony, and from the mad hatter's 
 tea-party to the debates on Olympus. 
 
 There is no need for profound thinking to answer this ob- 
 jection. Certainly we have to conduct intellectual experiments 
 in order to find out what 'redness' is, and we have complete 
 liberty concerning the things that we choose to compare, but 
 these experiments, like any others, are a means of discovery, 
 and the results of the experiments are made for us and not 
 by us. The air in ancient Corinth contained its twenty per 
 cent, (or thereabouts) of oxygen before Priestley discovered 
 'dephlogisticated air\' or Van Helmont distinguished ga,^ 
 from bias. Similarly, redness is discovered to be a common 
 property of blood and of sunsets whenever we choose to make 
 this experiment in comparing. 
 
 A fourth objection still remains. Judgment, we are told, 
 confounds the unity of things by dividing their substance. It 
 gives us a series of items in an inventory, like a grocer's list. 
 In fact vermilion is red and heavy and so forth, but it is all 
 these in one. In our judgments we take these items separately, 
 like the rosy cheek and coral lips and snowy brow of a 
 Jacobean poem. The dead hand of analytic separation is 
 
 1 Stephen Hales had priority in this discovery, but his researches were not 
 followed up.
 
 v] THE WORLD OF COMMON BELIEF 91 
 
 heavy on all judgment, and this paralysing analysis, like a 
 speck of rust, corrodes all the linen. 
 
 Now it is true that judgment selects items, but the selection, 
 unless it is wholly misconceived, does not impair a thing's 
 integrity. To say that vermilion is scarlet and heavy does not 
 mean that it is two things, a scarlet thing and a heavy one, 
 somehow connected together. It is just one thing all the time, 
 scarlet and heavy and much besides; and the analysis of these 
 properties never divides the substance. 
 
 The point is fundamental, for judgment is incomparably the 
 most important means of apprehending things, and if things 
 are not really and ultimately what they are judged to be, then 
 farewell to serious thinking. The implications of perception 
 and memory are worth very little unless they are followed out 
 into a chain of judgments, and the chain itself is worthless 
 unless each several link in it can be proved. He who trusts 
 himself to logic must trust altogether. He cannot seriously, 
 like the instrumentalists or Mr Bradley, step into the stream 
 with one foot and keep the other on the bank ; for the bank 
 is not firm enough and the stream too masterful. According 
 to Mr Bradley, logic is forced to assume what metaphysics is 
 forced to reject. That is marching with the Pretender and 
 investing in the Funds. All thinking must assume what logic 
 assumes, and realism, at bottom, is just the assertion of this 
 principle. Indeed, we might define realism as the theory of 
 knowledge which literally accepts all the logical assumptions 
 which Mr Bradley sees to be involved in logic. The most im- 
 portant of these assumptions are the following. Logic, 
 Mr Bradley says, is forced to assume that the processes of 
 distinguishing, comparing and constructing do not modify 
 their data \ that attention, retaining and holding together be- 
 fore the mind do not alter the content apprehended", and that 
 the Identity of Indiscernibles is true in the sense that 'so long 
 as an ideal content is identical no change of context can 
 destroy its unity'."" It would not be easy to give a better 
 definition of our thesis. 
 
 1 Principles of Logic, p. 506. "^ Ibid. p. 502. =* Ibid. p. 264.
 
 92 THE WORLD OF COMMON BELIEF [ch. 
 
 Perhaps we should linger to notice a consequence of our 
 position before proceeding to a further description of the world 
 of common belief. Just as the judgment that vermilion is red 
 and heavy seems, on its first aspect, to disintegrate the one 
 perceived thing into a list of separate items, so a chain of 
 judgments may seem to separate the world itself into so many 
 separate rigidities. It is important to notice that this conse- 
 quence does not follow. It is logically possible, to be sure, that 
 the world is a loose, disjointed, strung-along affair, but this 
 logical possibility is certainly not a logical certainty, and even 
 if Caird and his followers were right in maintaining that the 
 world is a ' seamless unity,' the chain of logical inferences 
 would not therefore be impugned. It does not follow that 
 linen is rent into whiteness and glossiness merely because we 
 are able to judge truly that it is both white and glossy. It is 
 white, glossy linen, and the predicates of these two judgments 
 characterise one and the same piece of linen without doing 
 violence to its unity. Each is partial but neither falsifies. 
 Similarly any chain of true beliefs which characterises different 
 portions and features of the world is partial but does not 
 falsify. The world need not be a mere collection just because 
 it can be parcelled into sorts, and it need not be strung along 
 in a chain simply because true beliefs are linked together chain- 
 wise. The world would have room for this chain of beliefs 
 even if it were a seamless unity, and neither the links nor the 
 chain need be false just because the world itself is more 
 intimately knit than any chain of items. 
 
 Summing up, then, we may say that the world of daily life 
 is a believed thing rather than a perceived or remembered one, 
 although some of the things in it are perceived or remembered. 
 It is a set of connected things having the characters recognised 
 in judgment; and this explains the sense in which the pro- 
 positions in which we believe are about things and the sense 
 in which things control our beliefs. 
 
 We have also seen that the world contains the structure 
 and implications of logic, precisely because it is thinkable. Its 
 empirical properties, however, are not merely logical. The 
 logical connection of subject and predicate, for example, does
 
 v] THE WORLD OF COMMON BELIEF 93 
 
 not explain the yellowness of ochre or the blueness of the sky, 
 and the logical relation of difference does not account for the 
 empirical fact that Warsaw is not Paris or that geese are not 
 swans. This empirical connectedness and difference, therefore, 
 which is not a merely logical thing, ought to be considered. 
 
 We know that ochre is yellow because we perceive yellow 
 ochre. Perception, therefore, is our warrant for believing in 
 this empirical proposition, and perception similarly entitles us 
 to believe in many other empirical propositions which assert a 
 connection between things. When we believe that the pea is 
 under the thimble, or that the report comes after the flash, or 
 that Lalage has torn her pinafore, our beliefs are either in- 
 ferences from perception or directly based upon it, and if they 
 are inferences, these inferences, in their turn, depend upon the 
 direct perception of empirical things empirically related. We ' 
 may be wrong about the pea and the thimble, but at some 
 point we perceived these two things in a certain spatial re- <:^ 
 lation. We may have to remember the flash when we hear 
 the report, but some intervals of time may be apprehended in 
 a single span of perception. We may have misjudged Lalage, 
 but we have often perceived children and puppies tearing 
 things to pieces. 
 
 We interpret the world spatially, then, and we have to 
 distinguish between the space of direct perception and the 
 space of belief. As we have seen, the things we perceive are 
 obviously extended, or, at any rate, things which are seen or 
 touched are extended, and also toothache and bruises when 
 we feel them. Perceived things, moreover, are suffused with a 
 perceived meaning, and this, in its turn, indicates a wider 
 spatial context. The order of space in Avhich we believe is an 
 interpretation of the sign-facts perceived or remembered. It 
 is based on the properties and the meaning of these sign -facts. 
 Common sense, it is true, does not think out these interpreta- 
 tions to their full conclusion, and consequently it is puzzled 
 by much that philosophers and physicists say about space. It is 
 offended by I'latland, and amazed by Einstein, but the truth 
 is that the common-sense belief in a single spatial world in , 
 three dimensions is too little developed to be able to discuss '
 
 94 THE WORLD OF COMMON BELIEF [ch. 
 
 such points. It is not in a position to argue whether space is 
 absolute or relative, whether ' the space of the real world is a 
 space of six-dimensions', whether the older physics or the 
 Quantum theory is in the rights and so forth. The world is in 
 one space for common sense because there is a general spatial ■ 
 order at least within the stellar universe, and it has three di- 
 mensions because one man can go east and another north and 
 another go up in the air. These meanings are found in per-., 
 ception, and common sense carries this perceptual meaning very 
 much further than perception can. If its orbit is too narrow 
 for science it is wide enough for most ordinary purposes. 
 
 The perceived meaning of extended things, then, cries out 
 for and receives the interpretation of judgment, and the 
 beliefs so arising commit us to the belief in a general spatial '" 
 order. We are constrained to interpret things in terms of 
 spatial continuity. The margin of perception, to mention no 
 other circumstance, forces us to conclude that any perceived 
 thing is of a piece with its surroundings, and that these sur- 
 roundings do not come into being when we attend to them 
 closely, although they are perceived but dimly on the margin 
 and beyond the margin are not perceived at all. The spaces [ 
 we perceive are filled spaces, that is to say, there is continuity ] 
 within them ; and when our attention passes from one thing to 
 another the interval is also filled space. We interpret these ^'^ 
 indications of perception when we believe that the world as a 
 whole is spatially continuous in the same sense as any per- 
 ceptible portion of it, and our spatial explorations in the way 
 of perceiving are subject to this interpretation. 
 
 Similar arguments hold of time. The world of common 
 belief is a world of continuants which are either simultaneous 
 or successive in the general order of time. The perception of 
 transience, with its order of earlier and later, is the basis and the 
 empirical warrant for this belief, and the common order of 
 time is the correlation of all perceived simultaneities and suc- 
 cessions. Common sense, it is true, has not reflected very 
 
 ^ Eussell's Mysticism and Logic, p. 138. 
 
 2 Cf. J. W. Nicholson's paper in Prohlenns of Science and Philosophy, Aristo- 
 telian Society, Supplementary Vol. ii, 1919.
 
 y] THE WORLD OF COMMON BELIEF 95 
 
 deeply upon time, or upon its continuity and its common 
 order. It does not cross-examine the Man with the Scythe 
 very sharply, and therefore it is not prepared to say whether 
 the structure of time may not be corpuscular^ whether the 
 measurement of time does not always move in a circle, whether 
 the time of the stellar universe may not be only a sort of 
 local time, whether time and space are not indivisibly united in 
 the fact of motion, and so on. It does assume, however, that 
 every perceptible event is either simultaneous or successive as 
 compared with every other perceptible event, that the inter- 
 mittence of perception does not annul the temporal continuity 
 of any continuant, and that any continuant has just one 
 history. These assumptions, taken together, sufficiently define 
 the beliefs of common sense, and the last of them sums up the 
 position. If we want to know whether Kaspar Hauser was 
 really a wild man of the woods we have to trace his biography, 
 and if that can be done the problem is solved, just because he 
 had only one biography. A ship has only one history, and 
 therefore we should be likely to discover what argosy was 
 wrecked at Tobermory if we could trace the history of every 
 other ship in the Armada. 
 
 Our belief in the world, then, is an interpretation of the 
 things and events we have perceived or remember. It is 
 reflection following out and giving full weight to perceived 
 meanings. The world, it is true, is richer than our judgments, 
 because we perceive so little and reflect so ill ; and it is far 
 richer in qualities than its logical structure and the order of 
 space and time imply of themselves. For these are only the 
 skeleton of a body which has flesh and blood and is clothed 
 besides. Common sense, however, uses at least one other 
 general principle of interpretation, since it interprets things 
 causally, and we must therefore examine this principle. 
 Under causal interpretation I mean to include causal correla- 
 tions like the growth of stamens and pistils in a poppy, causal 
 laws like the law of gravitation, and causes and effects like the 
 stroke of the bat that sends the ball to the boundary. 
 
 As all the world knows, Hume made the difficulties of 
 ^ Eussell's Mysticism and Logic, p. 129.
 
 96 THE WORLD OF COMMON BELIEF [ch. 
 
 causal interpretation so very clear that no philosopher coming- 
 after him has any excuse for neglecting theni\ The principal 
 counts in Hume's indictment are that causation is never per- 
 ceived, and that it cannot be inferred from the perceived 
 phenomena without an immense fallacy. Although we say 
 that bread nourishes, we never observe any mysterious tie 
 between the bread and the formation of tissue. We can only 
 observe a uniform sequence between the swallowing and 
 digesting of the bread and the subsequent formation of tissue. 
 This uniform sequence maybe uncontradicted in our experience. 
 Bread always nourishes if we assume that when it does not 
 nourish it is either not bread or rejected by the stomach. But if 
 the causal action of the bread cannot be observed in any single 
 instance, it plainly cannot be observed in a series of instances, 
 and the problem, therefore, is how the number of instances 
 (granting that there are no known exceptions) can guarantee 
 the causal interpretation when no single instance does so. 
 
 According to Hume, this inference from repetition of in- 
 stances to their necessary connection is quite unjustifiable. 
 The repetition neither discovers nor produces anything new 
 in the phenomena, and yet we infer that bread must always 
 nourish and tire always consume, because we have repeated 
 experience that they have done so in the past. This inference 
 can scarcely pretend to be valid. If we say that bread will 
 nourish in the future because it mtist nourish, we have begged 
 the question; and if we say that fire will continue to consume 
 in the future because it has always consumed in the past, our 
 inference manifestly outruns the evidence. 
 
 Those philosophers who had wit enough to see the force of 
 Hume's difficulties have usually attempted to answer him in 
 one or other of two ways. The first answer is that causality 
 is an a priori law somehow involved in the possibility of things. 
 Hume answered this contention when he pointed out that 
 there is no contradiction in denying causal laws. There is no 
 logical contradiction in supposing that bread may cease to 
 nourish without any reason whatever; and that seems final, 
 
 1 Treatise, bk i. pt iii., and Enquiry concerning the Human Under- 
 standing, sect. vii.
 
 v] THE WORLD OF COMMON BELIEF 97 
 
 even granting that Kant gainsaid it. According to Kant 
 there is some sort of intrinsic connection between causation 
 and the hypothetical judgment in the logical category of rela- 
 tion i. This contention, however, is only a tour deforce inspired 
 by Kant's belief that every 'principle of the understanding' 
 must have a twin brother in the logical table of judgments. 
 
 The other attempted answer to Hume maintains that we 
 can directly perceive certain causal connections in our own 
 persons. We feel that we can enforce, and also that we can be 
 compelled against our wills. This reply, however, does not 
 wring Hume's withers. These feelings of compulsion and 
 spontaneous enforcement are highly capricious and irregular, 
 and they cannot justify any inference to that necessity for 
 uniform behaviour which Hume took to be the meaning of 
 causality. Hume might well have thought this argument 
 sufficient in itself, but he supplied many other arguments in 
 case any one should want them, like the nine and twenty 
 excellent reasons which the Mayor of Coventry gave for 
 refusing to ring the bells in honour of Queen Elizabeth 
 after he had stated ^Imprimis we have no bells' by way of 
 preface. 
 
 These arguments are set forth in his Enquiry, §§ 52 and 53, 
 where he asks whether we know the secret union between 
 mind and body, whether we know why the will controls the 
 fingers and does not control the liver, whether we can really 
 discern the connection between volition and the nerves, muscles 
 and tendons of the body, whether we know the precise manner 
 in which the soul creates images in the fancy, whether we can 
 tell bow the mind commands itself at some times and fails to 
 do so at other times, or how men have greater self-control in 
 health than they have in sickness. Only experience, Hume 
 argues, can tell us what we can do and what we cannot do, 
 and this experience is never the perception of necessity. That 
 is surely clear, and we might add (if it were permissible to 
 labour the point further than Hume himself) that even if, per 
 impossibile, we had this direct acquaintance with necessary 
 connection in our own persons, we should have very little reason 
 
 ^ Critique of Pure Reason, Analytic of Conceptions, Sections ii. and in. 
 L. 7
 
 98 THE WORLD OF COMMON BELIEF [ch. 
 
 for extending it by analogy to the inanimate world. Do we 
 seriously suppose that we are acquainted with the dynamics of 
 impact just because we can belabour a punching-ball? Because 
 we sympathise with Tantalus, have we any right to maintain 
 that the tides are due to the moon's unavailing thirst for the 
 sea? 
 
 Hume did not deny, of course, that we believe in causes. 
 He was too indolent, he said, not to believe that fire burns, 
 although he had proved that this belief had no rational 
 grounds^ He peopled his world by judgment (including 
 causal interpretation) like other folk. The problem is there- 
 fore whether this peopling of the world has any grounds which 
 Hume overlooked ; and this question is clearly of the first im- 
 portance. Causal interpretation, indeed, is nine parts of the 
 world of common belief We read a letter and infer that it 
 came over the seas, we see the sun and infer that rays from it 
 must reach us through some medium. The postal system and 
 the solar system are believed things based on the causal in- 
 terpretation of a few vestiges of perceived fact. Causation, in 
 a word, is our clue to the continuous filling of space and time. 
 
 Hume"'s analysis was defective because he overlooked the 
 perceived meaning of perceived things. He admitted that we 
 perceive succession directly, as well as figure, colour and sound; 
 but he maintained that all perceived things are loose and 
 separate, and that we always perceive bare conj unction without 
 any hint of connection. Both these assumptions are false in 
 fact. We perceive things within a context, and all perceived 
 things have a meaning when they are apprehended. Wliat 
 is more, this meaning is not mere conjunction (or bare to- 
 getherness in space and disconnected succession in time). The 
 context which signified in perception is not a mere skeleton of 
 conjunction, because we never perceive empty space and empty 
 time; and the meaning directly perceived in the filling of space 
 and time has the seeds of causality in it. 
 
 1 Treatise, bk i. pt rv. sect, vii.: "My natural propensity reduces me to 
 this indolent belief in the general maxims of the world." " If we believe that 
 fire warms, or water refreshes, 'tis only because it costs us too much pains 
 to think otherwise."
 
 v] THE WORLD OF COMMON BELIEF 99 
 
 Causal interpretation implies the irrelevance of mere space 
 and mere time. Time and space, to be sure, are not irrelevant 
 in one sense. A plant takes time to grow, and needs room to 
 grow in, but the causal interpretation of its growth depends 
 upon the assumption that the rate of its growth and the 
 stature which it attains depend upon the kind of thing which 
 it is, the degree in which it can keep itself alive and utilise 
 surrounding things, and so forth. Things are in space and in 
 time, and therefore their actions and reactions are spatial and 
 temporal too; but empty space and time are subordinate in 
 all causal determinations because the determining factor in the 
 behaviour of things is taken to be the properties of the things 
 themselves. Neither position by itself, nor succession by itself, 
 nor these twain together can explain the causal connectedness 
 of anything; and Hume's argument shows very clearly why it 
 must be so. His argument also shows that causal connected- 
 ness means nothing without perception and experience. 
 
 In point of fact, however, perception does contain a causal 
 meaning, and so does the experience of voluntary movement. 
 This perceived causal meaning, to be sure, is only a perceived 
 meaning, unverbalised and very easy to interpret falsely when 
 we reflect upon it, but it contains a presumption which is the 
 nucleus of a principle. To state it broadly, this presumption 
 is that anything which occupies a place thereby keeps other 
 things out and makes a difference to other surrounding things, 
 and that any continuant changes or remains identical because 
 of the kind of thing it is and because it is set in a certain 
 environment. In particular cases we can perceive things play- 
 ing their part, making a difference to other things or clinging 
 temporarily to their individual being ; and this fact of percep- 
 tion is the ultimate basis of causal interpretation. 
 
 This broad presumption, it is true, does not of itself justify 
 the transition to unvarying causal laws. From the principle 
 that everything makes some kind of difference to other things, 
 or that every happening has some kind of explanation in its 
 surroundings, it is (juite impossible to infer that all causation 
 is regular and uniform. Simplv because a thing plays its part, 
 it does not follow that the same thing always plays the same 
 
 7—2
 
 100 THE WORLD OF COMMON BELIEF [ch. 
 
 part in the same surroundings, still less that the same kind of 
 thing always plays the same part. We never perceive necessary 
 connection, although in some cases, after long experience, we 
 come to expect some res'ults with very great confidence indeed. 
 On the other hand, if we grant the presumption (even with- 
 out the certainty) that everything which exists plays a part 
 in the world, it is reasonable for us, and even necessary, to 
 consider what part it plays; and if the result of this enquiry is 
 to show, on the whole, that the same kind of thing always 
 seems to play a uniform and regular part, there is no coercive 
 reason for rejecting this specific interpretation of the causal 
 principle. 
 
 When we are scientifically minded we assume that all 
 physical events are regularly determined in this way, and 
 probably that psychological events are causally determined 
 too. Indeed we argue, not only that every physical event is 
 a member of some uniform causal series, but that it is highly 
 probable that we can discover by our methods of elimination 
 ivhat causal series it belongs to; and we explain apparent ex- 
 ceptions by supposing that there always is a cause for them 
 even although we have not had the luck to discover it. 
 Common sense, however (unless it is content to echo the 
 trumpets of science), is not nearly so positive in its experi- 
 mental determinism. It is quite ready to admit that some 
 things are as capricious and irregular in their behaviour as a 
 woman's wit. Indeed, common sense often supposes that the 
 human will is a citadel of irregularity, though never that it is 
 disconnected with other things or that it is not even an 
 irregular cause. Still, unless changes are regular there is no 
 use in expecting, and common sense has its expectations ful- 
 filled often enough to continue expecting with a good heart. 
 The belief in these detailed expectations is just the world of 
 common sense, and therefore it is very important to consider 
 whether these expectations can or cannot be justified. 
 
 Necessary connection is not a perceived meaning, and neither 
 are the causal laws of science. Galileo had to drop the weights 
 from the tower at Pisa in order to prove that the light one 
 and the heavy one took the same time to fall. The tested
 
 v] THE WORLD OF COMMON BELIEF 101 
 
 uniformities of science do not lie on the surface and do not 
 leap to the eyes like colour or shape. Per contra^ some of the 
 acquired meanings of adult perception are of the same type 
 as scientific generalisation, and even the prelogical generality 
 which M. Levy-Bruhl attributes to savages is very often an 
 instance of the same kind^ The savage belief that foreigners 
 bring pestilence or bad luck is an obvious application of the 
 Method of Difference, though it may be very hard upon some 
 of the visitors. Savages believe that fire burns and that 
 water quenches it for the same logical reasons that constrain 
 civilised thinkers to believe that prices must rise when goods 
 are scarce or that dominants and recessives will mix according 
 to the proportions of MendeFs formulae. These common- 
 sensical expectations, it is true, are less carefully sifted than 
 the scientific ones, for common sense does not show any in- 
 temperate zeal in searching for negative instances. But the 
 principle is the same, although the applications of it may be 
 hastier than they ought to be. 
 
 Our contention is that perception always has a general 
 precausal meaning, that it often has a specific precausal 
 meaning, and that this precausal meaning is of the same type, 
 logically speaking, as any causal generalisation in science. 
 Hume was wrong, therefore, in his analysis of the observed 
 facts; and he may also have been wrong in his inference that 
 causal laws are only irrational habits of expecting. I suggest 
 that he was wrong in maintaining that repeated experience 
 discovers nothing new in the phenomena. We all know that 
 familiar things are not the same to our perception as they were 
 when we perceived them first, and this truism also holds of 
 causal meanings. Adam may not have known what fire would 
 do, but his children have a very shrewd idea. Fire has 
 acquired a meaning for them, and it has acquired a specific 
 causal meaning. Why therefore should we deny that the 
 repetition has discovered something new ? It does not discover 
 
 ^ Les Fonctions nientales dans les Socu'U's infirieures, passim. M. L^vy- 
 Bruhl believes that this prelogical mind of the savage is different in kind 
 from the reasoning mind of civilised people, but his grounds for drawing this 
 distinction are extremely flimsy.
 
 102 THE WORLD OF COMMON BELIEF [ch. 
 
 a 'mysterious tie,' to be sure. Causes are not magic filaments, 
 and anyone who looks for such filaments may expect to be 
 disappointed. He need not expect to find more than causal 
 connection since there is nothing more to find. 
 
 Indeed, in the last analysis, we seem to have the same kind 
 of evidence for believing in the causal properties of things as 
 for believing in any of their attributes. If we perceive that 
 this water is limpid, may we not also perceive that this water 
 cleanses.? And if we cannot perceive that fire always burns, 
 we have surely no right to say that all fire is ruddy just 
 because some particular flame is. This meaning also has to be 
 acquired. The ultimate difference in the case is that properties 
 belong to things while causal connection is a relation between 
 them. Related things are perceived, however, and their rela- 
 tions are causal as well as spatial and temporal. If perception 
 is evidence that a flame is ruddy it is also evidence that fire 
 burns. 
 
 Such, then, is the world we believe in. It has a logical 
 structure, and a general order of space and time. There is 
 always some connection in it, and frequently there is regular 
 connection. That is all we discover at the plane of common 
 experience, and there is no need to discredit this discovery so 
 far as it goes, although there is utter need and unlimited 
 opportunity for going further in the way of reflection and 
 experiment. Some philosophers argue, it is true, that our ideas 
 must really be more penetrating than common sense ones if 
 the common sense point of view is itself possible. We could 
 not have any idea of causality, they say, even a halting idea 
 and irresolute, unless the law of universal causation were true 
 to its marrow. That seems an overstatement. We could not 
 have the idea of regular causes, I suppose, unless there were 
 some approximately regular connections in past experience, 
 but if vermilion is usually heavy and politicians usually de- 
 ceitful that in itself would be a sufficient psychological basis 
 for supposing necessary heaviness or necessary deceit: and if 
 this approximate regularity were all the regularity that exists, 
 our ideas, in all probability, would be very much what they
 
 v] THE WORLD OF COMMON BELIEF 103 
 
 Perhaps I should append some remarks on error. Those who 
 maintain that we judge objectives and not facts, usually say 
 that there is no room for error except upon this assumption^; 
 and they point out, very justly, that error is the familiar spirit 
 of true belief. Truth is not a mint with an image and super- 
 scription that falsity lacks. It has not even a cachet which 
 only the learned can descry. On the contrary, false propositions 
 are believed as well as true ones; and they are the same, to our 
 inspection, as the true ones. Most of us believe that James IV 
 perished at Flodden, but many believed in the legend that he 
 died of old age in a monastery, just as many believed in similar 
 legends concerning the death of Barbarossa, or Gordon, or 
 Kitchener. If judgment characterises the real James IV^, how 
 can it characterise him falsely.'^ Yet there are certainly false 
 objectives, and propositions, in themselves, do not guarantee 
 their own truth or falsity. We have to suppose, therefore 
 (according to this argument), that objectives are judged and 
 not fact itself. If fact itself were judged there would be false 
 facts; and that is nonsense. On the other hand there are false 
 objectives, and it is reasonable to hold that true objectives 
 correspond to fact while false ones do not. 
 
 The trouble about explanations of error is always in the 
 beginning. Once it is admitted that true judgments are in- 
 distinguishable from false ones, there is just as much difficulty 
 in explaining why false objectives should appear to correspond 
 to reality as in explaining how reality can appear falsely. 
 Either we know this correspondence directly in some cases or 
 we do not. If we never know it, truth is mere supposal. If 
 we know it, we must know both terras of the correspondence. 
 We must compare the fact with the objective, and this com- 
 parison is itself a judgment. If facts cannot be judged, 
 therefore, the relation between facts and objectives cannot be 
 judged; and, in that case, what becomes of the theory? 
 
 It is impossible to explain error. We must simply accept 
 this eternal possibility and try to be as careful and consistent 
 
 ^ Cf. Kussell, 'On the Nature of Truth and Falsehood,' in his Philosophical 
 Essays,
 
 104 THE WORLD OF COMMON BELIEF [cii. v 
 
 as we can. Apprehension always has the seed of misappre- 
 hension in it, and judging the chance of misjudging. 'Ex- 
 planations' of eiTor only stave it off for a little. We may, if 
 we choose, try to explain the mistake about James IV by 
 saying that all the elements in this false judgment may be true 
 in some other judgment. James IV did not die in a monastery, 
 but Charles V did. It is true, then, that there are monasteries, 
 that some monarchs have died in them, and that James was 
 a monarch. But James did not die another man's death, and 
 his death in a monastery is not less of a sham because Charles 
 really died in one. Our only comfort is that the eternal risk 
 of error is not an everlasting presumption in its favour.
 
 CHAPTER VI 
 
 PRINCIPLES 
 
 Let somebody now demonstrate this Triangle described in the Matter 
 to have its three angles equal to two right ones; Why yes, saith the 
 Soul, this is true, and not only in this particular Triangle but in all 
 plane Triangles that can possibly be described in the Mutter. And thus, 
 you see, the Soul sings out the whole Song upon the first hint, as 
 knowing it very well before. 
 
 Henry More, An Antidote against Atheism. 
 
 Philosophers may be divided into two classes, the class for 
 whom facts are just facts and the class for whom facts are only 
 suggestions. Those who belonged to the first class used to be 
 called empiricists, and the second class, in its extreme form, 
 includes Hugo of St Victor, for whom physical things were 
 literally nothing but symbols of the Christian revelation, and 
 Professor Macran, who says that "the first step in philosophy, 
 though by no means the whole of philosophy, is idealism or 
 the denial of the fact\" Now realists, I suppose, are as em- 
 pirical as they dare, but however empirical one may be, it 
 stands to demonstration that the most empirical thinking is 
 logically bound to accept general truths as well as particular 
 truths of fact. The proof is simple and Mr Russell has put it 
 so simply that the best thing I can do is just to quote him: 
 
 Of course it is clear that we have general propositions Wq have 
 
 such propositions as 'All men are mortal' and 'Some men are Greeks.' 
 
 But you have not only such propositions; you have also such ./rtc^* You 
 
 cannot ever arrive at a general fact by inference from particular facts, 
 however numerous. The old plan of complete induction, wliich used to 
 occur in books, which was always supposed to be quite safe and easy as 
 opposed to ordinary induction, that plan of complete induction, unless 
 it is accompanied by at least one general proposition, will not yield you 
 the result tliat you want. Suppose for example that you wish to prove 
 in that way that ' All men are mortal,' you are supposed to proceed by 
 complete induction, and say 'A is a man that is mortal,' 'B is a man 
 
 1 Hegel's Doctrine of Formal Logic, Introduction, p. 13.
 
 106 PRINCIPLES [CH. 
 
 that is mortal,' ' C is a man that is mortal,' and so on until you finish. 
 You will not be able in that way, to arrive at the proposition 'AH men 
 are mortal' unless you know when you have finished. That is to say 
 that, in order to arrive by this road at the general proposition 'All men 
 are mortal,' you must already have the general proposition 'All men 
 are among those I have enumerated.' You can never arrive at a general 
 proposition by inference from particular propositions alone. You will 
 always have to have at least one general proposition in your premises^. 
 
 AVe must accept principles, therefore, even in an empirical 
 inventory of fact; and there is nothing unusual in this pro- 
 cedure; for the world of common belief, as we have seen, is 
 just an interpretation of perceived or remembered fact in the 
 light of certain principles. Common sense, however, uses 
 general principles without reflecting very deeply upon them. 
 It is a world of axiomata media, not of 'Ap-^ai; and therefore, 
 as a Hegelian would say, the reach of common sense is greater 
 than its grasp. The destiny and the ineluctable privilege of 
 philosophy, on the other hand, is to pursue these principles 
 so far as thinking can, to reach categories (or first principles), 
 and to be an untiring critic of these categories. 
 
 I shall not attempt, of course, to deploy the categories in 
 full array. No one but a fool or a demi-god would try to do 
 so in a few pages. The time has passed since courtiers wanted 
 thumb-nail sketches of ultimate metaphysics, since sovereigns 
 requested the quintessence of truth before breakfast, or since 
 Mme de Stael asked Fichte for a complete revelation of the 
 Ich and the Anstoss in five minutes. Instead of treating 
 metaphysics in this princely style, I shall be content if I can 
 answer a few questions about principles, and especially if I 
 can sketch in outline what kind of being a principle or a 
 category has. 
 
 To say that a principle holds of certain facts is to say that 
 these facts are instances of the principle; and an instance is 
 something which can be logically derived from its principle. 
 Such principles, plainly, may themselves be instances of more 
 fundamental principles or they may not. If they are, they are 
 axiomata media in Bacon's language. They are dependent 
 
 1 The 3Ionist, vol. xxix. No. 2, pp. 198-199.
 
 vi] PRINCIPLES 107 
 
 principles, intermediate in a possible hierarchy. If they are 
 not, they are logically primitive; that is to say, they are ^Kpyai^ 
 or first principles, or categories. 
 
 There is always considerable difficulty in knowing for certain 
 whether a principle which seems to be primitive may not really 
 be derivative. On the other hand, it is usually comparatively 
 easy to have an inkling, at least, of the whereabouts of a first 
 principle. Indeed, we know this whenever we find that we 
 always need to make a certain kind of assumption in order to 
 avoid a fallacy, however ingeniously we may twist and turn 
 our arguments: for then it is certain that there really is some 
 dogged and elusive principle which is always present although 
 it is sometimes hidden and often disguised. The notion of 
 value, for example, is logically primitive in any system of 
 ethics. If you say that one thing is better than another be- 
 cause it is more highly developed, you must first assume that 
 development is necessarily improvement, i.e. that there is always 
 greater value in a thing in proportion to the degree of its 
 development. If, like St Thomas, you believe that a thing is 
 good in the measure in which it is 'natural,' you require 
 St Thomas's premiss: Qiiod omne agens agit proj)ter bonum^. 
 If you argue that the good is what ought to be desired, you 
 need a premiss to the effect that everything that is better than 
 some other thing ought therefore to be desired before it, and 
 conversely. Value or goodness in a word must always be 
 included in the ultimate premises of ethics. 
 
 Granting, then, that there are principles and first principles, 
 we may proceed to consider certain questionsabout them, and we 
 may begin by considering how they are discovered in experience. 
 One of the favourite arguments of historical empiricism is that 
 experience begins without principles and that nothing can be 
 apprehended intellectually unless it has previously been sensed. 
 Now it is true that the reflective recognition of principles 
 comes late in psychological development, if, indeed, it ever 
 
 1 Summa Philosophica, liber iii. caput iii. St Thomas tries to derive this 
 proposition from in. ii., Qitod omne agens agit propter finem, on the ground 
 that ihe finis of anything is conveniens ei. But then he quietly begs the point 
 at issue by saying 'Quod autem conveniens est alicui est illi bonum' (iii. ibid.).
 
 108 PRINCIPLES [cH. 
 
 comes; but principle and instance are correlative terms, and 
 the use of principles comes very early indeed. We begin, in 
 fact, with inarticulate principles and come to find out what 
 they are, and, in the same sense, we begin with inarticulate 
 instances and come to find out what they are. Prelogical 
 generality has a double aspect affecting principle and instance 
 alike, and it is not difficult to see that prelogical generality 
 comes very early indeed. The child who calls all men 'daddy' 
 classifies all men together, and at the same time does not fully 
 appreciate the difference between his father and all other men. 
 The savage who tries to express all his thoughts with a 
 vocabulary of some four hundred words, generalises when he 
 speaks, but the reason is that he ignores important differences 
 in things or, at least, cannot express them. Those who speak 
 indifferently of trees when they mean to refer to beeches, oaks 
 and poplars, may never have noticed the differences carefully, 
 and in that case they are not more reflective than botanists but 
 less so, because they have not examined the facts. MilFs village 
 matron who supposed that the medicine which had cured 
 Lucy of whooping-cough must also cure Mary's chilblains', 
 generalised; but she generalised very badly, because she had 
 not examined the instances closely enough. The process of 
 psychological development is the passage from prelogical 
 generality to the logic of principles, and not from an explicit 
 knowledge of particular facts to certain mysterious 'high 
 priori "■ categories. Principles are discovered pari passu with 
 their instances. 
 
 When this is granted we are plunged at once into a far 
 more perplexing and important set of problems. What are 
 these general facts.'' What kind of being have principles or 
 categories ? 
 
 It seems to be evident that general facts do not exist; for 
 whatever exists is particular, and principles are universal. 
 When I learn on the authority of Euclid, for example, that 
 'Of all rectangles which have the same perimeter, the square 
 has the greatest area,' my information may indeed apply to 
 this rectangle and to this square, but the general fact itself 
 1 System of Logic, bk ii. chap. iii. § 3.
 
 vi] PRINCIPLES 109 
 
 concerns the rectangle and the square. It is applicable to par- 
 ticular existing things but it does not concern them directly, 
 because the existence of instances in reriim natura is no part of 
 the proof or of the being of the principle. Universals are the 
 stuff of general facts and universals do not exist. Red things 
 exist, but not redness. 
 
 In view of these reasons we are commonly told that universals 
 do not exist but merely subsist ; that particular things like 
 tables or churches or trades unions exist ; and that both have 
 being. Being, therefore, is a general term which describes both 
 existence and subsistence, and it describes the common relation 
 which existing things and subsisting ones have to the mind. 
 Being is always determinately so-and-so and confronts the 
 mind in this determinate character ; it always constrains the 
 mind ; it is always objective in the sense that this or that is 
 true of it without appeal. In respect of knowledge, therefore, 
 the status of existing things is very much the same as the 
 status of things which merely subsist; indeed the only im- 
 portant difference is that the former may be perceived or felt 
 and that the latter cannot. There is the same kind of constraint 
 and the same kind of confronting when I judge that sugar is 
 sweet and when I judge that a'" x a" = «'"+". The general 
 proposition and the existential one are discovered by the mind 
 in the same sense, both propositions are true independently of 
 our thinking, both are equally binding, equally indisputable, 
 and equally incorrigible. 
 
 This contention seems very just, but it plainly gives rise to 
 a great many problems. Even if a dualism between existence 
 and subsistence is the last word in this important matter, it 
 is clear at least that the relation between these two divisions 
 of being needs to be considered very closely indeed ; for some 
 general facts hold of all existent fact. To take the most 
 obvious point, existing things would not be thinkable at all 
 unless logical principles at least were applicable to them. 
 Existing things are what they are and are not other things, 
 and the laws of identity and contradiction are the most general 
 of all general facts. Let us consider, therefore, whether it is 
 possible to avoid this dualism.
 
 110 PRINCIPLES [CH. 
 
 In the first place, an attempt may be made to dispense 
 with universals altogether. The appearance of universality, of 
 course, is beyond question, and we have already seen that 
 there are general facts. None the less, attempts are frequently 
 made to show that universals have no independent status, and 
 that they are only parts of the world of existence. Most of the 
 materialisms which delight tough-minded gentry are committed 
 to this view, and most of the evolutionisms which attract 
 ' advanced thinkers.'' This problem therefore may even have 
 some general interest despite the austerity of its subject. 
 
 Those who try to reduce universals to a variety of existent 
 fact have to choose one or other of three roads. They may 
 hold either that universals are only attenuated existents, 
 or that thev are functions of certain privileged existents, or 
 that they are a kind of organisation of existence. Let us con- 
 sider these views in turn. 
 
 The first view is the most usual. Universals, we are told, are 
 mere abstractions and they are nothing else. In this character 
 they are treated with contempt, or pity, or sorrow, according to 
 the critic's mood. Your full-blooded critic is always scornful. 
 He prefers a man to his wraith, lusty things to anaemic ones, 
 Tom Jones to poor Tom's a-cold. Other critics are more 
 merciful. These poor abstractions, they tell us, are as true as 
 abstractions can be, and they are often very useful indeed; for 
 although we always want a full draught of reality we have often 
 to put up with the small beer of abstraction for our stomachs' 
 sake. These full draughts of reality take too long to digest. 
 
 There may be some comfort, therefore, in finding that 
 universals are not abstractions at all, since the Abstraction 
 Theory is utterly unable to account for them. According to 
 this theory, we are supposed to pare away the distinctive 
 peculiarities of things, and so to be left with their common 
 elements. This elimination, however, to use Hegel's metaphor, 
 is only peeling the coats off an onion ; and such a process can 
 never generate a universal from a particular. The effect of it 
 is only to get less onion and in the end no onion at all. 
 Elimination takes something away, and leaves the rest. There 
 is nothing else for it to do. And even if elimination in this
 
 VI] PRINCIPLES 111 
 
 sense sometimes seems to be possible, and sometimes seems to 
 leave a common residue, it is abundantly evident that many 
 universals have not the faintest semblance of being such a 
 residue. It might be possible to arrive at redness by this pro- 
 cess after eliminating the distinctive shades of red, but it 
 would be interesting to know what colour is when the redness 
 of the reds and the greenness of the greens have been abstracted 
 from it. The residue might be a neutral grey, but neither 
 redness nor greenness is a kind of grey. Again, the mutilated 
 figure of a triangle which is neither right-angled, acute-angled 
 or obtuse-angled, is plainly not triangular, and one would fain 
 know what man is in the abstract when he is neither dusky 
 nor fair, neither short nor tall, neither male nor female nor 
 hermaphrodite, or how abstract man, thus denuded, could also 
 contain all the males and females and hermaphrodites that 
 ever were or will be. This theory begins by stripping a man of 
 all that he possesses and then requires him to support the race. 
 
 There is no road this way, therefore, and so we must try 
 the next. Universality, according to this second theory, is a 
 function of certain particulars, i.e. of words and images and 
 other signs. These words and images are particular in them- 
 selves. A word is just a noise and as particular as the boom 
 of ordnance or the whistling of reeds. Its universality lies in 
 its function, and this function is twofold. The particular 
 things which have this function are labour-saving substitutes, 
 and they lead the mind to the same terminus as the particulars 
 for which they are substitutes. It saves me a great deal of 
 trouble to know that a dog's dentition is so-and-so, for then 
 I do not need to examine Tray's mouth or Fido's every time. 
 And if I do not bring my eggs to market, the substitute signs 
 in a letter about them may still find a purchaser. 
 
 Now it is clear that words and other signs certainly have 
 this function, since they may be used as substitutes for things ; 
 and they save a great deal of trouble. On the other hand, it 
 is equally clear that these signs save trouble precisely because 
 they signify universals and because these universals therefore 
 apply to their particular instances. Why is the sound 'red- 
 ness' a substitute for the detailed examination of red things.'*
 
 112 PRINCIPLES [CH. 
 
 The sound, surely, is only the symbol for the quality and there- 
 fore not identical with it; and if we need not examine red 
 things when we are sure of their redness the reason is simply 
 that we know in advance that 'redness' must hold of all red 
 things. If Stevenson's style resembles Sterne's, and if Gibbon's 
 style does not resemble Kipling's, there really is resemblance 
 or difference in these cases, and consequently the universals 
 'resemblance' and 'difference' apply directly. These state- 
 ments use words, but they also express general facts because 
 the words themselves have a general meaning. It is possible, 
 to be sure, to attend to the signs only, and to practise the 
 manipulation of symbols for its own sake. We do this when 
 we work out examples in school arithmetic or in the differential 
 calculus, for then we concentrate our attention upon certain 
 counters and manipulate them according to the rules. Even 
 in that case, however, the counters have a meaning if we 
 choose to consider it, and when words are used to express 
 general facts this meaning, usually, is clearly before the mind. 
 It is simply absurd to say that words are identical with general 
 facts. Some words express particular facts, and other words 
 express general facts, and the difference, in both cases, can be 
 seen in the facts themselves. 
 
 We may pass, then, to the third view according to which 
 general facts are only the way in which existence organises 
 itself. This contention may take very different forms, but its 
 primary significance is clearly in terms of the mind, and this 
 primary significance is far more important than any other. 
 For it is clear that we organise our experience, and that, in 
 a certain sense, we organise our world. Some enthusiastic 
 realists, it is true, tell us that the laws of gasoline engines 
 were just the same in the days of the ancient Athenians as 
 they are no\v\ These laws, however, were certainly not con- 
 tained in the organised experience of the ancient Athenians, 
 and, in a way, had nothing whatever to do with the ancient 
 world. 
 
 1 Montague, Studies in the History of Ideas, p. 236. Quoted by J. E. 
 Kantor, ♦ Instrumental Transformism and the Unrealities of Realism,' Journ. 
 of Philosophy , etc., vol. xvi. No. 17, pp. 452-453.
 
 vi] PRINCIPLES 113 
 
 Indeed, if things were our thoughts about them, and if any 
 man's world were just the group of ideas which he has come 
 to learn and recognise, we might say very truly that what our 
 logic (or our causal interpretation) does for us is simply to put 
 our worlds into order. Principles undoubtedly regulate our 
 thoughts, and, through our thoughts, our actions. A logical 
 mind is one which organises its experiences in a logical way; 
 a scientific mind arranges its observations in a quantitative, 
 causal way; and so on. These statements, however, are surely 
 incomplete, just because the problem is taken too narrowly 
 when it is restricted to the mind's arrangement of its ideas. 
 It is true that I organise my ideas when I think, but it is not 
 true that this organisation is the whole of the matter. When, 
 for example, I organise my ideas about the physical world, 
 my endeavour is to discover the characteristics of the physical 
 world itself. I have to arrange my ideas consecutively in order 
 to appreciate the logic in the facts, but this arrangement is 
 not itself the discovery to which it is a means. Logical think- 
 ing, indeed, is a habit of following the logical structure of 
 things. This structure controls and determines the habit, and 
 the habit, once formed, has a certain momentum in it. But 
 that is a different matter. 
 
 Even the most uncompromising idealist (and his pragmatic 
 brother) would admit the justice of this criticism, if nothing 
 could be taken into account except the individual mind, on 
 the one hand, and the world itself on the other. This theory 
 would entrap me within myself for ever if my logic were only 
 my private habit of organising; and no one wants to be a 
 solipsist. By general admission, therefore, the theory nmst be 
 stated more broadly. According to the Absolutists, the Ex- 
 perience which is organised is not really anyone's private 
 experience, for private experience, as we call it, is only a part 
 (and, in some ways, a spurious part) of Absolute Experience; 
 and Cosmic Experience includes all private worlds as well as 
 all finite selves. There is no opposition, therefore, between my 
 ideas and the world ; and even my private habits, as I call them, 
 have a new significance when I realise what they truly are. 
 These private habits are tiny wavelets in the current of the
 
 114 PRINCIPLES [cu. 
 
 universe, and this current itself is the sweep of principles 
 which, like the Ideas of Plato, are always supposed to be 
 dynamic and productive, although they act, of course, in the 
 majestic timeless fashion of Absolute Reality itself. The 
 pragmatists, to be sure, cannot go all the way with the 
 absolutists, but they have gone to school with them, and most 
 of them, by substituting the life-process, or the Zeitgeist, or 
 the intelligence of a great people, for the Absolute contrive to 
 retain some of the momentum of the Platonic Ideas and yet 
 to dress the world in workmen's overalls, or to credit it with 
 the overwhelming vitality of a gendering bull. Both sets of 
 theories, however, presuppose that truth is a kind of con- 
 struction and that thinking somehow produces the world. 
 When this assumption is granted, principles may very well be 
 the self-organisation of thinking and nothing else whatever; 
 but the falsity of the assumption is apparent of itself, and 
 that is a bad augury for the conclusion of the argument. This 
 theory, indeed, is in conflict with itself. It maintains that all 
 principles and all thinking is just a kind of constructiveness, 
 and yet professes to describe the facts of Experience (with 
 a capital E). This description, however, is also a piece of 
 thinking, and so, on the theory, would have to be a fresh piece 
 of constructiveness and nothing more. That is nonsense. If the 
 Absolute or the Zeitgeist or creative life organises experience, 
 this experience surely is organised and, if so, the recognition of 
 the orderly result is surely different from the production of it. 
 Even an organising principle has its own character, and this 
 character cannot be another organising principle. It, at least, 
 is simply found, and there is no room for this naked discovery 
 in terms of the constructive theory itself. 
 
 The description of an orderly mental product, in a word, is 
 logically in pari materia with any other piece of description, 
 and so implies that general facts hold of particular ones in the 
 same sense as they hold on any other theory. Nothing is 
 gained by stating the problem in terms of mental organisation, 
 and so we may leave the theory of dynamic impulses on one 
 side. The important logical problem remains precisely where it 
 was. The being of principles is guaranteed by the logic of any
 
 vi] PRINCIPLES 115 
 
 theory that has faced the facts; and our original problem 
 concerning the kind of being which principles have is still 
 with us. 
 
 As we have seen, there seems to be a dualism between the 
 mode of being which general facts have and the being of 
 particular facts. For general facts merely subsist, and par- 
 ticular facts (unless they are the universal instances of universal 
 principles) also exist. On the other hand, a dualism of these 
 modes of existence is a difficult theory to sustain, since all 
 particular existences logically require some general principles 
 to hold of them. 
 
 Perhaps we may find some light in our perplexity if we 
 consider the famous pact between universals and particulars 
 according to which universals are in re and neither ante rem 
 nor post rem^. 
 
 Part of this contention is so manifestly true that it has only 
 to be stated to be accepted. Particular things are determi- 
 nately so-and-so; that is to say, they have universal characters; 
 that is to say, universals are in them. Again, red things are 
 not redness. The universal is onlym its instances, if it is in them 
 at all. It is not identical with them, and it is not identical with 
 any part of them, since any portion of a particular thing is just 
 as particular as the particular thing itself. All particular 
 things, therefore, logically require universals; and we should 
 avoid any taint of dualism if we could also show that all uni- 
 versals logically require particular instances which actually exist. 
 
 This supposed requirement seems certainly to be fulfilled 
 in the case of some universals. What kind of being could 
 redness have if nothing were red, or what could sweetness be 
 if there were no toothsome things .f* What would baldness, or 
 sententiousness, or sleepiness be if they were not found in the 
 world.'' Any adjective, indeed, has a universal corresponding 
 to it, and some adjectives seem to be utterly and intrinsically 
 
 1 In the mediaeval controversy concerning the existence and potency of 
 universals, the extreme realists like Bernard of Chartres or William of Cham- 
 peaux held the ante rem theory, and nominalists like Eoscellinus held the post 
 rem theory in a most uncompromising form. Universalia in re is the mediating 
 position. See Mercier, Criteriologie ginirale (5th edition), pp. Z2Qsqq.
 
 116 PRINCIPLES [CH. 
 
 empirical. Universals of that kind, therefore, are surely in 
 their particulars, and have no other conceivable mode of 
 being. 
 
 Indeed, there seems to be only one limitation to this account 
 of the status of universals like redness or sententiousness. 
 Logic is satisfied if there are some existing instances of these 
 universals; it cannot deduce all the particular instances which 
 happen to exist. Kiilpe says, for example, that there are about 
 150 discriminate colours ^ I do not know whether that is the 
 right number, but I can be quite certain that the right number, 
 whatever it may be, cannot be deduced from the universal 
 'colour.' Granting then that 'colour' logically requires to have 
 some varieties in the world, it does not logically require to have 
 any determinate number of varieties. The universal ' man ' may 
 logically require mankind, but this circumstance does not 
 relieve the census officials. 
 
 This limitation is important, but there is a still more im- 
 portant point to notice. Even if the universal 'sententiousness'* 
 seems to be logically derived from existence, many universals 
 do not. It seems perfectly clear, for example, that two and 
 two would be equal to four if no couples existed in the world. 
 Pure mathematics, to continue the argument, is logically in- 
 dependent of its application to existence, and so is the pure 
 logic on which pure mathematics is based. For logical prin- 
 ciples are a pylori, and they apply to any thinkable being 
 whether such a being could exist or not. Logic applies to all 
 possible worlds as well as to the actual one, and if there are 
 thinkable beings which could not exist in any possible world 
 logic applies to these beings too. 
 
 It is true, of course, that a pj-iori principles, like those of 
 logic and number, do in fact apply to existence. All existing 
 things have a logical structure. Eggs, again, can be counted. 
 Therefore they have a number. Therefore they have all the 
 characteristics and implications of numerable things. If pure 
 geometry, as some maintain, can be completely arithmetised, 
 its application to existence needs no argument. If not, its 
 application to existence is not imperilled, for if the geometrical 
 1 Outlines of Psychology, p. 127.
 
 vi] PRINCIPLES 117 
 
 relation of betweenness, for example, is not a mere logical 
 relation, it is extra-logical simply in so far as it is tinged with 
 a meaning observed in empirical existence ; and this may per- 
 haps be inferred from the fact that the betweenness of right 
 and left is logically identical with the betweenness of before 
 and after, although these, most manifestly, are not the same. 
 These facts, therefore, do not affect the argument. It is 
 possible to maintain, it is true, that the general facts of num- 
 ber logically require some application to existence, although 
 this application to existence is taken for granted throughout 
 any demonstration in pure mathematics, and therefore does 
 not enter into the demonstration itself. If this contention 
 were sound it would meet the argument that two and two 
 would plainly make four even if no couples and no quartettes 
 existed; but it is impossible to see what grounds can be 
 adduced in its favour; and it is clearly absurd to argue that 
 pure logic or pure mathematics logically require existence just 
 because they apply to existence. 
 
 We must conclude, therefore, that some general facts are 
 logically independent of existence although existence itself 
 cannot be independent of general facts. In other words the 
 ultimate difference between existence and subsistence re- 
 mains. 
 
 It must remain, I think; and yet there is an excuse, at least, 
 for dallying with a very old conjecture. If subsistence cannot 
 be shown to be only a species of existence, may not existence 
 be only a species of subsistence? This contention has a long 
 history behind it. Plato's account of the participation of the 
 world of generation and corruption in the hierarchy of Forms 
 is an illustration of it. It is assumed in Anselm's form of the 
 Ontological Argument. It is seen in the Cartesian and Leib- 
 nizian theory that sensations are only confused reasoning, and 
 music, for example, a piece of unconscious arithmetic. Kant, 
 it is true, by insisting on the ineradicable difference between 
 sensation and understanding, led philosophy on to another 
 track, and razed the Ontological Argument to the ground. 
 But Kant himself strove to find a common font for these two 
 jets, and HegePs Absolute Idealism claimed to have overcome
 
 118 PRINCIPLES [CH. 
 
 the dualism. In our own days Mr Holfs 'neutral monism^' 
 has very obvious affinities to the older view. 
 
 Existence, to be sure, seems to differ toto coelo from mere 
 logical subsistence. A hundred dollars are very different from 
 the mere idea of the same. Living men are not merely the 
 logical properties of vitality. We shed no tears over death in 
 the abstract, unless we are worthless sentimentalists. Only 
 existing things can do any work, and no other things can, 
 properly speaking, be idle. 
 
 And yet it is possible to argue the question. 
 
 In the first place there is a well-known argument of Hume\s, 
 "To reflect on anything simply," he says, "and to reflect on it 
 as existent are nothing different from each otherl"" And that, 
 in a sense, is true. Existence, so to speak, happens to the 
 things we contemplate. When we think of an angel and j udge 
 that an angel exists, the angel has the same characteristics in 
 both cases. The only difference is that we judge that it exists 
 in the second case and not in the first. This difference, how- 
 ever, is all-important. Indeed, it is so important that it refutes 
 Hume's argument. Anyone can think of an angel stirring the 
 waters, but not everyone is constrained to believe that a real 
 angel really stirred them. The only way of constraining us to 
 this belief would be to produce evidence that some trustworthy 
 person had actually perceived the angel. In other words a 
 special sort of evidence is required of all propositions which 
 assert existence in rerum natura. Anyone can think of 
 Gaunilo's perfect island^ but no one would believe that the 
 perfect island existed unless some one had seen it; and if any 
 mariner brought the news of it, we should think ourselves 
 bound to test his story by sending an expedition to the spot. 
 
 In the second place, it might be argued that existence is just 
 
 ^ In The Concept of Consciousness. 2 Treatise, bk i. pt n. sect. vi. 
 
 2 Gaunilo, a monk contemporary with Anselm, disputed the Ontological 
 Argument of Anselm's Monologium in his Liber pro insipiente. In terms of 
 Anselm's reply, the ' insipiens' was unconvinced ' quia non magis consequitur 
 hoc, quod dico, quo mains cogitari non possit, ex eo quia est in intellectu, 
 esse et in re, quam perditam insulam certissime existere ex eo quia cum 
 describitur verbis, audiens eam non ambigit in intellectu suo esse.' Liber 
 apologeticus contra Gaunilonem.
 
 VI] PRINCIPLES 119 
 
 the kind of subsistence which spatial and temporal things have. 
 Existence is the asserted being of spatial and temporal things, 
 as opposed to their contemplated being. When we make judg- 
 ments about things which are also perceptible, our assertions 
 claim to be time of existence. Otherwise, they should not. 
 
 This argument seems to be an ignoratio eletichi, because it 
 is clearly no answer at all to the contention that existence is 
 a different order of being from mere subsistence. The fact 
 that perception (or introspection) is the only possible evidence 
 for existence is common ground, and therefore does not tell in 
 favour of this argument. Moreover, there is no contradiction 
 in believing in the possibility of non-spatial and non-temporal 
 existence, although we do not encounter such things in 
 ordinary life. Orthodox theology, it is true, maintains that 
 God is non-spatial and non-temporal, but these orthodox 
 tenets may very well rest on a mistake. It is a mistake, for 
 instance, to suppose that God's existence must be fleeting and 
 perishable just because it is temporal; for God is not mocked if 
 he endures while infinite time endureth. Again, it is a mistake 
 to argue that changelessness is more noble than change, or that 
 God, being changeless, is out of time. For change in itself is 
 not a defect, and the unchanging is just as temporal as the 
 changing. The everlasting hills are not less temporal than a 
 poppy's petals. They only endure longer. And I cannot see 
 that there is anything derogatory in spatial existence whether 
 or not it is totum in toto ac totum in qualibet parte. But enough 
 of this digression. We have no certain evidence for the 
 existence of anything which is neither spatial nor temporal; 
 and we have no right to deny the possibility of such existence. 
 
 AVe conclude, therefore, that there are general facts as well 
 as particular ones; that these general facts may apply to 
 existence, but that their validity is logically independent of 
 existence; and that the subsistence of general facts cannot be 
 reduced to any characteristics of existence, nor conversely. If 
 this dualism seems lamentable, there is no way of avoiding it. 
 
 And there are compensations. The existent world, it is true, 
 is a very wide parish, but even the immensities and the 
 eternities are parochial in comparison with the discoveries of
 
 120 PRINCIPLES [CH. 
 
 the logical intellect. The freedom of thought is not a mean 
 thing, and the wings of the logical intellect are clipped if logic 
 is hampered by persistent appeals to sensory fact. Logic and 
 pure mathematics are more certain than inferences based on 
 sense, and they are not less worthy of being followed to a con- 
 clusion. There is a genuine basis in reality for the distinction 
 between what used to be called the verites Hernelles and the 
 veritea dejait, between the Forms and Becoming, between the 
 relations of ideas and the relations of matter of fact. The new 
 discoveries concerning infinity, or the analysis of arithmetic 
 and its logical basis which Mr Russell, stimulated by the 
 researches of Cantor, Peano and Frege', has forced upon 
 philosophy's attention, are not retrograde steps, although they 
 come nearer to Leibniz or Plato than to Heg-el. On the con- 
 trary they bring, as Mr Russell claims, 'a sense of power and 
 a hope of progress ^^ 
 
 It is highly important, too, that the status of general facts 
 should be recognised to be what it is. Classical idealism in 
 the grand marmer of Hegel sought to correct empiricism by 
 giving the understanding its due as well as the senses, and then 
 found a home for both of them in the higher synthesis of 
 Absolute Spirit. Now it is true that the first step in philo- 
 sophy is frankly to see that the senses cannot be the basis of 
 all knowledge, but it is a serious mistake to conclude that 
 what is not sensory must be the work of the mind. That is 
 only another fetter due to this arbitrary restriction of thought 
 to mere existence. Principles, it was argued, must be mental 
 because existence itself is either sensory or mental, and because 
 principles, plainly, are not sensory. We must strike off this 
 fetter too. General facts, as we have seen, are just as objective 
 as the facts of existence although they do not exist. They 
 confront the mind and reveal themselves to it. They are in- 
 dependent of our thinking and they are literally discoverable 
 
 1 Mr Eussell, who always makes a point of explaining his indebtedness to 
 other authors, explains that he himself amved independently at Fi'ege's 
 definition of number. See e.g. his Introduction to Mathematical Philosophy, 
 p. 11. 
 
 2 The External World, p. 30.
 
 vi] PRINCIPLES 121 
 
 as they are in themselves. For that reason they do not need 
 a higher immediacy. They are not aimless migrations of 
 wandering adjectives but ascertainable facts and relationships. 
 And we should accept them as we find them. 
 
 That is one side of the story. The other side shows that 
 there is insight as well as prudence in rendering unto Caesar 
 the things that are Caesar"'s. The general truths which depend 
 upon sensory observation are only probable conjectures al- 
 though they require a prioi'i principles. Scientific induction is 
 not strict demonstration; and the point is fundamental. 
 
 Even granting that causal interpretations have a basis in 
 perception, it is plain that our inferences to particular causal 
 laws are neither certain nor demonstrable. In an inductive 
 proof we argue that any A must be a B because all observed 
 A's have been B"'s. As it stands, this inference is a fallacy. 
 The most we can infer is that such a connection is likely. 
 And induction by means of the hypothetical method is also 
 fallacious on the face of it. We are never in a position to 
 prove that any given hypothesis is the only possible one, and 
 therefore we commit the formal fallacy of affirming the con- 
 sequent in all such inductions. 
 
 Obviously, therefore, we must conclude that inductive con- 
 clusions are matters of probability at the best. Probability, 
 however, is a branch of logic. It is the logic of relevant but 
 inconclusive evidence, and its axioms and assumptions are 
 a priori. It follows therefore that induction requires a priori 
 principles, and it is plainly not completely a priori since the 
 evidence for it is also, in part, the observation of certain 
 particular instances. But there is a further difference still, 
 since, as Mr Broad has shown recently^, there is no known 
 principle of mere probability which justifies our belief in 
 scientific inductions. For a full proof of this conclusion I must 
 refer the reader to Mr Broad's article, but I can sketch some 
 of the main points here. 
 
 If I infer that the atomic weight of nickel is 58*68 the 
 quantity of nickel that has been weighed is wholly insignificant 
 in comparison with the quantity of nickel in the universe, 
 1 Mind, N.S. vol. xxvn, No. 108, pp. dS% sqq.
 
 122 PRINCIPLES [en. 
 
 Now the probability in this case is measured by the fraction 
 
 ^ where m is the number of observed cases and n the 
 
 number of cases in the universe. This probability, therefore, 
 is too slight to be worth believing. It is true that the proba- 
 bility of the next case of nickel having the atomic weight 58'68 
 
 is ~ , but this formula will not help us very much. In the 
 
 first place we always infer the general law, and not merely the 
 probability of the next case; and in the second place the 
 formula for the next case is far too feeble for the inference 
 we want to make. If an inexperienced student with a poor 
 balance found the weight to be 61 "23 four times running, then 
 the odds would be five to one that the weight in the next 
 case would also be 61*23. Such a principle, therefore, is not 
 sufficient for our purposes. No one would put his trust in 
 induction unless he supposed that he had obtained enough 
 relevant evidence to justify assertions about the atomic weight 
 of nickel. 
 
 Induction, in a word, is worthless without certain special 
 assumptions about nature itself. We always suppose, in fact, 
 that the weight of nickel is constant when no one is weighing 
 it, and that fair samples of the element can be obtained so 
 that the experimental results in a few instances are evidence 
 for any instance. We assume, in other words, that there are 
 parallel cases, and that we can recognise the species of things 
 if we take the trouble to examine them carefully. Without 
 these assumptions induction would not yield any likelihood 
 worth considering, and with them it is only probable; for we 
 all admit that the increase in the number of instances makes 
 the evidence better, and our admission would be meaningless 
 if the conclusion stood to demonstration. 
 
 This difference in the character of inductive evidence as 
 compared with pure logic, is interesting in many regards, and 
 it is particularly important in view of certain metaphvsical 
 theories. The ideal of all rationalism is to show that every 
 feature of the realm of being can be shown to be a consequence 
 of a few comparatively simple principles, and we may conclude
 
 vi] PRINCIPLES 123 
 
 this chapter by considering, very briefly, whether this ideal is 
 well founded. 
 
 As we have seen, the a priori laws of logic, number and the 
 like, cannot be the principle of the empirical connectedness of 
 existence, although they hold of all existence. We have to 
 observe nature in order to discover her laws, and we have to 
 argue from these observations inductively and at our peril. 
 On the other hand, it is at least possible that nature behaves 
 in conformity with a comparatively small number of relatively 
 simple principles, and that these may be discovered. We saw 
 in an earlier chapter that the continuity and connectedness of 
 nature is given to the mind in perception in a fragmentary 
 but not in a negligible fashion. The inductive assumptions 
 that there are parallel cases and that fair sampling is possible 
 are extensions of this perceived meaning — too sweeping, indeed, 
 to be indisputable consequences from it, but, on the whole, 
 legitimate and even moderate. Is it not possible, then, to 
 include the interpretation of perceived meanings together with 
 the foundations of inductive inference concerning nature in 
 some general Law of Analogy or Principle of Sufficient 
 Reason .P Even if this principle is not wholly certain, it is 
 supported by very many facts and contradicted by none. And 
 it is, perhaps, the metaphysical ground of that law of parsi- 
 mony to which, as many consider, all sound thinking conforms. 
 
 The Law of Parsimony — the Friistrajit per plura quodjieri 
 potest per pauciora of Occam''s razor — is manifestly just and 
 highly important. In logic it means that 'analysis is to be 
 carried as far as possible^'; in physical science it means that 
 'as hypothesis increases necessity diminishes^'' Again it is a 
 principle of elegance. The fewer the assumptions the neater 
 the proof. It is an appeal for economy, too, and sumptuary 
 edicts have excellent intentions, although luxury is a hard thing 
 to define and an impossible thing to curb. On the other hand 
 the limitations of a finite intellect compel us to be sparing in 
 our assumptions if we hope to understand anything of im- 
 portance; the power of the intellect is simply its grasp of a 
 
 1 P. E. B. Jourdain, The Monist, vol. xxix. No. 3, p. 451. 
 
 2 Whitehead, The Organisation of Thought, p. 176.
 
 124 PRINCIPLES [ch. vi 
 
 multitude of details in a single principle, and no intellect can 
 grasp very many principles in this thorough fashion. Those who 
 count knowledge precious may therefore be expected to believe 
 that nature responds to a principle of this kind. Natura Jiorret 
 superjluum. The structure of reality is simple and so our 
 thinking should be simple too. 
 
 This final step seems more seductive than solid. True, we 
 cannot understand things unless we can bring them under a 
 comparatively simple principle, and we are bound to believe 
 that there are principles, and that explanations can be given 
 in terms of them. It does not follow, however, that the 
 principles which are most important for our understanding of 
 things are also most important in the order of nature, still less 
 that nature as a whole is peculiarly adapted to our limitations. 
 If maxim can be met with maxim, we might follow Kant by 
 opposing Entium varietates no?i temere sunt minuendae to 
 Occam"'s razor*. The complexity of the universe is always with 
 us and we have no right to set bounds to it. We may hazard 
 the guess that logic cannot be less fundamental in the world 
 than in those selected portions of reality which we are capable 
 of apprehending; and we should not go further than that. 
 
 ^ In the Appendix to the Transcendental Dialectic.
 
 CHAPTER VII 
 
 VALUES 
 
 The world that earth-born man, 
 
 By evil undismay'd, 
 
 Out of the breath of God 
 
 Hath for his heaven made. 
 Where all his dreams soe'er 
 
 Of holy things and fair 
 
 In splendour are upgrown, 
 
 VVTiich through the toilsome years 
 
 Martyrs and faithful seers 
 
 And poets with holy tears 
 Of hope have sown. 
 
 Robert Bridges. 
 
 The value of things, I think, does not raise any special prob- 
 lems in the theory of knowledge. Value is a quality which 
 things may have or may not have, and it can be recognised 
 by the mind like any other quality. On the other hand, the 
 problems of value are important and distinctive enough to 
 repay separate discussion. According to many philosophers, 
 value and existence are somehow one, and this principle, if it 
 were true, would plainly have most important metaphysical 
 consequences. Again, even apart from the perplexities of that 
 majestic theme, there are many humbler difficulties in the 
 conception of value, and some of these difficulties, if they were 
 insurmountable, would have very important results ; for they 
 would show either that there is no such thing as value, or else 
 that values cannot be known. And if knowledge is impotent 
 concerning all questions of worth there might perhaps be 
 neither use nor worth in considering it. 
 
 The orthodox tradition in philosophy classifies values into 
 truth, beauty and goodness, and then sets about to establish 
 a legislative and executive union of logic, aesthetics and ethics. 
 This procedure, I think, is mistaken. It omits certain im- 
 portant values from the list (happiness, for example), and it 
 includes truth which is not, properly speaking, a value at all.
 
 126 VALUES [CH. 
 
 True knowledge is a value, and perhaps the highest of all 
 values, but truth itself means only that the facts are so, and 
 it is hard to see how that can fortify anyone's soul. There may 
 be value and importance in krwwing that Popocatepetl is higher 
 than Shooter's Hill, but this true proposition itself is only a 
 state of affairs. 
 
 It is otherwise with beauty and goodness, for these are 
 valuable in themselves, whether or not anyone appreciates the 
 fact. The justice of Aristides was independent of the suffrages 
 of the Athenians, and if the man was just he was therefore 
 worthy. In the same way, beauty need not wait upon the 
 fashion. A romantic revival may be needed to reveal the stateli- 
 ness of gothic cathedrals or the serene splendour of Alpine 
 summits, but this beauty, and the worth of it, belonged to 
 the Alps and the sanctuaries all the time. 
 
 These statements, I am well aware, require a defence, and 
 I hope to give one (with certain reservations concerning beauty). 
 The most formidable lion in my path is the common and 
 most pertinacious attempt to prove the subjectivity of all 
 values. Aesthetics gives the strongest case for this argument, 
 but the subjectivists also attempt to annex the whole province 
 of morals. We must therefore consider these topics in their 
 order. 
 
 Mr Saintsbury, defending Castelvetro's doctrine that delight 
 is the sole end of poetry, writes as follows: — "That Burns 
 administers, and has a right to administer, one delight to one 
 class of mind, Shelley another to another: that Beranger is 
 not to be denied the wine of poetry because his vintage is 
 not the vintage of Hugo : that Longfellow, and Cowper, and 
 George Herbert are not to be sneered at because their delight 
 is the delight of cheering but not of intoxication : that Keble 
 is not intrinsically the less a poet because he is not Beddoes, 
 or Charles Wesley because he is not Charles Baudelaire — or 
 vice versa in all the cases — these are propositions which not 
 every critic — which perhaps not very many critics — will admit 
 even in the abstract, and which in practice almost every critic 
 falsifies and renounces at some time or other\" 
 1 History of Criticism, vol. ii. p. 87.
 
 vii] VALUES 127 
 
 The reason for this inconsistency among the doctors is not 
 far to seek. Common sense and the King's English renounce 
 subjectivism, and so the critics are bound to contradict them- 
 selves verbally unless they are intolerably prolix, and are cer- 
 tain to contradict themselves really whenever they relapse into 
 common ways of thinking through the weakness of their critical 
 flesh. And there may be another reason. Perhaps the critics 
 cannot keep faithfully to their own standards, because these 
 standards are false, and because the truth of the opposite 
 catches them unawares whenever they try to state their posi- 
 tion fully. Be that as it may, there is a strong personal 
 obstacle, at all events, in the way of complete subjectivism 
 among the critics. If they were consistent subjectivists, none 
 of them could be right and none of them could be wrong. 
 This consequence would be fatal to their pretensions ; and 
 perhaps their pretensions, in this instance, are better than 
 their creed. 
 
 Delight, I take it, is simply a feeling, and feelings are 
 neither right nor wrong. It is illogical, therefore, to dispute 
 about tastes, if tastes are only feelings; and those critics who 
 follow the subjectivist theory have only the right to say that 
 they like Baudelaire or Mrs Aphra Behn with some special and 
 peculiar relish ; and either that they are not alone in their 
 taste or that they are proud to be alone. The critics, to be 
 sure, may explain the way in which this or that offends them, 
 and point out what they like best, just as Meredith complained 
 of smug Victorianism in the Idylls of the King, or as Francis 
 Jeffrey drove about Edinburgh declaring there had been 
 'nothing as good as Nell since Cordelia^' Delight in detail, 
 however, does not differ in principle from delight in the whole, 
 and there is as much logic in Stevenson's blunt dislike of 
 M. Anatole France as in Meredith's contrast between the 
 * gentleman Boccaccio' and the 'Sir Pandarus public' which 
 liked 'the Euphuist's tongue, the Exquisite's leg, the Curate's 
 moral sentiments, the British matron and her daughter's 
 purity of tone' in The Holy Grail-. It is all one to delight in 
 
 1 On the authority of Forster's Life of Dickens, vol. i. p. 226. 
 
 2 The Letters of George Meredith, vol. i. pp. 197-198.
 
 128 VALUES [cH. 
 
 a jewel five words long and to delight in a quarto from frontis- 
 piece to colophon. The only difference is that the latter delight 
 needs more space. Anthologies may give delight like the 
 flowers in a crystal vase, but the flowers are as sweet in the 
 garden, and they are better in the field. Excellence should not 
 be rent into patches. The author has to struggle for the mot 
 rat/onnant, but his travail should be the reader's ease. 
 
 Very few of the great critics, I fancy, hold that delight is 
 mere liking without any qualitative differences in it. Mr 
 Saintsbury plainly does not; and, indeed, he adds an aggrieved 
 and petulant footnote to the statement quoted above. " It is 
 perhaps well," he says, " to meet a possible though surely not 
 probable objection. 'Do you deny ranks in poetry.''' Cer- 
 tainly not, but only the propriety of excluding ranks which 
 do not seem, to the censor, of the highest^" Mr Saintsbury's 
 view is that there are different kinds of delight, and different 
 ranks in it, and our problem is whether his theory is sufficient, 
 and what is the logic of it. Meanwhile we must certainly 
 admit that there are, in fact, different kinds of delight. From 
 the aesthetic point of view pushpin is not as good as poetry 
 when the quantity of pleasure is equal ; and Yvetot is not as 
 good as Constantinople. The delight in one of Vermeeren's 
 paintings is not the delight in Bach, and the gourmefs delight 
 in plovers' eggs is not comparable to our delight in a Delia 
 Robbia. Each of the arts, indeed, has its appropriate delight, 
 and each division within each art. The delight in Theocritus 
 differs from the delight in Lucretius, and the delight in Donne 
 is other than the delight in Heine. Indeed, there is no good 
 reason for stopping at this point. There are many strings in 
 Erato's lyre, and each has its appropriate charm. Melpomene 
 has many masks, and each mask has its own delight. The 
 greatest poets give the greatest range of delights, and critics 
 excel when they respond to them all. 
 
 These facts, however, do not prove that beauty is only a 
 feeling of delight ; and that is the subjectivist's case. Let 
 us ask, therefore, whether he can support his position by 
 argument. 
 
 1 History of Criticism, vol. ii. p. 87 n.
 
 vii] VALUES 129 
 
 It is commonly argued that beauty is plainly a thing of 
 the mind and therefore subjective. The reply to this argu- 
 ment is that the fact is doubtful and the inference false. 
 
 There is beauty, I take it, in sky and cloud and sea, in 
 lilies and in sunsets, in the glow of bracken in autumn and in 
 the enticing greenness of a leafy spring. Nature, indeed, is 
 infinitely beautiful, and she seems to wear her beauty as she 
 wears colour or sound. Why then should her beauty belong 
 to us rather than to her ? And why should the beauty of art 
 need a different explanation ? A melody can be heard, and 
 temples, paintings, and statues can surely be seen. If every- 
 thing we perceive is mental, then beauty, of course, is mental 
 too. But where is the necessity otherwise ? Literary art, it 
 is true, may seem of a different order, but this appearance 
 may be a deceit after all. The cadence of words is just a kind 
 of music, and the magnificence of numbers can be heard. It 
 is idle to argue that this cadence and this magnificence are 
 the vestures of thoughts and that thoughts are mental ; for 
 thoughts, to be sure, are mental, but literature does not pre- 
 sent us with thoughts. It presents us with things that we 
 think about. Anyone, I think, who takes the trouble to con- 
 sider what precisely is before his mind when he reads a page 
 of Tom Sawyer or of Blake's Songs of Experience finds the 
 problem very puzzling. Images are blended with words, and 
 these in their turn with present questionings. But whatever 
 the pattern of the fact may be, fact of some kind is certainly 
 before the mind. The printed page is a turnstile which clicks 
 us into things. We are confronted with fact when we read, 
 and this fact is not our thoughts about it. 
 
 It is difficult to see, therefore, why so many philosophers 
 and critics hesitate to ascribe beauty to things in the same 
 sense as they ascribe colour or shape, but even the most deter- 
 mined realists incline towards compromise at this point and 
 admit that the status of the ' tertiary qualities "' is somewhat 
 dubious. Probably they are swayed by several considerations, 
 but we may consider two at this stage of our argument. The 
 first is the intimate connection between beauty and delight, 
 and the second is the extent to which personal meanings 
 L. 9
 
 130 VALUES [CH. 
 
 and personal associations are involved in the appreciation of 
 beauty. 
 
 The first consideration is not convincing. Feeling is always 
 blended with our thinking. A man has his feelings even when 
 he thinks of the multiplication table, but that does not prove 
 that the multiplication table is part of his mind. Certainly, 
 it is far easier to distinguish the feeling from the multiplication 
 table than to distinguish the delight in poetry from the beauty 
 of it. That is no proof of identity, however; and in certain 
 palpable instances the judgment of beauty seems to differ very 
 notably from any feeling of pleasure. Indifferent art is not 
 always unpalatable. There are many who prefer Mr Standfast 
 to Colonel Jack, although they know perfectly well that Colonel 
 Jack is far the better. If pleasure were the sole criterion, this 
 discrimination would be out of the question, and it is difficult 
 to see how the discrimination can be due to a qualitatively 
 different feeling of delight in the two cases; for both Defoe's 
 story and Mr Buchan's are just vigorous pieces of narrative. 
 Indeed, the fact seems to be, quite plainly, that we judge 
 Defoe's work to be the better, but that we may find more 
 interest in Mr Standfast for the time being. 
 
 The second circumstance is equally unconvincing. True, the 
 beauty of literature, for example, is permeated with associa- 
 tions. Without these, wit and simile, irony and allusion would- 
 be utterly wasted, and these things are the attic salt of good 
 prose and of good verse. Pathos and anti-climax, again, or the 
 sermo pedester in place of superbia canji'momm, construct ionis 
 elatio and excellentia vocabulorum, offend us precisely because of 
 the discord which their inevitable associates bring. In the same 
 way, the expressiveness of chiselled stone or shining canvas is im- 
 personal only in the sense that it is catholic. The eye of the 
 beholder must bring experience with it in order to appreciate 
 the charm and the meaning of these things. It is impossible, 
 however, to extract a convincing argument from this circum- 
 stance. All our perception has an acquired meaning, and even 
 the allusiveness of literature is not on a different plane from the 
 meaning of perception or judgment. It is subtler, more recon- 
 dite, more palpably acquired, but the essence of it is the same.
 
 vii] VALUES 131 
 
 It would be most illogical to argue that nothing can be 
 objective if it is acquired and not primitive, but even this 
 argument is not open to the subjectivists. For by what right 
 do they deny that a sense of beauty really is primitive ? The 
 noble savage, it is true, is so often a stupid fiction that it is 
 useless to speculate about him, and the zeal of missionaries to 
 discover whether there are savages without a code of morals or 
 a belief in unterrestrial potentates, has left us very sparse data 
 for guessing how far the untutored races appreciate the beauties 
 of nature. On the other hand, art is nearly as old as man. 
 It is a very long time since the first cave was rudely adorned, 
 since barbaric taste thrilled with the joy of glittering things, 
 since song and dance appealed to men, and since sagas were 
 told over crepuscular embers. We may safely conjecture, then, 
 that the sun-god was seen in his beauty as soon as he was seen 
 at all, and that Neanderthal man had something of the artist 
 in him when he rejoiced in the spring. Men have fancied that 
 molluscs sing praises to God. It is easier to fancy that they 
 respond to beauty. Peter Bell is not the only type of peasant, 
 and yellow primroses are only yellow to many who are not 
 peasants. Have we not heard, indeed, of a great philosopher 
 who likened the stars to a rash ? 
 
 Let us pass, then, to the most familiar argument on this 
 topic. Beauty, we are told, must be subjective because it 
 varies from man to man. Fire burns here and in Persia, but 
 the standards of good taste vary with time and climate and 
 training. There is a pleasant tale of certain anthropologists 
 who visited the Malay Archipelago, and took records of the 
 native music. They found a sort of rhythm in the head- 
 drummer's measures, but no music ; and they rejoiced when an 
 understudy took his place, for then they heard something like 
 music. The natives thought otherwise, however, and were as 
 disappointed as a music-hall audience when a popular comedian 
 falls sick. Who can arbitrate, then, between the taste of the 
 anthropologists and the taste of the natives ? And the same 
 thing holds universally. The precious ones who cannot admire 
 Raphael differ from the great public, but which of them is 
 right ? And if anyone prefers I^andseer to Cezanne why must
 
 132 VALUES [CH. 
 
 he be wrong ? There are different fashions in these affairs, and 
 what else is there ? 
 
 These contrasts, I think, are more striking than just. There 
 are great differences, to be sure, between the age of Solomon 
 and the age of Queen Anne in all matters of sensibility, and 
 between Constantine''s taste and the demon-trappings of some 
 patron of ju-ju. Such contrasts, however, need not be about 
 the same things. Perception depends so much upon what we 
 have learned to look for, that it is not surprising if primitive 
 taste differs from civilised. A savage does not perceive the 
 same thing as a modern European, and if the pair could pass 
 an eye round, like the Graiae, each might perceive the same 
 beauty. Again, the contrast is not so absolute as it seems. 
 The Indian''s beads and wampum have a certain beauty; and 
 we can understand his delight and even share it if we choose. 
 Beauty need not be denied just because it does not seem the 
 best, and if the golden splendour of Solomon''s temple with its 
 palm-trees and cherubim and open flowers was too ostentatious, 
 and the brazen pillars of his house with their chapiters of brass 
 and their two hundred pomegranates were too sumptuous for 
 modern taste, the temple and the palace were magnificent all 
 the same, and would have seemed so to anyone. Indeed, it is 
 very easy to exaggerate this clash in the standards of beauty. 
 The return to old standards is as patent a fact as the alternation 
 of new ones. Greek beauty has been the perpetual privilege of 
 all beholders, though sometimes a Winckelmann was needed 
 to point it out. Great art has a catholic and an abiding 
 appeal. 
 
 This argument, to be sure, is easily overstated, but there is 
 enough of truth in it to show that the clash of standards is 
 not the whole truth. This clash of standards, moreover, only 
 proves that judgments of beauty are often wrong, and that it 
 is difficult to be certain when any of them is right. True, 
 the difference between maintaining that there is beauty al- 
 though no one can ever know when he has found it, and that 
 there is no such thing as beauty, is so slender that it is not 
 worth quarrelling over; but the difference is crucial if (as seems 
 to be the fact) taste is not wholly wayward, if its development
 
 vii] VALUES 133 
 
 is intelligible, if its early stages are not superseded utterly, and 
 if the great epochs of beauty in art can almost always be ap- 
 preciated by later ages, although each age must be doubtful 
 of its own art until time has taken its measure. 
 
 Beauty, in fact, is something that is judged, not something 
 that is merely felt. None of the arguments of the subjectivists 
 is able to impugn this fact, and our judgments of beauty are 
 either right or wrong simply because judgment itself must 
 be either true or false. Indeed, the doctrine of ranks in 
 beauty already admits this conclusion by implication. The 
 comparative must have the same logic as the positive, \\Tiat- 
 ever is placed in a higher rank must be placed there because 
 it really is better ; and if it is better it must either be good or 
 bad in itself. 
 
 At the same time it is possible to argue that nothing is 
 beautiful save our delight itself, and that we call things 
 beautiful just because they cause beautiful delights. Our 
 judgments of beauty, on this view, would refer to our feelings, 
 and not to things. Our delights really are valuable, some 
 delights are really better than others ; and that is the sum of 
 the matter. We must consider this theory, therefore, for 
 something more than completeness' sake. 
 
 Is it true in fact that there is no beauty in comeliness or 
 majesty save only the feeling of delight.'' It is plain, of course, 
 that we do not usually say this even if we mean it, for we say 
 that Fiesole, or Leonardo's Last Supper or Beethoven's Ninth 
 Symphuny is beautiful without further ado. Language, how- 
 ever, has a short way of dealing with intricate matters, and its 
 crude methods should not, perhaps, affect our conclusions. The 
 problem is not about what we say but about what, on reflection, 
 we really mean. 
 
 The most extreme contention on the opposite side would 
 be that the Ninth Symphony would be beautiful if no one had 
 ever heard it, and that the frozen seas would still glisten with 
 loveliness after all life has departed from the earth. This 
 view, I think, is not nonsense, and it is, of course, extremely 
 simple, but most philosophers recoil from it for better reasons 
 than mere repugnance. This contention, they think, plainly
 
 134 VALUES [cH. 
 
 goes beyond the evidence ; for beauty, in our experience, is 
 never appreciated without delight, and therefore it is illegiti- 
 mate to argue that beauty would still be beauty in the 
 absence of any possible delight. The fact of inseparability, 
 to be sure, would not prove very much in itself, for many 
 things are inseparable which are plainly distinct, but when 
 the companionship, as in this instance, is peculiarly close and 
 peculiarly relevant the inseparability may mean a gi'eat deal. 
 The fact is not simply that we find charm wherever we find 
 beauty, just as we find scales wherever we find fish. It is easy 
 enough to distinguish between the fish and their scales, but 
 the union between beauty and delight seems to be as close as 
 the union between a man's enjoyment and the fruition of his 
 wishes. 
 
 On the other hand the delight in beauty is a very complex 
 thing, and it is far richer than mere feeling. If the sense of 
 beauty must be suffused with delight, this delight, in its 
 turn, presupposes a very subtle harmony of the mind as a 
 whole. The delight needs vision and insight and under- 
 standing. It is not merely superadded to these, but it blends 
 with them; and they in turn blend with it. Words like 
 'charm' or 'delight,' to be brief, signify, not mere feeling, 
 but an intricate sentiment of the soul in which feeling, how- 
 ever predominant, does not extinguish all else. If our minds 
 are charged with feeling whenever we appreciate beautiful 
 things, our feelings are also enlightened by our knowledge of 
 the significance of these things. Beautiful things, therefore, are 
 much more than occasions of beautiful feelings, and if beauty 
 is not a predicate of non-mental things simplic'iter, it cannot 
 hold of anything less than the whole complex ' thing-that-is- 
 felt-with-delight.' A perceiving and comprehending delight 
 is no mere feeling. 
 
 We may conclude, then, that delight enters into the 
 recognition of all beauty, but that beauty is not merely de- 
 light, even if we have no right to maintain that things can 
 be beautiful apart from a mind. They might be, but we 
 cannot tell. On the other hand, things may certainly be 
 beautiful when they bring delight, and the beauty (and there-
 
 vii] VALUES 135 
 
 fore the value) of these delightful things is a predicate of 
 them just as certainly as their lustre is a predicate of my lady's 
 diamonds. 
 
 The subjectivists can unmask the same batteries against 
 moral standards as against aesthetic ones. Shaftesbury, for 
 example, spoke of a taste or relish for virtue^ although he 
 recoiled from any determined logical pursuit of his own 
 theory. Most people who do not make a business of reflecting 
 on these matters are content to believe that conscience is only 
 a feeling, and that wicked imaginings ought to be destroyed 
 simply because they are felt to be defiling. For conscience is 
 an oracle, and oracles love the twilight. Again, the variations 
 of ethical standards are as striking as the variations in the 
 fashions of beauty. Those puberty rites that seem so foul to 
 us are the first obligation in Central Australia. We have no 
 moral copybooks for the ethics of a strike, and war convulses 
 the duties of the moral world as thoroughly as any cataclysm 
 convulses the face of nature. Indeed, nothing is easier than the 
 dialectic of naturalism which sets about to prove that any 
 commandment in morals is a mere convention like the rule of 
 the road, and that the abiding thing in conduct is just life 
 and its instincts. 'Thou shalt not steaP — but what is robbery ? 
 If a man's ancestors have been allowed to keep a piece of 
 land, why should the community pay tribute for it.^* And what 
 is wrong in promiscuity of ownership except custom, con- 
 venience and strong prejudice ? What are the rights and 
 wrongs of polygamy, and where is the pretence of inter- 
 national morality when there is no international police ? The 
 whole thing is an affair of sentiment and convenience (is it 
 not ?), and moral standards, like all religions in Gibbon's epi- 
 gram, are equally true in the eyes of the populace, equally 
 useful in the eyes of the magistrate, and equally false in the 
 eyes of the philosopher. Hunger and lust, ambition and 
 comradeship, love of ease and joy of battle are the only 
 enduring forces in a human polity. 
 
 Even if these arguments were sound, however, they would 
 not prove their conclusion unless they could explain the fact 
 1 'An Inquiry concerning Virtue, or Merit,' Characterititickn, vol. ii.
 
 136 VALUES [CH. 
 
 of moral obligation ; and they cannot do that. Moral obli- 
 gation is neither habit nor fear, neither custom nor obedience, 
 neither fashion nor deference, and it is not the obscure feeling 
 of these. For what do we mean when we say that we ought 
 to do certain things, and ought not to do certain other things ? 
 Clearly we do not mean that we are forced to do the one and 
 forced not to do the other. Our zeal for righteousness may 
 perhaps constrain us but nothing else need ; and righteous- 
 ness itself includes obligation. Neither the enactments of our 
 rulers with their sanctions in the police court, nor the com- 
 mands of God with the carnal and spiritual penalties which 
 the lawyerly minds of priests have devised, are moral obliga- 
 tion itself. And obligation is neither custom nor habit. 
 There are customary obligations, it is true; and just men, I 
 suppose, have formed the habit of walking uprightly; but 
 there are obligations to non-conformity too, and these are felt 
 and acted on. The Hobbists among our moralists, therefore, 
 have reached a most desperate pass. They have to maintain, 
 by hook or by crook, that obligation is really the constraint 
 of fear although it does not seem so to us. Obligation, they 
 say, is an obscure sort of fear, which we take for an ultimate 
 principle, and it would vanish if we knew our own emotions 
 thoroughly. A theory which leads to such a morass as this 
 should be avoided at all costs, and yet Hobbes's attempt to 
 base obligation on fear\ was better argued and more plausible 
 in itself than most of the modern attempts to base it on 
 tribal ritual or a nation's habits. 
 
 What, then, is obligation ? 
 
 It is tempting to suppose that obligation is an ultimate 
 of ultimates, an autonomous imperative of which it is enough 
 to say that he who cannot discern it of itself need not borrow 
 a lantern elsewhere. Moral obligations, in a word, like 
 
 ^ Leviathan, pt ii. chap. xvn. : "The Lawes of Nature. ..of themselves, 
 without the terrour of some Power, to cause them to be observed, are contrary 
 to our naturall Passions.... And Covenants, without the Sword, are but Words, 
 and of no strength to secure a man at all. Therefore, notwithstanding the 
 Lawes of Nature... if there be no Power erected... every man will, and may law- 
 fully rely on his own strength and art, for caution against all other men."
 
 vii] VALUES 137 
 
 Luther's Ich kann nicht anders, or the modem Englishman's 
 and the ancient Persian's ^ pride in keeping his word, plainly 
 are imperatives which take precedence over all else. On the 
 other hand, there are reasons for any imperative, gi'anting 
 that moral obligation is rightfully lord over the will. The 
 good does not wait upon the ought, and the ought is en- 
 lightened by the good. For the will always seeks an end, 
 and it is contrary to all sound reasoning to suppose that the 
 imperative justifies the end. On the contrary, the end justifies 
 the imperative. The reason for obligation, then, lies in the 
 value of the end. It is always legitimate to ask why anyone 
 ought to do this or that, and always sufficient to answer 
 (if we can) that he ought to do it because it is the best thing 
 to do. 
 
 The analysis of obligation, therefore, shows, on the one 
 hand, that obligation cannot be resolved into a mere feeling 
 or sentiment, and, on the other hand, that there are reasons 
 for obligation based upon judgments of value. Morality 
 stands or falls, therefore, with the validity of judgments of 
 value; and if these are not true, the grounds of right action 
 cease to hold. We have already considered most of these indict- 
 ments in principle, but the subject is so important that it is 
 better to consider them again from the point of view of moral 
 theory. 
 
 Those who resolve a judgment of beauty into a species 
 of appreciative feeling commonly maintain that moral judg- 
 ments are also a kind of appreciative feeling. Virtue pleases 
 us in a certain Avay, and vice displeases us in its appropriate 
 emotional fashion. Inequality or injustice offends us some- 
 times and arouses our resentment. At other times it pleases 
 us, for our sympathies may be with the princes, or the pluto- 
 crats, or the other superiors which God or our own indolence 
 have set above us. This analysis, however, palpably fails to 
 describe what we mean by injustice or any other moral evil. 
 We know very well that A has no moral right to prefer him- 
 self to B unless A's value is really greater than B's, and we 
 know that the justice or injustice of princes is not an affair 
 1 Cf. Herodotus, History, i, 138.
 
 138 . VALUES [CH. 
 
 of sympathy or resentment. Indeed, those who take this line 
 of argument speedily give the case away by the modifications 
 which they are compelled to make. Sympathy, they admit, is 
 not moral in itself. What is moral is sympathy with a 'flavour 
 of generality ^^ Resentment is not moral ; it has to be the 
 resentment of Adam Smith's ' impartial spectator' ; it has to be 
 watered with sympathy and enlightened knowledge. These are 
 desperate expedients, for they are intended to explain why 
 moral judgments claim to be binding on all and to hold irre- 
 spective of persons, although sympathy and resentment make 
 no such claims. 
 
 What these theories really do is to give an analysis of 
 moral feelings which is as nearly accurate as may be, granting 
 the false assumptions that there are no moral judgments and 
 that moral judgments have no connection with moral feelings. 
 These moral sentiments certainly exist and we should be 
 grateful to Hutcheson, Hume and Westermarck for their 
 psychological analyses of them ; but they are not the whole of 
 moral theory. Conscience is not merely the abhorrence of foul 
 deeds, although foul deeds are abhorrent ; and the head con- 
 demns Heliogabalus or Alexander Borgia, granting that the 
 heart loathes them too. These sentiments, indeed, like any 
 other sentiments, are allied to judgments, and, what is more, 
 are allied to judgments of value as well as to judgments of 
 fact. Righteous anger, for example, is enlightened by, and 
 derived from, the discovery of unrighteousness. Sentiments 
 are never blind, and reason is neither quite cold nor quite 
 aloof, although all the subjectivists in morals from Hume- 
 to Mr Bertrand Russell' have supposed so. These writers 
 
 1 Westermarck, The Origin and Development of the Moral Ideas, vol. i. 
 pp. 104 sqq. 
 
 ■■* Treatise, bk n. pt ni. sect. iii. : " Eeason alone can never be a motive to 
 any action of the will," "Eeason is and ought only to be the slave of the 
 passions, and can never pretend to any other office than to serve and obey 
 them "; and the whole section. 
 
 •* Principles of Social Reconstruction, p. 12: "Only passion can control 
 passion, and only a contrary impulse or desire can check impulse. Eeason, as 
 it is preached by traditional moralists, is too negative, too little livmg, to 
 make a good life. ...All human activity springs from two sources: impulse 
 and desire."
 
 vii] VALUES 139 
 
 maintain that morality must be an affair of feeling or impulse 
 just because moral ideas stir our emotions and lead to action. 
 They admit, to be sure, that the head plays some part in 
 action however cold it may be, for knowledge of the means 
 to an end clearly guides our actions, and blind feelings are as 
 useless in practical affairs as they are in a boxing-bout; but 
 they arbitrarily assume, none the less, that knowledge of the 
 values of the ends of action is impotent and aloof. Why ? 
 The only possible proof of this contention would be a demon- 
 stration that there is no such knowledge ; and this proof has 
 never been given. 
 
 We may pass, then, to other arguments. If beauty is 
 sometimes supposed to be subjective because of its close con- 
 nection with personality, one would suppose that this argument 
 holds a fortiori of morals. For persons are the bearers of 
 morality, and so far as we know, the only bearers of it. God 
 and the angels are moral beings if, and only if, they are persons ; 
 and theologians, recognising the difference between God's per- 
 sonality and ours, feel bound to hold that God is good per 
 alium modum eminentiorem^. Again, if animals have rights 
 (as most of us believe in a way) they must also have duties, 
 but moral obligations among the animals are so primitive 
 and so rudimentary that they are rightly excluded from serious 
 consideration in ethical theories. 
 
 Persons, then (together with their character, actions, and 
 capacities), are the final subjects of moral obligation, and, as 
 all Kantians know, the subjects of this law are also the law- 
 givers. These lawgivers, indeed, must not only be conscious 
 but must have foresight, choice of alternatives, some know- 
 ledge of means, some knowledge of values and some recogni- 
 tion of the independent worth of other subjects in the king- 
 dom of ends. These requisites are very stringent and they 
 exclude most of the animals and all of the plants, to say 
 nothing of ions and engines. Those, indeed, who maintain 
 
 1 Smnma Contra Gent. , lib. i. caput xxx. : " Quia enim omnem perfectionem 
 creaturae est in Deo invenire, sed per alium modum eminentiorem quaecum- 
 que nomina absolute perfectionem absque defectu designant, de Deo predi- 
 cantur et aliis rebus: sicut est bonitas, sapientia," etc.
 
 140 VALUES [CH. 
 
 that there arc no morals without Sittlichkeit deny morality to 
 anyone who is not a member of a relatively organised human 
 community, but they go beyond their record. A man's duties 
 in the concrete, to be sure, are so intimately connected with the 
 rights and aspirations of his fellows that there is no sphere of 
 private existence which the community does not touch ; and 
 even the most austere anchorite is no exception to this truth. 
 But the community is not a moral being. It is a community 
 of moral beings, and a man owes duties to himself irrespective 
 of the rest of society. Robinson Crusoe did not have to wait 
 for man Friday's coming in order to have duties, and a man's 
 morals are not entirely submerged in his social station. 
 
 There is nothing subjective in this view, despite Hegel. 
 The thesis of the Phenomenology^ and its successors is that 
 gifts and capacities are not valuable in themselves because 
 they may be either good or bad, that their value lies in the 
 rational organisation of them, that the state is just this 
 rational organisation, and so that the bearer of value cannot be 
 anything less than the state. It is an odd thing, perhaps, that 
 a theory which persistently contrasts the unity of the cosmos 
 with the imperfections of the sciences should be so ready to 
 find Reason writ large in politics. Reason must be strong 
 indeed to shine through the nmddle and the meannesses, the 
 intrigues, rancour and pretence of political history ; and the 
 philosophers of Hegel's school shut their eyes so often to the 
 facts of history, and so often descry Utopias by faith alone, 
 that their thesis has to be accepted blindly before it can be 
 followed wittingly. But enough of that. Gifts and capacities 
 have a value just because they are either good or bad. These 
 values imply a world of claims and counter claims, and the 
 claims are worth meeting just because they are genuine. 
 Although the conflict of values may be pernicious, the settle- 
 ment of them is worthless unless there are values to settle. 
 Values, in a word, are not subjective. They really pertain to 
 individual characters and capacities and do not merely seem 
 to do so. A man's capacities belong to him : his needs require 
 his neighbours ; and this objectivity is so thoroughgoing that 
 1 Professor Baillie's translation, vol. i. pp. Z&'dsqq. and especially p. 373.
 
 vii] VALUES 141 
 
 it constrains some and dismays others, that it wrecks lives as 
 well as makes them. Even if truth were unknown and impo- 
 tent, it might still be objective; but the values of morals and 
 of conscience are known and felt. 
 
 The strongest arguments for the subjectivity of moral 
 judgments are varieties of the plea that moral judgments are 
 so much expediency and opportunism, capricious, wavering 
 and temporary. It is held that some moral rules are plainly 
 arbitrary, and therefore that all may be : that some moral 
 rules are things of fashion or social status and therefore that 
 all may be : that moral rules are nothing unless they are uni- 
 versally binding, whereas it is admitted that every command- 
 ment has its exceptions. 
 
 The first argument need not detain us long. There is 
 nothing sacrosanct about many rules of conduct; but even if 
 every particular rule were arbitrary, it would not follow that 
 it is arbitrary to have any rules at all. Even the rule of the 
 road is not arbitrary in this sense, for the road would be 
 worthless if there were no rules. The laws of property, again, 
 as we find them in any given community, may not have a 
 very secure basis in ethics. A family may own estates because 
 the ancestors of its second cousin's great-uncle helped a weak 
 sovereign to despoil an arrogant bishop. Even if all the legal 
 titles to property were worthless, however, no society could 
 continue without rules for its security in possessing the goods 
 required for food, production, and comfort. Moralists, indeed, 
 usually try to prove too much. If they were content to prove 
 that there must be some rules of property or some rules for 
 the relations of the sexes, it would be impossible to gainsay 
 them. Instead of that they often argue as if there were no 
 alternative between chaos and primogeniture, or between 
 Christian wedlock and promiscuous sexual intercourse. It 
 should not be necessary to point the moral. Rules of property 
 and rules of wedlock are justified, pro tcmto, if they work. 
 These affairs must be regulated to prevent utter disorder. 
 And we know from experience that disorder is barbarous, 
 wretched and vile, even if a starving renniant of redeeming 
 virtues fllourishes sporadically in it.
 
 142 VALUES [CH. 
 
 The second argument can .summon innumerable examples, 
 but it is enough to refer to a few modern instances. Why 
 should uprightness be less of a virtue in the press than on 
 the bench, why should it be pardonable to cheat the revenue 
 or to steal from a hotel, why should blacklegs be judged so 
 variously, and why should a fallen woman be treated so 
 differently from a male libertine ? Some flimsy reasons may 
 be given in these cases, it is true, just as reasons may be given 
 to prove that duelling was once a duty and is now a crime, 
 that debts of honour should be met before tradesmen's bills, 
 or that lynching is not really the thing that it is. But these 
 reasons are mere excuses. Nothing is more precious than 
 honour ; and nothing, it would seem, more wayward. 
 
 Evidence of this kind, however, is far too slender to justify 
 the sweeping conclusion that there is neither reason nor truth 
 in morals. It would be a strange thing, indeed, if moral 
 rules were less capricious than they are ; for duty is just as 
 complicated as life. If we could be reasonably certain that a few 
 moral rules are generally binding, that is the most that could 
 be expected even of a society that reflected seriously on its duty. 
 And there never was such a society. Intuition takes the place 
 of reflection with nine men out of ten, and the moral reflec- 
 tions which abound in our novels, our pulpit and our press, 
 are only scattered suggestions summoned by a few notable 
 perplexities. When a man's standards of honour depend as 
 much as they do upon hearsay evidence and the expectations 
 of his friends, when conditions change so much that circum- 
 stances perpetually alter cases, when the ends of action vary 
 so nicely with the conditions of living, it would be strange 
 indeed if anyone could be sure of more than a very few moral 
 rules. 
 
 But can we be certain of any ? Are any moral rules uni- 
 versally binding ? One wonders what is meant by this question. 
 It is true, perhaps, that any code of rules like the ten com- 
 mandments conflict here and there, and that there is always 
 the chance of a hundredth case which contradicts one of the 
 rules flatly. I admit I cannot think of a case in which wanton 
 cruelty or sexual perversion is a plain duty ; but perhaps
 
 vii] VALUES 143 
 
 there are such eases. Again it may be true that very general 
 rules like Sidgwick's Axiom of Benevolence, or Kant's in- 
 junction to treat humanity always as an end, or the 'golden 
 rule' of our childhood, are either too abstract to apply deter- 
 minately to any particular case or else are faulty in some 
 respects. The principle that my neighbour's welfare ought to 
 count as well as mine never tells me by itself how to treat 
 my neighbour; I may have to treat engine-drivers as mere 
 means for my purposes : and the golden rule is false unless I 
 do unto others as they ought to do unto me, and even then 
 would fail signally if there were relevant differences between 
 us. But to grant all this leaves the main question untouched. 
 
 Every rule of conduct must be justified by the values it 
 subserves. When persons and circumstances differ as much as 
 they do, therefore, it is not at all remarkable that a course of 
 action that would be the best for most people under most 
 circumstances should not be the best for all people under all 
 circumstances. The rule ought to differ in its applications if 
 there are relevant differences between its instances ; and what 
 is generally right may sometimes lead to disaster. The logical 
 thing to do is to rejoice that so many rules hold in nearly all 
 cases, for this enables us to follow the rule with as little risk 
 as may be, and generally to dispense with the pains of hard 
 thinking. What is more, it puts the onus prohandi on any one 
 who holds that the rule does not apply in such and such a 
 case. But if the rule can ever be proved to be inapplicable 
 (as when the rule against suicide is absurd if the fortress 
 must fall and the women in it have something worse than 
 death to expect at the hands of their conquerors) then the 
 rule ought not to be followed in these cases for precisely the 
 type of reason which determines that it ought to be followed 
 in other cases. Suicide is a wanton destruction of values and 
 therefore wrong ; but if life must certainly be worse than 
 death, then suicide is a duty. 
 
 This circumstance, therefore, so far from proving moral 
 judgments invalid, may prove precisely the contrary. That 
 treachery is generally unjustified and promises generally 
 binding is as firm a rule as anyone ought to wish for ; and
 
 144 VALUES [cu. 
 
 there is no laxity in maintaining that reasons can be shown 
 for exceptions in war or in civil commotion. It would be an 
 accident if any rule were always binding so that we could 
 always 'damn the consequences'' and never needed to consider 
 the case on its merits. This accident may sometimes happen, 
 and one would be interested to meet a case of justifiable in- 
 gratitude, or praiseworthy lust, or admirable malice. The 
 argument, however, does not stand or fall because of the 
 certainty of rules without any possible exceptions, and it is 
 better to leave it so. 
 
 Summing up this long discussion, we may say that human 
 actions, human character and human dispositions have value 
 or worth. They are good or bad in a moral sense, and value 
 or its opposite belongs to them in the same sense as redness 
 belongs to a cherry. For similar reasons, the values of beauty 
 or its opposite belong to certain things in certain connections 
 just as objectively as any other qualities. And the reasons 
 which seek to prove the contrary are not well founded. 
 
 This result may seem to be very disappointing, for value is 
 the crux of most of the larger issues concerning the relation 
 of the human spirit to the world, and if philosophical enquiry 
 shows only that some things are valuable, some opposed to 
 value, and others indifferent, what grounds have we for hope 
 in our destiny or for reverence in the order of existence .'* 
 Religion seeks a metaphysical basis for the conservation of 
 values ; and beauty or happiness or true knowledge or moral 
 worth indicate, though they are not identical with, a certain 
 harmony between man and his environment. If the physical 
 order as a whole is indifferent to values, bringing tornados 
 as well as zephyrs, and earthquakes as well as sunshine, 
 values seem to be but fortunate accidents with little promise 
 of stability and no secure basis in the constitution of things. 
 
 Our argument, to be sure, has not attempted to prove that 
 the values which we can discern empirically are the only values 
 there are, but it has given no hint of the kind of premiss 
 which is sought by those who wish to establish a larger union 
 between values and existence. A survey of the empirical facts 
 seems to show conclusively that value and existence are not
 
 vii] VALUES 145 
 
 the same, since many things are neither good nor bad ; and it 
 finds no excuse for denying or mitigating the facts of suffering 
 or ugHness or sin. Again, it is impossible to prove from 
 empirical data by themselves that the universe has a bias to- 
 wards the good, that the human spirit is all but certain to 
 have ample opportunities in the world, or that there is always 
 a balance of good over evil. If these things are true, the 
 reason, it would seem, must be extrinsic to the character of 
 existence and of value as we find them. The world, perhaps, 
 is the work of a Creator whose aim is the good of his creatures 
 or his own glory; and the stubbornness of our environment 
 may really be our school and our opportunity. This extrinsic 
 connection, however, is not enough for most idealists, and 
 idealism appeals to many of us precisely because of its faith 
 in the reality of ideals. It takes its stand upon the superior 
 might of ideals in comparison with fact. The ideal, we are 
 told, is reality itself, and what we call fact is only the tarnished 
 surface of the ideal deceiving the foolish and rewarding the 
 wise. The universe is transfigured good because it is trans- 
 figured mind ; and without some premiss of this kind we are 
 yet in our sins. 
 
 According to Dr Bosanquet^ and those who think with him, 
 the thesis of the idealists depends on the unassailable premises 
 that there is no individuality short of complete existence, and 
 that the harmonious organisation of existence is just what we 
 mean by value. Anyone, we are told, who has fully realised 
 that existence is one and individual, therefore knows that all 
 is well with the world ; and he may even entertain a certain 
 chastened optimism concerning human life. Our private in- 
 terests, it is true, and our temporal and contingent aspirations 
 must expect a short shrift and a perfunctory hearing. Finite 
 personality is not a serious matter for this transfigured good. 
 Length of days, the reverence for human beings in a kingdom 
 of ends, the love of one man for another and the joy of the 
 lark in its freedom are only the fruits of finitude, and so are 
 without true being. If Absolutism gives us hope, therefore, 
 it gives little hope for those personal ties and temporal goods 
 1 The Principle of Individuality and Value. 
 L. 10
 
 146 VALUES [en. 
 
 which most men hold so dear. It is enough for us if values 
 exist, and absurd to ask who has them. Indeed it is kinder 
 to pluck the eye of finitude out than to let finitude pervert a 
 philosophy. Even the removal of suffering, the unmasking 
 of hypocrisy, the suppression of infamy, crime and pollution, 
 are in a way, less urgent duties according to this philosophy 
 than according to most others, for if all is well with the world 
 then the poisonous stench of iniquity and the unspeakable 
 horrors of lust and plague and war are not ultimately evil, 
 and so are either too superficial for genuine philosophy or else 
 are necessary elements in the dramatic fitness of things. ' Pity 
 would be no more if we did not make somebody poor, and 
 mercy no more would be if all were as happy as we.' On the 
 other hand, this optimism which is only another name for the 
 necessary perfection of the universe, has an impersonal and 
 timeless splendour of its own. For the sum of being is enough 
 for any man, and our finitude is a sorry thing to yearn after 
 when we participate in the whole. 
 
 The temper of realism is the reverse of this. Realists need 
 not deny, it is true, that the universe as a whole is a sublime 
 unity sempiternally perfect, but realism does not imply this 
 conclusion, and it does imply the full reality of good and evil 
 as we find them. For the realists, Borgia was a villain and 
 Francis a saint. What is more, Borgia's wickedness was Borgia's 
 affair, and not the work of the universe. The qualities of the 
 whole need not be the qualities of the parts; and many of the 
 parts may be very bad indeed even if the whole is good. 
 Realists, therefore, may logically accept the facts which they 
 find without referring to the whole which they do not know ; 
 and when they fight against real abuses, they are not compelled 
 to enquire into the perfection of reality as a whole. 
 
 It is difficult, no doubt, to avoid prejudice in these matters. 
 Very few philosophies stand in the way of anything a man 
 has a mind to do, and there is nothing to prevent an absolutist 
 or a mystic or any other metaphysical idealist from becoming 
 a crusader against sin and suffering and crime. Even if wicked- 
 ness is an illusion, that in itself is not a command to refrain 
 from meddling with it. Its half-being may disturb the half-
 
 vii] VALUES 147 
 
 being which we call life, and there is nothing illogical in 
 setting our own half-being against it. Indeed, it is possible 
 that philosophers could devise a theory according to which 
 time, although an illusion, is less of an illusion in its illusory 
 early stages than in its later ones, and thence infer that the 
 universe has summoned us to be instruments of progress, and 
 has enabled us to guess at the nature of the progress from 
 the character of the impulse to improve. This inference from 
 idealistic premises is not more illogical than quietism or in- 
 difference, but it is not less illogical; and there can be little 
 doubt that any theory which impugns the reality of time or 
 sin or suffering lessens the importance of removing evil. If 
 nothing but the whole matters ' in the end,' any finite enter- 
 prise and any human crusade is thereby belittled. It can only 
 ameliorate finite conditions; and there is a great difference 
 between theories which maintain that finite abuses and finite 
 remedies are real, and those which maintain the contrary. 
 
 " But you go too fast,'' I think I hear. Even if the penalty 
 of believing that ' the ideal only is actual ' were a certain 
 carelessness concerning the apparent rottenness of apparent 
 institutions, a certain acquiescence in the established order of 
 things, and a certain tardiness and conservatism in the appli- 
 cation of remedies, does not idealism contain a stimulus 
 to progress which realism must wholly lack ? Struggle is 
 pathetic without the conviction of victory, and how is this 
 conviction possible without the knowledge that the universe 
 itself is on the side of righteousness ? Idealism encourages 
 hope just because it gives a reason for this conviction, and the 
 realistic temper has no such prop to sustain it. When the 
 actual is taken to be finally true, nature is taken to be in- 
 different to values, and man, waging a precarious and unequal 
 struggle on the crust of the earth, has no prospect before him 
 save the temporary alleviation of a few evils. He cannot cope 
 with the greatest of all evils, for that is just the purposeless- 
 ness of human existence itself, and the littleness of mere 
 humanity. 
 
 Is it really so ? To begin a struggle without hope is indeed 
 a pitiful thing, but is it pitiful for man to struggle even if 
 
 10—2
 
 148 VALUES [CH. vii 
 
 he cannot count on the succour of the universe ? If human 
 life can be made worth living, is not that worth striving for ? 
 If wrong can be worsted, is not that worth doing ? Is duty 
 nothing because it is only a man's duty, and suffering nothing 
 because it is only human suffering ? On the contrary, if these 
 things are trivial, human existence itself must be trivial. If it 
 is, why should the universe trouble to help us ? It might 
 surely find something better to do. And if human existence 
 is not trivial, where is the worthlessness of the struggle ? 
 Throw a man on his own resources and he may do something 
 worth while. Make a pensioner of him and he will repay your 
 alms with feeble dependence.
 
 CHAPTER VIII 
 
 THE MIND 
 
 These postulata being admitted, it will follow in due course of reason- 
 ing that those beings, which the world calls improperly suits of clothes, 
 are in reality the most refined species of animals ; or, to proceed higher, 
 that they are rational creatures or men. For, is it not manifest that they 
 live, and move and talk, and perform all other offices of human life? are 
 not beauty, and wit, and mien, and breeding their inseparable proper- 
 ties? in short, we see nothing but them, hear nothing but them. 
 
 Swift, A Tale of a Tub. 
 
 It is time to pass from the objects of knowledge to the 
 process of knowing and to the mind itself. When things 
 confront the mind, the mind has to accept them, and realists 
 should be able to explain what they take this knowledge to be 
 
 Realists are committed to a doctrine of logical pluralism. 
 They maintain that there is nothing in the nature of knowledge 
 to prevent any given judgment from being wholly and finally 
 true, irrespective of the conditions of existence and of the 
 truth of other judgments, however closely the judgment may 
 be connected with these in fact. And realists accept a kindred 
 theory of perception, for they hold that physical things may be 
 revealed to the perceiving mind as they really and truly are 
 in their own proper character. 
 
 Many things, that is to say, can be known bi/ us. The 
 knowledge and observation which realists set out to defend is 
 human knowledge and human observation, not the celestial 
 apprehending of some impersonal cosmic intelligence. Realists, 
 it is true, should not deny knowledge of some sort to the ape 
 and the python on the one hand, or to the angels in heaven 
 on the other, but their principal concern is man's mind as 
 they find it. 
 
 Our knowledge is a temporal affair happening in common- 
 place life-histories, passing rapidly from one thing to another, 
 overcome by a drowsy nod, a whiff of ether or a gust of
 
 150 THE MIND [ch. 
 
 passion. At the same time it is acquired continuously, it is 
 fostered by education and experience, and it is not quite un- 
 systematic or quite capricious. Realists, therefore, have to 
 defend the reality of a mind which grows and then grows old, 
 which learns and forgets, which struggles, and studies, and 
 takes its ease. Thus far they are committed by their theory 
 of knowledge; but they are not committed further; and so 
 they may differ from one another very sharply indeed, in their 
 views of the mind in detail. Any psychology, indeed, which 
 does not implicitly or of set purpose deny that Smith or 
 Jones may know this or that finally and without qualification 
 is consistent with realism. 
 
 On the other hand, realism would be only a torso without 
 some philosophy of the mind, and so I intend to pass to this 
 question without further preamble. To be brief, my thesis is 
 that we must look to psychology if we wish to know what the 
 mind is, and that there are no sound metaphysical principles 
 which prove that psychological results must be merely pro- 
 visional. No psychologist or philosopher pretends, of course, 
 that he has obtained complete insight into the mind and its 
 workings, but no sane enquirer makes this pretence in any 
 of the sciences. It is enough in any science if we can know 
 part of the truth and if we are justified in believing that the 
 methods and principles which have already obtained this 
 partial success are also the remedy for our present ignorance. 
 
 If this thesis were not denied so frequently, and sometimes 
 so bitterly, there would be excellent reasons for supposing it 
 an innocuous truism, cautious to a fault. In reality, however, 
 there is no such thing as caution in these matters. For it is 
 meaningless to defend psychology without explaining lohat 
 psychology is defended ; and psychologists are so radically and 
 so acrimoniously divided upon the meaning, the scope and the 
 methods of their science that any one who explains his position 
 in detail has to face the certainty of disagreeing with most 
 of them. 
 
 Psychology, we may suppose, seeks to obtain a certain body 
 of knowledge, and such knowledge is about the mind. This 
 inevitable supposition, however, is dangerous enough to ruffle
 
 viii] THE MIND 151 
 
 the waters of itself. For what is the mind, and how can we 
 know what it is? If psychologists were agreed in their general 
 opinions on these matters, they might cheerfully leave the 
 answer to the detailed results of psychological investigation. 
 In fact, however, they are not at all agreed upon these funda- 
 mental questions, either before they begin their enquiries or 
 after they have finished them; and so it is idle to expect 
 that psychological problems will solve themselves by patient 
 psychologising. Too many false starts are possible, and there 
 is not even a consensus of opinion among the experts to guide 
 us in determining which of these starts are the false ones. 
 
 It has always been recognised that human beings are mind- 
 bodies, or (if the reader prefers) body-minds, and again that 
 the empirical study of the mind trenches upon the metaphysics 
 of the soul. At the present time, however, the trend of 
 speculation has profoundly altered the perspective of psy- 
 chology. It is not merely that souls have gone out of fashion. 
 Psychologists, rightly or wrongly, seldom took souls very 
 seriously. What has gone out of fashion is traditional psy- 
 chology itself. The doctrine of evolution has conquered biology 
 so thoroughly that biologists can afford to be critical of it. 
 Psychologists, on the other hand, write as if they were thrilled 
 with the novelty of the notion, and as if they were bound to 
 accept it with the faith of a little child. Armed with this 
 confidence they seize the flail of evolution, demolish the cob- 
 webs of theology, stifle philosophy in the filmy ruins and then, 
 with the lust for destruction hot within them, turn and 
 belabour consciousness itself. The mind of a man may seem 
 very wonderful but, look you, it has the pedigree of the ape. 
 And a man's ancestry is the stuff of him. The continuity of 
 the germ-plasm has seen to that. Therefore man is the ape's 
 brother under his skin, just as Judy O'Grady is sister to the 
 colonel's lady. Our simian ancestors, it is true, have unfor- 
 tunately perished (I think we developed too fast for the 
 truth of these theories) and so we should look to the gorilla 
 and the chimpanzee, or better still, simplify our theorv by 
 declaring that the most prominent characteristics in the higher 
 animals are also most fundamental in mankind. The higher
 
 152 THE MIND [ch. 
 
 animals are dominated by instinct. Therefore man is also a 
 bundle of instincts. And when this stage of the argument is 
 reached it seems a pity to stop. Animal intelligence is a risky 
 thing to think about, for animals are not introspective and 
 do not connnunicate with us. Let us therefore renounce 
 introspection, and then we shall know where we are in psy- 
 chology. Instincts, again, are conscious processes and conse- 
 quently elude the biologist and his methods. Let us therefore 
 deny the existence of consciousness and study behaviour 
 instead. When we track behaviour behind the reflexes down 
 to the simplest conceivable response, we shall then, at long 
 last, turn our backs upon superstition and discover what the 
 mind really is. Thus all will be ad maiorem naturae gloriam. 
 
 Those of us who believe that Newton did a little thinking 
 before he wrote his Principia, and that his protoplasmic 
 ancestry was incapable of anything of the sort, are not likely 
 to be impressed by these rhapsodies. If we were pressed for 
 an argument, we might reply that continuity of development, 
 interpreted in this preposterous sense, cuts both ways. If 
 Newton cannot be more than his pedigree, his pedigree must 
 be at least as good as Newton. That being so, it is not at all 
 unscientific to take the mind of man as we find it. The 
 question is. What do we find ? 
 
 In our own persons, at all events, we find consciousness ; and 
 if our theories deny consciousness to other people, so much 
 the worse for our theories. What then is this conscious- 
 ness? 
 
 The only possible answer to this question, it is plain, is 
 just a description of fact. Colour is proved to exist when it 
 is pointed out, and not otherwise. And so with consciousness. 
 We are aware of our own consciousness, and we can usually 
 detect the signs of consciousness in others. When a man, as we 
 say, regains consciousness after chloroform, he feels, perceives 
 and attends where he could not do so before. The evidence 
 in these cases, it is true, is sometimes deceptive. The ' ether 
 cry ^ has been heard in cats whose cerebral hemispheres have 
 been removed, and these pitifully mutilated creatures have 
 been known to show an anger-mimesis though never the
 
 viii] THE MIND 153 
 
 pleasure-mimesis of purring ^ This occasional uncertainty, 
 however, proves nothing. It is impossible to prove that no 
 Englishman enjoys roast beef just because we cannot always 
 be certain whether our guests like their dinner or not. 
 
 It seems clear, however, that something more than bare 
 feeling is needed if consciousness can be said to exist in the 
 usual sense. Etymology, indeed, is a perilous guide, and so it 
 may be irrelevant to point out that the word's history implies 
 a certain togetherness of experiencing in co?isciousness. On the 
 other hand, consciousness, in ordinary usage, implies a certain 
 organisation of experiencing and at least a hint of memory. 
 The total absence of memory from moment to moment in 
 cases o^ petit mal inclines us to deny consciousness altogether; 
 and the twilight sleep under scopolamine is doubtfully con- 
 scious precisely because the patient's retentiveness is so re- 
 markably fugitive. 
 
 On the other hand, it is easy to exaggerate the minimum 
 of togetherness which consciousness involves. The unity of 
 consciousness, it is true, has been a favourite theme with 
 philosophers. Kanfs account of the unity of apperception, 
 for example, is a pre-requisite of the possibility of scientific 
 experience. At a humbler level, again, it is fair to point out 
 that there cannot be disappointment unless the disappointed 
 person is also the person whose hopes have been frustrated, 
 and that there cannot be inference unless one and the same 
 mind is aware of the premises and draws the conclusion. 
 Unity of this kind, however, is far closer than the minimum 
 required for consciousness to exist. A dog's consciousness, or 
 a child's, and sometimes a man's, is much more loosely united 
 than this. True, there is some togetherness of experiencing 
 in any consciousness, and a mens rnomentanea is not a mind 
 at all; but there need not be very much togetherness. 
 
 There is still less justification for the view that conscious- 
 ness implies self-consciousness, or that this togetherness of 
 experiencing must not only exist, but must also be known 
 to exist, ere consciousness occurs. The primary function of 
 
 1 Sherrington, The Integrative Action of the Nervous System, pp. 254-255.
 
 154 THE MIND [ch. 
 
 consciousness is to refer beyond itself, and for the most part 
 consciousness fulfils this function without any nrriere pensee 
 towards the self. No one, therefore, has a right to deny con- 
 sciousness to an animal or to an infant simply because they are 
 probably not self-conscious except in a most rudimentary 
 fashion. Deliberate self-cognition, again, requires a special 
 effort which is seldom made; and very little can be inferred 
 from the circumstance that a man who attends to other things 
 can attend to himself if he chooses, and that he is, for the 
 most part, inattentively self-conscious. 
 
 Many philosophers, it is true, are at pains to point out that 
 consciousness tends to disappear altogether when intense 
 interest fills the sails or utter concentration takes the helm. 
 Self-forgetfulness, they say, is the law of crisis, of spiritual 
 awakening, of deeds of valour in mettlesome emergencies, of 
 love and joy and supreme skill. The wrestler is his tottering 
 opponent, the hunter is the trigger of his Winchester. T^Tiis 
 argument, however, is beside the point; for what is absent in 
 these cases is not consciousness but self-cognition. Archimedes 
 was not unconscious when the soldier killed him, but he was 
 so intent on geometry and so inattentive to himself that he 
 was as defenceless as a penguin. Absent-mindedness of this 
 notable kind is only the mind's absence from practical con- 
 cerns. It is not the absence of consciousness; and there is 
 nothing miraculous in the fact that the mind, in moments of 
 crisis, is often too busy to attend to itself. 
 
 Another argument is sometimes added to this one. Ac- 
 cording to it, we know from experience that attention (and even 
 consciousness) tend to cease when the need for them has gone. 
 The tyro at the pianoforte has to attend to each movement 
 and to each note. The finished pianist attends to the score, 
 and his fingers, as it were, act of themselves. This argument 
 certainly shows that consciousness, regarded as the guide to 
 action, is not always indispensable. We breathe too well to 
 need consciousness, and consciousness is often a hindrance 
 when we have it. Consciousness comes on the scene when our 
 breathing is disturbed; and even then it is more of a nuisance 
 than a help, like an old maid weeping when someone is hurt.
 
 viii] THE MIND 155 
 
 But although this argument may disprove the usefulness of 
 consciousness in certain contingencies, it has no further bearing 
 on the issue, and it has very little bearing even on the con- 
 nection between consciousness and self-consciousness. The tyro 
 need not be self-conscious though he may be. He learns the 
 better if he is not ; and if he is self-conscious when he is skilful 
 enough to dispense with consciousness, his self-consciousness 
 need not do any harm although it may sometimes disturb the 
 routine of action in its perverse attempt to probe into the 
 action''s machinery. 
 
 Although these arguments are designed to belittle con- 
 sciousness, they admit, at any rate, that we know what 
 consciousness is; and anyone who denies this is beyond the 
 pale of argument. On the other hand, it is plausible to urge 
 that our knowledge of consciousness carries us but a little way. 
 Our experiences, we are told, are a motley crew of contingent, 
 evanescent, superficial events. Consciousness is only the iri- 
 descent surface of bodily life, and the best way of studying the 
 mind is to leave consciousness alone. In the last analysis, indeed, 
 consciousness is only a kind of response which a living body 
 makes when it has reached a certain level of integration, and the 
 real business of psychology is to study the nervous system. 
 
 The general outline of the argument which is deduced from 
 these considerations is now so familiar that it can be indicated 
 very briefly. The study of the human organism shows, in the 
 first place, that it is a selective instrument. The eye is attuned 
 to light, the ear to sound. In the second place, this selection 
 from the environment is only the beginning of a process, not 
 the consummation of it. Stimuli are selected, and every 
 stimulus issues in movement. The schematic outline of this 
 process is expressed by the conception of a simple reflex. Such 
 a reflex is the conduction of a stimulus through a chain of 
 neurons, passing through the central system and issuing in 
 movement. What we find in experience, however, is an inte- 
 gration of compound reflexes. We are the creatures of reflex 
 patterns, and the root conception of the whole enquiry is the 
 alliance of certain reflexes, the inhibition of others, and, most 
 important of all, the regulation of alternating reflexes in time.
 
 156 THE MIND [ch. 
 
 Reflexes reinforce one another when a succession of weak stimuh 
 in the same region, or a number of stimuli from different 
 organs, debouch simultaneously on a final common path and 
 lead to a single strong thrust. This implies, of course, that 
 there is inhibition of antagonistic reflexes. If all the reflexes 
 were stimulated together we should have the strenuous impo- 
 tence of tetanus or strychnine poisoning. Most reflex patterns, 
 however, are alternating. Food is first chewed, then swallowed. 
 Walking is the alternating contraction of the flexor and the 
 extensor muscles. And so on. Coordination in time, therefore, 
 is the fundamental requirement for the integration of reflexes, 
 and this, we may suppose, is the distinctive office of the 
 central nervous system ^ 
 
 This theory explains much more than consciousness. The 
 movements of developed animals are integrated through the 
 nerves and the central nervous system, and this ample kingdom 
 includes unconscious reflexes like breathing, and subconscious- 
 ness at all its levels, as well as our intermittent consciousness. 
 It would be highly illogical, however, to infer on this account 
 that consciousness is an otiose affair. Even if the same kind of 
 work, broadly speaking, could be done without consciousness 
 as with it, it would be very unlikely, in terms of the argument, 
 that the special work of consciousness could be done quite so 
 well, or quite in the same way, if consciousness were absent. 
 Reflexes may be integrated through the spine or the bulb, but 
 the hemispheres, we are bound to suppose, are better than the 
 spine or the bulb for certain kinds of response; and if con- 
 sciousness, as the argument indicates, is the highest level of this 
 integrating process, we must conclude that it is better for cer- 
 tain necessary purposes than any infra-conscious integration. 
 
 The argument, indeed, is usually developed along these 
 lines. Sometimes, it is true, the consciousness accompanying 
 a reflex seems merely to register an occurrence in our bodies. 
 The sneeze goes off, and we feel it going off, and that is all. 
 Even in this case, however, consciousness is not really inert, 
 for we can delay the sneeze if we try; and in the general case 
 consciousness does not seem to be merely a spectator of an 
 1 The outlines of this account follow Sherrington, op. cit.
 
 viii] THE MIND 157 
 
 organic disturbance. It seems to play its part in the process, 
 and the evidence suggests that it really does so. When we 
 distinguish between conscious and unconscious response, the 
 chief differences seem to be the presence of pain and pleasure, 
 of the one part, and an extension of the range of response, of 
 the other part. The biological utility of pain and pleasure 
 has been noticed so frequently that it is needless to dwell upon it 
 here. Pain is the signal of utter need, sometimes mischievous, 
 but salutary on the whole. And the extension of the range of 
 response which we find in conscious reactions is even more 
 striking. The skeletal muscles (which are the voluntary ones) 
 are specially connected with the senses which have the longest 
 range, and all the evidence shows that consciousness is a pre- 
 venient thing, anticipating movement, and permitting more 
 delicate adjustments When the peril is near, it is true, there 
 is no time for consciousness. The eye, unless it is a baby's eye, 
 closes quicker than the branch that meets it, and we marvel at 
 our cleverness (if we are wise) apres coup. But when there is 
 time we need all the wits we have. The cat stalks the mouse 
 and crouches before it springs. 
 
 If this statement of the case confined itself to the primary 
 biological function of consciousness, there would be no occasion 
 to dispute it. The dispute begins with the magisterial an- 
 nouncement that the primary biological function of con- 
 sciousness is all that consciousness is; and this announcement 
 is a mere non sequihir. To say that consciousness helps the 
 nervous system and that, in certain selected cases, it does the 
 same kind of work as the nervous system might do without it, 
 is not even the beginning of a proof that consciousness is only 
 a species of nervous process. It is highly important, for 
 example, to show that there is a certain continuity between 
 lifeless and living things, but such arguments can never prove 
 that life itself is not an emergent, radically novel in comparison 
 with its antecedents, although requiring these and using them 
 for its own ends. Similarly there is nothing against believing, 
 and a great deal in favour of believing, that consciousness is 
 an emergent, and not simply a modification of antecedent 
 ^ Cf, Sherrington, op. cit. especially Lecture IX.
 
 158 THE MIND [ch. 
 
 nervous processes. If so, a certain continuity of function with 
 the nervous systeni and a certain solidarity with it is not to 
 the point, for that may be admitted without touching the 
 argument; and it would even be permissible to argue that 
 consciousness, beginning as humble menial, gradually became 
 the governor and even the tyrant of man's life. The thing has 
 been known to happen. 
 
 The statement of this possibility, it is true, is not a proof 
 of its truth, and there are plenty of modish arguments which 
 try to show that the encroachments of consciousness are either 
 negligible or non-existent. Man, we are told, is a bundle of 
 instincts and the consciousness of instinct is buried in the 
 nervous system. Instincts are more subconscious than con- 
 scious; and it does not matter in principle which they are. 
 
 The discussion of this theory in detail would need a long 
 argument. Here it must suffice to say that if an instinct be 
 defined in the usual way as a racial habit, relatively little 
 educable, and serviceable, on the whole, after very little ex- 
 perience on the part of individuals, then man is the least 
 instinctive of animals. Human beings acquire most of their 
 habits, and it is foolish to argue that 'direct action,' or the 
 habit of drawing cheques, or the movements of armies, are 
 strictly comparable to a moorhen's instinct for diving, or to 
 the weary journey of Fabre's caterpillars^ when the thread 
 they had spun had been made a closed circle instead of a trail 
 to lead them homewards. Of course it is easy to exaggerate the 
 fixity of instinctive routine, on the one hand, and the reflec- 
 tiveness and initiative of human behaviour, on the other; but 
 human behaviour, whether in society or in the cloister, is 
 reflective, adaptive and opportunist to an extent which no 
 other species of animals can match. Those who deplore the 
 herd-like irrationalism of the masses cannot have listened to 
 working men discussing their own proper business, and they 
 ignore the extraordinary adaptiveness of modern society. Ten 
 years ago, no one would have predicted that the British 
 working man would show a business-like respect for ration- 
 
 1 For a short account see the Essay in Fabre's The Wonders of Instinct, 
 EngUsh translation.
 
 viiij THE MIND 159 
 
 cards, but when ration-cards were needed the very children in 
 the streets clung to them with the tenacity of limpets. When 
 Florence Nightingale tried to induce the soldiers in the Crimea 
 to save some of their pay, Lord Panmure said bluntly that the 
 British soldier was not a remitting animal. The British soldier 
 became a remitting animal as soon as he saw the need for it. 
 
 These considerations cannot be discounted by exploiting 
 biological arguments. Certainly it takes a very cool philo- 
 sopher not to see the necessity that a man must live, but 
 arguments concerning the biological utility of consciousness, 
 however important they are, cannot be conclusive of themselves. 
 Civilised man must be cunning enough to avoid motor cars 
 and runaway horses, and in that respect he is on a par with 
 the animals; but he lives the better if he has his children 
 vaccinated and his house disinfected, if he pays policemen to 
 protect him, and sees to it that those who bring him his food 
 in ships are commanded by expert navigators. There is more 
 biological utility in the discovery of salvarsan or chloroform, 
 in the sextant, and in the town's water-supply, than in all the 
 instincts of fear and flight and protective mimicry put together. 
 Granting that life must be preserved and fostered, intelligence 
 at the helm is woi'th a whole cargo of instinct. If these per- 
 versely narrow biological theories were right, the lion and the 
 tiger would rule the world, and man, bereft of fire and ships, 
 of clothes and gunpowder, would cling forlornly to some tree- 
 top, cursing the loss of his tail, and gibbering morosely at the 
 di'yness of the nuts within his reach. 
 
 The moral of these arguments is that any psychology worth 
 the name must take knowledge very seriously indeed. Man, 
 to be sure, is a psycho-physical being sprung from lowly 
 origins, and there should be no dispute concerning the im- 
 portance of his muscular apparatus or of his nervous system. 
 On the other hand, we must frankly recognise his intelligence 
 as we find it, either in the subtle theorising of Laplace or in the 
 humbler workings of a skilled artisan's mind. Even if a theory of 
 consciousness could be devised which accounted for perception 
 and instinct with reasonable completeness, this theory would 
 be only the beginning of psychology and not the end of it.
 
 160 THE MIND [ch. 
 
 Every theory of the type we are now considering tries to 
 explain what consciousness is by describing what it does; and 
 so it is important to point out that even a complete account 
 of the functions and of the effects of consciousness would be 
 a very poor substitute for a description of consciousness itself. 
 Even if our consciousness always did the same kind of work 
 as the nervous system it would not therefore be the nervous 
 system ; and even if it were the eye wherewith the universe 
 beholds itself there would still be a question concerning the 
 kind of eye which the cosmos selected for this purpose. Both 
 the naturalistic and the idealistic theories of consciousness, 
 however, fail to notice this important point, and their failure 
 is worth examining. 
 
 The naturalists, as we have seen, start from the nervous 
 system and maintain that consciousness is just the central 
 part of a delicate neural adjustment comprehensively organised. 
 What is more, they hold that consciousness makes precisely 
 the same selection as the body does. Its material, they say, is 
 just what the sense organs select, and consciousness, so far from 
 being distinctive and peculiar, is simply a certain selection 
 from things and a certain arrangement of them. Consciousness 
 has no peculiar stuff in it, and the same things may be either 
 conscious or not. Bells are associated with books and candles. 
 That is one of their relationships. They also agitate the sur- 
 rounding atmosphere when they are struck by a clapper. That 
 is another relationship which they commonly have, and the 
 truth is that out of the infinite variety of relationships which 
 bells have, one set or pattern is called physical and another is 
 called psychical. The same bell appears in both the patterns. 
 To say that there is consciousness is to say that a thing is 
 selected and organised in a certain way ; and that is the whole 
 mystery. 
 
 A more ancient and more magnificent theory of the same 
 type states that the function of the mind is to be a microcosm 
 of the universe. The mind is a chameleon of the cosmos perpetu- 
 ally mirroring the totality of things in its subtle, tiny trans- 
 lucence. It is needless to cite authorities in support of this 
 conception since it is, on the whole, the orthodox opinion of the
 
 viii] THE MIND 161 
 
 classical tradition in philosophy, but I cannot resist the 
 temptation of quoting from an author who is but little read 
 since I do not see how the point could be better put. In his 
 Synthetica^ the late Professor Laurie explained that he took 
 the universe to be a whole of things "which find their truth 
 in the last term of a continuoifs and unbroken system ; that 
 is to say, as presented to conscious subject which makes its 
 appearance in the evolution of the world-organism, for the 
 mere purpose (so to speak) of gathering up the universal record 
 into itself as that record is therein written; man himself being 
 the concluding chapter of that record — the individual into 
 whom the whole is poured^" " Colour,"" hesaid again, "demands 
 me for its own purposes. Colour and I are fellow creatures in 
 the same related system, helping each other's full reality out-." 
 And once more: "If I might indulge in rhetoric I would even 
 say : The natural world of flowers and stars might be regarded 
 as waiting patiently for the emergence in the system of a con- 
 scious entity that they might fully realise themselves. The 
 said consciousness, however, adds nothing to what they truly 
 are, save the awareness in feeling of what they ti'idy are. And 
 you might even imagine a dim thrill of joy in the star- world 
 when a conscious subject first beheld them in their reality, and 
 again when Copernicus and Newton revealed their ordered 
 motions. The stars then sing together. Any other view is, to 
 my mind, crude dualism ^^ 
 
 The objection to these theories is that they deal only with 
 a part of the problem. Granting that consciousness is the 
 means which the universe has chosen for self-revelation (it is 
 common to choose one's biographers oddly), we have still to 
 consider precisely what this chosen means is, since the universe, 
 for aught we know to the contrary, might have selected very 
 different means. It is very important, to be sure, to try to see 
 consciousness in its true perspective, and therefore to show its 
 relations to its bodily conditions on the one hand, and to the 
 putative needs of the universe on the other. But that is an 
 account of the setting of consciousness, not of its character,^ 
 
 1 Synthctica, vol. i. p. 79. '^ Ibid. p. 92. ^ j^jV/. p, ng. 
 
 L. 11
 
 162 THE MIND [ch. 
 
 and the only way of discovering this character is just to 
 observe consciousness itself. 
 
 Such observation is introspection, and common sense is fully 
 convinced of the feasibility and reliability of that process. 
 The plain man knows from his own experience what pleasure 
 or sorrow or excitement feels like. Indeed, he knows these ex- 
 periences infinitely better than he knows his brain or the 
 needs of the universe. He has only a hearsay knowledge of 
 grey matter or the Fissures of Sylvius, and he does not dare to 
 guess at the needs of the universe; but he knows directly in 
 his own person what joy or pity or desire is. Introspection, to 
 be sure, is not infallible. There is no pontifical magister'mm 
 about it, and it is easy, I daresay, to set it too many riddles, 
 and to base ex cathedra encyclicals upon it which have no 
 better warrant than inclination or theory. These admissions 
 however do not affect the issue as the plain man understands 
 it. The character of his consciousness, he thinks, is not hid 
 from him; for he can observe it introspectively. 
 
 This point is disputed, however, and, clearly, one of the most 
 pertinent objections to it is the theory that consciousness 
 is not really a distinctive, peculiar thing. Those who main- 
 tain this view challenge the plain man's interpretation of 
 introspection, and hold that the facts which his introspection 
 reveals can be studied better in another May. 
 
 We have already seen the main outlines of this theory, and 
 now we may consider it more narrowly. Up to a point it is 
 eminently successful, since it criticises most effectively the 
 natural, reverend and most mistaken doctrine which asserts 
 that consciousness consists of an inner world of mental images 
 and sensations mirroring the outer world of things. Berkeley, 
 it is true, exploded this fallacy long ago; but superstitions 
 linger, and the inverted Berkeleianism of the American 'new 
 realists' is at least a most interesting experiment in exorcism. 
 
 As we have seen in earlier chapters, the principal objects of 
 our consciousness are things perceived, remembered, or imaged, 
 together with principles and universals; and the analysis of 
 the American new realists is fully in accord with our con- 
 clusions concerning most of these objects. The things we
 
 viii] THE MIND 163 
 
 perceive may also be remembered, expected or imaged, and we 
 have seen reasons for believing that precisely the same things 
 may be studied in the science of physics. Different men, it is true, 
 make different selections from perceived reality, and the same 
 men make different selections from things according as they 
 perceive, remember, or image them. It does not follow, however, 
 that perceived things or imaged ones belong to an inner mental 
 world. We may grant, then, that if consciousness were a name 
 for the 'inner"* world, and if the 'inner' world consisted of 
 sense-data, images and the like, then, in all probability, this 
 'inner' world would be only a selection from the stuff which, 
 otherwise selected, is the 'outer' or physical world. This 
 theory, it is true, might have greater difficulties in explaining 
 the status of universals and of general facts, but it would be 
 sufficient for the others. Sense data are not timeless or colour- 
 less or unextended; they do not have a peculiar non-physical 
 temporality or spaciousness or colour; and if introspection 
 were the observation of sense data and images, it might, indeed, 
 perform a useful office but it would not differ in kind from the 
 observation of physical things. Its results, therefore, might be 
 attained more simply and more fruitfully in some other way. 
 It is a great thing to have avoided this confusion between 
 'inner' and 'outer,' and yet it is not enough, for these perceived 
 and imaged things which confront our consciousness are not 
 themselves consciousness at all. When I see a blue sky on a 
 winter's day and notice sadly how different the trees are from 
 their budding greenness in spring, the blue is not conscious of 
 the green, or the green of the blue, and the principle of differ- 
 ence is not conscious of either. It is I who am conscious, I who 
 apprehend the green and the blue and the difference. My 
 consciousness is not a character of the things I observe or think 
 or imagine, and it is very doubtful whether it is ever written 
 on the faces of the things I know. My consciousness is my 
 awareness or apprehension of these things, together with the 
 feelings and strivings which accompany my apprehension. 
 Nothing, surely, can be plainer than this, and yet stress must 
 be laid upon it because it is neglected in so many theories of 
 knowledge. 
 
 11—2
 
 164 THE MIND [ch. 
 
 We may say with Dr Strong that these theories of the 
 American realists neglect the fact of givenness^ When a thing 
 is given to the mind it does not therefore acquire a new and 
 peculiar mode of being; but things may exist without being 
 given, and it is useless to argue about the given without ad- 
 mitting the ultimate fact of givenness. Or, again, we may say 
 that it is one thing for an object to exist and quite another 
 thing for that object to appear. When a thing appears, it need 
 not appear otherwise than it is; but it cannot appear unless 
 there is awareness of it, and it can exist without that. When 
 it appears, moreover, it must appear to something, and this 
 something must be or contain awareness or consciousness. The 
 theory of the American new realists, in a word, deals onlv with 
 the objects of consciousness, and so it is not a theoiy of con- 
 sciousness at all. 
 
 Here is the fatal gap in the theory, and the point, perhaps, 
 may become clearer when it is approached in another way. 
 Dr Ward's Psychological Principles is rightly regarded as the 
 chief systematic work on psychology which any living English- 
 man has produced. We may therefore consider his analysis of 
 the mind^. 
 
 According to Dr Ward, psychology is the study of individual 
 experience^, and experience itself is the commerce between 
 subject and object*. In the next place, he distinguishes the 
 'objective"" from the 'subjective' aspect of experience'. The 
 objective aspect of experience, he maintains, consists of sense 
 data, images, and the like — in a word, it consists of presenta- 
 tions^. The subjective side consists of feeling and attention '^. 
 Dr Ward interprets 'feeling' very narrowly, for he restricts it 
 to pleasure and pain^, and he uses 'attention' very broadly, for 
 he means by it any sort of perceiving, inferring, desiring or 
 striving, and even what we should usually call inattention. 
 
 1 The Origin of Consciousness, chap. i. pp. 31 sqq. 
 
 - The reader will see that I differ from Dr Ward on a great many points. 
 My object is to elicit certain conclusions from his analysis of psychology, and 
 neither to accept nor to criticise his view au pied de la lettre. 
 
 3 Psychological Principles, p. 28. * Ibid, chap i. especially § 3. 
 
 5 Ibid. pp. n sqq., SO sqq. 6 j^^^^^ pp^ 46 sqq. 
 
 7 Ibid. e.g. General Analysis, pp. 55 sqq. ^ Ibid. p. 45.
 
 viii] THE MIND 165 
 
 According to him there is attention whenever we are active 
 enough to receive impressions \ 
 
 It is clear that the analysis of the American new realists 
 deals Avith what Dr Ward calls the objective aspect of ex- 
 perience, and omits the subjective aspect altogether. Anyone, 
 indeed, who maintains that there is no distinctive stuff of 
 consciousness has to argue that the subjective side of ex- 
 perience is really part of the objective side. Let us see then, 
 what these philosophers say. 
 
 They begin with the argument that feeling is a sensation 
 and therefore in pari materia with other sensations ; and they 
 can claim distinguished authority in support of this contention. 
 Indeed, we should all agree with Stumpf- in believing that 
 bodily pain and bodily pleasures, like the comfort of a warm 
 fire or the smoothness of underwear, are sensations; and perhaps 
 we should follow him in his further contention that the 
 pleasures and pains of the special senses are very often sen- 
 sations. Toothache and nausea, for example, are organic 
 sensations, and the ache of the one and the diffused disagree- 
 ableness of the other seem to belong to the tortured organism. 
 The pleasures of sight, again, are blended with the smoothness 
 of ocular adjustment and that is an affair of kinaesthetic 
 sensation. There is a bodily resonance, indeed, in all emotion, 
 and the most refined delight has its own thrill of organic 
 harmony. On the other hand, our delight in the neatness of 
 an argument or in the point of a jest is not merely the smile, 
 or the laugh, or any other bodily accompaniment. The bodily 
 movements, indeed, may be repressed (perhaps by a rather 
 painful effort), and the delight still be felt. Such delights, 
 therefore (and the pains of remorse or failure), are certainly 
 not sensations, even if organic sensations accompany them; 
 and therefore the theory falls. 
 
 Indeed, there is no difficulty in distinguishing mental pains 
 and pleasures from the pleasures and pains of sensation. The 
 latter are organic sensations which are localised within the 
 
 1 Psychological Principles, p. 49 and p. 60. 
 
 2 Zeitschrift fur Psychologic, xliv. 1906, 1 sqq. Cf. Titchener, The Psy- 
 chology of Feeling and Attention, Lecture III.
 
 166 THE MIND [ch. 
 
 body. The former are not. They have no habitation in muscle 
 or eye or ear, because they are not bodily at all. 
 
 Passing, then, from feeling to attention, we may ask whether 
 attention belongs to the objective aspect of experience. The 
 effect of attending to anything, it is generally agreed, is to 
 increase the clearness of that thing and also to secure a certain 
 dominance for it. Attention lays hold on a thing, brings its 
 outlines into relief and tends to keep irrelevant things from 
 appearing. There are differences, therefore, on the side of the 
 object according as it is attentively or inattentively regarded, 
 and the problem is whether these differences on the side of the 
 object can be all that attention means. 
 
 This discussion may seem very strange, but the strangeness, 
 after all, may be due to our own misconceptions. It seems 
 absurd to ask whether things themselves are clear or obscure, 
 because we always suppose that clearness or obscurity depends 
 upon us. We are fully convinced that things themselves can- 
 not differ in point of clearness, and therefore infer that these 
 differences in clearness are manifestations of our activities. 
 Still, this interpretation, spontaneous and inevitable as it 
 seems, may be mistaken. To parody a famous saying of 
 Hume"'s, " Wlien we exclude consciousness, we really do exclude 
 it." Is it not possible that this enhanced clearness, steadiness 
 and dominance in the objects of attention is all that attention 
 is, and that the attentive consciousness which we take to be 
 the cause of these characteristics is only a myth.'* 
 
 In point of fact we should have no business to infer that 
 these differences in clearness and the rest were due to attention 
 unless we were acquainted with attention itself. If the attentive 
 process can be observed it is possible to apply the usual logical 
 methods and to infer that clearness results from attention 
 because it increases as attention increases and is absent when 
 attention is absent. Without this acquaintance the hvpo- 
 thetical cause of the increase in clearness might be anything 
 under the sun. The plain man's certainty, then, is due to the 
 fact that he is directly acquainted with attention itself, and 
 not merely with its effects. But since this point is disputed I 
 prefer to approach it gradually.
 
 viii] THE MIND 167 
 
 As we have seen, Dr Ward takes 'attention'' to include per- 
 ceiving, inferring, desiring and striving. This is not a common 
 use of the term, and it is not a very good one. But that may 
 pass. The point which I wish to put before the reader is the 
 difference between these varieties of 'attention.' No one denies, 
 I suppose, that perceiving is a different process from inferring, 
 and that both perceiving and inferring differ from desiring. 
 To infer an ecHpse is very different from seeing one. It is one 
 thing to see the Kaiser hanged and another thing to desire 
 his execution. 
 
 My question is: How do we know these differences? We are 
 as certain of them as of anything in the world, and there must 
 be some explanation. What is it, then.-^ 
 
 Plainly, the certainty cannot be due to any difference in the 
 objects. There is generally a difference in the objects, it is 
 true, when there is a difference in the mental experience. We 
 do not perceive what we desire. The fox saw the grapes, but 
 he desired to eat them ; and eating is not seeing. No one, again, 
 is dazzled by an inferred corona. On the other hand, some of 
 these differences are independent of any differences in the 
 objects. If one man tells me that Yorick is dead and another 
 questions the statement, both of them refer to the death of 
 poor Yorick, but one of them believes it and the other doubts. 
 It is clear, I think, that the difference between believing and 
 merely supposing, or again between desiring and inferring, 
 can be readily discerned by inspection; but if the reader jibs 
 he may be invited to consider what happens when we doubt 
 and when we believe precisely and numerically the same thing. 
 
 It is quite certain, therefore, that some of the differences in 
 what Dr Ward calls the 'subjective aspect of experience'' do 
 not belong to the 'objective aspect,' and surely there is no need 
 to prove that doubting, supposing, believing, and the like, are 
 conscious processes and that their differences are diffei'ences in 
 consciousness. Any theory, therefore, which identifies con- 
 sciousness with the objects of consciousness has failed in most 
 elementary fashion. These objects of consciousness are not 
 consciousness at all, except in the special case of introspection. 
 Consciousness is the awareness of them, the striving for them,
 
 168 THE MIND [ch. 
 
 the joy in them, not the things striven for, apprehended or 
 enjoyed. 
 
 This finding changes the course of the debate. Conscious- 
 ness, it appears, is different from the objects of consciousness, 
 and we are directly acquainted with it. The problem is there- 
 fore, how we are acquainted with it, and what we learn from 
 this acquaintance. 
 
 The simplest answer to the first of these questions is Locke's. 
 The mind can notice its own operations. This simple answer, 
 I think, is the right one, and I propose to defend it; but it is 
 contradicted so flatly by so many eminent philosophers that 
 its defence, unfortunately, is rather a lengthy business. For 
 some maintain that we do not observe in this way, and others 
 that we do not need to do so, and others that we could not if 
 we would. 
 
 The first objection, I think, is contrary to fact and based 
 on misconception. We can and do observe our own conscious- 
 ness, and anyone who doubts this statement may perhaps see 
 the truth of it when he considers the alternatives. We have 
 seen already that our knowledge of our consciousness cannot 
 be an affair of inference. If it were, there would not be any 
 radical difference in experience between our knowledge of our 
 own anger and our knowledge of the Apostle Paul's. In fact, 
 however, we are directly acquainted with our own anger and 
 not with the Apostle's, and our acquaintance with our own 
 consciousness must depend upon observation unless it is a 
 different species of awareness from any other. For knowledge 
 is either observation or inference. 
 
 As might be expected, therefore, we have to meet the 
 argument that our acquaintance with our own minds is 
 altogether sui generis. This conclusion, we are told, is inevi- 
 table, since observer and observed are one in the case of 
 introspection. We have to observe a saucepan in order to be 
 aware of it, but we do not need to observe our anxiety in order 
 to apprehend it. Anxiety is a mode of consciousness, and we 
 are conscious of anxiety whenever we are anxious. There is no 
 difference between the consciousness and our awareness of it, 
 since consciousness is itself awareness.
 
 viii] THE MIND 169 
 
 This argument is thoroughly fallacious. In the first place, 
 anxiety and the awareness of anxiety are not the same; for 
 awareness is a kind of knowing and anxiety is an emotion. At 
 the best, therefore, this argument would hold only of the 
 awareness of awareness, and not of the awareness of emotion 
 or striving. In the second place, it is abundantly clear that 
 attention to our own minds is only an occasional process which 
 need not be very efficient. Some psychologists even affirm that 
 attention to ourselves always disturbs the current of our con- 
 sciousness, and TurgeniefF went further, for he said: "When 
 my sufferings are unendurable I follow Schopenhauer's advice. 
 I analyse my sensations, and my agony departs for a period." 
 Be that as it may, it is plain that attention to our consciousness 
 is not the same thing as being conscious. When we are con- 
 scious we are usually aware of our consciousness at least dimly, 
 but we need not always be ; and we have to attend very hard if we 
 wish to discern the features of our consciousness accurately. 
 What is more, we often make mistakes in this enterprise. 
 
 This obvious reflection, however, is often forgotten. M. Berg- 
 son forgets it in his theory of intuition^ and all the mystics 
 forget it when they argue that knowing and being are one in 
 the case of the self. Now it is true that the self can behold 
 itself, but this self-observation is never the identity of observ- 
 ing and being. If it were, how could there be any occasion 
 for attending to ourselves ? We cannot help heing ourselves; 
 and if our conscious being were identical with this knowing, 
 we should always have a complete answer to all psycho- 
 logical questions through the mere fact of existing. The 
 mystics and the intuitionists, indeed, can give no reason for 
 the pains and labour which they think intuition requires. 
 They tell us that anyone who would learn of them must make 
 an unusually resolute effort to sink into himself: but if being 
 conscious and knowing one's consciousness are one and the 
 same, where is the need for this effort .'' We always are what 
 we are without anv effort whatever, and we do not have to 
 struiiiile in order to become ourselves. 
 
 1 Introduction to Metaphysics, passim.
 
 170 THE MIND [ch. 
 
 Granting, then, that consciousness and the awareness of 
 consciousness are not the same, it remains to consider whether 
 this awareness of consciousness is a kind of observation. If it 
 is, it has to dispense with eyes and ears, but that is no ob- 
 jection. Indeed, to cut a long story short, the difficulties that 
 have to be met are really a priori difficulties. Berkeley has 
 told us that there can be no awareness of activity^, and the 
 Kantians, following their master's lead% assert that there can 
 be no awareness of awareness. 
 
 Berkeley's argument rests on confusion. We can certainly 
 observe conscious processes as they occur, and this is the deter- 
 mining circumstance here. Activity, to be sure, is not observed 
 if it is taken to mean the hidden spring of change, the mystery 
 that makes motion move, but activity in that sense is a will 
 o' the wisp like the 'force' of unreflective dynamics. On the 
 other hand we can certainly observe our consciousness playing 
 its part in bodily adjustments and in subsequent conscious- 
 ness; and the experiences of striving, willing, and the like, 
 may well be called active since they are peculiarly bustling and 
 purposeful. Such consciousness, then, 'has hands and feet,' in 
 the classic phrase, and we can see it at work^ 
 
 We are left, therefore, with the celebrated dogma that 
 there can be no awareness of awareness. The principal argu- 
 ments in favour of this contention seem to be thi'ee, and we 
 may consider each in turn. It is argued, imprimis, that such 
 awareness is never a fact of experience; deinde, that whatever 
 we know is an object, so that if consciousness is made an object 
 it is therefore transformed utterly ; adhuc, that knowledge 
 always refers beyond itself, and consequently that it cannot 
 refer to itself. 
 
 " I have to confess," Mr Russell says, "that the theory which 
 analyses a presentation into act and object no longer satisfies 
 me. The act or subject is schematically convenient, but not 
 
 empirically discoverable It seems to me imperative, therefore, 
 
 to construct a theory of presentation or belief which makes 
 
 ^ Principles, §27. 
 
 ^ Metaphysische Anfangagriinde der Naturicigsenschaft. 
 
 ^ Cf. James, Essays in Radical Empiricism, pp. 155 sqq.
 
 viii] THE MIND 171 
 
 no use of the subject or of an 'acf as a constituent of a 
 presentation ^^ 
 
 It must be admitted that it is difficult to observe acts of 
 knowledge. Our thinking processes, it is plain, are thoughts 
 of something, attention to something. It is unlikely, therefore, 
 that they could be isolated and set up for inspection by them- 
 selves ; and certainly they are not isolable in this way. None 
 the less, these acts of attention and belief can be discovered 
 empii'ically for the reasons already given. When we consider 
 doubting, believing and supposing, for example, we can observe 
 very clearly that these attitudes may refer to the same object 
 and yet differ intrinsically. This difference is plainly a differ- 
 ence in the character of the mental processes themselves; and 
 we cannot observe these differences in the acts without ob- 
 serving these very acts. Tlie trouble is that when we look for 
 knowledge we often expect to find something more than the 
 process of apprehension, and therefore may be inclined to 
 dispute the existence of knowing just because we find nothing 
 except knowing. And that is not a reasonable objection". 
 
 The objection that the observation of consciousness trans- 
 forms it into an object may be dismissed very briefly. The 
 primary function of knowledge, it is true, is to know something 
 that is not itself, but this does not prove that knowledge itself, 
 or any other form of consciousness, cannot be known. As I 
 have said in another place, " To be directly acquainted with 
 anything, and to be directly acquainted with that thing ' as 
 an object,' express precisely and numerically the same fact. 
 The subject 'as known' or 'as an object' is just the subject 
 itself. If we are acquainted with it, then we are acquainted 
 with it, and no qualification of this statement is permissible 
 
 1 ' Problems of Science and Philosophy ' {Aristotelian Society, Supple- 
 mentary vol. n. 1919), pp. 25, 26. 
 
 2 I assume that Mr Kussell means that the knowing subject is not empiri- 
 cally discoverable. It is nonsense to say that the feeling and striving subject 
 is not. And perhaps I may add that the analysis of act and object has suffered 
 some harm by Mr Moore's description of ' acts ' as ' diaphanous ' (' The Kefu- 
 tation of Idealism,' Mind, N.S., vol. xii. (1903)). The diaphaneity of an act 
 of knowledge, as I understand the description, only means that the charac- 
 teristics of the act do not appear in the object, and not that the act has no 
 observable characteristics when attention is paid to it.
 
 172 THE MIND [ch. 
 
 unless the acquaintance is mistaken, or the word ' object,' for 
 purposes of technical convenience, is defined in some restricted 
 sense. Nothing can be transformed in any sense whatever 
 simply owing to the fact that it is known. To suppose the 
 contrary is scepticism ^"'"' 
 
 The third argument is equally inconclusive. The process of 
 knowledge, it is true, refers beyond itself, and therefore an act 
 of knowledge can never be aware of itself. This fact, however, 
 does not justify the inference that acts of knowledge cannot 
 be observed. Such acts cannot observe themselves, but why 
 should not another act observe them ? Introspection is a de- 
 liberate inspection of awareness, and the empirical evidence 
 strongly suggests that it is always a different act from the act 
 of which it is aware. Our minds, however, are rich enough to 
 contain a multitude of awai'enesses almost at the same moment. 
 For thought is quick. In a word, observer and observed are 
 one in introspection, because the act of introspection and the 
 experience which it observes form part of one and the same 
 mind, but this circumstance does not imply the absurdity of 
 an act of attention attending to itself. 
 
 We conclude, then, that consciousness can be directly ob- 
 served, and this result is of the utmost importance for psy- 
 chology. Psychology is the study of the mind, and the 
 phenomenal mind is just the living continuity of desiring, 
 choosing, perceiving, and similar experiencings. It is these 
 processes in their union, and it is nothing else. What is more, 
 there is no good reason in metaphysics for maintaining that this 
 phenomenal self differs from the real self. It would be other- 
 wise if our acquaintance with it could be discredited ; but that, 
 we have seen, is not the fact. It would be otherwise, again, if 
 any phenomenal thing had to have a noiimenal basis which 
 always eludes observation, but this metempirical nucleus of 
 thinghood is only a sort of transcendental blessing upon a union 
 which God has already decreed without the laying on of philo- 
 sophical hands. Substance is always a descriptive term indi- 
 cating a unity which exists de facto. It does not make the 
 connectedness of properties; it only describes their connected- 
 1 Article, 'Introspection,' Mind, N.S., No. 112, pp. 396, 397.
 
 viii] THE MIND 173 
 
 ness. It is not even an accessory after the fact. Minds are 
 substances simply because desiring, willing, and knowing do 
 not float about loosely. They always unite in a personality, and 
 this united fact is the spiritual substance just as the cohesion 
 of certain molecules is the whole substance of a pie-dish. The 
 self, to be sure, is not the same sort of thing as a pie-dish. But 
 that is another story. 
 
 What objections are there to this conclusion ? 
 
 It may be said, in the first place, that the self, after all, 
 may only be the body. The plausibility of this argument 
 disappears, however, when the argument is sufficiently precise 
 to be worth considering. As we have seen, it is easy enough to 
 show that con sciousness continues the work of the nervous system 
 and even that it does the same kind of work in certain cases. 
 These suggestions, however, are not proofs of identity. Does 
 the brain discriminate, judge or infer, does it choose or resolve, 
 does it feel and enjoy ? To ask these questions is surely to 
 answer them. We know what these processes are because we 
 can observe them, and we know that we could not observe them 
 if we turned our lenses upon the brain after some delicate 
 operation of trepanning. What is more, we know that this 
 consciousness which we observe is different in kind from any- 
 thing that could be observed in the brain. The only reasonable 
 conclusion, therefore, is that consciousness is not cerebral move- 
 ment. If the brain is the coloured, irritable, convoluted pulp 
 that physiologists study, then this quivering indented thing 
 is not the mind ; and to say that it may also be conscious is 
 only a quibble. For anything one can prove to the contrary, 
 some pebble on the side of Ararat may have spent the days and 
 nights of the Flood in working out differential equations, but 
 then it was not the sort of pebble which Dr Johnson kicked 
 or physicists consider. If the brain means what physiologists 
 mean by it, it is not a mind. If not, you may ascribe to it any 
 properties you choose, but it is mockery to call it only a brain. 
 
 It may be said, in the second place, that the work of the 
 mind is far too arduous and too intricate for mere conscious- 
 ness to perform. At the most obvious empirical level, con- 
 sciousness is nothing without memory or retentiveness, and
 
 174 THE MIND [ch. 
 
 consciousness itself, fleeting and evanescent, is not even a 
 permanent condition of its own retentiveness. Here, surely, is 
 a singular argument. If consciousness is really continuous and 
 retentive it cannot also be less retentive than it is ; and if it 
 is fleeting it must be at least as retentive in its fleetingness 
 as we find it to be. Nothing ever accounts for itself, and the 
 only sense in saying that a thing accounts for part of itself 
 is to show the interconnectedness of its parts. It is useless, 
 therefore, to suppose that the self, or anything else, could 
 account for memory otherwise than by having the function 
 of memory deeply implanted in it. And it is plain that 
 memory belongs to the self in this sense. Systematic de- 
 scription is the only possible explanation of such matters. 
 
 A systematic description of the self, it is true, is incomplete 
 in many particulars. For the self is a continuant, and we only 
 observe fragments of it in introspection. In this respect the 
 self is like any other empirical thing. Our conscious lives are 
 infinitely richer than the casual records of introspection; and 
 introspection, perhaps, is not very thorough, even when it is 
 careful. The conscious self, therefore, is very largely an in- 
 ferred thing, but it is not merely inferred, since its principal 
 features and the outlines of its connectedness can be observed 
 introspectively. True, the fragments of consciousness which we 
 observe in this fashion do not do the whole work of conscious- 
 ness. But then they are only parts of a conscious mind. 
 
 There is no difference in principle when mind is taken at 
 its highest level. Kant's proof of the unity of apperception 
 showed, once and for all, how subtle knowledge is ; but even if 
 the most commonplace judgment implied all the principles of 
 logic and most of the categories, that in itself would not show 
 that empirical consciousness is incapable of the work of know- 
 ing ; and Kanfs other arguments on this question make quite 
 unnecessary assumptions. As his most recent commentator 
 points out, Kant's theory implied the consequence (which 
 Kant himself was very loth to accept) that " the activities 
 generative of consciousness have to be recognised as themselves 
 falling outside it. Not even in its penumbra, through some 
 vague form of apprehension, can they be detected. Only the
 
 viii] THE MIND 175 
 
 finished products of such activities, not the activities them- 
 selves, can be presented to consciousness ; and only by general 
 reasoning inferential of agencies that lie outside the conscious 
 field can we hope to determine them^" Kant held, therefore, 
 that the work of thinking was performed by an unknowable 
 and indispensable faculty which he called productive imagina- 
 tion, and his argument has appealed to many philosophers. 
 It rests on the assumption, however, that consciousness re- 
 quires 'generating activities' outside itself, and there seems to 
 be no good reason for denying that consciousness does its own 
 generating. Conscious knowledge is just knowledge at work — 
 a piece of being, strenuously alive. 
 
 It may be objected in the third place that the real and the 
 phenomenal self cannot be identical because personal identity 
 belongs to the real self and because it is not found in our 
 consciousness. There is no identity between our childish esca- 
 pades and our present dignified and important pursuits; and 
 even when childhood is left out of account, there is little identity 
 between Lord Braxfield on the bench and Lord Braxfield in his 
 cups, or between John Newton the pirate and John Newton the 
 hymn-writer. Indeed, it is needless to consider these striking 
 instances, or the still more extreme cases of William Sharp and 
 his feminine personality "Fiona Macleod," or of multiple per- 
 sonality in the Hanna case or the Beauchamp family. The same 
 kind of abrupt variation, we are told, occurs in every life al- 
 though we are too little reflective to rate it at its proper worth. 
 
 There are two questions to consider here, a question of logic 
 and a question of fact. The logical question concerns the 
 meaning of identity. If nothing is identical unless it persists 
 unchanged then, plainly, there is no such thing as personal 
 identity, and there is no identity of body or brain. Neither 
 mind nor body persists unchanged, and it is foolish to speak 
 of their identity in this sense. If there is any pitiful remnant 
 of unchanging consciousness in us, this dubious residuum is 
 certainly not ourselves, just as our bodies are not those scraps 
 of tissue, if there are any, which have never been renewed. 
 
 ^ Mr Norman Smith's Commentary on Kant's Critique of Pure Eeason, 
 pp. 263, 264.
 
 176 THE MIND [ch. 
 
 Personal identity, therefore, is either a mere fiction or else it 
 does not mean unchanging existence. If it is a fiction, it shares 
 the illusion with any other existing thing. If not, identity 
 must be sought, not in unchangeableness of material, but in 
 continuity of character and function. And this brings us to the 
 question of fact. Our bodies remain the same, not because they 
 are composed of the same material, but because they retain the 
 same type of organisation in similar material ; and identity in 
 this sense also holds of personal consciousness. Any self has a 
 certain typical organisation. The emotions are organised into 
 sentiments; inherited habits unite with acquired ones to make a 
 formed will and a formed character; and knowledge is organised 
 too. Our knowledge is complex in its simplest manifestations. 
 It is charged with meaning and servile to a host of principles. 
 It has to look before and after in order to look at all. 
 
 That is personal identity in general. In particular, each 
 self is born different, and acquires its own individual organi- 
 sation. What is acquired may lessen the original differences 
 or it may accentuate them. The army or the public school 
 may standardise men, and a bohemian existence may foster 
 the same sort of imitative eccentricity. On the other hand, 
 each member of the same family or school or nation sets about 
 living in his own. way, and each makes something different of 
 his capacities and opportunities. These personal differences per- 
 sist despite the most radical changes of outlook or opportunity. 
 When a timid sinner is converted, he usually becomes a timid 
 saint; and even when he is made strong out of weakness, his 
 strength has quite a different fibre from the strength of those 
 who were strong before. A man's feelings, again, are not a 
 child's. The man outgrows his childhood, as he outgrows 
 sailor suits and a taste for Henty, but we can see the marks 
 of his childish selfishness in his manhood's considerateness, 
 and we can feel his boyish shyness quivering through his sophis- 
 ticated aplomb. If the man himself is too blind to trace this 
 continuity, his mother is not. ' 
 
 Personal identity, then, is not a superstition. It is a reality ; 
 and the most serious difficulty in it is not the variations 
 of personality, but the recurrent annihilation of conscious
 
 viiij THE MIND 177 
 
 personality in sleep, and its occasional annihilation in a trance 
 or in an accident. That, of course, is the standing argument 
 in favour of materialism. It takes courage to defend conscious 
 identity when it evaporates with a breath of narcotile, and 
 when a brickbat sends it spinning. On the other hand, it takes 
 foolhardiness to deny it. Peter wakes up the self-same Peter 
 every morning. He rises earlier than usual if he has to catch 
 a train, he starts at a burglar's stealthy tread, and he is the 
 same old Peter in his dreams. Perhaps, then, it is unlikely that 
 Peter's consciousness vanishes utterly during sleep, but there 
 is certainly very little empirical evidence of it during profound 
 slumber, and none at all when Peter is under chloroform. 
 Peter's identity, therefore, may have gaps in it, for sometimes 
 there seems to be no Peter at all. And why not ? Peter is Peter 
 when he exists. When he does not exist, there is naturally no 
 Peter. 
 
 Conscious personality, therefore, is not unbroken existence, 
 but it is distinctively individual in a way that is matched by 
 nothing else that we know. Some have supposed, even, that 
 there is an ultimate metaphysical principle to the effect that 
 no part of a self can also be part of any other self; and the 
 facts of experience, with a few dubious exceptions, would 
 cei*tainly support this metaphysical principle. Castor and 
 Pollux may have similar thoughts when they think of the 
 same thing, but Castor's thoughts do not pass into Pollux. 
 All this is commonplace, to be sure. I mention it — apolo- 
 getically — because it is true. 
 
 We are told that this is precisely the principle which all 
 good Platonists are most concerned to deny^ It ossifies the 
 self into a repellent unit, it is perversely ' linear ' where it 
 should also be 'lateral,' it exaggerates the unity of finite self- 
 hood and substitutes a supposititious entity for a description 
 of fact, it loses sight of the truism that the state is more 
 individual than any of its members, and it ignores the funda- 
 mental canon of all true philosophy that nothing can be 
 
 ^ W. K. Inge, ' Platonism and Human Immortality,' Aristotelian Society 
 Proceeding}!, 1918-1919, p. 286, referring to Bosanquet (Proceediiu/K, 1917- 
 1918, pp. i8'2sqq.)., and to his criticism of the present writer. 
 
 L. 12
 
 178 THE MIND [ch. 
 
 individual short of the whole. These are high matters and I 
 must forbear to debate them at length. But I shall hazard a 
 few reflections. 
 
 If the cosmos be indeed a unity whose all-encompassing 
 extent is welded in an indivisible harmony, its members 
 may still have an individual office and themselves be worlds 
 within a world. A monistic metaphysic, on this interpretation 
 of monism, has room for Leibniz as well as for Spinoza, and 
 neither 'good Platonists' nor good Hegelians should take sides 
 on the question. No one claims, however, that mortal man can 
 discern either the cosmos or its members in their fulness of 
 harmonious being; and it should not be at all astonishing if 
 some parts of the whole exhibit a greater degree of unity than 
 other parts or than the whole itself. If conscious selves are the 
 best examples of individuality that we know, neither monists 
 nor pluralists should be surprised. 
 
 The statement that anyone who believes in the reality of 
 finite selfhood therefore considers the self a bare or repellent 
 unit is a piece of scandalous sophistry. Nothing is repellent, or 
 isolated, or shrunken, or dreary, simply because it is itself. A 
 man does not lose his individuality by cooperating with others, 
 or by sympathising with them, and those who worship 'organic 
 unities' have no reason to draw this consequence from their 
 beloved metaphor. Heart and brains and liver do not transfer 
 their characters from one to another. On the contrary, a 
 differentiated organism details its functions to specific organs. 
 There is nothing in our theory to prevent the most strenuous 
 belief in the impossibility of ' self-realisation ' apart from social 
 influences, although, equally, there is nothing to compel this 
 inference unless the facts prove it of themselves. 
 
 It is the same with 'linear' and 'lateral' identity \ The 
 ' lateral ' side of the self apparently consists of everything 
 which personality touches, the lives of others, the fruits of the 
 earth, and anything which a man's interests concern. If our 
 view is correct, this ' lateral ' aspect of personality is no part 
 of the person, but we certainly do not deny that we really have 
 these interests and possessions and influence, or that our lives 
 1 Bosanquet, op. cit. p. 498.
 
 viii] THE MIND 179 
 
 may be spent in their service. A general does not cease to 
 control his soldiers just because he and they are different 
 beings, and those who know Pan and old Silvanus and the 
 nymphs of the field do not therefore become part of the soil. 
 
 We know enough of the self to be able to scan its main 
 features, and to be justified in believing that the hidden mind 
 is of a piece with the mind that is known. Indeed, we know 
 what we are, sufficiently well to be much more confident of the 
 reality of our conscious personality, as a finite discoverable 
 thing, than of the reality of the transfigured selfhood which so 
 many philosophies proclaim. We are not adjectives of the 
 cosmos any more than a dog's tail is an adjective of a dog, 
 and the characteristics which belong to us need not belong to 
 the universe. And, as we have seen, we are not adjectives of 
 the state, even if that institution, by some light-fingered 
 subreption, is first of all identified with the community and 
 thereafter with the Athenian 7roXt<?. 
 
 Those who look at nature from the side of physics commonly 
 incline towards monism, while biologists and psychologists are 
 usually pluralists because of the striking individuality of the 
 things they study. These tendencies, however, need not coerce 
 anyone, and many biologists take the race to be more in- 
 dividual than the organisms which mark its passing, and dream 
 of a world-organism intercellularly diversified into livingthings. 
 Indeed, there is no way of proving that the universe is either 
 fundamentally one or fundamentally many, and realists need 
 not enter into these lists. For the pluralism in which realists 
 believe is a logical pluralism, not necessarily a pluralism of 
 existence. On the other hand, any monism which seeks to dis- 
 credit the empirical unity of empirical things, or to cast doubt 
 upon the possibility of any knowledge unless the knower is 
 also the whole of existence, should be disputed to the hilt. 
 We may be parts of a stupendous whole, but at any rate we 
 are om-selves; and the reality of self-reliance, responsibility, 
 personal freedom and individual judgment are worth fighting 
 for. Realists need not deny that the self is also a part of tlie 
 cosmos and knit with it. But it is at least a self. 
 
 12—2
 
 CHAPTEK IX 
 
 THE LARGER OUTLOOK 
 
 I am very sensible that on such subjects arguments fall short of 
 
 evidence I shall nevertheless go on as I have begun, and proceed, by 
 
 reason, by conjecture, and by authority, to cast the best light I can on 
 the obscure paths that lie in my way. 
 
 Berkeley, Siris. 
 
 Classical philosophy has always striven to include a con- 
 spectus of the achievements of the human spirit. It has dealt 
 with these achievements in the large, of course, and from a 
 standpoint of its own. For philosophy is more analytic, more 
 critical, more synoptic, than any science or art or religion, and 
 it is a cosmology rather than a cosmophany. None the less, 
 philosophy searches all these regions in its own philosophical 
 way. 
 
 It is a common complaint, then, that realism is unfit for 
 the burdens of this office, and so that it is not philosophy at 
 all. It is only a cobbler without large designs, a seeker of 
 trifles, not the spectator of all time and of all existence. It 
 is a temple without a Shekinah, workmanly enough in its 
 outer courts, and, for the rest, empty nothing. It is a poor 
 drudge, stolid, flat, ponderous, obstinate, clumsy and rude. 
 And if these names are too hard, the best that can be said for 
 realism is that philosophy has a place for it, just as art needs 
 realism for its own artistic purposes although, in itself, it is 
 above realism. Realism in philosophy, we are told, has no 
 'central standpoint,' so that its criticism is only a sort of 
 guerilla warfare without any strategy, and its conclusions 
 valueless, except accidentally, since they lead nowhither. It 
 'has little capacity for solving ultimate problems \' And so 
 forth and so on. It is impossible to set a period to these 
 
 1 See e.g. a review in The Times Literary Supplement, Sept. 18, 1919, 
 p. 492.
 
 CH. ix] THE LARGER OUTLOOK 181 
 
 sententious and wary sagacities ; for the patient critics of 
 realism are as contemptuous as they dare, and the others are 
 as contemptuous as they feel. 
 
 Now it must be admitted that many idealistic philosophies 
 have at least proved their mettle by grappling resolutely with 
 the high problems they have set themselves. They have 
 attempted to give a synthesis of art, morals, religion, and the 
 sciences; and they may reasonably claim that intransigeant 
 denials and a few forays in the way of pithy argument are 
 not enough to refute them. Moreover, they may claim with 
 some show of reason, that realists have never done their own 
 work thoroughly. Realism claims to be a theory of knowledge, 
 and yet realists have seldom made any systematic attempt to 
 bring the whole of knowledge under review. Realists have 
 analysed logic and mathematics very fully, and have also dis- 
 cussed the problems of perception, of applied mathematics 
 (especially physics), and of ethics, as well as many nice psy- 
 chological problems in the theory of knowledge, but even if 
 these problems are cardinal for any philosophy (and some of 
 them certainly are) the sum of them is manifestly incomplete 
 in comparison with what a theory of knowledge ought to be. 
 If realism is a genuine theory of knowledge, it must be a theory 
 of all knowledge, and so it should deal with art and history 
 and religion and biology as well as with logic and physics. 
 Rational reflection can deal with these matters; and the 
 idealists maintain that they have tried to be realists in their 
 investigations, and that they have learned from their failure 
 that realism is not enough. 
 
 The assumptions of realism, as I understand the theory, 
 are that knowledge is always the discovery of something: that 
 anything discovered is distinct from and independent of the 
 process of recognising it: that nothing which is known is 
 therefore mental except in the way of being selected by a 
 mind: and that if any selected thing is mental or mentally 
 tinged de facto ^ this circumstance does not affect the kind or 
 validity of our knowing of it. The opponents of realism argue 
 that these assumptions prove to be quite inadequate when 
 anyone comes to take them seriously, since there is a kind of
 
 182 THE LARGER OUTLOOK [ch. 
 
 imaginativeness and constructiveness in all important thinking 
 which is only mocked unless realism is superseded by some 
 more penetrating theory. 
 
 Even the physical sciences, it is urged, need much more 
 than observation and logic to have any body to them. Science 
 would not advance a step without the sagacious use of hypo- 
 theses, or without the trained imagination that frames them. 
 And hypotheses are imaginative constructions. What is more, 
 observation itself, scientifically regarded, is only the occasion 
 for constructive theory ; probability is a tissue of construction 
 which leaves the given just because it is merely probable; and 
 the laws of the sciences are flights of the logical imagination. 
 The world is what we make, and science is what the wit of 
 man has made and assimilated at any given epoch. 
 
 It would be easy to expand this argument almost to any 
 length now that the instrumentalists, the pragmatists and the 
 absolutists have shown the way; and I have not the space to 
 defend realism in more than a very summary fashion. The 
 most that can be expected in any discussion short of a treatise 
 is an enquiry into the most critical respects in which realism 
 is found wanting, especially on the ground that it cannot find 
 a place for constructive imagination in the physical sciences ; 
 and the chief of these respects are probability on the one hand, 
 and hypothesis on the other. 
 
 It is said that things are always determinately what they 
 are, so that they are never only probable, or only probably so 
 and so ; and we are asked to infer from this that probability is 
 only a conjectural construction and not the discovery of fact. 
 This consequence, however, does not follow at all. Probability, 
 as we have seen, is the logic of relevant but inconclusive 
 evidence. The probability of any proposition, therefore, is 
 the logical conclusion from the evidence that supports it; 
 and certainty is only the special case in which the evidence 
 is quite conclusive. 
 
 To be sure, we have to distinguish between probability and 
 the numerical measure of probability. The latter may be 
 inapplicable in some cases of probability, and it always pre- 
 supposes certain conventions which are at least remotely
 
 IX] . THE LARGER OUTLOOK 183 
 
 disputable. But probability itself need not be disputed just 
 because probability-fractions sometimes may be. Again, we 
 have to be careful to notice the respects in whish probability 
 seems to be a subjective affair. It often happens that some 
 proposition becomes increasingly probable as our knowledge 
 increases, and it is natural to conclude from this that pro- 
 bability itself is mental. That is a mistake. Probability never 
 changes when the evidence remains the same; and when we 
 have more relevant information than formerly, the increase in 
 the probability is a consequence of the difference in the evidence. 
 A man who knows that the dice are loaded, therefore knows 
 that double sixes are more likely than chance would suggest; 
 but, when he knows this, he still knows what the probability 
 would be if they were not known to be loaded; and this piece 
 of information remains unaffected by his subsequent discovery. 
 Relatively to any given piece of evidence, probability does not 
 vary at all. 
 
 Conclusiveness of evidence is a matter of degree. The 
 evidence may preclude other possibilities or it may not, and, 
 if it does not, its inconclusiveness can be conclusively proved 
 and also the degree of its inconclusiveness. Anyone, therefore, 
 who maintains that probability is constructive or imaginative 
 must also maintain that all inference is imaginative or con- 
 structive; and that, precisely, is what we deny. 
 
 But what of hypotheses, postulates, assumptions, et id genus 
 ornne? The hypothetical method pervades the sciences, and 
 hypotheses themselves vary from a slight analogical extension 
 to an admitted makeshift or a mere ballon d'essai. Hypotheses, 
 surely, are made and not found. There is will in them as well as 
 intellect, and if they are not imaginative constructions, what 
 in the world are they ? A sailor noticed the peculiar properties 
 of Iceland spar, and brought a piece of it to Bartholinus. 
 Bartholinus observed what the sailor observed, but he guessed 
 more; and Huyghens, setting his imagination to work, proved 
 the laws of double refraction. Realism, we are told, might 
 account for what Bartholinusand the sailor observed, but it can- 
 not account for Huyghens's imagination, and that is a proof of 
 its hopeless inadequacy as a theory of knowledge. The inward
 
 184 THE LARGER OUTLOOK [ch. 
 
 eye is the light of science. Even those who, like KirchofF and 
 his school, take science to be only the compendious description 
 of phenomena, do not mean by 'description'' what their realistic 
 phrase naturally suggests. Their 'descriptions' are not narra- 
 tives or accumulated observations. They are the inv(nition or 
 construction of imaginary things which are substitutes for per- 
 ceived ones, and description, according to this way of it, is the 
 economical manipulation of these substitutes in such a way as 
 to find the smallest set of symbols which corresponds to the 
 unmanageably intricate phenomena. 
 
 This argument is striking and it reaches as far as any science 
 worth the name. But there are holes in it. 
 
 liCt us grant, for the sake of argument, that all hypotheses 
 are mere figments of the mind, mental products through and 
 through. What follows .'* The products are what we choose 
 to make, but after we have made them they are discoverable 
 things. Making" a thins; is not the same as recognising it. It 
 is one thing to do something, and quite another thing to know 
 what we have done; and even when we have made things con- 
 sciously there may still be a great deal to learn about them. 
 To do anything consciously is only to be aware of it, in some 
 measure, while we are doing it, and it is common enough for 
 a man to make suggestions or hypotheses which carry him 
 further than he would, or are fully intelligible to his successors 
 only. That, indeed, is the history of all fertile ideas, and this 
 fertility of ideas, it would seem, is something which has to be 
 discovered in the same sense as the colour of a cat's eyes. 
 Whatever the mind constructs, in a word, must also be 
 apprehended, and the apprehension of mental constructions 
 or imaginings does not require any peculiar mode of knowledge. 
 The construction confronts the mind, and the mind is directly 
 aware of it. 
 
 That is one point, and another point leads to similar re- 
 flections. It is usual in the logic-books to distinguish between 
 hypotheses concerning laws and hypotheses concerning causes. 
 Hypotheses of the former class are suggested fornndae of 
 correlation, and those of the latter class are either agents (to 
 use popular language) or else the medium of some agency. All
 
 ix] THE LARGER OUTLOOK 185 
 
 ' models ■" of the type of the kiminiferous ether or Faraday's 
 * tubes of force' belong to this class, and so do Robinson 
 Crusoe's dismayed belief that the footprint meant a man, or 
 any suggestion like *•' Cherchez la femme."" In addition to 
 these two classes we may perhaps add a third class of merely 
 conceptual models. According to Hertz, for instance, the 
 ultimate problem of dynamics is the invention of images 
 which are thought-substitutes for phenomena, obeying the 
 same laws because they have the same logical properties ^ 
 
 In all these cases, it would seem, knowledge has the same 
 task of observation and inference before it, and the principles 
 of such knowledge do not seem to differ from those discussed 
 in the earlier chapters of this book. A formula, plainly, must 
 be apprehended like anything else, its implications depend 
 upon the laws of deduction, and its correspondence with the 
 facts is something which the mind simply finds. Hypothetical 
 causes, again, may be merely supposed or constructed, but 
 their mode of working must be given if these conjectures have 
 any plausibility at all, and they collapse of themselves if any 
 other agent can be shown to have done the work. Robinson 
 Crusoe's belief was an inductive inference of the ordinary type 
 in which the ground of belief rests primarily upon repeated 
 observations, and he would have rejected it utterly if he had 
 found that an imp or some carved wreckage from the ship 
 had made the footprint. And so of the third class. These 
 conceptual models must be apprehended by the mind, and the 
 mind has no option about them once they are formed. If 
 their logical properties correspond to those of the phenomena, 
 these properties of the images and of the phenomena, together 
 with the precise correspondence between the two, are the same 
 in principle as any other piece of logical correspondence ; and 
 the very simplicity of the images is also a detenninate and 
 discoverable property of them. 
 
 These arguments, therefore, prove at the best that the mind 
 can construct as well as know, that it may know its own con- 
 structions, and that its knowledge of these constructions is 
 often more serviceable for action and for speculation than 
 1 The Princij)les of Mechanics, at the beginning.
 
 186 THE LARGER OUTLOOK [ch. 
 
 direct perception of the phenomena. Moreover, these state- 
 ments are true under a most significant limitation. The con- 
 structions must be known to have the same general properties 
 and type of connectedness as the phenomena they represent, and 
 the adequacy of their representation must be constantly checked 
 by direct observation. Still, this limitation does not alter the 
 fact that we often think of symbols and substitutes instead of 
 their originals, and if these symbols are our own constructions 
 the positive thesis of the argument is sound. On the other 
 hand, it does not contradict I'ealism; for realism does not 
 imply that the mind cannot construct or that its constructions 
 cannot be known. And yet, when we examine these construc- 
 tions, we find that this making is at least three parts finding. 
 Formulae, and hypotheses too, are selections from reality. 
 They are schematic facts, or facts considered in an unfamiliar 
 connection ; and scientific imagination itself may not prove 
 to be anything more. 
 
 Let us pass, now, to the biological sciences which have 
 recently received so much attention. Biology, to be sure, 
 is still empirical and opportunist in a sense in which physics 
 is not. On the other hand, it is mere bigotry to assume that 
 physics and chemistry are the only genuine sciences and that 
 biology is mere guess-work. Undoubtedly the event may 
 prove that the physics of osmotic pressure, or of the formation 
 of sand-ripples and other symmetrical patterns, or the chem- 
 istry of colloids, may also include the main principles of 
 organic assimilation, restitution, reproduction and regulation. 
 That is the dream of the ' mechanists,*" though it is hard to 
 see why anyone should suppose that chemistry, to say nothing 
 of biology, is nothing but mechanics. Indeed, the event may 
 prove that the organisation and behaviour of living things has 
 principles peculiar to itself; and it would certainly be odd if 
 these extremely complex and unstable compounds had no 
 peculiar properties or peculiar principles. If water has pro- 
 perties which cannot be reduced to the properties of hydrogen 
 and oxygen taken separately, is it not improbable to the last 
 degree that the principles of inorganic and unorganised things 
 are fitted, by themselves, to explain organic life?
 
 ix] THE LARGER OUTLOOK 187 
 
 But we may leave this dispute. Even if it could be proved, 
 to take the most extreme case, that organisms have a type of 
 unity which lifeless things do not possess, that their character 
 (despite the Mendelists) cannot be calculated, that the re- 
 petition of an errand-boy's rap or of a miner's pickaxe is not 
 comparable to anything in mechanics, that the substance of 
 the living is sharply divided from the substance of the dead, 
 and that all biological development is irreversible in time 
 whereas physical laws are not, there would still be no good 
 reason for denying that the mind discovers and accepts these 
 biological facts and principles in the same sense as any others. 
 The whole problem, indeed, might be dismissed as irrelevant 
 to the present enquiry were it not that certain authox's loudly 
 proclaim its relevance. Biological facts and principles, they 
 allege, are not merely profoundly different from physical or 
 chemical facts or principles, but they differ in such a fashion 
 that it takes a different kind of knowledge to understand 
 them. In these affairs, logic and observation have to give 
 place to sympathy and intuition. 
 
 This thesis is commonly supported by such arguments as 
 the following: Life, it is said, ex vi termini, can only be felt, 
 and dare not be intellectualised. Living things are ultimate 
 centres of spontaneity, creative and themselves uncreate, 
 radically novel and therefore superior to rules. There is a 
 continuity in life which defies the intellect, and time is lived 
 through, not disarticulated like the successive movements of 
 an electric clock. To answer these arguments in detail would 
 lead too far afield, but I should like to point out that the pro- 
 posed substitute for thinking and observing is utterly futile, 
 and that observation and inference are the only means of 
 attaining truth in the biological sciences. 
 
 The first point may be proved very simply. Intuition, in 
 M. Bergson's sense^ is cabined within the self, and sympathy 
 is only a projected feeling which may perhaps be attuned to 
 the same pitch as the feeling of the person or animal to whom 
 our sympathies go out, but is never identical with that feeling. 
 If life, then, can only be felt, the lives of others cannot be 
 1 See his Introduction to Metaphysics.
 
 188 THE LARGER OUTLOOK [ch. 
 
 felt ; and nothing could be more perversely intellectual than 
 an argument by analogy from our own feelings to the feelings 
 of a wasp or a caterpillar. It is very doubtful, indeed, whether 
 life (itself) can ever be felt, or whether, as some suppose, feeling 
 is just life at a certain degree of tension. Even if it were, 
 however, we can feel our own lives only. One man does not 
 become another even in the contagious gloom of a common 
 defeat or in the spreading hilarity of Armistice Day. 
 
 Our second point is also manifestly true. There is no way 
 of studying living things except by observing them and 
 inferring from our observations. We observe beetles in the 
 same way as we observe balloons, we classify lepidoptera by 
 the same logic as we classify crystals or alkalines, we see to the 
 hatching of chickens by the very rules which we use in making 
 water boil at a certain temperature. Biologists, to be sure, have 
 need of imagination. As Mr D'Arcy Thompson reminds us, 
 " It has taken great men to discover simple things^"" and the 
 motto he quotes from the Vegetable StatisticTtS of Stephen 
 Hales is singularly apposite. That pioneer among physio- 
 logists and admonisher of gin-drinkers sagely opined that " the 
 reasonings about the wonderful and intricate operations of 
 nature are so full of uncertainty that, as the Wise Man truly 
 observes, hardly do we guess aright at the things that are 
 upon earth, and with labour do we find the things that are 
 before us.*" That is true universally in the natural sciences 
 and particularly in biology. Karyokinesis does not show itself 
 like the drift of a summer cloud. If cell-division depends 
 upon the asymmetry of surface-tension, it takes a very pene- 
 trating vision to discover how and why. Even the problem 
 of correlation, as Cuvier conceived it, needs the master s skill 
 and not the journeyman's. Again, the development of the 
 individual is not a thing that can really be traced continuously, 
 and the interpretation of gametes and their doings is nearly 
 as constructive as the gametes themselves. The observer has 
 to fill in far more than meets his eye ; and if ontogeny is con- 
 structive in this sense, phylogeny is still more constructive. 
 To pass with Darwin to the origin of species from Malthusian 
 1 On Growth and Form, p. 8.
 
 IX] THE LARGER OUTLOOK 189 
 
 theories of population and the breeding of domestic pigeons 
 would naturally be described as a tremendous synthesis, and 
 von Baer's idea that phylogenetic evolution is recapitulated 
 in the embryo was certainly not a mere transcript from the 
 surface of things. There is no disputing these facts, therefore; 
 but there is interpretation of them, and our interpretation 
 must wait for a little. The problem of development is sub- 
 stantially the same as the problem of history, and imaginative 
 construction in the sciences is so nearly akin to creation in 
 the arts that it is best to consider it in that connection. 
 
 Let us turn, then, to the human sciences and begin with 
 economics which (except for history) is the most important 
 and the most highly developed of them. Economics, indeed, 
 is so highly developed that it has had time to pay considerable 
 attention to its scope, assumptions and methods. Conse- 
 quently there has been a sharp division in the schools between 
 historical, descriptive and realistic economists (as they call 
 themselves), and deductive, abstract ones. The best modern 
 economists, to be sure, do not subscribe to the articles of 
 either creed, and even those economists who, like Ricardo, 
 tend to be as severely abstract and deductive as they can, or, 
 like SchmoUer and Menger, as sociological and historical as 
 possible, belie their occasional professions in their practice, 
 and sometimes in their general discussions. Schmoller declares, 
 indeed, that the defects of economic dogmatism can only be 
 remedied by consulting the whole body of historical and 
 statistical material now existing i, but he does not try to depose 
 deduction in his general account of economic methods, and 
 he maintains that the descriptions in economics are tested 
 and determinate observations whose logic is thoroughly sounds 
 He holds, it is true, that there are limits to logic, and that 
 all reflection, in the end, needs the support of general and 
 genial intuitions concerning the ends of God or history or 
 creative nature^, but he does not suppose that these limitations 
 invalidate logic or observation. 
 
 1 Zur Litteraturgeschichte der Staats- und Sozialioissenschaften, p. 279. 
 
 2 Gnmdriss der allgemeinen Volkswirtschaftslehre, vol. i. p. 101. 
 » Ibid. p. 111.
 
 190 THE LARGER OUTLOOK [ch. 
 
 These disputes concerning the method of political economy 
 have a special importance for the subject of this chapter. To 
 be sure, we may neglect the claim of the historical school to 
 be ' realistic,"' because realism in this sense means concreteness, 
 and philosophical realism need not be peculiarly concrete. Our 
 interest in the controversy is different. The ultimate claim of 
 the historical school is that a theory of economics which relies 
 wholly upon logic and detailed observation is necessarily in- 
 adequate, even granting the truth of its specific observations 
 and of its individual chains of reasoning. Political economy 
 as a whole, we are told, is too large for these methods, too 
 complex to be observed in detail or to be made the subject of 
 experiment, and too human a thing to be dissected or to be 
 followed out into its logical structure. Other methods are 
 needed, therefore, and a different ideal of truth. 
 
 We need not consider the complaint of those who reject 
 the 'dismal science"' of economics on the ground that it studies 
 wealth irrespective of welfare, or that it is pledged to that 
 'vulgarest saw"' which states that time is money. The study of 
 exchangeable commodities does not imply that a man's soul con- 
 sists of the abundance of his goods. Again, it does not follow 
 that an attempt like Ricardo's to consider the effects of free 
 competition without reference to other conditions is worthless 
 in itself or useless in practice. Competition, it is true, is not 
 free in fact, but it is free enough to make its freedom worth 
 selecting for study. The only inference that can properly be 
 drawn concerning such attempts is that they are not, and never 
 should pretend to be, full descriptions of the whole countenance 
 of complex society. No one ever thought so, although a few 
 may have neglected to state clearly that they did not think 
 so. It would seem, then, that if economics had to select its 
 grounds in this way and had no other alternative, the only 
 consequence would be that the science of economics could not 
 give a full account of social happenings ; and that is not an 
 objection to economics. It is only an objection to the over- 
 weening presumption of ignorant economists who mistake their 
 science for general sociology; and no science is responsible 
 for the mistakes of its zealots.
 
 ix] THE LARGER OUTLOOK 191 
 
 The real objection to ' abstract economics ' is different from 
 this one. Selection of principles, induction and deduction, are 
 not only necessary for economics but its salvation. The trouble 
 is that the principles of economics are themselves insecure, and 
 that its deductions limp in their logic. Parts of the science, 
 indeed, may be relatively free from these difficulties. The theory 
 of banking, apart from special circumstances of legal enactment, 
 may be able to make a fair show of autonomous logic, and 
 Mr Withers, perhaps, can justly claim that "above the war- 
 flood which has drowned so many of the landmarks by which 
 the students of exchange were taught to steer, there still rise, 
 serene and stronger than ever, the pillars on which the chief 
 laws of this science are engraved ^" But it is otherwise with 
 economics as a whole. The methods of economics can never 
 succeed in studying quite exactly the doings of "mankind in 
 the ordinary business of life^,"" or investigate "all lawful ways 
 of making a living^,"" or "that part of social organisation which 
 is related to human wants and human efforts directed to the 
 satisfaction of these wants^." Yet this is what economics claims 
 to do. Indeed, abstract economics, as it is currently defined, 
 assumes a certain typical psychology of mankind and professes 
 to draw its conclusions from this source. Mr Flux, for example, 
 says that he adopts the a priori or deductive method, but he 
 bases his reasoning, in fact, upon psychology. He assumes the 
 principle " that men desire wealth and endeavour to secure it 
 at as little cost to themselves as possible^"; "the recognition 
 of the tendencies which lead to the multiplication of the 
 human species, and the conditions which limit the increase of 
 numbers""; the principle "that men are capable of judging of 
 the efficacy of means to an end, and that the easiest means 
 will be chosen to reach any desired end'^"; and "the principle 
 of the satiability of wants®."" 
 
 1 Money - Changing , Preface. 
 
 2 Professor Marshall's definition of economics, Principles, Vol. i. p. 1. 
 
 3 Mr Henry Clay's definition, Economics for the General Reader, p. 1. 
 * Mr A. W. Flux's definition, Eco7iomic Principles, p. 5. 
 
 5 Op. cit. p. 14. « Ibid. p. 15. 
 
 ' md. p. 15. 8 Ibid. p. 15.
 
 192 THE LARGER OUTLOOK [ch. 
 
 These professions seem very hollow to any serious student, 
 not because they are abstract or even because the psychology 
 of them is superficial, but because economics is not really a 
 rigorous piece of deduction from principles of this kind. Eco- 
 nomists rather discuss how the workmen and the capitalists in 
 a civilised community (supposing them to be reasonably intel- 
 ligent) should set about to further their own interests. This 
 science of ways and means, however, cannot possibly be a piece 
 of rigorous demonstration, since the logic of ways and means 
 presupposes a matter-of-fact description of economic conditions, 
 and since the consequences which are inferred are effects which 
 are only probable and not certain. Economists select certain 
 important facts out of the welter of social happenings, and 
 draw probable inferences from these selected facts. If they did 
 not profess to do more than this, they would have nothing to 
 fear on the score of logic or truth; but, in fact, they make 
 larger claims and fail to carry them out. 
 
 When economists forget their general definitions, they suc- 
 ceed, very well, in describing the skeleton and the arteries of 
 business; and they made trustworthy predictions before the war. 
 Their science, therefore, is of the first importance; and it is not 
 easy. As Professor Marshall says, " The economist needs the 
 three great intellectual faculties, perception, imagination and 
 reason ; and most of all he needs imagination, to put him on 
 the track of those causes of visible events which are remote or 
 lie below the surface^""; and he goes on to say that "economic 
 studies call for and develop the faculty of sympathy, and 
 especially that rare sympathy which enables people to put them- 
 selves in the place, not only of their comrades, but also of other 
 classes^." These remarks are most just, and they are valuable, 
 not only on account of their justice, but also because they 
 show very precisely what sort of imagination and sympathy 
 is needed in economics. Imagination of this kind is insight 
 and deeper perception, not volatile fancy. It burrows into 
 things instead of trying to escape from them. It discovers 
 rather than invents. The sympathy that is needed, too, is just 
 
 1 Principles of Economics, vol. i. p. 43. ^ lUd. p. 45.
 
 ix] THE LARGER OUTLOOK 193 
 
 insight coupled with its appropriate feeling and conviction. 
 Anyone can see that miners want higher wages and shorter 
 hours, but that is neither insight nor sympathy. What is 
 lacking in it is the appreciation of detail, the recognition of 
 what the daily routine in a mine is like, of sordid housing, 
 grimy leisure, exhausted slumbers, shortened days and inade- 
 quate enjoyments. When these specific points of detail are 
 realised the appropriate feeling comes of itself; and this sym- 
 pathetic insight should also dispel sentimental illusions. For 
 the knowledge of detail also shows that the miner''s toil is not 
 unaccustomed, so that his life is not utter torture, as it would 
 be to those who are not miners ; and it suggests that his 
 grimy leisure is not quite so unbearable as it would be to a 
 Congreve or a Halifax. Sympathy of this kind is the result 
 of understanding, not its pilot. It shows the power of the 
 intellect, and not its weakness. 
 
 The historical method in economics gives the science a wider 
 range of facts, and makes it easier to distinguish general and 
 lasting conditions from transitory and local ones. It also brings 
 the principle of development to the front, and these matters 
 are best considered by considering history itself. For history 
 is the queen of the human sciences, a reigning consort on the 
 throne of man's larger outlook. 
 
 Historians attempt to describe the development of some- 
 thing in a connected narrative. Clio's scroll has room for the 
 biographies of individual men, and there may be histories 
 of rocks or of planets. Historical development, therefore, 
 is not confined to man in society, but history is nothing 
 unless it is a connected narrative, and unless it describes 
 development. 
 
 A narrative is a description of events which pays special 
 attention to their sequence in time. It differs, therefore, from 
 logical reasoning; for logical reasoning deals with the connec- 
 tion of principles and not with the consecution of happenings. 
 Indeed, the aphorism that chronology is one of the eyes of 
 history is only half the truth. Chronology belongs to history 
 and does not merely suggest it, and the attempt to prove that 
 chronology should be surpassed and done away in the philosoj)hy 
 L. 13
 
 194 THE LARGER OUTLOOK [ch. 
 
 of history is sure to result either in a still-born history, or in 
 a changing logic. 
 
 NaiTative, therefore, might seem to be as realistic an enter- 
 prise as any mind could undertake; but that, perhaps, is a 
 mistake ; and we all know something of the difficulties of 
 historical evidence. The notorious circumstance that eye- 
 witnesses seldom agree among themselves does not prove very 
 much, it is true. When the caretaker in Mr Chesterton\s story 
 said that no one had called because he did not count the post- 
 man, all we can infer is that it is very hard to notice the right 
 things, and those who argue that Mary Magdalene and Salome 
 and the other Mary did not hear anything at the empty tomb 
 because an angel spoke to them, according to S. Matthew : and 
 a young man according to S. Mark : and two men according to 
 S. Luke: do not seem to be on very firm ground. Discrepancies 
 in the accounts of eye-witnesses only prove that it is easy to 
 be mistaken in these matters. On the other hand, there are 
 many cases in which the accounts of eye-witnesses prove very 
 little because the witnesses themselves do not understand the 
 events that are passing. A sailor at the battle of Gravelines, 
 for example, could not be expected to give a connected account 
 of the fight unless he understood the tactics and the strategy 
 of the Tudor admirals. 
 
 If that is true of the contemporary observation of a parti- 
 cular event, how much the more is it true of the narrative 
 of a people's doings? The best contemporary historian has to 
 piece together what he has heard rather than to describe what 
 he has seen with his own eyes. Procopius or Gregory of Tours 
 had to do this work of patching, as well as the historians of 
 the late war ; and when history is written some generations 
 after the event, the narrative is valueless unless it reveals a firm 
 gi'asp of principle and a thorough sifting of the evidence. 
 
 The difficulties of the latter enterprise are severe enough to 
 test the mettle of the best historian. Historians have to deal 
 with probabilities of the second order. Even if the testimony 
 were certain, the inferences from it would be, for the most part, 
 only probable, and the evidence, in point of fact, is not certain, 
 since the witnesses themselves may be mistaken or insincere.
 
 ix] THE LARGER OUTLOOK 195 
 
 There seem, therefore, to be no universal canons of historical 
 evidence, although, of course, any particular inference is value- 
 less unless it is logical. Contemporary narratives are not always 
 more trustworthy than later ones, for the later ones may pre- 
 serve a tradition which is really very accurate indeed ; and a tale 
 need not be true just because it seems too strange to have 
 been invented. The rules of evidence in such cases are pre- 
 sumptions at the best, and the most successful surmises of 
 historians are the wonder of untrained observers. Freeman 
 says that he found the key to the whole legal theory of 
 the Norman Conquest in the statement of the Peterborough 
 Chronicle that men bought land from the Conqueror, taken 
 in conjunction with the statement of Domesday that the Abbot 
 of Saint Edmundsbury was given certain lands ' quando Anglici 
 redimebant terras suas\'' 
 
 Historical narratives, then, have to be pieced together; and 
 history, before it is written, is like a jig-saw puzzle where 
 most of the pieces are missing. Even so, however, this species 
 of logical ingenuity can take only a few short steps ; and, on 
 any theory, it is but a small part, although an indispensable part, 
 of the historian's equipment. The whole life of a man cannot 
 really be pieced together, and the life of a nation is still less 
 tractable. History cannot attempt to record everything or to 
 conjecture all details. It has to find a principle in the nar- 
 rative, and to show that the whole story turns on a few salient 
 facts. There is more history in Sir Walter Raleigh's single 
 sentence, "Whosoever commands the sea commands the trade; 
 whosoever commands the trade of the world commands the 
 riches of the world and consequently the world itself-," than 
 in the recoi-d of half a hundred battles. Historians have to 
 find seminal principles and to trace their development. When 
 they find them they find history; and if they miss them they 
 are only a warning to others. 
 
 Accordingly, if historians are tied down to descriptive nar- 
 rative and the chances of testimony, it is not astonishing that 
 some of their friends should try to save them from this dungeon 
 
 1 Methods of Historical Study, pp. 185 sqq. 
 
 2 The motto of Sir Julian Corbett's Drake and the Tudor Navy. 
 
 13—2
 
 196 THE LARGER OUTLOOK [ch. 
 
 of realism ; and so we are told, on the one hand, that histoiy 
 should abandon her realism and betake herself to art, or, on 
 the other hand, that she must cease from gi-ubbing in the soil 
 and become a philosophy of history. 
 
 What is needed in history, it is said, is a pictorial method 
 of representation. The historian's function is to suggest the 
 spirit of an epoch by vivid and telling illustrations. Macaulay's 
 account of the state of England in 1685\ for example, relies 
 wholly on this method. He consulted the records of the time, 
 to be sui'e, and he was infinitely better acquainted with them 
 than most of his critics, but his method was pictorial. He gives 
 us Gregory King's estimate of the population of Great Britain, 
 the returns of the excise, the regiments in Charles's standing 
 army, the output of tin in Cornwall, and Sir William Petty 's 
 account of labourers' wages, but this information is essentially 
 illustrative, part of a picture, and it is of the same order, in 
 point of method, as the rest of his picture. Macaulay looks 
 for flashes of fact to illuminate the nation's life. The dalesmen, 
 he tells us, kept the secret of the road from Borrowdale to 
 Ravenglas, and every traveller in that region first made his will. 
 Monk ordered his sailors to wheel to the left. Queen Anne 
 saw a herd of five hundred red deer as she travelled to Ports- 
 mouth. A rural clergyman nailed up the apricots and curried 
 the coach horses, left the squire's table with the beef and the 
 carrots, and was lucky if he married a waiting-woman. The 
 coffee-rooms smelt like a perfumer's shop. Claude Duval danced 
 a coranto for a lady's ransom. Dryden foretold that the Royal 
 Society would 'reach the globe's last verge and on the lunar 
 world securely pry.' And so on. The only reason for multiply- 
 ing these examples is that this heaping up of illustrations is 
 an essential part of the method. 
 
 Illustration is not reasoning, and yet it enables the mind to 
 connect facts and think of principles. For illustrations are 
 chosen with an eye to their suggestiveness, and they may serve 
 the purposes of historians even if they are fanciful. Squire 
 Western or Humphrey Clinker may teach history as well as 
 John Wilkes or Titus Gates. This manner of portrayal, then, 
 1 History, chap. iii.
 
 ix] THE LARGER OUTLOOK 197 
 
 needs special study, and we shall meet it again. For the time 
 being, it is enough to say that the use of pictorial methods 
 does not prove that history needs a special mode of knowing. 
 Macaulay's illustrations, numerous as they are, are only a small 
 selection from the facts; but any narrative must select, and 
 thought, as Hegel says, is the great epitomiser. For the rest, 
 the point is psychological rather than logical. To think by 
 way of illustrations is to consider some of the consequences 
 of principles without examining the principles themselves ; and 
 this is often more vivid as it is generally less tedious. On the 
 other hand the illustrations are pointless unless they suggest 
 connection ; and suggestion, itself, as we have already seen, does 
 not need a peculiar theory of knowledge to itself. Historical 
 illustrations have an objective meaning, and this meaning is 
 their historical value. Let us consider, therefore, whether the 
 meaning is of such a kind that only a philosophy of history 
 can satisfy it. 
 
 The philosophy of history has suffered at the hands of its 
 friends. For example, those who know very little about it, and 
 many who would never dream of reading Hegel, are still able to 
 quote some hearsay jesting concerning his remarkable section on 
 the geographical basis of history ^ They know that he held that 
 world-history must be confined to the temperate zone, and to 
 the old world ; that America and Australia are physically and 
 psychically immature because some of the Australian rivers 
 lose themselves in marshes and because the Jesuit friars had 
 to ring a bell at midnight in order to remind the simple Indians 
 of their matrimonial duties ; that Americans have to be busy 
 and self-seeking but cannot have a history since their land has 
 only a dream future ; that Afi'ica must always be a land of 
 childhood, partly because of its geography and partly because 
 its inhabitants are without self-control, as is proved by the 
 fact that the wives of the King of Dahomey (there were exactly 
 3,333 of them) are slaughtered en masse when the king dies. 
 It is all very ludicrous, to be sure, and HegePs view that history 
 did not exist before the records of if-, or that the history 
 
 1 In the concluding section of the Introduction to the Philosophy of History. 
 - Philosophy of History, Sibree's translation (Bohn), p. 64.
 
 198 THE LARGER OUTLOOK [ch. 
 
 of the world need not mention individuals^ are prima ^acie 
 absurd. Hegel had a word of praise for great men, it is true, 
 because he was careful to point out that he had anticipated 
 Goethe in adding ' because he is a valet ' to the adage that 
 no one is a hero to his valet de chambre^ But even in that 
 passage he regards the hero as only the interpreter of advancing 
 Reason, and in general he maintains consistently that history 
 is only the spirit of a people reflectively expressing itself. 
 
 Those who smile at these ideas, however, usually feel the 
 need of their own philosophy of history, and perhaps they can- 
 not afford to be quite so contemptuous of Hegel. Hegel tried 
 to incorporate the geographical theory of history in his account 
 of the History of Spirit, and he became absurd by taking it too 
 seriously. He would have been still more absurd, however, if 
 he had accepted it outright. Geography undoubtedly plays 
 a critical part in human history. Climate tells on a people, 
 the sea may be their life, and the great road from Sardis to 
 Susa was the key to the fate of Europe. Geography makes 
 military history repeat itself, and a mountain range may be 
 more despotic than any monarch. But geography, as Hegel 
 saw, cannot be the essence of history, because history, at its 
 lowest rating, is man's use of geography. The economic theory 
 of history, again, is a better philosophy of history because it 
 takes human needs into account ; and the migi'ations of wan- 
 dering hordes in search of sustenance, or the lust of great 
 communities for naval bases, are probably far more forcible 
 than any other constraining impulse in the life of a people. 
 Still, the economic theory of history is a one-sided affair after 
 all, just because economics is not sociology, and those who follow 
 Augustine, or Bossuet, or Hegel, and regard history as a theo- 
 phany or as the unfolding of Absolute Spirit may be nearer the 
 truth than the others. Every great history is a philosophy of 
 history too. It traces events to their sources and discovers the 
 seeds of later things. Nothing is philosophy, if this is not. 
 
 Those who deny the possibility of a philosophical history 
 maintain either that there can be no pervasive order and prin- 
 
 1 Philosophy of History, Sibree's translation (Bohn), p. 70. 
 
 2 Op. cit. p. 33.
 
 ix] THE LARGER OUTLOOK 199 
 
 ciple in the development of societies, or else that any appearance 
 of this order is only superficial. The lives of men and women, 
 it is argued, are ultimately irrational and their behaviour a 
 thing of caprice and hazard. Or, again, it may be held that 
 social history must be without general principles, becausesociety 
 itself is only a specious aggregation of human beings, and be- 
 cause the man always makes the hour, not the hour the man. 
 
 The first of these arguments is a gross overstatement. How- 
 ever obscure human personality may be, it at least shows a very 
 marked aggregate similarity in individual dissimilarity. Even 
 opportunism shows few traces of accident except for the good 
 fortune of those who use their chances. The opportunities 
 themselves are certain to occur. It is another thing, of course, 
 to pass from this aggregate regularity to specific prediction 
 and to maintain, for example, that Capital must accomplish its 
 own overthrow in some particular fashion at some particular 
 time. History does not work so. Historians find their prin- 
 ciples after the event, and can only grope a little towards the 
 future. But sometimes they can understand after the event, 
 and that is a sufficient rejoinder to this contention. 
 
 The second argument is more plausible. A community is 
 not individual in the same sense as its members are, and it is 
 often necessary to expound this truism. Communities, again, 
 follow their leaders quite as often as their leaders follow them, 
 and even the idiosyncrasies of these leaders sometimes have a 
 profound effect. Sidgwick remarks somewhere, I think, that it 
 would have made a great difference to mankind if Mohannned 
 had been fond of wine and indifferent to women; and other 
 instances might easily be cited. On the other hand, the members 
 of a comnmnity are so imitative, and so profoundly influenced 
 by one another in their training and aims, that communal life 
 is certainly a reality. For that reason it is idle to discuss 
 whether the hour or the man is the proper object of historical 
 study. The man gives his own bent to the hour. The hour 
 needs the man and finds him sooner or later. British seaman- 
 ship was what Drake began and Nelson continued. It had the 
 stamp of these seamen upon it, and it would have been dif- 
 ferent under other leaders. On the other hand, it might not
 
 200 THE LARGER OUTLOOK [ch. 
 
 have been so very different under other admirals ; and we can 
 discern its outlines without enquiring too minutely whether 
 England owed more to Drake than Drake owed to England. 
 
 History, then, is possible, but how is it possible ? Is it, in 
 the last analysis, a narrative of the development of fact, or is 
 fact only the body of history, and history itself the soul of 
 this body ? Hegel and Croce' take the latter view, and so 
 they reinterpret the nature of knowledge. Knowledge, they 
 argue, is life and not discovery, and history is just man's life 
 become self-conscious. That is why Hegel maintained that 
 history is coeval with annals ; for annals are the self-conscious- 
 ness of spirit; and that is history^. According to this view, 
 therefore, the aim of any historian is to grasp the nature of 
 development by living it in his own person. Contemporary 
 narratives are valuable, not on account of the events they 
 record but on account of the life they reveal, and the first duty 
 of a historian of the past is to enter into the life of the past, 
 and to be present in spirit with Socrates at Chaeronea, with 
 Burghley at the Council table, or with some lonely British 
 sentry on the Afghan frontier. History is the torrent of life, 
 not the empty bed which the torrent has left. 
 
 This theory is full of confusion. In the first place, life and 
 the record of it are not the same. Life may be lived consciously 
 (and even self-consciously) without annals and without history; 
 and the recorder of life need not participate in the deeds he 
 records. He needs imagination and sympathy, to be sure, but 
 sympathy and imagination participate in life by metaphor 
 only, and they are needed in all the sciences, not only in history. 
 In the second place, a people is not really self-conscious. Men 
 and women are ; and if a nation were self-conscious one would 
 suppose that soldiers and statesmen would be the organs of 
 its self-consciousness instead of historians. Nothing is gained 
 by confounding the man of action with his biographer. In 
 the third place, self-consciousness, as we saw in the last chapter, 
 is not a peculiar variety of knowledge in which knowing and 
 
 1 Logic, pt II. chap. iv. Cf. Dr H. Wildon Carr, The Philosophy of Bene- 
 detto Croce, chap. xi. 
 
 2 Op. cit. pp. 83, 64.
 
 ix] THE LARGER OUTLOOK 201 
 
 being are one. If it were, HegePs philosophical conception of 
 history might be profoundly true despite its subsidiary diffi- 
 culties ; but because it is not his theory falls. 
 
 We may conclude, then, that observation and inference are 
 the only instruments of the human sciences, and that a realism 
 of this kind does not need to be supplemented or transfigured 
 by any other variety of knowledge. At the same time, doubts 
 may reasonably linger in the reader's mind, and especially 
 doubts which concern the constructiveness and imaginativeness 
 of scientific knowledge. It is advisable, therefore, to treat 
 these questions somewhat more generally than hitherto. 
 
 It must be admitted that the contrast between knowledge 
 as making and knowledge as finding, seems peculiarly hollow 
 on its first aspect. Often, indeed, the choice between these 
 two conceptions appears to be quite indifferent. One can 
 speak equally well of making the right synthesis and of finding 
 the true connection, and it looks as if there were very little 
 philosophy in meeting one's opponents with the remark : " Foi' 
 'constructing,' where possible, ixad 'finding.'" Yet realists 
 must admit, I think, that this curt notice is one of the most 
 compendious and one of the least inadequate w^ays of putting 
 their main contention. Again, it might seem that the op- 
 ponents of realism are at liberty to accept this amendment 
 and still to retain their theory. The mind, it may be said, 
 finds the reality of things by assimilating them. Quarrels on 
 this head have a verbal look about them, and, perhaps, not 
 even that. Flaubert, I suppose, was a good judge of the 
 meaning of words, and he wrote : " External reality must enter 
 into us so that it makes us almost cry out for its worthy repro- 
 duction. An author always writes well when he has his model 
 crisply before his eyes\" The mind, in other words, perceives 
 what it has assimilated, and that is the whole of its business. 
 
 On the other hand, the consequences of this theoretical dif- 
 ference are very important indeed. AVhat the mind organises, 
 quite plainly, is just itself, and its constructions, we are told, 
 are ideal constructions, that is to say, constructions of its 
 own ideas. The metaphysical importance of this theory is 
 ^ Correspondance, IP Serie, p. 269.
 
 202 THE LARGER OUTLOOK [ch. 
 
 surely plain enough, and its interpretation of knowleri^e is 
 most radically distinct from the common one. If knowledge 
 is only fabricating, it is clearly idle to conjecture whether its 
 fabric corresponds to any reality other than itself, for the 
 very thought of any such reality is just another fabrication ; 
 and the discovery of things as they are, is even more absurd. 
 For that reason the pragmatists, in their consuming zeal for 
 Formless Logic, disclaim any final or ultimate knowledge of 
 things, and profess to be content with the temporary fabric 
 which M or N has contrived to weave at some given time; and 
 the absolutists tell us that knowledge is just coherence of ex- 
 perience, although they do not usually explain what sort of 
 coherence it is. Indeed, according to all theories of this kind, 
 knowledge is only the regulation and organising of its own 
 material. It mav perhaps be controlled by stimuli which affect 
 it ah extra, but this control is unknowable simply because it is 
 not itself assimilated or known. According to this philosophy, 
 knowledge always transforms what is given to it, and therefore 
 we can never know the given as it is. 
 
 In view of these consequences, it is manifestly essential to 
 examine this ' constructiveness ^ of knowledge as thoroughly 
 as possible, and I make no apology for calling the reader's 
 attention, once again, to a most elementary point. Even if 
 knowledge is always a construction, this account of it would 
 be incomplete. For knowledge implies the recognition or 
 apprehension of constructions as well as the making of them. 
 When this requisite is clearly understood (and not till then) 
 it is possible to set about to enquire what the marks of con- 
 structiveness are; whether everything apprehended has these 
 marks; and, particularly, whether this ' construction ' is not 
 usually a mistaken expression, and most of it simply discovery. 
 
 ' Construction ' in theories of knowledge is antithetic to 
 ' givenness,"' and philosophers, like black and white artists, 
 work with these two notions as well as they can. The orthodox 
 procedure, indeed, is something like this : We are given very 
 little, and so we have to doctor it ; this doctoring is more than 
 a necessary expedient, indeed it outweighs the importance 
 of givenness ; it is worth so much more that the givenness
 
 ix] THE LARGER OUTLOOK 203 
 
 does not count appreciably ; the doctoring's the thing, and 
 the given can be doctored away altogether. Some idealists 
 stop at one stage of this philosopher's progress, and others at 
 others ; but although the caution of some of them and the 
 boldness of the others are always interesting and sometimes 
 valuable, the most important question of all is the truth of 
 the first of these statements. It is common to argue that the 
 given is sensation at best, and probably something still more 
 obscure ; that it must be so primitive as to be pre-natal or 
 pre-human, and therefore must be indescribably meagre; and 
 the like. Our thesis, on the contrary, is that anything which is 
 known is therefore given, and so that it is as complex (or as 
 rudimentary) as knowledge itself. That has been the principal 
 contention of this book, and the design of the present chapter 
 is to put it to the test in the least promising departments of 
 knowledge. 
 
 The need for imagination in the sciences seems an insur- 
 mountable objection but is not really so. Scientific imagination 
 is fundamentally insight, and, in the human sciences, insight 
 allied with sympathy. This captive imagination if we may 
 call it so, is either a profounder analysis than is common, or a 
 greater genius for detecting analogies, or both combined; but 
 unless all analysis is opposed to discovery, it is difficult to 
 sustain the thesis that profound analysis must be opposed to it; 
 and unless every analogy is a spiritual flight which leaves the 
 earth of discoverable fact behind, it is very hard to show 
 convincingly that remote or subtle analogies must be of this 
 order. In fact, the reverse is true. Analogy is a fumbling, 
 shrouded, tentative discovery of connection, and therefoi'e 
 differs from a connection which has been tested and proved, 
 but this very opposition shows that the two are the same in 
 principle; and if imagination in the sciences is not opposed to 
 realism, the sympathy which ought to accompany it in the 
 human sciences is not opposed to realism either. For sympathy 
 is the feeling which accompanies the appreciation of detail. It 
 is the lively interest which stimulates, and in turn is stimulated 
 by, the minute knowledge of the applications of a social 
 theory.
 
 204 THE LARGER OUTLOOK [ch. 
 
 It may be argued, however, that this captive imagination 
 of the sciences is opposed to the free imagination of art and 
 literature. Don Quixote made a half-beaver of pasteboard, 
 and cleft it in two with the first blow of his sabre. Then he 
 mended it with thin sheets of iron. But he never tested the 
 helmet again, precisely because he was a knight of romance 
 and not an armourer by profession. And there are parables 
 in La Mancha. 
 
 The best critics would not argue precisely in these terms, 
 for this way of putting it confuses between imagination and 
 fancy; and the flights of fancy, as we have seen in an earlier 
 chapter, are only a whimsical embroidery of artistic imagina- 
 tion, the sportive hem of its garment. Even so, however, 
 those who accept Coleridge''s distinction^ between imagination 
 and fancy might use a somewhat similar argument. Coleridge 
 defined the imagination as the 'esemplastic' power of the mind 
 which fuses the life of things into one-, and this faculty, it might 
 seem, is completely overlooked in any realism. Most of the 
 great critics have said the same. For example, Wordsworth's 
 famous description of poetry as ' the breath and finer spirit 
 of all knowledge' assumed, as Coleridge did, that the highest 
 knowledge is an esemplastic thing; and even Flaubert, so often 
 called a realist, probably did not mean anything very different 
 when he said that "poetry is just a way of perceiving the 
 things outside us, a special instrument which sifts material 
 things, and, without changing them, transfigures^."" 
 
 Plainly, the perennial disputes concerning realism in art have 
 no meaning unless the disputants have settled whether or not 
 realism may include imagination. Flaubert, for example, inter- 
 prets the artistic imagination most realistically, taking it to be 
 a piercing of the veil and the discovery of truth in its essence. 
 " What sustains me," he says, " is the conviction that I am 
 
 within the truth \n author must write as he feels. He must 
 
 be sure that he feels well; and then he may snap his fingers at 
 anything else on the earths" Meredith, on the other hand, 
 either denies that realism can be imaginative or else allows it a 
 
 1 Of. chap. IV. ^ Biographia Literaria, chap. xin. 
 
 3 Op. cit. pp. 193, 194. 4 ma^ p. 204.
 
 ixl THE LARGER OUTLOOK 205 
 
 pedestrian imagination only. "Between realism and idealism," 
 he says, "there is no natural conflict. This completes that. 
 Realism is the basis of good composition : it implies study, 
 observation, artistic power, and (in those who can do no more) 
 humility. Little writers should be realistic. They would then, 
 at least, do solid work. They afflict the w orld because they will 
 attempt that it is given to none but noble workmen to achieve. 
 
 A great genius must necessarily employ ideal means Men to 
 
 whom I bow my head (Shakespeare, .Goethe ; and, in their 
 way, Moliere, Cervantes) are realists aufoncl. But they have 
 the broad arms of idealism at command. They give us Earth ; 
 but it is earth with an atmosphere ^^ And as a third alter- 
 native I may quote a passage from a lesser writer of the 
 present day. " The whole book is a shout of discovery, horri- 
 fied discovery, of the ugliness of life. Ifs as if she said : 
 ' Listen ! Listen ! These things actually happen to some 
 people. Isn't it awful .'*'.. .Til grant you that imagination, is so 
 essentially a quality of youth that the merest rootlet of a 
 reality is enough to set a young artist beanstalk-climbing. But 
 the older he grows, the wiser, the more versed in reality, the 
 less he trusts his imagination, the more in consequence his 
 imagination flags and withers, till he ends... as the recorder 
 
 merely of his own actual experiences and emotions Remember 
 
 how much more Madala dwelt on the sense of loneliness and 
 lovelessness, on the anguish of the loss of something to love 
 her, than on what one might call the — er — official emotions of a 
 betrayed woman. Didn't it strike you? Doesn't that show that 
 she was depending on her experience rather than on her imagi- 
 nation, fitting her own private grief to an imaginary case^.'*" 
 If realism is restricted to the description of fact, and if we 
 mean by fact what is commonly meant by it, then realism 
 can only supply the artist with some of his material. Realism 
 in romance, for example, would mean those tiresome mbiut'iac 
 which every good writer has to reckon with. " I had miserable 
 luck with St Ives,'''' Stevenson wrote to Colvin. " Being already 
 half-way through it, a book I had ordered six months ago 
 
 1 Meredith's Letters, vol. i. pp. 156-157. 
 - Legend by Clemence Dane, pp. 89, 9'2, 93.
 
 206 THE LARGER OUTLOOK [cii. 
 
 arrives, and I have to change the first half of it from top to 
 bottom. How could I have dreamed the French prisoners 
 were watched over like a female charity school, kept in a 
 grotesque livery, and shaved twice a week ? And I had made 
 all my points on the idea that they were unshaved, and clothed 
 anyhow." Writers of romance have to attend to these matters 
 of fact for fear of destroying the illusion in their narratives 
 if they do not. We are sometimes annoyed, to be sure, when 
 facts are thrust upon us clumsily, and it is distressing to know 
 that a lady wrote to Dickens justly complaining that he had 
 no right to describe her physical peculiarities so minutely 
 when he gave Miss Mowcher to the world', but there is no 
 good reason why an author should not be a reporter if he 
 likes, or why he should not describe his neighbour's furniture 
 or his father"'s bathroom as meticulously as he chooses. It is 
 only an accident, however, if this is art, even granting that 
 art may use no other materials and still be art. If Dickens 
 had written a life of his father, or if Scott had, these bio- 
 graphies might have been genuine art, quite as unmistakeably 
 as the picture of Mr Micawber in David Copperjield or of 
 Saunders Fairford in Redgauntlet. But it is foolish to debate 
 whether an author should confine himself to literal fact, when 
 the purpose of art need not be literal description at all. 
 
 The aim of this essay is to consider a certain theory of 
 knowledge. There is no room in it, therefore, for a theory of 
 art unless art and knowledge either overlap or coincide. That, 
 however, is precisely what many philosophers take them to 
 do, and since art is certainly expressive and certainly conveys 
 truth, it is impossible to avoid the problem of artistic pre- 
 sentation without leaving the tale of knowledge half told. 
 Per contra^ it would be irrelevant to consider art except from 
 this angle. Art touches the theory of knowledge very nearly, 
 but art itself is not a species of knowledge. From the stand- 
 point of knowledge we have to consider what sort of reality 
 is reached in art, and how this reality is a vehicle of knowledge. 
 The other problems of art should be left by the wayside. 
 
 The simplest answer to the first of these problems is to say 
 1 See a letter from Dickens in Forster's Life, vol. iii. pp. 2 sq.
 
 IX] THE LARGER OUTLOOK 207 
 
 that artistic presentation is a state of the artist's mind reflected 
 into stone or canvas or print. Scott tells us that he wore the 
 wishing cap and built castles with an unreal trowel ; Dickens 
 said that " he could hardly hear his own ideas as they came 
 into his head, and say what they meant ^ ;"" and Flaubert said 
 of his Saint Antohie that " he found himself well within his 
 own nature, and so had nothing to do except to go on^" It 
 would be a mistake, however, to accept all these statements 
 literally, and to suppose with G. H. Lewes that Dickens's art 
 is only a description of Dickens's private hallucinations''. 
 Dickens thought, indeed, that his stories developed with an in- 
 dependent growth of their own. " As to the way in which these 
 characters have opened out," he says, "that is, to me, one of the 
 most surprising processes of the mind in this sort of invention. 
 Given what one knows, what one does not know springs up, 
 and I am as absolutely certain of its being true as I am of the 
 law of gravitation ^" That is the psychology of Flaubert's 
 Saint Antoine, except in so far as Dickens, by referring to the 
 truth of his imaginings, claims more than subjective reality for 
 them. And it is interesting to notice that Flaubert consistently 
 thought that the subjectivity of his mood in Saint Antoine 
 was fatal to its art. " Tout doit se faire a froid, posement," 
 he said^ And again : " The more a writer is personal the 
 weaker he is. My besetting sin has been to put myself into 
 everything I have done. For example, I am where St Anthony 
 ought to be.... The less we feel a thing the better we are able 
 to express it as it is, but we must be able to make ourselves 
 feel it, and that is a kind of vision ^" 
 
 If these accounts can be trusted, it would seem that artistic 
 
 ^ Forster's Life, vol. i. p. 106. 
 
 2 Correspondance, IP Serie, p. 70. 
 
 3 Fortnightly Review, Feb. 1872. This is one of the curiosities of literary 
 criticism. 
 
 * The reference is to the characters in Martin Chuzzlewit. See Forster's 
 Life, vol. 11. p. 58. 
 
 5 " Je connais ces bals masques de I'imagination d'oil I'on revient avec la 
 mort au cceur, epuis^, ennuy^, n'ayant vu que du faux et d^bit^ des sottises; 
 tout doit se faire k froid, posement." Correnpoiidance, II' 8erie, p. 175. 
 
 « Ibid. p. 82.
 
 208 THE LARGER OUTLOOK [ch. 
 
 imagination does not seem a thing of merely subjective imagery 
 or private feeling to those who have it, but that they consider 
 it an independent growth which they themselves watch and 
 tend. Art, to be sure, is plastic. Pigments and words are 
 put together, often with labour, and this labour is as con- 
 structive as anything could be. Indeed, imagination is plastic 
 because it is esemplastic. Art, then, is creative, and it is a 
 product of the spirit. The guiding idea of the artistic whole, it 
 is true, may be vision and nothing else, but the artist must plan 
 and construct in terms of his guiding idea, and that is a kind 
 of creation, however 'inevitable'' the development of the theme 
 and of its incidents may be. On the other hand, we dare not 
 overlook the belief of so many artists that they feel themselves 
 ' dans le vrai ' in all their imaginings, and that they breathe 
 the spirit of something larger than themselves which beckons 
 to them and forces them to utter it. Is it not possible, then, 
 that the ' constructiveness ' of art is only a secondary thing, 
 and that the first commandment in art is just to accept the 
 beauty that is revealed to a man ? There is the highest 
 authority for this view, for it is written in the Symposium 
 that " if a man behold beauty with the eye of the mind he 
 will be enabled to bring forth not images of beauty but 
 realities, and bringing forth and nourishing true excellence 
 to become the companion of God and be immortal if mortal 
 man may^." 
 
 Art, because it is representative, may be a vehicle of know- 
 ledge as well as a discovery of beauty, and this representative- 
 ness, as we all know, is a very vexed problem. The perfect 
 work of art, it is clear, should be reposeful and complete 
 within itself, but art is also a medium of knowledge, and it 
 may have this function without ceasing to be art. It is well 
 that there need be no conflict here, since the products of art 
 always have a meaning and cannot be denuded of this potency 
 without disaster. Mansfield Park is a picture of the times, 
 Sarah Gamp the immortal picture of a type. It is our mis- 
 fortune that we never met Colonel Dobbin, but from another 
 point of view we might be privileged to meet him every day. 
 ^ Symposiwn, 212 (Jowett, vol. i. p. 582).
 
 ix] THE LARGER OUTLOOK 209 
 
 These problems of representation have been somewhat 
 neglected by realists because of the violence of their reaction 
 from theories which take knowledge to be nothing but re- 
 presentation. Realists see so plainly that representation in 
 knowledge cannot be the foundation of it, that they are 
 preoccupied with the non-representative basis of knowledge. 
 It is manifest, as we have seen, that representation cannot be 
 the whole of knowledge, for no one can know that anvthing 
 represents anything else without apprehending the represen- 
 tative non-representatively, and without apprehending the 
 connection between representative and original non-represen- 
 tatively. On the other hand, the perception of this vital truth, 
 and the inadequacy of any theory which neglects it (as so 
 many theories do) do not absolve realists from dealing with 
 the problem of representative knowledge. Our knowledge 
 must ultimately be based on direct acquaintance with things, 
 but most of it is indirect and works through the medium of 
 signs. Words themselves are signs, and they are also the 
 vehicles of knowledge. And illustrations are signs. It is futile, 
 therefore, to argue as if all knowledge were wordless knowledge, 
 or as if the selection of appropi'iate illustrations were not a 
 way of conveying a meaning which is true or false. 
 
 When symbols express fact, as they must at some point 
 unless they are mere counters, this expressiveness can be 
 recognised and known only if symbol, fact, and the relation 
 between them is recognised and known. Knowledge by 
 symbolic representation, therefore, is the apprehension of 
 symbols subject to the condition that the equivalents of these 
 symbols in fact and the character of their significance can be 
 apprehended at will. This knowledge, to be sure, is very hard 
 to decipher precisely in many cases. It was said of a certain 
 surgeon that he would bind up anything in a Thomas's splint 
 from a broken heart to the break of day, and I defy anyone 
 to tell me what precisely is before his mind when he hears 
 this remark. It is clear, however, that the sounds are not the 
 only objects of our apprehension. Even if Flaubert's dream 
 came true, and it became possible to Avi-ite books containing 
 nothing but phrases, the allusiveness and suggestiveness of 
 L. 14
 
 210 THE LARGER OUTLOOK [ch. 
 
 these phrases would not be utterly verbal. In reading a book, 
 for example, we are aware of the sounds and of the black 
 forms which signify them, but we commonly neglect the 
 letters. We do not neglect the sounds to the same degree, 
 it is true, and our mind is changed as well as our mood 
 according as the sounds are liquid or harsh, but we also attend 
 to the things signified by the sounds, and, indeed, we attend 
 to them chiefly. 
 
 What, then, of pictorial representation with brush or pen, 
 imaginary word-painting like Mr Conrad's Typhoon, or his- 
 torical word-pictures like Macaulay's third chapter.'' The 
 principle in these cases is fundamentally the same as in the 
 others. The skilful selection of incidents and illustrations, to 
 be sure, is only partially symbolic, since these, up to a point, 
 may be directly selective of perceptible fact in the most literal 
 sense. These incidents and illustrations, however, are chosen 
 for their suggestiveness rather than on their own account, 
 and so they are symbols first and foremost. To understand 
 them, therefore, we have to understand the character of 
 meaning; and here we may draw upon eai'lier discussions. The 
 meaning of all signs is an objective connection. When the 
 connection between the sign and the thing it signifies has 
 been apprehended the sign refers beyond itself, although there 
 need not be any detailed apprehension of the thing which is 
 signified. The reference of the sign, therefore, seems vague 
 and ill-defined, and it may remain so unless a peculiarly 
 resolute effort is made to discover what the sign signifies in 
 detail. Thus in artistic representation the mind lingers over 
 the incidents without following them out, and yet it knows 
 that they do lead somewhere. This attitude is easier than 
 the attempt to follow principles to their roots, and the 
 artist's goal is the ease of the reader or observer (although, 
 as we have seen, that is also his despair). And there is another 
 point. A work of art need not ' stick in its incidents ' and 
 illustrations. These are parts of a whole, and the whole may 
 signify as well as its members. But the principle of significance 
 is the same for the whole and for its parts, and this circum- 
 stance, consequently, does not need separate analysis.
 
 ix] THE LARGER OUTLOOK 211 
 
 Any discussion of the larger outlook in philosophy — even 
 a mere outline — must consider religious experience, for that, 
 even more than art, is the chosen way in which the human 
 spirit seeks to free itself from parochial shrewdnesses and the 
 cenotaph of a blind alley. It may be conceded, indeed, that 
 religion is fundamentally more emotional than reflective, and 
 that religious dogma and religious history are rather the result 
 of loyalty to the universe and of emotional acquiescence in 
 the value, unity and rhythm of being than the reason for them. 
 Even so, however, this loyalty and acquiescence may be better 
 guides than any parade of dialectics. Just as love and fellow- 
 ship, on the whole, may be better indications of the character 
 of our brothers whom we have seen than any recital of their 
 actions or analysis of their dispositions, so the love of the 
 Spirit we have not seen and fellowship with Him may reveal 
 more than libraries of argumentative wisdom. It is probable 
 then (it may even be certain) that God did not choose the 
 rationalists to bring salvation to His people; and the ration- 
 alists, for their part, may justly deserve censure for many of 
 their efforts to show that reverence can never clarify, and that 
 mysticism is always pathological or necromantic. Indeed, why 
 should we deny that religious experience, and particularly 
 mystical experience, is fundamentally a new way of knowing.'' 
 This claim is often made, sometimes, it is true, with diffidence 
 (since mystical descriptions so often are metaphors), but still, 
 on the whole, with a sort of deprecating firmness. Mysticism 
 need not claim to be a philosophy, but it often does\ and 
 when it does, it claims to be a philosophy based upon the 
 validity, and indeed upon the supremacy, of a distinctive way of 
 knowing. Such knowledge, we are told, is open to everyone 
 
 1 Cf. Inge, The Philogophy of Plotinm, vol. i. pp. 3-4. "Mysticism is the 
 pursuit of ultimate, objective truth, or it is nothing 'What the world 
 calls mysticism,' says Coventry Patmore, 'is the science of ultimates, the 
 science of self-evident reality.'... Thus it soon became clear to me that 
 mysticism involves a philosophy and at bottom is a philosophy." Per contra, 
 Miss Underbill: "Not to kiivw about but to Be is the mark of the real practi- 
 tioner" {Mysticism, p. 86). "Mysticism, then, is not an opinion ; it is not a 
 philosophy" (Ibid. p. 97). But Miss Underbill also quotes Coventry Patmore 
 with approval, and chooses the self -same passage as the motto for Part i of 
 her book. 
 
 14—2
 
 212 THE LARGER OUTLOOK [ch. 
 
 although it is cultivated by the few only, but mysticism is a 
 distinctive kind of knowledge whether it is catholic or esoteric. 
 
 Philosophical mysticism, therefore claims to be a distinctive 
 way of knowing ; and it is by far the most important contri- 
 bution which religion has made to this branch of philosophy. 
 It is necessary, then, to examine these claims which are made 
 so explicitly. 
 
 'Mysticism' itself is a vague word, and it is commonly used 
 to describe so much that it describes very little in particular. 
 To some it means the medicine-man and his wizard progeny, 
 the orendo of the Iroquois, and the levitations of ecstatic 
 Moslems ; to others the strange visions of ascetic vestals, the 
 umbra viventis lurninis of St Hildegard of Bingen, or the 
 bridal passion of Mechthild of Magdeburg ; and to a third 
 party it means the Zohar, the Kabbala, and the Rosicrucians. 
 Psychologists discuss the stigmata of hysterical ecstasy, and 
 the psycho-physics of the trance, or smudge the symbolism of 
 a celibate's dream with a prurient finger. Others, again, re- 
 gard mystical literature as a record of abnormal experience, 
 curious perhaps, but curiously regular since it shows well- 
 marked periods of spiritual growth whose stages correspond 
 very closely, not only between Catholic and Protestant, man 
 and woman, mediaeval and modern, but also between Orient 
 and Occident. 
 
 The psychology of mysticism, however, is not the philosophy 
 of it. Philosophical mystics, to be sure, gladly accept the 
 spiritual biographies of the subjective mystics (as they call 
 them) and maintain that these personal recoixls embody per- 
 manent truth with a universal meaning. On the other hand 
 they go to Plotinus (or, perhaps, to Lao-tse) for the philosophy 
 of mysticism, and not to Suso, or Tauler, or Macarius or 
 Ndvikov. Philosophical mystics, it is true, tread the mystic 
 way, even if they do not always time their progress according 
 to the quaint itineraries of the handbooks of devotion. They 
 may show a certain contempt for being in the body, as Plotinus 
 did when he refused to speak of his ancestry, his parentage 
 or his birthplace^ they may claim esoteric knowledge because 
 
 1 See the Life of Porphyry, 1.
 
 IX] THE LARGER OUTLOOK 213 
 
 of the thoroughness of their purification, they may empty 
 their mind to find the Godhead in it, and they may be con- 
 templative to the point of quietism. These characteristics, 
 however, are common to all mystics. They are not pecuHar to 
 philosophical mystics. Philosophical mystics pass through the 
 same experiences as the others, and welcome the records of this 
 companionship; but philosophical mystics are not content with 
 that, for they find knowledge according to a first principle — a 
 cirrus cloud that casts no shadows — where the others find 
 only a wonderful sweetness and a garment of rapture. 
 
 " The central doctrine of mysticism," Dr Inge says, " is not 
 that we can see God only in a state of swoon, but that we can 
 see only what we are\''"' That is a negative way of putting it. 
 The positive thesis of philosophical mysticism is that true 
 knowing is a way of being, so that in becoming such and such 
 we therefore know what we are. When these assumptions are 
 granted, the outlines of philosophical mysticism show them- 
 selves very clearly. In the first place, it is plain that the only 
 way to see God is to become Him. The mystic must reverse 
 the Incarnation, and become absorbed in the One, In the 
 second place, it is clear that personality, as we commonly in- 
 terpret it, is incapable of this absorption, and mystics therefore 
 hold that what we call personality is a temporary, superficial, 
 illusory makeshift. We really are identical with the One and 
 the Eternal whatever our commonplace imaginings may declare, 
 and the mystic's task is ultimately to find what he truly is. He 
 finds this, it is said, with the 'logic of the whole personality-,*' 
 purifying his soul from the dross of time and things, and find- 
 ing that it expands infinitely when these hindrances are cast 
 out into the wilderness of make-believe. Some of the mystics 
 find their goal after an intellectual catharsis, others by prayer 
 and fasting, others by a sage and wilful ignorance, others in 
 a tempest of emotion. But the unitive life beckons to them all. 
 
 It seems to be clear that this principle, as the Dean of 
 St Paul's states it, is not an arbitrary gloss on the mystic's creed, 
 
 1 Article ' Neo-Platonism ' in Hastings' Encyclopaedia of Religion and 
 Ethics, vol. IX. p. 316. 
 
 2 Dr Inge's phrase in his lectures on Christian Mysticism, p. 19.
 
 214 THE LARGER OUTLOOK [ch. 
 
 but a terse summary of its cardinal principle. " We can only 
 behold that which we are," Ruysbroek said\ and Plotinus him- 
 self expressed the doctrine with great exactness, for he said : 
 " He who then sees himself, when he sees, will see himself as a 
 simple being, will be united to himself as such, will feel him- 
 self become such. We ought not even to say that he will see,hut 
 he will be that which he sees, if indeed it is possible any longer 
 to distinguish seer and seen, and not boldly to affirm that the 
 two are one. In this state the seer does not see, or distinguish, 
 or imagine two things ; he becomes another, he ceases to be 
 himself and to belong to himself. He belongs to God and is one 
 with Him, like two concentric circles ; they are one when they 
 coincide, and two only when they are separated. It is only in 
 this sense that the Soul is other than God. Therefore this 
 vision is hard to describe. For how can one describe, as other 
 than oneself, that which, when one saw it, seemed to be one 
 with oneself-.''" 
 
 Plotinus admits, just as Hegel admits^, that we seem to be 
 confronted, in our common knowledge, with something not 
 ourselves which we simply accept and try to decipher ; but both 
 of them hold that such knowledge is spurious. Knowledge, 
 they maintain, is not genuine until it is literally the union of 
 mind and thing ; and realism, to put it succinctly, is the direct 
 uncompromising denial of this ideal. Realists take that to be 
 final which Hegel and Plotinus take to be marks of inadequacy ; 
 for knowledge, according to the realists, is always the discovery 
 of something with which the mind is confronted. The mind 
 is therefore distinct from its object, and an object is not known 
 the better because of its resemblance to mind. The dogma 
 that like can only be known by like is the first fallacy which 
 realists try to expose*. The doctrine that auT/ knowledge can 
 be the identity of mind and thing is a second and a greater 
 fallacy of the same kind. 
 
 ^ Quoted by Miss Underbill, The Mystic Way, p. 20. 
 
 2 Enneadg, vi. ix. Dr Inge's translation, op. cit. vol. n. p. 140. 
 
 3 Cf . the Preface and the Introduction to the Phenomenology. 
 
 * Cf. Plotinus, Enneads, i. vi. 9 (Mackenna's translation, vol. i. p. 89), 
 " Never did eye see the sun unless it had first become sunlike, and never can 
 the soul have vision of the First Beauty unless itself be beautiful."
 
 IX] THE LARGER OUTLOOK 215 
 
 In the face of an ultimate conflict of principle, argument is as 
 useless as soft words before a tempest. There can be no via media 
 between the doctrine that knowing is a kind of identity with 
 the thing known, and the doctrine that this is precisely what 
 knowing is not. But if concessions are impossible, intran- 
 sigeant denials are of little account unless they are accompanied 
 by an attempt to consider what may be said on the other side. 
 The mystics and the idealists have a cloud of witnesses to-day, 
 and most of them are very patient in explaining how their 
 opponents have grasped a part of the truth, and yet have 
 fallen short of the truth itself. This patience, therefore, should 
 be met by equal patience on the other side. 
 
 It may be argued, then, firstly, that union with and absorp- 
 tion in an object is the ultimate end of emotion and endeavour, 
 and so, by analogy, the ultimate end of knowledge; and, secondly, 
 that when mind attains a certain level of tension it is always 
 self-conscious, so that the mystics, when they attain this stage 
 of being, also and inevitably attain this stage of knowing. 
 
 The end of our willing, we are told, is the fruition of desire. 
 Such fruition is just absorption, and the mystic attains his end 
 when he is absorbed into the All. Emotion, again, is the same 
 in principle. Strong emotion, as we say, carries us away. The 
 mind is wholly absorbed in its interest, and mystical feeling is 
 just absorption in the beauty, harmony and life which envelop 
 and encompass the mystic. The existence of such emotions 
 show at one and the same time what the intellect is and how 
 inadequate it is. Knowledge is like literary criticism, a sort 
 of voluntary detachment from complete enjoyment. The critic 
 has to write in a calm mood, contenting himself with the fjiint 
 memories of his previous enjoyment of beauty. His criticism 
 does not count unless he has lived through the experience of 
 absorption and self-forgetfulness, but he cannot be a critic 
 while he is pulsing with the joy of beauty. Reflection is only 
 the pale ghost of full apprehension and absorption. It is a 
 parasite upon the real experience. It is the lingua franca of 
 relaxed minds. And it is important in connnon affairs pre- 
 cisely because we keep our minds relaxed for nine-tenths of 
 our lives.
 
 216 THE LARGER OUTLOOK [ch. 
 
 Arguments of this kind abound in general literature, and 
 they are certainly held with intense conviction. One would 
 suppose, therefore, that they are better than some of us think. 
 Be that as it may, there are certainly grave difficulties in the 
 premises of them, and the gravest doubt of the conclusion 
 granting the truth of the premises. This theory of the will 
 outstrips sober description. When we resolve to do anything 
 we set ourselves to bring about a change which we believe we 
 can effect, but we may desire and strive for other things than 
 changes in our own person, and we can effect more than a 
 private transformation. We can struggle for the good of others, 
 or for the happiness of posterity, or for the honour of the 
 realm. We really seek these ends, and not merely our personal 
 satisfaction or absorption in them ; and those who live for 
 posterity neither expect nor desire absorption in posterity. 
 
 The mystic's analysis of feeling, again, is less securely founded 
 than many suppose. Mysticism is the apotheosis of love, and 
 the records of experience testify with surprising unanimity that 
 love is a godlike search for literal union. But is it so ? Love 
 looks for harmony; it is careless of itself; it seeks its other. 
 Yet it stops short of absorption, and that is its salvation. If 
 a lover became his beloved, or if creaturely devotion became 
 divinity, the excellence of the lover would be lost and only the 
 excellence of his beloved remain. It is better to think that 
 love achieves its perfection, not by relinquishing a lover's ex- 
 istence, but by attaining a complem.entary being in which 
 neither the lover nor his beloved is absorbed. 
 
 Even granting, however, that absorption is the end of seeking 
 and the consummation of emotion, it is strange to maintain 
 that it is also the end of knowledge. Analogies are manifestly 
 impotent to override a contrast in the facts, and nothing could 
 be plainer than the contrast between knowledge and emotion 
 when emotion is interpreted after the fashion of this argument. 
 If the reflective attitude, detached, freed from tension, rechmiffe, 
 is just the antithesis of living absorption, surely the conclusion 
 ought to be that reflection is always inadequate and incom- 
 plete, and not that reflection, at its highest, forgoes every- 
 thing that we mean by it. To say that reflection dare not be
 
 ix] THE LARGER OUTLOOK 217 
 
 impassioned, and also that passion and reflection (at its 
 highest) are one, is too like a flat contradiction to be tolerated. 
 And so we may pass to the other argument. 
 
 According to it, there is no difficulty in seeing that the 
 mystic's way of being is also a way of knowing. It is not our 
 knowing, to be sure, nor even self-consciousness after our pat- 
 tern, and therefore it has to be described either negatively or 
 figuratively. The One is above Soul and above Existence. 
 It is ineffable and incomprehensible. In knowing it we have 
 to ' unite ourselves in unknowing,' like the pseudo-Dionysius, 
 or, as St Bernard says, *' When something from God has 
 momentarily shed its ray upon the mind, immediately, whether 
 for the tempering of this too great radiance or for the sake 
 of imparting it to others, there present themselves certain 
 imaginary likenesses of lower things suited to the meanings 
 which have been imparted from above," None the less, if con- 
 sciousness at a certain level always includes self-consciousness, 
 it is reasonable to argue that to be one with God is to be 
 united with His Self-omniscience, and so to be one with His 
 being and with His knowing. 
 
 This argument is precisely the reverse of the former one. 
 That maintained that absorption means loss of self-conscious- 
 ness, this maintains that absorption in God is transfigured 
 self-consciousness. On the score of fact, if fact can be spoken 
 of in this connection, the second argument seems nearer the 
 truth, but even if union with an Omniscient Self-consciousness 
 implied participation in Its self-knowing after the manner of 
 Its perfect knowledge, the problem would still remain whether 
 God's self-knowledge must be different in kind from other 
 knowledge. These exalted matters can scarcely be put to the 
 proof ; and so we have to consider what we find in experience, 
 and to argue from experience as well as we can. Self-conscious- 
 ness is plainly a fact of experience, and our former analysis of 
 it ought to help us. If self-consciousness differed in kind from 
 any other form of knowing, the mystic might have very good 
 grounds for his belief that, in the end, knowing and being are 
 one. But since it does not, his contention falls. 
 
 14—5
 
 EPILOGUE 
 
 "Qffwep Kal QaKrji' AcTTpouofiovvra, w Qeddupe, Kal dvw ^Xeirovra, ireuhvTa. eiy 
 <f>piap, Opg.TTd rtj ififieXris Kal xa/'^«c<''a depawaivU avoaKQipai X^yerai. ws ra fikv 
 fv ovpavw irpodvfj.oiTO eldivai, ra 5' ^ixirpoadev avrou Kal irapa 7r65aj \avdavoi 
 avrbv. Taiirhv 5k dp/cet cTKCofifia iirl irdvTas 6(toi iv <pi\o(Jo<piq. 8idyovai. 
 
 Plato, Theaetetus, 174 a. 
 
 Long ago, as Thales stood gazing at the stars, he fell into 
 a well ; and his handmaiden laughed, saying he was so rapt 
 in the things of heaven that he could not see the earth at his 
 feet. Then Thales called her, for she was witty and comely, 
 and he bade her bring a salve for his hurt, and wine and some 
 bread. 
 
 So she tended his hurt, and Thales spoke with her, and said : 
 " Pretty one, dost thou jest at this star-gazing like the rest 
 of the Thracians ? " Then the maiden laughed again, and 
 answered, saying : " O subtle Ionian ! Canst thou find this 
 salve in the moon, or draw wine from the stars; and how 
 couldst thou live if thine handmaid did not serve thee ? " 
 
 And Thales said : " Pretty one, how did I find this salve ? 
 Hast thou not heard how I foresaw that the harvest of olives 
 would be great beyond all expecting, and how I hired all the 
 olive presses in Miletus and Chios, paying the earnest-money 
 for them, so that, when the harvest came, I could ask for what 
 I would, and take my toll of all the oil that was pressed? 
 And the maiden answered : " Verily thou art a cunning man, 
 yet I would thou didst not cheat the poor. For all men must 
 have oil." 
 
 Then Thales said: "The poor need not lack till they die, and 
 we who are not born to riches must seek for riches as we find 
 the means. Yet I would fain have thee see that my star-gazing 
 is not a thing of no account. For because of my learning in 
 the stars I knew that the day would turn into night, and so 
 it befel while the Lydians fought with the Medes, Then the 
 slaughter was stopped, and peace was made, and fathers had 
 not to bury their sons any more. Tell me, was not this a great 
 thing, and did I not find great honour in it ? ""
 
 EPILOGUE 219 
 
 And the maiden answered lightly : " Verily thou art wiser 
 than any in Ionia, and thou hast gone into far countries, and 
 hast learned to tell how far away the ships be from thy steep 
 tower, and by thy learning thou didst turn the Halys from its 
 course so that the army of Croesus passed it in safety. For 
 these things I honour thee, although I am but a Thracian and 
 thine handmaid. Yet thou art foolish and fallest into wells, 
 so that I must tend thee as a mother tends a child. And this 
 thing passes my wit, and thy wit too, O sapient infant ! And 
 now I must leave thee, for thy star-gazing does not bake bread, 
 and thou wilt chide me an I bring thee not thy bread at sun- 
 rise, even if thy stupid stars pay no heed. And, prithee, keep 
 away from the well." 
 
 So the maiden left him, and Thales mused awhile and fell 
 asleep. And as he slept he saw one coming to him with glass 
 over his eyes and clad in leather. Then the stranger said : 
 *' Arise, and come with me." So Thales arose and went with 
 the stranger, and they entered into a strange chariot with 
 great sails fastened to it. And lo ! they rose in the air, and 
 sped over the land, and the noise of their going was as the 
 droning of countless swarms of bees. 
 
 And the stranger showed Thales the source of the Nile, 
 whereon Thales had often pondered, and huge armies, and 
 many other marvels. Then Thales rejoiced because he saw 
 what men had done when they looked beyond their feet, and 
 sought out cunning devices, even as he had done. 
 
 And he spoke to the stranger, thanking him, and said : 
 *' I would fain know more of this knowledge, so that the oracles 
 can learn of me and kings do my bidding," And the stranger 
 said : " I shall take thee to a far country beyond the Middle 
 Sea, and the sages of that country will tell thee these things. 
 But, first, I must consult my own oracle," 
 
 Then the stranger put a band over his ears, and Thales heard 
 the sound of tapping. And after a little while the stranger 
 told him that the sages in the far country were ready to receive 
 him. And Thales marvelled greatly at this oracle, but he kept 
 silence. 
 
 So they journeyed many hundreds of leagues through the
 
 220 EPILOGUE 
 
 air, and the stranger told Thales how he kept the course. And 
 Thales understood him and was glad. And when they came 
 to the far country they lighted on a hill above a river. And 
 there was a palace on the hill and ti'ees around it. Then the 
 stranger left Thales at the gates of the palace, and Thales 
 went to the door and asked if he might see the king. But the 
 chamberlain told him that the house was no palace, only the 
 abode of a sage, and that the sage would rejoice to welcome 
 him. So Thales entered, and the sage told him how he read 
 the stars, and showed him glossy pictures of the stars, and 
 many charts, and great tubes to see them withal. 
 
 Then Thales marvelled and enquired concerning the stars, 
 and how men could foretell their courses. And the sage showed 
 hira many figures and diagrams, and explained the proofs of 
 them. And Thales said : " Often did I dream of these things, 
 and now thou hast shown them to me. And this knowledge 
 I count higher than any invention. For it is godlike. So now, 
 I pray thee, tell me of the gods, and show me how men fare. 
 For the knowledge of the gods is the most excellent of all 
 knowledge ; and yet I am a man, and I would learn how men 
 are made happier by this wonderful knowledge and these many 
 inventions." 
 
 And the sage said : " I can show thee mankind, but thou 
 must seek other sages to learn of the gods." So he showed 
 Thales a great city, and the men in it. And the men seemed 
 to Thales to be weary, and dingy, and busy without zest, and 
 clad in ugly raiment. And Thales said : " These be slaves, 
 and I would fain see the rulers of the city." But the sage told 
 him that these were no bondmen, but free, and that there 
 were rulers among them. Then the heart of Thales was heavy 
 within him, and he held his peace. 
 
 So the sage took Thales to see those other sages who might 
 tell him of philosophy and of the gods. They journeyed in 
 a chariot without horses, and the sages received Thales 
 gladly. 
 
 Now the first sage was bearded and courtly, and the second 
 sage was ruddy and hard of countenance and quick in his move- 
 ments, and the third sage was lean and tired and wistful. And
 
 EPILOGUE 221 
 
 all the sages were clad like the men in the city. And they 
 burned strange incense. 
 
 Then Thales enquired of them concerning the gods. And 
 the first sage said that all things were one, and that this was 
 the sum of our knowing. And the second said that knowing 
 was doing, and the truth what men had made. And the third 
 sage said that nothing could be known concerning the gods, 
 and that all knowledge was but a learned fable whereat a wise 
 man should be merry. 
 
 Then Thales was sad because these sages had learned so little 
 concerning the gods. For the first, he thought, did not reason 
 closely, though the sound of his words was good; and the third 
 was too weary to think ; and the second was little better than 
 the Thracian handmaid. So Thales bowed to the sages and 
 withdrew himself from them. 
 
 And after he left them, he mused very sadly, thinking that 
 cunning inventions had increased, and much knowledge con- 
 cerning numbers and the stars, and yet that he had seen men 
 to be no happier than they were in Miletus, and that he had 
 found they knew very little of the gods. But Thales knew that 
 this knowledge was the most precious of all excellent things. 
 And he knew that if men walked warily in their knowing there 
 was nothing they might not find out. 
 
 Then the vision faded ; and when Thales awoke the stars 
 were pale in the sky, and all was still save for the note of a 
 grasshopper near at hand. So Thales called his handmaid and 
 bade her bring him much wine. For he was weary, he said, and 
 had seen strange matters in a vision. And the girl laughed 
 again.
 
 INDEX 
 
 Absolute, the, and absolutism : op- 
 posed to realism, 2; attitude to 
 logic, 12 sqq. ; and perception, ABsq. ; 
 and the organisation of experience, 
 113 sqq.; concerning reason and 
 sense, 117, 120*7.; concerning in- 
 dividuality and value, 145 sqq. ; de- 
 fect in the logic of, 179; and con- 
 struct! veness, 182, 202 sq.; and the 
 philosophy of history, 198 ; in sum, 
 221. 
 
 Absorption in an object: 215 sqq. 
 
 Abstraction: 61, 110. st^. 
 
 Act and object: 170 sq. 
 
 Activity, awareness of: 170. 
 
 Analogy : Law of, 123 ; and sympathy, 
 188; and discoveiy of connection, 
 203. 
 
 Analysis: legitimacy of, 90 sq. 
 
 Anselm, St.: 117, 118 n. 
 
 Apperception : unity of, 153, 174. 
 
 Arnauld : 1 sqq. , 13 sq. 
 
 Art : and fancy, 60 sq., 66, 68, 82 ; and 
 realism, 180 sq., 204 sqq.; in pic- 
 torial representation, 195 sq. 
 
 Association: in symbolism, 34 sq. ; in 
 memory, 55; character of, 72 sqq. ; 
 in beauty, 130; according to be- 
 haviourism, 160. 
 
 Attention: movement of, 48; Dr 
 Ward's interpretation, 164 sq.; 
 character of, 16Qsqq.; to ourselves, 
 7 sq., imsqq. 
 
 Awareness of awareness : 169 sqq. 
 
 Bartholinus: 183. 
 
 Beattie, James : 4 s^. ; on Berkeley's 
 'ideas,' 63. 
 
 Beauty: a value, 125 «g. ; whether 
 subjective, 126 sqq.; conclusion con- 
 cerning, 134 s(/., 144. 
 
 Behaviour: distinguished from con- 
 sciousness, 152 sq. ; and behaviour- 
 ism, 155 sqq. 
 
 Being: technical meaning of, 109; 
 the 'being' of general facts, 114 sqq. ; 
 beingand knowing (see ' intuition '). 
 
 Belief: 83 .sggi. (see 'judgment') ; ex- 
 perience of, 167, 171. 
 
 Bergson : his theory of intuition, 8sqq., 
 169, 187: his influence on psycho- 
 analysis, 75. 
 
 Berkeley: Reid's criticism of, 2sqq., 
 13; his Hylas, 28; his 'divine 
 visual language,' 34; criticised by 
 Beattie, 63; on 'inner' and ' outer,' 
 162; on the idea of activity, 170; 
 his Siris, 180. 
 
 Bernard, St. : 217. 
 
 Biology: critical of evolution, 
 
 151 sq.; and consciousness, 
 155sqq.; methodsof and knowledge 
 in 181, 186 sqq. 
 
 Blake : on fairy funerals, 68. 
 Bosanquet : on individuality and value, 
 
 145 sqq.; on the self, 177 sqq. 
 Bradley, F. H. : and Mr Moore, 2 ; on 
 
 the assumptions of logic, 91. 
 Broad: on probability and induction, 
 
 121 sq. 
 
 Causal interpretation : 86 ; its divi- 
 sions, 95; Hume's view of, 95 sqq. ; 
 in perception, 99; the laws of, 
 100 sq.; conclusion concerning, 
 102. 
 
 Chalmers, DrT.: on German philo- 
 sophy, 6 7). 
 
 Champeaux, William of : 115 n. 
 
 Chartres, Bernard of : 115". 
 
 Chronology: and history, 193>s(2. 
 
 Clay, E. R. : on the specious present, 
 45 ?!. 
 
 Clay, Henry: his definition of econ- 
 omics, 191. 
 
 Clearness : and attention, 166. 
 
 Coleridge : on fancy and imagination, 
 60; an the images in Kubla Khan, 
 64 sq.; on ' esemplastic ' imagina- 
 tion, '204. 
 
 'Complication': 24 8^., 40. 
 
 Conrad: on imagination, 60«(;.; his 
 word-painting, 210. 
 
 Consciousness: and unconsciousness, 
 
 152 sq.; togetherness of, 153; and 
 self -consciousness, 153 sq . , 169 sqq . ; 
 its utility, 154 sqq. ; and nervous in- 
 tegration, 155 sqq.; its functions, 
 157 sqq.; and psychology, 159; its 
 nature, 160 ; and introspection, 162 ; 
 and its objects, 162sqq.; and the 
 brain, 173 ; its capacities, 173 sq. ; 
 its 'generative activities,' 174 sg. ; 
 its temporary annihilation, 176 s<7.
 
 224 
 
 INDEX 
 
 Construction: in memory, 51 sg . , 56*7. ; 
 in the fancy, 74:sqq.; in logic, 91; 
 of organising principles, 112 igqq. ; 
 in probability and hypotheses, 
 182 .tqq.; and the theory of know- 
 ledge, 201 sqq.; artistic, 205 sqq. 
 
 Continuants: explained, 27 sqq.; and 
 sign-facts, 36; relation to perception 
 and judgment, 42; of common be- 
 lief, 94 s(/.; in self-hood, 174. 
 
 Continuity: a descriptive term, 26; 
 spatio-temporal-causal, 94sqq.; of 
 development, 152; of nerves and 
 mind, 157 sq. ; and the identity of 
 things, 176; and the intellect, 187. 
 
 Continuum: thepresentational, 25x^7. 
 
 Correspondence: in truth and error, 
 103. (See also • representation ' .) 
 
 Cousin, V. : his Philosophie Ecossaise, 
 5.s(/. 
 
 Croce : on history, 200. 
 
 Dane,Clemence: on imagination, 205. 
 
 Dating: of memories, 5i sq., 71 sq. 
 
 Descartes: 3, 117. 
 
 Development: of perception, 16, 
 25 sqq., 40 sq. ; and acquired mean- 
 ing, 31; and evolution, 151 sq.; in 
 history, 19S sqq. 
 
 Dickens: Jeffrey on, 127; hismethods, 
 206*5-. 
 
 Discovery: knowledge as, 14: in per- 
 ception, 30877. ■> in memory, 51 S77.; 
 in fancy, 69 S77. ; in belief 86 S77. ; 
 of constructions, 81, 184, 202; and 
 organisation, 113. 
 
 Dreams: as presented fact, 61; and 
 waking, 65, 68; and psycho- 
 analysis, 75 sqq.; illusion in, 82. 
 
 Economic theory of history : 198. 
 
 Economics: type of knowledge in, 
 189 S77. 
 
 Einstein: 93. 
 
 Emergent: 157 S7. 
 
 Empiricism: historical, 107 87. 
 
 Error: 103 S7.; in perception, 41 .S7. 
 
 Existence: and subsistence, 50, 
 109 sqq.; organisation of , 112 S77.; 
 of mathematical facts, 116 S77. ; in- 
 cluded in subsistence? 117 .S77.; 
 non-spatial, non-temporal, 119; and 
 value, 125, 144 S77. 
 
 Expectation : primary and secondary, 
 49 «77. ; compared with memory, 
 53; scepticism concerning, 58; 
 justification of, 100. 
 
 Experience: 'pure,' 8S7. ; organisation 
 of, 112 S77. 
 
 Fabre: on beetles, 51; on pine-cater- 
 pillars, 158. 
 
 Fancy: and memory, 57, 69 .S77. ; and 
 imagination, 6OS77., 192, 204 877.; 
 general theory of (see images). 
 
 Feeling: and judgment, 130, 133*77. ! 
 moral, 137*77. ! nature of, 164 .S77. ; 
 in economics, 192 «7.; in history, 
 200; in mysticism, 213 S77. 
 
 Flaubert : his art, 201, 204, 207, 209. 
 
 Flux : on economics and psychology, 
 191. 
 
 Freeman : on the Norman Conquest, 
 195. 
 
 Frege: on arithmetic, 120, 120?i. 
 
 Freud : on association, 75. 
 
 Galton: on images, 62. 
 
 Gaunilo : on the ontological argument, 
 118, 118 7i. 
 
 General facts : 105 S77. (and see prin- 
 ciples). 
 
 Geography: and history ; 197 «7. 
 
 Givenness: in perception, 3I.S77.; 
 overlooked by the 'new realists,' 
 164; in knowledge, 202*7. 
 
 Grote, J.: on perception, 15. 
 
 Green: on Synthesis and Eelating, 32. 
 
 Hales, Stephen: and oxygen, 90 7i.; 
 quoted, 188. 
 
 Half-recollection: in association, 73. 
 
 Hamilton, Sir W. : on the past, 49. 
 
 Hegel : his Phenomenolgy, 8, 140, 
 214 ; on abstraction, 110; on reason 
 and sense, 117, 120; his monism, 
 178; on the philosophy of history, 
 197*77. 
 
 History : and knowledge, 181, 200 S7. ; 
 and economics, 189, 193 ; and nar- 
 rative, 193*77.; pictorial method 
 in, 196 ; the philosophy of, 197 S77. 
 
 Hume : criticised by Eeid, 2 S77. ; and 
 by Beattie, 63; on images never 
 sensed, 69; on 'judgment,' 83, 
 85*7.; on cause, 95*77.; ^^ exis- 
 tence, 118 ; on morals, 138; parodied, 
 166. 
 
 Hutcheson : on moral sentiment, 138. 
 
 Huyghens : his scientific imagination, 
 183. 
 
 Hypothesis : and realism, 182 S77. 
 
 Idealism: and realism, 2, 13*7., 
 204*77.; Anglo-Hegelian, 9; inter- 
 pretation of images, 61, 64, 73*7. ; 
 on universals, 112 S77. ; on reason 
 and sense, 117*77. > '^^ individuality 
 and value, 145*77-; on mind as
 
 INDEX 
 
 225 
 
 Idealism (cont.) 
 microcosm, 160 sq. ; on personality, 
 177»(/.; and monism, 179; scope 
 of, 180 s</. ; on construction, 202 sq. 
 
 'Ideal theory': Reid's criticism of, 
 3 sqq. 
 
 Illustmtion : and meaning, 196s'/. ; 
 as representation, 209 sq. 
 
 Images: in memory, 51, 54 sqq., 
 69 sqq. ; function of, 61 ; types of, 
 61 sq. ; mimicof sense, 62; status of, 
 63; contrasted with perception, 
 65 sqq. ; meaning of, 69, 82 ; theory 
 of 69*i^(/.; in association, 72 ; con- 
 structive, losqq. ; in literature, 129 ; 
 and the 'inner' world, 162 sg. 
 
 Imagination: and imaging, ^0 sq., 
 192, 20\sqq.\ in science, 1%2 sqq. , 
 203 ; in economics, 192 ; in history, 
 200 sq.\ 'captive' and 'free,' 
 2Qisq. ; in art, 207 sqq. 
 
 Individuality: and value, 145 sqq.; 
 for idealism, 111 sqq. 
 
 Induction: theory of , 121 «(/</. 
 
 Inge: on personality, 177 re. ; on 
 mysticism, 211 «., 213, 214 re. 
 
 Instinct : and behaviour, 152 ; and in- 
 telligence, 158 s(/. 
 
 Introspection: a description of fact, 
 152 ; plain man's view of, 162 ; not 
 concerned with the objects of con- 
 sciousness, l%bsqq.; objections to, 
 \QSsqq. ; importance for psychology, 
 172. 
 
 Intuition : and knowledge, 10 sq. ; and 
 introspection, 169; inapplicable in 
 biology, 187 s</.; and mysticism, 
 213 sgr/. 
 
 James, W. : on ' pure ' experience, 8 ; 
 unverbalised meanings, 29 ; on the 
 'specious present,' 45 re. ; on mem- 
 ory-images, 54; and psycho- 
 analysis, 75; on activity, 170 re. 
 
 Jourdain, P. E. B. : on the Law of 
 Parsimony, 123/). 
 
 .Judgment : and perception, 15, 27, 29, 
 42, 84; of nature, 83; and the 
 world, %'isqq.\ in Hume's sense, 
 85 re., 86, 98 ; analysis of, 86 sqq. ; 
 and 'objectives,' 87, 89; selective, 
 88 ; objections to theory of, 88 sqq. ; 
 incomparable importance of, 91 ; 
 chainsof, 92 ; of value, 125, VAl sqq., 
 143; and feeling, 130, 134; of 
 beauty, 132 s</.; moral, 138s(/(/. 
 
 Jung : on association, 72 ; and sym- 
 bolism, 75; on dream analysis, 
 77. 
 
 Kant : on synthesis, 32; on causation,, 
 97; on I'eason and sense, 117; on 
 the principle of variety, 124; on 
 moral obligation, 139, 143 ; on 
 unity of apperception, 153, 174 sg.; 
 on awareness of awareness, 170 sq. 
 
 Kirchoff: on description, 184. 
 
 Kiilpe: on the discriminable colours, 
 116. 
 
 'Lateral' aspect of the self: 111 sqq. 
 
 Laurie, S.S.: on conscious subject, 
 161. 
 
 Leibniz : and Arnauld, 3 ; on reason 
 and sense, 117; and mathematics, 
 120; and monism, 178. 
 
 Levy-Bruhl: on pre-logical generality, 
 101. 
 
 Lewes: on Dickens, 207. 
 
 Locke: and the 'ideal theoiy,' 3; 
 programme of his Essay, 7; toler- 
 ance in religion, 13; on introspec- 
 tion, 168. 
 
 Logic: and realism, 12, 181 s^r.; in 
 signs, Msq.; andassociation, 12sq. ; 
 Mr Bradley on, 91; and the structure 
 of things, 92 sq., 102, 116 sqq.; of 
 primitive propositions, 106 sq. ; and 
 'being,' 109 sqq.; in our habits, 
 113; a priori, 116; freedom of, 
 120 ; ' pure ' and inductive, 121 sqq. ; 
 of identity, 175; and hypotheses, 
 184 sq. ; and life, IHlsq . ; in political 
 economy, 189*7/.; 'of the whole 
 personality,' 213. 
 
 Lotze : on synthesis and relating, 32. 
 
 Lucas, E. V. : quoted, 72. 
 
 Macaulay : his third chapter, 196 sq., 
 210. 
 
 Macran, Professor: on idealism, 105. 
 
 Malebranche: his representative 
 theory, 2sq. 
 
 Margin, of perception : 22 sqq., AOsq., 
 msq., 61 sq., 94. 
 
 Marshall, Professor: Idl sq. 
 
 Materialism : and universals, 110 ; 
 and consciousness, 177. 
 
 Matter: whether perceived directly, 
 36 sqq. 
 
 Meaning: in perception, 16; and sen- 
 sory atomism, 21, 40; and the 
 presentational continuum, 26 sqq. ; 
 temporal, 48; of images, 68sqq., 
 82; in dreams, 18 sqq.; and judg- 
 ment, 93 sqq.; causal, 98.sr/7.; pre- 
 causal, 101; in beauty, 129 sq.; 
 representative, 209 sq. (See also 
 'sign-facts,' 'signs'.)
 
 226 
 
 INDEX 
 
 Mechanists: 186 s^. 
 
 Meinong: on 'objectives,' 84, 87. 
 
 Memory: and perception, 45 sr/^., 
 55 .sq-^., 8'dgqq. ; and direct acquain- 
 tance with the past, 49877.; and 
 reproduction, 51, bQaqq.; and re- 
 collection, 51 xqq-; memory-images, 
 51, 54 sqq., 69s(/7.; illusions of, 
 51 nq.; physiology, of 58 .s^. ; and 
 judgment, 91 sq. ; and conscious- 
 ness, 153, 162, 173 sgg. 
 
 Menger : on economics, 189. 
 
 Meredith: on Tennyson, 127; on 
 realism and idealism, 204 sq. 
 
 Mill, J. S. : on generalisation, 108. 
 
 Monboddo, Lord: quoted, 45. 
 
 Moore, G. E.: his realism, 2, 13 «g.; 
 on 'diaphaneity' of consciousness, 
 171 n. 
 
 Morals: and value, 12b sq.; whether 
 subjective, ISSsgij.; for realism, 
 181. 
 
 Multiple personality : 175. 
 
 Mysticism: described, 211 sg.; on 
 knowing and being, 54, 169, 213 
 gqq. 
 
 Necessary connection : 97sqq. 
 
 Nervous system, the: and perception, 
 17, 30; and memory, 59; and 
 dream-images, 66; and conscious- 
 ness, 155 sqq.; and the self, 173. 
 
 'Neutral monism ' : Mr Holt's, 118. 
 
 Newman, Cardinal: on belief in 
 dogma, 83. 
 
 ' New realists ' : on ' inner ' and ' outer, ' 
 162; Dr Strong's criticism of, 164. 
 
 Newton, Sir Isaac: glamour of, 7. 
 
 Nicholson, J. W. : on the Quantum 
 Theory, 94n. 
 
 Objective aspect of experience : 
 
 16Asqq. 
 Objectives: according to Meinong, 87; 
 
 and sensoi-y atomism, 89; in error, 
 
 lOB sq. 
 Object of consciousness: 167 sq. 
 Obligation: moral, 135 sgg. 
 Occam's razor: and imperceptibles, 
 
 37; and the Law of Parsimony, 
 
 123 sq. 
 Ontological Argument, the: for An- 
 
 selm and Kant, 117. 
 Oswald, Dr James: and common 
 
 sense, 4. 
 'Outer ' and ' inner ' : concerning con- 
 sciousness, 162 sq. 
 
 Parsimony, Law of: 123sg. 
 
 Past, present and future : in a stretch 
 of transience, 47. 
 
 Patmore, Coventry: quoted by Inge 
 and Underbill, 211 n. 
 
 Perception: and contact, 3.sr/.; of 
 external world, 15 sqq.; what? 15. tq. ; 
 whether direct, 17; immediate ob- 
 jects of, 17 sqq. ; and sensory 
 atomism, 18 sqq., 38 sqq. ; and dis- 
 crimination, 20 sq . ;neglect in, 21 sq.; 
 margin of, 22 sqq., 40 sq.; and the 
 presentational continuum, 25 sqq.; 
 fragmentariness of, 26; of physical 
 things, 27 sqq. ; meaning of, 29 sqq., 
 40. ^q.; discovery in, SO sqq.; of 
 matter, 'SQsqq.; improvement in, 
 40 sq. ; theory of, 42 sqq.; and judg- 
 ment, 15, 2d sq., 41 sq., 83 sqq. ; and 
 realism, 149; and the 'new realism,' 
 162 sq.; and attention, 167; in 
 economics, 192. 
 
 Percepts : in relation to perception 
 and judgment, 42, 56, 88 sg.; and 
 images, 67; and sensory atomism, 
 67. 
 
 Persons: and moral obligation, ISlsq. ; 
 as spiritual substances, 172 sq. ; 
 identity of, 175 sqq. 
 
 Physical things : howperceived, 27 sq., 
 36; meaning of, 38, 74; and the 
 world, G7 sq., 84 sqq.; and- belief , 
 8'6 sq., 86 sqq. 
 
 Pillsbury: on perceiving types, 22. 
 
 Plato : tiie ideas of, 114, 117, 120: ac- 
 cording to Inge, 177 sq. ; on beauty, 
 208; his Theaetetus quoted, 218. 
 
 Plotinus: a philosophical mystic, 
 212 ; his Enneads quoted, 214. 
 
 Pluralism: logical, 14, 149; in judg- 
 ment, 90 sg. ; and monism, 178 sg. 
 
 Pragmatists, the: on knowledge, 
 10 sqq., 202; on the function of 
 principles, 113 sqq.; on constructive- 
 ness, '202 ; in sum, 221. 
 
 Prelogical generality : 101, 107 sq. 
 
 Priestley: on the Scottish triumvirate, 
 4sq., 5n.; and oxygen, 90. 
 
 ' Primary ' and ' secondary ' qualities : 
 in perception, 17, 3687. 
 
 Principles : and particular facts : 
 105s77., 121 sqq. ; primitive and de- 
 rivative, 107 ; how discovered, 
 107 sq.; nature of, 108 S77.; regula- 
 tive function of, 113 ; a priori, 
 116; whether mental, 120 sq. 
 
 Probability: and induction, 12IS7.; 
 and realism, 182 sq. ; in history, 194. 
 
 Propositions: believed, 84; analysis 
 of, 86 sqq.
 
 INDEX 
 
 227 
 
 Psycho-analysis: and dreams, 67, 
 75sqq.; and association, 72 »</. 
 
 Psychology: and realism, 150 «</., 
 172, 179, 181 ; and knowledge, 179 ; 
 in economics, 191. 
 
 Quantum theory: in physics, 94. 
 
 Ealeigh, Sir W. : on sea-power, 195. 
 
 Realism : the word, Isq.; mediaeval, 
 2, 1 15 7i . ; modem and contemporary , 
 2sqq.; a phenomenology, 8, 12; 
 assumptions of, 8sqq., 181«(/.; in 
 perception, SOsqq. ; in memory and 
 expectation, idsqq. ; in imaging, 65, 
 81 sq.; and the physical world, 
 84sqq.; and judgment, 91; and 
 beauty, 129 »<;(/. ; and value, 146; 
 and knowledge, 149 sq., 214; and 
 logical pluralism, 179 ; and classical 
 philosophy, 180sqq. ; in economics, 
 190; and narrative, 195 ifqq.; and 
 imagination, 204 sg.; and repre- 
 sentation, 209. 
 
 Eeason : non-sensuous, 32 ; and sense, 
 in sqq. ; and politics, 140; and sal- 
 vation, 211. 
 
 Reflexes: 155 sg. 
 
 Reflection, 'way of: Reid's, &sq. 
 
 Reid : his philosophy, 2 s(/f/., Idsq.; 
 on memory, 49; quoted, 83, 
 
 Representation: in knowledge, 3 sg., 
 11 sq., 186, 196 sq., 206, 208 sq. 
 
 Ribot: on images, 62. 
 
 Ricardo : and abstract economics, 
 189; on 'free' competition, 190. 
 
 Roscellinus: 115 n. 
 
 Russell : and realism, 2, 13 ; and con- 
 tinuity, 26; on 'perspective space,' 
 94; on time, 95; on general facts, 
 105 sq.; on the basis of arithmetic, 
 120; a subjectivist in moral theory, 
 138; on actand object, 170 sgr., 171 n. 
 
 Ruysbroek: on knowing and being, 
 214. 
 
 Sainte-Beuve : on Arnauld, 3, 3n. 
 
 Saintsbury : on delight in aesthetics, 
 126 ; on ranks in poetry, 128. 
 
 Scepticism : general, 43, '221. 
 
 Schmoller : on method in economics, 
 189. 
 
 Scholastics: on realism, 2; on repre- 
 sentative perception, 3 ; on signs 
 in perception, 25 ; on non-sensuous 
 Reason, 32. 
 
 Scott, Sir W.: his memory, 51; and 
 Saunders Fairford, 206 ; and the 
 wishing-cap, 207. 
 
 Selection : in perception and judg- 
 ment, 88 ; by the body, 155, 160 ; 
 and realism, 181. 
 
 Self, the : 172 sqq. 
 
 Self-consciousness : and conscious- 
 ness, 15B sqq.; and intuition, 169; 
 its character, 170 sqq. ; in history, 
 200; for mysticism, 217. 
 
 Sensation: and perception, 15 sg.; 
 andthegiven, 31, 203; and feeling, 
 165 sq. 
 
 Sense data: perceived? 18sqq.; and 
 things, 38; in 'complication,' 40; 
 and error, 41 sq. ; and belief, 89sqq. ; 
 'inner' or 'outer'? 162. 
 
 Shaftesbury: on a relish for virtue, 
 135. 
 
 Sherrington : on anger-mimesis, 
 152 sq.; on reflexes and their in- 
 tegration, 155 sqq. 
 
 Sidgwick, H. : his Axiom of Bene- 
 volence, 143 ; on men in history, 199 . 
 
 Sign-facts: in perception, 24 sq., 
 29sg.,36,42,83, 88, 93;inmemory, 
 48, 55 sq. 
 
 Signs: meaning of, 33 sqq. ; and uni- 
 versals, 111 sq. ; and representation, 
 129, 186, 209 sq. 
 
 Smith, Adam : his ' impartial spec- 
 tator,' 138. 
 
 Smith, Professor Norman : on Kant, 
 174 sq., 17 5 n. 
 
 Soul, the : out of fashion, 151 ; as the 
 spiritual substance, 173. 
 
 Space: of sight and touch, 18; in 
 physics , 26 ; of perception , 29 sq . , 39 ; 
 of dreams, 66; of imagery, 67 sq., 
 70, 74; of the physical world, 85, 
 93 sg . , 102 ; in causation , 99 ; ' outer ' 
 or 'inner,' 163. 
 
 ' Specious present,' the: analysis of, 
 45 sqq. 
 
 Spinoza: his monism, 178. 
 
 Stevenson: on Anatole France, 127; 
 and St Lves, 205 sq. 
 
 Strong, C. S.: on the 'new realists,' 
 164. 
 
 Stumpf: on discrimination, 21; on 
 feeling, 165. 
 
 Subject: and predicate, 87 sqq., 92sq. ; 
 the, 171 sqq. 
 
 Subjectivity: of beauty, 126 sqq.; of 
 morals, 126, 135 sqq. ; of experience, 
 164 sqq. 
 
 Substance: the spiritual, 172»'/. 
 
 Sufficient Reason: Law of, 1'23. 
 
 Suggestion: for Berkeley, 34 sq.; in 
 representation, 88, 196, 209. (See 
 also association, meaning, sign.)
 
 228 
 
 INDEX 
 
 Sympathy: in biology, 187 sq.; in 
 economics, 192 sg. ; in history, 200 ; 
 of imagination, 203. 
 
 ' Tertiary qualities ' : and realism, 129. 
 
 'Thing': general meaning of, 9 sq. 
 (For its special meaning see 
 'memory,' 'perception,' 'physical 
 things,' and 'images.') 
 
 Thomas Aquinas, St : on ' natural ' 
 good, 107, 107 n.; on God's good- 
 ness, 139. 
 
 Thompson, D'Arcy : on simple things, 
 188. 
 
 Time: in physics, 26; of perception, 
 45sqq.; of dreams, 66; of images, 
 70 sqq.; associative response in, 
 76; of belief, 84, 94 sq., 102; in 
 causation, 99; and progress, 147; 
 coordination of reflexes in, 1-56 sq, ; 
 of sense-data, 163. 
 
 Titchener: and Stumpf's argument, 
 165h. 
 
 TurnbuU, Geo : and Newton's method, 
 7; Hogarth and, 7n. 
 
 Underbill, Miss Evelyn : on mysti- 
 cism, 211ri., 214n. 
 
 Uniformity : in causation, 97, 99 sqq. 
 
 Unity: of apperception. (See Kant.) 
 
 Universals: in mediaeval realism, 2, 
 11.5?i. ; in propositions, 87 sqq. ; 
 'being' of, 109 sqq.; for the 'new 
 realists,' 163. 
 
 Values: 125 sqq. 
 
 Victor, Hugo of St : and the symbol- 
 ism of physical things, 105. 
 
 Ward, James: on the presentational 
 continuum, 2bsq.\ on improvement 
 in perception, 40 ; on the ' aspects ' 
 of experience, 164s</(/.; on atten- 
 tion, 167. 
 
 Whitehead : on space and time, 26?i. ; 
 on the Law of Parsimony, 123. 
 
 Withers, Hartley: on the laws of ex- 
 change, 191. 
 
 CAMBRIDGE: PRINTED BY J. B. PEACE, M.A., AT THE UNIVERSITY PRESS