(Cornell Ittitieratta Hibrarg BOUGHT WITH THE INCOME OF THE SAGE ENDOWMENT FUND THE GIFT OF HENRY W. SAGE 1891 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924088998269 PROBLEMS OF THE SELF , MACMILLAN AND CO., Limited LONDON ■ BOMBAY ■ CALCUTTA • MADRAS MELBOURNE THE MACMILLAN COMPANY NEW YORK • BOSTON • CHICAGO DALLAS • SAN FRANCISCO THE MACMILLAN CO. OF CANADA, Ltd. TORONTO PROBLEMS OF THE SELF AN ESSAY BASED ON THE SHAW LECTURES GIVEN IN THE UNIVERSITY OF EDINBURGH MARCH 1914, BY JOHN LAIRD, M.A. PROFESSOR OF LOGIC AND METAPHYSICS IN THE QUEEn's UNIVERSITY OF BELFAST MACMILLAN AND CO., LIMITED ST. MARTIN'S STREET, LONDON 1917 A 377?57 COPYRIGHT PKEFACE This volume makes the attempt to consider a group of problems which offer an unceasing challenge to philosophers, and the catchword ' Psychology with- out a Soul ' gives the best general indication of the type of problem discussed. The aim of the enquiry as a whole is to show why there must be a soul, and in what sense precisely .this soul should be under- stood. A complete or final answer to such a question is, of course, unattainable until the day when all speculative problems have found their solution ; but the labours of philosophers and psychologists in recent years have made it possible to appreciate most of the important questions at issue with a clearer understanding than at earlier times, and permit the hope that an enquiry which keeps both the earlier and the more recent literature constantly in mind, need not be altogether fruitless. One of the principal difficulties of this investiga- tion is that so much of the relevant literature con- sists either of merely negative arguments, or else of discussions which are undertaken from a somewhat different point of view. It is not enough to prove the absurdity of a ' Psychology without a Soul,' vi PROBLEMS OF THE SELF even if the proof be conclusive. The still more important question is : ' What is Psychology with a Soul ? ' Most of us feel that the ' pure ego ' is a barren fiction, and the 'empirical ego' a formless mass. We want a synthesis which is concrete and definite, so far as any metaphysical synthesis can be. And the concrete treatment of the question is often to be discovered by the way, especially in the arguments for the primacy of will or feeling, or in disquisitions on the unity and continuity of the self, on the subconscious, or on multiple personality. An adequate discussion, therefore, must cast its net somewhat widely, and sometimes may seem not to have a very direct bearing on the central problem. The sketch of the plan of the argument in this book, which is given on pp. 42-44, together with the intro- ductory chapter, may lessen the difficulties of the reader in this respect. It was my privilege, as holder of the Shaw Fellow- ship in the University of Edinburgh, to give a course of lectures there in March 1914, and this volume expresses the argument of these lectures in a fuller . and, I hope, a more adequate way. I should like to thank my audience for the courteous attention they gave to the lectures, and, in particular, to thank ihy former teachers in that University for the encourage- ment they have given me, both at the time of the delivery of the lectures and upon so many other occasions. I owe them a debt which I cannot repay. By a happy fortune I am similarly indebted to my PREFACE vii teachers at Cambridge, and although an obligation of this kind cannot be specified in detail, and may appear but imperfectly in the result, the sense of it remains undiminished. Every man's work is stamped by the training he has received, and, without that, he can accomplish nothing. In the preparation of this volume for the press, I have received very valuable assistance from my col- league, Mr. M. W. Robieson, from Mr. John Baillie, Edinburgh, and from Mr. A. J. Dorward, St. Andrews ; and my grateful thanks are due them for this assistance. JOHN LAIED. Queen's Univeksity, Belfast. CONTENTS CHAPTER I PAGE iNTEODtJCTORY 1 Interest of the question (1) ; its importance for metaphysics (2-4) ; objections to the feasibility of the enquiry, and to the necessity for it, answered (4-7) ; terminology (7-8) ; statement of the principal theories of the nature of the self with a view to the discovery of the best plan of procedure (7-11) ; proof that it is necessary to begin with experiences (11-15). CHAPTER II Psychology and the Self 16 Impossibility of defining psychology, negatively, by reference to the non-physical (16-18) ; the distinction between ' act ' and ' object ' explained in the instance of cognition (18-22) ; the sources of psycho- logical evidence, and the nature of proof attainable through introspec- tion (22-30) ; introspective analysis of cognition (30-31) ; transition to the tripartite division of mind, and discussion of (1) meaning of terms, (2) common characteristics of conscious elements, (3) logical principle of division (31-36) ; reply to objections (36-38) ; formula for the positive principle of the constitution of mind (at any moment) derived from the above (38-40) ; problem whether a ' soul ' is also required (40-42) ; outline of plan of argument in this book (42-44) ; points to remember (44-45). CHAPTER III The Self and the Body ....... 46 Point of view of discussion (46) ; necessity for considering the implications of the distinction between body and mind by direct analysis (46-47) ; discussion of the 'internal sense,' in its traditional meaning (47-52) ; its literal meaning defended (52-53) ; the objects of the ' internal sense,' on this interpretation, although subjective in a ix PEOBLEMS OF THE SELF special sense, agree with the objects of the other senses in being present to mind, but not parts of it (53-57) ; reasons why there is a peculiar difficulty in distinguishing experiences from the objects of the internal sense- shown in a discussion of feeling (including emotion and pleasure-pain), multiple personality, James's view of personal identity, and activity (57-75) ; the sense, if any, in which oonsciousness occupies position (75-77) ; reply to a fundamental objection (77-79) ; summary (79-80). CHAPTER IV The Self as Feeling Plan of argument in Chapters V.-VIII. (81) ; the usual view that experiences of feeling or will have the primacy in the self (81-82) ; meaning of the term ' feeling ' (82-83) ; discussion of the primacy of feeling as defended in the argument from origin (83-87) ; from the ' subjectively-subjective ' character of feeling (87-92) ; from ' interest ' (92-93) ; from the significance of moods (93-94) ; Mr. Bradley's view (94-100). CHAPTER V The Self as Will ........ 101 Course of discussion in Chapters V.-VII. (101) ; common-sense belief in the primacy of will (101-102) ; 'will ' may mean either volition or activity-experiences generally (102-103) ; plan of argument in this chapter as preliminary to its successors (103-104) ; introspective analysis and description of activity experiences (104-110) ; discussion of views of Berkeley in this connection (110-113) ; and of Munster- berg (113-115) ; problem whether activity experiences are causes, and Hume's arguments on this question (115-123) ; more precise enumeration of activity experiences, and definition of will in the narrower sense (123-126). CHAPTER VI Psychical and Purposive . . . . . . .127 Purposive behaviour probably the best criterion of the existence of mind, but this no proof of the primacy of will (127-128) ; meaning of 'purpose,' 'end,' and 'teleology' (128-133); psychology and biology (133) ; the meaning of 'purpose' in biology (133-134); the inconclusiveness of the negative arguments in favour of neo- vitalism (134-138) ; insufficiency of the positive arguments owing to the fact that they rely upon an unduly strained analogy between ' purposive behaviour ' and conscious purpose (138-144) ; meaning of ' teleology ' in biology (144-149) ; fundamental distinction between psychical and organic categories (149-151). CONTENTS xi CHAPTER VII PAGE The Primacy of the PeactIoal Reason .... 152 Relation of the Practical Reason to the Will (152-153) ; the phrase indicates certain metaphysical arguments (163) ; criticism of Kant's proof of the primacy of the Practical Reason in the OrUique of Practical Reason (153-160) ; and of his arguments in the Critique of Judgment (160-165) ; criticism of Fichte's arguments (165,-174) ; and of Schopenhauer's (174-183) ; and of Bergson's (183-189) ; summary of results of Chapters IV.-VII. , denying the primacy of the will, and defending the analysis of CJhapter II. (190). CHAPTER VIII The Self as Knoweb .... . . 191 Reasons for adopting a somewhat different plan of procedure in this chapter, and account of the scope of the chapter (191-193) ; summary of preceding argument, and of its presuppositions (193-195) ; the problem of the content of cognitive acts (195-198) ; is an intro- duction to the fundamental problem of this chapter (198-200) ; problem whether the direct analysis of an act of knowledge implies a knower other than the act (200-201) ; criticism of Lotze's arguments (201-206) ; self-consciousness and self-cognition (206-210) ; criticism of Mr. Bertrand Russell's view (210-212). CHAPTER IX The TJnitt and Continuity of the Self . . . .213 Importance of unity and continuity in the theory of self-hood, and justification of delay in discussing it (213-215) ; plan of this chapter (215) ; preliminary considerations concerning (1) the preponderance of cognition, conation, or feeling in personalities of different types, and (2) concerning the danger of exaggerating the unity and con- tinuity of the self (215-217) ; cognitive unity (217-218) ; Kant's proof and further explanations (218-223) ; criticism of Mr. Joachim (223- 229) ; limitations to cognitive unity (229-231) ; development (231- 232) ; logical character of cognitive unity (232-235) ; limitations to cognitive unity further explained (235-237) ; unity and continuity of feeling (237-238) ; the sentiments (239-240) ; reason in feeling and the logic of the sentiments (240-242) ; the unity and continuity of conation (242-244) ; summary of argument (244-245) ; personal identity (245-247) ; and the body (247-248). CHAPTER X How is this Unity possible? 249 The problem of the conditions of reteutiveness (249-250) ; the physiological theory (250-256) ; psychical dispositions (256-257) ; xii PKOBLEMS OF THE SELF and subconsciousness (257-268) ; arguments in favour of the existence of subconsciousness derived from unconscious cerebration, psycho- analysis, acquired dexterity, the summation of stimuli, sense-percep- tion, and stored knowledge, are intended to explain the fact of continuity (257-263) ; but are inconclusive (263-265) ; the direct arguments more convincing (265-267) ; the range of explanation in terms of subconsciousness (267-268) ; the term applies to acts (268- 269) ; possibility that the observable continuity of the self need not require any further explanation (269-271). CHAPTER XI MuLTiPiiE Personality 272 This kind of evidence supplies an invaluable test of the truth of a theory of the self (272) ; general discussion of dissociation (1) in evanescent, and (2) in more permanent forms (272-277) ; division of the types of cases (277-278) ; consideration of the evidence for the dissociation of personality with special reference to memory and retentiveness, organic sensation, the unity of endeavour and of feeling, (278-289) ; the problem of estimating the evidence for dissociation inio personalities considered (289-293) ; probabilities suggested by the Beauchamp case (293-297) ; discussion of the problem whether a plurality of personaljj^ies can co-exist in the same body (297-300) ; difficulties raised by multiple personality especially important in connection with the soul substance (300-301) ; concluding comparison with normal personality (301-303). CHAPTER XII Discussions of the Sbbf as Substance in Modeen Philo- sophy 304 Necessity for considering certain historical theories on this question (304-305) ; great divergence of views (305-306) ; the cogito of Descartes and its implications (306-310) ; Locke's ambiguous position and the conflict between empiricism and rationalism (310- 319) ; Locke's interpretation of spirit (319-321) ; and of personal identity (321-324) ; Hume on personal identity (324-326) ; and on substance (326-328) ; and on the permanent ego (328-331) ; Kant's argument concerning the paralogisms of Pure Reason (331-334^ ■ Hegel (335-336). CHAPTER XIII The Soul 331, Relation of this chapter to its predecessors (337-338) ; definitions of substance following Descartes, and interpretation (338-339) ■ -0. merely logical definition of substance impossible (339-341) ; substance implies the existent and not merely objectivity (339-343) ; imaginary CONTENTS xiii PAGE substances (343-344) ; the existent is subjectively known through the senses, and is objectively the particular (344-346) ; this implies stuff or SX17, (346) ; virhich is further explained (346-348) ; the unity or identity of substance (348-349) ; examples showing the apparent arbitrariness of our standards of this unity (349-351) ; the difficulty of the margin (352-3S6) ; independent existence of substances and uselessuess of transcendent explanation (356-357) ; application of the general doctrine of substance to organisms (357-358) ; and to the soul as the necessary unity of experiences (358-362) ; the soul pre- eminently a substance (362-365) ; objections (365-367) ; immortality (367-370). INDEX 371 CHAPTER I INTRODUCTORY It is a truism that no study is more perplexing and, at the same time, more interesting to a man than the study of mankind and, in the end, of himself. Even if the pressure of the day's business leaves the average man but little time for self-reflection, he is still intensely interested in the personality of others, and the most obstinate questionings which beset him concern his soul and theirs. Moreover, the great objects of human interest affect personality and are tinged with personality. It is unnecessary to prove this statement by referring to the drama, the novel, history, biography. The thing is too obvious to require comment, and it is enough to illustrate it by mentioning a curious fact. Even those who in general have no great fondness for the study of biography are more keenly interested in the personal history of the great writers in literature than in their works, or, at any rate, are interested in a degree out of all proportion to the intrinsic interest of the careers of those authors. How else is it possible to explain the mass of literature and the years 'of discussion devoted to the shadowy author of the Odyssey, or to the stray hints which are all that is known of the career of Shakespeare ? Ivfor is the reason very far to seek. As Samuel Butler says, ' Every man's work, | whether it be literature, or music, or pictures, or ^ 1 B 2 PEOBLEMS OF THE SELF chap. architecture, or anything else, is always a portrait of himself, and the more he tries to conceal himself the more clearly will his character appear in spite of him.' ^ That is the truth unless, perhaps, such sciences as mathematics or physics are excepted. It may be a rare thing for the artist to be more interesting than the whole body of his work, but his character and career usually excite more attention than those of any one of his creations, and thus it is that the self is central among the things which touch the spirit of man. It does not follow, of course, that the majority of mankind are strongly attracted by the philosophical problems which cluster round the self. There are some minds to which any metaphysical discussion is stale and unprofitable. They may admit that philo- sophy began with wonder, but they marvel still more at the fact of its continuance. In the broad sense in which every man is a metaphysician he is occupied, probably, with the metaphysics of self to a greater degree than with any other philosophical problem. He would be less than human if he never asked himself, What is a man ? What is the soul ? What is this miracle of three, or thirty, or threescore years ? Nor could he fail to ask such questions so long as either theology or religion continues. Eeligion must raise the question whence man is and whither he goes, and it does not require a professional philosopher to point out that there is but little sense in asking such a question unless there is some comprehension of what it is that has come and what is about to depart. But the sense in which every man is a metaphysician is, after all, a very shallow one. Every man dallies with metaphysical conjectures, but there are very few who try to think philosophical problems out, resolutely and to the end, and unless a man make this attempt he has not begun to be a philosopher. ^ The Way of all Flesh, p. 62. INTRODUCTORY 3 At the same time, if and so far as metaphysical problems are really felt and earnestly attacked, the problems of the self deserve to take precedence of all others. If the self is central in point of human interest, it is also central in point of metaf)hysical importance. Nearly every philosophical problem has some bearing on the self, and conversely the most distinctive of the problems of the self are among the widest and the most important in philosophy. Pluralism and monism, individuality and value, realism and idealism are each, in a very distinctive fashion, problems of the self. The relation of selves to the Absolute is the hardest problem of idealistic monism, for, on that theory, the self is the most obstinate of all appearances. The ultimate reality of selves has been the basis of idealistic pluralism from the days of Leibniz onwards. And it is needless to elaborate in the other cases. These are perennial problems, but it is also fair to claim that the problems of the self bear a peculiarly close relation to much that is most characteristic of contemporary philo- sophical discussion. The importance of volition and the relation of body to mind are certainly among these problems, and both have a direct and obvious connection with the self. There are others besides the pragmatists who stoutly contend that action, conation, will are the basis of reality, and cling to the belief that ago, ergo sum deserves to succeed the Cartesian cogito. Again the contemporary dis- cussion of the problem of Body and Mind tends more and more to exalt the importance of the self. We hear less of the conservation of energy, and more of the growth of living energy : less of dynamics, and more of purpose : less of atoms, and more of cells : less of natural selection, and more of progressive creation. These arguments, indeed, may prove mistaken in the end, but their tenor, at all events, is significant. The tendency, a generation ago, was to explain the 4 PROBLEMS OF THE SELF self in terms of something else. The increasing tendency, nowadays, is to explain other thmgs m terms of the self. ' Briefly, then, if any metaphysical problems are interesting, those of the self are most likely to be so : if any are important, those of the self must be among them. And there can be no serious objection to the feasibility of discussing these problems. There are, of course, superficial objections to the discussion of any metaphysical problem whatever, but these objec- tions have not deterred metaphysicians in the past, and will not deter them in the future. They are easily refuted, and, in any case, have no peculiar relevance to this enquiry. The objections which may be raised, in limine, to a specific discussion of the self hardly require detailed consideration, although they certainly deserve a passing mention. It may be maintained that the self, as it were, shines by its own light, that every one knows what it is, and that instead of finding, or inventing, problems in it we should restrict ourselves to the task of showing how other things may be explained in terms of it. But has this argument ever been seriously held ? On certain metaphysical theories, it is true, the self has been proclaimed the clue to the riddle of existence. It is a ' transparent unity,' and there- fore we cannot see into it ; we can only see through it. Certainly it would be the rashness of mere folly to deny that the self, when understood, may prove itself the master-key of philosophy. But, if it is so, it owes its position not to its simplicity, and not to the ease with which it will fit, but to the fact, already mentioned, that most philosophical problems reach their climax in it. There are no idealists who believe that the self is transparent to the casual observer. A veil of obscurity intervenes before it can show itself as transparent as it really is, and that is enough for a beginning. In any INTEODUCTORY 5 case, no one can really believe that there are no problems of the self. The writers ^ who contend that the self must be ultimate (because it is so extremely- paradoxical) and those who contend that it cannot be ultimate (for precisely the same reason) do not argue wholly at random, and it does not require a believer in the Hegelian dialectic to see that this is a case which demands enquiry. Although the argument that a man must know himself better than anything else is frequently maintained by philosophers, neither Descartes, nor M. Bergson, nor any other exponents of it would conclude that discussion of the self does not involve problems. After all, the better a thing is known the more enigmatical it becomes. It gives rise, not to fewer, but to more intelligent questions. And why is it easier to understand the self than to understand other things ? There is no reason why knowledge should know its like more easily, or better, than anything else. If, in point of fact, the surest knowledge is that of the self, the reason lies, not in the necessity of the case and not ih the nature of knowledge, but in the superior interest of the psychical and the greater preoccupation with it. And this fact, after all, may be dispi^ted. It is just as important for any one to know himself and his feUows as it is for him to know anything else. He is as dependent on society as he is on bread ; he requires a knowledge of stocks and shares as well as of stocks and stones. But the science of psychology is in its infancy when compared with the physical sciences, and that, at least, is a fact very hard to reconcile with the other ' fact ' that is presumed. For the rest, it is unnecessary to consider the futile statement that we must know what the self is, because we are selves. We have opportunities, that ' Of., e.g.. Dr. M'Taggart {Studies in HegeUcm Cosmology) and Mr, Bradley {Appearance and Reality) respectively. 6 PROBLEMS OF THE SELF is all; and opportunities for knowledge are not knowledge. , ^i ^ ^.i, ij? Indeed, for every one who contends that the sell is not a fit subject for study, since it contains no mystery and involves no problems, there are many who maintain that it is useless to study it because it is the most inexplicable mystery in the world. 'One may understand the Cosmos but never the Ego: the Self is more distant than any star.'^ So Mr. Chesterton, and so, I do not doubt, many who agree with him, and many who do not. This extreme, however, is scarcely more Ukely to be true than the other. Mr. Chesterton himself does not scruple to maintain that while it is possible to understand causes and effects in our own persons it is never possible to understand them in a laboratory, and those who argue in his vein might easily bring forward plausible reasons for believing that the self is less mysterious than the multiplication table. There are problems of the Cosmos and special problems of the millions of Egos which, as Mr. Chesterton seems to forg'et, are included in it. And there are no a priori reasons which condemn in advance the attempt to solve such problems. Problems of the self are on the same footing, in this respect, as others in philosophy. A philosophy can never be judged by its fruits, and therefore it is unwise to raise the question what fruit it can be expected to bear. No philosophical discussion, except one of a very limited scope, can hope for finality or maturity. But if there is no season of harvest in philosophy, there is at least the possibility of perpetual growth. A philo- sopher is like an explorer. He never knows his luck. If there is something to be discovered, after all he may find it. If he fails he may at least inform others that certain routes are impracticable or require better equipment. There is need for the venture " G. K. Chesterton, OrtJiodoxy. INTRODUCTORY 7 and there is room for hope. That is all that need be said. For purposes of clearness it is necessary to make an explanation with regard to terminology. Generally speaking, the words ' person,' ' soul ' or ' mind ' may be regarded as synonyms for the self, and it would be mere pedantry to avoid using them as synonymous, unless there is some special liability to ambiguity in the particular context in which they are employed. Indeed, the words ' consciousness ' and * mentality' might sometimes be used in a similar sense, although with some inappropriateness and, in the latter case, with a wilful disregard of euphony. It is well to remember, however, that these terms are only approximately synonymous with the word * self.' The two latter are abstract rather than concrete, if strictly employed, and, even if they are employed loosely, they may still imply a conception of the content of the self which might well be disputed. They imply that the self is nothing but consciousness, and this, whether true or not, is very far from being univer- sally admitted. ' Person,' ' soul ' and ' mind ' come nearer to the meaning of 'self,' but each of them has a shade of significance which it is desirable to avoid at the outset of an enquiry like the present. The 'soul' and the 'self may mean the same thing, but the latter is humbler and less steeped in extraneous associations ; and, consequently, it ought to be used by any one who desires an unaccented reading of the problem. The word ' soul ' seems too aristocratic to have its ancestry scrutinised or its income assessed. It breathes the rarefied atmosphere of poetry and theology. It leads, perforce and at once, to questions of vitalism, and the meaning of the indiscerptible substance. I am far from wishing to assert that the self is not a noble thing, but when we speak of the soul we are apt to forget that it may also be ignoble, and, what is worse, occasionally dull. Nor do I believe 8 PROBLEMS OF THE SELF that the problems of the spiritual substance and the vital principle are irrelevant to a discussion of the nature of the self. On the contrary, I intend to dis- cuss them. The question is one of emphasis, and that emphasis may be all-important. 'A discussion of the soul makes it necessary to proceed at once to these questions. A discussion of the self need not approach them until after a more prolonged and more careful survey. Similarly, the term 'self is broader ^ and freer, though not looser, than either 'person' or ' mind.' ' Mind,' in common parlance, refers almost exclusively to the intellect, and, although the intellect is a part of the self, it is but a part, in the end, and possibly not the most important part. 'Person,' again, lays special emphasis upon certain ethical and legal implications {e.g. upon the degree of memory implied in the? continuance of responsibility), and, although these characteristics certainly deserve con- sideration, it is desirable to avoid a terminology which brings them into unduly high relief. The word ' self,' then, includes what these other words include, and is preferable because it does not dictate the road which the discussion must follow. It would clearly be futile to begin to discuss problems of the self without making a preliminary attempt to define what is meant by that term. The sage remark that a definition is the culmination rather than the starting-point of an enquiry is very properly disregarded even by those who make it. A preliminary definition is only an indication of the route to be followed, and therefore it is indispensable. The chief difficulty, in this as in so many other cases, is^ to find a definition which merely asks questions without begging them, and, probably, it is never possible to ask the right sort of question without begging or assuming something. Certainly, the self cannot be defined in such a way as to satisfy all parties, but it does not follow that there can be no consensus of INTRODUCTORY 9 opinion as to the right line of enquiry into its nature. Different schools may agree in this, though their conclusions are as the poles asunder. Let us consider how that is possible. The accounts of the self which are found in the works of different philosophers are very various and frequently conflicting. In "proof of this it is sufficient to cite a few instances. According to Hume, the self is a succession of impressions and ideas which, fleeting and perishing, are in a perpetual flux and movement. ThiSi view, variously amended but not transformed, has its adherents to-day, and is one of the current interpretations of the theory which Lange has de- scribed as 'Die Psychologic ohne Seele.' Others, again, who are by no means followers of Hume are in substantial agreement with him at least in one respect. According to Mr. Bradley, 'the Ego that pretends to be anything either before or beyond its concrete psychical filling is a gross fiction and mere monster, and for no purpose admissible.' ^ Such a statement is very far from implying the way of impressions and ideas, but it clearly expels any Arch- Ego or other intruder who is not an ordinary guest in the mind's presence-chamber, entering in the usual way. And it is irrelevant whether the ' monster ' has the shape of a permanent impression, as it had for Hume, or whether it adopts some other form. Many, however, would deny that the vehemence of Mr. Bradley's assertion gives it any claim to acceptance, and they would dispute it with at least equal vigour. It is true that Lotze^ declares that when he calls the self a ' simple and indivisible substance,' he is speaking ' in all innocence,' and merely means to express the fact of its indivisible unity. Such a theory is indeed incompatible with Hume's, for Hume meant to imply that a self is only ' Appearance and Reality, p. 89. ^ Metaphysic, English translation, vol. ii. p. 173. 10 PROBLEMS OF THE SELF ohai.. indivisible because it is divided. But it is not necessarily incompatible with Mr. Bradley's. The fact remains, however, that many would agree with Lotze in a sense that is not so innocent. They would take him to mean, as he certainly said, that a unity of consciousness must have a ' centre,' ' and, with or without warrant, they would take this 'centre' to imply something either beyond or before the concrete psychical fiUing. They would assert that knowing without a knower is a contradiction in terms. They would maintain that no single experience can be understood without the assumption of a self, and that the self must be more than any or all such experiences. This interpretation may be wrong, but it is prevalent, and, at least on the surface, it is not nonsense. It is in fact the usual view, whether or not it is capable, in the -end, of withstanding logical criticism. I do not know precisely what Dr. M'Dougall ^ means by the ' non-mechanical teleological factor ' which he finds requisite for life in general and the soul in particular, but I do know that, in the form in which he means it, neither Hume nor Mr. Bradley could possibly accept it. There is, in other words, a pro- found opposition between those who maintain that the self cannot exist except within or between experi- ences, and cannot have any other content, and those who maintain that it .must also be something more, if, indeed, it does not he wholly outside such experiences. And there are many other views. The late Mr. P. W. H. Myers, in his Human Personality, defends the theory that the self which is known is but a fragment which, when supplemented by the subcon- scious and by other hitherto unexplored tracts of mind, will be seen to belong to an astonishing and supra-personal whole. The grounds for this theory ' Appearance and Reality, p. 169. '^ Body and Mind, p. 364. I INTRODUCTORY 11 are lacking, but its possibility is beyond dispute. Similarly there is the ' transmission theory ' in its two forms. In the first of these forms the body is regarded as a sort of prism which breaks the white beams of the eternal soul into the variegated person- ality we are accustomed to consider ours. In the second form the body is regarded as an organ of concentration. Consciousness is really diffused. The body collects it, and so creates those fickle, partial, paradoxical unities which are the men who see before and after. Either theory magnifies the office of the body, and neither stands alone in doing so. In one of his Essays in Radical Empiricism Professor James remarks : ' So far as we are " persons," and contrasted and opposed to an " environment," movements in our body figure as our activities ; and lam unable to find any other activities that are ours in this strictly personal sense.' ^ The plain implication, from which there is no evidence that James would have shrunk, is that the personal and individualised self is the body, and although this view is rarely held (at any rate upon psychological grounds), it cannot safely be neglected. Moreover, there is the doctrine that the self consists of that in which interest is felt, a theory which seems to imply a peculiar view of the relations of meum and tuum. But the theories mentioned are enough to illustrate the great disparity between the different philosophical theories of the nature of the self, and in view of this disparity it might well appear hopeless to try to begin this enquiry in a way that all parties would recognise to be fair and just. The disparity of the conclusions, however, does not necessarily indicate that there is no common ground in the way the problem is attacked. In the theories which have been mentioned there are two main lines of cleavage, ^ Essays in Radical Empiricism, p. 170 n. For justification of the inter- pretation see the whole note. Italics mine. 12 PROBLEMS OF THE SELF chap. and the theories themselves are but variants within these lines. The first question is whether the Ego is anything more than its concrete psychical filling, the second the degree in which the body must be regarded as constitutive of the self. Let us take one of the arguments and consider its logical bearing. Dr. M'Dougall says that ' the facts of our conscious life, especially the fact of psychical individuality, the fact of the unity of consciousness correlated with the physical manifold of brain-processes, cannot be rendered intelligible without the postulation of some ground of unity other than the brain or material organism.' ^ This proof relies on a basis which is negative in two respects. It maintains that a study of psychical processes, such as that undertaken by psychology, cannot be completely intelligible without a reference to something beyond the observed facts of mind. Psychical facts and psychical acts cannot supply their own explanation : they require a com- plement. And the second part of the argument is that this complement cannot be the body. Therefore, there must be a soul. This proof may or may not be valid, but, at any rate, one of its implications is clear. Even if the self is something beyond its psychical ifiUing, it is necessary to begin with that psychical filling, and this is common ground. If the argument is that the soul is needed to explain the accounts of mental processes which psychology affords, the first step in the proof must be that of considering whether or not the processes thus described can be regarded as self-explanatory. When it is maintained that the body is the self, or the permanent ground of the self, an alternative hypothesis to that of the soul is defended, and whether it can or cannot be defended is of course a matter for argument. The only argu- ment which does not fit naturally into this scheme is the view of Professor James that the individualised ' ' Body and Mind, p. sSe. INTRODUCTORY 13 self is the body, and his argument depends upon Introspection only. In other words, it is nothing but a description of what mental processes are and appear to be. The best way of describing this result is to say that any account of the nature of the self must begin with an analysis of experiences. By experiences I mean the subject-matter of psychology, acts of knowledge, and acts of will, passions^ emotions, strivings. Experiences, as the Germans say, are Erlebnisse, bits of conscious life which must be lived through, if haply they are to be understood. There is no better word in English to express this meaning than the word ' experiences ' in the plural. Indeed, it is unfortunate that English philosophy, especially in latter years, should have spoken so consistently of Experience in the singular, Experience with a capital E, But no other translation of the word Erlebnisse, whether ' states of consciousness,' ' mental processes,' or any other, is so adequate or so con- venient, and therefore it is necessary to make the best of an indifferent case. The statement that an adequate account of the nature of the self must begin with experiences is very far from implying that the self is nothing but a collec- tion or bundle of experiences. It is in fact compatible with the belief that the self is something beyond and above them, of which they are not even parts but only consequences. The statement means only that the one argument which can possibly convince any one of the necessity for the belief in a self over and above its experiences is an argument which is able to show that the experiences could not be what they are unless they depended upon this self. Experiences, it may be maintained, unite in a self, and this unity requires something more than the experiences them- selves in order to be intelligible. Or again, the experiences are fleeting and perishing, and require the 14 PROBLEMS OF THE SELF; chap. assumption of a permanent self to explain their regularity and their continuity. Or again, it is clear from the analysis of even a single experience, that it would be nothing without a self. An act of resolve, without a subject, is a contradiction in terms. The advantage of beginning with experiences is that this starting-point is compatible with all the theories and does not presuppose any one of them. It is equally consistent with the view that these experiences, when thoroughly investigated and properly under- stood, a/re self-explanatory, that together they are the self, or that the self falls within them. The relation of the self to the body involves special problems, and will be discussed in Chapter III. Except for these special problems, there can be no serious objection to the choice of this starting-point. The first task, therefore, and the object of the next chapter i^ to consider, as precisely as possible, what experiences are, and then to discuss their relation to, or their union in, the self. This is, in the first place, an enquiry into the subject-matter of psychology, but it leads further of necessity. The problem of the nature of selfhood is never a problem of the mere description, or analysis, of experiences. Taken at its lowest terms, it is the question of how these experiences conspire together to be a self If the subject-matter of the self, in a way, is that of psychology, the problem of selfhood is by no means coextensive with . psychological investigation. From one point of view the nature of the self is one particular psychological question. From another point of view it is something too fundamental to be considered by psychology at all. For the present purpose, however, it is irrelevant whether this discussion is best described as meta- physical or psychological. Metaphysics has no subject-matter which the sciences have not or, at any rate, could not have. It is only an historical accident if there are still some realms of existence 1 INTRODUCTOKY 15 which science has hitherto refused to explore,- and has left as a sort of residuum to the metaphysician. Apart from such accidents metaphysics has the same subject-matter as the sciences, including psychology. The only difference is that metaphysics adopts a point of view which is wider than that of the sciences and, to put it paradoxically, mpre thorough and more searching. Further discussion is required in order to discover whether metaphysics must, or need not, re- interpret the results of science in order to secure their intelligibility or its own, and the sequel will consider whether it is necessary to abandon this relatively psychological standpoint, or whether it can be in- corporated without substantial modification. A Japanese gymnast poises himself on an insubstantial pile of loose bricks, and then knocks them down in order to show how clever he is. A metaphysician, perhaps, does not require this Oriental dexterity. CHAPTER 11 PSYCHOLOGY AND THE SELF The definition of the -subject-matter of psychology is not, unfortunately, free from controversy. Indeed, the very descriptions of this subject-matter may arouse dispute. It is frequently said, for instance, that psychology investigates states of mind, but such a phrase, obviously, is an open invitation to bickering. What is the relation^ of mind to it^ states ? A state must be a state of something, as Locke's ' modes ' had to be modes of something, and consequently the defenders of ' psychology without a soul ' might reasonably contend that the paradoxical character of their conclusions is increased without any warrant by the purely verbal difficulties into which this description brings them. The phrase ' states of con- sciousness' is equally objectionable: If consciousness is, properly speaking, an abstract term, it ought to mean a characteristic belonging to each of a number of events. Psychologists are probably innocent of any deep designs in their preliminary definitions of their subject-matter, but innocence, like Nature, has consequences. Let us begin by asking how the subject-matter of psychology differs from that of other sciences. This problem is too often attacked as if it were merely a question of distinguishing psychical from physical, and therefore it leads to the conclusion — a fatal one 16 CHAP. II PSYCHOLOGY AND THE SELF 17 if uncritically accepted — that whatever is not physical is psychical. The physical world, however, exists continuously throughout a uniform time in a homogeneous space of three dimensions. This, at least, is its current interpretation, and there may be, and are, many objects which are neither physical, in this sense, nor psychical. It is plain, for instance, that the direct objects of perception are not physical in precisely this way. No reflective man can be so utterly blind to the known differences between tactual and visual space, to the facts of optical illusion, to the differences arising from perspective and tha blendings and contrast of colours, as to suppose tha]^ what he perceives, as he perceives it, is necessarily this stable, orderly, physical world. The presenta- tions of sense may correspond to this world, roughly and in the main; they may even be distorted glimpses of it, but it is to the last degree improbable that the physical world itself, uninfluenced by the subject, is the direct and invariable object of sense-perception. And if, in any instance, it is not, then there is an instance of an object which is not, strictly speaking, physical. But, on the other hand, there is no reason for asserting that such objects are psychical or parts of mind. A dream table, for instance, is not part of the dreaming mind, any "more than the table used ia wakiag life is part of the mind of the per- cipient. These are objects . for mind, not parts of mind. They may, indeed, be subjective objects — that is to say, the special characteristics of what is pre- sented, the dream table or any other object, may depend, in part at least, upon the particular mental condition of the subject who perceives. But that is beside the point. To be partially dependent upon mind is not the same thing as to be part of mind. To argue in that way would be as futile as to main- tain that because a steamer cannot move without coal it follows that the coal is part of its motion. o 18 PROBLEMS OF THE SELF chap. Or, again, consider certain judgments. Suppose, for instance, that we judge that similarity depends upon identity in the end, or that relations cannot subsist without terms. The objects of these judg- ments are not psychical. There is nothing specifically mental about similarity, or identity, or relation. On the contrary, they are universals, and each mind, and each act of each mind, is particular. But these universals are surely not physical. They may hold of particular physical objects existing in space, but they are not themselves spatial \ and physical objects are. It is impossible, therefore, to define an ex- perience, or any part of the self, by saying that it is that sort of existent object which is not physical. Such a definition would include, as part of the self, much that is not part of it at all. It is necessary, therefore, to adopt some other mode of enquiry. Let us consider the previous instances again. In both cases, those of the percepts of sense and those of the universals which form the objects of certain judgments, there are two elements to consider, and these, though inseparable, are none the less distinct. The act of being aware of this or the other object, whether that object be perceived or intellectually apprehended, must be dis- tinguished from the object itself. I may judge to- day that the diameter and circumference of a circle are incommensurable, and I may judge it to-morrow. The acts of judgment are distinct because they occur at difierent times, but the object in each case is one and the same. That particular relationship may be apprehended at any time, and the time at which it is apprehended is irrelevant to it. The distinction, in this instance, is particularly obvious. In other instances it is not so obvious, but it is, none the less, necessary and important. Acts of perception and judgments of the intellect necessarily refer to an object, and the latter are either true or false : they are never PSYCHOLOGY AND THE SELF 19 identical with the object to which they refer, and there cannot be knowledge unless there is both act and object. The temporal reference, implied in this example, -may seem only a special case. It is not really so. The act of knowledge is part and parcel of the stream of time : it is conditioned by what goes before, and leaves its mark on what succeeds, but the knowledge itself must be timelessly true, and error is timelessly false. That is true even when the object of judgment is not a universal but a particular event in time. It is timelessly true that King George V. visited the Potteries in April of 1913. Time cannot impair the truth of that statement. It was true before the stars began : it will be true when the moon is no more. It will not, in all human probability, be always true that he is reigning now. But the reason is that the word ' now ' is ambiguous, since it may refer to a variety of dates. Perhaps this obvious distinction is less manifest in the case of acts of perception than in that of acts of judgment. For the object of perception may be contemporaneous with the act, and exist only when and so far as the act exists. Let it be so. The distinction between the act and the object still holds. A cathedral, as it is presented to the mind, is grey and made of stone. The mind is neither coloured nor stony. Simple as these considerations are, they still require some explanation, but the reader will probably com- prehend the explanation more readily if he understands, in a general way, the conclusion to which this analysis points. We have been dealing with a special class of instances, those, namely, of cognition, and have found that in any cognition two elements must be distin- guished — the act of awareness, and the object of that act. In many cases, so far as the argument has gone, there is no reason for supposing that the object itself is specifically mental. On the other hand, the act 20 PROBLEMS OF THE SELF of awareness is always mental, always an experience and a part of the self; and the subject-matter of that part of psychology which deals with cognition is just these acts and nothing more. The acts are conscious acts, they are experiences,^ and their being is to refer to an object. Cognitive acts, therefore, as thus inter- preted, are parts of the self But it is necessary to offer some further explanations. : ( 1 ) To distinguish between an act of knowledge and the object known does not imply any divorce between knowmg and being, nor does it imply that any act of knowledge is conceivable which does not refer to some object. On the contrary, the act has its being as a mode of reference to an object. If it did not so refer it would be nothing. None the less, the act must be distinguished from the object, since, to mention only one reason, different acts may refer to the same object. It is irrelevant whether any objects can exist towards which there is no such act of conscious reference. This analysis, by itself alone, supplies no reason for denying that they may. It is hardly a paradox to say that Neptune may have existed before any one, terrestrial or super-terrestrial, was aware of the fact. But such a possibility (or probability) does not affect the intimateness of the cognitive relation. When it exists it is as intimate as any relation can be. When it does not exist the question of intimacy does not arise. (2) The use of the term ' object ' also requires ex- planation. It is clear, of course, that the word ' In the previous chapter (p. 13) I used the German term Erlebnisse as if it were synonymous with ' experiences ' in the sense of this essay. It should be noted, however, that the German usage is not quite uniform, since the word Erlebnisse sometimes implies the whole complex act-and- objeot, and this, of course, is very different from the narrower sense in which I am employing the terra 'experiences.' I cannot understand how we can be said, in any strict use of language, to ' live through ' anything except conscious acts, and therefore the extension of the term beyond such acts seems unjustifiable. The fact of usage must be noted, however. This point is mentioned by Dr. G. E. Moore in his paper on The Status of Sense- data, given to the Aristotelian Society in the meeting at Durham in July 1914. PSYCHOLOGY AND THE SELF 21 ' object ' does not mean, simply and solely, a physical thing. A physical thing may be an object of cogni- tion, but so are presentations of sense, and so are universals, and so are experiences when any one thinks of them. I have used the word ' object ' in the sense which implies nothing but the complement of the act of reference, whatever a man is aware of in so far as he is aware of it. It may be objected, however, that this usage is very misleading. Is the ' object,' existentially, merely the complement of the act of reference, as those who believe in the ultimateness of the subject-object relation seem to maintain ? Is it not more usual to think of an ' object ' as something which may exist on its own account, and possess, perhaps, a wealth of qualities which most minds observe only in part, and some minds scarcely at all ? If so, no one can suppose that subjective presentations are ' objects ' in this sense. I agree that there is a difference, and shall try to illustrate it by an example. One man may think of Henri IV. as a French monarch who flourished some centuries ago, a second may think of him as an apostate Huguenot, a third as the husband of La Eeine Margot. We may suppose that such scattered pieces of information are all that John Eoe or Eichard Doe possesses, and that each of them has only one piece of information. In that case there is clearly a distinction between Henri IV. himself, the son of Jeanne d'Albret, many of whose qualities were known only by himself, while others cannot now be rescued from the debris of history, and, to put the matter popularly, the various ' ideas ' of him in the mind of Mr. Doe or Mr. Eoe. The former, to use a convenient distinction, may be called ^ the ' object,' the latter the 'material' of an act of reference. The 'material' ^ For further explanation of the terms 'material' and 'quality' vide, e.g., Husserl, Logische Vntersuchungen, Zweiter Teil, pp. 399 if., and Mesaer, Empfindung und Denken, pp. 50-53. 22 PROBLEMS OF THE SELF chap. belongs to the object, each part of it is characteristic of Henri IV., but it need not exhaust his nature, and it is limited, not by Henri IV., but by the imperfect information of those who are thinking of him. It is best, I think, to use ' object ' in the widest possible sense, and to call the 'material' of a cognitive act an 'object,' instead of restricting that term to the narrower meaning I have indicated. The distinc- tion, ultimate or not, is doubtless important in many cases, and those who will may remember it. But it has no peculiar importance in this connection, except in so far as it may resolve a possible ambiguity and misunderstanding in some minds. In either sense of the word there is a plain distinction between an act of knowledge, which is an act of reference, and that to which the reference is made. The psychologist is concerned primarily with the former, and not with the latter. To proceed. The object of this chapter is to analyse the subject-matter of psychology in the hope of gaining some insight into the nature of the self. In other words, it attempts to explain the nature of experiences. Such ' explanation,' of course, does not imply that these experiences can be defined in terms of anything else. That would be absurd, for no one can know what an experience is like unless he has had it. It may, however, be possible to discover some property common to all experiences and belonging to nothing else, and it is advisable to try to make this discovery in order to know precisely where we stand. Hitherto only one class of experiences has been considered, i.e. acts of cognition, and these, certainly, are very far from being the whole furniture of the self. Still, they are parts of that furniture and, therefore, are important. Their being is to be a mode of reference to an object. On what evidence was this result obtained? Clearly by direct analysis, that is to say, by intro- spection. And here we may pause to consider the PSYCHOLOGY AND THE SELF 23 sources of evidence at our disposal. Any psycho- logical enquiry (and this enquiry, up to the present, has been primarily psychological) has two sources of evidence, direct and indirect. The indirect evidence may be called interpretative. Languages, institu- tions and, for that matter, flint heads and ancient pottery indicate the existence, present or past, of certain kinds of experiences, and are interpreted on the strength of this belief. Direct evidence, however, in matters psychological can only be obtained by introspection, and that, it may appear, is very un- satisfactory. There are arguments which try to prove it impossible. It could hardly satisfy a court of law because it can never muster two independent witnesses. If a man can have introspective knowledge of his own experiences, he cannot have such know- ledge of the experiences of any one else ; and even if two men have similar experiences, who can verify the fact directly ? As this question is of considerable importance, it will be well to consider it. There are three questions to ask : (1) Is intro- spection possible ? (2) If it is possible, is it really the only direct evidence attainable ? and (3) Can it form the basis of a fruitful enquiry ? (1) A well-known objection to the possibility of introspection is that of Auguste Comte.^ Briefly stated, his argument is that the activity of knowledge cannot turn round and catch itself. To this it is sometimes replied that introspection is really retro- spection, and that conscious acts can be remembered although they cannot be observed. That is a view which seems contrary to experience : at any rate the act of introspection and its object seem so nearly contemporaneous that it is arbitrary to insist that in all cases there must be an interval of time between them. Moreover, in the vast majority of cases, that only is remembered which has been observed, and ' Cours de pMlosophie positive, i. 34-8. 24 PROBLEMS OF THE SELF chaf. introspection on this theory would be a startling and widespread exception to the general rule. The difficulty, however, is a manufactured one.^ In the first place the Comtian argument could only hold of cognitions and not of other experiences. The cognitions of feelings, desires, and volitions would not be subject to the objection. For introspection is a species of cognition, and an act of cognition could not be said to' catch itse'lf if it were aware of a feeling or a desire. And the difficulty is manufactured even with respect to cognitions. Every act of cognition must refer to an object other than itself, and there- fore no act of cognition can be aware of itself But why cannot one act of cognition be aware of another act, and why, even, should not the two processes be contemporaneous ? The act of cognition which is the object of the act of introspection will, it is true, refer to some other object. If we catch ourselves thinking, we catch ourselves thinking about something. But this is only a complication, not an objection. When the object of our introspection is an act of cognition the act will be a part of a complex including the object to which it refers. And why not ? (2) It may be questioned whether introspection is the only direct evidence of the nature and existence of psychical processes, but there is no doubt whatever that it is by far the mo^t important evidence. The question at issue is the interesting and much-debated one of the possibility of a direct acquaintance with the experiences of other minds. The accepted theory, strongly opposed by believers in telepathy and others, is that any one self is acquainted with others by a process of inference only. It is interesting to note that the writers who oppose this view frequently base their case upon arguments which, to say the least, are fully consistent with its truth. Professor ' This is true, even omitting the important qiiestion of the sense in ■which cognition can he sai(l to be ap activity. PYSCHOLOGY AND THE SELF 25 Alexander, for instance, writes as follows : ' We may- press a yielding object and become aware of its soft firmness and have besides an experience of our own effort of grasping. But there is all the difference between this and the experience of a hand which in any degree returns the pressure of ours ; and that is why we so much dislike an unresponsive hand which seems to us inhuman and disappoints expectation.' ^ This, surely, is no argument for Professor Alexander's case. The whole point of the opposite contention is that a human hand really behaves differently from a wooden hand or a stuffed hand, and we feel that we can only explain the differences by supposing that the motions of a human hand are directed by a con- scious self in the same way as we beheve that the motions of our own hands are directed. Strictly speaking, this is a side issue with reference to the main question we have in view, but its interest will excuse a slight delay. Let us consider the accepted position. Darby and Joan, in the evening of their days, know one another better than they know themselves, but this intimate acquaintance, ac- ' cording to the theory, is only a complicated inference. It seems direct and spontaneous for no other reason than the potency of habit. The child, too, comes to know that his brother or his nurse is, like himself, a conscious being through inference from their bodily actions and words. A little boy in the nursery knows that his brother is in a passion, because his brother's gestures and contortions are the same as he himself is wont to make when angry. But surely such a theory inverts the order of psychological development and implies a ridiculous degree of sophistication in the infant mind. The plain fact is that a man can recognise and construe the physical symptoms of expression far more accurately in the case of others than in his own, and there is no reason > Mind, N.S. vol, xxii. No. 85, p. 18. 26 PKOBLEMS OF THE SELF chai-. to believe that adults are different from children in this respect. The angry child seldom knows that he has clenched his little fist, still less does he know the aspect of his countenance. But he can easily perceive these facts in the case of others. Accordingly, it is monstrous to maintain that we infer the existence of consciousness in others from the similarity of their behaviour to our own. A more adequate psychology of development would explain the facts in a different way. The child dis- covers that his nurse and his mother will respond to his wants in a way that inanimate objects will not respond, and therefore he comes very early to dis- tinguish between human behaviour and other kinds of behaviour. This type of explanation is certainly consistent with the theory that there is no direct acquaintance with other minds. Through the senses and experience ;the child comes to distinguish between responsive and unresponsive beings, and when he comes to distinguish himself as himself he is able, by a gradual and unconscious logic, to believe without a question that responsive beings have a like nature to his own. And this theory is tenable. It is, of course, an expression in highly intellectual terms of a process of inference which is so obscure and involved that it should not, perhaps, be dignified by the name of in- ference. But, on the other hand, it is not the only tenable theory, and it may not be the most probable. .Even granting the premises of the theory, there is no reason to believe that the child recognises his own self- hood before he recognises the selfhood of others, and unless the impossibility of a direct acquaintance with . other minds is assumed from the outset, there are no grounds for asserting that there is a fundamental distinction between the logical and the chronological orders of discovery. There is something repugnant to the ordinary mind in the doctrine that even those who are nearest PSYCHOLOGY AND THE SELF 27 and dearest to him are known to him only as moving pieces of matter, although he can infer with a high degree of probability that they have thoughts and feelings like his own. Nearly any one would admit that it is easier to describe the actions and words of his fellows than any other part of their being and that it is usual to justify any assertion of their states of mind by referring to their behaviour. Indeed the doubters would be silent if they re- flected that they know their own experiences better than the physical expression of them, while they can describe minutely, or at least recognise easily and readily, the expressive behaviour of others. It is easy to recognise the voice of a friend on the gramophone, but very easy to mistake one's own, and such examples could easily be multiplied. But the current philosophical theory is an illegitimate inference from this fact. Our direct acquaintance with other minds, if it exists, is too fragmentary and ambiguous to be the sure foundation of a theory, and no one should build upon it. But to deny it in toto is equally unjustifiable. Let us put the issue in another way. If knowledge begins with the senses, it is surely rash to maintain that the senses which acquaint us with matter are the only ones. That is begging the question. If there were direct acquaint- ance with other minds there would be an additional sense to those which are usually recognised. And, again, matter or physical objects are not the direct objects of the senses even if they are known through the senses, and therefore it is fair to argue that we may know our fellows as directly as we know physical things. That would not be a legitimate argument if the fact were that we Jirst discovered them to be things, and subsequently inferred that they were things with minds. For that would be a double inference and, perhaps, neither step in it is mathematically certain. But there need not be two 28 PROBLEMS OF THE SELF chap. steps in the argument, and I am convinced that the repugnance of common sense to the doctrine I am discussing is based upon an instinctive rejection of the doctrine that there are two steps. The sound of the human voice means a mind just as surely and just as directly as any complex of sensory presentations means a physical object, and errors in the case of the former are no more frequent than in that of the latter. (3) At the same time it is necessary to admit the importance of the distinction between interpretative and introspective evidence. Introspection, if not the only, is the best and the surest direct evidence which can be obtained of the nature of experiences, and there is no sufficient ground for denying that it is good evidence. It is a sort of observation, akin to perception. Now it is a characteristic of any object which is observed that its distinguishing features cannot be adequately defined in terms of anything else, but can only be pointed out. Moreover, it can only be understood by those who have already observed something similar. Let us consider, for example, the distinguishing characteristics of redness. No one can understand what red means until he has seen something red. He may define its causal con- ditions and enumerate the rate of vibrations in the ether. But a blind man can understand these and yet he does not know the meaning of red. Thus if we rest our case on the results of introspection we can only point out these results and cannot prove its truth by direct observation of others. But that holds of all observation whatever, and where would science be without observation ? It is no real objection that the objects of introspection are, in the sense already explained, subjective. Probably the presentations^ of ' I have used the word 'presentation' both here and in other places, although I am well aware that there is no real agreement between psycho- logists as to its precise meaning. In a sense there is agreement. A 'presentation' is what Locke called o,n 'idea,' and it is used with an PSYCHOLOGY AND THE SELF 29 sense -perception are equally subjective. And yet science continues. There may be no reason to sup- pose tbat two or ten men ever see the same thing. There is every reason to suppose that they have similar presentations and can thence infer the nature of the physical object which occasions the perceptions of both of them. In the same way there is every reason to suppose that the experiences of one mind are substantially similar to those of another and there- fore that the results of introspection have much more than private significance. There is an interesting passage in Locke's Essay which bears upon this point : ' What perception is every one will know better by reflecting on what he does himself when he sees, hears or thinks, than by any discourse of mine. Whoever reflects on what passes in his own mind cannot miss it. And if he even more exclusive reference to sense-perception than his ' ideas. ' In the terminology of this chapter, a ' presentation ' is the ' material ' of any act of sensory apprehension, and I do not think there is any serious objection to the use of the term in this sense. The passage in the text does not require any further refinement of interpretation, however important that might be from some other point of view. It is necessary, however, to indicate briefly the nature of the controversies which cluster round the term. Since a presentation is that which is directly presented, or given, to sense-perception at any moment, it is clear that there will be disagreement about the precise meaning of the term if there is any difference of opinion concerning the nature of what is directly given. And there is a very marked disagreement upon this point. Mr. Bertrand Russell and Dr. G. E. Moore generally use the term sense-data in this connection, and Dr. Moore, still more recently, has adopted the term sensible!. The word, of course, does not matter, but the meaning attached to the word may matter a great deal. A sense-datum, or a sensible, is, as nearly as possible, synonymous with the miniimtm sensibile of Berkeley or Hume, and the natural inter- pretation of the writings of those who adopt this terminology is that the direct object of perception at any moment is a collection of sensibles. But, although all psychologists would agree that presentations are fragmentary, most of them would argue that they are relatively extensive, and most would certainly maintain that they cannot he rightly regarded as a collec- tion of Tninima sensibilia. On the more usual view, a presentation is a more or less fragmentary part of a contmuum, and is felt as such. The incompleteness of a presentation is part of its being. Some writers, indeed, go further and contend that every sense-presentation implies, in some way or other, a reference to a thing. Not only is it felt to be incomplete, but its completeness and its incompleteness are felt to be dependent on some more permanent condition, and the true analysis of what is directly given in sense must iinvolve all these elements. I agree with this last view, but it would be irrelevant to pursue the subject further. 30 PROBLEMS OF THE SELF chap. does not reflect, all the words in the world cannot make him have any notion of it.' ^ But Locke's dis- course was more than a stimulus to reflection. It was an aid and a guide to it. Accordingly, the direct evidence in the present connection depends principally upon introspection. Let us examine our experiences by introspection, and see whether they have any common and characteristic features. We began by considering cognition, and that element in the self requires still more careful examination. When that task is completed we must consider experiences other than cognitive. What is called the ' quality ' of an act of cognition must be carefully distinguished from its ' material ' and its ' object,' and the meaning of this term is best described by reference to an earlier example. Let us suppose that our knowledge of Henri IV. is derived solely from reading Dumas's novel. La Heine Margot. In that case we should think of Henri as the husband of Margot, and the master of La Mole, as a bold yet cautious schemer, as the rescuer of ■ Charles of Valois. We may judge that Henri was all these things. But we may also doubt it, or we may merely entertain the supposition without actually believing that the facts were thus. Novels, we know, are not history, and Dumas wrote scores of them. In such cases the acts of believing, doubting, judging, entertaining, are different psychical acts whose material is the same, viz. Henri IV. as portrayed in this novel of Dumas. They are all modes of cognition, but they are different modes, and it is to this difi"erence that the term ' quaUty ' is applied. It is unnecessary in this place to give a precise or exhaustive account of the various ways in which an act of cognition may differ in 'quality.' It is enough to indicate the nature of the term. Every act of cognition, then, has its being as a ' Essay, Book II. chap. ix. § 2. PSYCHOLOGY AND THE SELF 31 mode of reference to an object. But it is a mode of reference having a distinct quality, and the quality of the act can be discovered by introspection. As it is a distinctive experience, it cannot, however, be defined in terms of anything else, and therefore a great variety can be discovered in the analysis of those cognitive experiences which form so important a part of the self. There is a certain unity in our subject-matter, but there is also variety ; and the unity and the variety minister to one another. The analysis of cognition will furnish a clue to the investigation of the other main classes of experience which are parts of the self, and if we consider the connections of these other experiences with one another and with cognition, we shall have done some- thing towards understanding the self, and something more than merely describing its parts in fictitious isolation. The subject-matter of psychology, so fundamental in a discussion of the nature of the self, has, since the time of Kant (and even before his time),^ been distinguished into the three main divisions of cognition, feeling, and endeavour. This tripartite division has frequently been challenged, but the defence has proved itself stouter than the attack. Let us, then, accept it provisionally and consider what it means. There are three principal questions to ask : (1) What is the scope' and meaning of the terms employed? (2) Is there any common character- istic of the three classes which are thus distinguished ? If there is, and the division is exhaustive, we have reason to hope that this common characteristic wUl be the distinguishing mark of the psychical. (3) What is the principle of the division, its funda- mentum, to use the old logical term, and how are the three classes related to one another ? (1) Cognition has already been considered, but we have not considered feeling or endeavour, and psycho- ' By Tetens and Mendelssohn. 32 PROBLEMS OF THE SELF chap. legists are by no means agreed as to the exact sense in which these terms should be used. Feeling, for instance, is sometimes restricted to pleasure and pain, and at other times is applied to any vague massive experience, such as organic sensation or coenaesthesia. And the controversies with regard to endeavour, or activity, are endless. Is activity an experience, or is it not ? Is it, or is it not, a ' psychological scandal ' ? We cannot avoid these controversies altogether, but to consider them fully would lead beyond the limits of a general survey. It will suffice to indicate the sense in which the terms are used in this essay. By feeling is meant the way in which the subject is affected by any object of which he is conscious. Generally, he is either attracted or repelled by it, and consequently all ' qualities ' or modes of attraction or repulsion are qualities of feeling. Not only pleasure and pain, but the thrill of excitement of which Bain spoke, and the emotions of love and fear, anger and disgust, and the rest, should be classed under this head. So should desire and aversion, despite the fact that most psychologists call them modes of conation, that is to say, of endeavour. That there is a close connection between desire and conation is undoubted. There always is an intimate connection between feeling and endeavour, but desire and endeavour are, none the less, easily distinguishable. A man may desire very keenly to play a game of billiards after dinner, and his guest may desire it equally. All the endeavour that may be required of either is to walk to the table and begin. By endeavour is meant a characteristic mode of conscious experience whose being is to seek an object. Striving, seeking, and the choice in volition, are among its modes or ' qualities.' I shall try, in a later chapter, to explain and defend this statement. (2) The common characteristic of all three classes II PSYCHOLOGY AND THE SELF 33 in the tripartite division is that they are conscious experiences which are not primarily objects for con- sciousness, but modes of reference to an object. This statement agrees with the scholastic doctrine that the differentia of a state of consciousness is its 'intentional inexistence.' Its being is to intend or refer to an object. It means the existence of that object, and not its own. An act of consciousness may, indeed, be the object of another act, i.e. of an act of intro- spection. But that is irrelevant and accidental. The primary and fundamental characteristic of a conscious experience is its reference to an object, whether or no it is itself the object of a further act of reflection. This, I think, is the only common characteristic of that which is psychical, and it seems a meagre result of so much paiaful analysis. It does not tell us what experiences are, but only something about them. It gives no hint of the richness and variety of the self. That information can be given by introspection, and introspection only. But any enquiry of the present sort must be equally meagre. No analysis of the common characteristics of that which is psychical can hope to present a full picture of conscious life, and yet the result need not be altogether useless, just because it is meagre. It may show the plan and structure of a conscious self, and if it does so it is justified. Indeed, a more plausible objection would be that the analysis is false. It holds of cognition, no doubt, but in what respects does it hold of the other divisions of consciousness ? Cognition must be of something. But does seeking, or striving, always refer to an object? May there not be a blind striving? This alternative analysis, although plausible, must be re- jected. Striving implies a prospective attitude, which may be vague, but must be present, and be part of the being of the process of striving. Volition clearly implies an object, for in all cases of volition we 34 PROBLEMS OF THE SELF chap. choose to do something, and a similar account holds for any other species of endeavour. The mode of reference is different from that of cognition. But the reference itself is an essential constituent of the process of endeavour. The case of feehng, however, is not so clear. Feeling, according to Hamilton, is ' subjectively subjective,' and that, being interpreted, means that feeling has no reference to an object, but is, simply and solely, a state of the subject. This position is certainly not unreasonable, and I shall try to consider it more carefully later, but the presumption is against it. It seems plain that the feelings of love, or hate, or desire imply reference to an object. They are ways in which the subject is consciously affected by an object, and to be affected in this way implies reference to the object. . Indeed, the very instances of pleasure and. pain on which Hamilton based his case lend themselves more naturally to an analysis which is precisely the reverse of his. Is it not true, as a general rule at least, that we are pleased with something, and pained at something ? If that is usually the case there is a strong presumption that feeling, also, refers to an object, and is similar to other experiences in this respect.^ (3) What is the fundamentum of the tripartite / division ? If the division is logically sound it must have a principle, and that principle cannot be merely the difference between one class of experiences and another. The assent of judgment, for instance, is much more closely allied to the choice of volition than to the attitude of doubt. But doubt and judgment are instances of cognition, and choice pertains to endeavour. Indeed the three members of the tripartite division are so closely connected with, one another, so indispensable to one another, that it is much harder to distinguish them than it is to 1 For a fuller discussion see Chap. IV. PSYCHOLOGY AND THE SELF 35 distinguisli assent from mere supposition, or desire from aversion. The funda/mentum, therefore, must be sought in a different direction. If every experience is a mode of reference to an object, and if this characteristic is important, then it ought to be of use in helping us to distinguish the great classes of experiences. And examination shows that it does. Cognition is a mode of reference to an object. It includes perception, doubting, believing, affirming, denying, and these have, at least, certain negative characteristics in common. They do not attempt to change their object, and they cannot occur simul- taneously when there is reference to one and the same object. A man may doubt one part of the ,story which he hears, deny a second part, and believe in the rest, but he cannot both doubt and believe the same part of the same story at the same time. The first characteristic, however, is of more importance, and consideration of the second may be postponed. Cognition is adynamic. It seeks to know its object, not to change it. If it changed the object it would frustrate its own aim. For then it would know (or doubt, or deny), not that object, but something else. Cognition neither changes its object, nor is affected by it. It is, by its essence, impartial, unbiassed, disinterested. It is not an impression, but an act, and it is an act which is adynamic. On the other hand, the two remaining members of the tripartite division are dynamic, though each in a different way. Feeling is the way in which the subject is consciously affected by the object. It is essentially passive. It expresses the action of the object on the subject. In this respect it is manifestly different from cognition, and endeavour is equally different. Endeavour is the conscious action or re- action of the subject towards the object. Its being is to affect the object, to change it or to sustain it 36 PEOBLEMS OF THE SELF chap. when, without endeavour, the object would disappear. This is the aim of volition and striving. The result of this analysis, therefore, is a division of experiences which meets all logical requirements, and is based upon a principle, or fundamentum divisionis, which is necessary and important. That is clearly true of the subdivision of the dynamic experiences. Dynamic influence must be either passive or active. It must express the way in which something is affected, or in which it affects something else, and we find, not by an d priori or arbitrary procedure, but by a direct interrogation of introspection, that there are ex- periences which fall within each class. The distinction between dynamic and adynamic may seem, indeed, to be on a different footing. Logically it must be exclusive and exhaustive, because it is a division by dichotomy ; but divisions by dichotomy are often merely formal amd yield no important information. In the present instance, however, that criticism is not pertinent. To describe cognition as adynamic may seem to be a mere negation, but is not really so. The description seizes on one of the most fundamental characteristics of cognition as it appears to intro- spection, and gives as much definite and positive information as any general description could ; for it emphasises the most vital distinction between cognition and other experiences in a way which no other distinction does. It is hardly too much to Say that this distinction, if clearly recognised at the outset, saves a world of barren controversy. For the proof, consider the discussions concerning the passivity of sense -impressions and the activity of spirit. Two objections may occur to the reader. He may say, in the first place, that cognition is dynamic because it involves active attention. That is true, for the most part, although there are numerous ex- ceptions unless the word ' attention ' be misused, and u PSYCHOLOGY AND THE SELF 37 defined so as to include all degrees of inattention. It is not infrequent for ideas, and sometimes import- ant ideas, to occur without any conscious effort of attention, and they sometimes dwell with a man when he makes no active effort to keep them before his mind. Apart from such exceptions, it is true that cognition implies attention, but cognition is never identical with attention. Attention is an attitude of endeavour, not of cognition. In it the subject seeks to fixate an object, to keep it before his mind, in order that he may see or understand what it is. Accordingly, attention ministers to cognition, but is not identical with it. If proof be required it is enough to point to the very exceptions mentioned above. How could a man ever entertain ideas with- out active attention, if attention were identical with knowing ? The objection only illustrates how closely our different experiences are blended in the self, and how impossible it is to disrupt the self into isolated experiences. The consequences of that implication will be seen more fully in a moment. The reader may object, in the second place, that to speak of experiences as dynamic, implies causal agency, and that it is quite impossible to understand experiences as either causes or effects. In what sense, he may ask, can we be sure that experiences of en- deavour affect anything ? If they do, as in the cases in which the striving for movement results in actual physical movement, it is clear, at all events, that no one can explain how a mental endeavour can affect a physical event (such as the movements of the brain), and therefore it is absurd to claim as a fundamental characteristic of certain experiences what is, to the last degree, doubtful and problematical. Or, again, he may say that when, in feeling, we maintain that an ' object ' affects us, we are blind to the sense in which we have defined the term 'object.' If an object may be a presentation, or a universal, how can 38 PROBLEMS OF THE SELF ohap. these cause anything? A physical object, perhaps, can be a cause and can have effects, but physical objects are clearly not the only ones which affect our feelings. A mathematician may be delighted with his proofs, a philosopher with his arguments, an opium-eater with the visions revealed to his imagina- tion. But neither proofs nor arguments nor dreams are physical objects, however intense the enjoyment they may cause. To these objections it is enough to reply that the analysis of this chapter is merely introspective. No claim is made to decide the ultimate questions of what causation means, or what can be a cause. The problem is far simpler and is only this ; what do experiences /ee? like, and how do they reveal themselves to introspection ? In that restricted sense the being of a feeling is to be passive, the being of an endeavour to act towards an object, and, in the same 'restricted sense, the analysis is not open to objection. If the principle of this division is really funda- mental for the analysis of mind it ought, at the same time, to furnish a positive basis for the constitution of mind. And it is manifest, on closer inspection, that it does. The members of the tripartite division are not isolated, though they are distinct, and they are not unrelated. Each has a positive function to perform in the economy of the self, none is un- necessary, all are complementary. The proof of this statement would require an exhaustive examination in detail of the characteristics of all experiences, and cannot be attempted within the limits of this essay since the matter of fact is infinite in richness and variety. But the sequel will afford numerous tests of the accuracy of the statement, and will show that it has not been arbitrarily or dogmatically assumed. Indeed it is possible to supply, in a tentative and provisional way, a formula for the general type of connection which exists between the 11 PSYCHOLOGY AND THE SELF 39 members of the tripartite division. Endeavour is guided by cogriition and prompted by feeling. The principle of division involves, at the same time, a principle of connection, as every good principle of division should, and therefore it serves as a clue to subsequent enquiry. We have now completed the analysis of the subject-matter of psychology, and have been led to results of considerable importance. In the first place we have seen that the subject-matter of psychology consists of acts of reference to an object ; in the second place, that these acts are connected together by a principle. Since psychology is the positive science of mind, it might seem that the self can have no content other than that studied by psychology, and therefore that the only feasible line of enquiry into its nature is to pursue the clue of the connection of experiences, so far as we may. The sequel will decide whether such a course could be justified. Meanwhile, it is possible to affirm with some certainty that a * cross-section ' of the life of the self at any moment shows a plurality of experiences, fused and blended together. Broadly speaking, it consists of experiences of cognition, feeling and endeavour re- ferring to some object and related together according to the formula described. There may be a preponder- ance of one of these elements or attitudes, e.g. the state of mind of a man who returns to consciousness from a fainting fit may be one in which feeling predominates over cognition or endeavour. But no one element is ever entirely absent. Not only Tnay the elements be connected ia the way described, they must be so connected ; and this fact, for such it is, is an additional proof of the importance of the tripartite division. I have already remarked that it is impossible to have different cognitive attitudes simultaneously with respect to the same object, and that is also true of the ' qualities ' of 40 PEOBLEMS OF THE SELF chap. feeling or endeavour. It is otherwise with the members of the tripartite division themselves. Not only may they occur simultaneously but they must so occur. Now it may be true that a self or a soul over and above the experiences is required to explain even the unity of a momentary cross-section of its life. In any case the unity of the self is much more than this momentary unity. The cross-section is not self- subsistent. It has its roots in the consciousness of the past, it looks forward to the consciousness of the future. And it may seem that the existence of a soul is implied whenever, such phrases as ' the way in which the subject is affected,' ' the life of the self,' ' our mental experiences ' are employed. To have avoided these phrases would have been ridiculous pedantry, and therefore there is at least a verbal diflBculty in the way of those who advocate the claims of psychology without a soul. 'Mind splits up into consciousnesses,' says Professor Titchener. ' A consciousness is a mental present . . . a bit of mind that is occupied with a single, however complicated, topic. Thus to put the matter crudely we begin the day with a getting-up consciousness : this is followed by a breakfast consciousness . . . etc' ^ But who are the ' we ' who begin the day, and does not the phrase ' a bit of mind ' itself imply that very soul which Professor Titchener is so anxious to ignore ? I do not wish, in this place, to defend the theory of psychology without a soul. On that point, indeed, I find myself in substantial agreement with Husserl : ' The attempt to defend a psychology with- out a soul, corresponds to the theory of a science of nature without bodies. The first theory speaks of a psychology which abjures every metaphysical assump- 1 Article ' Psychology ' in Encyclopaedia Americana. Quoted by Giuender Psychology without a Soul, p. 29. PSYCHOLOGY AND THE SELF 41 tion with regard to the soul: the second rejects in advance every theory that touches the metaphysical nature of the physical world.' ^ It is useless to pre- tend that any analysis can proceed without making assumptions, but that, in itself, does not show the futility of analysis. Perhaps a soul is required in addition to the psychical processes whose nature has been described. Perhaps these processes must, if they are thought truly, be thought as the states or the activities of an ego. Whether that be so, or not, it is certain that consciousness, even the conscious- ness of the moment, is not a mere unit or a collection of units, but involves a very intimate connection of different modes of reference to an object. And the unity and continuity of the self throughout the whole tract of time during which it exists is as ultimate and compelling a problem as this unity of a ' cross- section ' at any particular moment. If the self were nothing but the unity of these experiences, it could not be explained without detailed considerations of the respect in which the experiences form a unity, and of why they must do so. The theory of a psychology without a soul must certainly be rejected if it implies the neglect of any of these considerations. At the same time it is false that the use of personal pronouns, and the like, compels the inference that the self is more than a unity of experiences or that the stuff of the soul includes more than the subject- matter of psychology. The inference is not a necessity of thought or, even, an implication of speech. It depends merely on a misconception of the nature of analysis. ' Following life in creatures we dissect, we lose it in the moment we detect.' If analysis of the self really did dissect, then it would be impossible to reconstruct a self out of those disjecta memhra. If consciousness really split up into consciousnesses, then the splitting would doom this enquiry or any ' Logisehe Untersmhwngen, Bd. II. S. 339. 42 . PROBLEMS OF THE SELF chap. similar one to fatuity. But analysis has no such im- plication, and if we could not analyse the nature of the self, we could only declaim about it, and could not think about it. It is possible, though not, perhaps, necessary or probable, that the term 'self expresses nothing but the fact that any experience whatever forms part of a connected individual whole of experiences. There is no good evidence for any other sort of experience, for any ' floating ' psychical state. And because of that, it is impossible to avoid the personal pronouns and other tell-tale terms. The terms express a profoundly important fact about the nature of consciousness. They do not assert the existence of a peculiar entity, or determine the answer to any particular metaphysical problem. This problem, the problem, of the soul, is the culmination of this enquiry. But it ought not to be approached witheut a careful and prolonged prelimin- ary survey, and the aim of the earlier part of this essay is to perform that task. The broader aspects of the problem are plain enough, and are matters of general agreement. There are three principal questions at issue : ' What are experiences ? ' ' How are they united ? ' and ' What are the presuppositions of this unity ? ' The first of these questions has been discussed ta a general way in the present chapter, the second and third are discussed continuously from the ninth chapter to the end of the volume, and the interval between the present chapter and the con- cluding chapters is occupied by the consideration of a range of problems which are too important to be neglected, and must be considered very fully if they are to be considered at all. There are risks in this procedure, and the chief of them is that the unity of the whole discussion may seem to be sacrificed to an undue preoccupation with mere detail. But that is a risk which must be taken. First of all there is the problem of the relation of n PSYCHOLOGY AND THE SELF 43 the self to the body. The plain man considers that his body is part of himself. Is he right, or is he wrong ? And, again, the body is frequently regarded as the permanent ground of the unity of the self That problem, in various forms, is bound to arise at many points in this discussion, and the principal features of the problem are the subject of the next chapter. The discussion from the fourth chapter to the eighth (inclusive) has a somewhat different object, and the questions discussed in it include the majority of the problems which are usually considered to be, par excellence, the problems of the self If it is granted that experiences afford the clue to the nature of the self, it is possible to argue from that basis to very different conclusions. For it may be maintained that some one particular kind of experiences reveal this clue with an adequacy which no other kind can approach. Cogito ergo sum; sentio ergo sum; ago ergo sum. Most of the arguments under this head fall naturally into the division of the primacy of feeling, will and cognition respectively, and I shall consider them in this order. The arguments for the primacy of will must obviously receive more attention than the others because they are more numerous, more persistently defended and, at any rate in appear- ance, more important. I shall try to show that none of the great divisions of experiences has an invariable or essential primacy. Not merely is each of them essential and irreducible to any other, but the particular arguments in support of the primacy of any one of them cannot bear the test of critical scrutiny. Unless the question of primacy is settled, there is no possibility of an adequate discussion of the unity or the substantiality of the self An answer to the question which is not precise and detailed is no answer at all, and too many of the current accounts of the self owe all their plausibility to a hasty assumption of the 44 PROBLEMS OF THE SELF chap. primacy of will, or the primordial importance of sensation. If I succeed in avoiding a fatal error of this kind I shall at least have accomplished some- thing. In a way the discussion in these chapters is a vindication and an explanation of the point of view which, is outlined in the present one. But it is also more. In conclusion, it is necessary to call attention to two points that must be constantly remembered in any discussion of the nature of the self In the first place the self, as we know it, grows and develops in time, and different selves are widely dissimilar in the degree of their development. It would be compara- tively easy to give an account of the self which would hold for a mature, harmonious and consistent person- ality. But difficulties thicken when we remember that even such fully-rounded personalities were once in their cradles *and yet are supposed to form the same individual psychic centres throughout their lives. Indeed, if the arguments of Semon ^ or Samuel Butler,^ on the subject of organic memory, have any weight, it is necessary to suppose that the beginnings of the psychical life of a single human self stretch back through many generations. Similarly it is necessary to give an account of garrulous selves as well as of silent ones, emotional selves as well as restrained ones, contradictory selves as well as logical ones. The general theory must be exceedingly elastic. ' Finally, we must submit a theory which is able to take account of capacities and dispositions ^ as well as of actual events. I do not mean to say that a capacity in itself is an existent entity, but I must insist that the phrase is not meaningless. It is a commonplace that character is the most important constituent of personality, and character is most ' In Die Mneme and Die mnemischen Envpfindungen. ^ In Life arid Habit and Unconscious Memory. ' For a further discussion of the precise sense in whiph these terms are used see Chap. X. PSYCHOLOGY AND THE SELF 45 accurately defined as the capacity for responsible behaviour. But it is needless to suppose that a man's character is completely revealed to any one who is aware of all his actions, or even, per impossihile, of all his secret thoughts. Suppose, for instance, that we were acquainted with the inmost feelings of the man George Jeflfreys when he browbeat offenders at the city sessions, when he embraced his favourites in his cups, as well as with the facts that he ordered Alice Lisle to the stake or that, when in prison, he was presented by his enemies with a barrel containing a halter instead of a barrel containing oysters. We should argue from these facts to his brutality, his sottishness, his cruelty and injustice, and we should thus understand the gift of the halter. But in so doing we should argue not only to the ways in which he had acted, and the thoughts which actually crossed his mind, but also to the ways in which he would have acted, had different circumstances presented themselves. In other words, we should include capacities, as well as actual behaviour. And this, accordingly, must be another feature of any true account of the self. CHAPTEE III THE SELF AND THE BODY Probably no single question lias been more fully dis- cussed in modern philosophy than that of the relation between mind and body. Treatises upon the question appear continually, and some of them, Dr. M'Dougall's Body and Mind, for instance, are so admirably lucid, accurate and comprehensive that they hardly leave more to others tlian the useless opportunity of re- counting a well-worn tale. There is still work to be done on the traditional lines, but such work would not be relevant in the present connection. In any case I do not wish to attack the question upon these lines. Instead of asking how we are to understand the relation of body and mind, I wish to ask whether upon careful reflection any parts or features of the body ought to be considered parts of the self This road has, I think, been less trodden than the other. It deserves to be trod. Unquestionably the body is the most important or, at least, an essential part of the self on many theories and for many men. ' That complex of fre- quent associations,' says Miinsterberg, ' a complex which at first embraces only the presentations of our own body and its immediate environment, but in later years annexes the whole circle of our interests and ideals, is our very self, our personality.' ^ This ' Die WilUnshamdlung, S. HI. 46 CH^P. in THE SELF AND THE BODY 47 analysis agrees very closely with James's account of the empirical self to which reference has already been made, and there can be no doubt that it represents a stage in the development of the idea of self, and corresponds closely to the popular, unreflective idea of self. But the reason f&r this is that early thought and popular thought cannot distinguish with sufficient accuracy between the physical body and the psychical seK, and I do not see how any one, when he is fully aware of the meaning of the question at issue, can seriously maintain that his body is part of himself. It is natural and intelligible to say, ' Body, thou hast grown old along with me, and my infirmities have almost kept pace with thine ' ; but, unless we were Egyptians and believed in a double soul, it would be meaningless to substitute the word 'soul' or 'self for ' body.' When we address our souls in soliloquy we are addressing ourselves. It is otherwise, how- ever intimate the connection between body and mind, when we think of our bodies. But while many would accept the principle, most remain blind to its implications, and in this chapter I wish to investigate these implications as precisely as possible. It seems to me that many bodily states are accounted parts of the self because there is a failure to recognise that they are really bodily. The investigation will lead to strange para- doxes — indeed, perhaps, to paradoxes so strange that it may seem simpler to deny the original assumption of this chapter. Be it so. The implications deserve to be considered. I propose to begin this discussion by considering a well-worn phrase, now out of fashion, ' the internal sense.' The phrase is slightly equivocal as it in- cludes 'innere Beobachtung' as well as 'innere Wahrnehmung,' to use Brentano's terms. 'Innere Beobachtung ' is equivalent to introspection, and that is the meaning of the 'internal sense,' as the 48 PEOBLEMS OF THE SELF chap. term is used e.g. in Locke's Essay. ' The mind,' he says, 'receiving the ideas mentioned in the fore- going chapters from without [ideas of sensation, etc.] when it turns its view inward upon itself, and observes its own actions about those ideas it has, takes from thence other ideas, which are as capable to be the objects of its contemplation as any of those it received from foreign things.' ' On the other hand ' the internal sense,' in the sense of ' innere Wahrnehmung,' means, not the acts of introspec- tion, but the experiences which are the objects of such acts. The equivocation, however, if it be an equivocation, is easily understood. When the mind turns its eye inward, it contemplates those objects which are inward. The criticism to which the phrase has been sub- jected seems very pertinent. ' Internal ' refers to space, and consqiousness is not extended nor does it occupy position. No doubt the analogous word ' introspection ' has a metaphorical reference to space, but so has the phrase ' a close connection.' The metaphor in these cases is ignored. It is only an accident of etymology. But the word internal is more than metaphorical. It tends to be a literal state- ment, and if it is taken literally, what can it mean ? Does 'internal' mean that which is enclosed within the periphery of the body, or some kernel yet more deeply hidden, the cortex, say, or some spot which philosophers or anatomists have seen fit to describe as the probable seat of the soul? In that case, we shall be told, it is utterly irrelevant to genuine intro- spection. And the word 'sense' is also misleading. The senses which we know are conditioned by some stimulus, and are correlated both with a peripheral and with a central organ. In what respect does the exercise of an act of introspection require a stimulus ? And where is the organ of the internal sense ? ' In- ' Essay, Book II. chap. vi. Ill THE SELF AND THE BODY 49 ternal,' in fact, ought to mean ' mental,' and should not refer to a ' sense ' at all. The phrase lingers only as a decrepit anachronism, deriving its sole significance from antithesis to an obsolete theory of sense-impressions. When the word impression implied Democritean phantasms, or Aristotelian ' sensible species,' which flitted into the mind's presence- chamber like airy sprites bestriding the gossamer, then it was necessary to invent a phrase to describe the contents of that presence-chamber itself. These were called internal because they never came in and could not go out. But when this cause of error is laid, the effect should also cease. At the same time it may be doubted whether the objections to the phrase ' internal sense ' are really so cogent as they appear. In the first place, the objec- tion which rests upon the absence of a specific organ is irrelevant and external. The nature of sense- perception is never made clearer by physiology. A knowledge of peripheral organs and nerve fibres and central organs may help us to understand the con- ditions of sense -perception, but not its specifically mental features. And why, for that matter, should there not be a special organ for the internal sense, even if we use the term in the sense of introspection ? That organ could not be a peripheral organ, but why should not acts of introspection have a special correlate in the cortex ? I do not maintain that it must be so. On the scholastic theory the intellect has no cerebral correlate although all other experiences have, and the exception may also hold of acts of introspection. But why should it ? The usual hypothesis is that there is a neurosis for every psychosis. The pre- sumption, accordingly, is that acts of introspection, being a specific class of experiences, have a specific brain centre. The internal sense, on its current interpretation, should have as its province whatever is ' in mind.' E 50 PEOBLEMS OF THE SELF chap. The use of this phrase, however, is misleading. The antithesis between inner and outer is a heritage of 'the way of ideas,' and. the word 'idea' is probably the most ambiguous in the whole realm of philosophy. An idea was defined by Locke as 'whatever is the object of the understanding when a man thinks ' ; and even Locke was too apt to consider that such objects are invariably of the type of sense presentations. This confusion became worse confounded when it was further assumed that every ' idea ' was a modification of mind. In that case every idea was necessarily ' in mind,' siuce every idea was a modification of mind ; and every modification of mind must, of course, be mental. It would follow thatjevery object whatever was in mind, since ideas are the only objects of the understanding, and accordingly there would be no difierence between internal and external. It is true that Berkeley, at least, did not, or did not always, use the phrase ' in mind ' in this mis- leading sense. ' Those qualities,' he says [i.e. ex- tension and figure], ' are in the mind only as they are perceived by it — that is, not by way of mode or attribute but by way of idea. And it no more follows that the soul or mind is extended, because extension exists in it alone, than it does that it« is red or blue because those colours are on all hands acknowledged to exist in it, and nowhere else.' ^ The meaning of this statement is very clear. Berkeley has been dealing with the presentations of the external senses. These, he says, are 'in mind' in a certain sense. They are, in part at least, mind-dependent existences, and their being is to be perceived. But they are not parts or modifications of mind. They are not mental in the sense in which the acts and operations of mind are mental. Berkeley, indeed, although his psychology of the 'internal' sense is obscure, maintains that the acts of mind, those at ' Principles, § 49. Ill THE SELF AND THE BODY 51 least which, following the example of Locke, he can classify as instances of ' perceptivity ' and ' motivity,' are not presentations at all, although they are 'things' or ' objects ' whose nature can be understood. Even granting, however, that Berkeley was right in his interpretation, and that the ' internal sense ' may be used legitimately with his meaning, the fact still remains that ' introspection ' is the better term because it is the less likely to mislead. On the other hand there remains the possibility that the literal meaning of the phrase (and, especially, its spatial reference) may have considerable importance. It seems to me that it has, and that the literal con- notation should never have been discarded in favour of the metaphorical. This is not merely a question of terminology, for there is no other term to express the precise spatial distinction which is in question, and that distinction is the crux of the problem of the self and the body, as it is considered in this chapter. I wish, then, to maintain (1) that the term * internal sense ' has a clear and legitimate meaning, although that meaning is neither synonymous with introspection nor with the objects of introspection ; and (2) that the objects of the ' internal sense,' in this new signification, have no right to be considered parts of the self at all. That is the question at issue. Most of us would agree that the self must be distin- guished from the body, and that anything revealed to introspection is part of the self; and the argument of this chapter presupposes the truth of that principle. I hope' to show that the objects of the internal sense (in the signification I am defending) are really parts of the body, and therefore are not parts of the self. But they are liable to be confused, in many ways, with the objects of introspection, while theix bodily features are quite unmistakable ; and therefore both the psychologist and the plain man are apt to be hopelessly confused when they try to determine the 52 PEOBLEMS OF THE SELF chap. precise boundary between the body and the self. The ambiguity of the words 'internal sense' is a standing ^lustration of the ambiguity attaching to the things, and that is why the phrase was selected for discussion. (1) If the words be taken literally the 'internal sense ' ought to refer to that which is perceived within the body of the percipient, and the phrase ' external sense ' to that which is perceived outside it. These are the specific objects of the 'internal' and 'external' senses respectively. It may be objected that this distinction, while real, is utterly unimportant. Careful reflection, however, will show the contrary. What is it that is localised within the body ? Is it not the pleasures and the pains of certain parts thereof, together with organic sensations, thirst and hunger ? These form a different class of perceived objects, a different class of presenta- tions, from those of the external senses. When we perceive the body of another, even under the dissecting knife, we have no presentations which differ in kind from those of the inorganic world. The body, like other presented things, is coloured, odorous, and the like. We know, theoretically, that our bodies, even within the mask of the epidermis, would be perceived in this way by others, and even, in the case of a minor operation, might be similarly perceived by ourselves. But, in general, the interior of the body is opaque to sight, or touch, or smell, and yet we are aware of it, and can localise sensations within it. Why should we deny that organic sensations, and the rest, are as truly parts of the body as anything which we, or others, can recognise through the other senses ? There really is an internal sense whose objects are certain real states of the body. The acts of sensing these objects are parts of the self, but the objects themselves are not. The throbbing volume of an aching tooth is as truly a state of the physical THE SELF AND THE BODY 53 tooth as anything which the dentist can perceive through the aid of his instruments. Consider, for instance, organic sensations,^ feelings of fatigue, effort or strain, feelings of physical pain, of general nervous excitement, or the bodily unrest which often accompanies desire or expectation. In what way do these differ from the presentations of external objects if the word 'feeling' is used in its popular and not in its technical psychological sense ? Both have quality, position in space, intensity, durati'on. Both are essentially presentations, material for acts of perception, not themselves psychical acts. They are objects for consciousness, not acts of reference to an object. A careful examination will show three distinctions that can be drawn, and will also show that no one of these distinctions is sufficient warrant for maintaining that the two classes of objects are in a fundamentally different category. (a) In the first place it may be said that the objects of the internal sense, especially in respect of ' The distinction between act and object ought to be drawn in the ease of sensation as much as in any other, and the usual accounts of sensation suflfer from neglecting the distinction. There is the act of sensing and the object sensed, and these are not one, but two. In the present chapter I have used the term sensation by itself, although with some risk of ambiguity. Any other course would have led to an irritating degree of repetition, and, besides, the term sensation is quoted so frequently in the discussion that no other course seemed practicable. The term sensation is usually employed as referring to the object rather than to the act. My contention is that acts of sensing are mental and their objects not, and the reader should remember this throughout the discussion. It should be un- necessary to add that the word ' object ' is used in the broad sense explained in the previous chapter, and does not necessarily, or usually, mean a physical thing. The view is sometimes maintained that the distinction between act and object applies to the level oi judgment only, and that there is no difference whatever between the act of sensing and the object sensed, just as there is. 'no difference between the feeling of a pleasure and the pleasure itself. The insuperable objection to such a view has already been mentioned. It is impossible to maintain that the mind is blue whenever there is » sensation of blue, while it is clear that the mind is pleaSed whenever there is a feeling of pleasure, in the strict psychical sense. Probably there is no instance of an act of sensing occurring apart from some degree of judgment, but, in that case, there is also no object sensed which is not also judged. And if the objects of judgment include, in any way, that which is sensed (and, plainly, they must include it) then the true analysis is that anch acts of judgment include acts of sensing. 54 PROBLEMS OF THE SELF chap. localisation, are vague and indefinite, while those of the external sense are definite and exact. We often feel toothache in the wrong tooth, and organic sensations are diffused, voluminous and massive, not definitely localised. This distinction may be true, but it is irrelevant. No one need maintain that the internal sense yields as definite information as the external. The question is only whether it has an equal claim to be called a sense. In our contact with the world we rely principally on the sense of sight. Instruments of measurement depend upon sight in the end. If we touch an object in the dark, we straightway attempt to visualise it. But the fact that sight is the most definite of the senses does not prove in any way that it is the only sense. The instances of erroneous localisation, similarly, only impugn the accuracy of the internal sense, and there is no reason to believe that such inaccuracy is very serious. Let us grant that the viscera are insensitive and that, when they are diseased, pain is felt in some other part of the body. If they are really insensitive it would be an error to localise pain in them. The disease causes pain — somewhere else. And that is where the pain is felt. (6) In the second place it may be maintained as an objection that the object perceived by the external senses is a real thing in real space. The objects of the internal sense are not real in this way. They are symptoms or indications of the real state of the body. The surgeon and the anatomist know the real nature and construction of the human body. When the patient relates his symptoms, he is only furnish- ing indications of that state. Now if ' real thing ' means only that there really is an object for consciousness, then that is true equally of the internal and external senses. In both cases there is something of which the mind is aware. But the statement, probably, means something diff'erent. in THE SELF AND THE BODY 55 In all probability it means that the external senses take cognisance of a permanent, systematic, orderly, physical world whose character does not depend upon the fluctuating, private moods of particular per- cipients, and that the localisation^ of objects of the external senses is a direct awareness of the spatial characteristics of this world. In that case the local- isation of objects of the internal sense would be only secondary. We know or believe that organic sensa- tions, and the rest, are connected with the real state of the body, and so we attribute to them, though falsely, specific spatial qualities. The question of the natute of objects of sense-per- ception is clearly too large a one to receive adequate discussion in this place. Suffice it to say that if, in perception, there is direct awareness of a permanent, orderly world, that awareness is very imperfect and inadequate. It is at least equally probable that there is no such awareness, and that the objects of per- ception are subjective presentations which indicate the existence of a physical object in the sense of that term which has already been described in an earlier chapter of this essay. Each man, in all probability, sees a slightly different space, and for each man the space which he perceives by touch differs in important respects from that which he perceives by sight, though it may broadly correspond. On either theory there is no reason for maintaining an ultimate distinction between the external and internal senses in respect of the reality of their objects. If we believe that what we are aware of directly is a presentation indicating certain qualities in a physical object that is never directly perceived, then the presentations of the internal sense may certainly in a similar way indicate the state of our physical bodies. It is possible to ^ In using the word ' localisation ' I do not mean to imply the special sense in which it is (falsely, I think) distinguished from projection. I mean merely the attribution of spatial position in any kind of space. 56 PROBLEMS OF THE SELF chap. argue from the pain of a patient as well as from his pallor, though the latter inference depends on sight, and the former on the internal sense. If, on the contrary, we believe that we directly perceive a real physical object, then why should not the pain in the tooth of an organism be as real a quality of the tooth as its colour ? That is the natural assumption. (c) The really important difference in the case is that the objects of the external sense are common to many percipients while those of the internal sense are peculiar to a single percipient. The nurse, and the surgeon, and many others may see the wound, only the patient feels the pain. That is a proof of the subjectivity of intra-organic percepts, but it is no disproof of the possibility that they are really states of the body in the sense in which any percept is real. They are percepts because they involve the spatial co-ordination and the objective reference implied by that term. To prove them subjective does not prove them psychical or parts of the self. It only proves that they are objects which cannot be perceived by any other self The fact on any theory is doubtful, for on any theory a certain taint of subjectivity clings to the objects of sense -perception. "What we perceive, on any theory, depends partly upon us. It is not a purely impartial view of the object. After-sensations are blended with the objects of vision since the chemical processes in the retina are comparatively slow and the eye, in this respect as in many others, is by no means an adequate instrument of vision. It is not merely santontne powder or jaundice which perverts colour vision. There are minute individual idiosyncrasies which affect the act of every percipient and similar individual differences occur in the case of every object perceived. But this circumstance em- phatically does not prove that what is perceived is part of mind. It is only mind-dependent in certain HI THE SELF AND THE BODY 57 respects. This consideration leads to the second question to be discussed. The objects of the internal sense should not be confused with the self. Although they differ in many respects from the objects of the external senses, they are at one with these in being objects for the conscious self, not parts of it. But it is also advisable to consider which of the objects of the internal self are commonly accounted parts of the self, and why this confusion is made. (2) The question has very considerable importance in view of many of the /issues raised by modern psychology, and a consideration of these issues will include the whole range of this debatable ground between the self and the body. Let us examine them from the standpoint already indicated. There is a reference to recent controversies with respect to the somatic resonance implied in emotion. There is the implication of one of James's numerous accounts of the nature of activity, accounts in which even the ' spiritual self,' the self of selves, is said to consist of a collection of movements in the head and throat. There is the problem of the distinction between psychical and physical ^ pain, and that of the im- portant part which many purely bodily conditions appear to play in dissociations knd alterations of personality, and in many of the delusions of the insane. True, the psychologist need not attempt to cover the whole ground traversed by the modern alienist. It is as little relevant for him to enquire whether every sort of insanity is due to toxic influence as to attempt to discover an anti-toxin for paranoia, or dementia praecox, or folie circu- laire. His proper task is to distinguish carefully between states of the body and mental experiences proper. Defective analysis in this respect frequently ' The word 'physical' is here used somewhat loosely to indicate that which is referred to as pertaining to the body. The reader must remember the distinctions drawn in Chapter 11., but no serious ambiguity need arise from this lack of precision. 58 PROBLEMS OF THE SELF ohap. leads to false conclusions, and the idea which many a man has of himself is really an idea of his body to an extent that would astonish him greatly were his attention called to the fact. This, in itself, is not surprising.- Analysis is notoriously difficult, and the obvious subjectivity of all objects of the internal sense leads easily and naturally to a confusion with selfhood. Moreover, there is yet a further respect in which the difficulty of introspection is aggravated in this case. Certain characteristics of objects of the internal sense seem to link them' inextricably with the objects of genuine introspection. It is really an old question, this, though it is perennial ; the issue, almost in the sense in which it arises here,^ can be clearly seen in the writings of Descartes, Locke and others. All readers of Descartes know that he believed, despite his dualism, in a very intimate relation between body and mind.^ The mind, he said, could not be merely a spectator of the doings of the body. The relation between the two [to use a metaphor already employed by Aristotle] must be more intimate than that of a pilot to his ship. The soul and the body, for Descartes, formed a melange confus, and the fact appears most clearly, if not solely, in those puzzling states of pleasure and pain, which the Creator implanted in man in order that he might maintain his bodily estate. But the examples which Descartes chooses are not merely those of pleasure and pain, but such states as those of hunger and thirst, and it is clear from the context that he regarded these states as pertaining equally to the soul and the body (if he did not believe them to be a third species of existence which partook, in an indeterminate manner, of the nature of each). They yield a confused acquaintance with the body in much the same sense as colours and tastes yield a confused knowledge of the qualities of the real ^ Meditations, vi. THE SELF AND THE BODY 59 physical object. Indeed it is difficult to see why Descartes thought that these states partook more of the nature of thought than other secondary qualities. The fact that they aid and abet the welfare of man- kind is nothing to the purpose. For on Descartes's theory, and on that of Locke and Berkeley too, one of the chief functions of every sense is to teach man- kind what to pursue and what to avoid. On Locke's theory we do not know, by our senses, the minute constitution of bodies, and on Berkeley's theory sense is but a symbol of the divine reality. In both cases it is part of the being of sense -perception to guide us in the conduct of life. But Locke is much more emphatic than Descartes. ' We have some kind of evidence in our very bodies that their constituent particles, whilst vitally united to this same thinking conscious self, so that we feel ^ when they are touched, and are affected by, and conscious of, good or harm that happens to them, are a part of ourselves ; i.e. of our thinking conscious self. Thus the limbs of his body are to every one a part of himself; he sympathises and is concerned for them. Cut off a hand, and thereby separate it from that consciousness he had of its heat, cold and other affections, and it is then no longer a part of that which is himself, any more than the remotest part of matter.' ^ Locke maintains, therefore, that the limbs and other parts of the body are thought to be parts of the self because of our interest in them and our sympathy for them. But it seems clear, despite Professor James and others, that this sense of the word is metaphorical. Because a man is interested in philately it does not follow that he is, in any sense, composed of stamps. Locke, however, and Descartes also, were influenced by other considerations, prin- • Italics Locke's. ^ Essay, Book II. chap, xxvii, § 11. 60 PEOBLEMS OF THE SELF ohap. cipally by those concerning the nature of feeling. Indeed the passage from Locke seems self-evident until the question of the meaning of feeling is care- fully considered ; and even then, as we shall see, there is something to be said in favour of his argument. We feel our body, and feeling is dis- tinctly and characteristically psychical, but the word feeling is ambiguous. When we speak of feeling our limbs, or the like, we refer to a sort of perception or sensation, and this implies a psychical act of refer- ence and an object presented. ' Feeling' in a strictly psychical sense means something different. It is wholly and entirely an experience and is not a sort of cognition as acts of perception or sensation are. At the same time there are close analogies between ' feeling ' as a psychical experience and ' feeling ' as including objects of the internal sense. We can only understand the pogition of Descartes or Locke if we take account of those analogies. Feeling, as a psychical experience, is passive in character ; it is the affective aspect of consciousness. And it is also subjective in a marked degree even , if it be not 'subjectively subjective' in precisely Hamilton's sense. But although passivity or activity form no part of the attitude of cognition, and there- fore no part of the attitude of sensation or perception, none the less our percepts and the objects of sensation seem thrust upon us from without and are independ- ent, of the control of the will. This is one of the respects in which they are distinguished from the objects of imagination or intellection. If I am in Paris and open my eyes I must see the town in a certain way, but if I imagine Paris to be the New Jerusalem I can picture the Seine as clear as Abana or Pharpar, or yellow like the Tiber, I can picture the Panthdon as composed of rubies and amethysts, and place it by the side of N6tre Dame in the Champs Elysdes. Thus in respect of passivity any object of Ill THE SELF AND THE BODY 61 sensation or perception (and the objects of the internal sense are included in this class) is ranked along with feeling, although the passivity of psychical feeling may differ from this passivity in important respects. And, again, the objects of the internal sense are certainly subjective, since they can only be perceived by a single person. That is the characteristic which is common to the ' internal sense ' and to intro- spection, and it occasions much of the confusion between them. But we have already seen that it is logically irrelevant, however natural the mis- conception to which it gives rise. In the third place psychical feeling plays a part in the life of the self which is analogous to the part which physical feeling plays in the life of the organism. In accordance with our previous account, though without any implications of temporal order, we may say that feeling looks behind to cognition and before to endeavour. Similarly the internal states of the organism are a connecting-link between the stimulus received from the environment and the reaction to- wards that environment. It is a rough analogy, no doubt, but in one way or another it influences dis- cussions on this question. Let us take this statement, for instance : ' It was said above that every instinctive process has the three aspects of all mental process, the cog- nitive, the affective and the conative. Now the innate psycho-physical disposition, which is an instinct, may be regarded as consisting of three corresponding parts, an afferent, a central and a motor or efferent part, whose activities are the cognitive, the affective and the conative features respectively of the total instinctive process. The afferent or receptive part of the total disposition is some organised group of nervous elements or neurones that is specially adapted to receive and to elaborate the impulses initiated in the sense-organ by the native object of the instinct ; 62 PROBLEMS OF THE SELF ohap. its constitution and activities determine the sensory content of the psycho-physical process. From the afferent part the excitement spreads over to the central part of the disposition ; the constitution of this part determines in the main the distribution of the nervous impulses, especially of the impulses that descend to modify the working of the visceral organs, the heart, lungs, blood-vessels, glands and so forth, in the manner required for the most effective execution of the instinctive action ; the nervous activities of this central part are the correlates of the affective or emotional aspect or feature of the total psychical process.' ^ The phrases afferent, efferent and central can only refer to the physical part of the psycho-physical disposition, and even in that case there would be no special reason for supposing that the afferent nerves (or their central nerve-endings) were correlated with cognition, the efferent with conation, and the central with feeling. On the contrary, there is every reason to believe that there is no Innervationsgefuhl, and consequently that our cognition of motor activity is sensory in origin. Nor, again, should the argument be used that the feelings connected with speculative thought are differently related to it than other feelings to other kinds of cognition. Yet in the case of this kind of cognition there is no reason to believe in any afferent nerve process, as there is no reason to believe that any external stimulus whatever enters. Dr. M'Dougall's analogy, therefore, has no especial im- portance even in the case of instinctive actions to which he refers. I have mentioned it only because the analogy seems to weigh with some who consider bodily feelings, or some special class of bodily feelings, parts of the self. Dr. M'Dougall's argument refers only to cerebral localisation, not to the awareness of states of the body ; but the same analogy might be ' M'Dougall, Social Psychology, pp. 32,, 33. Ill THE SELF AND THE BODY 63 applied equally well to the body as a whole. The body as a whole is obviously intermediate, even in point of time, between stimulus and reaction. The passage is also an excellent introduction to a discussion of the distinction between bodily 'feelings' and psychical feelings. It is because the two are confused that the body is considered a part of the self. Let us begin by considering emotion. That surely is a part of the self. What is more intimately part of our being than our love and our hate, or the rarer states in which poetry or music flood the soul with feeling ? The older theories of emotion treated emotion from the point of view of psychical feeling, and they were right in doing so. But the modern treatment lays chief or sole emphasis upon an entirely different aspect, the bodily aspect, and this aspect is also relevant. I cannot consider this question as fully as it deserves, but it is impossible to avoid dealing with it in considerable detail. Modem interest in the question of the nature of emotion was greatly stimulated by the arguments of Professor James. His theory of the emotions is closely connected with his views upon instinct. Under the term instinct he seems to include every impulse, at any rate he maintains that every instinct is an impulse. The meaning of the term impulse is not so clear. He seems to mean that every bodily adjust- ment, simple or complicated, is an impulse, and also that any kind of ultimate preference or liking is an impulse. This usage is very misleading, for what is there in common between a cough or a sneeze and a preference of champagne to ditchwater?^ It is probably true that every instinct involves physio- logical reflexes and impulses ; and is connected with ultimate likes and dislikes, but it does not follow either that these preferences are simply the awareness ^ These examples are all chosen from Professor James himself. See his Prmciples of Psychology, vol. ii. pp. 403 and 386 respectively. 64 PROBLEMS OF THE SELF of physiological events, or that a conscious instinct is only the awareness of such events. Professor James is more concerned with psychological analysis than with coherent psychological theory. The view which he defends in connection with the instincts and emotions is that psychology depends upon physiology, and that consciousness is to be explained by refer- ence to a complicated system of physical reactions together with the connected reflex arcs. It is usual to maintain that an instinct is a sort of racial habit belonging to every member of a given species ; that it is complex and co-ordinated like the actions of a bird in building its nest ; that it is biologically advantageous to the species concerned and tends towards its maintenance and development, although the individuals who act instinctively need not have any conscious awareness of that end ; and that, although it may be perfected through experience, it does not require previous experience as a condition of its occurrence. These characteristics, collectively (and for the most part severally), are sufficient to distinguish instincts from impulses. On Professor James's definition it is unnecessary to draw this distinction at all. He proposes, however, to consider an emotion as almost indistinguishable from an instinct. The ex- pressions of the two are almost identical and, as we shall see, it is the awareness of the expression of emotion which is the differentia of the total emotional state. 'Instinctive reactions,' he says, 'and emotional expressions thus shade imperceptibly into each other. . . . Emotions, however, fall short of instincts, in that the emotional reaction usually terminates in the subject's own body, whilst the instinctive reaction is apt to go farther and enter into practical relations with the exciting object.' ^ James's analysis of the emotional state is as follows : ' Principles of Psychology, vol. ii. p. 442. THE SELF AND THE BODY 65 Owing to the extreme range and variety of instincts, or, as we should say, of instinctive bodily reactions and impulses, an indefinite number of bodily changes may occur in connection with particular thoughts, and it is a universal law that consciousness is necessarily correlated with bodily movement. ' The whole neural organism, it will be remembered, is, physiologically considered, but a machine for con- verting stimuli into reactions ; and the intellectual part of our life is knit up with the middle or central portion of the machine's operations. . . . Every im- pression which impinges on the incoming nerves produces some discharge down the outgoing ones, whether we be aware of it or not. Using sweeping terms and ignoring exceptions, we might say that every possible feeling produces a movement, and that the movement is a movement of the entire organism, and of each and all its parts,' ^ The whole organism, therefore, may be compared to an indefinitely complex and varied sounding board. When we consider its reactions, and to these we are restricted in the case of emotion, we have before us complex waves of ' somatic resonance.' This somatic resonance is the crucial feature of his account of emotion, and its complexity is at least equal to the complexity of man's emotions — the most complex of all his experiences. Now in any emotional state there is both an act of cognition and a feeling of this somatic resonance. We perceive some object and our emotions are aroused, or we think of some idea associated with this object and the same result happens. On reflection, how- ever, we do not find that this perception or this idea difiers from other perceptions or other ideas. Acts of cognition are, invariably, cold and neutral, whether they occur in a state which, as a whole, is emotionally tinged, or whether they do not. The differentia of emotion, accordingly, cannot be found in this element 1 Principles of Psychology, vol. ii, p. 372. F 66 PROBLEMS OF THE SELF ohap. of the emotional complex. James's theory is that it belongs to the other. ' My theory is that the bodily changes follow directly the perception of the exciting fact and that our feeling of the same changes as they occur is the emotion.' ^ ' The emotion, therefore, is, strictly speaking, a sort of sensation. The emotions are sensational processes, processes due to inward currents due to physical happenings.' ^ And it is undeniable, not only that there is this somatic re- sonance, but also that it is felt, obscurely or acutely, the moment it occurs. ' Our whole cubic capacity is sensibly alive ; and each morsel of it contributes its pulsations of feeling, dim or sharp, pleasant, pain- ful or dubious, to that sense of personality that every one of us unfailingly carries with him. It is surprising what little items give accent to these complexes of sensibility. When worried by any slight trouble, one may find that the focus of one's bodily conscious- ness is the contraction, often quite inconsiderable, of the eyes and brows. When momentarily embarrassed, it is something in the pharynx that compels either a swallow, a clearing of the throat, or a slight cough ; and so on for as many more instances as might be named.' ^ Although there is a difference between the older and the more recent treatment of emotion, it would be false to suppose that most modern psychologists agree with James. On the contrary, the balance of authority is against him. But his friends and his foes both admit that the sensational elements to which he accords so much prominence are really important and profoundly important constituents of the total emotional state. His opponents, how- ever, argue, like. Professor Stout in his Manual of Psychology,* that James's argument ' lacks logical stringency.' He rests his case upon the assertion ' Principles of Psychology, vol. ii. p. 449. 2 Ibid. p. 453. Italics James's. ^ Ibid. p. 451. ■■ P. 302. THE SELF AND THE BODY 67 that without the somatic resonance there would be no emotion, and it no more follows that the rei^onance is the emotion than that a man's vision is his spectacles because he cannot see without them. That criticism is unfair because James's argument does not require logical stringency. It is not a logical demonstration, but primarily an appeal to intro- spection, and it should be met on this ground. I shall, accordingly, avoid other arguments, some of which appear to me sound, and others mere mis- understandings. Let us take the point raised in the last quotation from the Principles of Psychology. It is probably true that there is a contraction of the eyes and brows when we are annoyed, aujd that we feel something in the pharynx wh^n momentarily embarrassed. But we are not usually aware of these facts until our attention is drawn to them. They may conceivably be felt obscurely, but certainly not acutely. At the same' time the annoyance or the embarrassment is felt acutely and is certainly a real state of feeling, not merely a cold and neutral intellectual apprehension. This, I think, is obvious on reflection, and I cannot find, in my own case at least, that the result of the analysis is appreciably altered by the remembrance that the contraction of the eyes and brows is not the whole of the somatic resonance in question. It is true that James's account refers primarily to the coarser emotions, but even in their case it is possible to detect the same difierence. Fear is accompanied by dryness of the mouth, by goose flesh, by altered circulation, and so forth. But it will be found that we are frequently aware of these before we are really afraid, just as we also find that these manifestations may continue long after the psychical excitement has passed. Although, then, the bodily sensations and the psychical feelings are blended together and frequently indistinguishable, still, in the 68 PEOBLEMS OF THE SELF chap. main, we can distinguish the two even in those cases which are most favourable to James's case. The attentive reader of the Principles of Psychology cannot fail to notice that James habitu- ally supposes consciousness to be synonymous with cognition. On that assumption there is no other analysis open to him than the one he actually adopts. The emotional state is certainly not merely cognitive, and therefore the differentia of emotion must be sought elsewhere. If that differentia cannot be found in consciousness, organic sensation is clearly its probable source, especially when the concomitance of organic sensation with emotion, and its relevance for emotion, can be demonstrated. But if consciousness and cognition are not synonymous, the burden of proof rests with James, and he has not examined that burden fully. His most characteristic arguments with regard to activity are open to the same criticism, but it is convenient to consider the distinction between psychical and physical pain,^ before proceeding to the question whether activity is really bodily and nothing else. The question of the distinction betw'een pain as a psychical experience and as an organic sensation ought to be decided by reference to two character- istics : (1) whether there is localisation within the body ; (2) whether there is conscious reference to an object. If the former, then the reference' is to sensa- tions which are objects of the internal sense but no more parts of the self than any other presentations, and are ' in mind ' only in Berkeley's sense, if at all. If the latter, the reference is to experiences proper. Analysis in this question is certainly difficult. There seems a balance of probability that pain spots exist on the surface of the body, in the same sense as spots peculiarly sensitive to heat or cold, although • This is a narrower question than that of the distinction between bodily and mental feeling, just as feeling (including the emotions, etc.) is broader than pleasure and pain (or ' hedonic tone ')• THE SELF AND THE BODY 69 one possible analysis is that pain is always psychical, but so intimately connected with certain bodily sensations that it is localised where they are localised. The localisation would, in this case, be an error, but it would be an error most incident to psychology. ' It is through the character of the accompanying sensation that we are able to distinguish different kinds of organic pain and pleasure. Thus we dis- criminate from each other stinging, piercing, gnawing, crushing, beating, shooting and other innumerable kinds of pain. ' ^ These characteristics are undoubtedly physical, and the class of pains which is relevant in this connection is so intimately bound up with them and so definitely localised, that the presumption is that the pain also is as much a matter of sensation or perception as its features of shooting or burning. On the other hand, aesthetic pleasures, and the like, are not localised in the body, and seem to refer to an object, although this latter characteristic has not been sufficiently discussed hitherto. The pain which is a sen- sation cannot be said to refer to an object in any sense. The very fact that the question can be raised shows how kindred in nature and similar in efi"ects certain psychical feelings and certain states of the body are, and it is because of this blending and this similarity that there is so much truth in Descartes's position. Descartes had reason for maintaining that the relation of a man to his body is something sui generis, something he may naturally believe a part of himself. The facts previously adduced show this clearly, and when we remember also that the body is a constant seat of interest, a constant instrument for action, we see, more and more clearly, the reasons which determine the plain man to adopt the view he does. Some bodily sensations are Very like real 1 stout, Manual of Psychology, p. 226. (The references, here and else- where, are always to the first edition, unless otherwise stated ; and any difference between the editions is noted.) 70 PROBLEMS OF THE SELF oHAr. psychical feelings. Moreover, they are usually much alike in the way they influence conduct. The philosopher is quite as impatient of toothache as he is of the pain of a contradiction. He is thwarted by both in a way that pierces his being, and although, if he is strong in will, he can dismiss either for a brief space, he is impelled to seek every means in his power to bring the unsatisfactory state of affairs to an end. Let us consider some other ways in which bodily sensations are apt to be confused with psychical experiences, and so with the self. In the tangled mass of evidence relating to alternation and dissociation of personality, one fact stands out clear and indis- putable. The principal factor which leads to such disintegration is not merely a change in emotion and mood, but the organic sensations connected with it. Loss of sight or hearing does not have this effect, nor do hallucinations except when conjoined with changes in coenaesthesia and organic tone. The attentive student of the Beauchamp case^ will have noticed this fact in every one of Miss Beauchamp's alternating phases, as also in her distinction from ' Sally.' As this evidence falls for more detailed consideration in a later chapter, it need not be fully considered here. It is enough to remind the reader of the connection between the various personalities in the case of Louis Vive, and the various degrees of the anaesthesia of their common body, or to ask him to consider those cases which led M. Eibot to formulate his ' colonial ' theory of the sel£^ In such instances (occasionally) there seems to be a contradiction. The subject declares that he knows that he has become a different person. But even a normal subject is apt to con- tradict himself, and the only inference is that the subject, in these instances, seizes upon a feature which he is wont to consider an important, or the ' As recounted by Morton Prince, The Dissociation of a Personality. ^ Maladies de la personnaliti. THE SELF AND THE BODY 71 most important, part of himself, viz. a certain con- tinuous tone of bodily sensation. When that oscillates between two states which have each a characteristic coherence and continuity of its own, then the subject is at a loss. He retains his memory, but an essential part of what he is accustomed to call him- self ha§ suddenly disappeared. And therefore he contradicts himself It is far more important to discover the grounds of this contradiction than simply to point it out. And in those cases of hysteria in which various apparently different personalities refer to one another as I'autre, or Vidiot, or le scelerat, there is, at least, no contradiction. Professor James's account of the nature of personal identity relied chiefly upon this class of facts. He was certainly in error when he assumed that the question of the nature of the pure ego was the same as that of the nature of personal identity. The ' pure ego,' if it exists, is nearer of kin to the 'spiritual self,' although hardly in the sense in which he uses that term. This, however, does not necessarily affect the accuracy of his analysis of personal identity, which he states as follows. A judgment about personal identity is logically on the same plane as any other judgment of identity. As in any other instance, the judgment refers to the identity of the object about which it is made. What is identical is not 'an act of appropriation, but that which is appropriated. The object of this thought, that which it consciously appropriates, must be part of the empirical self, and the nucleus of this self is feeling. Thought (James argues against the associationists) is not itself a presentation, and personal identity must be sought in the identity of those presentations which the thought appropriates as specifically its own. Certain bodily feelings are the most significant of such presentations, and form the real core of personal identity. It is the identity of these presentations, 72 PROBLEMS OF THE SELF chap. especially of the adjustments which accompany affirma- tion, denial or endeavour, that form the only sort of activity which introspection can discover. To speak of the identity of these presentations is to speak of the 'warmth and intimacy,' or the 'animal aroma,' which invariably clings to them. 'Warmth and intimacy leads us to the answer sought. For what- ever the thought we are criticising may think about its present self, that self comes to its acquaintance, or is actually felt, with warmth and intimacy. Of course this is the case with the bodily part of it : we feel the whole cubic mass of our body all the while, it gives us an increasing sense of personal existence. Equally do we feel the " inner nucleus of the spiritual self," either in the shape of yon faint physiological adjustments, or (adopting the universal psychological belief) in that of the pure activity of our thought taking place as such. Our remoter spiritual, material, and social selves,' so far as they are realised, come also with a glow or a warmth ; for the thought of them infallibly brings some degree of organic emotion in the shape of quickened heart-beats, oppressed breath- ing, or some other alteration, even though it be a slight one, in the general bodily tone.' ^ This all- important factor of ' warmth and intimacy ' is, there- fore, a bodily factor. It is because of the evaporation of animal heat that the grown man feels that his life is foreign to that which he lived when a child. ' No sentiment of his little body, of his emotions, of his psychic strivings as they felt to him, comes up to contribute an element of warmth and intimacy to the narrative we hear, and the main bond of union with our present self thus disappears.' ^ It is true that personal identity, being the identity of a person, is logically on the same footing as any other sort of identity. It is true, also, that at least ' Principles of Psychology, vol. i. p. 333. 2 Ibid. p. 335. in THE SELF AND THE BODY 73 the best starting-point for a discussion of personal identity is to be found in the identity of the self as it appears to introspection. But if the previous argument holds, the body is not the self at all, and its animal aroma, however interesting and however profoundly connected with the hfe of the self, is not part of that life. 'Warmth and intimacy,' then, if literally understood (and Professor James means it to be taken literally), is at most an indication of personal identity and not itself part of the person at any time. Moreover, James himself lays undue stress upon it, as is clearly implied in one of his examples. The psychic strivings of the child, his juvenile breaches of decorum and the like, have neither warmth nor intimacy when the child becomes a man. But the grown man, none the less, continues to believe that he is the same psychic centre which, twenty or more years before, teased his comrades and plagued his schoolmaster. And even if we concede for the sake of argument that the body, or part of it, is part of the self, still we must also maintain that personal identity includes more than a continued warmth and intimacy. What of the identity of cognition revealed by memory ? Do we never remember cognitive and other experiences directly ? Do we always base our remembrance, which is, and must be, personal, upon this identity of animal aroma ? There is another oversight in Professor James's account. It is not very easy to reconcile the statement (p. 340) that * the thought never is an object in its own hands' with the careless grace with which the concluding paragraph of the chapter (p. 401) leaves the question entirely open whether or no ' we have any direct knowledge of thought as such.' The problem, once more, is the possibility of introspection, and it is defective analysis on this point which has led James into error. He saw that psychical processes, or at least processes of cognition, are not primarily presentations or 'objects,' 74 PROBLEMS OF THE SELF chap. and he also perceived that the attempt to discover the nature of personal identity implies the considera- tion of the identity of a certain kind of object in the same sense as in any other case of identity. But psychical processes, though not primarily objects but references to an object, may themselves be objects of another psychical act. And if these psychical acts are not the sole constituents of the self, they are at any rate among its most important constituents, and their identity, accordingly, among the most im- portant questions to be discussed. A few words will suffice on the question of activity. According to James, if we ask what activity feels like, we find that it consists in a feeling of some bodily processes taking place, for the most part within the head. ' The self of selves, when carefully examined, is found to consist mainly of the collection of these peculiar motions in the head, or between the head and throat.' This is the spiritual self, according to James, and if his account is true it is hard to see why the consent of mankind assigns the bodily self to the bottom of the scale of values and the spiritual self to the top. James mentions the fact, and does not even protest.^ But the experience of activity should not be confused with these accompanying states of the body. Feelings of effort may be bodily, and so may feelings of strain ; but these are difierent from psychical endeavour. Congenial work gives scope to much endeavour, but not necessarily to appreciable efibrt or appreciable strain. EflFort and strain refer to the overcoming of obstacles, not to the degree of endeavour. I shall discuss this question more fully later. At present it is sufficient to mention it. I have said enough to indicate which bodily pre- sentations are frequently accounted parts of the self, and I agree that, in many instances, it is paradoxical ' Principles of Psychology, vol. i. p. 313. in THE SELF AND THE BODY 75 to deny that they are such parts. At the same time it seems clear that these sensations, the objects of the internal sense, stand on no different footing from other sensations, and, if that be true, they ought not to be considered parts of the self. They have no right to be considered parts or modifications of mind, and that is the crux of the question. They may be subjective, they may be analogous to feeling, they may be always present, their absence may lead to doubt of the continuance of personal identity. All these considerations are irrelevant to the main point at issue. Although this chapter has already extended to too great length, it is necessary to mention two other points. First of all, there is the problem of the sense in which consciousness occupies position. Consciousness is not extended, but most men would consider that one self is in a difiierent place from another, and, to use Locke's illustration, can ' travel in a coach from Oxford to London.' There is nothing unnatural when Macaulay says of Byron that ' he carried his exhausted body and his wounded spirit to the Grecian camp.' I was in Dresden, and am now in Paris. Would it seem equally natural to say, ' The physical body which I call mine, or have reason to believe is closely connected with my self, was in Dresden and is now in Paris ' ? Is it not the plain, unvarnished truth that a given self is in a given place, and was not Mrs. Crisparkle right in asserting that 'a man must be somewhere'? ' To be in a place ' (I quote from Lotze) ' means simply and solely to exert action from that place and to experience the actions or effects that reach that place.' ^ Or, again, ' Wherever there are elements with which the nature of the soul enables and compels us to interact, there it will be present and active ; wherever there is no such summons to action, there ^ Metaphysic, English translation, vol. ii. p. 284. 76 PROBLEMS OF THE SELF chap. it will not be, or will appear not to be,' ^ These state- ments have frequently met with assent, but do not seem true if carefully considered. To occupy position is one thing, to exercise causal influence quite another thing. It is perfectly conceivable that something should have position and yet should exercise no causal influence whatever. It is true that light does work certain physical efi'ects, but no contradiction is involved in supposing it not to do so, and yet in ascribing position to it. Conversely, there is the theoretical possibility that some being, not in space, should exercise causal influence even upon beings who are. Let us suppose that God, at sundry times and in divers manners, has wrought miracles. Does it follow that He is in space ? There is a paradox which results from Lotze's theory. If the theory is true it follows that on parallelistic theories, since the sole giver or receiver of spatial effects is the body, the mind cannot be anywhere. On interactionist theories, on the contrary, since the mind has direct dynamical influence upon the brain if not upon the body at large, it would follow that the mind is extended ! If cognition is regarded as a light which sheds its beams without imparting or receiving effects, then it is true that cognition occupies the position of the object to which it refers, and a similar account might hold good of other experiences. In that case, if a man looked from the sands towards a lighthouse far out to sea, the position of his mind would be the whole stretch from his eyes to the lighthouse. But the fact that in such perception the body is a pivot, that if he turns round he perceives not ocean but sand, would be enough to lead to the reflection that the mind, in a special sense, has its abode in the body. Moreover, certain acts of physical adjustment, which are phenomena of the internal sense, form the chief ^ Metaphysic, English translation, vol. ii. p. 289. THE SELF AND THE BODY 71 data for estimating the spatial position of objects. In short, if position must be assigned to conscious- ness, then the only possible course is to maintain that an experience occupies the position of the whole range of objects to which it refers at any given time. It should never be localised in the brain, for that is only the instrument of consciousness. And if it be a paradox that consciousness may extend from the earth to the stars, because of the power of human vision, the paradox disappears on closer reflection. The range of the other senses is very small, and in touch, the most important of them, extends only to the horizon of the sweep of the arm. The range of endeavour is similarly restricted to the body, so far as action is concerned. And many experiences, on this theory, occupy no position whatever, since they refer to the universal or, perhaps, the imaginary. The complications, in fact, are so great that it is best to avoid them altogether. To ascribe position to consciousness is more metaphorical than useful. In the second place, there is a deeper problem still. It might seem, indeed, as if this whole enquiry had taken a wrong direction. Why distinguish body from mind at all ? May not the living body be the self? This conjecture, under the form of the double- aspect or double-attribute theory, has always been considered, but certain modern writers maintain it in a new form, which does not seem identical with these theories or with materialism. Professor James has written a powerful essay on the question, ' Does Consciousness exist ? ' ^ and he argues that the word consciousness does not stand for an entity but only for a function. ' I could perfectly well define, with- out the notion of consciousness, what the knowing actually and practically amounts to — leading towards, namely, and terminating — in percepts through a series of transitional experiences which the world supplies.' ^ ' Essays m Radical Empiricism, pp. 1-38. ^ Ibid. p. 25. 78 PROBLEMS OF THE SELF ' The I think which Kant said must be able to accompany all my objects, is the " I breathe " which actually does accompany them. There are other internal facts besides breathing (intra - cephalic muscular adjustments, etc., of which I have said a word in my larger Psychology), and these increase the assets of " consciousness," so far as the latter is subject to immediate perception ; but breath, which was ever the original of " spirit," breath moving out- wards, between the glottis and the nostrils, is, I am persuaded, the essence out of which philosophers have constructed the entity known to them as con- sciousness. That entity is fictitious, while thoughts in the concrete are fully real. But thoughts in the concrete are made of the same stuff as things are.' ^ This view has an engaging simplicity, but it achieves this simplicity, at least in the form in which it is stated by Pjofessor James, by ignoring some fundamental distinctions. Consciousness is essentially reference to an object, but it does not follow that anything which leads to or terminates in a percept is eo ipso aware of that percept, or strives towards it, or feels with regard to it. If a derelict leads towards or terminates in a reef, the derelict is not therefore conscious. And if percept, by way of reply to this objection, be distinguished from thing, what is this but the flat denial of the truth of James's contention ? No philosopher is constrained to construct either an ' er),tity ' consciousness or a ' function ' consciousness out of such results. A percept is not an experience but an object of experience, and the world does not supply transitional experiences between percepts, while it does supply transitional percepts. I shall be told, no doubt, that this criticism is the product of philosophical sophistication. In a world of pure experience the distinction between a conscious act and its object does not apply. The reply is obvious ; ' Essays in Madical Empiricism^ p. 37. THE SELF AND THE BODY 79 unless the distinction exists there can be no experience at all. To avoid misunderstanding, it is necessary to state the precise conclusion of this chapter. The first and aU-important consideration is that it is the essence of the self to be conscious, and consciousness is not an object, however localised and individualised, but an act of reference to an object, and the aim of the chapter was merely to follow this principle to its logical conclusions. Nearly any one would admit that the body is not the self, if he mean the body as it is known by the external senses, but it is otherwise if he includes the objects of the internal sense in its connotation. He certainly ought to do so, and I submit that, if he desires consistency, there is only one conclusion open to him. All that can be done by way of proof is to analyse the question carefully and in detail, and the distinctions, at all events, are worth drawing, whether or no the conclusion meets with dissent. There is no question of robbing ourselves, by psychological analysis, of something that clearly belongs to us of right. Our bodies belong to us, but they are not our actual selves. A man may rightly call his appearance his own, since it belongs to himself exclusively, even if it be coveted by others. But although it is his exclusive possession, it is not, therefore, a part of himself, and that is the respect in which it differs most profoundly from experiences. And there is no question of the importance, or even the necessity, of the body to the self. It may be that a self must be embodied, and embodied in a distinct particular organism. That important tenet is quite consistent with the argument of this chapter. We may even agree with Professor James that ' the body is the storm centre, the origin of co-ordinates, the constant place of stress in all the experience train. Everything circles round it, and is felt from 80 PROBLEMS OF THE SELF its point of view.' ^ Tiie body is a constant centre of action, it is the constant focus of perception, and organic sensations are the only ones constantly present to us. It is the cause of the subjectivity of our presentations, and thus goes far to differentiate personalities. These considerations are enough to explain the confusion between the self and the body, especially on the intricate points which have been reviewed in this chapter. But to be a cause of sub- jectivity is not to be a part of the self. 1 Essays in Radical Empiricism, p. 170 n. CHAPTER IV THE SELF AS PEELING In this and the four succeeding chapters I shall deal with a set of questions which are somewhat different from those already discussed. The object of the pre- ceding chapters was principally to clear the ground, and although the part of the general enquiry into which we are entering is, in a way, the amplification and the test of what has previously been said, it also involves questions which have hitherto escaped notice. I have said enough at the close of the second chapter to indicate the general plan and purpose of this part of the subject, and to show why it is necessary to consider in detail the numerous and somewhat perplexing varguments which attempt to prove that the citadel of selfhood consists of some one particular class of experiences. The self, it may be said, is primarily a knower, and all other elements in its composition, if there are any such, are depend- ent upon its acts of knowledge. According to others, still more numerous, the essence of selfhood is its activity, and the proof of the primacy of will is at the same time an adequate account of the structure of the self. According to a third party the self radiates out from feeling, clings to feeling, ' exists because of its connection with a fundamental basis of feeling. \. There seems to be a consensus of opinion in favour 81 G 82 PROBLEMS OF THE SELF of one or other of the two latter views, or, possibly, in favour of the two conjointly. Take, for instance, the following passage from Professor Lipps : ' As ex- perience increases,' he says, ' we separate, by gradual stages, what was originally an undivided unity, the content of the world and that of our personality or, more shortly, the world and the self. In this process the contents of direct self-feeling show themselves, more and more prominently, as the , specific basis, or the specific kernel, of personality, or the self, and these are the feelings of pleasure or displeasure, of striving or resisting which accompany all other modes of psychical life. Rightly so ; for in them is found the direct meeting-point of consciousness with the relations of psychical contents to one another and to the whole life of the soul, with the particular kind and the whole economy of these relations and, more generally, with . our own private being, and not merely our peripheral existence. It is to the con- tents of this self-feeling that the consciousness of constraint belongs, a consciousness which judgments, even mere judgments of perception, carry with them.' ^ For purposes of clearness it is advisable to separate the question of feeling from that of striving, and I shall begin with feeling. I have already explained the sense in which I propose to use the term feeling, a sense which seems, in different ways, both narrower and broader than that of Professor Lipps. It is narrower because it excludes striving and resistance. It is broader because it includes other feeling elements than those of pleasure and displeasure. Pleasure and displeasure, however, are regarded, not infrequently, as common constituents of all feeling, rather than as the whole stuff of feehng ; and no exception need be taken to this view provided that there is sufficient recognition of those neutral states which are neither predomin- ^ Grwndtatsaehen des Seelenlebens, pp. 408-409. IV THE SELF AS FEELING 83 antly pleasant nor predominantly painful. Any kind of excitement, any kii^d of emotion so far as that is psychical, in a word the whole aiFective side of consciousness, is best described as feeling. Although this meaning of the term is in conformity with the best psychological usage, it is hardly necessary to mention the serious ambiguity of the term 'feeling.' Tactual sensations of hardness, smoothness, and the like may be described as feeUng, and even this sense of the word is not entirely irrelevant to the present question, since it is connected with some forms of the feeling of constraint. A more important ambiguity is that between bodily and psychical feeling. Even if the argument of the preceding chapter be accepted in the main, a lurking suspicion may still persist in the mind of the reader that the vital feelings are the nucleus of personality. As Jodl says, 'Respiration, alimentation and sex, as well as mobility, refer not merely to complexes of sensations, but are the most primitive, the oldest, the specifically original needs not only of mankind, but also of the organic world in general,' ^ These feelings are the origin of the idea of self, and of its distinction from soulless things. Nor does their importance cease when the earliest stages of development have passed. To feel fresh or tired, to feel well or ill, these surely are primarily feelings of the body, but they, and the moods which they induce, are, as it were, the tonicity of the self, its vital balance. Together they form that general feeling of bodily tone which we call coenaesthesia, a general mass of feeling which, because of its diflfused generality, is at once fundamental to selfhood, and difficult to analyse precisely. There are two moments in this argument, the first being a proof from origin, the second an appeal to introspection, and the second supplements the first by showing that the nucleus of development persists, • Lehrbuch der Psyehologie, Bd. II. S. 29. 84 PROBLEMS OF THE SELF chap. and that the self, as it rises in the scale of develop- ment, does not climb over its own corpse. But the second moment of the argument has been met in the preceding chapter, and we must abide by our decision. The importance of these bodily feelings to the sense of continuous personality is not in dispute, and will be more fully illustrated in the sequel. But so far from being the nucleus of selfhood, they are not even parts of the self, and consequently the argument from origin is the only new one which enters in this connection. The question of the origin of the idea of self, and of its distinction from that which is not a self, is a very intricate one, and is, fortunately, not directly relevant to our present enquiry. While it is necessary to consider the development of the self, it need not be necessary to consider the development of the idea of the self. The process by which men come to under- stand the meaning of what actually exists, the long series of instructive errors which beset them before they come to adequate comprehension, is fascinating enough, but is only relevant by way of illustration. That the first stage of this development treats the self as the body, and the second as some part of the body ; that the soul comes gradually to be regarded as a film, or a vapour, or a shade, and that these unsub- stantial beings must drink blood in order to be really alive — these and similar beliefs are matter of history. ■And the gradual process by which these different stages are reached is full of interest, though accounts of it are speculative. But it is hardly necessary to point out nowadays that an argument from origin is, strictly speaking, only an argument about origin. It traces the path of development, and development, doubtless, requires some identity and some continuity. But no one can specify in advance the degree of identity and continuity, and it would be as logical to maintain that a chicken is the yolk of an egg because THE SELF AS FEELING 85 it develops from that nucleus, as that the self is feeling because it develops from feeling. Indeed, as I have already said by implication, the inference in this instance would be still more precarious. If the primitive self is feeling, then feeling may constitute a self, and the further question is only that of the precise continuity of the process of development. But all is altered when the question is that of the development of our knowledge of what the self is, and it is surely not absurd to maintain that, blindly groping, we seize on some elements connected with selfhood, and liable to be confused with it, instead of, from the first, discovering and clearly understanding the self as it really is. The argument from origin, then, is very likely to prove a mere ignoratio elenchi, and there is no reason to suppose it otherwise in the present instance. In the previous chapter I have explained the important analogies between bodily feeling and psychical feeling, and it is hardly surprising that these analogies should add considerable strength to this particular argument from origin. Only psychical feeling, however, is really relevant to the issue. It may be that bodily feelings are the chief conditions of psychical mood and tem- perament, and that the latter are the core of the self. That question will be discussed in due time. For the moment I desire to point out that the argument from origin may derive seeming plausibility from its use of that perplexing word * feeling ' in yet another sense. Eeaders of Mr. Bradley know that when he speaks of feeling he means vague, indeterminate, marginal experience, too indefinite to be known or analysed. To the same purpose Croce argues that philosophers have no right to regard feeling as a distinctive part of spirit. ' Feeling, in fact, has been the indeter- minate in the history of philosophy, or rather the not yet fully determined, the half-determined.' ^ As he ' Philosophy of the Practical, English translation, p. 25. 86 PROBLEMS OF THE SELF explains, with reference to one of the psychological uses of feeling, ' It has happened that, with various times and authors, all the most rudimentary, tenuous and evanescent manifestations of the spirit have been called "feelings," slight intuitions (or sensations as they are called) not yet transformed into perceptions, slight perceptions, slight tendencies and appetites, in fact all that forms, as it were, the base of the life of spirit.' ^ This sense of feeling clearly supports the argument from origin, since development must pro- ceed from the indeterminate to the determinate. But the support is ambiguous. To argue that the self is feeling in the determinate sense, because it arises out of feeling in the indeterminate sense, is an obvious instance of a purely verbal fallacy. There are none, I suppose, who would draw this inference explicitly, but the nerve of the inference has not been without influence. It is plain, of course, that indeterminate experiences are either indeter- minate feelings, endeavours and cognitions, or else that they are too indeterminate to permit any such distinction to be drawn. If the former, there is no reason to hold that the developed self is more dis- tinctively a feeling entity than a willing or a knowing entity. If the latter, the development to the self proceeds per saltum without any marked identity or continuity in its stages. That is possible ; but it does not help the argument from origin, since it also permits of the interpretation that the self, and not merely the idea of self, may have developed from something which is not a self at all. Indeed, there is only one way in which this argument from the indeterminate supports the belief in the primacy of feeling. If it be true that indeterminate or marginal experiences can be seen by introspection to be qualitatively similar to developed feeling, in a sense in which they are not similar to developed conation ' philosophy of the Practical, English translation, pp. 22-23, THE SELF AS FEELING 87 or cognition, then, at least, the continuity between the original nucleus and feeling is noteworthy ; and that, I take it, is Mr. Bradley's opinion. He speaks constantly of the ' felt mass, akin to pleasure and pain.' But must this analysis be true ? It is plain that aesthetic feelings, at any rate, are as definite, as developed and as specifically directed as the cognitions or the endeavours correlated with them ; and the same is true of love, hate, sympathy and other feelings. Our strongest feelings, perhaps, are connected with organic sensations, but the feelings are not more indeterminate than these sensations themselves. Perhaps, however, it is hopeless to look for agreement when a term may be used in so many distinct senses. All that I can say is that I cannot find any introspective evidence of analogy in this respect. When it is maintained that psychical feeling is the core of selfhood, the burden of the argument is usually that feeling either is, or is the index to, the only kind of experience which characterises the private and peculiar domain of self. Most of us would admit that our feelings are the most intimately personal of our experiences. We might assent to the theory that our acts of knowledge are but instances of the universal mind thinking within us ; we might be driven to the view, albeit reluctantly, that our striving is but the activity of a cosmic force that envelops and impels us, but we should have much greater difficulty in admitting that our feelings are not our very own. The Cosmos may think for us, and act through us, but we feel our own feelings in its despite. This argument purports to be founded upon a direct analysis of the nature of feeling. Indeed, it implies the theory that feeling is ' subjectively subjective,' to use Hamilton's terms, and it is necessary to consider this analysis with greater care than the earlier sketch permitted. Hamilton's 88 PEOBLEMS OF THE SELF phrase is lucid, if unattractive, and his discussion of the question is the more pertinent as his general view of the nature of the tripartite division of mind agrees broadly with that of this essay, and his use of the word feeling, although he restricts it to pleasure and pain, comes much closer to ours than some other uses which have been mentioned in this chapter. Hamilton believes that feeling plays a part in the life of the self which is intermediate between cognition and endeavour. ' The mere cognition leaves us cold and unexcited ; the awakened feeling infuses warmth and life into us and our action ; it supplies action with an interest, and, without an interest, there is for us no voluntary action possible. Without the intervention of feeling the cognition stands divorced from the conation, and, apart from feeling, all conscious' endeavour after anything would be altogether incom- prehensible.' ^ But Hamilton does not believe that feeling is a kind of reference to an object. ' In the phenomena of Feeling — the phenomena of Pleasure and Pain, — on the contrary, consciousness does not place the mental modification or state [i.e. the object] before itself; it does not contemplate it apart — as separate from itself — but is, as it were, fused into one. The peculiarity of Feeling, therefore, is that there is nothing but what is subjectively subjective ; there is no object different from self, — no objectification of any mode of self. We are, indeed, able to con- stitute our states of pain and pleasure into objects of reflection, but in so far as they are objects of reflection, they are not feelings, but only reflex cognitions of feelings.' ^ It must be admitted that the essential reference to an object is less obvious in the case of feeling than in that of cognition or endeavour. That cognition ' Lectures on Metaphysics, vol. ii. p. 426. The passage in question is quoted from Biunde, but states Hamilton's own view, - lUd. p. i32. IV THE SELF AS FEELING 89 involves reference to an object goes without saying, and it is equally clear that striving involves an attitude towards an object. Feeling seems on a different plane. It is not an attitude towards an object but the way in which the subject is affected by the object. Now if the meaning of this last state- ment were only that the object causes feeling, then it would in the first place be false, and in the second would not prove any conscious reference to an object. We may and do experience both grief and joy at the fate of fictitious characters in a novel, and. these, in any ordinary sense of the word causation, could not cause anything. Similarly, some totally irrelevant circumstance may be the cause of very real feelings. A teacher may be annoyed with his pupils because he has missed his breakfast, but he is, none the less, really annoyed. To say, then, that the attitude of feeling is the way in which the subject is affected by an object does not mean that the object causes this feeling but refers to the way in which the subject is consciously affected. It is the receptive attitude of the subject in respect to the object, and therefore would seem to imply reference to the object. Another analysis, however, is possible. To say that the subject is con- sciously affected in feeling is to say that feeling is the attitude in which the subject is affected when and because. he is conscious of an object, and this statement is perfectly compatible with the view that there is direct reference to an object only in cognition and endeavour. To decide between these views on introspective grounds is exceedingly difficult, but there are no other grounds on which the question can be decided. Both theories can account equally well for the part which feeling plays in psychical life. Indeed it might be possible to maintain that feeling, cognition and endeavour are not really distinguishable elements in the life of the self, but only distinguish- 90 PROBLEMS OF THE SELF able aspects of a single psychical process. But this theory is only tenable if it is possible for one ' aspect ' to predominate, in certain cases and not in others, over another ' aspect.' If, however, the ' three aspect ' theory is possible, there is no reason whatever why every aspect should refer directly to an object, even granting that they are all concerned with an object. I do not say, then, that the interpretation of feeling as a reference to an object can be conclusively proved, but there are reasons for deciding in favour of that view. There are many instances in which it is the best explanation, and none in which there is any good reason to deny its possibility. Pleasure and pain seem at first sight to support Hamilton's analysis. In examining them we frequently fail to find a reference to an object, in any sense which could not be explained on this theory. But there are other instances in which his analysis is, to say the least, forced and unnatural. I have already mentioned the case of aesthetic feelings, and I may mention other instances. If I am pleased with Tony Weller and bored with Mark Tapley, if I love Colonel Newcome and hate his nephew Barnes, I cannot help thinking that my feelings have a direct reference to these gentlemen. There are other cases, certainly, in which this reference is less marked, but rlone in which it may not be supposed to hold, and in the instances qf love, and hate, and desire and many other forms of excitement, this reference is so marked as to be compelling. Yet these are all attitudes of feeling, desire no less than the rest of them. Conclusive proof may be impossible, but the balance of evidence is against Hamilton, not for him. And there are some who not only disagree with Hamilton but find the contradictory of his view to be self-evident. ' Pleasure,' says Husserl,^ ' is unthinkable without an ^ LogiscJie Vntersuchungen, Bd. II. S. 368. Cf. the whole passage. THE SELF AS FEELING 91 object in which pleasure is taken . . . and this because the specific nature of pleasure involves the relation to the pleasant object. In the same way the experience of conviction is impossible unless there is something of which we are convinced.' If, then, feeling be a reference to an object, as other experiences are, it is hard to suppose that it is a private possession in any sense in which they are not. And this , conclusion, so far from being para- doxical, is the plain verdict of ordinary reflection. Our acts of knowledge and our acts of will are our own. No one can resolve for us, no one can supply us withiunderstandings. Two men may think of the same thing, but they are not, therefore, identical beings even in respect to that thought. The objects may be the same, but the acts of cognitive reference to them are not the same. I do not see that it is necessary to labour this point. It could not be dis- puted except by overthrowing the whole analysis of experiences on which this essay proceeds. The more we reflect, the more we come to see the co-ordinate necessity of all three elements of mind. No doubt the fact that all are necessary does not in itself prove that all are equally important, and therefore it is necessary to examine the claims of each of them for predominance. But, so far as the argument has gone, there is no ground for assigning this predomin- ance to feeling, even in respect of its private and particular character. Feeling may be the usual index of such particularity in the self, but it is not, intrinsically, more obviously ' ours ' than other ex- periences. Are not a man's acts of memory, of association, of judgment, as characteristic of him as his feelings ? Is not the difference in this respect a fair criterion of the differences of personality ? And does not the same account hold good of his efforts, and aims, and endeavours ? It is probable that the sense of 'feeling' which 92 PROBLEMS OF THE SELF identifies it with organic sensations (including kinaesthetic sensations) is chiefly responsible for this argument as well as for its predecessor. Husserl, for instance, who denies, as we have seen, that psychical feeling is subjectively subjective, is prepared to admit, in the same passage, that physical feelings probably are. And there is no need to dispute such a theory. Organic sensations are peculiar to the individual subject, and it would be misspent subtlety to argue the point further. Indeed it is clear from the argu- ment of the previous chapter that organic sensations are specially apt to be confused with the self because they are more distinctively subjective than any other sensations. Let it be so. The conclusion is irrelevant to the present question. The doctrine that the self should be defined by interest is a variant of the same theory, and requires fuller mention than it has received hitherto in this discussion. It is, of course, totally incapable of affording an exact definition of selfhood. The self does not consist merely of interest, however widely that term be employed. There are characteristic ex- periences which are not interest at all, and if the self be something over and above its experiences, then it is also above and beyond that particular class of experiences which we call interest. Similarly many things in which we are interested are by no means parts of ourselves. If Torquemada was interested in his victims, his interest would have been rather difi"erent had they been parts of himself. It may be argued, however, that if we speak of an object as ours we mean, unless we are lawyers, that it interests us. One man directs his attention to poetry, another to philosophy, a third to engineering. Each makes these objects his own, and the reason is that these different things appeal to him in a special sense. His individual preferences are the root of his selection, and the nature of the self is best seen in the things THE SELF AS FEELING 93 it selects. Because the self is revealed in its selection, its most distinctive element is that which determines this selection. The contention may be true enough, but it fails to prove the point in dispute. Feeling is the way in which the subject is affected by the objects presented to htm, and this, in its turn, determines his reactions towards them. But the reactions themselves are just as characteristic of the self as its feelings, and neither is more characteristic than the cognition which guides and informs them both. King Lear loved Goneril and Eegan when he divided his kingdom between them, and he was offended with Cordelia. His feelings and his actions were both characteristic of the man, but not less characteristic was his lack of foresight and wisdom. He was old and venerable, but very far from wise. When he knew more he loved Cordelia only. Feeling is not the only relevant feature of the case, it is symptomatic of the degree of a man's knowledge of the nature of the object which excites it. Feeling and interest may, and do, indicate a special relation to the self, 'but any other of the principal classes of experiences might serve this purpose equally well. Finally, in this connection, we may mention the argument that we find ourselves as we really are in those more prolonged trains of feeling which are called moods. There are certain moods in which we sink within ourselves to the neglect of the outside world. We are aware of the voice of the singer, the strains of the accompaniment, the hall, the audience. But these things seem to drag us from ourselves. They are less to us than the private internal mood of mere enjoyment, and that is where we should look for the essence of the self. When we are released from the tyranny of the world, and from the efforts and attention it requires of us, we become ourselves, and it is clear that in these cases the mood itself 94 PROBLEMS OF THE SELF chap. determines the chain of thoughts, associations and desires, even if these moods are more than half compact of bodily feeling. So runs the argument ; and- 1 do not wish to renew the discussion whether feeling, or trains of feeling, are, in an unusual degree, private and particular. I have tried to show reasons to the contrary, but let us grant their insufficiency. Even if feelings are distinctively private and particular, it does not follow that they are the most fundamental parts of the self. That would only be the right conclusion if the self is real precisely in proportion as it can detach itself from all else and creep within its own shell, and this is not the true account of human personality. Are the men whose lives radiate out towards other things and other persons less really selves than those who try to shrink into some un- approachable crevice of private being ? Surely the facts are otherwise, To understand the self it is best to go outside it and consider its influence and the range of things which it contemplates. To argue that these ought not to count, or ought not to count very much, is like arguing that the most essential parts of a lighthouse are those which are not constructed for the diffusion of light. It remains to consider the most important argu- ment of all, and that is found in Mr. Bradley's illuminating discussion of the meanings and the reality of self.^ No two chapters have done more, or are, likely to do more, towards unveiling the obscurity of the subject than these. Their interest and their value have compelled attention to the problem, and neither the interest nor the value is lessened by the fact that the theory does not commend itself to every one. It is therefore in no spirit of mere con- tentiousness that I should like to give some reasons for dissenting from Mr. Bradley's conclusions. The principal merit of Mr. Bradley's analysis is ' Appewrance and Reality, Bk. I. chapters ix, and x. THE SELF AS FEELING 95 that he insists on a precise answer to a precise question. ' The fact of one's own existence, in some sense, is quite beyond doubt.' ^ But in what sense precisely ? And he distinguishes a number of mean- ings, most of which are inadequate. The self cannot be the body, in any ordinary sense of these words; and its essence can never be discovered by the mere consideration of ' the total filling of the man's soul at this or that moment.' This, of course, must be admitted, but such inspection need not be so valueless as Mr. Bradley supposes. For in this in- spection we may distinguish psychical acts of reference from the ob_[ects to which they refer, and may restrict our further enquiry to these acts alone. Such a mode of procedure, however, would be impossible for Mr. Bradley. He would refuse to draw the distinction in this sense, or to regard the acts as mental and the objects as non-mental. It is self-evident to him that nothing can exist except Experience, and he interprets this as meaning that act and object cannot be con- sidered distinct existences, but only moments in a whole. But although Mr. Bradley rejects such an analysis, he eventually arrives at something similar. His reasons for doing so, however, depend upon the rejection of certain other possibilities, and it is necessary to consider his reasons for that rejection. Since the self is a continuous unity, he argues that it is impossible to discover its essence by analysis of any given state, and that there is no constant average mass of experience which deserves to be called the essence of the self because it alone is present in its totality at any given moment of the existence of that self. This type of problem will occupy us later. At the moment it is enough to assent to Mr, Bradley's argument that any such constant average remaining identical 'from the cradle to the coffin' would be ' Appearance, and Reality, p. 76. 96 PEOBLEMS OF THE SELF chap. such a pitiful residuum that no one could dare to call it a self. And in the same way there is no inner nucleus of coenaesthesia, or the like, which is more than a wretched fraction or a poor atom. Nor does the supposition of a monadic pure ego appreciably alter the case. ' If the monad stands aloof, either with no character at all or a private character apart, then it may be a fine thing in itself, but it is mere mockery to call it the self of a man.' ^ Nor should, the self be defined as that in which interest is felt. Mr. Bradley argues that if these theories are rejected, the problem must be regarded in a difi'erent light, and the proceeding which promises best is the distinction between subject and object, between self and not-self. This relation is either theoretical or practical : theoretical in the case of perception or intelligence, prac.tical in that of desire or will. Mr. Bradley, then, after a process of elimination, seems driven to attack the problem on the lines we adopted in our second chapter, and the chief difference, it might seem, is that he has refused to include feeling, in any sense, among the experiences which refer to an object. We shall find, however, that his analysis differs radically from ours, and that he maintains that feeling, in his sense, not in ours, is the sole constituent of the subject side of the subject-object relation. In any given case, Mr. Bradley proceeds, both of the terms which enter into the subject-object relation have definite contents. But when we consider either term we cannot say that it is such that it must in- variably be subject or invariably object. Thus we have reason to believe that the subject -term can always be made an object, or, at least, we cannot specify the exceptions and we know them to be few. The most intimate features of the subject-term at any given moment may be set before the self at any other ' Appearance and Reality, p. 87. THE SELF AS FEELING 97 moment, may be considered as object, and so become part of the not -self at that moment. The same account holds of desire and volition. 'As intro- spection discloses this or that feature in ourselves, can we not wish that it were otherwise ? May not everything that we find within us be felt as a limit and as a not -self, against which we either do, or conceivably might, react ? ' ^ And a similar argument holds of the object part of the relation. There is nothing in the object which cannot become part of the subject, or if there is anything, it is so trivial and unimportant that it may, for practical purposes, be neglected. When I direct my attention to some objept I do not straightway pass out of all relation to the object I attended to a moment before. What happens is that this object ceases to be an object and becomes part of the self It passes into a general background of feeUng from which it was detached as an explicit object, and to which it returns. ' And the fact of the matter seems this. The whole psy- chical mass, which fills the soul at any moment, is the self so far as this mass is only felt. So far, that is, as the mass is given together in one whole, and not divisible from the group which is especially connected with pleasure and pain, this entire whole is felt as self. But, on the other side, elements of content are distinguished from the mass, which therefore is, so far, the background against which perception takes place. But this relation of not-self to self does not destroy the old entire self This is still the whole mass inside which the distinction and the relation falls.' ^ It is clear that the meaning which Mr. Bradley attaches both to subject-term and to object-term is so fundamentally distinct from ours that a whole treatise on epistemology would be required to make thorough discussion possible. To speak of the object of know- ' Appearance and Reality, p. 91. ^ Jbid. p. 95. H 98 PEOBLEMS OF THE SELF chap. ledge or will as a content which is now rendered explicit and opposed to the self, and anon becomes implicit and part of a felt mass akin to pleasure and pain, is to speak an entirely diflferent language from that which has hitherto been employed in this essay. Any criticism, therefore, is bound to appear inept, and yet Mr. Bradley's theory is much too important to be passed over. But is it not clear that we know the self too intimately for Mr. Bradley's view to be true ? Let us consider our knowledge or our volitions at any particular moment. Is it an adequate analysis of either of these processes to say that a felt mass, akin to pleasure and pain, is somehow confronted with, or opposed to, an object ? How can we say that feeling knows or that feeling wills, and how can we avoid saying that the self knows and that the self, wills ? It is true that Mr. Bradley, to be consistent, is pre- cluded from telling us what feeling means. For if he could regard it as an object, he would drag it from its background, and the shock would be fatal. All we can know about it is that it is, and that it is indissolubly connected with pleasure and pain. Surely much of the task of psychology consists in analysing the different characteristics of knowing, willing and the like, and this from the point of view of the subject, not of the object. We know more than a background of feeling. We know psychology. , To such an argument Mr. Bradley has a very obvious reply. ' These arguments,' he might say, ' only refute mine by ignoring my meaning. We know too much about the self to say that it is but a mass of feeling at any given moment. How ? By intro- spective analysis, and by that only. But my argument was that introspection is a kind of knowledge or perception. It is therefore aware of an object or not- self, and from this doom it cannot escape. Hence the objection is futile. The self is never an object.' IV THE SELF AS PEELING 99 We have already answered this by implication. Acts of knowledge or acts of will are not primarily objects. They are primarily references to an object. To use Professor Alexander's terminology, they are primarily enjoyed and not contemplated. It is false, however, to maintain that they can never be contem- plated.^ Introspection is just this contemplation, and, as we have already argued, there is no good reason for denying that introspection may succeed, i.e. that we may discover by introspection what experiences are in themselves. Certainly we may be mistaken about our experiences. Cognition is no more infallible in this respect than in any other. Everything is what it is, and is not some other thing. Similarly, experiences are what they are, and are nothing else. But we may fail to understand what experiences are, just as we may be unable to understand what other things are. Cognition of an experience is as liable to error as any other cognition. The whole question at issue is whether the object of cognition must be only an appearance. That is part of Mr. Bradley's meta- physics, and his metaphysical system may be true for all we have shown to the contrary. But if it is permissible to neglect these ultimate points of epistemology (and they do not enter into the dis- cussion at the level at which Mr. Bradley conducts it in this place), we are entitled to maintain that when we consider the self as an object we are really considering it, and that acts of cognitioli when intro- spectively examined have distinct features and are not sunk in a mass of feeling, much less constituted by that mass. Indeed, it may very well be maintained that the introspective awareness of experiences is less likely to deal with mere appearance than, let us say, sense-perception is. Why should not these experiences be real parts of mind as it is in itself? The question is one of introspection, and, in the ' As Professor Alexander does. 100 PROBLEMS OF THE SELF chap.iv last resort, of introspection only. It is agreed that there is always some content of mind other than the objects before it at any given time, but Mr. Bradley maintains that this content cannot be explicitly contemplated, or coherently described and analysed. It is true that an experience in vacuo is unthinkable. Every experience is united to its specific object, and would not be an experience unless it were So united ; but that does not afi'ect the question of the nature of the experience when it is united to its object. Our argument has been that the contents of mind can be discovered by introspection, and can then be seen to be determinate. In that case the self cannot be mere feeling in Mr. Bradley's sense ; and, plainly, it is not mere feeling in any other sense. CHAPTER V THE SELF AS WILL The arguments in favour of the primacy of will are so varied and important that they require what may seem a disproportionate amount of space to be adequately treated. Indeed, they might very well have a whole volume devoted to them. In default of this, it will conduce to clearness if the subject is subdivided, and I propose to devote three separate chapters to its discussion. The first of these will be psychological in the main. The question, ' If the self is essentially will, what are the experiences in- volved ? ' seems to me a prerequisite of any further enquiry. It will form the subject of this chapter, together with questions of interpretation which arise immediately and directly from it. In the second of these chapters I shall consider the various meanings of purpose. Perhaps the self should be defined as purposive, and, in any case, it will be found that purpose and its implications are the chief warrant for maintaining, rightly or wrongly, that the self is will. The third chapter will deal with what, for want of a better name, may be called 'The Primacy of the Practical Reason.' It will refer to certain meta- physical arguments which have been prominent in the history of philosophy. To assert that the will is the essence of the self is as familiar to Common Sense as it is to philosophers. 101 102 PROBLEMS OF THE SEL-F ohap. It is enough to give an illustration from that kindliest of works, Sir Walter Scott's Journal : ' A touch of the morbus eruditorum. . . . Fighting with this fiend is not always the best way to conquer him. I have always found exercise and the open air better than reasoning. But such weather as is now without doors does not encourage la petite guerre, so we must give him battle in form, by letting both mind and body know that, supposing one the House of Commons and the other the House of Peers, my will is sovereign over both.' •" Truly, Sir Walter was an example of the predominance of will, if ever there was one, and not the less so at the period when he wrote these words. For it was then that he faced ruin boldly, and gave her the noblest of battles. It is natural to believe that the will is sovereign. It is not unusual to deny that its monarchy is limited. A word of explanation is necessary with regard to terminology. Hitherto I have used the word ' endeavour ' or, more rarely, ' conation ' to express the genus of psychical facts which fall for dis- cussion here, and these, on the whole, are the best expressions for describing the active side of human consciousness. ' Will,' strictly speaking, has a narrower denotation. It should mean volition or voluntary action. It implies conscious consent or resolve, and not only are many human actions in- voluntary or non-voluntary, but much seeking and striving lacks the specifically volitional element of resolve. This strict usage is, without a doubt, the correct one, but there is a broader usage in which the word ' will ' is used to include, not only volition, but also any sort of endeavour or conation. The ambiguity is not usually very serious, and it is so firmly rooted that it requires at least lip service. I used the word ' will ' in the broadest sense when I selected the title of this chapter, but I intend to give ' Pp. 45, i6. V THE SELF AS WILL 103 it that broad signification as seldom as possible, and, in the sequel, to distinguish it carefully from the other ' active ' experiences which are relevant to this discussion. The concept of mental activity has probably excited more heated discussion among modern psycho- logists than any other. To some it is a ' scandal,' and to others merely an obvious fact. But even the acerbity of the discussion has some redeeming features : it has forced the issue into relief, and has begotten much searching and valuable analysis. ' Perhaps the most elaborate work ever done in descriptive psycho- logy has been the analysis by various recent writers of the more complex activity situations.' ^ Accord- ingly it is difficult, if not impossible, to say anything both new and important on the question, and I cannot hope, nor do I wish, to do more than cover the ground already traversed by others. But the question must be discussed in any serious attempt to grapple with the problems of the present essay or the present chapter. And therefore it must be faced now. There are three questions which excel all others in importance, from this point of view. The first is, ' In what respects are certain experiences called active, and what are these experiences ? ' In other words we must try to discover and to Analyse the active elements of consciousness. And this might seem the sole aim of the present section. But a little reflection will show that two other questions are forced upon us. It is necessary to ask not only what experiences are called active but whether they are rightly so called. Do we mean anything by calling them active and, if we do, can our meaning be in- telligently expressed ? It is here that the ' scandal,' if there is one, shows its ugly head. No one can reasonably object to analysis, but the explanation of 1 James, Essays on Radical Empiricism, p. 163. The reader will fre- quently perceive my debt to this essay on the Experience of Activity. 104 PROBLEMS OF THE SELF chap. the analysis is quite a different matter, and the critics cry out in horror, not because there has been analysis, but because that analysis, though susceptible of divers interpretations, is straightway regarded as an oracle. We must, therefore, define our position in this respect also. And in the third place we must en- deavour to discriminate as accurately as possible the various subdivisions of the genus ' active experiences.' For when it is said that the self is primarily will, the word ' will ' is often used ambiguously, and it is im- possible to conduct any argument until the meaning of the terms employed is made explicit. (1) The plain man would say that wherever any- thing happens, wherever change takes place, wherever there is anything doing, any life or any process, there activity in some sense is implied. But in saying so he certainly would not mean that every process and every change, just because it is a process or change, is necessarily active.* He would recognise that some processes are active and others passive, and con- sequently his assertion would only mean that the existence of any process or change indicates the presence of something active somewhere. A man is drugged and murdered. He is not active, but the opiate and the murderer are. We can infer the presence of some activity somewhere from the events which happen in the most passive subject. Now when activity is regarded as an aspect of, or element in, experience, it is clear that there is re- ference to those facts in which activity, in the plain man's sense, is somehow involved. There may be disputes concerning the ultimate meaning of the term, but the fact of the presence of change and process in psychical life is beyond all cavil. Every act of consciousness is a process, and the totality of these processes, as seen in the life of the self, flows cease- lessly through time. More than that : this change is experienced, and is not merely an inference. We THE SELF AS WILL 105 may infer change from the knowledge that one and the same thing has occupied different positions or possessed different qualities at different times. But we may also perceive a body moving, or changing, and such changes are directly perceived, not merely inferred. There is a clear difference in the two cases, and it is only because change can be perceived in some instances that it can be inferred in others. Similarly we may merely infer a change in our psychical lives. A lover, meeting his mistress after an interval of time, may be surprised that she no longer attracts him, although she is, to all appearance, the same as she was, and so he may infer that he himself has changed. But he may also perceive conscious process going on, he may directly experience change in him- self, and therefore change, directly perceived, is an important and distinctive feature of mind. This, however, is a very small step in analysis. The perception of mental process indicates that there is activity somewhere, but the activity, so far as the analysis goes, need not even be mental. The ex- periences in question might merely be passive, as many experiences are. It is necessary, therefore, to distinguish those perceived psychical processes which are active, from others which are passive or, like cognition, are neither active nor passive. Activity experiences are only one class of experiences. That is clear to introspection, and no other evidence is relevant. Some experiences feel active and others passive, whether or not it is possible to define the meaning of activity or passivity precisely. Let us ask, then, whether it is possible to give an intelligible account of the experiences which we call active. The term activity may be a misnomer, but the experiences are real and can, perhaps, be analysed. There are certain concepts which may be employed for the description of activity experiences, and the chief of these are direction, initiation and immanent 106 PROBLEMS OF THE SELF chap. causation or, as Mr. Bradley puts it, self-caused change. It is plain that these notions are, at least, relevant. Without the concept of direction activity would be meaningless. When we seek anything, strive after it, aim at or intend it, the very being of these pro- cesses is to be in some direction. This is more than the fact that all conation must be guided by some cognition, that ignoti nulla cupido, that there is always some awareness of the end. It means that the being of conation is to seek — something. It is, being directed. This, then, is a fundamental characteristic of activity experiences, but it is hardly sufl&cient to distinguish them. After all, every re- ference to an object is directed towards that object. It is the manner of the direction that counts. And the manner of the direction in the present case is just to seek or strive. In particular, it is not enough to say that the differentia of conation consists in the fact that ' the stream of consciousness feels its own current,' ^ that it ' is, eo ipso, experienced transition,' ^ or the like. To feel oneself moving, even in a certain direction, is not to feel active, and is not enough for activity. When the crew of a sinking ship experience a transition into the waters, they are not experiencing an active transition, and although the analogy of the stream may be useful, it also is insufficient. A stream has a source and it flows towards the sea, but it does not seek the sea, and it would not seek it even if, per impossibile, it were aware of its own current. It is possible to experience transition in a certain direction without any implication of conscious activity. The obsession of an id^e fixe would tally exactly with the description. And we must be prepared to face the possibility that no description will really suffice, and that seeking is nothing but seeking. But descriptions, ' stout, Analytic Psychology, vol. i. p. 160. » Ibid. p. 159. V THE SELF AS WILL 107 none the less, have their value, and so the quest should not be abandoned at once. Let us turn to the concept of immanent causation, with the proviso already mentioned, i.e. that the con- cept is regarded as nothing but a description of the differentia of activity experiences as they appear to introspection. Activity, perhaps, means self-caused change, and if, to return to the metaphor of the stream, we regard its current as driven by its own momentum, then it may be right to maintain that this sort of experienced transition is mental activity. It is clear, as Professor Stout has argued,^ that mental activity is, in one sense of the words, not merely immanent. Seeking and striving are active experiences, but they may, and usually do, involve a constant reference to external conditions, a constant conflict with obstacles which restrain, and a constant striving for a goal which is sought by the process but need not form any part of it. A man, for instance, may seek distinction among his fellows. It is they, and not he, who can accord this distinction, and their existence, their aims, their very whims and caprices, are the constant conditions of his striving. If im- manent is used as the contrary of transient, and this is how it ought to be used, then it is clear that activity experiences are not really immanent. It is unnecessary to add that activity experiences have little in common with the momentum or the inertia of a particle. The concept of immanence, however, is used in another sense when it is applied to psychical activity. It does not deny the forward or the outward reference of activity experiences. What it does mean to deny is causation a tergo. It is meant to be synonymous with self-initiation, and that is our meaning when, for instance, we speak of the freedom of the will, or distinguish activity from passivity. A process which ' Analytic Psychology, Book II. chap. i. 108 PROBLEMS OF THE SELF chap. initiates, or seems to initiate, will be called active, a process which depends upon the initiation of some- thing else will be called passive. Thus we may resolve to remember something which we have forgotten, and, for that purpose, give rein to the play of association in the fond hope that something will turn up to give us the clue we want. Only the experience which initiates, in this instance the resolve, is an activity experience. The rest depend upon it. This feature, characteristic of activity experiences, has been variously described as spontaneity or inde- termination, and, of course, appears to involve all the difficulties attaching to the freedom of the will, the adequacy of mechanism to account for bodily behaviour, and the like. But these difficulties need not disturb us at present. The question at issue now is merely one of introspection. Do we find this apparent feature of initiation when we introspect, and, if so, is it a characteristic common and peculiar to activity experiences ? The answer to this question is that activity is a characteristic of any experience which is called active. There is no mystery about the matter. We begin processes. We make fresh starts. And even if we repeat what we have done before we need not repeat it in a mechanical fashion. We may begin it again. The process does not roll on pushed by some extraneous impulse. We start it- afresh. We make up our minds, give it onr fiat, say ' let it be done.' Here there is activity and initiation. And such experiences feel quite diff"erent from those which ensue upon them. Initiation, then, is a fact which demands recognition, and is characteristic of some experiences and not of others. ' The pity is that its reality to introspection, and its precise meaning for introspection, have been obscured by the introduction of irrelevant considera- tions. I may refer briefly to two such obscurities. In the first place, the presence or absence of repetition THE SELF AS WILL 109 is irrelevant. Repetition is not necessarily unspiritual. Spirit, like other beings, may never repeat itself exactly, but would it be spirit if it never repeated itself at all ? The fundamental fact is that we may begin again, and that when we begin again we really begin, even when and so far as there is repetition. If a Cabinet Minister resolves to bring in a bill, in precisely the form in which he introduced it before, the fact of repetition need not make any difference to his responsibility or his initiation. In the second place, initiation does not imply a rupture of connection with the past. It is, of course, compatible with the introspective evidence that the real and complete cause of an act of initiation may be the brain or the subconscious. These are not revealed to introspection, and consequently are irrelevant. But the more im- portant point is that initiation is not mere immanence, and does not imply discontinuity with those previous events which are so revealed. Now, whether it be Bestial oblivion, or some craven scruple Of thinking too precisely on the event, A thought which, quarter'd, hath but one part wisdom And ever three parts coward, I do not know Why yet I live to say ' This thing's to do ' ; Sith I have cause, and will, and strength, and means To do't. Examples gross as earth exhort me. Had Hamlet acted according to these exhortations, had his too precise thinking borne fruit in deeds, there is surely no question that his resolve would have been connected with those thoughts and ex- hortations. And there is just as little question that he did not undertake this initiation of action, which is surely a proof that the initiation of action is not identical with the motives towards it which throng into the mind. Initiation is compatible with con- tinuity, and does not arise ex nihilo. It is the moment of birth in the soul, and birth is a fresh start, rooted in the past. no PEOBLEMS OF THE SELF- chap. We may fairly claim, then, that this characteristic of initiation is common to all activity experiences. It may be most obvious in resolve, but is also true of all seeking or striving whether they involve conscious resolve or not. But it is not peculiar to activity ex- periences. It holds of adynamic experiences as well as of dynamic, it is as marked a characteristic of the assent of judgment as of the consent of will, and that is why the two are often confounded, as in Descartes's theory of truth and error. If this initiation be the meaning of freedom, then the freedom of judgment is as important and as unmistakable as the freedom of will, and yet the two are not identical. Accord- ingly, the differentia of activity experiences is still to seek, although some of their indispensable char- acteristics have been discovered. Seeking, striving, choosing, and the like, are the only activity experi- ences proper. They alone are dynamic without being passive, and the array of descriptive concepts we have mustered in the attempt to analyse activity apply to certain adynamic experiences also. Nothing except introspection can gauge the unanalysable residuum. This may seem a very meagre answer to our first question, but meagre as it is, it is sometimes disputed on psychological grounds. It will be well to notice some of the more important of these objections before proceeding to the second, and much more difficult, question of interpretation. The first objection is familiar to all students of Berkeley.^ ' Such is the nature of Spirit or that which acts, that it cannot be of itself perceived, but only by the eflFects which it produceth. If any man shall doubt of the truth of what is here delivered, let him but reflect and try if he can frame the idea of any power or active being. ... So far as I can see the words will, understanding, mind, soul, spirit do not stand for different ideas, or in 1 Principles, § 27. THE SELF AS WILL 111 truth for any idea at all, but for something which is very diflferent from ideas, and which, being an agent, cannot be like unto, or represented by, any idea what- soever.' But Berkeley speedily saw that his position required modification, and in the second edition of his book he set to work to amend it. Though we have not an idea {i.e. a presentation) of soul, spirit or the operations of the mind, we have a ' notion ' of them ' inasmuch as we know or understand the mean- ing of these words.' These operations may be called ' things ' or ' beings,' though not ideas. In our own case we comprehend them by ' inward feeling or re- flection ' ; their existence in other minds is matter for inference. And he seems to give the case away in the Dialogues between Hylas and Philonous.^ ' My own mind and my own ideas I have an immediate knowledge of ; and, by the help of these, do mediately apprehend the possibility of the existence of other spirits and ideas.' This cognition, he further informs us, is cognition ' by a reflex act.' In its broad issues Berkeley's original position Yas simply the denial of the possibility of direct intro- spection, and as this view has already been considered more than once, it need not be considered again. Berkeley's special reason for maintaining his view, however, is (at least in form) somewhat difierent from Gomte's. According to Berkeley an idea is the object of knowledge and is passive, and any possible object of knowledge is passive. But activity, ex vi terminorum, is active and therefore no active ex- perience, qua active, can ever be the object of know- ledge. But this analysis is fundamentally mistaken. Ideas of sense are perhaps passive since they come from a foreign source and bear the marks of their origin with them. But neither activity nor passivity play any part in the analysis of knowledge or per- ception. Knowledge is aware of its object, and this 1 Dialogue iii., Fraser's 4 vol. edition, vol. i. p. 448. 112 PEOBLEMS OF THE SELF chap. awareness is neither active nor passive although activity of some sort and passivity of some sort may be implied among the conditions of the existence of knowledge. Accordingly, Berkeley's difficulty is entirely manufactured, and he has not even succeeded in establishing a presumption in favour of the view that the activity element in experience is opaque to introspection. The object of knowledge need not be passive, and therefore there need not be any contra- diction in knowing an active experience as active. If such a contradiction existed it would also be fatal to Berkeley's ' notions.' It must be admitted, however, that many of the ' effects which it produces ' are often erroneously ascribed to the concept of activity itself. When strain, e.g., is said to be a feature of activity there is an example of this confusion. Strain is essentially a sensation, and so, it seems to me, is effort. They are qualia of the conflict between the activity and the obstruction it encounters. Similarly the feeling of release from strain is also a sensation. It is not activity itself. These sensations are bodUy, and this fact introduces a very real difficulty into the analysis. In the main there is no greater difficulty in dis- tinguishing active experiences from collateral bodily accompaniments than in distinguishing assent or denial from a nod or a shake of the head. But in some cases the difficulty is considerable. It is often difficult, for instance, to distinguish psychical activity from the sense of unconstrained bodily movement. The reason for this is, I think, that our seeking and striving is usually directed towards change in the physical world and our bodies are our only constant instruments for effecting such change. And from this point of view it is true to say, as Professor James does in a passage which I have already quoted in part, ^ that ' the body is the storm centre, the origin of ' Essays in Badical Empiricism, p. 170. THE SELF AS WILL 113 co-ordinates, the constant place of stress in all that experience-train. Everything circles round it and is felt from its point of view. The word " I," then, is primarily a noun of position, just like " this " and " here." Activities attached to " this " position have prerogative emphasis, and, if activities have feelings, must be felt in a peculiar way.' But it is false in fact that the body is invariably a centre of this sort. When a man sets himself to think out a problem in ethics, how is his body the centre of his action ? We can choose to do and strive to do things in which bodily movements have no place or, at any rate, no conscious place. And if we reflect carefully we shall find that the activity characteristics of choice or striving are in no way affected thereby. Hence the bodily sensations are only collateral accompaniments of activity and are not parts of its essence. A second objection takes the form of saying that introspection, carefully performed, shows no trace of any distinctive class of activity experiences. Thus, according to Professor Miinsterberg,^ ' a special collection of sensations is just what we call will.' We find in every volition that ' the clear conscious- ness of a certain presentation a is preceded by another state of consciousness which contains the presentation a as part of its content.' ^ This view, Miinsterberg contends, is inevitable because the alter- native view of an InnervationsgefUhl can be proved to be worthless. And that would be true if there were no other alternative. There is no good evidence for the InnervationsgefUhl. James and Miinsterberg have slain it.^ Miinsterberg's analysis would have been still more convincing had he expressed it thus : ' In every ' Die Willenshcmdlung, S. 96. The account in the author's GrundzUge is different. « Ibid. p. 97. ' See Miinsterberg, ibid, passim, ; and James, Principles of Psychology, vol. ii. pp. 494 ff. I 114 PROBLEMS OF THE SELF chap. instance of will the expectation, clear or obscure, of the possession of a certain object, or of the actual experience of a certain sort of bodily sensation, invariably precedes that possession or that actual sensation. This fact of expectation, which may be erroneous, together with our experience of the con- tinuity of the transition in the cases in which it is not erroneous, is all that we mean by will.' And he would add one further requisite. There must be a connection with our own bodies, and the presentation of their successive phases must be the perceived link between foretaste and reality. There are other variants of a similar doctrine. The theory of ideo -motor action, for instance, is one of them. It, too, denies that there are any specific experiences of striving or willing. We think of rising from bed in the morning, and lo ! we have begun to dress, without knowing why or how. That is the whole psychology of the matter. Striving is analysed away. It is a blend of two factors, (1) the expectation of the result of certain processes due to previous experience or, perhaps, merely the thought of this result, and (2) a preference or liking for the result. And sometimes the irrelevant remark is added that seeking or striving cannot be an original element. We cannot want anything without know- ing that we like it, and without supposing that we have a chance of getting it, all of which presupposes previous experience. This contention may, or may not, be true, and it is very often disputed, but it is clearly impotent to prove that striving is not now a distinctive element of consciousness. If the appeal be to introspection, then to intro- spection let us go. Is it really possible to explain away the experiences of seeking, endeavour, or resolve in Miinsterberg's fashion, or in any similar fashion ? Surely the analysis omits an essential point. Do we not all drift frequently into situations THE SELF AS WILL 115 which we anticipate with pleasure and yet do not actively seek after ? In these cases all the elements of Miinsterberg's analysis enter, and yet there is no seeking. I have already said a good deal in illustra- tion of this point, a,nd shall have something to say at the close of the chapter, so further reiteration is useless. The experience of endeavour may be too ultimate to be adequately described, but at least it contains elements which are ignored in this description. (^) Let us pass to our second question. There are activity experiences. We can point them out if we cannot describe them completely. But are they really active? Do they really do anything, and, if so, what do they do, and how do they do it ? May not the real agent in the case be something behind the so-called activity experiences, just as a feel- ing of freedom may be due to an absence of tension ? Are the experiences to be taken at their face-value ? And what precisely does this face-value mean ? These are much more difficult questions, but it is impossible to ignore them. They may be asked in a sense which is absurd and in a sense which is legitimate. If the question be, ' How does doing do ? ' or ' how does pulling pull ? ' it is certainly absurd. To ask how pulling pulls is to ask for a precise analysis of the process of pulling. Otherwise it is mere nonsense. But the question may be put it a form which is legitimate. It is legitimate to ask whether these activity experiences are causes, and, if so, what they cause. And it is also legitimate to ask whether they are causes in some sense in which otffer experiences are not. Let us address ourselves to these questions. The problem of the meaning of causation is one of the most intricate, and one of the most disputed in metaphysics. For our present purposes it is enough to distinguish some of its principal inter- pretations. According to some authors the only 116 PEOBLEMS OF THE SELF chap. intelligible meaning of causality is that it states a rule of the succession of events in time so that inference from one set of events to another set of events is possible. It is a principle of calculation (not merely of prediction) and it means nothing more. Nor is it used in any other sense in the sciences, e.g. in dynamics. But many would contend that this doctrine robs causation of its real meaning. Causation must mean something more in order to be intelligible, and this additional element of meaning is sought in various directions. The principal attempts of this kind are two in number. The first seeks to connect causation with ground and consequence, the second appeals to the very experiences which we are now considering. Activity, on the second view, is some- thing more ultimate than causation. We experience it in the self, and extend it by analogy or ' project ' it into our conceptions 'of causation in other things. We may begin by considering Hume's famous objec- tion to this latter contention. ' It may be pretended,' ^ he says, ' that the resist- ance which we meet with in bodies, obliging us frequently to exert our force and call up all our power, this gives us the idea of force and power. It is this nisus, or strong endeavour, of which we are conscious that is the original impression from which this idea is copied. But, first, we attribute power to a vast number of objects, where we never can suppose this resistance, or exertion of force to take place. . . . Secondly, this sentiment of an endeavour to overcome resistance has no known connection with any event : what follows it we know by experience but could not know it a priori. It must, however, be confessed that the animal nisus which we experi- ence, though it can afford no accurate precise idea of power, enters very much into that vulgar, inaccurate idea, which is formed of it.' And Hume also objects to ' Enquiry, Selby-Bigge's edition, footnote p 67. THE SELF AS WILL 117 the argument that we have an intuition of power in the case of voluntary action or, in other words, that we have only to raise an arm to understand causation, and so to refute the most astute dialectician.^ ' We learn from anatomy that the immediate object of power in voluntary motion, is not the member itself which is moved, but certain muscles, and nerves, and animal spirits, and, perhaps, something stiH more minute and more unknown, through which the motion is successively propagated, ere it can reach the member itself whose motion is the immediate object of volition.' In his more elaborate discussion of causation in the Treatise Hume remarks that ' the terms of efficacy, agency, power, force, energy, necessity, connexion and productive quality are all nearly synonymous,' ^ and he proceeds throughout on this assumption. If so, the first of his arguments is fully justified, and time has given it clearer proof. Dynamics, for instance, does not require the idea of force, although there are comparatively few who would deny that it employs the idea of cause. But we cannot suppose that the experience of acting freely or of being constrained is in any way applicable to the behaviour of a particle. On the other hand, ' force,' ' power ' and ' efficacy,' as usually understood, imply something analogous to this psychological experience of activity. Connection, correlation, causation as applied in dynamics do not. If, then, Hume insists on interpreting the former set of terms as equivalent to the latter, his criticism is justified, and it has clearly an important bearing on the question. The specific features of the animal nisus would be irrelevant to the general question of cause. Again, there is considerable value in Hume's con- tention that the efiects of striving are known only through experience, and have no more a priori ^ Enquiry, Selby-Bigge's edition, p. 66. ' Treatise, Selby-Bigge's edition, p. 157. 118 PROBLEMS OF THE SELF chap. evidence about them than any other kind of cause. His general position, of course, is that there is never any a 'priori evidence for particular causal laws. We find out by experience that bread nourishes, that morphia depraves, that a combination of two poisons results in common salt. Such knowledge depends upon observation. The principle of ground and con- sequent is helpless. The efi"ect8 of any substance cannot be deduced from the qualities of the substance until experience has shown what effects the substance works. These effects may be strange and unnatural : they may seem, in the end, a standing miracle, even when we know what does in fact result. In causal laws there is the mere brute fact of regular occurrence. There is nothing like intelligible connection. It is usual to reply to this that when striving and seeking is a cause there is an intelligible connection with the effect, that striving and seeking are therefore causes par excellence, and that any other kind of cause differs from them at its peril. We strive after an end and attain it ; we wish, for instance, to visit Florence and we go there. Surely in these cases there is an intelligible connection. The end for which we strive is present throughout : it craves completion, and it receives this completion when the process is fulfilled. But even granting that there sometimes seems an intelligible connection of this sort, there are negative instances in plenty. We try to be healthy and succeed in becoming so. But is the seeking the cause of the result ? On the contrary, its success depends upon the observance of the ordinary rules of health, and these are proved, if they are proved, by the collected and compared observations of physicians. Or, again, no one can, by taking thought, add a cubit to his stature, though some try, and no one has volitional control over any muscles other than those attached to the skeletal system. Experience is necessary to limit our seeking THE SELF AS WILL 119 to paths in which it can be successful. There is no a priori reason why it must be successful in these paths and not in others, and consequently our know- ledge of this species of causation is just as much dependent upon experience as any other. These arguments are sufficient to prove that activity experiences have no apocalyptic message to overcome the difficulties and perplexities of causation, but they do not prove in any way that such experiences are not causes at all, and that remark holds of Hume's third argument also. It is true that when we try to raise our arms the endeavour seems an immediate antecedent of the desired effect, whereas, in point of fact, an exceedingly complicated physiological process intervenes. The implication, to Hume's mind, is that these physiological processes are the real cause, and that their precise nature is unknown to us when we try to raise an arm. But, in the first place, Hume does not give a perfectly fair account of the facts. If we try to lift a finger we know that we must contract many muscles of the arm. We include these means in our striving for the end, or, rather, the end is a con- tinuous process which includes these movements of the arm, shoulder and neck. It is false, therefore, that we are ignorant of the intervening bodily chain. We do not, indeed, know it in the way an anatomist or physiologist knows it, but we are aware of it in the only possible way for us, viz. through kinaesthetic sensations and the like ; and I have already argued that these kinaesthetic sensations have as good reason to be considered real parts of the body as anything that the anatomist can observe.^ And in the second place, if the volition is not the sufficient cause of the movement of the arm, it may, none the less, be an indispensable part oi^ the process and perhaps the most important determining factor. Hume's argu- ' They are characteristic objects of the ' internal sense ' as that is interpreted in Chap. III. 120 PROBLEMS OF THE SELF ment, therefore, is inconclusive if it be intended to prove that there is no reason for supposing activity experiences to be causes. They have at least as good a right as anything else to be considered part-causes. Unless it is maintained that activity experiences have some peculiar tidings for the metaphysical prob- lem of causality, it is clearly absurd to demand a full explanation of all that causation means in order to justify the statement that activity experiences really are causes. The principle of causation must mean this at least, that events in time are connected according to rules, so that we have reason to suppose that one event makes a specific kind of difference to other events. None of them would exist in the way they do if the others did not exist at their own determinate time, in their own determinate way. Causation also implies a selective determination in a special sense of these words. Mill's insistence on the ' sum-total of conditions ' has misled logicians into, making the attempt to seek their causes in the total state of the universe at any given time. They might just as well, and they frequently do, relinquish the whole notion in despair. What we seek to do, in the attempt to discover causes, is not to expand our enquiry until it embraces all that is, but to narrow it down to those points of presumed connection which are really relevant. If we want to know why Jones has caught typhoid fever we try to discover from what particular spot the germs of the fever have probably come. We are not concerned with the rest of the universe. If we were, there would be little chance of preventing Smith and Robinson from falling victims in the same way as their friend Jones. In this sense of the words there is no good reason for denying that activity experiences really are causes and may be taken at their face value. Perhaps, how- ever, they cannot claim more than this. The experi- ence of ' force ' does not explain causation in general. THE SELF AS WILL 121 It is hard to see wbat it could explain. In any case it does not explain regular or necessary connection. It is irregular and capricious as far as experience goes, and it sometimes succeeds and sometimes does not. And we have already seen reason for supposing that it is very doubtful whether there really is an intelligible nexus between cause and effect even in the case of activity experiences. Not only are activity experiences causes, but they are peculiarly instructive because of the sense in which they imply selectiveness, initiation and the possibility of genuine novelty. That our seeking is selective is too obvious to require comment. We never seek everything all at once. And enough, perhaps, has been said with respect to initiation. I have tried to show how seeking and choosing seem to initiate, and now we can see that there is no good reason for denying that they really do initiate. It is false, of course, that they alone initiate, and absurd to argue that they are the only causes in the self or in the world. Such a view cannot be seriously maintained. They may have preponderating im- portance in some selves at all times and in all selves at some times. That is the most that can be claimed for them. And if either initiation or the possibility of novelty meant absence of continuity with other elements, or implied that these other elements were not past causes of the ensuing process, then their claims would have to be rejected. But neither of these implications holds in fact, as has already appeared in the discussion of initiation. Let me give an example of what I mean by the possibility of genuine novelty and the way in which it does not imply complete, or important, discontinuity. How often do we find, in reading the biography of, let us say, a great poet, that his youthful flights were singularly timid and uninspired? He is the sedulous ape of some other poet, his Pegasus is 122 PROBLEMS OF THE SELF chap. between the shafts and has cast a shoe. There are no symptoms of genius, but only an attempt to manipulate metres. And then, of a sudden, a new poet is born into the world. The sedulous ape has become a master, never again to return to his old level. In this case there is a real beginning which is not independent of previous events. Without the early trials and failures, without the travail of pains- taking apprenticeship, the real poet could not be born. Novelty, in this sense, is a plain fact of which any theory of causation must take account. But it is not independent of previous events, nor could it occur without them. I have purposely chosen this example because it has no special connection with striving or will. There was as much seeking and choosing at the stage of the sedulous ape as at any other. The new poet is born and does not majce himself. And this leads to an important reflection. When Berkeley and others maintain that spirit is essentially active their mean- ing is plain. They wish to show the distinctive difference between spirit and a moving panorama , of sense impressions. The history of philosophy sub- sequent to Berkeley has emphasised the necessity for recognising that distinction. Spirit is not a resultant of anything else. It lives and moves on its own account. But to say this is very far from saying that spirit is all compact of the particular class of experiences which we call active, or even that other experiences are more dependent on these than vice versa. That is an interpretation for which there is no warrant. Cognition and feeling play their part in psychical life as much as endeavour. They are not merely bye -products of endeavour; they are causes in the life of mind as much as endeavour. They have an independent function in that life ; they are existent causes. Sometimes an act of will may predominate and seem to sweep the others in its V THE SELF AS WILL 123 train. Even then it is not the sole cause, though, perhaps, the most important. But the other elements of consciousness may also predominate in other dis- tinctive phases of the life of the self. To pass from the independence of the self to the primacy of experiences of conation is a salto mortale. This point is so clear that it is needless to defend it. If I were to try to do so I should begin by pointing out that cognition is as selective as conation, that it initiates just as clearly, that it lives because of the possibility of novelty, that it exhibits a closer con- nection between plan and fulfilment than conation itself. But such an argument would not really make the issue plainer. (3) It remains to give a more precise enumeration of this special class of activity experiences and, in particular, to supply an analysis of will in the narrower, which is also the more accurate, sense of the word. To illustrate the first point I shall quote a passage from the earlier edition of Stout's Manual of Psychology. ' Such words as interest, craving, longing, yearning, endeavour, desire, purpose, wish and will, all mark this characteristic of the process of consciousness. All of them imply an inherent tendency of conscious states to pass beyond them- selves and become something difi"erent, an inherent tendency which continues to operate, unless inter- rupted by interfering conditions, until a certain end- state is reached, which is called the satisfaction or fulfilment or realisation of the interest, craving, longing, yearning, endeavour, desire, purpose, wish or will.' ' It may be true that all the states mentioned by Professor Stout have an inherent tendency to pass beyond themselves, but some of them are attitudes of feeling, not of activity, and the tendency to pass ^ Manual, p. 64. In the latest edition the whole passage has been rewritten. 124 PROBLEMS OF THE SELF chap. beyond themselves is really their tendency to arouse activity proper. We have already seen that desire is a feeling attitude, a way in which the subject is aflfected, though it is so closely connected with activity that the two may easily be mistaken. This is still more obvious in the case of interest, and it is also clear in the instances of craving, longing and yearning. We have not, therefore, such an exuber- ance of examples as Stout mentions here, but we have the facts of consciousness indicated by such words ag striving, seeking, endeavour, appetition. There are some slight distinctions between these, but the distinctions {e.g. in the case of appetition) refer chiefly to the characteristics of accompanying states of the body. We may pass, then, to will itself The exact range of the term voluntary is a matter for dispute. Indeed it is perhaps most convenient to define the term negatively and say with Mr. G. B. Moore ^ that voluntary actions are those which an agent could have done otherwise had the choice been open to him. Even on this definition, however, the fundamental feature of will is choice or its possibility, and the nature of will is seen most clearly in choice after deliberation. The question of resolve need not receive separate treatment, for it is either identical with choice or else means the choice to choose at some date other than the immediate present. When we deliberate, then, we entertain certain alternatives of action as possible objects of choice. And we must also believe that these objects can be attained through our choice, either directly or in- directly. We cannot, strictly speaking, choose, unless we believe that the object of our choice is within our power. The thought of will implies the thought of ' can ' ; otherwise there is not will but wish. But when we have summed up the advantages on either side, and compared them in any way in which such ^ Ethics (Home University Library), p. 13. THE SELF AS WILL 125 comparison is possible, we have not yet reached the essence of will. There still remains the element of choice or consent — that fiat of the will which is being interpreted, ' Let it be done, and done through this my choice.' As the assent of judgment is always more than the entertainment of the meaning of a proposition, so the choice of the will is more than the preliminary deliberation which balances the books. Assent of judgment and choice of will have much in common, but they are also essentially distinct, since the one is a theoretical and the other a practical attitude. If anything can claim to be an experience, choice certainly can. If any reliance can be placed on observation, then, surely, choice influences conduct, determines the current of our psychical life, helps or hinders bodily movement. Both for introspection and from the standpoint of achieved results choice is an active process. I am glad, in a matter of this sort, to be able to adduce the weighty testimony of Professor James. The passage which I am about to quote deals with the relation between the assent of judgment or belief and the fiat of choice. The analysis of the former seems to m'e to exaggerate the importance of emotional and volitional characteristics, but that of the latter is, to my judgment, true in almost every line and syllable. ' To the word " is " and to the words " let it be " there correspond peculiar attitudes of con- sciousness which it is vain to seek to explain. The indicative and the imperative moods are as much ultimate categories of thinking as they are of grammar. The " quality of reality " which these moods attach to things is not like other qualities. It is a relation to our life. It means our adoption of the things, our caring for them, our standing by them. . , . And the transition from merely considering an object as possible to deciding or willing it to be real; the change from the fluctuating to the stable personal 126 PEOBLEMS OF THE SELF chap, v attitude concerning it ; from the ' don't care ' state of mind to that in which we mean business is one of the most familiar things in life. We can partly enumerate its conditions, and we can partly trace its consequences, especially the momentous one that when the mental change in question is a movement of our own body, it realises itself outwardly when the mental change in question has occurred. But the change itself, as a subjective phenomenon, is something which we can translate into no simpler terms.' ^ What is true of the imperative mood is also true of the whole range of endeavour, and analysis can proceed no further. But enough has been said to furnish some indications of the nature of conation, and so to provide a basis for the discussion of its place and function in the self. Let us, therefore, proceed to the consideration of the familiar doctrine that the essence of the self is to be purposive. ' Principles of Psychology, vol. ii. p. 569. CHAPTER VI PSYCHICAL AND PURPOSIVE Although the division between psychical and physical is not exhaustive/ it is of great importance, and the usual course is to maintain that the psychical is purposive and the physical non-purposive. When we examine the realm of being we find that there is a broad distinction in the way in which things behave. Behaviour which results from conscious purpose is very different, even to the external observer, from the behaviour of mere matter, and there is much in the behaviour of organic beings which is so similar to purposeful movements, and so unlike merely mechanical action that it is usual to call such behaviour * purposive ' from its analogy to actions inspired by conscious purpose. The extension of the term in this neutral sense is, of course, legitimate and does not call for comment. The fact that the presence or absence of indica- tions of purpose is the best criterion of the existence of mind does not prove in itself that purpose has the primacy among experiences. Indeed, the advantage of the criterion is chiefly that it is an objective test which the external observer can apply, and this is clearly an indirect method of describing conscious processes. None the less the mere emphasis upon the significance of purposive behaviour tends to be 1 Cf. Chap. II. pp. 16-18. 127 128 PROBLEMS OF THE SELF chap. an argument for the primacy of purpose, and is constantly present in the minds of Voluntarists, whether obscurely or explicitly. The claims of pur- pose for primacy must therefore be considered from this point of view. It is unnecessary to challenge the assumption that the marks of purpose are really the best indications of the existence of mind, and their absence the sign of the existence of mere matter. There are some monadistic theories, it is true, which maintain that everything which exists is a self, and consequently that there is nothing in the universe which is non- purposive. But those theories must ascribe properties to matter which, in the ordinary acceptation of terms, it emphatically does not possess. The sun and the planets, we believe, and the whole realm of inorganic nature pursue a magnificent, if sightless, course with unfailing regularity and precision. Their action seems at the opposite pole from that of mind. Psychical beings strive ceaselessly to protect the little corner where they abide, to save themselves from extinction, and to perpetuate the species. They select what will achieve these ends, and reject all else. They adapt themselves to their conditions as well as they can. If their movements are ordered by general laws, the laws must take account of their aims and their interests. To keep their feeble micro- cosms intact they will make all sorts of shifts which would be meaningless unless nature existed for them as well as through them. Their action thus frequently appears incalculable, and the unity of their lives a unity of individual aims instead of a mere instance of general laws. If you put obstacles in the way of a lover, he may laugh at you for your pains. Deflect a particle, and it will obey without a murmur. What wonder, then, that this very marked difference should be seized upon as the real differentia of selfhood ? '' De facto purpose,' as Professor Bosanquet assures VI PSYCHICAL AND PURPOSIVE 129 us, 'is a psychological, temporal and ethical idea,'^ and he seems to imply that its claims may, therefore, be slighted. But purpose, even at the humble level of psychology, time, and morals, is sufficiently ambiguous to invite discussion, and so are ' end ' and ' teleology,' They also are ambiguous, as Professor Bosanquet ably argues, when they are taken at his ' higher ' level, and he is probably right in asserting that the principle of Teleology, when applied to the Absolute, vanishes in favour of something else. For this reason, as for so many others, it is imperative to begin this enquiry by examining the various possible meanings of teleology, end, and purpose, and the relations between them. The most usual and the most natural sense of the word purpose is that which is implied in conscious striving or conscious choice. What we .seek to *do, or what we strive to do, is our purpose. In this case we are consciously aware of the end, and en- deavour to bring it to fruition. But ' purpose ' and ' purposive ' are also used in a wider sense, not always identical with the natural extension of the term 'purposive' which has already been mentione'd. In the first place the terms are frequently applied to any process which appears to be directed to an end whether that end be in fact present to consciousness or not. Thus instincts are said to be purposive because they appear, to an external observer, to be like processes which consciously strive after an end. But the animal which acts instinctively need not be conscious of the end at all. On a still more extended usage, life, of any kind, is sometimes called purposive. Or, again, the claim is frequently made on behalf of certain idealistic theories that they are teleological. In this case the reason is that idealists attach so much importance to values and ideals in their inter- pretation of existence. That is one of the implications ' The Prmeiple of Individuality and Value, p. 127. K 130 PKOBLEMS OF THE SELF chap. of the simile that the universe depends, in the end, on the idea of the good, as sensible things depend on the sun. Indeed, those who are idealists usually base their belief on this ground. They welcome idealism rather because it is a city of refuge for ideals than because it exalts ' ideas ' at the expense of matter. And there is yet another respect in which idealism may be called teleological or purposive. Purpose refers to final causes, and perhaps there is no final cause except the whole. Idealism insists that the cosmos is a perfect whole. In this sense teleology means the explanation from whole to part. How is it that one and the same term can have such diverse interpretations and may be used to justify so many theories ? The fact is not very difficult to explain even with regard to those cases of ' finite and outward design ' which Hegel ranked so low. Even in their case it is clear that end may expand into something like system, and is closely connected with value. There are analogies to both in explicitly conscious purpose, and extension by analogy seems the essence of the argument here. It is usual to draw a distinction between end and means. Mr. A wishes to cross the Atlantic in order to see his friends, and so he books his passage on some liner. To distinguish end from means is easy enough in such an instance, and it is not difficult to specify the characteristics of either. The end is distinct from the means, and the means are directed towards it ; the means are selected with a view to the end, and are adapted to the conditions which render the attainment of the end feasible ; and both means and end are restricted, specific, particular. But even this analysis soon leads us to perplexities and intricacies. What precisely is the relation of means to end from the point of view of psychical analysis ? When the means are external to the end, as in the choice vi PSYCHICAL AND PURPOSIVE 131 of a steamship or a railway train, the relation seems only that the means are instruments for attaining the end. But even in this instance Mr. A plans, not merely to see his friends, but a whole expedition, and the voyage or the journey is an integral part of this whole. And how, when we consider a long process of striving towards a goal with the hindrances incidental thereto, are we dealing solely with an instrumental relation, and in what sense is the end really distinct from the means ? Or let us take another instance. A student wishes to take his degree and prepares himself for examination. If his end is only the right to append certain letters to his name, and wear a hood, and receive an academic benediction, then, no doubt, there is a clear and well- defined distinction between end and means. But if he has sufficient intelligence to desire the degree as a token of efficiency and knowledge, then there is at least no temporal distinction between end and means, between the proficiency and the preparation, for he becomes proficient in so far as he prepares. And, again, the degree is not the be-all of his existence. It may be merely a means towards obtaining a com- petency or it may be that and also an end in itself. This example is sufficient to show that means and end are often closely interwoven, and many writers are driven by these difficulties to deny that the ends which we choose are really ends at all. The proximate end of any particular choice or striving is often far from being an end in itself. Psycho- logists usually describe these proximate ends by saying that they are the end states of particular conative processes, and that, when they are attained, one conative process ceases and another succeeds. But this account is mistaken. Suppose our proximate end is to eat a good dinner. The end state of that dinner is the last mouthful of dessert or the last sip of coffee. When this is attained the conative process 132 PEOBLEMS OF THE SELF chap. ceases, but the end-state is not the end or aim of the process. It is not the object of our striving, but only the last state in time. We might as well maintain that the aim of a hungry man is satiation. He aims, not at satiation, but at a good square meal. Accordingly it is maintained that the end of a con- ative process should in no respect be regarded as the last state in time of that process but as the completion or completeness of that process, and in that case every part of the process will form part of its completion, since all are necessary to it as a whole. Similarly it may be held that the end of a process is not satiation, but satisfaction, and this again has no specific refer- ence to the last state, in time, of the process. These views are clearly much more adequate than the former (even to express the nature of finite particular cona- tions), and I have but one criticism to offer. The satisfaction is net the direct object of a conative process. It is a feeling attitude which accompanies attainment and prompts to fresh attainment, but it presupposes that attainment and it is found, not sought directly. The principal arguments against psychological hedonism are fatal also to the doctrine that satisfaction is the object of choice. We choose objects, not satisfaction, except in and through their attainment. To get satisfaction we must forget it. It is clear, therefore, that the consideration even of particular plans brings us near to the question of totality, and this implication is still clearer when we remember that even the ordinary man does to some extent subordinate his particular aims to the plan of his whole life. Accordingly any explanation which proceeds from whole to part is sometimes called teleological ; and wholeness and completeness is often said to be the principal characteristic of purpose. In this sense anything which acts as a whole in any marked or peculiar way is sometimes said to be there- fore purposive. VI PSYCHICAL AND PUKPOSIVE 133 And the implication, or the possible implication, of value is also manifest. When we choose to do anything we ought to choose the action because it will bring about more value, on the whole, than any other which is open to us. No doubt we frequently act otherwise than as we ought. We sometimes know- ingly choose to do things which we have no reason to suppose will conduce even to our own private and particular welfare. But we always consider them worth doing from some aspect and from some point of view, and this is what appeals to us at the moment of choice. Any process, therefore, which is directed towards the attainment of value is sometimes called purposive. Is the self, then, distinctively a purposive entity in some one of these senses or in all of them ? Probably every one of these implications occurs at some one time or other in arguments upon the subject. In modern days, however, most of the discussion centres round biology. The behaviour of living beings, it is argued, requires the conception of purpose or some- thing analogous to it, in order to become intelligible. There are two great classes of existent beings, those which act mechanically and those which act pur- posively, and purposive behaviour either implies a self in every instance, or selves form a sub-class of purposive beings. ' We may then define psychology as the positive science of the behaviour of living things. . . . We all recognise broadly that the things which make up our world of perceptible objects fall into two great classes, namely, inert things, whose movements and changes seem to be strictly determined according to mechanical laws, and living things, which behave or exhibit behaviour; and when we say that they exhibit behaviour, we mean that they seem to have an intrinsic power of self-determina- tion, and to pursue actively or with effort their own welfare and their own ends or purposes.' ^ Let us > M'Dougall, Psychology (Home University Library), pp. 19, 20. 134 PROBLEMS OF THE SELF chap. consider this question more precisely in order to discover, if possible, why life of any sort should be called purposive. The problem of the meaning and adequacy of mechanical categories in biological ex- planation is, of course, a technical one which can only be decided by experts on the subject, if, indeed, it can be decided by any one. The layman must walk warily and has no right to expect that his personal views are entitled to any particular respect. But he is entitled to discuss the general nature of the arguments adduced and to consider what they could prove if they were shown to apply. The range of the term ' purpose,' as used in this sense, is clearly a very wide one. It must apply to physiology as well as to psychology, to plants as well as to animals, to digestion as well as to ratiocination. This, for instance, is the wide meaning in which Schneider uses the term. ' When oxygen combines with iron and thus produces rust this is not a purposive process. But we have behaviour according to purpose when oxygen is combined with the carbon in our blood and so promotes our conservation.'^ The term purpose has here no special implication of conscious striving. It means only conduciveness to an end which has value. In this sense, any feeling, organic or psychical, any impulse or instinct that helps the animal in the struggle for self-preservation or the preservation of the species should be called purposive. The problem is whether explanation in terms of purpose is really significant, when analogies are pushed so far, and in so many directions. The arguments in favour of neo- vitalism — for it is these which we are discussing — have a negative and a positive side. They state (negatively) that the behaviour of living beings cannot be explained in terms of mechanism, and (positively) that the ex- planation which mechanism cannot afi"ord must be ' Der thierische Wille, p. 24. VI PSYCHICAL AND PURPOSIVE 135 sought in something analogous to immanent purpose. External teleology, the kind of purpose invoked by the Argument from Design, is, of course, irrelevant to the question since it is perfectly compatible with mechanism. God made a machine which works according to the Laws of Motion. Let us, then, consider the negative arguments. The behaviour, even of the lowliest organisms, exhibits characteristics to which there is no real analogy in the inorganic world, and consequently it is a mistake in principle to attempt to explain them by categories which have proved their worth in that world only. No machine, and no collection of particles, exhibits the phenomena of restitution, reproduction, adaptation, selection, or persistence in a certain direction despite all obstacles. A ship cannot regenerate a lost propeller. A newt can regenerate a lost leg. ' Thus we see that, at the very bottom of the evolutionary scale, animal behaviour exhibits the two pecuUarities which at all higher levels also distinguish it from the movements of inorganic things, namely, (1) the "total" or unitary nature of reaction, i.e. the reaction of the organism as a whole with co-ordination of the movements of its parts in response to a stimulus directly affecting one small part only ; and (2) the persistence of the effect of the stimulus, a persistence closely analogous to that persistence of varied move- ment which in ourselves and our fellows we recognise as the expression of a persistent effort after a desired end. And to this it must be added that these persistent and varied and total or unitary reactions of the whole organism are in the main adaptive, i.e. of such a nature as to promote the welfare of the creature.' ^ ' M'Dougall, Body and Mind, pp. 260, 261. This is really a negative argument, and it occurs in the chapter entitled ' Inadequacy of Mechanical Conceptions to explain Human and Animal Behaviour.' 136 PROBLEMS OF THE SELF Now before considering the positive side of the contention of the neo-vitalists (and tkey are disagreed upon it), we ought to remember that a negative argument is of two kinds, and that these kinds carry very different degrees of weight. It may be a disproof of possibility, or it may be merely a proof of failure. If the neo-vitalists could prove con- clusively that mechanical categories could not possibly explain animal behaviour, then biologists would be compelled to seek some other type of explanation. If, on the contrary, they could only prove that biology is still, in many respects, an occult science, and that the principles of nineteenth-century biology have not succeeded so well as was expected, then the biologist of the twentieth century might reasonably seek some other hypothesis, but would not be compelled to do so. The lacunae and the failures of science are apt to prove themselves far from unsurmountable. Professor Driesch is one of the best-known ex- ponents of neo-vitalism, and he is fully aware of this difference in the probative value of negative arguments. Indeed, he discriminates between his own arguments from this point of view, as a brief reference to the first of his recent lectures on The Problem of Individuality^ will show. The point in dispute is whether a machine can or cannot be the source of life, and Driesch admits the theoretical possibility that the machine theory could explain some of the instances which are prima facie favour- able to the opposite view. It is not intrinsically absurd, he maintains, that adaptation and immunity could be mechanically explained, although there is no known machine which can, e.g., produce an anti- ^ These lectures give a condensed and very clear summary of the argument of the earlier Gifford Lectures, entitled The Science anid Philosophy of Che Organism. The shorter form is better fitted for the purpose of a rapid survey like the present. It is also the more recent presentation of the argument, and the logic of it is just as clear. VI PSYCHICAL AND PURPOSIVE 137 toxin to rust. The same is true of regeneration. We know of no machine which is capable of restoring itself, a'nd so the probabilities are in favour of neo- vitalism. But there is no direct contradiction in the conception of such a machine. Driesch contends that the facts are otherwise with regard to morphogenesis. He was able to show, by experiment, that in certain cases artificial interference with the embryo will not affect the normal course of the development of the organism, except in the way of reducing its size. These experimental results were varied and striking, but the mention of his early discoveries with regard to the development of the sea-urchin will sufiice to indicate the logical bearing of his argument. ' The so-called " cleavage" of the egg . . . ends in the formation of the hlastula, i.e. a hollow sphere built up of about a thousand cells, forming an epithelium. If you cut this hlastula with a pair of very fine scissors in any direction you like, each part so obtained will go on developing — provided it is not smaller than one quarter of the whole — and will form a complete larva of small size.' ^ Facts of this kind, Driesch maintains, stand in plain logical contradiction to the very meaning of a machine. A machine is ' a given specific combination of specific chemical and physical agents,'^ and an arbitrary and random disarrangement of a machine could not leave its equilibrium unaffected. This argument seems inconclusive. In the case of the hlastula of the sea-urchin the smallest frag- ment which is capable of developing as a whole is one quarter of the original hlastula, and therefore must contain at least 250 cells by Driesch's own computation. Similarly, in the other cases, the fragments must be relatively large and complex. If, then, there is no logical contradiction in the mechanistic account of the normal development from 1 The Problem of Individuality, p. 11. ^ Ibid. p. 17. 138 PKOBLEMS OF THE SELF a complete blastula, it is surely not absurd to explain Driesch's results by saying that fragments of the blastula may develop into complete, though smaller, organisms in those cases in which the arbitrary dis- section of them results in leaving a sample which contains all the requisite physical and chemical ingredients. The laws of averages would make such sampling not intrinsically improbable. It is scarcely surprising, therefore, that many of the neo-vitalists themselves dissent from Driesch with regard to this argument.^ It is impossible to refute mechanism absolutely, although some statements of the mechanical theory may be intrinsically absurd. On the other hand, the negative arguments certainly prove that there are great difficulties in the mechanical conception, and the principle of the economy of hypotheses may be in favour of neo-vitalism. The positive formulation of the vitalistic or neo- vitalistic argument differs greatly in different authors and sometimes does not appear at all. Indeed, the negative side of the argument is much stronger than the positive. Still, the positive formulations when they occur are in strict conformity with the rules of inductive procedure. A hypothesis must be framed upon analogy ; and it is maintained that the true analogy in this case is derived from psychical purpose. Driesch, for instance, calls the teleological factor in action, which he postulates, a ' psychoid.' ' I propose the very neutral name of psychoid for the elemental agent discovered in action. , " Psychoid " — that is, a something which, though not a " psyche," can only be described in terms analogous to those of psychology.' ^ ' Cf. J. S. Haldane, Mechanism, Life, and Personality, p. 27. 'There is no evidence at all that each cell, in growing and dividing in the one particular manner which constitutes normal development, is not determined by special physical and chemical stimuli peculiar to its position relatively to the other cells, and to the external environment. We do not yet know what these stimuli are ; but probably no physiologist would doubt that they exist, and will be discovered when our methods are fine enough. Hence Driesch's argument for an independent vital force breaks down entirely. ^ The Science amd Philosophy of the Organism, vol. ii. p. 82. VI PSYCHICAL AND PUEPOSIVE 139 He distinguishes this ' psychoid ' of action, however, from the teleological factor in morphogenesis or restitution. For it he reserves the still more neutral name of entelechy. Dr. M'Dougall, however, does not draw these subtle distinctions. ' The embryo seems to he resolved to acquire a certain form and structure, and to be capable of overcomihg very great obstacles placed in its path. . . . This power of per- sistently turning towards a particular end or goal, manifested in these two ways, namely, in growth and bodily movement, is the most characteristic feature of the life of organisms, objectively regarded. It seems to involve essentially teleological determination, that is to say, it seems to be essentially of the same nature as the striving towards a goal or end that runs through all our inner experience, the goal being present to consciousness with extremely different degrees of clearness and fulness. . . . The processes seem to be essentially teleological, that is to say, they seem analogous to the behaviour of organisms, which from analogy with our own experience of purposive striving we believe to be prompted by psychical impulse and, in the more highly developed organisms at least, governed and guided by some prevision of the end to be achieved.' ^ Hypothesis rests on analogy ; but some analogies are good, and others not so good. Our question is. On what grounds of analogy are we led, or compelled, to believe that purpose is the essential characteristic of all life ? The best way to approach this question is, I think, to consider those cases in which purpose unquestionably occurs, and then to attempt to dis- cover how far there really is an analogy in cases which seem more doubtful. And this, as we have seen, was the way in which biology quite legitimately came to use the word purposive. Granting that some behaviour is not a mere reflex but implies conscious • Body and Mind, pp. 242, 243. (Italics mine.) 140 PROBLEMS OF THE SELF direction and purpose, then any behaviour which resembles this in important respects may well be called ' purposive ' whether there is proof of conscious purpose or not. We have every reason to believe that the choice of will influences action. Choice and resolve, especially after deliberation, demand the presence of rational reflection : they are antecedents of the ensuing actions : and there is no good reason for denying that they are determining antecedents. Here, then, is a case of action which is really purposeful and not merely purposive, which is due to the psyche and not merely to the psychoid. Similarly, as was argued in last chapter, purpose in the sense of conscious striving after a conscious end does work effects. And it is also true that these processes may persist for long in face of obstacles, and may try all means of attaining their end. They.may also be consciously subordinate to a general plan of life. But they need not be. When, however, we come to the case for the psychoid we come to what is really a very weak analogy. Let us take, for instance, the whole range of instinctive action. The great instincts, no doubt, involve conscious striving, a ' unitary ' process and the rest ; and their striving is directed towards an end. But so far as introspection goes they are consciously directed towards a proximate end only, and not towards the welfare of the organism as a whole or of the species as a whole. Even in human experience the instinct of sex at the time when it is strongest is not consciously directed towards its biological end — the perpetuation of the species. The youth seeks the maiden because he loves her and wants her, and not for the sake of his duty towards posterity. He does not choose his mate for the reasons that actuate the breeder of pedigree stock. If he did, there would be no need for the apostle of eugenics. And if this is true of the human species. PSYCHICAL AND PUEPOSIVE 141 is it not even more profoundly true of other animals ? The very perfection of many animal instincts seems to show that they cannot, properly speaking, be conscious of their end. The male and female larvae of the stag-beetle are of the same size, but the male larva builds itself twice as big a hole. There must be room for its horns to grow, but is it possible to believe that the larva has any consciousness of this fact ? ' Instinct,' says Hartmann, from whom the above example is chosen, ' is conscious willing of the means to an unconsciously willed end.' ^ But the means are not consciously willed as means. Certain ends are chosen which do, in fact, lead to a farther end which is biologically useful. But this does not imply the vaguest degree of knowledge of what the biological end is. To say that it is known dimly or implicitly is as important a piece of information as to say that we all know, dimly and implicitly, what life is or what are the implications of Euclid's axioms. And if this holds of instiuct surely it holds, a fortiori, of growth. Each of us, with individual variations, grows to the form and stature of a man, and each wishes to do so. But we do not, strictly speaking, choose or try to do so. Profiting by the experience of others we may aid and abet the process in various minor ways. We may play games and eschew cigarettes. But we do not grow to be men by taking thought. The strong and healthy man does not try to be strong and healthy in any sense in which his weaker brother does not make the same attempt. He becomes strong and healthy. It is not surprising that conscious purpose should exhibit some analogies to bodily behaviour, since the body is the instrument of action, and its behaviour is our only or our most important clue to the existence of any conscious purpose other than our own. But to 1 The PMlosophy of the Unconscious, vol. i. p. 88. 142 PROBLEMS OF THE SELF suppose that digestion, for instance, or other bodily processes of a similar kind, participate in conscious- ness even in the vaguest degree is to mistake fable for fact. We might as well believe that Menenius Agrippa related an actual historical conversation between the belly and the other members. And when we leave the higher animals and come to lowlier animals and to plants, what reason have we for attributing anything like conscious purpose to them ? The ' psychoid,' doubtless, has too much in its favour to be summarily dismissed. What I am urging at present is that there is a difference in kind between that which de facto makes for the conservation of an individual or a species, and therefore may be represented as a means to this end, and that which is consciously chosen in order to bring about this result. Nobody, maintains, of course, that a beaver builds a dam or an infant seeks the breast with a conscious foreknowledge of the biological utility of such actions, and an explicit resolve to promote that utility. If it were so, the infant would have a surer and more reasoned knowledge of the nature of things than the majority of adult men, and a casual assembly of politic rooks in a field have the collective wisdom of the British House of Commons. It is true that the range of conscious purpose extends further than is frequently supposed, and also that there is a wide domain of subconscious purpose. The evidence which, e.g., has been collected by Freud ^ and others concerning the rdle of repression in for- getting seems to leave no doubt on this head. But the realm of the subconscious, by its very definition, exhausts the cases in which there is any justification for supposing that anything analogous to conscious- ness (in an important sense of analogy) enters ; and, therefore, it is necessary to remember that there is , ' Vide his Zwr Psychopathologie des Alltagslebms and Traumdeutung. PSYCHICAL AND PURPOSIVE 143 much in the sphere of the organic in which there is no reason to suppose that any subconscious purpose plays its part. When that is the case and there is a diflference in kind, it is futile to speak of purposive determination, and, consequently, there is no justifica- tion for believing in the primacy of will. Instinctive action is an instance in point. Conscious purpose enters frequently into such action, and may even be present, through inheritance, on the first occasion in which the instinct is called into being ; for some believe that meanings are inherited. There is, there- fore, no need to return to the ancient doctrine that Instinct must difiier toto coelo from Reason, or to say, with Pope, that when Reason is at fault , . . . honest Instinct comes a volunteer, Sure never to o'er-shoot, but just to hit ; While stiU too wide, or short, is human Wit. But, on the other hand, conscious and subconscious purpose, by themselves, are quite incapable of ex- plaining every instinctive action. To illustrate this point I may refer to Professor Stout's very interesting discussion of the inheritance of meanings.^ It is generally agreed, nowadays, that instincts are modifiable through experience of the success or failure attending their action. But it is still maintained that such complicated actions as nest-building, migrating, and the like are performed blindly on the first occasion without any prevision of the end. That is much more than a denial that there is any awareness of the biological advantage of performing the action prompted by instinct. The chances are that no animal, other than man, is ever aware of this. The argument is that the animal, at the first time e.g. of nest-building, has no awareness whatever of what it is about. It acts from merely biological causes. To this Professor Stout answers, ' Maimml of Psychology, 3rd edition, Book III. Part I. chap. i. 144 PROBLEMS OF THE SELF I think conclusively, that the process has all the marks of intelligence on its first performance. The animal is alert, attentive, and not merely aware of something happening inside it, as it ought to be on the purely biological view. Moreover, the fact that there is adaptation of instinct through experience seems to show that the development is continuous. It is a modification of experience through further experience, and consequently there is not a difference in kind between the first performance and the sub- sequent ones. It would seem, then, that instinctive action, even at its first performance, implies a degree of intelligence which cannot be completely accounted for by the previous experience of a particular animal. But one of the merits of Professor Stout's discussion is the careful way in which he limits the presumption, if not the proof, that meanings are inherited. The factors, analogous to conscious purpose, which it is necessary to assume as part of the being of instinctive action, are very restricted. ' It is only necessary to assume an awareness of the present state as transitional — as something which not merely is but is to be. Such rudimentary reference to the future is not wholly indeterminate ; it is specific inasmuch as it is concerned with the further development of a specific situation and, more particularly, of certain selected factors within it. It is vague inasmuch as the animal has no clue to the particular nature of the changes which are to take place. The important point is that the situation is apprehended as alter- able. This is enough to make conation possible.'^ If then the part which purpose, even in the dimmest sense, plays in instinct is very restricted, how much more is it restricted in ordinary processes of growth, alimentation and the like. Why do we need the ' psychoid ' for all biological explanation ? Indeed, if the behaviour of organisms is to be ^ Manual of Psychology, 3rd edition, Book III. p. 355. PSYCHICAL AND PURPOSIVE 145 explained by the concept of purpose, it is by purpose in the sense of explanation from whole to parts — which is another matter, though not totally devoid of analogy to the purposes we know. In that case three questions are implied. (1) In what sense is explanation from whole to parts legitimate ? (2) In what sense is it required for the explanation of the organic and not required for the explanation of the inorganic, and may not the former sense make it unnecessary to believe in the existence of a distinct and peculiar 'vital' or 'purposive' factor? (3) Is is not possible that the principles of biology need not be identical either with those of psychology or with those of mechanism ? (1) To explain anything is simply to give a systematic account of it, and if a reference to the whole introduces system in our comprehension of it, then explanation from whole to parts is, eo ipso, legitimate. In this sense it is legitimate to say that inflammation means an increase of the blood pressure in a certain area of the organism in order that a foreign body detrimental to the organism may be expelled or destroyed. Or, again, it is legitimate to explain the fact that the brain of a man who dies from starvation is still well-nourished as compared with the rest of the organism, by saying that the brain is pre-eminently important from the point of view of the organism as a whole. But it is a different thing to examine the matter more closely and explain what this teleological factor is. The main point to notice is that it cannot be a factor or element co-ordinate with the parts. It is irrelevant whether it exercises energy or, as is the case on Driesch's theory of the entelechy, it only exercises guidance without work. The latter con- ception may be useful, even essential, in explaining the r61e of consciousness in the production of move- ment. If the law of the conservation of energy must L 146 PEOBLEMS OF THE SELF chap. apply to the organic as well as to the inorganic — and that cannot be proved, though certain experiments may show it to be probable — then we must employ this conception. But it is always meaningless to assert that the whole, as an element distinct from or additional to its parts, can either work or guide without work. The idea of the whole may guide conscious action, because it is possible to think only of the general characteristics of the whole, and this idea is particular like other ideas, or, rather, the act of reference to such a general idea is as particular as any other act. But to speak of the whole as an entity co-ordinate with other parts is to make the whole a particular — as Hegel emphatically maintained. Wherever there is systematic connection of any sort, there the conception of whole is as necessary as that of parts. Whole and part are in fact correlative notions for explaining a given set of facts, and neither of them are factors or elements in those facts. The real question is how far elements in a whole can be said to remain the same elements when they enter into some other combination. Have we any right to believe that the same substances can be parts of different wholes at the same or different times ? As we shall see when we come to discuss the nature of substance, there are several senses in which they can. The question itself is one of degree. Transfer a brick from one heap to another and it seems obvious that the brick remains the same and that the two heaps are also, to all intents and purposes, identical. Take a brick from its place in a building, however, and the building, at least, will seem to have changed. Take a cell or collection of cells from an organism, and the cells will certainly change. They will become mere matter, while the organism, until it can either restore the cells or otherwise compensate for their loss, will also be affected by the change and cease to be precisely the same whole. In every one of these PSYCHICAL AND PUKPOSIVE 147 instances a part is taken from a whole. The only question is the intimacy of the relation of parts to whole. It is futUe to argue that the question hinges upon quality and relation, or to say that when anything becomes part of a new whole all that happens is that it changes its relations. That may, or may not, be the case, but it is just as essential to include relations in treating of part and whole as it is to include quality. The whole is the parts as they are related together, and nothing more. It is not the parts considered as an aggregate to the neglect of all save numerical relations, nor is it the parts considered in other relations. It is not even true that we can ever consider the parts separately from the whole. That which we call a part when we consider it as a member in a whole may be considered ' separately,' provided it remains relatively the same when considered in itself or as a part of some other whole. (2) Accordingly the real difference between organic and inorganic categories is only that, in the sphere of the inorganic, that which can be considered (at the same or at different times) part of different wholes remains relatively the same, while, in the sphere of the organic, there is not the same relative identity but a very marked difference. And as parts and whole are correlative, the arguments which hold of the parts will also hold, mutatis mutandis, of the corresponding wholes. To postulate an additional factor, as vitalism does, is at least not logically necessary. It may be true that each organism acts as it does because it is impelled by some sort of conscious self. But, unless this factor can be proved to exist, it is not called for in order to explain the different behaviour of organic and inorganic in the phenomena of restitution, elementary morphogenesis, and the rest. But, it may be argued, there is a relevant differ- ence. Organisms can assimilate. Grass assimilates 148 PEOBLEMS OF THE SELF ohap. the inorganic soil, sheep assimilate grass, and men assimilate sheep. How is it possible that the same material when assimilated behaves so diflferently from the way it behaves when not assimilated ? If organic bodies were formed of a different kind of substance from inorganic things then the foregoing argument might be justified, but if they are not, and if, as we have reason to believe, every body is recruited from its environment in such wise that, after a certain interval of time, no part of the original substance of the organism remains, how is it possible that the parts of an organism should behave so very differently from the parts of inorganic things ? There is no mystery here if a single vital principle continues to actuate the organism. Unless it does so there is not merely mystery but miracle. But is there really any greater mystery here than in many other f^iCts which science accepts without question ? Is there really anything mysterious (in the sense of unusual or irrational) in the fact that entities forming part of a certain whole exhibit differ- ent characteristics from those which they show jas members of some other whole? If a silk handker- chief and a piece of amber are rubbed together, the result will be that electrical phenomena appear. But neither the amber nor the handkerchief is itself electrical, and yet there is electricity when they are connected in a certain way. It is not un- usual to talk of the latent electricity which appears on occasion of the rubbing of these two, but to talk in this way is not to think. And even if the theory were true it would not prove that a single vital principle animates each particular organ- ism, but only that when certain particles of m9,tter become connected in the form and fashion of an organic body there latent life will become actual. This difiiculty, accordingly, is not really serious. In short, although the behaviour of organisms is PSYCHICAL AND PURPOSLVE 149 very different from that of inorganic matter, and may very well require distinctive principles of explanation, it does not follow, in any way, that these new prin- ciples are absolutely heterogeneous from the old. They need not differ more from the old than, let us say, tropism differs from gravitation or electricity from the laws of motion. And if they do differ, if it is false to call the body, as Descartes did, ' cette machine compos^e d'os et de chair,' if they even, in sojne sense of that much-abused word, deserve to be called ' teleological,' it does not follow that this * teleology ' can only be explained by postulating the influence of a permanent psychical being. If such a being were really present neo-vitalism would not differ from vitalism except in so far as it expresses itself in a more exact and guarded manner. A dis- tinct new entity would be presupposed, viz. conscious- ness. But there is no sufficient ground for supposing that such an entity is always present where life is. The reasons for believing so rest on very vague analogy, and on analogy which fails precisely at the points where it ought to be most helpful. The ' psychoid ' is a mongrel which skilfully conceals any conscious strain. There is no evidence that it is conscious in any degree. (3) If these arguments are sound, it will follow that it is a mistake in principle to maintain that the existent universe contains two classes of beings, the mechanical and the purposive, if the word purpose implies anything that can truly be said to be con- sciousness. If we must subdivide in this way we should speak of the inorganic, the organic,^ and the conscious. The conscious, however, when it is found, is always found in alliance with an organism, and it ^ ' The organic,' in this sense, means the whole realm of living things. A distinction, however, may be drawn between their organisation and their organic constitution, A more accurate usage would therefore be to define ' organic ' as meaning carbon compounds, and to mention the characteristic of being organised when describing living beings. 150 PROBLEMS OF THE SELF chap. is therefore not unreasonable to suppose that con- sciousness is only a sub-class of life though not co-extensive with all life. We shall conclude this chapter by discussing the question. But we need not discuss it at great length, because we have already been discussing it frequently.' We were discussing it when we distinguished the self from the body, when we distinguished psychical endeavour from bodily sensa- tions which accompany it, when we showed that the central role of the body in action was only remotely analogous to that of feeling in the self. The nature of consciousness is to be a reference to an object. That is no part of the nature of the body. It is not, as such, conscious, and so far as it enters into relation with consciousness its being is to be an object for consciousness and an instrument of consciousness. The existence of the body may be a necessary con- dition for the existence of consciousness, but the nature of consciousness is fundamentally distinct from that of the body. Another distinction which might be mentioned is the all-important one that cognitive processes may be valid and volitions may be right (from the point of view of ethics), while there is no sense in attribut- ing validity or moral rightness to processes of the body. Any one except a pragmatist or a materialist must assent to this statement, and the pragmatist would only dissent because he mistakes the way in which a process may lead up to and terminate in another process for the true awareness of the meaning of that process. I do not say that this is a universal characteristic of all consciousness. Conscious logic is often faulty, conscious action often wicked. The laws of logic are not the ways in which men always think, or the laws of ethics the ways in which they always act. It is not necessary to agree with Descartes when he says, ' As to the Reason or Sense, inasmuch as it is that alone which constitutes us PSYCHICAL AND PURPOSIVE 151 men, and distinguishes us from the brutes, I am dis- posed to believe that it is to be found complete in every individual.' ^ But the fact that truth or falsity. Tightness or wrongness, are characteristic of some conscious processes though not of all, while they are never applicable to bodily processes, shows at least that some conscious processes are not bodily processes, and supplements the independent arguments which prove that the two are distinct. When we come, in a later chapter, to consider the unity and continuity of psychical processes we shall find still further confirmation of the distinctions we have drawn here. Meanwhile, at the risk of repeti- tion, I should like to point out the bearing of the argument of the present chapter on the general question of the primacy of will in the economy of the self. If it is maintained that the differentia of selfhood, at any level, is that it is purposive, and that the absence of purpose distinguishes mere matter from the rest of the universe, then it is not unnatural to conclude that this distinguishing feature is the real essence of any being which is not mere matter. And if it is held further that purpose is the spring of life, then, because the self is alive, it is easy to draw the inference that all experiences are simply various manifestations of purpose. The arguments of this chapter, if they are sound in any respect, will suffice to prove that this interpretation is needless ; and so we may pass to the consideration of some other arguments in favour of the primacy of wiU. ' Discourse on Method, Part I. CHAPTER VII THE PRIMACY OF THE PRACTICAL REASON The two previous chapters have shown that the proofs of the primacy of will are drawn from a variety of sources, and that the ambiguity of the word 'will' is partly responsible for this result. It may seem gratuitous, therefore, to introduce a phrase which is possibly still more ambiguous, i.e. the practical reason. When Schopenhauer maintains that the will is the thing in itself and everything else an appearance derived from it, he means by the will an insatiable impulse to life, something, in fact, very like Bergson's elan vital. This sense of will, whatever it is, is clearly different from volition in the narrower sense which is peculiarly relevant to ethical choice. On the other hand, when Kant tries to prove the primacy of the practical reason over the speculative he is referring to this narrower sense strictly. Closer inspection will show, however, that the difference is not so great as might be supposed, at least with regard to the way in which these authors define their terms. The wUl, for Kant, is synonymous with the practical reason. ' Everything in nature works according to laws. Rational beings alone have the faculty of acting according to the conception of laws, that is, according to principles, i.e. have a will. Since the deduction of actions from principles requires reason, the will is nothing but practical 152 CHAP. VII THE PRACTICAL EEASON 153 reason. ' ^ And Kant also says, ' The appetitive faculty is the faculty of being by means of one's ideas the cause of the objects of these ideas. The faculty which a being has of acting according to its ideas is life.' ^ It is strange that any one should have adopted a definition of life which implies that honeysuckle, being alive, is, through its ideas, the cause of the object of these ideas. But thus it was. And although causation through ideas is not identical with action according to the conception of law, still the gulf between them is not impassable. For the rest, I have chosen the title of this chapter in order to emphasise the precise nature of the subject considered in it. I wish to deal with the metaphysical arguments for the primacy of will, and I shall keep in mind principally the treatment of the question during the great constructive period of German philosophy. From this point of view it is most convenient to follow our authors in chronological order. The reader will remember the well-known saying that the history of philosophy is thought itself taking its time. The primacy of the practical reason is the keystone of the arch of the Kantian system, at least so far as that appears in the first two Critiques. Kant's thesis is that the practical reason has the first place in comparison with the speculative, and this assertion of the secondary importance of the intellect as compared with the will forms the thread of continuity between aU the arguments that call for mention in this chapter. Kant did not, indeed, contend that there are two distinct entities, the theoretical and the practical reason. On the contrary, he held that one and the same reason has both a practical and a theoretical use. But the practical use of reason has ' Fv/nda/mmital Prmdples of the Metaphysic of Morals, Abbott's translation, p. 29. 2 Introduction to the Metaphyaie of Morals, Abbott, p. 265. 154 PROBLEMS OF THE SELF chai-. the prerogative of being the first determining principle. In ethical choice we are confronted with a fact of pure reason, the fact that it is practical. To act rightly is to act in accordance with the conception of universality, and such action must be universal in a double sense, since it must apply to every rational being, and also in every possible circumstance. Any exception, in either sense, destroys the universality. Nor is it enough that the action should, in point of fact, be universal in these ways. A man may keep all the ten commandments from a wrong motive. The important point is that the agent should choose the act because of the universality. He must do it because he wills it to be universally practised ; and so it is permissible to argue that action is morally right if and so far as the concept of universality is the determining principle. When that occurs we have the ' fact of pure reason,' and the existence of such a fact must be admitted. True, it may be impos- sible to prove that any given action is wholly and completely right. Some empirical and, therefore, non-rational motives might have helped to determine it. Some alloy of self-interest might have entered, as in the cases when a man knows that doing his duty will also conduce to his own advantage. Such an act, according to Kant, would not be a perfect instance of ethical action. Universality would not harve been the sole determining principle. But it would clearly be absurd to argue that because we can never prove that a given action is completely disinterested, we should therefore conclude that action is never disinterested in any degree. We may, then, admit this fact of pure reason. Since the basis of Kant's argument is the fact of moral choice, it seems a clear duty to consider whether he really gives an accurate account of that fact. Fortunately, however, the most serious objections VII THE PRACTICAL REASON 155 refer only to the sufficiency of his account, and its insufficiency need not affect the validity of his further argument. It is true that his conception . of universality may require reinterpretation. There are no good grounds for maintaining that any action must be right for all agents in all circumstances. It is generally a duty not to commit suicide, but a community has a right to anticipate its fate at the hands of a savage foe if further resistance is hopeless, and capture would involve something worse than death. It may be A's duty to take a holiday after working all the summer. It need not be B's duty, for B may have had his holiday, and in that case there is a relevant difference. But if Kant mis- interpreted the universality implied in right action, the necessity of universality, rightly interpreted, remains unaffected. Right action is universal in the sense that if it is right for A to perform a particular action it is also right for B, unless there is a relevant difference in B or his circumstances. Right action, then, must be universal, but Kant was wrong in subordinating value (or goodness) to universality, although it is a misunderstanding of his position to censure the categorical imperative because it is ' formal.' Kant maintained explicitly that every act has its material or object.^ He did not treat -of willing in general, but of the willing of this or that. What he held was that the specifically moral element of an action lies in the fact that it was performed, not to attain some particular end,, but because of its universality. But any action can be universalised. It is consistent, whether or not it is psychologically possible, to choose to act so as to bring about the maximum amount of misery to ourselves and others. Universahty, then, is only one characteristic of a moral act, and value is the other. The act must be chosen ' Vide Critique of Practical Reason, Book I. chap, ii., 'Of the Concept of an Object of Pure Practical Reason.' 156 PEOBLEMS OF THE SELF chap. on account of its value as well as on account of its universality. This, however, is only a qualification, not a denial, of the ' fact of pure reason.' The universality, and probably the value also, imply the presence of reason, and -consequently the logical grounds of Kant's argument remain as they were. What, then, are these grounds ? The fact of pure reason, Kant contends, means that pure reason deter- mines action in time or, negatively, that there is freedom, since freedom implies the initiation of an action in time in a way that is not entirely explicable in terms of preceding events in time. Pure reason be- longs to the intelligible world, it is the thing in itself, it is a noiimenon. Events in time are phenomena be- longing to the sensible world. How is it possible that the noiimenal world or the world of pure reason, the ' intelligible character,' can determine the sensible, especially in view of two difiBculties : (1) that cause and effect, strictly speaking, refer to phenomena only ; and (2) that they hold universally of the phenomenal world, so that ' freedom,' in the sense defined above, seems impossible. How can these two worlds unite together in the mysterious bond of the moral judgment? The Critique of Pure Reason had shown that pheno- mena are the only things which can be known as objects. The intelligible world can, therefore, never be known as an object. But the fact of freedom (to which we have already referred) makes the Practical Reason give a clue to the explanation of the world which the speculative reason by itself could not furnish. The speculative reason shows that freedom (or, positively, determination by the noiimenal) is possible, but cannot show that it is actual. The fact of moral freedom, or of right action, does more than this. It shows that such determination is actual. The practical reason, therefore, has primacy over the theoretical inasmuch as it goes further than vii THE PKACTICAL REASON 157 the speculative. It gives us fuller light upon the nature of the universe and, in particular, upon that portion of the universe which we call ourselves. The argument is not an easy one to follow, and the principal difficulty attaching to it is the very perplexing problem of the precise sense in which the possibility of noiimenal determination (acknowledged by the speculative reason) is relevant to particular cases of moral choice ; and, clearly, the performance of a man's duty is always specific and determinate. Many of Kant's critics, indeed, maintain that the speculative reason (on Kant's definition of it) cannot logically admit even the possibility of freedom, but Kant's consistency, on this head, may be readily defended. While we cannot know what nolimena are or, in other words, can never demonstrate the precise nature of the ' intelhgible character,' we can know that they are. We can prove that phenomena are conditioned, and therefore know that they have conditions, although the intrinsic nature of these con- ditions baffles the understanding. And so we have not merely the noumenon, but the causa noiimenon. Noiimena are the ground^ of phenomena, and determine the character of the existence of phenomena, although we do not, and could not, know the precise manner in which they do this. There is even a certain analogy between the determination of noiimena and empirical causality. For the noumenon is the ground of the phenomenon, determining how it occurs. It is not strictly an antecedent, since it is not in time, but it is a condition. And, in this instance, one term at least is phenomenal, namely, that which is determined. In causation, in the strict sense, both cause and effect are phenomena. It is possible, then (indeed we may have reason to believe), that the intelligible determines the pheno- menal. We cannot tell how it does so, but we may ' Cf. Critique of Judgment, Bernard's translation, p. 39, note. 158 PROBLEMS OF THE SELF know that it does so. But it is much more difficult to see how we can know that this determination makes a special difference in particular cases. That the whole realm of phenomeija (so far as it can be called a whole) should be determined by noiimenal conditions does not affect particular causal laws. By their means we spell out phenomena and phenomena only : the noiimenal determination is a general con- dition of all phenomenal connection and consequently is irrelevant to the special connections between special classes of phenomena. But when we come to freedom all seems changed. Kant has clearly particular cases in his mind ; ^ he speaks of the act of rising from one's chair ; and, as we have seen, no other sense of freedom would be relevant to morals. It would seem, then, that the specifically ethical facts upon which Kant bases his doctrine of the primacy of the practical reason are the weakest part of the foundation of his system. At the same time it is not unintelligible that, although all phenomena are determined by the intelligible, some phenomena should show this determination more clearly than others ; and a man's acts in the concrete, so far as determined by con- siderations of duty, might possibly be of this kind. Indeed, a possibility of this sort may obviate an objection which, doubtless, has already occurred to the reader. The objection is that there is never any meaning in saying that reason determines anything, determines it to act, that is to say, or causes any event to be what it is. An act of volition may cause a subsequent movement or train of thought, and this act may choose to follow the dictates of reason. But reason, in the sense of the intelligible connection, which is the ground of the choice, does not determine at all. The intelligible connection is there, whether we are aware of it or not, whether we choose it or not, and if it is a non-temporal connection, as Kant ' Vide the Them of the Third Antinomy. VII THE PEACTICAL REASON 159 supposes it to be, it cannot determine at one time rather than at another. But this is exactly what the actual volition does. The volition occurs in time and determines subsequent action. All causation, therefore, even that involved in right action, occurs at definite points in time and is therefore, on Kant's theory, phenomenal and not noiimenal. But, even granting this, it is clear that reason, on Kant's conception of right action, would determine such action in a way in which it does not determine other actions, e.g. the actions of a somnambulist. It is quite true that this intelligible connection, this universality, this reason, cannot be said to be a cause in the ordinary sense of the words. But Kant never maintained that it was a cause in the ordinary sense. On the other hand, reason does determine. If we choose a certain course because it is rational, then the rationality is a condition of our choice. The choice is a conscious choice and refers to the rationality of that which is chosen. Unless the choice referred to this rationality, and unless the rationality were the reason for the existence of the choice, that particular choice would not occur. There is sense, therefore, in maintaining that reason determines the choice of right action. And it is theoretically possible that such action could be explained by psychological rules of cause and efiiect, while at the same time it would be true that some particular psychical processes had also the quality of validity (ethical or speculative), and therefore were determined by reason. I cannot see, however, that this admission proves the primacy of the practical reason. For, in this sense of the word, reason determines true belief just aS surely as it determines right choice, and, indeed, the true belief that such and such an action can be universalised is a necessary pre-requisite of the, right choice itself. The right choice may certainly deter- mine our behaviour with respect to beings other than 160 PROBLEMS OF THE SELF chap. ourselves. It will make a difference to the rest of the universe, and the true belief (apart, of course, from its expression) will not. But so far as the question of determination by reason goes, belief and choice are on the same level, and if there is any primacy of volition in the life of the self the reasons for that primacy are only those reasons for the primacy of the experience of activity which we dis- cussed in the fifth chapter. On the whole, then, Kant has failed to prove that the practical use of reason has primacy over the theoretical. If his aim were to discover indisputable evidence of the way in which reason determines the life of the self, he would find this evidence as clearly in true knowledge as in right action. And even if reason does appear in a purer form in right action than in right thinking, since reason dictates in morals while in knowledge it only interprets and is indis- solubly linked with sense, that in itself would not prove the primacy of will. The fact of morality would give us more striking evidence than the fact of true knowledge, but it would not prove that the practical use of the reason is in fact constitutive in a sense in which the theoretical is not. It only proves that we cannot obtain our evidence as clearly in the latter case as in the former. It is possible, however, that the Critique of Judgment throws further light on the question, from the point of view of will in the broad sense of purpose if not from the narrower one of ethical decision, and in view of Kant's relation to some of his successors, it is necessary to indicate, by means of quotations, the precise sense of the terms he uses. ' The Will, regarded as the faculty of desire, is, in fact, one of the many natural causes in the world, viz. that cause which acts in accordance with concepts. All that is represented as possible (or necessary) by means of a will is called practically possible (or THE PRACTICAL REASON 161 necessary) ; as distinguished from the physical possi- bility or necessity of an eflfect, whose cause is not determined to causality by concepts (but in lifeless matter by mechanism and in animals by instinct). Here in respect of the practical it is left undetermined whether the concept which gives the rule to the causality of the will, is a natural concept or a con- cept of freedom,' ^ Kant proceeds to argue that the two realms of Understanding and Reason {i.e. our knowledge of phenomena and of freedom) do not conflict with one another, but that each plays its distinctive part. In saying so he merely restates the conclusions of his previous Metaphysic, but he enters on new territory when he explains why, for the purposes of our knowledge, the two realms are not really one whole. ' That they do not constitute one realm, arises from this, that the natural concept represents its objects in intuition, not as things in themselves, but as mere phenomena ; the concept of freedom, on the other hand, represents in its Object a thing in itself, but not in intuition.' '^ ' There must, therefore, be a ground of the unity of the supersensible, which lies at the basis of nature, with that which the concept of freedom practically contains ; and the concept of this ground, although it does not attain either theoretically or practically to a knowledge of the same, and hence has no peculiar realm, nevertheless makes possible the transition from the mode of thought according to the principles of the one to that according to the principles of the other.' ^ The aim of the Critique of Judgment is, to find the bridge which spans these two worlds. Accordingly, it is easy to see, if we consider the question carefully, that the Critique of Judgment, despite its insistence upon teleology and purpose, can only throw light indirectly on the question of the ^ Critiqiie of Judgment, Bernard's ti-analation, p. 7. 2 Ibid. pp. 11, 12. 5 Ibid. pp. 12, 13. M 162 PROBLEMS OF THE SELF chap. primacy of will. It does not seek to prove that primacy but, on the contrary, presupposes it. The causa noiimenon, freedom, does determine the causa phenomenon, mechanism and desire. The question is only how any relationship is possible between things so disparate as the objects of sense and the objects of reason ? How can phenomenal nature be adapted to intelligible nature ? And Kant's answer is that some of our experiences serve to show, subjectively at least, that such a harmony occurs in fact, and therefore it is reasonable to assume that the general characteristics of this subjective harmony may furnish a clue to the general question of the relation of the sensible to the supersensible. In aesthetic judgments the beauty of the object contemplated spells a harmony between the mere form of that object and our sub- jective faculties ; and the purposiveness of nature, as seen especially in the adaptation of living things to their environment, seems analogous to the harmony between our purposes and our environment. Or we may put the question in another form. Granting that the practical reason has the primacy in the determination of the nature of the cosmos, have we any analogies in experience for this unity ? Is there any intuition which is adequate in any respect to symbolise the whole? The answer is that some intuitions may be adequate in some particulars, though all are imperfect. ' In his discussion of the organic realm Kant introduces many of the arguments which the writings of the neo-vitalists have made familiar to us, and he is much more acutely aware of the philosophical significance of his arguments than they usually are. To illustrate this I shall quote a passage of some length. ' In a watch one part is the instrument for moving the other parts, but the wheel is not the effective cause for the production of the others ; no doubt one part is for the sake of the others, but it THE PRACTICAL REASON 163 does not exist by their means. . . . Hence a watch wheel does not produce other wheels, still less does one watch produce other watches, utilising foreign material for that purpose ; hence it does not replace of itself parts of which it has been deprived, nor does it make good what is lacking in a first formation by the addition of the missing parts, nor if it has gone out of order does it repair itself — all of which, on the contrary, we may expect from organised nature. An organised being is, then, not a mere machine, for that has merely moving power, but it possesses in itself formative power of a self-propagating kind which it communicates to its materials though they have it not of themselves ; it organises them, in fact, and this cannot be explained by the mere mechanical faculty of motion. We say of nature and its faculty in organised products far too little if we describe it as an analogon of art ; for this suggests an artificer external to it. Much rather does it organise itself and its organised products in every species, no doubt after one general pattern but yet with suitable deviations, which self-preservation demands according to circumstances. We perhaps approach nearer to this inscrutable property, if we describe it as an analogon of life ; but then we must either endow matter, as mere matter, with a property which contradicts its very being, or associate therewith an alien principle standing in communion with it (a soul). But in the latter case we must, if such a product is to be a natural product, either presuppose organised matter as the instrument of that soul, which does not make the soul a whit more com- prehensible, or regard the soul as artificer of this structure, and so remove the product from (corporeal) nature. To speak strictly, then, the organisation of nature has in it nothing analogous to the causality we know.' ^ 1 Critique ofJudgrmnt, Bernard's translation, pp. 278-279. 164 PROBLEMS OF THE SELF chap. The latter part of this quotation shows that Kant was fully aware that the aims of explanation are not served at all by postulating a teleological factor (or a soul) to explain the behaviour of living organisms, while the former part gives an admirable resum^ of the type of argument which we have discussed in the preceding chapter. It is probable, I think, that the conclusions of the Critique of Judgment are too weak for the weight of evidence which he brought to bear. The difficulties of understanding the purposiveness of organic life — this purposiveness without a purpose — are so great that our intellects cannot cope with them. 'By the constitution and the principles of our cognitive faculty we can think of nature, in its purposive arrangements which have become known to us, in no other way than as the product of an Understanding to which it is subject.' ^ Purposive- ness in Nature (which, for Kant, means adaptation), whether specific or general, can only be thought, he maintains, according to the eflfete principles of the argument from design, and these are necessarily inadequate. Hence the moral proof of the existence of God is ultimately the only one in which our reason can rest. There is no way, in the end, in which we can adequately represent to ourselves the determina- tion of the sensible realm by the supersensible. The facts of morality are our only ground for believing in this determination, and we cannot, even in this case, know how the determination is realised. It follows therefore that the basis of the primacy of will, if there be such a basis or such a primacy, is to be found in the Critique of Practical Reason and not in the Critique of Judgment. I have introduced the question partly because it explains the historical origin of certain post-Kantian arguments in favour of the primacy of will, and partly because the argu- ments from purposiveness in nature, as we have ' Critique of Judgment, Bernard's translation, p. 369. (Italics mine.) THE PRACTICAL REASON 165 already seen in the preceding chapter, seem relevant to the issue. There is no warrant in Kant himself for supposing so. He is positive in maintaining that instinct at least is unconscious, and therefore no part of the self or of the will. As we have seen, however, his definitions of life and of desire seem very closely connected with his definition of will ; and so it may be true that if Kant had fully appreciated the force of his own arguments he would have remodelled his position and accepted much which he in fact rejected. Schopenhauer, for instance, accuses the Critique of Judgment, not of error, but of incompleteness, and he criticises this incompleteness so tartly that we cannot help suspecting the presence of motives of personal jealousy. ' In the Critique of the Teleo- logical Judgment, on account of the simplicity of the matter, we can recognise perhaps more than anywhere else Kant's rare talent of turning a thought this way and that way, and expressing it in a multitude of difierent ways, until out of it there grows a book.' ^ The work, he continues, is incom- plete, especially because it fails to seek for a principle which ' would recognise both in the mechanical (accordipg to law) and the apparently intentional efi"ects of nature one and, the same ultimate principle, which might serve as the more general ground of explanation of them both. Such a principle I hope I have given by establishing the will as a real thing in itself; and in accordance with it . . . especially in my work On the Will in Nature, the insight into the inner nature of the apparent design and of the harmony and agreement of the whole of nature has perhaps become clearer and deeper.'^ Schopenhauer, like the Psalmist, is wiser than his teachers. He was not, however, the only successor of Kant who believed in the primacy of will. An earlier, and ' The World as Will and Idea, English translation, vol. ii. p. 156. 2 Ibid. pp. 158-159. 166 PEOBLEMS OF THE SELF chap. a greater, was Fichte, who sought to show that the being of knowledge itself depends upon activity. His metaphysic, certainly, is very different from Schopen- hauer's. Indeed, it seems a historical travesty to link their names together, since Schopenhauer rarely missed an opportunity for sneering at Fichte, and even at Fichte's personal honesty. But both contended for the primacy of will, although on quite diverse grounds, and therefore they agree verbally at all events. It need not surprise us if we find that some Voluntarists agree in no other way. The best short statement of Fichte's views is found in his two introductions to the Wissenschaftslehre} These views are especially interesting because they purport to be an exact philosophical interpretation of the self. Fichte did not mean the individual self. He did not mean, as Heine unkindly suggested, the particular ego sailed Johann Gottlieb Fichte. But the self (more accurately self-hood) must be the basis of every idealistic system ; and every system which is not idealistic must fail inevitably, according to Fichte, because it cannot do justice to the reality of self. It soon finds itself forced to degrade the self to the status of an accident of the world. A philosophy which does not begin with the self can never reach it. A philosophy which begins with it can reach everything else. The Ich an sich must replace the Ding an sich ; and the proof lies in the completion of the idealistic system. Now the fundamental fact in the life of the self is the fact of freedom. It is true that we are conscious of a multitude of ideas, which are subject to the law of causality, and accompanied by the feeling of necessity. These we call Experience or, rather, Erfahrung. But it is the business of every philosophy to stand above experience in the sense of Erfahrung. It seeks to make Erfahrung intelligible, and exhibit its » Published iu 1797. vu THE PKACTICAL REASON 167 real ground. This ground ia the absolute spontaneity and freedom of the self. ' Intelligence as such is aware of itself, and this vision of itself is directly united to everything which enters into it ; and the nature of intelligence consists in this unmediated union of being and vision. Whatever is in intelli- gence and, in general, the very essence of intelligence, is to be for intelligence, and intelligence is only in- telligence in so far as it is, in this way, for itself I think this or the other object ; but what is this object, and how do I appear to myself in this act of think- ing ? Only in one way : I bring some determinations of myself into prominence when the object is a mere fabrication ; or they are present without my aid when the object is reality ; and I am aware of thus bring- ing them forward or of this being. The objects are in me only in so far as they enter into my self- cognition ; self-cognition (Zusehen) and being are indivisibly united.' ^ The original fact of the self — or rather, as we shall see in a moment, the original act thereof — is some- thing still more ultimate than self-consciousness, in the sense of Selbstbewiisstsein. The original datum is the return of the ego into itself, ' It is through this act first of all, and through it only, through an act upon an act which does not presuppose any anterior act, that the ego is originally for itself. It is only for the philosopher that it is present as a fact because he has already constructed the whole of ex- perience.'^ But what is the nature of this return upon itself ? ' It is not conceptual knowledge, for that comes into being with the opposition to a not- self, and with the determination of the ego through this opposition. It is therefore mere intuition.' * This original act is called by Fichte intellectual intuition, and he regards it as the true and proper ' Erste EinleiUmg, p. 435. 2 Zweite Einleitung, p. 409. ' Ibid. p. 459. 168 PROBLEMS OF THE SELF interpretation of Kant's unity of apperception. But although the phrase suggests the primacy of intellect rather than that of will, Fichte considers his doctrine akin to the practical, not to the theoretical reason. The self is first of all an act, and by its action it comes to know. The modern reader will find an interesting parallel between Fichte and Bergson in this connection, not merely with regard to the general dependence of intellect upon activity but even with regard to details. Let us take this passage : ' I should like to know how those who in mentioning intellectual intuition adopt the fashionable view of its nature think of the consciousness of ethical laws, or how they could construct the notions of right or of virtue which they doubtless possess. According to them there are two kinds of a priori intuition, Time and Space, and without a doubt they construct these notions in Time, the form of the inner sense : but the notions are clearly, on their view, not Time itself but a certain filling of Time. But what is this filling of Time which underlies their construction ? Nothing remains for them except space, and so their " right " must apparently be something rectangular and their virtue round like a circle, just as every concept of sense intuition which they construct (as for example a tree, or an animal, or the like) are nothing but limitations of space. Accordingly they cannot really think of right or virtue at all. What then is the (real) ground of their construction ? If they pay attention they would see that it consists of action or of freedom.' ^ In this passage there is an emphasis upon ethics which is not found in Bergson, but both agree in their insistence upon activity and in their denial that space by itself can be the true basis of reality. If Bergson is translated into terms of ethics, his argument reads as if it were Fichte's, a fact which ' Zweite Einleitung, p. 467.- THE PRACTICAL REASON 169 is not surprising when we remember the relation of both of them to Kant. The doctrine that the intellect is especially concerned with space and time in the sense in which these are employed in physics depends, directly, as in Fichte, or indirectly, as probably in Bergson,^ on that marriage between the categories and the forms of sensibility in the Under- standing on which Kant insists so pertinaciously. But life is wider than Ethics, and therefore, per- haps, Bergson is wiser than Fichte. Let us, however, consider Fichte's exposition of his fundamental position in fuller detail. As it is impossible to state Fichte's position more lucidly than Fichte himself, I shall content myself with giving two somewhat lengthy quotations from him : ' Now this whole procedure of the philosopher seems, to me at least, very feasible, very easy, very natural, and I can hardly see how it could appear otherwise to my readers, or how they could find any- thing strange or mysterious in it. It is to be hoped that each of them can think himself. It is to be hoped that he will become aware when he has come to this thought that he has come to something dependent upon his own activity, that is to say, that he will act. It is to be hoped that he will be able to distinguish this activity from that which he sets against himself when he considers objects outside him, and to find that in the latter the thinking and that which is thought are opposed to one another, so that his activity must go towards something different from himself, whereas in the former connection the thinking and that which is thought are one and the same, and therefore his activity returns inward upon itself. It is to be hoped that he will understand that, since the thought of himself arises in this way and in this way only, and because, as he finds, quite ' Cf. A. Lovejoy, 'Some Antecedents of the Philosophy of Bergson,' Mind, N.S. No. 88, pp. 465 ff. 170 PEOBLEMS OF THE SELF chap. a different thouglit arises when lie thinks of an object over against him — he will find that the thought of hinaself is nothing else than the thought of this activity, and the word self nothing else than the sign thereof : that the self and an activity which returns upon itself are identical concepts. It is to be hoped that he will comprehend this, even if he presupposes only problematically, as transcendental idealism does, that all consciousness depends upon self-consciousness and is conditioned thereby. That is a presupposition which he must make in any case, as surely as he but turns an attentive glance upon himself and rises to the level at which he requires a philosophy ; and its validity will be shown him categorically in philosophy itself through a complete deduction of the whole of experience from the possibility of self- consciousness. Accordingly he must think of this return upon hinaself as the presupposition of any other act of consciousness, as the condition thereof, or, which is the same thing, he must think that return upon itself as the most original act of the subjects. The reason is that there can be nothing for him that is not in his consciousness, for anything else in that consciousness is conditioned by this act itself, and therefore cannot condition it again in the same connection. It is, therefore, for him, and entirely unconditioned, and thus an absolute act. Accordingly this presupposition, and this thought of the self as originally conditioned through itself, are absolutely identical.' ^ The second quotation, which is happily shorter, is chosen for the purpose of making this activity which returns upon itself a little more intelligible. Fichte himself admits that it is something too primitive (though not in a temporal sense) and too fundamental to be adequately explained by intellectual categories. These refer to an object, and thus presuppose a stage ^ Ersle EinleUung, pp. 461, 462. vu THE PRACTICAL REASON 171 of thought which is derived from, and therefore not identical with, the original activity of self-conscious- ness. The reader who is to understand Fichte must put himself freely into Fiehte's point of view, and then the vision will be vouchsafed him. But certain descriptive phrases, however inadequate, and however cumbered by the implications of language, may set him upon the right track and stimulate him to exercise his freedom. The most telling passage occurs in the first introduction : ' Idealism explains, as has already been shown, the determinations of consciousness through the activity of intelligence. That on the theory is active and absolute, not passive. It could not be passive since it is the first and highest ground of what it postulates, and there is nothing anterior through whose influence it could be said to be passive. For the same reason no specific being or existence can be ascribed to this activity, since that is the result of an interplay of forces, and in the present instance there is nothing, and nothing can be supposed, with which the intelligence can be supposed to interact. Intelligence is for idealism a doing (Thun) and absolutely nothing more. It should not even be called an active thing (ein Thdtiges) since this latter expression implies a reference to some species of existence in which the activity dwells. . . . Accordingly this is the pre- supposition of idealism. Intelligence acts, but, because of its o? etc., we should naturally suppose that there is a ' one-one ' correspondence between the S, a, and 9 series respectively. If we examine the matter, however, we shall probably find that the intensities 9i sa ?8 are indistinguishable, while ?4 is recognised to be different from 9i ; and yet if we began with 93, we should judge it indistinguishable from 94 and should have to proceed to 95 or 95 before we recognised a difference. It is held that the best explanation of this fact is that 9i 92, etc., are all really different and they, of course, are present to consciousness. There are, therefore, conscious differ- ences not explicitly recognised as such. And this is the meaning of subconsciousness. Take, again, the analysis of sense-perception. It has been abundantly proved that in estimating the size and the distance of objects, to mention no other characteristics, we rely upon a multitude of signs. The plain man makes his estimate unhesitatingly and, as it were, instinctively. He does not know what signs he is using until they are pointed out to him. Is not this a clear instance of subconsciousness. 262 PEOBLEMS OF THE SELF chap. and does it not bear very directly upon tlie unity and continuity of the self? And so with our feelings. In one sense of the words feeling must be what it appears, for its being is to appear. But it is possible to have a feeling without being aware of having it. It is often clear to any observer that such and such an one is in love. But the lover himself may not be aware of the state of his feelings. He may even be ignorant of them until his opportunity has passed. But when on this, or some other, occasion the true state of the facts comes to his knowledge, he may recognise not only that he is in love but that he has heen m love all the time. In the same way our organic sensations are always with us, and in pain or in intense excite- ment even of a pleasurable sort, we are acutely conscious of them. But normally we neglect them. There must be some sudden change before we think of them at all* And the fact that we are aware of the change shows a continuity of consciousness between the new and the old. What is more, we can frequently remember our previous feelings and compare them with the present. We ignored them at the time they occurred, but they were not non-existent for us. We were aware of them subconsciously. The same account holds of other similar instances, of the miller who notices that his mill-wheel has stopped, though, from familiarity, he does not hear it when it is going ; and so forth. » The instances, indeed, might be multiplied in- definitely. Consider the case of the nurse or mother who sleeps through thunder or lightning but wakes at the slightest cry from the child. Every burglar knows that even a sleeping household has its dangers, for a stealthy step during the night is much more likely to arouse the household than louder noises which are familiar. The meaning of sounds is not entirely lost on the sleepers, and therefore we must suppose that they are subconsciously intelligent HOW IS THIS UNITY POSSIBLE? 263 although not consciously so. I shall, however, content myself with mentioning only one other type of instance, and thife is one familiar to the student of Greek philosophy. Let us consider the simile of the aviary in the Theaetetus. In what sense precisely can we be said to have knowledge ? We have it, of course, when we actually make a true judgment, but that meaning is much too narrow to satisfy the ordinary usage of terms. We have it also when it is at our command. The candidate for examination knows all the proofs of Euclid's propositions, because he can demonstrate any one of them on command. But he never consciously knows them all together. And, again, we have the kind of knowledge which is illustrated by the well-known saying that it is not necessary for a man to know the classics. It is enough if he has forgotten them. He has forgotten them in their details, but the classical department of his mind, so to speak, is not an utter void. His thinking is tinged by half-forgotten recollections and vistas subconsciously recognised. It may be too much to say with Hamilton that 'the greater part of our spiritual treasures' is thus present to us in subconsciousness. But the statement contains much that is true. Speaking generally, we may say that continuity is the beginning, middle, and end of these arguments. They have, however, very different degrees of cogency. If, for instance, we consider the facts of acquired dexterity, we shall see that it is possible that the details of the process may become physiological re- flexes although some general conscious control is also required. The fact that conscious attention had to be paid to each detail in the process of learning surely does not clinch the raatter. There may be con- tinuity, even if some processes, originally conscious, can be relegated to physiological functions. In the same way the psycho - physical arguments of the 264 PEQBLEMS OP THE SELF chap. Weber-Fechner tradition are not conclusive. There are many alternative explanations of the facts. For instance, the argument, as outlined above, does not pay sufficient attention to the o- series. There is no doubt that if the S series and the ? series were the only relevant facts, we should suppose it most probable that the differences in the 2 series corre- sponded precisely to differences in the ? series although we were not always conscious of the fact. But all is changed when we remember the a series. That is certainly the direct condition of the 9 series, but it is in no way necessary to believe in a precise correspond- ence, term for term, between the % series and the a- series. Something may be lost in the conduction from the peripheral centres to the brain. The brain may not be delicate enough to adjust itself to slight differences, and need not always react in the same way to the same stimuli. The stimuli need not be the only conditions which count, and the effect of the summa- tion of previous stimuli may be very important in each instance. It is therefore possible that Si S2 S3 should all yield a o- which is identical in all three cases {i.e. o"! = 0-2 = 0-3). There may be distinct gaps in the o- series, and the reason why 2^ has a different physio- logical affect from ^3 need not depend entirely upon the difference between Sg and ^4, but may also de- pend upon the cumulative effect of 2i and Ss. If so, it would not follow that if we began our series of trials with S3, we should judge that ?4 was different from 93. In the absence of the cumulative effect before mentioned, we might have to proceed to S5 or Se before we could discriminate the difference. Again, the argument that we can be in love with- out knowing that we are in love seems, at first sight, singularly irrelevant. It is a fact, of course, that we may have the actual feelings of a lover, without knowing, by introspection, that we have them, or what they mean ; but, if that is all, subconsciousness HOW IS THIS UNITY POSSIBLE? 265 presents no diflBculty whatever. So far from being a contradiction in terms (or, perhaps, in one term), it is merely a truism. To have an experience without reflecting upon it introspectively is the normal lot of most of us. The theory, consequently, must mean something more than that. It is sometimes put in the form that we have experiences of which we could not be aware by introspection, however hard we try ; biit that, in its turn, seems to make the evidence for subconsciousness merely the fact of continuity. The fact is admitted, and the desirability of an explana- tion in psychical terms. But if there were no direct evidence over and above these indirect reasonings the theory would not be entitled to particular respect. The probative value of instantiae praerogativae, like Leibniz's petites perceptions, has already been seriously impaired by our discussion of the summa- tion of stimuli. Aid the conclusions of psycho- analysis may be challenged. Psycho-analytic pro- cedure assumes the existence of subconsciousness so readily and so frequently that doubts may be raised whether it has sifted the grounds of the assumption with sufficient care. But, when we reflect, we shall see that there is some direct evidence which it is legitimate to extend by analogy. There is always an unexhausted margin in the field of consciousness at any moment, and this margin is commonly neglected, but it is consciously present, in some degree. When I say it is consciously present I do not mean merely that it afiiects the course of consciousness ; for many conditions which are not conscious at all may do that. I mean that it is consciously present without being discriminated, and present in such a way that when we do come to discriminate it, we know that we are not dealing with something wholly new. We have experience of this at every moment in which we come to focus what was previously on the periphery of vision. We may 266 PEOBLEMS OF THE SELF chap. look at some given object, but we cannot perceive it except as part of a setting, and this setting, in addition to its obscurity, has no precise bounds. At_ this moment I am looking at the words I am writing and am not noticing the books in front of me or the window at my side. But I recognise, on reflection, that I was not entirely unconscious of them. And, as I have said, there is no definite boundary to this field of vision. I know that I can see nothing above me or, behind me if I keep my eyes in their present position, but, however hard I try, I could not tell the precise point at which this ' above ' or ' behind ' begins. We are always conscious of more than what we notice, and this marginal consciousness, which is a fact, has very difierent degrees. I have been speaking in terms of cognition, but the argument holds for other experi- ences also. This margiu of consciousness has found such frequent and such emphatic emphasis in modern psychology that it is needless to illustrate it further. It supplies us with direct evidence of the existence of subconscious elements in a conscious state. More- over, it is connected with continuity. Any state of consciousness throws out subconscious feelers, and the continuity is recognised when part of the margin is made the centre of a new conscious act. Moreover, the margin, which is apprehended with very various degrees of clearness, is inexhaustible in the sum-total of its details, and hence the subconscious is in some degree opaque to the most minute introspection. And, consequently, we are justified in extending sub- consciousness beyond the direct evidence for it, and even in making it at least part of the explanation of the facts of continuity which have been mentioned. We have seen no evidence for the existence of a sub- conscious self, but we have seen evidence for the existence of subconscious elements in every conscious act. Every given act may be retained, and part of X HOW IS THIS UNITY POSSIBLE? 267 the retention is always subconscious. There is no contradiction in believing that there may be literal persistence of consciousness in a subconscious form. On the other hand, it is clear that we ought not to extend the range of subconsciousness indefinitely ; and that is an error which is not infrequent. There is no justification for believing that everything which tells on conscious life is subconsciously present to that life.^ The conception of subconsciousness has frequently proved itself, in James's words, a 'sovereign means for beheving what one likes in psychology, and of turning what might become a science into a tumbling-ground for whimsies.' ^ If the margin has no precise bounds it is not therefore unUmited : if its contents are, in a way, inexhaustible they are not therefore co- extensive with reality. The soul has not windows for everything, nor is it necessarily aware in any sense of all the implications of the things it knows. There are really three questions to ask in the present connection. The first is whether subconscious elements are present in any of our experiences ; and the a,nswer is that such elements are always present. The second is whether such elements may persist continuously, and conscious elements persist subconsciously ; and the answer is that they may. The third is whether the whole of the unity of conscious Hfe can be so explained. This may be possible, but carries us far beyond the available evidence. It is begging the question to say that revival, retentiveness and memory must be due to subconscious persistence, and it is very difiicult indeed to believe that nothing we have known or done can ever fade from us entirely. They may all leave traces, but need they persist in a bodily, or rather in a ghostly, form ? We cannot disprove the ' Unless the term ' subconsciousness ' be used in a much more general sense than in this chapter, as e.g. by Morton Prince. Vide his recent work, The Unconscious, p. ix. and passim. ^ Principles of Psychology, vol. i. p. 163. 268 PROBLEMS OF THE SELF chap. theory that subconsciousness is the sole basis of retentiveness. The evidence makes that possible. But we need not believe that it is, and we shall be wiser if we withhold our assent. There is a gulf between consciousness (including subconsciousness) and the unconscious, and we cannot make them con- tinuous although we know they are not disconnected.' It is necessary to make one further explanation with regard to the interpretation of subconsciousness. When the threshold of consciousness is mentioned, it is usual to speak of presentations as rising above or falling below this threshold, and, as I have hinted in a previous place, it is tempting to suppose that retentiveness is due to the persistence of presentations and their mutual influence, one on the other. If presentations were entities neither physical nor mental but somehow the joint product of both, then the facts might be explained in this way even without sub- consciousness. These presentations would owe their inception, in part, to a particular mind, and then, persisting, they would be there to be perceived at will, and would give mind its cue. But it is very hard to believe in the existence of presentations in this sense, and the manner of their origin would be unintelligible. Accordingly, despite the difficulties, we should strive to consider a presentation as merely the ' material ' of cognition in the technical sense I have explained. In that case a presentation might be a real thing incompletely apprehended and perhaps distorted, or a universal thus apprehended, and then it would be the acts of mind that are subconscious, and their objects would be subconsciously appre- hended. This theory makes no difference to the facts, for the acts are always correlative to their objects, and can only be distinguished through them ; and when we speak of retentiveness we naturally think of what is retained, i.e. the presentations. If the acts persist, the persistence of the objects of these HOW IS THIS UNITY POSSIBLE? 269 acts, in precisely the sense which the acts mean and intend, is, eo ipso, intelligible, while the persistence of the presentations themselves, as presentations, is not intelligible. The complexity of mind at any moment demands subconsciousness. The pulse of consciousness, throwing feelers into the past and the future, is infinitely rich and contains far more than the clear and determinate acts which a hasty survey would suggest. I have spoken already of the dangers of magnify- ing the office of subconsciousness and should like, in conclusion, to call the ?:ea(ler's attention to a possibility he may have overlooked ; and because this possibility may seem strange and unusual I shall begin by seeking the protection of authority. The youthful Berkeley pondered for long over the debate between Locke and the Cartesians whether the soul thinks always. He could not be content with Locke's naive assertion that ' every drowsy nod ' overthrows the contention of his opponents. And so he wrote in his Commonplace Book : ' Locke seems to be mistaken when he says thought is not essential to the mind. Certainly the mind always and constantly thinks : and we know this too. In sleep and trances the mind exists not — there is no time, no succession of ideas. To say the mind exists without thinking is a contradiction, nonsense, nothing.' ^ And Lotze, at the conclusion of his Metaphysic, makes a similar reflection : ' Thus we have not scrupled, any more than any psychology has so far scrupled, to use the supposition of unconscious ideas, or unconscious ■ states, which ideas left behind, and which become ideas again. , . . There was nothing to compel us to th^se suppositions but the observed fact that previous ideas return into consciousness : but is there no other way in which that which once was can be the determining ground of that which will be, except 1 'Berkeley's Works, Fraser's 4-vol. edition, vol. i. p. 34. Italics Berkeley's. 270 PEOBLEMS OF THE SELF by continuing to be instead of passing away ? And if the soul in a perfectly dreamless sleep thinks, feels, and wills nothing, is the soul then at all, and if it is, what is it ? How often has the answer been given, that if this could happen, the soul would have no being ! Why have we not the courage to say that as often as this happens the soul is not ? ' ^ ' In sleep and trances the mind exists not ' ; ' Why have we not the courage to say that, as often as this happens, the soul is not ? ' These reflections,' surely, indicate a possibility that is too little regarded. Let us grant that the content of the self extends beyond any experiences which can be scrutinised in detail. Even so, perhaps, Lotze and Berkeley were right. The substantiality of a physical thing, in the current acceptation, implies continuity throughout every moment of tiihe. We believe, and are probably right in believing, that the smallest temporal gap in its existence would annul its identity. But must the identity of the self be precisely of this type ? The unity of the self, at all events, does not require such an interpretation. It is enough if every pulse of experience must, by its being, look before and after. The self exists when and so far as there is this continuity. Without the continuity it is nothing, and if there are temporal gaps the inference may only be that the gaps do not count. Peter continues to be Peter if, when he awakes, his experiences link themselves to that system which existed at the time he went to sleep. They link themselves to Peter's thoughts and not to Paul's. Even the fact that Peter can, to a certain limited extent, prearrange his time of awaking does not necessarily show that there is any Peter during the interim. The unity and continuity of Peter exist when he exists, and, perhaps, mock at the interim. There is no need for regarding this unity as a literal transcription of the mode of ' Metaphysic, English translation, vol. ii. pp. 316-317. Italics Lotze's. HOW IS THIS UNITY POSSIBLE ? 271 t existence of a physical thing. And the same argu- ment holds of feeling and endeavour as holds of thought. When Peter awakes, his aims and his feelings link themselves in the same way to the same old Peter of yesterday. In a word, the permanence of the self may be only an expression of its unity and continuity in time. The unity is compatible with the existence of temporal gaps, and these may be irrelevant. Why complicate the discussion by seeking a permanent in any further sense ? If such a possibility be admitted, no man can set bounds to its scope. Without a doubt the brain is relatively permanent, and is one of the conditions of the retentiveness of mind, but we cannot conclude that it is the sole condition, nor do we know how it affects consciousness. Again, there is subconscious- ness ; but such subconsciousness may not extend far beyond the ' fringe ' where it is found by actual inspection. Similarly a ' psychical disposition ' may be only a descriptive phrase, and not an explanation. We must cling to what we find, and remember that entities should not be multiplied. The question of the sense in which the self is permanent will confront us later, for ' permanence ' is part of what we mean when we speak of the substantiality of the soul. Meanwhile let us consider how far the investigation of certain abnormal cases throws light on the problems of the self. CHAPTEE XI MULTIPLE PERSONALITY Multiple personality is no new thing. It is as old as demoniac possession, lycanthropy, the frenzy of the oracle, the superstitious reverence for epileptics. But the scientific study of it is new, and if science has allayed its terrors, it has hardly resolved its mystery. Despite the mystery, however, no serious discussion of the nature of the self can ignore the rapidly increasing mass of evidence which relates to the dissociation of personality and the birth of multiple personality. These abnormal cases are the best possible test of the truth of a theory of the self. They verify the account of the salient features of personality, for it is only a change in the essential that can arouse the suspicion of dissociation. In a word, these abnormal cases enable us to apply the test of the ' method of difference.' When all flows smoothly we are apt to ignore the existence of problems. The abnormal cases force us to consider what is really crucial. And, again, a discussion of the dissociation of the self is the necessary com- plement of any account of its unity. The instances are rare and occasional, and prudence, perhaps, demands that too great emphasis should not be placed upon them. It is impossible, in most cases, to doubt the good faith of the investigators, but their zeal may sometimes have outrun their logic, and their suggestions, unwittingly given, may account for some 272 MULTIPLE PERSONALITY 273 of the results. But there is a sufficient body of facts to challenge enquiry. Every schoolgirl who reads the newspapers has heard of cases of total loss of memory, and even of men and women who have begun fresh careers in complete ignorance of their previous history. She knows more about these facts than Macaulay's early Victorian maiden knew of the text of Marmion. The number and variety of the instances is sufficient to arouse attention, if not mis- giving. What is this personality which seems some- times to be united by the frailest of bonds ? What is it which alcohol may dissever for a tiihe, and an accident destroy ? Even if there never is complete dissociation, it is worth while considering why the dissociation is erroneously supposed to exist. It is almost an axiom of common sense that a self is one and indivisible, that it is united to a single body, and that it exists as long as the body exists, if not longer. And it is natural that we should use the persistence of the body as our criterion of the continuance of personality, if only because bodily behaviour is by far the most important clue to our knowledge of our fellows. If a jury were satisfied that identity of thumb-marks is sufficient evidence of bodily identity, it would certainly scout the suggestion that the personaUty connected with the body, at any time, might give place to a new and distinct self. Moreover, in the ordinary course of events, the jury would be right. It is true that the body changes, and perpetually creates itself anew, but there is, in the end, no greater difficulty in reconciling this change with the identity of substance than in any other instance of substance. The criterion is a reason- able one, and may be safely adopted in the majority of cases. But there is no inconsistency in denying its infallibility. The jury, however strongly convinced of the truth of the inference from the body to the self, would not maintain, after all, that the body is the self. 274 PEOBLEMS OF THE SELF cHAr. Let us pass, then, to the more general question of the dissociation of personality. The term covers a very wide range, and some of the instances included under it are beyond dispute. They are, indeed, so frequent that they tend to be forgotten. Every one knows of the phenomena of somnambulism, delirium, trance and ecstasy, masked epilepsy, hypnotic sug- gestion, and the like. It is a commonplace, that is to say, that, at certain times, and for relatively inconsiderable periods, we find a break in the normal current of personality. For the time being there is dissociation. The ravings of a fever imply conscious- ness, but they are chaotic, they do not linger in the memory, they do not appreciably influence the further life of the person as the experience of each normal day, duly garnered, influences it. They imply, on the whole, a breach of the continuity of the normal life of the person. The same account holds of the doings of the somnambulist. Angel Clare, in Mr. Hardy's novel, had no consciousness, on the morrow, of what he had done overnight, and, in this respect at least, he was much like other somnambulists. But, on the whole, common sense is right in attaching comparatively little importance to these phenomena. For the breach of continuity, although marked and often regrettable, is very far from absolute. The ravings are connected with the previous current of conscious life. It is frequently possible to elicit in- formation from them which otherwise would have been carefully concealed. Sometimes, no doubt, the character of the person seems to change suddenly and completely. The gentlest patient, undergoing an operation, may, just before the anaesthetic takes complete efi'ect, suddenly become violent, blasphemous, obscene. It is possible to maintain that a subcon- scious self, usually repressed and more than half a demon, suddenly assumes the reins of government. But this explanation is quite unnecessary. The fact MULTIPLE PERSONALITY 275 that tendencies usually repressed and usually, perhaps, subconscious, may, in certain defined circumstances, awake to abnormal activity, is no good ground for maintaining that they form part of a separate, organised, and relatively stable and independent personality. They are far more probably parts of the normal personality. And sometimes the argu- ment takes precisely the opposite direction.* The real personality, it is argued, is much more likely to be revealed in hypnotism, in delirium and the other cases than in normal intercourse. A veneer of convention and carefully acquired habit obscures a man's real self in the sight of his fellows ; the cases we are mentioning give them a glimpse into his being. That interpretation is, in all probability, unlikely and unjust ; but it is not without some foundation in truth and in fact. The previous life of the self is not irrelevant to these so-called dissociations. The story in Coleridge's Biographia Literaria of the illiterate servant who repeated in her delirium long passages of Hebrew which she had heard an old Rabbi read when she was in his service fifty years before, may be exaggerated, but is not entirely baseless ; and it indicates a fact which ought not to be neglected. We may speak of such cases as cases of dissociation, meaning thereby that a personality, while not ceasing, may be appreciably disintegrated. While preserving some unity and continuity, retrospectively if not prospectively, it exhibits much less continuity than normal waking life. But we may suppose other instances. A dream is, or may be, an instance in point. It is usually a dissociation, even if vaguely remembered. But dreams are so fragmentary and bear so little relation to one another that we tend to neglect them altogether: Let us suppose, however, that a man, when slumber seals his eyelids, invariably begins his dream-life at the point where it ceased the night before. The idea is not a new one. It has 276 PROBLEMS OF THE SELF chap. formed the basis of many excellent stories. In that case we have, at least, the possibility of dissociation of personality into personalities. The man might really be a carpenter in the daytime and an emperor at night. He would act as an emperor, have all his dignity, and probably be as responsible and as con- sistent as most emperors are. But he is only a dream- emperor, you say. I do not wish to deny that it is possible to distinguish dream-presentations, however apparently coherent, from perceived objects, but the thoughts of the man are surely real, even in his dream. He does form resolutions in his dream, he does argue, he does compare evidence, and so on. He is exercising his mind about imaginary objects, but, none the less, he is exercising his mind. He may, of course, be the same personality throughout. He may be really a carpenter with the soul of an emperor, or an emperor with the soul of ^ a carpenter, or he may have a soul of featureless type, not peculiarly adapted to any one life in particular. But, on the contrary, each life might be distinct, without mutual continuity or reciprocal memory. In that case would it really be absurd to believe that there were two personalities in one and the same body ? What I have said hitherto is only a supposition. If it were fact, who could muster the evidence ? But it is a supposition which has some value as an illustration, and there is more than supposition to go upon. We are discussing at present the more permanent and lasting dissociations, not the evanescent cases mentioned hitherto. The explanation of these lasting dissociations, if they can be proved to occur, raises three distinct problems, although the first two are so closely connected that it is unwise to separate them rigidly. The first question is that of the evidence for the alleged fact of this fundamental dissociation. The continuity of a normal self is often comparatively loose ; but this, in itself, is not dis- XI MULTIPLE PERSONALITY 277 sociation in the sense we are considering it at present. How can we prove that there is disruption of the self, and not merely, so to say, a loosening of its unity ? What degree of alteration is required, and is this degree of alteration ever found in fact ? As we shall see, it is not easy to give an answer, at least to the latter part of this question. The second problem is whether it is possible to prove, not merely dissociation of a personality, but dissociation into personalities, or the birth of a new self. It is one thing to prove that the original personality has become disintegrated to such a degree that there is no meaning in speaking of its continuance. It is quite another thing to prove that a new self has arisen, or new selves emerged. The result of the dissociation might only be that the self has given place to something which is not a self at all, perhaps even to a mere succession of dissociated processes. The third question raises a further point. If there may be dissociation of a personality into personalities, is it possible for a plurality of person- alities to exist contemporaneously in the same body, or can they exist only successively ? Let us consider, first of all, the types of cases which we find. Each instance, of course, has specific peculiarities of its own. The case of Fdlida X, studied by Dr. Azam, is different from that of Miss Beauchamp, studied by Dr. Morton Prince, and similarly the other classical cases of Ansel Bourne, Mary Reynolds, Louis Viv^ and the rest, are distinctive in their several ways. At the same time there are certain broad groups under which the instances are ranged, and a division of this sort, especially the fundamentum of the division, is of considerable importance. Memory is one such ground of division. In some of the cases the memory — on the whole the intimate personal memory — of the earlier self is retained, but the subject insists that this earlier self is foreign to him. He may express the 278 PROBLEMS OF THE SELF chap. facts by saying that he has become a different person, but that is only a verbal contradiction. More frequently he refers to this other personality which he remembers so intimately as ' the other fellow ' or ' it,' or he uses some descriptive title, as often as not uncomplimentary. In other cases there is, to all appearance, a complete break in memory. The new personality, if it is new and a personality, retains no memory of the old. The Eev. Ansel Bourne became A. J. Brown and forsook his vocation as an itinerant preacher in favour of the activities behind a con- fectioner's counter, but, until recollection returned and with it Ansel Bourne, there was no bridge of memory, and, apparently, no good reason for denying that A. J. Brown was really a distinct person. Or, again, the cases may be distinguished according as they are alternating or not. The personalities, so- called, may alternate rapidly or slowly. The smile of Sally may break through the woebegone counte- nance of Miss Beauchamp, the alternations may be induced by hypnotic suggestion, or they may occur periodically at intervals of months or years. On the other hand some dissociations seem permanent. There is no alternation. One self has gone, never to return. These are the most important divisions of the abnormal cases which suggest multiple personality. Let us now proceed to the discussion of our first question in somewhat fuller detail. On what grounds is it held that there can be a dissociation so profound that the self, eo ipso, forfeits its identity ? Personality has many attributes, but some of these are more distinctive and important than others, and we may assume that the dissociation is due to disorganisation in some one, or all, of these essential features. In the preceding discussion we have tried to discover what these attributes are, and the consideration of the abnormal cases will test the accuracy of our results. MULTIPLE PERSONALITY 279 The identity of the self means its unity and continuity, and, up to the present, we have seen no sufl&cient reason for denying that this unity and continuity is a unity and continuity of experiences and nothing elsa In any case we must argue from the unity and continuity of experiences. There is unity, then, in the experiences of cognition, feeling and endeavour, and each of these three requires the others. In the development of these essential elements of the self there is real growth and real novelty. Consequently the mere fact of change in any of them is irrelevant to the question of a funda- mental dissociation. If there is dissociation of this kind we must mean by that term a change so radical, sudden, and complete, that we cannot, as it were, graft it on to the earlier mode of existence. The unity and continuity also require retentiveness in a high degree. Without that, there could not be development, and instead of novelty, properly speak- ing, there would be mere disconnection. It is some- times held also that memory is implied, but that is more doubtful if memory be identified with recollection, and distinguished from retentiveness. There must be livrniT], but there need not, perhaps, be avdfivrja-Ks. Such is the identity of the self as it appears to introspection ; but we must also remember the role of the body. We are not, as the four beasts in the Apocalypse, ' full of eyes within,' but we are aware of our bodies through the internal sense as well as the external. While the body is never a part of the self it is, as we have seen, the constant companion of the self, its constant centre of movement, and the centre, also, which ties down all sensation and perception. So much of our conscious life, in fact, refers to our body and its needs, so closely is it connected with our experiences, that it is natural and, perhaps, inevitable to maintain that the in- sensibility of any part of the body, or any profound 280 PROBLEMS OF THE SELF chap. alteration in it, is bound to make a difference to the self and, perhaps, to disrupt it. Even witches and warlocks were supposed to carry a body with them in their unearthly rambles, and this body had a real physical connection with the normal body which they deserted for the time being and left in a state of coma. The reader need scarcely be reminded of the tale of Tod Lapraik in Stevenson's Catriona. The spiritual body gambolled devilishly on the Bass Eock : the earthly body nodded over the loom in the warlock's cottage on the mainland. But the silver bullet which found its billet on the Bass pierced the heart of the physical body on the mainland. The question of the distinction between memory and retentiveness, in addition to its intrinsic import- ance (to which we shall return), raises another question which has been neglected hitherto in this essay. We have seen that memory only discovers, and does not produce, personal identity. The self must be a unity if we are to recognise the unity by the aid of memory, and, clearly, a great part of the content of the self is never explicitly remem- bered at any given time. Consequently it is legiti- mate to argue that a self might be a self without any explicit memory. All that would be necessary would be a degree of retentiveness sufficient to insure continuity. Those who believe in the doctrine of reincarnation must take this line of argument, since they must admit that there is no > conclusive evidence for the recollection of a previous existence. On the other hand it is at least equally legitimate to take the contrary view. The fact that there must be some parts of the self which are never remembered in detail does not logically imply that it is possible for a self to exist in the absence of any memory. A being incapable of recognising itself as itself is, perhaps, not a self at all. This argument raises a new issue. Personality, in MULTIPLE PERSONALITY 281 the current acceptation, implies a certain degree of intellectual and moral development. A person is responsible, and cannot be responsible without the power of making a deliberate and reflective choice. Personality, in short, is a legal and ethical notion which applies only to beings of a complex and developed type of psychical Hfe. It is probable, how- ever, that the term self is wider and more inclusive than the term person. Only the adult self is legally responsible. Children have a lesser degree of re- sponsibility, and very young children may not be responsible at all. And similarly it may be true that when the chmacteric of life has receded far into the past and decay has supervened, personality, strictly speaking, has departed. But the same self, we commonly assume, has lived through all these stages. Personality may apply only to a stage in this con- tinuous process, although that stage, in point of value and length of duration, is best worth considering. We must remember this point in the course of our discussion. It is natural to insist that a self should at least be capable of developing up to the level of personality, but perhaps we cannot claim that it must have this power. If memory is an essential characteristic of the self, then there is no doubt as to the interpretation of some of the abnormal cases. Mere gaps of memory are not, of course, sufficient to prove a rupture of personality, for there are such gaps in normal life. But when, as in these cases, the gaps extend for long intervals of time, the problem is altered ; and it is need- less to give instances to prove that they may extend for years. The instances are so frequent that they have only to be mentioned for the point to be admitted. The ordinary laws of obliviscence may account for lacunae of memory, but not for a total absence of memory during long periods. And in the cases of alternating personality, like that of the 282 PEOBLEMS OF THE SELF chap. Rev. Ansel Bourne, the gaps make a difference when the old self is renewed. When the patient ' becomes himself again ' he takes up his life at the poinl3 where he left it off, but he has to resort to all sorts of shifts in order to fill the interval. And when the gaps are filled with consciousness opaque to the memory of the older self but with a bond of memory in itself, we begin to have real evidence of dissociation into person- alities. There are different systems of memory, each impervious to the other. In the Beauchamp case, for instance, we have an instance of two apparently different selves opening a correspondence. The one self knew of the, doings of the other indirectly from the testimony of others, from the notes which the other personality had left, and the practical jokes she had played. It is plain, in such an instance, that the march of events is very different firom the con- tinuous current.of normal personality, and we have, at least, good grounds for suspecting that the facts can only be explained on the hypothesis of multiple personality. On the other hand there does not seem to be sufficient evidence for maintaining that these pro- longed gaps of memory also imply a complete absence of retentiveness. Mary Eeynolds, for instance, when she passed for the first time into her second state, had to be taught the arts of reading and writing, apparently from the very beginning. But she learned them in a few weeks, and, therefore, it is fair to assume that she retained the effects of her previous education in some degree. Similarly the new person- alities, in these cases, can speak and walk. They do not have to begin from the cradle once again. There are great differences, of course. Some members of the Beauchamp family showed entire ignorance of French and some other accomplishments of the normal Miss -Beauchamp. But although some parts of her education were not retained, the most funda- MULtlPLE PERSONALITY 283 mental were. I do not wish to minimise the difference between the personalities in this respect. Some are stupid and some are clever, some babble and stammer while the speech of others is quick and coherent. Louis Viv6 in some of his states was an expert tailor. In others he could scarcely thread a needle. What I mean is that, on the whole, the personalities which are claimed to be different are not completely isolated in point of retentiveness. It is probable that little tricks of mannerism could be detected by careful en- quiry, and seen to be identical in all the personalities. And there is no sufficient reason for believing that only bodily habits and motor dexterities are retained. These must remain, but they do not account for all the phenomena. Accordingly, the evidence is inconclusive in these respects. Without retentiveness there could be no continuity, but there seems invariably to be some retentiveness. And, as we have seen, it is not un- reasonable to contend that something less than exiplicit memory is a sufficient minimum for the continuity of the self. There are few of us who can remember events in our lives which occurred in our infancy or early chUdhood, but our conscious lives did not begin suddenly at the date we can remember, and although the memories of extreme youth some- times recur in advanced age, they do not always do so, and are usually forgotten in the heyday of life. At the same time the balance of evidence inclines in the opposite direction. The continuity of the self seems to imply something more than that minimum of retentiveness which occurs in the recorded instances of multiple personality. And we must also remember the very important r6le which explicit memory plays in normal personality. Memory is the only direct evidence which a man can have of his own identity. There is indirect evidence also. A man may be justified in believing that he did, or said, certain 284 PROBLEMS OF THE SELF ohap. things in his childhood which he has now forgotten. His parents or teachers have told him so, and he\ has no reason to doubt their word. But even in these cases memory may enter in some degree. The man may not remember the particular incidents in question, but he' does remember circumstances which make the whole story probable. And, again, he often re- members incidents which would otherwise have escaped his recollection when they are recalled by others. A total absence of memory would make our earlier history meaningless to us, " even if it were really ours. Indeed, there is some truth in Locke's quaint suggestion that it would be unjust if we were condemned at the day of judgment for acts we could not remember, even granting that the recording angel had duly inscribed them in the book of life. These reasons, taken together, suggest that a self without explicit memotry is an impossibility. Let us pass to some other characteristics of self- hood. The abnormal cases are marked by a profound change in the unity of endeavour. ^ Dr. Morton Prince believes that, in the Beauchamp case, there is evidence of the simultaneous conflict of two wills, each of which is reinforced by the influence of a long- continued train of action. Whether that is true or not there is, at least, conclusive proof that the per- sonalities which are presumed to be diflferent strive to attain very difi'erent aims, in this instance and in others. Sally, in the Beauchamp case, arranges to go to Europe ; Miss Beauchamp desires to continue her studies in the States. Sally arranges to meet her friend Jones ; Miss Beauchamp would rather meet any one else. When F6lida X felt her crisis beginning she would write down the particulars of her immediate situation in order that there might be as much continuity in her actions as possible. Similarly in the case of Louis Viv^ and others there are all the symptoms of demoniac possession. Some of the MULTIPLE PERSONALITY 285 selves, assuming them to be different, are thieving and some honest, some shifty and others straight- forward. The difference in character between them is at least as remarkable as that between the re- generate and unregenerate days of many who have been converted from sin to saintliness. But the rejoinder may be that such a difference is irrelevant. Bunyan unregenerate was the same man as Bunyan regenerate. Sudden conversion only implies a new birth in a metaphorical sense, and it is all too certain that the convert may, at any moment, return to the flesh-pots of Egypt. Accordingly it may legitimately be argued that the differences in character and aim of conduct which are found so frequently in these abnormal cases, are nothing but exaggerated instances of a normal occurrence. This argument would be sound if such changes occurred alone. But when they are connected, as they are in these cases, with a complete break in memory and a relative loss of retentiveness it is clear that the hypothesis of a unitary personality has come perilously near to nonsense. There are different chai'acters allied to different systems of memory. What is that but multiple personality ? The unity of feeling is as distinctive a feature of character as the unity of endeavour, and in this respect also there is a significant difference between these abnormal cases and the instances of religious con- version. The hasty and irascible temperament rarely changes into the meek and long-suffering. Paul the persecutor does not lose his abundant vitality when he becomes Paul the missionary. If there is added restraint, that restraint is the fruit of long and careful self -surveillance. But these profound and radical changes in feeling and temperament are the most significant of all in the abnormal cases, and such a change is found in every single instance of importance. It is as if one of the personalities could not be 286 PROBLEMS OF THE SELF chap. despondent, and the other, if there is but one, could be nothing else. The lives of most of us are a mingled yarn, joy and sorrow together. But the unstable beings we are now considering seem to follow a different order of things. The sorrows unite in one life, and the joys in another. Neutral experiences may belong to both, but whenever the pendulum passes from one extreme to the other the unstable personality cannot bear the shock, and one self seems to give place to another. Indeed, in almost every instance, a great emotional shock was the original cause of the dissociation. Such a shock, primarily mental in character though appropriately accompanied by thunder and lightning, launched the Beauchamp family (with the possible exception of Sally) on its strange career. The shock of being stung by an adder led to the curious phenomena in the case of Louis Viv^, and so of the rest. It is an interesting confirmation of the importance of the sentiments that any emotion connected with these events was most . significant in determining the current of personality. These differences might not be sufficient in them- selves, but, again, they occur in conjunction with the other differences previously mentioned. And the evidence is still further strengthened by the close connection between such emotional states and states of the body. It is true that defects of the external senses seem to have no appreciable influence in the way of dissociation. A man becomes deaf or 'blind, and becomes deaf or blind suddenly, without giving rise to the suspicion, in himself or in others, that there has been a dissociation of his personality. But with organic sensation it is different, especially when there is a sudden and violent change of coenaesthesia. The case of Pfere Lambert has been frequently quoted in this connection, but is apposite enough to be quoted again. ' A soldier believed himself to have been killed in the battle of Austerlitz, MULTIPLE PERSONALITY 287 and he had, in fact, been grievously wounded. When he was asked how he was he would reply, " You want to know how Pere Lambert is ? He exists no longer : a cannon-ball carried him away. What you see is not he. It is a wretched machifie which has been made in his likeness. You ought to ask them to make another." In speaking of himself he never said " I " but " it ".' ' The change in organic sensation due to his wound led this poor man to suppose that his very personality had ceased. He may have been wrong, but his very mistake shows how important bodily sensations are for the sense of personality. It was on evidence such as this that M. Ribot based his ' colonial ' theory of the self. The importance of coenaesthesia, and of certain specific organic sensations such as those of sex, led him to maintain that the self is nothing but a colony of sensations, the most important of which arise from the internal condition of the body. The colony is usually compact and well organised. But there may be, perhaps from physical causes only, dissension and even a revolution within the community. When the revolution has become a fait accompli there is, strictly speaking, a new self, although the external appearance of the body is insufficient to show the fundamental change which it has undergone. I am mentioning this theory only by way of illustration of the importance of organic sensations ; enough has been said already to prove its inadequacy as a complete account of the phenomena. Not only does it ignore the fundamental distinction between experiences and their objects, but it also commits the fallacy of implying that all experience can be construed in terms of sense. The colony is only a colony of objects known through the senses. It is not the self at all. And I do not think it is necessary to believe that P^.re Lambert, or any others in the same case, 1 Cited by Kibot, Maladies de la personnalM, p. 36. 288 PROBLEMS OF THE SELF chap. are right in the inferences which they draw. The chances are that the presence of memory is almost always a sufficient proof of personal identity, whether the subject owns it or not, although, of course, it does not follow that the absence of memory is a proof of the disruption of identity. In this particular class of cases, where memory is present, there is usually comparatively little change in personal unity and continuity. There is only enfeeblement, in the majority of cases, and a dazed feeling of homeless- ness due to the change in organic sensation.' But in the other instances, where memory, temperament and plan of life change all together, we have invariably the additional factor of a concomitant change in organic sensations. In the case of Louis Viv^, for instance, there was a necessary correlation between each of the personalities and some specific localised anaesthesia or partial paralysis. If any one of these physical states were induced, the corresponding self, if it can be dignified by that name, appeared also. The Beauchamp family showed traces of the same phenomenon, and, in particular, the difiierent ' person- alities ' were liable to difiierent sorts of hallucination. Let us sum up the evidence. Personality may alter very profoundly in any of its fundamental features. Memory may cease, and the continuity of feehng, endeavour and cognition be broken. More- over, organic sensation may undergo what seems a complete revolution. Some bond of retentiveness may perhaps remain, but it is a feeble and tattered bond, with frayed strands. The changes in each particular respect may seem to diff'er only in degree from the necessary and normal changes in a healthy personality. They seem more sudden and abrupt, that is all ; and it is likely enough that a breach in any one of them would be insufficient to annul ' The relations between some of the quasi- personalities in the Beauchamp case are somewhat exceptional in this respect. MULTIPLE PERSONALITY 289 personal identity. But the facts, in a suflScient number of instances, show that the changes in these respects do not occur separately but together, leaving at best a pitiful and impotent residuum. How is it possible to maintain that the self can continue when that occurs ? It is futile to argue that, because there is a connection between all these elements, they must all change together. Perhaps they must ; but what is it except the self that is broken in such a case ? We may pass, then, to our second question. What is the evidence for the dissociation of personality into personalities, or, in other words, for multiple person- ality in the strict sense ? In the first place, we should have to prove the existence of consciousness or sub- consciousness. Without that there is no self, and in some of the instances already mentioned the absence of consciousness is probable enough. It is not un- likely that the behaviour of a somnambulist or an epileptic is unaccompanied by any sort or degree of consciousness. The opposite theory, no doubt, may be maintained also, but whatevet be the correct ex- planation in these special cases, there is no reasonable doubt of the presence of consciousness in the majority of the other instances. We have as good ground for believing its presence in these instances as in the case of any one other than ourselves. The second requisite is a negative one. We must be certain that the evidence does not owe its plausi- bility to some error in the observations. One such cause of error, as has already been mentioned, is found in the influence of suggestion. It is often difficult to put questions without suggesting the answer, and that not merely because the form of the question may itself furnish indications of the answer expected, but also because the observer himself may, quite unconsciously, suggest that answer. A subject in hypnosis is peculiarly susceptible to such suggestions, and there- fore the observer may, quite inadvertently, build a u 290 PROBLEMS OF THE SELF chap. fabric of his own construction. Fortunately, most of the observers are aware of this danger and guard against it as much as possible. We have, therefore, the right to assume that a great deal of the evidence is free from this defect. Similarly we have, in most instances, no reason to doubt the honesty of the replies made by the subject observed. Miss Beau- champ was not assuming an interesting pose. Her greatest misfortune was her ignorance of the designs of Sally and of the contents of the letters which Sally wrote to 'Jones.' Again, it would be hard to suggest a better test of the accuracy of memory than Mr. Grurney's offer of a sovereign to the waking self if it could remember the events which had happened to it during hypnosis. At the same time it is probable that a good deal of the evidence should be discounted for this reason; We need not suppose any conscious disingenuousness, but although the play of fancy is less conspicuous in adults than in children — partly because adults have a firmer grasp of reahty — it is part of the constitution of the normal mind. Some of the selves in multiple personality may be creations of the actor's instinct. It is easy for us to throw ourselves into a certain situation, to feel that our personality finds full play in it, while in others it is unduly repressed ; and the further step of believing ourselves different and even of ridiculing ourselves is not so strange as to appear impossible. We must therefore be sure either that these causes of error are entirely absent, or else that due allowance has been made for them. In the third place we must be able to prove that when dissociation occurs new organised personalities arise or begin to arise. This, however, is only a statement of the issue. Let us, then, define it more carefully. We must guard against the possibility that the unity of self has given place to a lesser unity, that, in other words, one and the same MULTIPLE PEESONALITY 291 self continues but in a form in which its unity is either absent or else very difficult to discover. That explanation probably suffices for many instances. It is impossible, in reading many discussions of the subject, to avoid noticing that the existence of a sub- conscious self, distinct from the primary, or normal, or waking self, is far too lightly assumed. Let us take, for instance, those cases in which there are post-hypnotic effects of suggestion. At a given time the subject, while engaged in ordinary duties, feels an irresistible impulse to perform some trivial act, — say, to toss a penny. This impulse, and the time at which it occurs, may be due to previous suggestion during hypnosis of which the subject retains no recollection. This does not prove that there are two selves in the case, a conscious and a subconscious. It rather proves a direct and intimate connection between the conscious and the subconscious life of one and the same person, since the connection is much closer and more intimate than that which subsists between two normal personalities. And the fact that certain trains of experience can only be recalled in hypnosis does not prove it either. We have all heard of the Irishman who lost his gun when he was drunk and could not find it so long as he remained sober. We might explain the event by saying that Paddy drunk and Paddy sober are two different persons, but there is no necessity for adopting this hypothesis. The kind of actions which an intoxicated man performs may have supplied the stimulus to Paddy's memory, and the same explanation may hold of the passes of hypnotism. Again, we must reckon with the possibility that the dissociation in question means that unity of personality has disappeared (at least temporarily) and that, instead of a self, we have a succession of detached psychical states, too loosely organised to 292 PEOBLEMS OF THE SELF chap. be called a self. On certain theories of the connection between mind and body, for instance, we should expect that such ' floating ' experiences would occur. That theory of parallelism which maintains that there is universal parallelism between body and mind must maintain that our minds are organised centres of experience corresponding to an organised physical centre in the body. A disturbance of this bodily centre would not destroy the physical elements organised but only scatter them, and we should expect, therefore, to find a similar scattering of experience. Such scattered experiences would be the ' floating ' states in question. The evidence which we have, however, seems opposed to this theory. If our investigation is only superficial we have no right to expect more than inconclusive results ; but when it is prolonged and careful we find that the experiences ensuing upon dissociation are not utterly fragmentary but are organised. We are therefore driven back upon the view that the original self persists in a disguised form, with the alternative of maintaining that there really is multiple personality. To prove the latter we should have to show that there is a system of memory, of feeling, of endeavour, of cognition, implying, perhaps, a certain degree of development and responsibility, and correlated with a distinctive tone or trend of bodily feeling. We should also have to prove that this system is distinct from that which existed prior to disintegration, and, if there be more than one such system, that each is distinct from the other. It might be added that we should also have to prove that such systems of experiences are not only distinct, but as distinct as one normal personality is from another. This requirement, however, is much too stringent. It assumes that the minimum of distinction between persoiialities is to be found in those personalities i MULtlPLE PERSONALITY 293 which are connected with different bodies, and it is improbable that cases of multiple personality could reasonably be expected to show such complete distinctness. Despite the differences in organic sensation, there is a greater bodily connection be- tween such personalities than between normal personalities correlated with different bodies. And, in theory, it is possible to go further. Multiple personalities might have a good deal in common, and yet be distinct personalities. They might share experiences, provided that the trend and the unity' of each personality was distinct. A difference in some experiences, and in the organisation of these experiences, might suffice to constitute different personalities. But who can say what degree of difference in this respect would really suffice ? The discussion of our first question has shown that there is, at least, some reason to believe that the second question can be answered in the affirmative. In alternating personality there seem to be different systems of memory. The gaps are not states of coma, but are filled with a new system of experiences apparently organised. The changes in character are changes in which one type of character becomes another. There is dissociation in all the essentials of personality conjointly, and then, apparently a ■ fresh synthesis. I have not, however, given any proof of these statements, and such a proof could only be given by an exhaustive examination of the instances. And it would take too long to review ' them all. Accordingly I think the best course is to examine one of them in order to see whether there is dissociation into personalities in that case. We may be reasonably certain that the sources of error enumerated above have been fully appreciated in the case of the Beauchamp family, and, accordingly, we may approach the consideration of this case with a fair degree of confidence. We may also follow 294 PEOBLEMS OF THE SELF chap. Dr. Prince's terminology in calling tie apparent personalities B L, B IL, etc. B I., the Miss Beauchamp whom Dr. Prince first came to know, was clever and cultured though hysterical. She was also puritanical, morbidly sensi- tive to the opinion of others, morbidly conscientious, morbidly anxious to obey, and to give no possible cause of offence. B IV., a later 'personality' which alternated with B I., was qilite different. B IV. had a good appetite, B I. had a bad one, and the taste of each, with reference to matters of the table, differed profoundly. B I. hated oysters and cigarettes, but liked ice-cream. B IV. was fond of oysters and ex- travagantly fond of smoking, but hated ice-cream, B I. wore her hair low, went to church, and neglected newspapers. B IV. wore her hair high, neglected religion, and devoured the newspapers. B I. was patient, dependent, emotional, fond of children and kind to the poor. B IV. was none of these things. Quick-tempered, self-reliant, conqueror of her emotions, regarding children and the poor as nuisances, she seemed the antithesis of B I. It is true that one and the same personality may exhibit revulsions of taste and feeling which are parallel to these. Some men have a craving for ■ tobacco in the morning or the evening, but at no other time of the day. A capricious fondness for ice- cream is no unusual phenomenon. Similarly, a weak 'character is sometimes unexpectedly obstinate, and at other times pliant, meek, and long-suffering ; and so of the other points of difference. It is not merely the differences and the caprice that are important in the abnorm'kl cases, but their organisation into systems. The opposing tastes and tendencies, instead of blend- ing together as they do in normal personalities, organise themselves systematically into different groups. Differences of taste and character, differences, to some extent, of intellectual attainment become MULTIPLE PERSONALITY 295 connected together into different unities. The im- portant points are the degree of this coherence and the difference of the two unities, and the reader will be satisfied on the question only if he reads the whole evidence for himself. Add to this that B I. had no memory of B IV., nor B IV. ^ of B I., but that the memories of each referred to their several characters, tastes and aims, and it seems almost impossible to deny the presence of multiple personality. I have purposely chosen this example because it is not one of the strongest instances in the Beauchamp family. B I. and B IV. were most nearly allied in this remarkable case. The education of B I. was present in B IV. The tones of her voice and, doubt- less, unnumbered mannerisms were the same. A synthesis of the two was obtained in the end, and there are better grounds for supposing that B IV. was only a phase or pose of B I. than in any other instance. Dr. Prince himself is inclined to believe that there were really only two personalities — Sally and Miss Beauchamp— -and that B I., B II., etc., were only dissociated conditions of the real Miss Beauchamp. None the less the fact remains that B I. and B IV., to mention none of the others, were distinct enough to be distinguished as different selves on the current interpretation of the term, and that they were organised in the way a personality is organised. It was not always so in the other cases. The personality called Mary Reynolds was so feeble in onef of its alternating phases that this was, perhaps, below the level of selflbood, and the same is true of some of the ' personalities ' in the case of Louis Viv6 and in some of M. Janet's cases. F^lida X, again, sometimes showed a personality so feeble that it scarcely rose to the level of development which we find in dementia praecox. Indeed, the personality was so feeble in these cases ' B IV. must be distinguished from B IV. a. 296 PROBLEMS OF THE SELF chap. (if it was not too feeble to be a personality at all) that it is plausible to argue that the phenomena were only • exaggerated instances of fatigue. The self had not the strength to remember, or to hold itself together continuously with its past and its future. This explanation is sufficient in the case of lapses of memory ; for memory requires a conscious effort, and fatigue may make this eflfort impossible. But it is inadequate in all other instances. It is false that the self, at any moment, keeps its whole life together by an effort. That is a metaphorical expression of the fact that the self is united and is continuous, and if the unity and continuity is broken through any cause, then personality, in the strict sense, has dis- appeared. There is more than the mere fatigue of a self which persists ; the self ceases, and dissociation takes its place. In these abnormal cases, if fatigue be the cause, wp are forced to admit tha,t the self has died of fatigue. If we require of personality the power of acting responsibly, and of thinking with a certain degree of clearness and consistency, then we must admit that some members in the families of multiple personality do not fulfil this requirement. They are mere travesties of personalities. The fragmentary intellect which they show may be only a perversion of some- thing borrowed from previous, though forgotten, experience. But it is not true that dissociation invariably results in the imbecility of the ensuing conscious processes, or that only one self can remain which is organised enough to be a personality. In the Beauchamp case, for instance, Sally seems a real personahty, though in some respects childish and unformed. She was not more childish than many persons we know. The others may only have been phases of a single personality, as Dr. Prince maintains. But if we take the evidence as it came to him, we may be confident that we should not have doubted XI MULTIPLE PERSONALITY 297 tlie existence of multiple personality, apart from tl^e natural assumption that there cannot be more than one self connected with a single body. And there is no doubt that B I. and the rest had attained the level of personality., If B I. was only a phase, and not a person, she had at least won a scholarship in modern languages and that is commonly supposed to be a test of ability. Let us leave this matter for the moment, and consider our third question. The only direct evidence for the simultaneous existence of multiple personalities is found in the case of Sally. Sally was a bright, impish, childlike spirit who frequently enlivened the proceedings of the Beauchamp family. Her memory, by her own account, extended back to the earliest events in Miss Beauchamp's history, and, although her independent life did not become prominent until later, she had, even in early times, a better right to be considered a personality than, say, B IV. No doubt she was, at all times, more of a child than a woman, and, in particular, she had none of Miss Beauchamp's accomplishments, none of her diligence, none of her seriousness ; and, again, she possessed an extremely limited range of bodily feeling as compared with a normal personality. She was almost insensitive to pain, and her organic sensations had not the intimate union with personality which such sensations usually have. She was aware of these sensations in the same way as she was aware of Miss Beauchamp's past and present experiences. They had a foreign aspect to her mind; she thought of them as belonging to another. That, of course, is some slight indication that SaUy and Miss Beauchamp were not really distinct personalities. But if Miss Beauchamp's memories were open to Sally, Miss Beauchamp herself had no memory of SaUy's doings. And the most important point is that of the simultaneous contest of wills. The fight between Sally and the other 298 PROBLEMS OF THE SELF chap. personalities — if we may be permitted to speak of them in the plural — was far keener than between these personalities themselves. The conflict in Sally's case was very stern. She played practical jokes on the other members of the family, requisitioned their funds, and even signed her hand to treaties. These cases, however, are cases of alternation, although the alternation was extremely rapid. They do not prove the simultaneous conscious presence of two personalities, and I do not think that such a contention can be proved absolutely. But the evidence points in this direction. Let us hear Dr. Prince : ^ ' It came about in this wise. I had endeavoured to change IV. into B II., but could obtain only 'the hypnotic state B IV. a, evidently prevented by Sally, whose hand was apparent from certain characteristic manifestations. It ended in Sally's coming instead of B II., and I puoceeded to lecture her on her conduct ; but while in the act of doing so she cleverly escaped by changing herself back to IV. (a conscious personality). To this personaUty an attempt was made to explain the situation. ' " Sally has been behaving very badly," I began. IV, repeated the sentence as she heard it, the words being transformed into others having an opposite meaning. " Sally has been behaving beautifully." " No," I said, " badly." " Yes," she repeated, " beautifully." " No, no ; badly." " Yes, I understand ; beautifully, beautifully." ' Thus, for the moment, I was circumvented. It was in vain that I sought to make her hear the word " badly." It became apparent that Sally twisted in her mind everything that I said so as to give it an opposite meaning. She became deaf to certain words and heard in their places other words of a different ' The Dissociation of a Personality, pp. 321-322. MULTIPLE PERSONALITY 299 signification. Everything that was said in criticism of S., she heard and understood in S.'s praise; she even said repeatedly that she liked S., had no fault to find with her, was perfectly satisfied with her, and so on. Finally she ended by refusing to obey, asserting that she was her own mistress, would go where she pleased, and do as she pleased. This, too, was plainly the work of Sally, who had taken possession of her tongue. But most dramatic was the assertion of her own personality in the midst of these sentences. Every now and then, like one pursued by an invisible demon, and as it momentarily she had broken away from the power that bound her, she would exclaim, "Don't let me speak like that," and then the next instant she would give utter- ance to Sally's words.' This is but one instance out of many, but it is hard to see how better evidence could be forthcoming. Any one who remembers the incessant feud between Sally and the other members of the family is bound to admit that the behaviour of the others, in this instance, cannot be a pose or a whim. Unless they were passive instruments of Sally the whole dialogue becomes inexplicable. And this fact, together with the other evidences of Sally's personality, makes the hypothesis of multiple personality far the most reason- able in the circumstances. But let us suppose that some other conclusion is possible. Let us suppose that Sally and all the other members of the family form but a single personality variously disturbed and disguised. Even in that case the data of psychical research have very considerable value. For they show that selfhood is compatible with a fragmentary minimum of unity and continuity. If the family, despite its organised diversity, were really a single personality, what can we mean by personality ? Character, attainments, memory, degree of responsi- bility may differ utterly at different times, the unity 300 PROBLEMS OF THE SELF ohap. and continuity which exists may be divided into distinct trains of unity and continuity, and yet we maintain that all these trains and all these diversities are really one. The data of psychical research are important precisely because they compel attention to one of the most fundamental problems of the self. We commonly assume that one self is connected with one body and we find in this self unity and continuity, and a distinctive trend of experience. That explanation suflfices for normal cases or, at least, seems to do so. Difficulties do not obtrude themselves. It is fair to assume that the unity and continuity are considerable, and any attempt to estimate the precise degree of this unity and continuity seems labour thrown away. We forget the enormous differences between youth and age, we neglect the temporary lapses of memory and the disconnected experiences in the life of a normal person. These seem insignificant in com- parison with the usual current of his existence. And, pursuing this train of thought, we are apt to attribute to the self, lightly and unthinkingly, a very high degree of unity. One single permanent self exists in childhood and in age. Each of us is an identical ego. The discussion of this ego, the theory of the self as a single substance, will occupy us before long. But it is essential, in this place, to take a retrospective glance at the course of our argument. We have considered experience and the unity of experience. Experiences seem the stuff of the self, and selfhood is the unity and continuity of those experiences. For experiences as we find them are only possible as part of a unity, as focussed in the life of a particular self. The explanation of these facts may lie in an over- soul, a substance, an ego, and we may be forced to admit the existence of such an ego. But ia that case we must remember that the reason for assuming MULTIPLE PERSONALITY 301 the existence of a soul lies in the unity and continuity of experiences. The soul is either the correlate required to make this unity possible or else it is the correlate and the unity together. Accordingly the unity and the continuity of personal experience is the "most important factor in the case, and it is essential to discover what kind and degree of unity and continuity is required to make a self. It is in this respect that the psychology of multiple personality is peculiarly instructive. The plain man is very loth to admit that such multiplicity is possible and he may reasonably deny that it has been proved. But consider the arguments that he uses. He can only present a plausible case by maintaining that the dissociations studied in these abnormal cases are but exaggerations of similar phenomena which occur in ordinary life. He must, in other words, concentrate his attention upon the lach of unity in the life of a normal self. Character, aims, intellect, memory, and bodily feeling all change, and change profoundly, in ordinary life. The unity of personality is compatible with the utmost diversity in any of these respects and even in all of them taken together. There is force ia this argument, since it rests on solid fact. Very likely it is the suddenness rather than the character of the changes which is the essential difference between the abnormal cases and the normal, and it may be argued that a difference in suddenness cannot possibly amount to a difference in kind. But if that be so we must admit that the normal lives of the men we know, the lives of Simpkins and of Clarke, are less of a unity and less coherent than we commonly suppose. To say that a single permanent self is necessary to account for the unity of the self may be a legitimate argument. But when there is so little unity as this argument implies, it is hard to see how such a statement can have much meaning. There may be a considerable degree of unity in some, 302 PROBLEMS OF THE SELF chap. or most, cases. But the argument, if it is valid, proves that there need be very little ; and therefore that the self need not mean much. If, on the contrary, it is maintained that the self must involve a high degree of unity and continuity, then it must be maintained that multiple personality is a fact, and this, perhaps, is the easiest solution of the problems raised by the evidence. If it be admitted that the self is not the body then there is no special difficulty, in theory, in maintaining that several selves may be connected with a single body. But it is very difficult to think that the theory is actually exemplified in fact, although easy enough to con- template it as a possibility. The multiple selves have too much in common to stand out clearly in splendid isolation, even if they agree only in respect of retentiveness. And such ' selves ' can be synthesised, often, into a personality which may be said to contain them all and to be continuous with them.^ This synthesising of many personalities into one is very paradoxical to the ordinary mind although certain philosophers and theologians have sometimes shown themselves ready to accept it. Such difficulties, perhaps, derive most of their weight from prejudice, but, to say the least, it is hard to be certain that the prejudice is a mere prejudice, without any foundation in logic. Moreover, there is too much truth in the other line of argument to make this one entirely satisfactory. If we maintain that a self must contain a very high degree of unity, then we shall be forced to the un- pleasant conclusion that multiple personality, instead of being the exception, is the rule. We are different men at different stages of life, and although the contiguous stages may show unity and continuity, ^the remote stages hardly do so at all. The conclusion of the whole matter shows the intricacy of the issues ' See, e.g., the final chapter of the Beauohamp case. MULTIPLE PEHSONALITY 303 involved. We are selves, and our life is a unity, but let us beware of laying too great emphasis on the degree of ^hat unity. Let us remember that in many cases it is less than we are accustomed to suppose. If this is remembered, it is easier to approach the discussion in the two succeeding chapters with a fair and unbiassed judgment. CHAPTER XII BISCUSSIONS OF THE SELF AS SUBSTANCE m MODERN PHILOSOPHY The term substance has a scholastic flavour which is repugnant to many minds at the present day, and it is natural enough to suppose that a discussion of the self as substance possesses little more than anti- quarian interest. That is a very superficial objection, however ; for jbhe substantiality of the self, and the existence of the soul, are, in reality, one and the same problem expressed in different words, and this problem is the inevitable culmination of our enquiry. It is a complex issue, and some of the difficulties have already appeared, but they cannot be adequately appreciated unless they are read in the light of history. The danger of a historical discussion is that argument may be subordinated to scholarship, and theory to the history of theory. But sometimes it is necessary to take the risk, and this is an instance in point. It .may be true that modern philosophy (by which I mean the movement that began with Descartes) lays ' less stress on substance than ancient or mediaeval philosophy, but that seeming objection is really the strongest argument on the other side. The beginnings of this period of philosophical reflection are rooted in substance. ' Omnia vel in se vel in alio sunt ' is an axiom standing in the forefront of Spinoza's Ethics, and expresses the belief, which he held in common with Descartes and Leibniz, that the principle of 301 CHAP. XII IN MODERN PHILOSOPHY 305 substance and attribute is the final one according to which existence should be construed. And if the conception of substance, in some of its interpretations, tended to be discarded later, the reason lay in the exceedingly careful and trenchant criticism which had been brought to bear upon it. If this prolonged discussion, where the disputants were of the first rank, had achieved no resjilt whatever, we should be driven to despair of the powers of the human intellect. It is my belief that the labours of these great men were not fruitless, and that they have shown for all time the most important meanings and the most im- portant misconceptions of substance. True, the question is too vast to be adequately discussed within restricted limits. It is arbitrary to begin with Descartes, as if implying that Plotinus and Averroes, to mention no greater names, had not been,^ or did not count. And if we begin with Descartes, there is so much matter for reflection that the only feasible course is to select here and there with a purpose. I wish to consider a few of the arguments of a few of the giants during this period, not in the vain hope of raising the issues exhaustively, but with the humbler aim of emphasising the most important of these issues, and so of paving the way for a more independent discussion in the next and concluding chapter. At the first blush, the prospect is far from alluring because of the extraordinary diversity of the results obtained. That the self is substance, that it is not substance, that ' it matters not at all ' whether it be substance or no, are the answers of Descartes, Hume, and Locke ^ respectively. That, surely, is sufficient disagreement without mentioning the further modi- fications and subtleties of Kant or Hegel. But results, after all, are of comparatively little moment in metaphysics. They are overshadowed by the ' Essay, Book 11. chap, xxvii. § 10. 306 PEOBLEMS OF THE SELF chap. reasons for the results, and there is enough community in the arguments, and even in the pre- suppositions, of these philosophers to give a real continuity to the history of this controversy, despite their alarming lack of agreement in the end. We shall see how and why it is so, as we go on. Mean- while let me point out one fundamental respect in which this community of principle appears. Every one of the discussions which I have in mind considers the problem of the interpretation of spiritual sub- stance, or the soul, in connection with the general problem of the meaning of substance. One substance is not another substance, nor is a spiritual substance identical with a material substance. But, inasmuch as both claim to be substances, it is necessary to keep the general problem of substance constantly in mind. Every one has heard of the famous argument, eogito, ergo sum, and knows that Descartes was its most lucid exponent, if he was not the first to discover it. A true philosopher, he held all things in pro- visional doubt until he should perceive some truth so clearly and distinctly that he knew he could proceed upon a basis of absolute certainty. In a way he had not far to look ; for he straightway perceived that this doubt and this readiness to suspend judgment was itself a kind of thought, and that thought implies a thinker. The very doubt of thinking is itself a proof of the existence of thinking and the thinker, ' And as I observed that this truth, I think, hence I am, was so certain and of such evidence that no ground of doubt, however extravagant, could be alleged by the sceptics capable of shaking it, I concluded that I might, with- out scruple, accept it as the first principle of the philosophy of which I was in search.' ^ It is an error to suppose that Descartes' philosophy is based upon the eogito. The method he adopts is analytic. Why is the eogito certain ? Because it is 1 Discourse of Method, Veitoh's translation, p. .33. xn IN MODERN PHILOSOPHY 307 perceived clearly and distinctly of itself, or is, in other words, self-evident. By the same warrant we must believe in other propositions which have the same kind of certainty.' These principles (some of which Descartes enumerates here) have an equal claim to be regarded as the foundation of his metaphysics. The cogito shows the type of certainty which a funda- mental principle of this sort must possess, and that is all. There is no suggestion that the other ultimate principles can be deduced from it. But, on the other hand, it is the fundamental self-evident principle refer- ring to existence, and this is the sense in which the importance of the argument is usually understood. Whatever else a man may doubt, he cannot doubt his own existence. He knows that he must be real, and on that basis he may proceed to interpret the rest of reality. In some sense the matter of fact is plain. The question is, in what sense precisely ? Clearly the self-evidence of the cogito cannot lie in the necessary connection between the meaning of thought and the fact of existence, for there is no such implication. If Elizabeth Bennett doubted the sincerity of Mr. Bingley's intentions towards her sister Jane, we are, unfortunately, not entitled to conclude that Elizabeth Bennett existed. The argument can only be an in- terpretation of the nature and implications of a psychical process which actually exists. The character- istic of existence does not follow from the argument, but must be given to it. And the interpretation is not a very simple or obvious matter. The word cogito implies personal thought, and if no more primitive kind of experience is possible, then the self must exist in every act of thought. But that is a statement of fact, and not an argument. If 'floating' experiences were possible, then the argument would only prove the existence of these floating experiences, 1 Discowrse of Method, Veitoh's translation, p. 34. 308 PROBLEMS OF THE SELF ohap. and not the existence of a self. In short, the argu- ment is only analytic. In whatever sense thought goes on, there we have a piece of existence. It is clear, then, that Descartes' interpretation of the cogito must involve other principles than that of the cogito itself, and these principles may be disput- able. There is no doubt as to the precision of his own answer. ' What, then, am I ? A thinking thing, it has been said. But what is a thinking thing ? It is a thing that doubts, understands, affirms, denies, wills, refuses, that imagines also, and perceives.' ^ All these properties, he proceeds, are clearly kinds of thought, and thought is the essence of the self. " For it is of itself so evident that it is I who doubt, I who under- stand, and I who desire that it is here unnecessary to add anything by way of rendering it more clear." ^ This may be the correct psychological description, and the correct metaphysical interpretation, but it goes far beyond the cogito. It is a matter for argu- ment whether the existence of a doubt implies the existence of a substance which not only doubts but also fears, chooses, denies ; and the unity of the self, in Descartes' sense, is very far from being an assump- tion which is so clear th,at it only requires to be stated in order to be accepted. Still less is there the necessary implication of an ego which persists un- changed throughout every pulse of experience. Similarly if a doubt implies a thinking substance, there still remains the problem of the interpretation of substance. Descartes has two definitions of sub- stance which he seems to regard as equivalent. The one is that a substance is a res per se suhsistens, the other that it is the supporter of accidents. The second of these is essentially a logical notion, apply- ing to subject rather than substance. A subject is that' of which qualities can be predicated but, as we shall see more fully later, there is an important ' Meditation II., Veitoh's translation, p. 109. " Ibid. IN MODEEN PHILOSOPHY 309 difference between the logical concept of subject and the ontological notion of substance. The difference, however, is unimportant in this connection since the ego, on Descartes' interpretation, is a substance in the ontological sense, and every substance is a logical subject, not a predicate. But it is clear that there is a long road to travel from the existence of a doubt to the existence of a res per se subsistens which doubts, especially if that substance is other than the experiences which it has at any given time. In short, Descartes assumes a great deal more than he has proved, and the same criticism must be passed on many other discussions of this question. Take, for instance, the following statement from Reid. ' A person is something indivisible, and is what Leibnitz calls a monad. My personal identity, therefore, implies the continued existence of that indivisible thing which I call myself. Whatever this self may be, it is something which thinks, and deliberates, and resolves, and acts, and suffers. . . . My thoughts, and actions, and feelings change every moment — they have no continued but a successive existence ; but that self or / to which they belong, is permanent, and has the same relation to all the succeeding thoughts, actions, and feelings which I call mine.'^ For Reid, who wrote after Hume had written, this assertion is mere dogmatism. One part of Descartes' interpretation aroused strenuous opposition from the first. From the assertion that the self is a thinking thing he pro- ceeded to the further assertion that its whole essence consists in thought, and that it is, therefore, completely distinct from matter. The transition seems startling; but it follows, I think, from his premises. Briefly stated, his argument runs as follows. It is logically possible to conceive the non-existence of matter without invalidating the certainty of the 1 Keid, Intellectual Powers,''Eisaa,j\lll.leha,Tp. iv. §6. 310 PEOBLEMS OF THE SELF chap. existence of the self. The essence of the self, there- fore, does not require the support of matter in any form, whether of the body or of some more subtle material, such as an aura, a breath, or a flame. It follows that matter is no part of the essence of the self, and therefore that the self is only a thinking thing. The self is a res per se suhsistens, the logical ground of its existence lies in its own essence (apart from the co-operation of God). If it once be granted that the content of the self depends upon its essence, and that the support of matter cannot affect the intrinsic character of this essence, then Descartes was justified in his conclusion. He would have thought it absurd to deny that the content depends logically upon the essence. But if the argument is conclusive the reason Ues partly in the barrenness of the content of the self on this interpretation, and partly iu a confusion between ground and cause. Unless cause and ground are identical the self may require physical conditions although these are no part of its essence, and this is the usual view. And when Descartes maintained that the presence or absence of the body does not affect the content of the self, it is doubtful whether he understood by this content or essence anything more than the permanent substance which he believed to continue unchanged so long as thinking persists. It was inevitable, then,S that this theory of the soul should meet with opposition, especially from 'those who, like the English empiricists, tried to hold fast to what is given in experience. Descartes' contention may be true in the end, but, if so, it requires further defence and fuller explanation, and the criticisms of Locke and Hume are valid, at least in so far as they illustrate and emphasise this necessity. In this as in so many other instances, the conflict between rationalism and empiricism paves the way for a better understanding ; and the empiricists also deserve praise for their courage in XII IN MODERN PHILOSOPHY 311 attacking the problem at all, since a position like the Cartesian is often comfortably accepted be- cause it supplies a ready basis for the doctrine of immortality. If the soul is an indiscerptible substance which is also sui generis, it is at least ' naturally ' immortal, i,e. there is no reason in the nature of things why the soul shbuld perish with the body ; for its essence is distinct from that of the body. It ia easy to understand how Locke's tentative sugges- tion that God might ' annex ' thinking to matter led to some ugly mutterings of the drum ecclesiastic. A brilliant writer has maintained that Locke, so far from being a mere empiricist, is really a sort of Kantian, born out of due time.^ That is an over- statement, although he was not a consistent empiricist. Indeed, his doctrine of substance is instructive precisely because it shows both the strength and the weakness of empiricism. Locke saw how empty the doctrine of substance may be on some interpreta!- tiona, and yet his criticism was tempered by the recognition that substance, in some sense, is a necessity of thought. Again, his discussion of substance is especially important from our point of view, for two reasons. In the first place, Locke is careful to apply his doctrine to the ' spiritual substance ' and to personal identity ; and our problem consists in a similar application. In the second place, his * historical plain method ' (in which his empiricism principally consists) is at bottom psychological, and ought to show how far the ' em- pirical self can be regarded as identical with the ' spiritual substance.' We may learn in this way how far the psychological method we have hitherto adopted requires to be supplemented. 'The ideas of substances,' according to Locke, ' are such combinations of simple ideas as are taken to represent distinct particular things subsisting by ^ T. E. Webb, The IntellectuaMsm of Locke, 312 PKOBLEMS OF THE SELF themselves ; in which the supposed or confused idea of substance, such as it is, is always the first and chief.' ^ These ' distinct particular things,' as we learn later, include minds as well as physical things, since ' ideas of reflection,' i.e. ideas obtained through introspection, imply this kind of combination as well as * ideas of sensation.' ' The mind being, as I have declared, furnished with a great number of the simple ideas conveyed in by the senses as they are found in exterior things, or by reflection on its own opera- tions, takes notice also that a certain number of these simple ideas go constantly together ; which being presumed to belong to one thing, and words being suited to common apprehensions, and made use of for quick dispatch, are called, so united in one subject, by one name ; which, by inadvertency, we are apt afterward to talk of and consider as one simple idea, which indeed is a complication of many ideas together ; because, as I have said, not imagining how these simple ideas can subsist by themselves, we accustom ourselves to suppose some substratum wherein they do subsist and from which they do result.' ^ This confused idea of substance was a great stumbhng- block to Locke. He never denies its necessity, although he sometimes appears to take refuge in nominalism. Indeed he admits in his controversy with Stillingfleet that any other sup- position than that which we ' accustom ourselves to suppose' would involve a direct repugnancy. In the present chapter (§ 5) he is scarcely less explicit. It is impossible, he argues, to deny the existence of matter or spirit on the ground that there is no idea of their substance, ' it being as rational to affirm there is no body, because we have no clear and distinct idea of the substance of matter, as to say there ^ Essay concerning Human Understanding, Book II. chap. xii. § 6. ^ Jbid. chap, xxiii. § 1. IN MODERN PHILOSOPHY 313 is no spirit, because we have no clear and distinct idea of the substance of a spirit.' The implication is that there is substance in both cases. Even when Locke speaks of our specific ideas of substances he does not exclude the general idea of substance. That idea is always the same (and therefore is impotent to explain particular problems), but it is always necessary. ' I say, our specific ideas of substances are nothing else but a collection of a certain number of simple ideas, considered as united in one thing.' ^ The union in one thing is always necessary and always confused. We must believe in it ' though it be certain we have no clear or distinct idea of that thing we suppose a support.'^ Philosophers assert that there are substantial forms in which qualities inhere, but this assertion is mere prating. The con- cept of substance never explains the manner of the inherence, and, although it is necessary, it does not aid any particular enquiry. ' It is the ordinary qualities observable in iron, or a diamond, put together, that make the true complex ideas of these substances, which a smith or a jeweller commonly knows better than a philosopher.' * This vacillation is exceedingly interesting. Clearly Locke has a strong case against certain scholastic theories, and against Descartes. It is easy to speak of substances or substantial forms in which qualities inhere, but to talk in this way is not to think, unless it can be shown how and why a particular substance has the particular qualities which it has, and no others. Why does my ego, for instance, possess the attributes of thinking, doubting, willing, and the like, and how does it support them in a unity ? No one has answered that question, for the very good reason that no answer is possible iu terms of this theory of substance. I should linger on the question longer ^ Essay, Book II. chap, xxiii. § 14. Italics Locke's. " Ibid. § 4. 3 ii,i^_ § 3. 314 PROBLEMS OF THE SELF were it not that it must reappear, again and again, in the course of this discussion. Sufl&ce it to say here that Locke's criticism is just and unanswerable. The concept of a substance which is the permanent bond of unity of its properties, including the qualities and ' powers ' of bodies if physical substances are in question, must be able to explain the way in which these properties cohere. If it does not do so it is use- less, because it does not explain what it has set out to explain. The Cartesian substance ought to be an aid to detailed scientific enquiry, since it should be able to explain the specific mode of union in particular things. In reality it is useless for this purpose ; and that criticism is sufficient. It is tempting, therefore, to suppose that Locke ought to have discarded the notion of substance altogether. Part of the reason why he did not do so was that he thought it absurd to say that things were collections of simple ideas. If Locke had been in earnest with his own definition of idea, 'that which is the object of the understanding when a man thinks,' he might have been less convinced of the absurdity. If we mean by ' ideas ' the sense-data or presentations which are given us, there is no absurdity in saying with Berkeley that things are collections of ideas, because each ' idea,' in this sense of the word, is substantial and a piece of existence. It is not an idea of something, but itself something which is before the mind and presented to it. On the other hand, our ' ideas,' in the more usual sense of the word, are of or ahout something, and this other meaning supplies the reason why Locke denied that a substance could be merely a collection of ideas. Accordingly, his insistence on the necessity for substance is partly occasioned by a fundamental ambiguity in the word 'idea,' but the fact is not wholly to be regretted. Whenever we speak of things around us we imply an objective reference IN MODERN PHILOSOPHY 315 to a reality other than ourselves, and this objectivity is an important part of the meaning of substance, and it is very doubtful whether a merely phenomenal view of matter (like that of Berkeley) can adequately explain the implications of objectivity.^ Even if it is impossible to find in things anything except sensory elements, a thing need not be a collection of such elements. The category of thinghood implies that there is a ground for the union of properties which we find. The mistake of Descartes lay, not in what he set out to find, but in the fact that he, as well as others, hypostatised a problem into an active entity, and then supposed that he had given an explanation. The fundamental error is to suppose that substance is simply one thing amongst other things, although of greater importance than they. Part of the difficulty of understanding Locke's argument is that he is not fully alive to the futility of such an attempt, and that he does not see that it leaves the metaphysical problem unaltered. Indeed, he frequently tends to regard substance, in an analogous way, as one idea amongst others. It is clear, however, that a substance cannot be a simple idea in the sense in which colours or sounds are simple ideas. If it were, it would only add another unit to the collection, instead of explaining the union of this set of ideas in a distinct particular thing. And it is plain that Locke was pursued by an ignis fatuus of this kind, and sought to evade the issue by saying that the general idea of substance was 'confused.' But his mistake, in another form, is a very common one, and is found, all too frequently, among his opponents. When the average man thinks of a substance he thinks of it as a thing holding other things together. Let me illustrate the point by quoting another passage from the Essay : ^ We shall see in the next chapter that substance implies a, particular kind of objectivity, and not merely objectivity in general. The aim of the present chapter is only to indicate the lines along which the problem of the substantiality of the self should be attacked. 316 PROBLEMS OF THE SELF chap. ' The little bodies that compose that fluid we call water, are so extremely small, that I have never heard of any one, who, by a microscope . . . pretended to perceive their distinct bulk, figure, or motion, and the particles of water are also so perfectly loose one from another, that the least force sensibly separates them. Nay, if we consider their perpetual motion, we must allow them to have no cohesion one with another ; and yet, let but a little sharp cold come, and they unite, they consolidate; these little atoms cohere, and are not, without great force, separable. He that could find the bonds that tie these heaps of loose little bodies together so firmly ; he that could make known the cement that makes them stick so fast one to another, would discover a great and yet unknown secret ; and yet, when that was done, would be far enough from making the extension of body (which is the cohesion of its solid parts) intelligible, till he could show wherein consisted the union, or consolidation, of the parts of those bonds, or of that cement, or of the least particle of matter that exists.' ^ This line of discussion is clearly wide of the mark. These heaps of loose little bodies are heaps of loose little substances. Therefore Locke's ' cement,' if it were discovered, would not affect the philosophical problem of substance, however useful it might be for science. It would not explain the substantiality of the ultimate atoms themselves. It could not be a cement for their qualities and their 'powers,' or potential influence upon other atoms. In short, it is impossible to regard substance as a thing which holds other things together. That way lies a vicious infinite regress. If the cohesion of particles cannot be explained by their own laws, if the universe must be bound together by some sort of transcendent steel, or cement, or glue, and if this necessity be the metaphysical problem of substance, then, plainly, ' Essay, Book II. chap, xxiii. § 26. IN MODEEN PHILOSOPHY 317 these transcendent entities are themselves substances which require a still more subtle bond, and so on infinitely. Locke came very near to a comprehension of this point when he complained of those who first ' ran upon the notion of accidents.' By giving the accidents a quasi-substantial existence they made the conception of a substance additional to them a tedious superfluity. If we remember consistently that substance and quality (or accident) are correla- tive, and that neither has any meaning apart from the other, we may succeed in avoiding this error. A quality, ex vi terminorum, is a quality of a substance. We dare not isolate the two in such a way as to obscure this reciprocal implication. Of course there are substantial bonds in nature, and the self may be such a bond. Experiences really exist. They are not mere qualities. And there may be an ego, also existent, which unites them in fact. If so, this ego must be discovered by the evidence of fact, and the question of its existence has nothing to do with the metaphysical problem of substance. The experiences are substantial in the same sense as the bond. But if Locke is to be blamed for this confusion, there are many who should hesitate before throwing the first stone. Most of the defenders of substance imagine it in this impossible way, and not a few suppose that they are really thinking in this exercise of their imagination. Let us sum up the results of this discussion. The empiricists were right in asserting that the concept of substance must afibrd a real explanation of the union of the properties of actual things, and their criticism of the rationalists was successful in so far as it showed that a transcendent doctrine of substance does not aflford any such explanation. At the same time there is an objective reference in our thought which cannot be readily explained- by empiricism of the traditional sort, and the principal problem for 318 PEOBLBMS OF THE SELF chap. empiricism is to give an intelligible account of this objective reference. There are two ways in which it is impossible .to find the desired solution. One of them involves the theory that substance is an idea amongst other ideas, or a thing amongst other things. If there is such a permanent substratum of the self or of any other substance, it must be discovered by empirical methods, and the discovery is irrelevant to the metaphysical problem of substance. The other impracticable route is that which separates qualities from substance in such a way that the qualities them- selves tend to be regarded as substances. This conclusion seems merely negative, but, another possibility emerges. Why should not substance, instead of being the bond of qualities or accidents, be simply an expression of the necessity for their union ? The simplest piece of existence has many qualities, and may have no content other than these qualities, but it' is not an aggregate of the qualities. The qualities must be united, and this necessity may be the principle of substance itself. Let us call this a mere suggestion, made at a venture. What is worth remembering is that the suggestion had occurred to Locke. It appears very plainly in some passages in which he discusses the distinction between nominal and real essence. By the nominal essence Locke means the collection of sensible qualities by which we distinguish one thing from another. A lump of gold is yellow, fusible and the like, and our selection of these properties seems arbitrary, nor are we able to account for their union. By the real essence Locke means ' that particular constitution which everything has within itself, without any relation to anything without it.' ^ He compares this essence to the defini- tion of a triangle, and believes that we could deduce ' whole sheaves ' of properties could we discover these essences. If these essences are the ground of the ' Essay, Book III. chap. vi. §. 6. XII ,IN MODERN PHILOSOPHY 319 union of the properties of a thing it is plain that the conception of substance has undergone a revolution. No one would maintain that the essence of a triangle is a thing, or a bond, distinct from particular triangles and somehow holding a triangle together. It is not a support for the properties of a triangle, but a formula for the kind of connection which the properties of a triangle must have. Locke's account of material substance holds, mutatis mutandis, of spirit. ' Thus by putting together the ideas of thiaking, perceiving, liberty, and power of moving themselves and other things, we have as clear a perception and nption of immaterial substance as we have of material. . . . For our idea of substance is equally obscure, or none at all, in both.' ^ Even the infinite substance, God, must be regarded in the same way. The natural iuference from such a doctrine is that the substance of any particular spirit is at least numerically distinct from the substance of any particular body. As Locke says in his first letter to Stillingfleet, ' The general idea of substance being the same everywhere, the modification of thinking joined to it makes it a spirit.' And this is his general view. It was in his mind when, in prov- ing the existence of the self, he used an argument essentially the same as Descartes' cogito. ' It is pa^t controversy that we have in us something that thinks ; our very doubts about what it is confirm the certainty of its being, though we must content ourselves in th« ignorance of what kind of being it is.' ^ But while this is the natural and probable inference, another inference is possible. It is unnecessary to suppose that a distinct immaterial substance exists at all, ' it being, in respect of our notions, not much more remote from our comprehension to conceive that God can, if he pleases, superadd to matter a faculty of thinking, ' Essay, Book II. chap, xxiii. § 15. ^ Essa/y, Book IV. chap. iii. § 6 ; cf. iUd. chap. Ix. § 3. 320 PROBLEMS OF THE SELF chap. than that he should superadd to it another substance with a faculty of thinking.' ^ This view is obviously compatible with Locke's theory, but its very suggestion aroused theological attack. The question in itself is casually introduced and of little importance. Its interest lies in the way it illustrates the futility of this conception of substance, and this point becomes still more manifest in Locke's treatment of identity. If substance is the concept which lays emphasis on the way in which ideas are considered as united in one thing, it should surely throw light upon identity, which is just the problem of ' union in one thing.' On the contrary, we find Locke arguing that the notion of substance does not in any way elucidate the notion of identity. The principium individuationis, according to Locke, is ' existence itself which determines a being of any sort to a particular time and place, in- communicable* to two beings of the same kind.' ^ ' And in this consists identity, when the ideas it is attributed to vary not at all from that what they were that moment wherein we consider their former existence, and to which we compare the present.' ^ It is cleaj, however, that the interpretation of identity will difi'er somewhat according to the subject to which it is applied. An atom, for instance, with its im- mutable size and qualities remains identical because of this immutability. A mass of matter, again, remains identical so long as the same constituent atoms remain in the same configuration. The identity of a living body, however, is of a different kind. It is not meaningless to speak of the identity of an oak or any other organism, but such an organism is com- posed of different constituent particles at different stages of its Life. The identity of an organism is really the identity of the organisation and disposition 1 Essay, Book IV. chap. iii. § 6. " Ibid. Book II. chap, xxvii. § 4. ■' Ibid. § 1. XII IN MODERN PHILOSOPHY 321 of its parts, so far as they partake of one common life. It is difficult to determine whether Locke maintained that the continuance of the organisation of the oak was its identity, or whether he believed that this organ- isation was due to the continued presence of a single unchanging vital principle. In the latter case organic identity would, in the end, be logically similar to material identity. It is the discussion of personal identity, however, which interests us most. Locke devotes considerable care to the problem and, in the first place, he sharply distinguishes the terms ' man,' ' soul,' and ' person.' When -W^e speak of a man we speak not only of his soul, spirit or mind, but also of his body. Man is not merely psychical, he is also physical. The soul of Heliogabalus might, upon the death of his body, descend into a hog, but we could not say that the hog was the same man that Heliogabalus was. For the rest Locke carefully distinguishes the person from the soul or spiritual substance. His account of personality is not without interest and importance ; his account of the soul shows how useless that conception may be, if understood as Locke under- stands it. As Locke's definition of personal identity, though apparently clear, presents several difficulties to more careful inspection, it will be well to quote it at length : ' To find wherein personal identity consists, we must consider what person stands for ; — which, I think, is a thinking intelligent being, that has reason and reflection, and can consider itself as itself, the same thinking thing, in difierent times and places ; which it does only by that consciousness which is in- separable firom thinking, and, as it seems to me, essential to it ; it being impossible for any one to perceive without perceiving that he does perceive. . . . Thus it is always as to our present sensations and perceptions : and by this every one is to himself that y 322 PROBLEMS OF THE SELF ohap. which, he calls self: — it not being considered, in this case, whether the same self be continued in the same or divers substances. For, since consciousness always accompanies thinking, and is that which makes every one to be what he calls self, and thereby distinguishes himself from all other thinking things, in this alone consists personal identity, i.e. the sameness of a rational being ; and as far as this consciousness can be extended backwards to any past action or thought, so far reaches the identity of that person ; it is the same self now it was then ; and it is by the same self with this present one that now reflects on it, that that action was done.' ^ To distinguish this continued self from the soul or spiritual substance, may seem a mere perversity to ordinary minds. It is an implication, however, of the theory which insists on regarding the soul as some- thing over and above its psychical filling, and this is the way in which Locke regards the question. What wonder, then, that he regards the nature of the spiritual substance as of small importance ? Our consciousness is interrupted by intervals of forgetful- ness, and therefore ' doubts are raised whether we are the same thinking thing, i.e. the same substance, or no. Which, however reasonable or unreasonable, concerns not personal identity at all. The question being what makes the same person ; and not whether it be the same identical substance which always thinks in the same person, which, in this case, matters not at all. . . . For, it being the same consciousness that makes a man be himself to himself, personal identity depends on that only, whether it be annexed solely to one individual substance, or can be con- tinued in a succession of several substances.' ^ The theory of reincarnation, as Locke remarks, supplies a very good test. Let us suppose that the neo-Pythagoreans admit the absence of any good y, Book II, chap, xxvii. § 11. " Ibid. § 10. XII IN MODERN PHILOSOPHY 323 evidence of the memory of previous incarnations, and consequently deny that the self which is reincarnated remains the same to itself. This is enough to make Locke deny that there can be any personal identity in these instances, since he contends that ' the same continued consciousness ' implies explicit memory. We have seen, however, that it may be theoretically possible to have unity and continuity of experiences without definite recollection. If so, we may admit that there is .an analogous continuity between succes- sive incarnations. If we say that Mr. Bernard Shaw is Sheridan reincarnated, our reason must be the extraordinary resemblance between Sheridan and Mr. Shaw. But both Locke and Kant maintain that there are several possible interpretations of this alleged fact in terms of the doctrine of substance, and that we have no means of distinguishing between them. The statement of fact is that there is a series of personalities agreeing in certain important respects. But what is the difference between supposing, on the one hand, that a single substance, having these pro- perties, continues, and, on the other hand, that there is a succession of substances whose properties agree in these respects ? The difference appears to be quite arbitrary, and to mean nothing more than the barren distinction between one and many, in a sense that is only numerical. The neo- Pythagoreans, I suppose, would reply that there might be a further distinction. If there is but one substance reincarnated, there is at least the possibility of closer community and, for that matter, there is a basis which might recall slumbering memory in course of time. Locke's rejoinder would be that even memory might be attached to a succession of substances, and the rejoinder is valid enough on the conception of substance against which he is arguing. Let me quote an analogous passage from Kant, which applies to a single incarnation as much as to a 324 PKOBLEMS OF THE SELF chap. succession of incarnations : ' An elastic ball whicli impinges on another in a straight line communicates to it its whole motion, and therefore (if we only consider the places in space) its whole state. If then, in analogy with such bodies, we admit substances of which the one communicates to the other representa- tions with consciousness, we could imagine a whole series of them, in which the first communicates its state and its consciousness to the second, the second its own state with that of the first substance to a third, and this again all the states of the former, to- gether with its own, and a consciousness of them to another. That last substance would be conscious of all the states of the previously changed substances, as of its own, because all of them had been transferred to it with the consciousness of them ; but for all that it would not have been the same person {i.e. substance) in all those states.'^ That is a fair comment on the uselessness of a transcendent doctrine of substance. We distinguish substances according to the unity of properties which we discover in a thing, and unless the theory of substance makes this unity intelligible it is nothing. Accordingly, the empiricists are right in calling attention to the specific kind of unity which we find in substances as we mean and intend them, and in dwelhng upon this unity. One of the principal merits of Locke's discussion is his recognition of the different senses of substantial identity, and when Hume compares the relation of personal identity to the other kinds of identity which we attribute to things, his argument, following as it does upon the footsteps of Locke, deserves still more careful in- vestigation. Hume maintains that, strictly speaking, there is only one kind of identity, and this he calls material identity. If anything persists continuously without ' Critique of Pv/re Season. Muller's translation, p. 316 n. xn IN MODERN PHILOSOPHY 325 any change whatever except in point of time, it is identical. Otherwise there is not identity, but diversity. Common language, however, uses identity much more loosely, and misapplies it to many objects. The merit of Hume's discussion is that he draws examples ' from daily experience and observation ' of these improper, uses of identity, and tries to explain how the mistake occurs. We frequently call things identical when their existence is really variable and interrupted. In the first place, we neglect changes which are inconsiderable in proportion to the whole object we are considering. ' The addition or diminution of a mountain would not be sufficient to produce a diversity in a planet ; though the change of a very few inches would be able to destroy the identity of some bodies.' ^ Again, even if the change be proportionately considerable, we are apt to discount it if it occur gradually and insensibly. We are the more inclined to do so if we suppose that the parts of the object are organised ■for a common end. ' A ship, of which a considerable part has been changed by frequent reparations, is still considered as the same ; nor does the difference of the materials hinder us from ascribing an identity to it.'^ And if we suppose, in addition, that there is a sympathy of the parts for a common end, we may continue to ascribe identity even if there be a total change in the parts of the object. ' An oak, that grows from a small plant to a large tree, is still the same oak ; though there be not one particle of matter, or figure of its parts the same. An infant becomes a man, and is sometimes fat, sometimes lean, without any change in his identity.' * Similarly we call a river the same during many ages although its parts may be totally altered in twenty-four hours ; we call a sound the same though we only hear it ' Treatise, Selby-Bigge's edition, p. 256. 2 Ihid. p. 267. ' Ibid. p. 257. 326 PKOBLEMS OF THE SELF ohap. intermittently ; we may even say ' that sucli a chureli, which was formerly of brick, fell to ruin, and that the parish rebuilt the same church of free-stone, and according to modern architecture. Here neither the form nor materials are the same, nor is there any- thing common to the two objects but their relation to the inhabitants of the parish ; and yet this alone is sufficient to make us denominate them the same.' ^ Hume's theory is that any of these senses of identity (save material identity) is fictitious or, as he puts it, ' an artifice.' The oak and the river are not identical, but we call them identical for two reasons.^ In the first place, they resemble presented objects which really are identical, and, in the second place, the mind which contemplates them feels a smooth transition resembling the continued con- templation of material identity. In general, there- fore, we attribute identity to objects from a natural mistake which is due to mere mental association, and the explanation of substance is, for the most part, a psychological matter of habit and custom. This is not the place to. offer a detailed criticism of Hume's general theory. We might ask him by what right he speaks of an ' act of the mind ' * when he also maintains that the mind is nothing but * a bundle or collection of different perceptions which succeed each other with an inconceivable rapidity, and are in a perpetual flux and movement ' ? * We might ask, again, where we have this perception of material identity with which we confuse the fictitious identities ? Hume believes, plainly, that the only possible identical object is a changeless atom, but it is impossible to find such objects among impressions and ideas. He is too consistent an empiricist to work with anything save what is presented to the mind in the form in ' TreatiM, Selby-Bigge's edition, p. 258. ' Ibid, footnote pp. 204-5. ' Ibid, same footnote. * Treatise, p. 252. XII IN MODERN PHILOSOPHY 327 whiclx it is presented, and it is clear, on reflection, that if we fix our attention for any length of time on an object which we believe to be constant, we are very far from having a constant presentation of it. If the reader is unconvinced, let him fix his attention on the vase in front of him, and he will know. Indeed it is safe to say that if the only legitimate sense of identity is material identity, then there is no identity in the things we are wont to consider substances. A changeless atom may possess material identity, but stars and planets, plants and animals, the eternal mountains and the soul of man do not. And that is enough to give us pause. At the same time it is doubtful whether Hume would have shrunk from this conclusion, even if he had been forced to admit the inadequacy of his psychological account of the apprehension of identity. He consistently refuses to attach any meaning to substance in the transcendent sense. There is no such idea. It is an ' unintelligible chimera.' Bodies we know, and minds we know ; they are what they appear to be, a union of presentations ; but the mysterious tie which is supposed to unite them is a fiction, and it is only an inaccurate expression of the fact that they seem to cohere together. Is a substance a distinct particular thing existing by itself? Every impression, and every idea, is a distinct existence in this sense, and what, then, becomes of substance ? The unity of any particular thing, be it a body or be it a mind, is a necessary part of what we mean by calling that thing a substance, and it may seem that, in view of the extraordinary variations of the manner of identity of difierent kinds of substances, there is no alternative between the transcendent view of sub- stance and the Humian view that substance, in the concrete sense, is only a psychological makeshift. It is true that Hume's scepticism rests, in the end, on 328 PROBLEMS OF THE SELF chap. the double assumption that every distinct perception is a distinct existence and that the mind never per- ceives any connection among distinct existences/ Our discussion of the unity of the self has shown that the first assumption may be understood in a sense that is false, and it is fair to assume that the labours of Kant and others have shown that the necessity for the second assumption can be refuted by reasoned argument. But it is always rash to suppose that Hume's difficulties can be answered completely. In the next chapter I shall try to show that an'immanent view of the unity of substance is tenable, admitting all the specific differences which Hume sets in review, and the reader can decide for himself whether such an attempt can be successful or not. For the moment I shall pass to another point. We saw in our dis- cussion of Locke that the discovery of some ultimate cosmic ' cement,' or some changeless material sub- stratum would not afi"ect the metaphysical problem of substance. At the same time it is usual to maintain that the discovery of such substances would explain the form and feature of nature, and the relative unity of the things we find. If, for instance, there are permanent indestructible entities, such as atoms, molecules or ions, then it is possible to hold that the 'fades totius universi,' to use Spinoza's phrase, is fully explicable in terms of these and their combinations and relations. The discovery of such entities is a matter for ordinary scientific investiga- tion. It does not enter the realm of the transcendent. And it is not an immanent explanation of substance, because it assumes the union of properties in a thing. We may ask, then, whether there is any such per- manent basis for the self If there is, it must be discovered by the ordinary psychological channels, just as the existence of atoms must be proved by ordinary scientific methods, whether or not the existence of ^ Treatise, Appendix, p. 636. xn IN MODERN PHILOSOPHY 329 sucli atoms can be verified singly or only in the aggregate. The most famous passage in Hume's discussion of the immateriality of the soul and of personal identity is that in which he raises this question from this point of view. To say that there is a permanent self of this kind is to say that there is some unchanging experience of which we are constantly aware. ' There are some philosophers, who imagine we are at every moment intimately conscious of what we call our self ; that we feel its existence and its continuance in existence ; and are certain, beyond the evidence of a demonstration, both of its perfect identity and simphcity.'^ Hume has two answers to make to metaphysicians of this kind. They are mistaken in their analysis ; and their analysis, if true, would make the concrete content of mind unintelligible. The first answer depends on introspection as it ought to do, since introspection is the only direct source of evidence in psychology, but it is fair to say that Hume has simplified the introspective problem unduly, by interpreting this supposed permanent intuition of the self in far too narrow a sense. He insists, in fact, that this permanent self must constantly stand before us as distinctly as a mountain peak, whether we are asleep or awake, whether y^ are examining ourselves in a reflective mood or engaged in occupations so strenuous that they would only be hampered, and perhaps destroyed, by anything like explicit self- cognition. It goes without saying that there is no permanent intuition of the self in this sense. If a man, to take the latter instance only, were fighting for his life, he would lose it speedily if he distracted his attention from his sword's point to himself But in our discussion of the unity of the self, and especially of subconsciousness, we have seen that the permanent experiences of selfhood, if there are such experiences, 1 Treatise, p. 251. 330 PEOBLEMS OF THE SELF chap. are of a much more subtle kind, and I do not think it is possible to deny, though it is equally impossible to prove, that there is a thread of permanent experi- ence in the self. Semper idem sentire ac non sentire ad idem recidunt — there is enough truth in that brocard to make it possible to hold that our very familiarity with this permanent makes it impossible for us to describe it accurately, or to notice it usually ; and, perhaps, as Samuel Butler suggests, that is the reason why the average man clutches at a straw like the cogito} What is certain is that this permanent experience, if it exists, is something far less than the self at any given time (still more at all times), and that it is impossible to regard the content of the self as merely the variable manifestation of such a per- manent. This latter is the crux of Hume's second argument. What, he asks, becomes of our particular percep- tions on the theories of those metaphysicians ? ' After what manner, therefore, do they belong to self; and how are they connected with it ? For my part when I enter most intimately into what I call myself, I always stumble upon some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe any thing but the perception. . . . And were all my perceptions removed by death. . . I should be .entirely annihilated, nor do I conceive what is farther requisite to make me a perfect non-entity.' ^ That is the truth, apart from questions of terminology, such as the adequacy of the term ' perception ' to express Hume's meaning. The self is not a pitiful residuum of unchanging experience. It must include all its experiences at all times in the unity in which they occur. And so there is no psychical permanent with ' See Butler's Life avd Habit, chap. ii. 2 Treatise, p. 252. xn IN MODERN PHILOSOPHY 331 shoulders broad enough to support these experiences, Atlas-like, from day to day. I shall conclude this chapter by discussing certain of Kant's famous arguments on this subject. They are stated most clearly in his account of what he calls the paralogisms of pure reason, A paralogism is a logical fallacy, a sophisma figurae dictionis, and Kant hunts the paralogism which attempts to prove the existence of spiritual substance through the four principal subdivisions of the categories. The dogmatists, i.e. those who follow the scholastic tradition, maintain that the soul is substance, that it is simple, numeric- ally identical, and in relation to possible objects in space. These characteristics are included in thcs categories of relation, quality, quantity, and modality respectively. As the soul is substance it must be an immaterial substance ; as it is simple substance it must be incorruptible ; as intellectually single it includes personality; and its relation to objects in space is that of immortality. It will be enough if we follow Kant's argument under the first and second of these divisions. The first paralogism ^ contains an illegitimate transition from logical subject to ontological substance. True, there is unity and continuity in the self. I must be able to say that all my ideas are mine, if I can ever think at all, or, as Kant puts it, the ich denke must be able to accompany all my ideas. We have seen the truth of this in our account of the unity of the self, but, according to Kant, the dogmatists have no right to use it as the minor premiss of the specious syllogism which they accept. That syllogism runs as follows. ' That which is represented as the absolute subject of our judgments, and cannot be used therefore as the determination of any other thing {i.e. as a predicate) is the substance.'^ The ^ This statement of the argument follows the first edition of the Critique of Pv/re Reastm. * Critique of Pure Reason, Miiller's translation, p. 303. 332 PEOBLEMS OF THE SELF chap. ego which appears in the ich denke is such an absolute subject. Therefore it is substance. The minor in this syllogism is true in the sense explained above, and the syllogism itself appears to be formally valid in Barbara. The fallacy really consists in the ambiguity of the middle term, and therefore Kant is right in calling the argument a paralogism. An ultimate logical subject is not necessarily a substance, and the ich denke is only subject, not substance. A substance must exist, and it is im- possible to prove that what is merely formal is an existent entity. The ich denke is only a formal condition of all thinking. It is not a permanent substance, nor is it a substance existing out of time. It is a form of thought recurring in every particular act of thought and nothing more. Consequently there is an ambiguity between- an ultimate subject which is a subs^iance and an ultimate subject which is merely formal, having no content as a substance must. The second paralogism deals with the supposed simplicity of the soul, from whence its indivisibility results, with the consequent impossibility that it can be dispersed with the corruption of the body. The soul, according to this argument is an absolute unity, and cannot be supposed to be merely the resultant of composite motions or beings. It is therefore a simple indivisible substance. But how can the argument sustain this contention ? Not from experience, for experience can never afford an example of an absolute and perfect unity. And not from pure reason, for the basis on which the argument proceeds is only that the formal proposition ich denke holds of every piece of knowledge. That is all the simplicity in the case, and it is illegitimate to argue that the only possible explanation of the fact is that a single indivisible substance controls or produces all our ideas. To call it a substance is to give it a content. But the logical structure of thought, the ich denk^, is only a formal IN MODEKN PHILOSOPHY 333 statement of unity, and consequently has no content in itself. It is possible that a being which appears as simple should really be the result of composition, and, con- versely, that a being which appears to be composite should really be simple. The object of the paralogism is to show how different the soul is from what is corporeal. Now, if bodies in space be what is meant by corporeal existence, then Kant has proved, in- dependently of the argument of the paralogism, that such bodies cannot be the substance of our thoughts, since they themselves are phenomena, and not sub- stances in the sense of this argument. But the noiimenal, or hyperphysical, substance of what appears as matter might, in fact, be not composite but simple, and the appearance of composition be due to the fact that it is revealed to us in space. Such a substance, Kant maintains (as Locke had hinted before, though from a different point of view), might be the substance both of thought and of matter, or of the self and the body. The suggestion of a counter-possibility of this sort may seem contrary to all reasonable probability, but Kant is arguing against a proof for which mathe- matical certainty is claimed, and mathematical certainty disappears with the existence of any counter -possibUity, even of the flimsiest and most fanciful kind. Mathematical proof claims complete certainty, and if that claim must be rejected in any given instance the proof is no longer mathematical, whether or not it is reasonable to reject the conclusion. To discuss the problem generally, it is plain that Kant has succeeded in the comparatively easy task which he set out to perform. He does not, of course, deny that there may be noiimenal or transcendent substance, or a pure ego. On the contrary, he believes that such substance or substances exist, although his reasons for that belief are expressed so 334 PKOBLEMS OF THE SELF ohaj.. generally that it is impossible to determine whether in his opinion the noiimenal realm should be interpreted monistically or pluralistically. Kant himself passes from substance to substances without, apparently, recognising the importance of the difference. But he denies, and rightly, that the existence of the soul, in this transcendent sense, is logically demonstrable by any argument of the type of these paralogisms,^ It is impossible to prove the existence of anything, unless some evidence for its existence can be derived from the senses, internal or external, and, lq the present instance, there is no evidence of sense which can support the weight of the transcendent super- structure which is placed upon it. Even if the ich denke were not the purely formal (and, indeed, analytic) principle which Kant main- tains it to be, and even if there were some permanent features of the. self which introspection could discover, it would be impossible to prove from these grounds that the self could, or must, persist, when experience had ceased, or that it was an indivisible unity which could account for the whole procession of experiences which we call the empirical self. The conclusion, in that case, would extend far beyond the available evidence. All that would be required would be a self which had enough unity and enough permanence to explain the facts of psychology. And, as Kant remarks, if the pure ego (not, of course, the empirical) .were an indivisible substance it might gradually disappear into nothingness in the same way as an intensive quantity may disappear ; and an intensive quantity is just as indivisible as a self. Or, for that matter, the . indivisible substance might cease to exist suddenly and all at once. To suppose that the indivisible must also be indestructible is to cheat ourselves with words. ' See e.g. the statements in defence of the First and the Third Analogies of Experience, respectively. IN MODERN PHILOSOPHY 335 • We may conclude by considering the Hegelian dictum that the self is ' subject and not substance.' The implication of this aphorism is that a substance must be interpreted after the fashion of a physical thing, and that it is a sort of Use majeste to apply the principles of corporeal thinghood to the self. Thought, expressed by the term I, is not an abstract' formal principle, as Kant supposed. Nor is it something which exists merely in itself. It is also for itself, an active universal. ' Mind is essentially active in the same sense as the schoolmen said that God is " absolute actuosity." But if the mind is active it must, as it were, utter itself. . . . The mind, of all things, must be looked at in its concrete actuality, m its energy and in such a way that its manifesta- tions are seen to be determined by its inward force.' ^ Or, again : ' It cannot be denied that predicates like simplicity, permanence, etc., are inapplicable to the soul. But their unfitness is not due to the ground assigned by Kant, that Reason, by applying them, would exceed its appointed bounds. The true ground is that this style of abstract terms is not good enough for the soul, which is very much more than a mere simple or unchangeable sort of thing. And thus, for example, while the soul may be admitted to be simple self-sameness, it is at the same time active and institutes distinctions in its own nature. But whatever is merely or abstractly simple is g,3 such also a mere dead thing.' ^ The Hegelian universal, splendid, active, self- completing, is fitted to arouse admiration in some minds, and something akin to despair in others. But the Absolute is not the human mind, nor the human mind the Absolute. If the Hegelians are right in contending that substance is too narrow a category to express the nature of mind, we must also ^ Hegel, Smaller Logic, Wallace's translation, p. 69. ^ Ibid. p. 97. 336 PEOBLEMS OF THE SELF chap, xn remember that they maintain that personality is too narrow for the truth. We may readily admit that, if the self is substance, its substantiality is not identical with that of a physical thing. And, again, if it is substance, it is not invariable to the point of tediousness, or barren to the point of simplicity. None the less the self is substance. We shall see in what sense. CHAPTEE XIII THE SOUL Theoretically there may be a 'psychology without a soul. It is never possible to predict the degree of abstraction which is legitimate in scientific enquiry, and, just as the student of physics may be justified in neglecting the philosophical problems of the nature and existence of matter, so the psychologist may be justified in neglecting the metaphysics of the soul. But problems, like men, do not die because it is sometimes convenient to ignore their existence, and, fortunately, if some societies cast them out, there are other companies to welcome them. Men will continue to ask whether the soul exists and what it is, and they have a right to ask the question quite apart from the conventions of science. In the present chapter I wish to maintain that there is a soul, and, what is far more important, to try to consider what the soul is. The historical investigations of the previous chapter kept this aim singly in view. If there is a soul it must be a substance, immaterial and existing in time. Where any of these features is lacking, there is no longer a soul but something else. We have seen already, however, and shall see more clearly in the sequel, that a mere definition of this sort can carry us but a little way. There are many senses of substance, and some impossible senses. The soul is a substance, but is it a permanent or in- 337 Z 338 PROBLEMS OF THE SELF chap. divisible substance, and in what sense is it superior to matter ? The soul is immaterial, but is not that, in the end, only a negative characteristic ? The soul is in time, but must time be regarded as absolutely real ? Above all, what is the relation of the soul to the procession of experiences which constitute the empirical self? We have a right to ask for a detailed and precise answer to these questions, and there is only one way of attacking them. Let us ask what substance is, and then we shall find whether the self be substance or not, i.e. whether and in what sense there is a soul. That is the plan and purpose of this chapter, and in it I shall assume the reality of time. It is convenient to begin with the two definitions of substance given by Descartes. According to one of them a substance is a res per se suhsistens, or, as Locke puts, it, a distinct particular thing existing by itself. According to the other it is the supporter of accidents. The latter phrase owes its origin to the logical doctrine of the predicables. Accidents are the non-essential qualities which differentiate particular things, and therefore the second definition refers to the essence or substantial form rather than to the substance itself. The substance Tray, for instance, has all the qualities which are common to the canine species, and also a number of others, such as his honesty or the colour of his coat, which mark him ofi" from other dogs. It is plain, then, that the latter definition is not equivalent to the former. There is no existent thing which has only the properties common to its species. We must not rob Tray or Fido of their distinctive characteristics as individuals — poor things, perhaps, but their own. But perhaps we can amend this definition by saying that a substance is the supporter of all its qualities and all its modes. A mode is defined as something which cannot exist by itself, but only as a state or condition XIII THE SOUL 339 of a substance. In this sense particular experiences may be modes of the soul. Even in this sense, however, the second definition is not equivalent to the first. A substance requires its modes or qualities as much as they require it. A substance possesses its qualities, but, without them, it is nothing ; and, in any given case, a distinct par- ticular thing is recognised as distinct and particular in and through its qualities. What would Tray be without his qualities ? And how could we know him as a canine, and not as an equine or a bovine substance, if it were not for these same distinctive qualities ? A substance without its qualities would be unrecog- nisable, and it would be impossible to explain how it has its qualities when it has them, or how it may or may not have some of them at different times. Unless the being of a particular substance is to support one particular group of qualities and no other, the whole conception of substance is either meaningless or totally useless for the purposes of explanation. This does not imply that a substance is merely its qualities, or the sum or the unity of them, but only that a substance distinct from its qualities is never an existent thing. And, again, we must distinguish clearly between the logical notion of a subject of which qualities may be predicated, and the ontological notion of substance. Some logical subjects are obviously not substances ; and if I say, for instance, that whiteness is pleasant to the eye, or sweetness agreeable to the taste, I cannot be supposed to imply that whiteness or sweetness are substances. Only ultimate subjects can be claimed to be substances, i.e. those terms which are always subjects, and never can be predicates. And, at first sight, substance and ultimate subject seem identical. An individual substance, when we refer to it, must be a subject. If I refer to Charles II., I must ascribe predicates to him, and say that he was clever, or witty. 340 PROBLEMS OF THE SELF ohap. or lax in morals. It is his inalienable privilege to be the subject of a proposition. He can never be a predicate of any other subject. But although all substances are ultimate subjects, the converse does not necessarily hold. There are ultimate subjects which are not substances. Let us take, for example, the demonstration in Euclid L and 5, known as the fons asinorum. This demonstration consists entirely of the implication of universals and so it is not a substance. But it is also not a predicate. The universals and their implications are real in their own right, independent e.g. of actually existent space, or existing things in space. Hence they are not, strictly speaking, predicates of anything. The demonstration deals not with qualities but with universals, and universals need not be qualities. Some readers, doubtless, will challenge this asser- tion, but, if they do, they must at least admit that a subject is ultimate, in this sense, because it is a substance, and not vice versa. Charles II. was an ultimate subject because he was an individual human being ; it is false to call him an individual, because he was an ultimate subject. And the point is of very considerable interest and importance. In every judgment there is a reference to reality or, more shortly, there is objectivity, and this objectivity is a great part of what we mean by substance. The hall- mark of objectivity is the constraint and control of thought. We believe in the reality of what is revealed to our senses because, will we or nill we, we can only control the order and connection of what appears in this way to a very limited extent. We may choose to open our eyes, or we may choose to keep them shut ; and if we choose to open them we may con- centrate our attention on some part of the field of vision to the neglect of the rest. So far we have the power of selection, and the selection rests with us. But when we have opened our eyes and con- XIII THE SOUL 341 centrated our attention we can select no longer. Something is given to us which we can only discover and follow. We cannot create it. Smith may choose to go to America, or he may choose to stay at home in Clapham. But when he has chosen, the independent reality of America or of Clapham is burned jipon his soul. Objectivity means the reference to a reality which is real on its own account and is not merely dependent on us. And that is why resistance, or counter-pressure, makes us believe most firmly in the existence of physical objects. If Dr. Johnson had not felt the stone when he kicked it, he would have been less satisfied with his refutation of Berkeley. In the same way the self has a reality independent of any particular act of introspection, although we can choose whether we shall reflect on ourselves or not. Accordingly, when we speak of a substance we regard it as independently real if we do not, indeed, consider it as a solid thing which can resist our bodies ; and we naturally suppose that anything which controls or limits our thought in this way, i.e. any- thing objective, is a substance or part of a substance. If we reflect a little more, however, we shall see that this interpretation is unnecessary. Objectivity, or the reference to reality, is not necessarily a reference to existence, but substance is an ontological term restricted to the existent. Some truths are in- dependent of the things which exist. It would be true that 2 + 2 = 4, whether or no any objects existed to which the concept of number could be applied. Or again, these non-Euclidean geometries which deny the axiom of parallels and seem otherwise astonishing and arbitrary are not really so. They do not pretend to be a complete accouiit of perceived space, or existent space in any sense. They are concerned oijly with the inferences which follow from certain axioms and assumptions, and although there is liberty in the selection of principles, there is no liberty as to the 342 PROBLEMS OF THE SELF meaning of the principles, or the inferences which follow from them. It is not the existent only which can control our thinking. Universals can do so also, and therefore have an independent reahty of their own. Reality and truth are wider than existence. The reason, probably, why this distinction is so frequently neglected is that the things which are known through the senses seem to control our thought, and especially our action, so much more definitely than the universals which are the objects of pure thought. There are comparatively few men who, in respect of action, tie themselves down to principles, and it sometimes seems as if those who bind them- selves in this way are cumbered by dry withs of their own fabrication. And although proof must certainly proceed according to principles and is very far from arbitrary, it may seem that our thoughts have such an unlimited domain in which to roam that it is absurd to speak of them as constrained or controlled at all. That is an error, but it is a natural error, and it obscures the fact that reality is wider than existence. We may take it, then, that while the reference to substance is a reference to reality, it is also more than that. It is a reference to existent reality, and every substance is a piece of existence. How, then, are we to distinguish existent reality from reality in the widest possible sense, including anything for which any kind of objectivity can be claimed ? Subjectively or, more generally, from the point of view of the sources of knowledge, the criterion is plain enough. We have no right to believe in the existence of any substance unless part of the evidence for its existence is derived from the senses. That is the principal function of the senses from an epistemological standpoint. They are not merely confused thought or obscure logic, as Leibniz and others supposed. They bring us into contact with existent reality, and we cannot refer to existence xm THE SOUL 343 without somewhere feeling the touch of the senses. The interpretation of existent reality is another question. We have a right to believe in much that is never directly perceived, and, possibly, the material universe cannot be perceived as it is. But our point of departure must be something which is given to the senses, however obscurely and inadequately. If we conclude that suns exist beyond the reach of the telescope, the reason is that these suns must exist if the behaviour of that which is within the reach of the telescope can be made intelligible. This fact may serve to dispel an objection which has doubtless occurred to the reader. If a substance must be a piece of existence, what becomes of imaginary substances ? Hercules, King Arthur, the Jabberwock, the philosopher's stone, never existed. But do they not belong to the same category as Julius Caesar, or radium, or a dinosaur ? And if they do, must not the category of substance have a wider range than that of actual existence ? Certainly, it is clear that the status of imaginary objects in the realm of existence represents a problem. They do not exist if we mean by existence the occupation of position in a single uniform world in space and time, but that is an interpretation of existence much narrower than our previous argument has given us any right to suppose. It is tempting to think that such imaginary objects are parts of the mind of him who invented them, but that is impossible since they are clearly something which he creates and to which he can, lingeringly, direct his thoughts. They have, in fact, at least the same status as his dream presentations, and these are not parts of the dreaming mind. But they need not exist in any further sense. More importantly, they appear to their author in the way of sense, and he clothes them with the body and the raiment of sense. They are a piece of con- crete sensory imagination, tricked out in the same XIII THE SOUL 345 the idea of what we conceive to be existent. To reflect on anything simply, and to reflect on it as existent, are nothing different from each other.' ^ In other words, existence is not a predicate. It is only a formal expression of the claim which every judgment makes to hold of reality. But while we have seen that there must be this objective holding in reality, if our thought is to be constrained and controlled, we have also seen that reality is wider than existence, and, therefore, that it is impossible to identify ex- istence with objectivity. Objectivity, not existence, is part of the formal claim of every judgment. The term ' particular ' is a way of describing this further element, which distinguishes the existent from the merely objective ; but it, also, may be ambiguous. When we speak of anything as particular we may only mean that it is specific, i.e. that we have enumerated the features peculiar to it, as well as the essence which it possesses in common with many other things. According to the principle of the identity of indis- cernibles, particularity can have no other meaning. It is impossible that there should be merely a numerical difference between two things. If there are two things there must be a qualitative difference between them which is more than numerical, and justifies us in calling them two. If they had precisely the same qualities they would be identical, and there- fore there cannot be any two things which differ solo numero. If the principle of the identity of indiscernibles were true and if particularity had no other meaning than this, then we would not explain the meaning of existence in any way by saying that whatever exists is particular; for universals are particular in this sense, i.e. they are specific. But there are very few who believe in the identity of indiscernibles, in a strict and ultimate sense. There may be no two things ' Treatise, pp. 66-67. 346 PROBLEMS OF THE SELF chap. in the world which are precisely similar, but that is no disproof of the logical possibility that there might be two such things. And particularity seems to have a further meaning, even ia a purely logical sense. There is a difference between universals and the instances of universals. Each universal has, logically, an infinite number of instances, and the instances are particular in a sense in which the universal is not. They are not more specific than the universal, but they are particular. Since substance and existence are ontological, however, and not merely logical, it is plain that particularity must have yet a further possible mean- ing if it is to be able to differentiate the existent from a mere logical instance. Nothing which exists is a bare particular, mere matter without form, but, on the contrary, everything which exists is particular in an ultra-logical way. The form and the qualities which anything has are necessary to its existence, but, being universal, do not explain the ultimate particularity of existence itself. There must be matter, iiXr], stuff, and our next task, accordingly, is to explain what this stuff csin mean. I have shown in the previous chapter that this stuff, or substratum, or support, cannot be regarded as itself a substance. It is not a distinct particular thing, but an element in any particular thing, in virtue of which that thing is ultimately particular. And we have seen at the beginning of this chapter that the form of anything is not its substance either. A distinct particular thing requires matter as well as form. It is unfortunate that the word matter should be so ambiguous. When inorganic substances are described as matter (and this is the usual sense of the word), the meaning of the term is quite different from the present meaning, which is that of an element correlative to form. That is why stuff or vXtj are terms which are less ambiguous, but even they are XIII THE SOUL 347 not wholly free from ambiguity. It is possible, how- ever, to give a precise expression to the meaning which is relevant here, and I shall do so by means of an example which is readily iatelligible. Discussions on Free Trade and Tariif Reform have called attention to the theoretical diflSculties involved in distinguishing raw material from manufactured articles. True, it is irrelevant for our present purpose whether the distinction can be drawn with sufficient accuracy for the ends of the Budget. The relative distinction drawn in the statistics of the Board of Trade between articles wholly or mainly manufactured and other articles, may be sufficiently precise for the Chancellor of the Exchequer. But the theoretical question is of some general interest, and is particularly appropriate in the present connection. Leather cushions, door handles, and so forth, are raw material when fittings of a motor-car are in question, but the upholsterer or the smith will justly regard them as manufactured articles. Again, the skins supplied to the tanner are raw material for him, but the farmer may surely consider them as manufactured or, at least, artificial products. He has to feed the beasts in such a way that their skins are available for sale, and some one has to flay the carcases. Similarly it is arbitrary to maintain even that the ore which is sent to be smelted is raw material, pure and simple. Why should we neglect the labours of the miner ? This kind of difficulty is the same in principle as the metaphysical question we are now discussing. The rawest of raw material, the ore as it exists in the bowels of the earth, or even the primitive nebular substance from which this ore has been manufactured in the enduring forge of the cosmos, is not mere vXr) or mere stuffi Raw material is stuff only in a relative sense. However raw it is, it has shape and form when it exists. Stufi" pur sang is utterly formless and utterly 348 PROBLEMS OF THE SELF ohap. featureless, and that is why it can exist only as an element in substance, not as substance itself. It is an ultimate element in any substance, but is not itself an ultimate thing, not even an atom, for that, if it exists, is a thing having definite properties. And, being formless, this element is unimaginable, but not therefore unthinkable. It is an element correlative to form and as essential to substance as form is. Only in the union of these twain is there substance, and neither can be neglected. We may use stuff in a relative sense, as when we say that linen and eider- down are the stuff of a pillow. This is not stuff in its ultimate meaning, — the meaning which is required for the metaphysics of substance. The simplest expression of the nature of a substance is to say that it is an xa or an xh, where x stands for the stuff, and a or 6 for the form of the substance. The x is not problematical. . There can be no substance without this X, and the existence of the x is certainly required. Moreover, although the x is indeterminate it need not be indifferent to its ' form.' It is unnecessary so to interpret it, that any a or any h can hold of it. We have been proceeding upon the assumption that a substance is a res 'per se subsistens, a distinct particular thing existing by itself, and have been occupied in explaining some of the characteristics of this definition in a quite general sense which holds of all substances, whether the soul or any other. To .sum up, we may say that thinghood implies the presence of the elenient of stuff" or v\r), and that there is no necessity, of any kind or in any case, for construing this element as something transcendent. Such a construction is always unnecessary, and it never explains. But the element of form is equally necessary and equally important. We recognise substances as distinct and independent because of their form, and qualities, and unity. Accordingly, there are two questions which remain as integral parts of this THE SOUL 349 general discussion. Wherein consists the distinctive unity of particular substances, and wherein consists their independence ? The form and the unity is not the whole meaning of substance, and there may be different substances having the same form because the stuff of them may differ. But a particular thing must have a specific unity of its qualities, and this problem is as little transcendent as the former. Let us proceed by the way of examples, and see whether there is any common kind of unity which we invariably ascribe to what we call particular things. If we begin with physical things, we find that the clearest examples are artificial products, such as chairs and tables, or cups and saucers. These have definite, precise boundaries in space, or we believe that they have, and we suppose that they continuously preserve the same shape, the same configuration of particles and so on. Indeed we might be inclined to agree with Hume in contending that a thing of this kind can only remain the same, strictly speaking, when no particle in it is destroyed or altered, and the position of each particle, relatively to the rest, remains the same. This belief, however, would imply that these things must, in the end, be composed of certain least units of matter, and there is no suflBcient reason for believing that such ultimate units exist. If matter is continuous, and there are as good reasons for supposing it continuous as for supposing the space which it occupies to be continuous, then the existence of actual infinitesimals is mere nonsense. The probabilities are that there is no least unit of matter, and that it is as arbitrary to subdivide matter into atoms as to sub- divide it into things of far greater bulk. It is only a question of big or little substances each containing their form and qualities, and each containing their stuff or vXt). a substance does not cease to be distinct and particular because it is possible to separate its parts. When it is dissolved it ceases to be that 350 PEOBLEMS OF THE SELF particular substance which existed before, although something may remain. Before it is dissolved it is a single substance, despite (or perhaps because of) its complexity. We do not believe, in practice, that the identity of a substance is annulled if there be some change in its constituent parts. But practice is hardly the arbiter of metaphysical problems, and our practical criteria of the identity of particular physical things seem to have very different degrees of stringency. As long as a table can be used as a table we should usually say that it has remained the same table. We should call it the same table although it had been revarnished, and repainted, chipped a little in places, and allowed to become rather unsteady. Indeed, if an old leg had been broken, and a new one substituted, we might still deny that there had been any absolute bar to the identity of the^ table. Until the table actually falls to pieces, or is broken up into firewood, it remains the same. Similarly a book would be considered to have remained the same even if the print had faded some- what, and the pages become mildewed. The reason is that any change which is gradual, or small in proportion to the whole, is not supposed to destroy the identity of a thing. And there is no reason why the mere fact of change should be a bar to identity. There is a counter-argument, familiar to students of metaphysics, but that argument may be easily refuted. A thing, we are told, is defined by its qualities : it remains the same only if it has the same qualities : and so, if it changes, i.e. comes to have different qualities, it must become a different thing. This paradox is a confusion which only appears to be logical because of a significant and all-important omission. The question is not that of having a quality and also not having it (which would be a contradic- tion). It is a question of having a quality at one time and not at another (which is plain matter of fact). XIII THE SOUL 351 To say that Simpkins has the quality of baldness is incomplete unless the time at which he is bald is specified. The fact that he was not bald between the ages of, say, one and thirty-one, does not show that Simpkins is a fiction and a mere contradiction, being both bald and not bald. In fact, it is always necessary to specify the time at which a particular substance has a particular quality, and when that is done the problem disappears, and we may rescue the reality of Simpkins from this particular danger. A substance has all the quahties which it possesses at any time during its existence, but it cannot have them all together. A thing is or is not the same, according to the unity and continuity of properties which it has at any one time and at any other. But while the theory is consistent enough, the standards in practice seem very arbitrary. The ordinary man might admit the identity of a book if it had lost its title-page, and be doubtful on the point if the book had been rebound. The book collector, on the other hand, would admit the rebinding in certain styles, but would consider the loss of the title- page the destruction of the identity and the value of the volume. The procedure is not entirely arbitrary in either case. Binding makes a diflFerence, as every schoolboy knows who receives his prize on closing day. But, on the other hand, some books must be rebound in order to be preserved at all, and the title- page is the only good evidence of their identity. There is usually some reason for our standards of continuous identity, but there is also a great deal of convention. The identity of an old painting is seriously impaired if it has been restored, more or less terribly. But the painting may become itself again if the restoring touches are removed. There is a reason for this, again, but there is also convention. If the original colours have faded the painting may be more nearlv itself when it is restored than when it is not. 352 PEOBLEMS OF THE SELF chap. When we come to consider gradual change a special type of difficulty occurs which we may call the difficulty of the margin. A table may change considerably in a century and still be called the same table. If it changed as much in a single night we should send it back to the cabinet-maker as a fraud and a delusion. But how can the suddenness of a change affect the identity of a thing except by an arbitrary convention ? Moreover, a further perplexity seems always to arise with regard to changes which are gradual. Let us suppose that something has existed continuously for ten years, and has changed gradually during that period. Let us describe its state at the beginning of this period as x, its state after five yea,rs as y, and after ten years as z. If we confront y with x we should say that both were states of the same thing. If we confront z with y our verdict would be the same. But it will frequently happen that, if we confront x with z directly, we cannot see that there is sufficient warrant for ascrib- ing identity. Let us consider the classical instance of Sir John Cutler's stockings in Arbuthnot's story. Through continuous darning these stockings became all green silk without one thread of the original black silk left. There is some identity if we compare any intermediate state of the stockings with either of the end -states. But what of the relation of the end- states to one another ? In this case there seems to be no identity, in any important sense. Yet, surely, things which are identical with the same thing are identical with one another. The difficulties of the margin are not so serious as they appear : in fact they may be overcome without any lack of precision. If identity is exclusive of change then none of the things we know are identical for any length of time. But if identity includes change then there will be identity where there is unity. The only question, therefore, is the degree of xiu THE SOUL 353 unity which is sufficient for the identity of substance. If, to take our previous example, we require the same degree of unity between x and z as between x and y, or between y and z, then we shall not be able to find it. This principle applies to any substance which we consider the same. Does a cloud remain the same cloud from the time it appears no larger than a man's hand to the time when it darkens the heavens ? Is a river the same when it is swollen with melting snow in spring and when it is but a trickle in summer? Is a ship the same when new turbines have been fitted into it and a new rudder ? The answer is that it is or is not, according as the degree of unity which is meant has been specified. Until that is done, the question cannot be answered. There can be no precise answer until a precise question is put. It might be replied that no one would maintain that clouds or rivers are substances, or distinct par- ticular things existing by themselves. They are clearly collections of substances, collections of drops of water or of something still more minute. I do not know whether these drops have a better right to be called substances than the cloud or the river, but if they have such a right they have it only in one sense. The drops have a closer unity and one which exists as a unity longer than the cloud or the river. In that case it is generally wise to call the river many rather than one. But the theoretical problem is precisely the same. A flash of lightning, so long as it exists, which is but for a moment, has a right to be considered a thing. And a river is a thing if we specify the unity between its volume at different times and mean no more than that unity by calling it the same. The river is either one or many, accord- ing as it has the kind of unity which is specified ; but it may be better worth our while, and there may be more lasting opportunities for investigation if, in 2a 354 PEOBLEMS OF THE SELF chap. this case, we consider the many rather than the one. And if an atom always retains the same shape or weight it remains a thing in the same sense, whether or not it is a more useful thing to consider, for scientific purposes, than the others. There is no doubt, then, about the precision of this reply. A substance is not merely the unity of its attributes, because any substance must also contain some element of stuff or vXr), but the problem of its identity is that of the unity of attributes in that which occupies time. The general idea of substantial unity, the minimum of unity required for the existence of any substance, need not mean very much. The specific unity of specific substances may mean a great deal, and when we ask whether a given substance has remained one and the same, or whether its identity is broken, we must always specify the kind and degree of unity of attributes which we mean in any given instance. When we say that a thing is one we mean that it has a certain type of unity, and we have a right to say that it is one if that statement, in any way, aids us for purposes of explanation. That is quite different from saying that substance is a practical makeshift dependent upon practical purposes. If, for any purpose, we have a right to consider a substance as one, then it must really be one. It need not be one in a more fundamental sense than it is many. But that is beside the point. The reader, however, may hesitate to assent to this theory, because he may consider it impotent to explain certain difficulties of the margin which still remain. Let us take the case of Sir John Cutler's silk stockings. Every one will admit that the end- states of these garments are not identical in any important sense, while the intermediate states are, perhaps, identical with either of the end-states in some degree ; but how is it possible to say pre- cisely whether the stockings are or are not identical THE SOUL 355 unless we can say precisely at what point the identity ceases ? And there cannot be identity in the series taken as a whole. I had almost said that this is an illegitimate demand for mathematical precision where that type of precision must be lacking. But I should be wrong in saying so, because mathematics, which is the soul of precision, can meet the same difficulty in a precise way. The mathematician saiys that ' in the neighbourhood of a the function /"(a;) approximates to c within the standard k. This means that some interval can be found which (i.) includes a not as an end-point, and (ii.) is such that all values of/ (a;), where x lies in the interval and is not a, differ from c by less than k. . . . For example, in the neighbour- hood of 2 the function x ^ approximates to 4 within the standard -5. For (1-9)' = 3-61 and (2-1) ' = 4-41, and thus the required interval 1-9 to 2-1, containing 2 not as an end-point, has been found.' ^ We may apply this precise mathematical conception to the case of the silk stockings. All we have to say, in order to obtain complete precision, is that the nearer any term is to an end term of the series the more closely does it approximate to that term. If we ask whether the stockings are identical we must specify the standard of identity and the degree of approxiination to it which is sufficient for identity. The precision of our answer will not be affected by the impossibility, if such there be, of pointing out any definite particular term where the break in identity occurs. This would always be impossible if there were continuity in the transition. We can always specify a poiat which is clearly identical with one of the end terms, according to the degree of identity which is meant, and similarly we can specify a point more remote from that end term where, ' Whitehead, Introduction to Mathematics, Home University Library, pp. 160-161 (with some verbal changes and omission of certain illustrations). It is unnecessary to consider any more elaborate discussion. 356 PROBLEMS OF THE SELF equally clearly, there is no identity. And there will be an intermediate range of values where the identity is doubtful. It is possible to state this fact clearly and precisely by choosing some point at random in the intermediate range, and saying that the passage from one substance to another occurs within the range in the neighbourhood of this point. Such an answer is precise and free from theoretical difficulty. So much for the identity of physical things. When and so long as this identity can be found, there a particular substance exists, and it may exist independently, i.e. on its own account. We have seen that anything objective is, in part at least, inde- pendent of us, because it controls and constrains our thinking ; and things may also be independent of one another. Independence does not imply absence of relation. It is like the independence of the free man, not of the hermit. The free man is not unrelated « to other men or to things, and he could not exist without them. But he is not part of them, nor can he be explained fully in terms of them. He acts and thinks on his own responsibility. Physical things are not responsible, but they may exist on their own account and be irreducible to other things, and this is all that need be meant by the inde- pendent existence of a substance. There is no need for any further explanation of the substantiality of physical things and, in particular, no need for any transcendent explanation. To suppose that a particular table-substance must exist to keep a table what it is, or a particular river- substance to keep the Ganges within its banks, or a particular atom-substance to preserve the identity of an atom, is mere folly. The unity of substance in this sense is as ultimate and as immanei\t as causation, and, however little it may mean in general, it may mean a great deal in particular cases. The reader may remember Stevenson's lines : XIII THE SOUL 357 ' O ! I wad like to ken,' to the beggar wife, says I, ' The reason o' the cause, and the wherefore o' the why, Wi' mony anither riddle brings the tear into my e'e.' ' It's gey and easy speirin',' says the beggar wife to me. I have discussed these problems very fully because of their great importance for the problem of the sub- stantiality of organisms and, more especially, of the self. The general theory of substance holds of these as well as of physical things. Organisms exist in rerum natura. They are distinct particular things and, therefore, they must contain stuff, and are not merely a unity of universal qualities. But the chief problem in their case is that of their identity. "We consider that the identity of a table depends, in large measure, upon the proportion of the original parts retained. Since we admit substitution of parts this criterion must be insuflScient, but there is no reason why it should not enter in the majority of cases. With organisms it is otherwise. Except for insignificant exceptions no part of the material of the embryo, or even of the adult organism, remains unchanged until death. There is constant remodi- fication, a constant give and take with the environ- ment. Indeed, it has been a commonplace of physiology, at least since the time of Mayer, that the whole of the energy liberated from the body can be traced to sources outside the body. But there is continuity of development following a certain plan, and this continuity is seen, sufficiently diversified, from the original nucleus to the last mortal breath. Because of this unity and organisation the organism is called one, and it has every right to be called one on our theory. We require a particular equine substance to explain the development of Bucephalus just as little as we require a cannon substance to explain the identity of Mons Meg. The fact of substance is that a certain organisation of matter tends to continue in that organised form in spite 358 PROBLEMS OF THE SELF of, or because of, its give and take with the environ- ment. When I say that an organism has a right to be considered as one, I do not mean to deny that it can also be rightly considered as many, or as a part of a larger whole. It is one, it is many, it is such a part. Organisms are composed of cells, and cells are the stuff of organisms, in the relative sense of stuff which means raw material. We have a right to call a cell a single substance. It is so. But we have an equal, and probably a better right, to treat an organism as one, because its unity is more remarkable and more important. What I have said of selection holds in this instance. We may select this or the other point of view, but each point of view must correspond to a real difference in fact, and it is the difference in fact which is relevant to the doctrine of substance. Simi- larly, although organisms are independent identical substances they are also continuous with a larger unity. There is continuity of the germ-plasm ; the individual is the bearer of the properties of the race. In some respects he is no more independent of the race than a blossom is independent of the tree, and it is possible and legitimate to regard all living things as but transient offshoots of living substance, whether that substance, in its turn, is ultimate or not. An organism is or is not one and distinct, if that unity and distinctness be precisely defined. Otherwise the statement is meaningless. Its distinctness is com- patible with its continuity with a wider whole and with its differentiation into parts. And no other meaning of substance is possible. And what of the soul ? Is it not surprising that while most of us scoff at the supposed necessity for an equine substance to account for the existence of Bucephalus, we are at one in demanding a rational or thinking ego to account for the psychical existence of a man ? The principal reason is that we are so THE SOUL 359 deeply impressed with the characteristic unity, and the importance, of human existence that we are afraid that a human personality would be dissolved unless there were such a substance to support it. And we are afraid to admit the possibility of such a calamity. I believe that there is a soul, and that this soul is a substance. What I deny is that the substantiality of this soul need be interpreted in a fundamentally different way from other instances of substance. When and so long as there is a characteristic unity of experiences, then a particular soul exists. Experi- ences are the stuff of the soul, not in the absolute sense of stuff, but in the relative sense in which timber is the stuff of Nelson's Victory. There is, however, an important difference in the cases. If the Victory became a hulk suitable only for target practice, her timbers would be thrown to the seas, and become fuel for a thrifty fisherman, or a source of danger to pleasure-boats. In what sense the timber of the Victory remains the same after this treatment is a matter of definition. There is a sense in which it does remain the same. But the unity of a self is more distinctive and important than the unity of the Victory. We have no evidence, or, at least, very insufficient evidence, to prove that any experience whatever can exist except as part of a self, and none that it can exist apart from a unity analogous to that of a self, and differing only in degree of organisation. Its being implies that it is part of such a unity, and, therefore, the facts compel us to maintain that it is a unity ex officio. Amj given experience may be considered as one, although it is continuous with other experiences, but the self seems to be an indefeasible unity, and that is why it is a substance in so fundamental a sense. That unity of experiences is the soul. Its substantiality is the ultimate fact that any given experience must form part of a distinctive unity of 360 PROBLEMS OF THE SELF experiences. It is therefore a substance in the same sense as other things are substances, though it is a distinctive kind of substance whose parts are experiences. We have alrea'dy seen abundant reason for reject- ing the transcendent doctrine of substance. It is never required. It solves no theoreticfil difficulty. Like a corpse that has been embalmed for ages, it crumbles to dust at the slightest touch. It has meaning only when it is regarded as a permanent thing which is somehow responsible for evanescent appearances. But experiences are real, and they are as they appear to careful introspection. They are a distinctive kind of beings. They are substances having stuff in them. They exist : and, as we have shown, they cannot be regarded as mere qualities of anything else, be that other thing matter or what you will. But, say you, if they are substantial, they are not self-existent substances ; and it is true that they are not. They must exist as parts of a unity, and the existence of all of them in a unity through time (though perhaps with intervals)^ is the soul, the psychical substance. There is no content of the soul other than experiences, and the permanent elements in experiences, such as they are, are too little to be a self. But the soul is neither an aggregate of experi- ences, in themselves loose and disconnected, nor is it a unity of qualities. It is a unity of experiences ; and there must be a soul, because it is part of the being of any experience to form part of such a unity. What that unity is, I have tried to show in a previous chapter. There is nothing mysterious in personal identity save for the intricacy of the facts. If we ask for the respects in which our souls are one we must analyse the unity and continuity of feeling, endeavour, and cognition, and the unity of these with respect to one another. That is the whole of our ' Of. the argument at the conclusion of Chapter X. THE SOUL 361 task. I do not mean, of course, that we can exhaust the entire content of the self by repeated acts of introspection. There is more in the self than can ever be the object of explicit self-Cognition. Much of it may be subconscious. And if the whole of our past selves could be revealed to introspection there would still remain the self of the future, which must be the same self as that of the past and the present. But, however large the drafts on subconsciousness, and even if there be overdrafts, the fact remains that subconscious experiences do not differ in kind from other experiences, and therefore that introspection showd'us the kind of being which the soul is. The difficulties of the margin apply to the identity of the soul as they apply in other cases of substantial identity. The unity of the self, from the cradle to the tomb, is less than the unity of many particular strands of its experiences. One of the classical objections to the doctrine of immortality is the simple question, ' What self is to be immortal ? ' Is it the self of old age, doting, perhaps, and trembling with decay, or the mature self of middle life, or the hope- ful self of youth ? But the difficulties of the margin are not insoluble. If we demand of a self the close- knit unity of the life of a Caesar during the Gallic wars, then, no doubt, we shall not be able to find it. We must be content with less in the lives of most men at the zenith of their powers ; and the unity of the life of the soul throughout its existence is, naturally, less than its unity at some particular time. There is enough unity for personal identity, and that is a very real unity, although it is all too easily exaggerated. And there may be enough unity for personal identity, even after the death of the body. Even the difficulties raised by the problem of multiple personality, that fertile mother of negative instances to all accepted beliefs, are not insoluble on this theory. If we say that personality requires a 362 PROBLEMS OF THE SELF chap. very close and distinctive unity of experiences, then dissociation into multiple personalities, and also the disintegration of trains of experiences below the level of personality, must be accepted as a fact. If we do not accept it as a fact then we must be prepared to maintain, unless we can challenge the evidence, that the minimum of unity and continuity implied in personality is very little ; and such a dispute tends to become a matter of definition. The facts of psychical research cannot do more than introduce qualifications into this general theory of the soul ; they do not affect its principle. And the important qualifications are only two. In the first place, we must admit the possibility that the same experience (or hmited train of experiences) may form part of several different selves, and it is possible to argue on this basis to a cosmic soul which includes all human personalities ; but the inference is very precarious and the evidence for it exceedingly slight. In the second place, the unity of experiences, while it must exist, may be less closely knit than personal unity, in the strict sense, requires. That possibility is exemplified, clearly enough, in some of the instances of the unity of animal experience. The peculiarity of the present instances is that some of the experiences, and trains of experiences, which are too disunited to form a self, are at the human level. But neither of these quali- fications affects the general principle of our argument. Experiences can only exist in a unity, though that unity may, in a few instances, be less intimate than the unity of personality. If these qualifications must be accepted in rare and abnormal cases, the plain implication is that they are not normally required. In normal cases the existence of the soul is beyond dispute, and the individual finite centre of experiences exists on its own account. If anything has a right to be called a distinct particular thing, the soul has such a right THE SOUL 363 pre-eminently. While the distinctions which we draw between things in the physical world are true and important, there seems to be no good reason, apart from momentary convenience, why we should fix on one boundary rather than another, and that is why scientific thought tends towards a monism of matter. It is otherwise with the self. Despite the difficulties of personal identity, despite the fact that no self is a perfect or fully-rounded whole, there is a greater independence and a more ultimate distinction between selves than between any other beings. The self requires society, but it is an independent member of society, and any political theory which neglects this fundamental truth is bound to be inadequate. I have said that the independent reality of selves is the greatest stumbling-block in the way of idealistic monism ; and, if the contentions of this essay are sound, it is irrelevant to seek to evade the difiiculty by pointing to the imperfections of the unity of finite selfhood. It is true that the soul is part of a wider spiritual realm. It is part of a social whole, to say the least, and its birth and its death (if it really begins and really ends) are not discontinuous with the rest of the cosmos. We are not fashioned, like Frankenstein's monster, from elements which have neither soul nor life ; indeed the beginnings of the existence of any individual soul are wrapped in mystery. Similarly there is nothing in the theory which I am defending in this chapter that is absolutely inconsistent with those doctrines of a supra-personal self which were mentioned at the beginning of this essay. There may be a cosmic spirit which shines in the lives of all of us. We may be but shoots of everlastingness, dim or bright. Our independence may be transient and we may, one day, return again to that spiritual whole from which, even now, we are not entirely separated. The transmission theories, or even such a theory as that of Mr. F. W. H. Myers, 364 PROBLEMS OF THE SELF cHAr. need not be regarded as impossible. Perhaps, as Fechner suggested, our personalities are distinct only in the sense in which the ribbon of white on the crest of the wave is distinct, and we, in the end, are only arbitrarily separable from the soul of God. But while a monism of this kind is possible, and not inconsistent with our theory, it is no part of that theory, and we should beware lest we accept it lightly. We know what our souls are, we know the meaning of their identity, we know the sense in which they are distinct and independent in the world. Because we know these things we should hold fast to them and insist, first of all, upon the reality of our person- alities as we find- them, and so long as we find them. These are not disconnected with other things, or with other personalities, and they may be part of a wider spiritual whole. But so long as they exist they dare not relinquish one tittle of their meaning. They are what they seem to be and must not be transmuted. The strip of white on the crest of the wave, so long as it exists, is not identical with the trough of the wave or with the rest of the ocean, and when it is swallowed up it ceases to be that strip of white. So the soul of man, as long as it exists, is not identical with the soul of the world, and it need not retain its personal identity when it returns to the soul of the world. Analogies of this kind are always felt to be strained and fanciful, and the reason is • that they insist on comparing the soul to the most trivial and evanescent substances. Even so they fail. But if they were more conclusive than they are, they would still require to be accepted with great reserve. The human soul, instead of being the most trivial example of substance, is the hest example of a distinct and independent substance which can be found in all the multitude of the things we know, and therefore the failure of these analogies is the point which requires most emphasis. There may be a world-soul, XIII THE SOUL 365 as pantheism asserts, but the unity, the significance, the value of this world-soul may be infinitely less than that of particular human souls. If it be true that personal identity means less and is poorer with regard to the self as a whole than with regard to particular periods of its life, how much more may the identity of the world-soul be poorer than that of our personalities ? The world-soul may be wider than the self, but it may mean so much less that it does not deserve to be called a soul at all. Accordingly, while we must not deny anything which is possible, our first duty is to accept that which is actual. We must accept and investigate the soul as we find it, insist on its reality, and refuse to barter that reality. The mistake in the past has been the assumption that the soul is more enduring and more perfect than it really is. Let us guard against the error of believing it less enduring or less perfect. If we can think the soul as substance, and I have tried to show that we can, we have a basis of certainty which we must not relinquish. Many objections have been made to the doctrine that the content of the soul is nothing but experi- ences, and some of them have already been stated and answered. There is, for instance, the argument from self-cognition, and Lotze's argument from the unity of consciousness.^ These aiTguments are not really in- consistent with the interpretation of the soul which is given in this essay, but the subject is too wide to permit of an exhaustive answer to all possible objec- tions. The most fundamental argument, I think, does not prove the necessity for a specific soul sub- stance distinct from experiences, but is in the fullest "possible harmony with our conclusion. Every act of judgment, it is said, is my judgment ; I must assent to it on my own responsibility, and this assent can- 1 Cf. Chap. VIII. pp. 201-212 (including the arguments concerning self- cognition). 366 PROBLEMS OF THE SELF ohap. not be given vicariously. Every resolve which I make is my resolve ; I am its author ; others may suggest the possibility, may educate and influence me, but they cannot resolve for me. Even Nature cannot. My resolutions are not hers to give. I should conform to her in most cases ; I should pre- scribe to her in some cases ; but to conform or to prescribe is mine in the end, whatever influences may be brought to bear. Similarly the world of feelings is my own. Nothing can compel my love or my hate, though these may be influenced. Hence is Stoicism, and hence the conviction of the substan- tiality of a particular finite self. You may lead a horse to the water, but you cannot make him drink. You may influence the resolve, the assent, the enaotion of others, but you cannot perform these acts for them. So far from denying this argument I wish to assent to it unequivocally, but the data on which it depends are features of the experiences of resolve, or assent, or emotion themselves. Any act of will or of judg- ment has this spontaneity and may not surrender it. Psychical acts are acts of reference to an object. They therefore belong to a different order of being from these objects, and, the mere presentation of the object cannot be the act of reference to it. What in this is inconsistent with our previous argument? But I shall be told that this objection cuts deeper. , My act cannot make your act, nor your act mine ; therefore the question is not merely that of the acts, but refers to the souls to which these acts belong. That, also, is part of my contention. The fact that there are selves is the fact that every experience forms part, and must form part, of an individual, specific, particular unity. If such experiences can be shared by different souls, they are very rarely shared so far as our evidence goes, and that statement of fact is a precise statement of this argument. When THE SOUL 367 I say that I resolve, I mean that any given resolve is part of that unity of experiences which is myself. What more can I mean ? What more can any one mean? It is hardly possible to avoid some reference to the question of immortality. If the view I am defending is true, some of the classical arguments for immortality must disappear. If the only ground for believing in immortality is that the soul is an in- discerptible substance, immune from change, genera- tion, or decay, and therefore indifferent to the fate of the body, then we must frankly admit that there is no reason for believing in the existence of such a substance, and draw the inevitable conclusion. In- deed, it would be fair to say that if the only reason for the belief in immortality is this antiquated piece of dogmatism, then the sooner we relinquish that belief the better it will be for our souls. And we might venture to add with Kant that the right to believe in immortality ' has lost nothing by this renunciation, for the merely speculative proof has never had any influence upon the common reason of men. It stands upon the point of a hair, so that even the schools have been able to preserve it from falling only by incessantly discussing it, and spinning it like a top. The proofs which have been current among men preserve their value undiminished ; nay, rather gain in clearness and power by the rejection of the dogmatical assumptions of speculative reason.' ^ It is not my purpose, or my wish, to add another to the countless volumes which deal with immortality. I wish merely to point out how the conclusions of this essay affect that question. The issue may be stated very shortly. If the soul is not the body then it may survive the body unless the body can be proved to be necessary for its existence. And if the soul may survive the body it may possibly be * Critique of Pure Reason, Meiklejohn's translation, pp. 250-51. 368 PROBLEMS OF THE SELF CHAl" immortal in the strictest sense, i.e. it may be inde- structible. If our argument has been sound, the soul is a substance which is not the body. It consists of experiences, not physical reactions. Its unity is a mental unity, not the unity of a physical thing, and not the unity of an organism. The soul exists, and exists on its own account. Consequently the death of the body does not imply the destruction of the soul, unless for further reasons. Our argument, on the other hand, has given no warrant whatever for the belief that the soul is indestructible. We have proved that the soul is a substance when it exists. There is no implication that it must exist for ever, and this perpetual exist- ence is immortality unless we deny the reality of time. The soul may exist for ever, but the fact that it is a substance does not prove that it must persist in saecula saeculorum. And it is also necessary to mention that the distinction between soul and body, even the existential distinction which consists in the fact that they are two diflferent substances, does not, by itself, prove that the soul can survive the body. The soul may require the body as a condition of its existence, although there is no convincing proof that it does. This question is beyond the scope of our enquiry except in one particular respect which has confronted us again and again. Although the body is not part of the self, it is so closely connected with human personality that it is difficult to realise what a dis- carnate personality would be. This difficulty may only be due to a deficiency of imagination. The poet, addressing his soul, can dream of a time when Thou shalt not peep through lattices of eyes Nor hear through labyrinths of ears ; he may insist that the body is a tomb, and that our selves may go forth freely and joyously into the un- charted land whose gates are the death of the body. XIII THE SOUL 369 It may be so, and death may be a new birth, not dis- continuous with our present personality but only, like physical birth, unfolding a new world. That is the hope of millions, and neither science nor philosophy can prove it to be vain. •One conclusion certainly follows from the argument of this essay, and that conclusion is so far from being a paradox that it is obvious without any argument. An immortality which is not personal is nothing. We are not immortal unless our personality can survive the shock of death. Unless there is unity and continuity of experiences, and the kind of unity which is personality, there is no soul. The im- mortality of the soul, therefore, is necessarily, and always, a personal immortality. There must be personal continuity between the incarnate and the dis- camate spirit, else the soul has disappeared either into nothingness or into something else. The body may not be necessary to this personality, and the question whether it is or is not is far from easy. What is certain is that most theories of immortality, especially pantheistic theories, fail because they do not honestly consider the question, and are content with trying to prove some continuance and some influence after death. That is not enough, and we know that it is not enough. The question (apart from supernatural revelation) must always remain open, because we could not prove the continuance of personality after death, unless we could compare the per- sonality which has survived with that which went before, and we cannot disprove the possibility of this continuance. It is a matter for empirical evidence as the existence of the soul upon earth is a matter for empirical evidence, and the life after death, whatever it be, cannot be ascertained -by empirical evidence. I cannot do more, and I cannot do better, than repeat what Simmias said, long ago, in the Phaedo : ' I feel myself how hard, or rather impos- 2b 370 PEOBLEMS OF THE SELF ohap. xm sible, is the attainment of any certainty about questions such as these in the present life. And yet I should deem him a coward who did not prove what is said about them to the uttermost, or whose heart failed him before he had examined them on every side. For he should persevere until he has achieved one of two things : either he should discover or be taught the truth about them ; or if this be impossible I would have him take the best and most irrefragable of human theories and let this be the raft upon which he sails through life — not without risk, as I admit, if he cannot find some word of God which will more surely and safely carry him.' ^ ' Phaedo, 85 (Jowett's translation). INDEX Abnormal psychology, 271 sqq. Absolute, 3, 129, 335 Acquaintance with self, 210 sgq. Acquired dexterity, 260, 263 Act and object, 18 sjg., 39, 99 sq. , 195 sqq., 211, 214, 223 ; act return- ing on itself, 167 sqq. Activity, 32, 57, 74 sq., 103 sqq., 184 sqq.; 'not an idea,' 110 sqq. ; as a cause, 115 sqq. ; prior to knowledge, 169 sqq. ; prior to experience, 173, 184 Adaptation, 128, 135 sqq., 144 ; of phenomenal to noiimenal, 162 Adynamic, 35 sqq., 110 sqq. Alexander, Prof. S., 25, 99 Alternations of personality, 57, 70 sq., 217, 276 sqq. Analysis, misconceptions of, 41 sq., 213 sqq. ' Animal aroma,' 72 sq. Ansel Bourne, 277 sq., 282 Anstoss, 172 Apperception, 198 sq., 207, 218, 224 sqq., 250, 261 Aristotle, 58 'Assent,' 34, 110, 125 Association, 198 sq., 218 sqq., 326 Azam, 277 Bain, 32 Beauchamp case, 70, 277 sq., 282 sqq., 293 sqq. Bergson, 5, 152, 168 sq., 183 sqq., 213 Berkeley, 50 sq., 59, 68, 110 sqq., 122, 269 sq., 314 sq., 341 Biunde, 88 n. Body and self, 11 sq., 43, 46 sqq., 83 sq., 112 sq., 150, 181, 184, 246 sqq., 273 sqq., 279 sq., 286 sq., I 292, 300 sq., 310 sq., 321, 367 sqq. Bosanquet, 128 sq. Bradley, F. H., 5 n., 9 sq., 85 sq., 94 sqq., 106 Brain, 253 sq., 271 ; see also Body and Self Brentano, F., 47 Broad, C. D., 200 m. Butler, Samuel, 1 sq. , 44, 330 Capacity, 44 sq. Causality, '61 sq., 89, 115 sqq., 157, 161 sqq., 175 ' Centre ' of consciousness, 10, 202 Change, perceived, 105 ; possibility of, 350 sq. Chesterton, 6 Choice, 34, 124 sq., 140, 154, 159, 183 Coenaesthesia, 70, 83, 96 Gogito, 43, 306 sqq., 330 Cognition, distinctions and divisions in, 19 sq., 24, 30 sqq., 35 ; position of, 76 ; compared with feeling, 91, 95 sqq. ; with will, 122 sq., 159 sq., 169 sqq., 179 sqq., 187 sqq. ; and the self, 191 sqq. ; unity of, 217 sqq., 359 sq. Coherence view of truth, 223 sqq. Comparison, 201 sqq. Completion and end, 132 Comte, A., 23, 111 Consciousness, abstract (?), 16; ex- tended (?), 75 sq. ; existence of (?), 77 ; not biologically explicable, 150 sq. ; see also Subconscious- ness, Introspection, Feeling, En- deavour, Cognition 'Consent,' 110 Conservation of energy, 145 sq., 357 Content of cognition, 193, 195 sqq. Continuity of self, 2i3 sqq., 2Zhsqq., 257 sqq., 276 sqq. Croce, B., 85 sq. Deliberation, 124 sq. Descartes, 5, 58 sq., 69, 110, 149 sqq., 253, 305 sqq., 338 371 372 PROBLEMS OF THE SELF Desire, a feeling attitude, 32, 123 Development of self, 44, 232, 243, 280 Dialectic, Hegelian, 5, 229 Ding an sich, 166 Direction and activity, 105 sq. Disposition, 44 sq. ; psychical, 250, 256 sqq. Dissociation of personality, 57, 70, 274 sqq. ; into personalities, 276 sq., 289 sqq., 361 sq. Dreams, 222, 259, 275 sq. Driesch, 136 sqq., 145 Duplication of ideas and things, 184 sqq. Dynamic, 35 sq. Dynamics, 116 sj. Effort, 74 Emotion, 63 sqq. ; emotional shook, 286 Empirical self, vi, 193 sq., 208, 244, 311, 338 Empiricism and Rationalism, 310 sq., 317 End and means, 129 sqq. Endeavour, 32, 34 iqq., 74, 81 sq., 101 sqq., 123 sqq., 129 sqq., 149, 159 sq., 175 sqq., 184 sqq. ; unity of, 242 sqq., 284 sq., 359 sq. ; see also Activity ErUbnisse, 13, 20 n. Error, 229 Essence, 310 ; nominal and real, 318 sq. Experience, 13 Experiences, 13 sq., 16 sqq., 79, 190, 365 sq. ; see also Consciousness ' Fact' of Pure Reason, 154 sqq. Eechner, 364 Feeling, 32, 34 sq. ,37 sq., 5S, 81 sqq. , 180 sq. ; physical and psychical, 52sg'3'., 63, 83 ; unity of, 237 sqq., 285 sqq., 359 sq. ; particularity of, 82, 93 sq. Felida X., 277, 284, 295 Terrier, 254 Fichte, 166 sqq., 174, 178, 190 Final cause, 129 sq. Flechsig, 254 'Force,' 117, 120 Freedom, 115, 156, 168, 172 Freud, 142, 259 sq. Galen, 237 Gall, 254 Growth, 141 Ground and cause, 310 Guidance, of cognition, 39, 180, 183, 192 ; without work, 145 sq. Habit, 251 sq. Haldane, J. S., 138 ». Hamilton, 34, 60, 87 sq., 90, 263 Hartmann, 141 Hegel, 130, 207, 229, 305, 335 sq. Hippocrates, 237 Hitzig, 254 Hume, 9, 116 sq., 119, 188, 305, 309 sq., 324 sqq. , 344 leh an sich, 166, 171 sqq. Ich denke, 331 sqq. 'Idea,' 50, 175 sqq., 195 sqq., 223, 314 sqq., 326 sq. ; see also Act and Object, and Presentation Idealism, 3, 129 sq., 166, 185, 188, 363 Identity (material), 320, 324 sqq., 349 sqq. ; (personal), 71 sqq., 245 sq., 273, 278 sq., 320 sqq., 360 sq., 364 ; of indiscernibles, 345 Ideo-motor action, 114 Image, 186 ; imagery and reasoning, 255 ; imaginary substances, 343 sq. . Immanent causation, 105, 107, 109 ; immanent substance, 329, 366 ; of the self, 358 sqq. Immortality, 311, 361, 367 sqq. Immunity, 136 Impersonal thinking, 236 sq. ' In mind, ' 49 sg'. Independence, of substance, 341, 356 Indeterniinate, 85 sq. Inheritance of meaning, 143 Initiation, 105 sqq., 121, 123, 156 Innervationsgefilhl, 62, 113 Instinct, 61 sqq., 129, liQsq., 143 sq., 165 Intellect, 8, 175 sqq., 188 sq., 255 Intelligible connection : in causation, 118 sqq. ; in rational choice, 158 ; ' intelligible character,' 156 Intensive quantity, 334 'Intentional inexistence,' 33 Interaction, 251 Interest, self and, 11, 59, 92 sq. ; see also Feeling ' Internal sense,' 47 sqq., 218, 220 Interpenetration, 186, 188 INDEX 373 Interpretative evidence, 23, 27 sq. Introspection, 13, 23 sq., 30, 47 sq., 51, 58, 73, 98 sgq., Ill, 114, 125 sq., 129, 187, 190, 194 sqq., 209, 329 sq. Intuition, 161 sq., 167 sq., 172, 189, 232, 234 sq. James, 11, 47, 63 sqq., 71 sqq., 77 sq.. 103, 112 sq., 125 sq., 215, 252 Janet, 295 Joachim, 224 sqq. Jodl, 83 Jonson, Ben, 237 Judgment, distinctions in, 18 Kant, 31, 78, 152 sqq., 173, 188, 190, 207, 218 sqq., 249, 305, 311, 328, 331 sqq., 344, 367 Kinaesthetic sensations, 119 Knower, self as, 191 sqq. Lange, 9 Leibniz, 202, 207, 260 sq., 342 Lindsay, A. D., 187 Lipps, 82 Locke, 16, 28 n, 29, 48, 50 sq., 58 sq., 75, 269 sq., 284, 305, 310 sqq., 328, 338 Logic, 232 sqq. ; of the senti- ments, 241 ; of endeavour, 245 ; laws of, 150 ; logical subject, 308 sq., 339 sqq. Lotze, 9 sq., 75 sq., 201 sqq., 269 sq., 365 Louis Vive, 277, 283 sq., 286, 288, 295 Margin, of consciousness, 265 sqq., difficulties of the, 352 sq. Mary Reynolds, 277, 282, 295 'Material,' 21 sq., 30, 53 Matter. See ' Stuff,' or ' Physical Object ' Mayer, 357 M'Dougall, W., 10, 12, 46, 61 sq., 133, 136, 139 M'Taggart, 5 n. Means and intelligence, 180 ; see also End amd Teleology Mechanism, 133 sqq., 161 sqq. Memory, 220, 246, 251, 277 sqq., 295, 323 Mendelssohn, 31 n. Meno, 233 Messer, 21 n. Metaphysics, and the self, 2 sqq., and psychology, 14 sq. Meynert, 253 Mill, J. S., 120 Minimum sensibile, 29 n. Modes, 338 sq. Monadic self, 96, 206 sq. Moods, 98 Moore, G. E., 20 n., 29 n.., 124, 195 Morphogenesis, 137, 147 Morton Prince, 70 n., 267 n., 211, 284, 294 sqq. Multiple personality, 217, 272 sqq. 361 sq. Munk, 254 Miinsterberg, 46, 113 sqq. Myers, F. W. H., 10, 363 Neo-vitalism, 134 sqq., 162 sqq., 190 'Notion,' 111 Not-self, 96 sq. Noiimena, 156 sqq., 221, 333 Novelty, 121 sqq. 'Object,' 20 sqq., 30, 37, 53 n., 96 sqq., 191 ; for categorical im- perative, 155 ; see also Act Objectivity, 219 sqq., 340 sqq. 'Organised,' 149 n., 182 Organism, 135 sqq., 357 Origin, of idea of self, 84 ; argu- ment from, 84 sq. ; speculations on, 232 Pantheism, 364 sq. Parallelism, epiphenomenaUstic, 251 Paralogisms of pure reason, 331 sqq. Particular, the, 344 sqq. Passivity, 104 sq., 107 ; of sensa- tions, 111 sq. Permanent self, 269 sqq., 328 sqq., 358 sqq. 'Person ' and self, 7 sq., 280 sq. Phaedo, 369 sg. Plato, 236 Plasticity, 250 sqq. Physical world, 17, 127, 188; physical object, 38, 55, 270 Practical reason, 152 sqq. Pragmatists, 3, 150 Prediction, 116 Preponderance of conscious elements, 39, chapters iv.-viii., 245 Presentations, 17, 28 n., 29 n., 37, 52 sq., 55, 188, 200, 203, 268 sj., 327 374 PROBLEMS OF THfi SELF Primacy, 43, chapters iv.-viii., 245 Projection, 55 «., 116 Psycho-analysis, 259 sq., 265 'Psyclioid,' 138 sqq., 142, 144, 149 Psychology, and the self, 14, 16 sqq, ; without a soul, v, 9, 16, 40 sq., 194, 201, 337 Pure ego, vi, 96, 193 sq., 208 Purpose, purposive, 101, 127 sqq., 161 sq., 244 ; purposiveness of nature, 162, 164 ' Quality,' 21 n., 30 sq., 191, 195 sq. Quality, and relation, 147, 228 ; and substance, 317, 339 Reason, a determining factor, 156 sqq.; reaction against, 240 sqq. Reference to object, 20 sqq., 31 sqq., 53, 95, 99, 106, 247 ; to Reality, 340 sq. Reflection, ideas of, 312 Regeneration, 137 Reid, 309 Reincarnation, 280, 322 sqq. Repetition, 108 sq. ' Reproduction, 135 ; in memory, 220 Res per se suhsistens, 308 sqq., 338 sqq. Resolve, 108 sqq., 124 Responsibility, 183, 281 Responsive behaviour, 26 Restitution, 135, 147 Retentiveness, 220, 249 sqq., 271, 279 sq., 282 sqq. Ribot, 70, 238 w., 287 Russell, Bertrand, 29 n. , 210 sqq. • Sally,' 70, 278, 284, 286, 290, 296 sqq. Satisfaction, 132 Schneider, 134 ■Schopenhauer, 152, 165 sq., 174 sqq., 190 ' Scientific mind, ' 225 Selection, in cognition, 123, 223 ; in causation, 120 ; in behaviour, 128 Self-cognition, 167, 207 sqq., 361, 365 Self-consciousness, 167, 206 sqq. Semon, 44, 254 Sensation, 53 n., 198, 201 ; not obscure logic, 342 sq. Sense-datum, 29 n. Sensible, 29 n. Sensible species, 49 Sentiments, 238 sqq., 286 Shand, A. F., 238 Simple ideas, 312 sqq. Somatic resonance, 57, 65 Soul, and self, 7 sq., problem of, 12, 42, 200, 260, 304 sqq., 337 sqq. ; for vitalism, 164 ; thinks always? 269 ; see also Spiritual Sub- stance, Pure and Empirical Ego Space and Time, 168, 175, 177 sq., 186 Spinoza, 216, 238, 304, 328 'Spiritual self,' 72, 74 Spiritual substance, 306 sqq., 317, 319 sqq., 328 sqq., 358 sqq. Spontaneity, 109 sqq., 167, 366 Stout, G. F., 66, 106 sq., 123 sq., 143 sq., 238 «. 'Stuff,' 3i6 sqq. Subconsciousness, 142 sq., 257 sqq., 291, 361 Subject-object relation, 96 sq. ; sub- ject of qualities, 195, 308 sq., 339 sqq. Subjective, meaning of, 17, 28 sq., 80 ; subjective presentations, 54, 56 ; subjective and objective, 218 sqq. 'Subjectively subjective,' 34, 60, 87 Substance, 195, 200 sqq.; 250, 273 sqq., 300 sq., 304 sqq. Substantial forms, 313, 338 Suggestion, errors due to, 289 sq. ' Supporter of accidents,' 308, 338 sq. Synthesis, of apprehension, 218 ; of reproduction and of recognition, 219 sq. ; of personalities, 302 Teleology, 10, 129 sqq. ; 139, 145, 149, 161, 164 Temperament, 237 sq. Terminating in a percept, 77 sq., 150 Tetens, 31 n. Theaetetus, 263 Timeless truth, 19 Tirelessness of wiU, 179 Titohener, 40 Transcendent substance, 324, 333, 366, 360 Transmission theory, 11, 363 Tripartite division, 31 sqq., 38 sq., 88, 190, 192, 215, 245 Unconscious cerebration, 257 sq. Unconscious will, 174 sqq. Unity of self, 9, 12 sq., 40, 43, 201 sqq., 213 sqq., 250 sqq., 290 sqq., 301 .sqq., 359 sqq. INDEX 375 See also Cognition, Feeling, En- deavour Unity of a thing. See Material Identity Universals, 18, 37 sq., 154 sq., 181, 335 sq., 339 sq., 346 Value, 129 sq., 133, 155 sq. Vicious infinite, 210 Vital principle, 321. See also Neo- vitalism. Volition, 102, 113 sqq., 117 sqq., 122, 124 sqq., 150, 166 sqq., 174 sqq. ; implies an object, 33 sq. Voluntarists, 128, 166, 183 Voluntary, 124 Wants, 179 Webb, T. E., 311 Weber-Fechner law, 261, 264 Whitehead, 355 n. Whole and part, 130, 132, 145 sqq., 213 sq. Will. See Volition and Endeavour. In nature, 174 sqq. THE ENP Printed by 'R. & K. Clark, L\mTE.i>, Edinburgh.