Logical and Philosophical Papers 1909-13 (Collected Papers, Vol 6) [Hardcover ed.] 0415084466, 9780415084468

The years covered by this volume of the Collected Papers of Bertrand Russell were among the most productive, philosophic

669 46 29MB

English Pages 632 [320] Year 1992

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Logical and Philosophical Papers 1909-13 (Collected Papers, Vol 6) [Hardcover ed.]
 0415084466, 9780415084468

Citation preview

BERTRAND RUSSELL

Logical and Philosophical Papers 1909-13

Edited by John G. Slater with the assistance of Bernd Frohmann

Bertrand Russell, circa

1910.

(McMaster University)

I~

London and New York

First published 1992 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge a division of Routledge, Chapman and Hall Inc. 29 West 35th Street, New York, NY 10001 Bertrand Russell's unpublished letters and Papers 4, 10, 11, © McMaster University r992. Papers 3, 5- 9, 13-14, 17, 20, 22- 36 and Appendixes 1v-v, vn-vm ') The Bertrand Russell Estate, r909, 1910, 191r, 1912, 1913. Paper 18 ,) Routledge & Kegan Paul Ltd. 1961. Papers 28, 33 © U.K. , George Allen & Unwin Ltd. 1966; " u.s., Simon and Schuster, Inc. 1967. Papers 15, 18 '' George Allen & Unwin 1932. Paper 16 George Allen & Unwin 1956. Editorial matter © John G . Slater 1992.



Fur,ds to edit this volume were provided by a major editorial grant from the Social Sciences and Humanities Research Council of Canada and by McMaster University. Typeset in 10/12pt Plantin by The Bertrand Russell Editorial Project, Mc.Master University Printed in Great Britain at the University Press, Cambridge All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented induding photocopying and recording, or in any information storage or retrieval system'. wnhout permission in writing from the publishers.

THE COLLECTED PAPERS OF BERTRAND RUSSELL

GENERAL EDITOR

John Passmore (Australian National University)

ADVI SO RY EDITORIAL BOARD

Sir Alfred Ayert I. Grattan-Guinness (Middlesex Polytechnic) Jock Gunn (Queen's University) F rancess G. Halpenny (University of Toronto) Royden Harrison (University of Warwick) Leonard Linsky (University of Chicago) H . C. G. Matthew (St. Hugh's College, Oxford) John Passmore (Australian National University) D.F. Pears (Christ Church, Oxford) John M. Robson (University of Toronto) Alan Ryan (Princeton University) Katharine Tait

British Library Cataloguing in Publication Data Russell, Bertrand Logical and Philosophical Papers, 1909-1 3 - (Collected Papers of Bertrand Russell; v. 6) I. Title II. Slater, John G. III. Frohmann, Bernd IV. Series 192

Librmy of Congress Cataloging in Publication Data Russell, Bertrand, 1872- 1970. Logical and philosophical papers, 1909-1 3 / Bertrand Russell; edited by John G. Slater with the assistance of Bernd Frohmann. p. cm. - (The collected papers of Bertrand Russell; v. 6) Includes indexes. I. Philosophy, M odern - 20th century. 2. Logic, Modern - 20th century. I. Slater, John G. (John Greer). II. Frohmann, Bernd. III. Title. IV. Series : Russell, Bertrand, 1872-1970. Selections. 1983;v.6. Bl649.R91 1983 vol. 6 192 s-dc20 •(1921 92-2380

ISBN 0-415-08446-6

t Deceased 1989

Contents

Abbreviations Introduction Acknowledgements Chronology

Xll Xlll

!xi !xiii

PART I. LOGIC AND THE PHILOSOPHY OF MATHEMATICS

1 2 3 4 5 6 7 8 9

The Theory of Logical Types [1910] The Philosophical Importance of Mathematical Logic [1911] On the Axioms of the Infinite and of the Transfinite [1911] What is Logic? [I912] Reply to Koyre [19I2] Review of Reymond [ l 909] Review of Carus [ l 909] Review of Mannoury [1910] A Medical Logician [I912]

3 32 4I 54 57 60 64 67 70

PART II. THE PROBLEM OF MATTER

10 On Matter [1912] Nine Short Manuscripts on Matter [1912-13]

11

PART III. METAPHYSICS AND EPISTEMOLOGY

12 On the Nature of Truth and Falsehood [I910] 13 The Basis of Realism [191 l] 14 Analytic Realism [19n] 15 Knowledge by Acquaintance and Knowledge by Description [1911] 16 On the Relations of Universals and Particulars [1912] 17 The Nature of Sense-Data: A Reply to Dr. Dawes Hicks [1913] 18 On the Notion of Cause [1913]

Vil

115 125 132 147 162 183 190

viii

CONTENTS

CONTENTS PART IV. ETHICS

19 20

The Elements of Ethics [1909] Spinoza [r9ro]

213 25I

PART V. CRITIQUE OF PRAGMATISM

21 22 23 24 25 26

Pragmatism [I909] The Philosophy of William James [r9ro] Review of James's Memories and Studies [1911] Pragmatism and Logic [r9I2] Review of James's Essays in Radical Empiricism [1912] Review of Boutroux [1912]

257 285 290 292 298

vnr. "Reponse a M. Koyre" [1912] and an English Translation of "Sur les nombres de M. Russell" by A. Koyre "On Mr. Russell's Reasons for Supposing That IX. Bergson's Philosophy Is Not True" by H. Wildon Carr

PART VII. CRITIQUE OF IDEALISM

31 32 33

Some Explanations in Reply to Mr. Bradley [r9ro] The Philosophy of Theism [1912] Hegel and Common Sense [1912] 34 The Philosophy of Good Taste [1912] 35 The Twilight of the Absolute [1913] 36 Philosophy Made Orthodox [1913]

349 359 363 366 371 374

APPENDIXES I.

F. C. S. Schiller's Replies to Papers 21 and 24

II.

VI.

Preface to Philosophical Essays [r9ro] F. H. Bradley's Criticism of Russell and His Reply to Russell [r9ro-11] Surles axiomes de l'infini et du transfini [r9rr] Le Realisme analytique [r9r r] G. Dawes Hicks's "The Nature of Sense-Data"

VII.

Remarks Opening the Section [1912]

[1909-12] III.

IV.

v.

[r9r r]

379 386 388 398 409 433 444

455

TEXTUAL NOTES BIBLIOGRAPHIC INDEX GENERAL INDEX

27 The Professor's Guide to Laughter [1912] 28 The Philosophy of Bergson [1912] 29 Metaphysics and Intuition [1913] 30 Mr. Wildon Carr's Defence of Bergson [1913]

450

ANNOTATION

305

PART VI. CRITIQUE OF THE PHILOSOPHY OF BERGSON

IX

551

Illustrations

frontispiece Bertrand Russell, circa

1910 ..

(McMaster University)

between pages 106 and 107 "Philosophy of Matter" (Paper Ile). n-v "Matter" (Paper lld). VI

"Matter" (Paper Ile).

vn

"Untitled" (Paper llf).

VIII

"On Matter" (Paper Ilg). All plates are photographs of documents in the Bertrand Russell Archives at McMaster University. They are shown reduced from their original size, which is given at the head of each set of textual notes.

Xl

Abbreviations

To

PROVIDE THE

Introduction

reader with an uncluttered text, abbreviations and

sy~bols have been kept to a minimum. The few necessary to the refer-

encmg system are as follows.

The papers printed in the volume are given a boldface number for easy

refe~e~ce. For_ example, "The Theory of Logical Types" is Paper

1.

B1bhograph1cal references are usually in the form of author, date and page, e.g. "Russell 1988, 122". Consultation of the Bibliographical Index shows th~t this reference is to p. 122 of Bertrand Russell, Essays on Language, Mind and Matter: 1919-26 (The Collected Papers of Bertrand Russell, Vol. 9; _London, ~oston and Sydney: Unwin Hyman, 1988). The locat10n of archival documents cited in the edition is the Bertrand Russell Archives at McMaster University ("RA"). File numbers of documents i_n the Russell A:chives are provided only when manuscripts of papers prmted here are cited or when files are difficult to identify. "RA REC. ACQ." refers to the files of recent acquisitions in the Russell Archives. Cross-references to annotations are preceded by "A" and followed by page and line numbers (as in "A44: 4"). Further abbreviations used in the Textual Notes are identified at the beginning of each set of notes. The numbers and dates of Russell's letters to Lady Ottoline Morrell refer to photocopies in the Russell Archives at McMaster. The original lette_rs are at The Harry Ransom Humanities Research Center, The University of Texas at Austin. The numbering was established while the letters were still in the possession of Lady Ottoline and it has not been revised.

XU

by this volume were among the most productive, philosophically speaking, of Russell'~ e1.u~re career. In. ad~itio~ to th_e papers reprinted here, he brought Principia Mathematica mto its fimshed form and wrote The Problems of Philosophy, Theory of Knowledge, and Our Knowledge of the External World. In October l9IO he began teaching at Cambridge, having accepted an appointment as lecturer in logic and the principles of mathematics at Trinity College for _a term of fiv~ ~ears. A year later Ludwig Wittgenstein beg:n to _attend ~1s lec~ur~s. Withm a few months he was influencing Russell s philosophical thmkmg as much as, or more than, Russell was influencing his. Russell's life was made even busier when he began, during his first year of teaching, an intense love affair with Lady Ottoline Morrell, which lasted for several years. An important consequence of this affair was a constant stream of letters to La~y Ottoline, which discussed his work almost as much as they professed his love. These letters are the single most important source of information for his activities from l9II until the end of the First World War. In order to place the papers printed here in their context, it is necessary to consider the various philosophical activities to which he devoted his energies between January 1909 and December 1913 with, as is inevitable, some reference to events both before and after those dates. His more political writings-during this period principally on women's suffrage-are ignored; they have been reprinted in Volume 13 of the Collected Papers. His philosophical activities were: (I) the completion of Principia Mathematica and the seeing of it through the press; (2) the drawing out of the implications of that work for philosophy; (3) his teaching, especially his philosophical relationship with Wittgenstein; (4) the writing of The Problems of Philosophy; (5) his attempt to determine th~ nature of matter; ~6) his search for a solution to the problems concernmg knowledge, which arose out of his thinking about matter, and which led to Theory of Knowledge; (7) the writing of Our Knowledge of the External World, after he abandoned Theory of Knowledge; (8) his controversies with other philosophers, especially pragmatists, idealists, and Bergson; (9) his first attempts, under Lady Ottoline's influence, to develop as a writer of philosophical works for the general public. THE YEARS COVERED

Xlll

XIV

INTRODUCTION I. PRINCIPIA MATHEMATICA

On I8 October I909 Russell wrote to Lucy Donnelly, a long-time friend and correspondent who taught English at Bryn Mawr College, with important news: Since your letter came I have been too busy to write until now but now I have time to write, having arrived at a great moment; tomorrow I go to Cambridge, taking with me the MS of the book for the printers. There is a certain amount at the end that is not ye~ finished'. but over 4,000 pages are ready, and the rest can be fi?ished e~s.Ily. I ha~e ~een working like a black to get the last bits of revis10n done m time for my visit to Cambridge tomorrow and now the MS is packed in two large crates, and now I feel more or less as people feel at the death of an ill-tempered invalid w~om they have ?ursed and hated for years. It is amusing to ~hmk how much time and trouble has been spent on small points m_ obscure_ corners of the book, which possibly no human being :Vill ever discover. Owing, I imagine, to the near prospect of takmg the MS to t~e Press, I have been lately in a state of strange an~ unusual excitement, very loud and bristling and argumentative. The preparation of the MS for the press had taken a few months longer tha? ~e had expected. In a letter to Donnelly dated 18 March 1908 he had anticipated another year's work on it: Since September, I have written about 2,400 pages of the MS of our book, and am still only in the third of eight Parts. I expect we sha.11 both die before any one reads it through, but people will read bits, and they will have to praise it, for the same reasons for which people praise Clarissa Harlowe, because otherwise they would have been wasting their time. I have a very firm belief in the importance of the book, and this belief will not be shaken by an unfavourable reception. There is more work in the writing out than I expected, and I hardly think the MS can be finished in less than a year from now; and then the printing will take about two years.

!

Judging from a letter which he wrote to her on 1 September 1907, it appea~s that he had begun the task of producing the manuscript on the followmg day. He then told her he had just returned from a holiday in Scotland, and that "tomorrow I must settle to work".

INTRODUCTION

xv

His collaboration with Alfred North Whitehead, who had been one of his teachers at Cambridge, on this immense work arose out of discussions they had held during the summer of 1901, when Russell was at work on The Principles of Mathematics (1903) and Whitehead on the second volume of his Universal Algebra, the first volume of which had been published in 1 898. By the time Principles had been published their agreement to collaborate was firm, a fact which Russell proudly announced in the Preface to that book. Their stated intention was to produce a second volume of Principles which would present a detailed proof of the derivation of certain branches of mathematics from the axioms and the rules of inference which had been formulated by symbolic logicians during the previous fifty years. Whitehead later wrote that they believed the job could be done in about a year, but as they worked at it their "horizon extended" (1941, IO) with, inevitably, an extension of the time taken to complete their work. As is well known, they also encountered difficult philosophical problems which had to be solved before the work could proceed. Russell's theory of descriptions and theory of types provided solutions to two of these problems. According to accounts offered by both authors, Whitehead left the solution of the philosophical problems to Russell. Whitehead's great strength lay in his extraordinary knowledge of mathematics and his skill in proving theorems. The actual writing of the book, once the philosophical difficulties were out of the way, was, according to a letter Russell wrote to Dr. G. Revesz on 6 December 1932, accomplished in this way: After we had agreed upon the general plan of the work, we apportioned the parts so that the first draft of any part was done first by one of us, and sent to the other, who worked over it and generally improved it, and then sent it back to the first .... The business of writing out the book in finished form for the Press fell to me, as Whitehead had his time much taken up by teaching, whereas I had no professional duties. We met frequently, but the most important part of the collaboration had to be done in writing, since the matters concerned were too difficult for verbal treatment. Frequently a draft of some portion would go backwards and forwards between us many times before we were satisfied. The work was a joint one in the strictest sense of that word, and it annoyed Russell when others gave him too large a share of the credit. In a review of J.M. Keynes's Treatise on Probability, he took Keynes to task for this sin: I take this opportunity to protest against Mr. Keynes's practice

XVI

INTRODUCTION

INTRODUCTION

of alluding to Principia Mathematica as though I were the sole author. Dr. Whitehead had an equal share in the work, and there is hardly a page in the three volumes which can be attributed to either of us singly. (r988, I22)

Mathematica. I shall be relieved when the old thing is all finished; it seems stale and flat now, though I still think just as strongly as ever that it was important to do it. (#357)

But Whitehead himself did temper this judgment by remarking in a footnote in Process and Reality (r929, IO) that the "introductory discussions are practically due to Russell, and in the second edition wholly so". In January I909, when this volume opens, Russell was about half-way through the writing out of Principia. The important philosophical problems had been solved to his and Whitehead's satisfaction, and it only remained for them to bring their results together in one system of axioms and theorems. Even so there was still a lot of work to be done. He had to complete the writing, and he did not find the time to do so until the Christmas vacation of I9IO. On 24 January 19I1 he wrote Donnelly: "I got very little holiday at Xmas, as I had to finish the writing of the end of our big book. In Term time, there are so many interruptions that original work is difficult." From late 1909 until early in I9I3 he corrected a steady stream of proof sheets. The last lot of them arrived on 3 February I9I3. The proof sheets when unfolded were very large, so he was obliged to spread them out on the floor and read them on his hands and knees. On I9 January 19I3 he wrote to Lady Ottoline: we have decided to publish what is printed of Principia Mathematica at once in a thin volume (about 480 pages), and I have had to make the table of Contents and List of Errata-such things as "p. 218, last line but one, for A readN', where only a very careful reader would notice there was any change. But although it is tedious it is soothing; work that involves no thought is always rather agreeable. What remains to be done to the book is Whitehead's affair-his MS. won't be ready till the autumn at the earliest. (#675a) From this one gathers that the original plan was to include some of Whitehead's work on geometry in Volume III and the remainder of it in a projected Volume IV. As Russell remarks, the work on geometry was to be carried out by Whitehead alone. Later Whitehead decided to publish the gist of his geometrical studies in his own books, and his surviving fuller manuscripts were burnt after his death, so Volume IV of Principia Mathematica never appeared. By 24 February 19I2 the printing of Volume III had begun: Today I have had the first proofs of the third Volume of Principia

XVll

The same mood surfaced on I3 March when he received his copies of the second volume: Volume II of Principia Mathematica was published today-it was a mild event getting it. But Volume I was much more. Odd how much passion goes into doing a thing and how cold it is when it is done. A vast amount of various people's solid misery is crystallized in the book, and I would have done almost anything to bring about the finishing of it, and now it is a mere moment's interest-because there is nothing more to do. (#380) The crucial role that passion played in his technical philosophical work is a common theme in his letters to Lady Ottoline. At about the time Volume III was published, on 19 April 1913, Russell received a statement from Cambridge University Press informing him that as of 30 December 1912, nearly 500 copies of Volume I had been sold, and nearly 300 of Volume II. "I can't think who buys it", he told Lady Ottoline, "it must be libraries" (#748). From the records of the Press we know that the print run for Volume I was 750 copies, and for each of the other two volumes it was 500 copies. These sales figures, therefore, must have surprised the officials of the Press as much as they surprised Russell. II.

THE PHILOSOPHICAL IMPLICATIONS OF

PRINCIPIA

MATHEMATICA

During the "years and years of absolute torture over Principia Mathematica" (#710, 26 Feb. 1913), Russell had, of course, appreciated that the philosophical work he was doing on that project had important consequences for other philosophical problems. In "On Denoting" (1905), for example, he had, in presenting his solution to the logical problem posed by expressions of the form "the so-and-so", alluded to some broader philosophical problems upon which his theory cast light. As soon as his duties on the big book were in hand, he began to think of ways in which he could communicate its philosophical innovations more widely. His first concern was to make his logical discoveries better known among logicians. Paper 1 in this volume, "The Theory of Logical Types", published first in French and then included in the Introduction to Principia Mathematica, is a prime example of this sort of effort. In it he calls attention to the troublesome importance of paradoxes which arise from

INTRODUCTION

INTRODUCTION

the use, nearly always unconscious, of vicious-circles and expounds his method of avoiding them. His discovery of the paradox that bears his name, first published in The Principles of Mathematics, had obliged him to attempt to find a way of taming it. The paradox concerns the class, call it O, of all those classes which are not members of themselves. Russell proved that 0 is a member of itself if and only if it is not a member of itself. In essence, then, Russell's paradox involves a vicious-circle by allowing a totality to be a member of itself. His first tentative solution, included in an appendix of Principles, was a theory of types, and so was his final one, after many attempts to discover an alternative that was closer to common sense. The next two papers, "The Philosophical Importance of Mathematical Logic" (2) and "On the Axioms of the Infinite and the Transfinite" (3) were both lectures delivered to audiences of French philosophers who had invited him to expound his latest discoveries. Paper 2 was also published in English. Although these papers, in general terms, expound doctrines he had already presented, he expands and develops as he expounds. They are of outstanding usefulness as a systematic exposition of the logical views which are central to his thinking of this time. In philosophical areas more remote from logic he also had new points to press. On 6 March I9I I he read "Knowledge by Acquaintance and Knowledge by Description" (15) to a meeting of the Aristotelian Society. This paper brings into the heart of epistemology his theory of definite descriptions. Using this theory he was able to offer an explanation of how we can be said to know (say) Napoleon, even though we are not acquainted with him, i.e. he does not fall within our experience. For Russell the list of items with which we are acquainted is short: "We have acquaintance with sense-data, with many universals, and possibly with ourselves, but not with physical objects or other minds" (I6I: 9-II). Napoleon we can know only by description as, let us say, "the French general who was defeated at Waterloo". At the same time Russell argues that "every proposition which we can understand must be composed wholly of constituents with which we are acquainted" (I54: 6-8), so in the final analysis the proposition which is apparently about Napoleon must be about certain sensedata (the noises or shapes which make up the word "Napoleon") and certain general concepts. In his first presidential address to the Aristotelian Society, "On the Relations of Universals and Particulars" (16), he drew his audience's attention to the importance of relations for philosophy, a subject which he believed had been neglected with far-reaching and sometimes disastrous consequences. The point of the address is to show his auditors the ways in which modern logic can be used to analyze an ancient metaphysical problem. His second presidential address, "On the Notion of Cause" (18), has a similar theme. While he was working on it he had toyed with the idea of substituting an address "'on the scope and method

of philosophy', which would be very controversial, a hattle-cry, with all existing philosophers against me" (#574, II Sept. I9I2), but had decided against it. In I9I4 he did write .such a paper with the title, "On Scientific Method in Philosophy", which is reprinted in Volume 8 of The Collected Papers (I987), and it did have the intended effect, at least among the older generation of philosophers. In the letter just quoted he went on to tell Lady Ottoline:

XVlll

XIX

I should say that philosophy can't deal with any questions of human interest, but reduces itself to logic, but in that sphere can attain real knowledge; also that most of the great systems have depended on elementary blunders in logic. This succinct statement of the thesis of his I9I4 paper shows that he had reached the conclusion that philosophy itself was a branch of logic long before he published "On Scientific Method in Philosophy". Actually a much earlier statement of it is to be found in A Critical Exposition of the Philosophy of Leibniz (1900, 8): "That all sound philosophy should begin with an analysis of propositions, is a truth too evident, perhaps, to demand a proof", because for Russell logic is the analysis of propositions. A number of the other papers in this volume argue for the importance of the new logic for philosophy. His reply to Bradley (34) and "The Basis of Realism" (16) both stress the importance of a correct view of the logic of relations in attempting to answer certain perennial philosophical problems. "On the Nature of Truth and Falsehood" (15) makes use of both the theory of descriptions and the theory of relations to offer solutions to some problems concerning the relation of a proposition to its Objective in Russell's version of the correspondence theory of truth. The new logic is also employed in this volume as a mode of criticism, for Russell often makes the point that a book or theory under scrutiny is defective in its logic. III. HIS TEACHING AT CAMBRIDGE

Russell accepted an appointment to lecture at Trinity College, Cambridge, on logic and the philosophy of mathematics, to begin in October I9IO, as part of that general project, to which we have already drawn attention, of making his ideas known to a wider general public. Whitehead, who was leaving Cambridge for London, had been instrumental in arranging the five-year appointment. Russell took advantage of the appointment to effect a physical separation from his first wife, Alys, by moving into rooms in Nevile's Court. Their marriage, as his autobiography reveals, had in fact terminated in all but name many years before. Living in rooms in

xx

INTRODUCTION

INTRODUCTION

college permitted him more opportunities for interaction with his pupils, who, to judge from his letters to Lady Ottoline, took great advantage of it. His letters to her are often interrupted because one of them has arrived for a visit. On one evening a week he was "at home" to students and his fellow dons, and they often came in droves, knowing that the talk and the argument, especially that of their host, would be both witty and brilliant. Russell became known for his ability to crack jokes, and he used them to enliven both his lectures and his conversation. Some of his jokes were included in The Philosophy of Mr. B*rtr*nd R*ss*ll (1918), a delightful little book which was edited by his former pupil, Philip Jourdain, later a leading historian of mathematics. As many others, both before and after him, have learned, Russell discovered that a full-time teaching position had an unwelcome dampening effect on research and writing. On 24 January 1911, less than four months after taking up his duties, he told Lucy Donnelly that he had found it necessary to alter his writing schedule:

of 1914, when he gave what he called his "popular lectures" at Cambridge, a preview of his Boston Lowell Lectures, later published as Our Knowledge of the External World (1914), he was astounded at the number of people who turned up to hear them. "Yesterday when I went to my lecture I found such a crowd that I thought I had come to the wrong place-there were quite 150! They mostly won't come again" (#975, 24 Jan. 1914). But he was wrong; the numbers stayed up. On 26 February, toward the end of the lectures, he reported meeting one of the auditors:

I got very little holiday at Xmas, as I had to finish the writing of the end of our big book. In Term time, there are so many interruptions that original work is difficult. The interruptions are pleasant in themselves-lecturing, and talking to pupils, and so on-and if one gives oneself up to them they make an easy life compared to writing. But if one tries to write at the same time, they make life difficult. So I only try to write regularly in the Vacation.

It is apparent from his letters to her that he pinned his hopes for the future development of his discoveries upon a very few of his pupils. During the six years at Cambridge he taught a number of such gifted students. C. D. Broad was working on his fellowship dissertation when Russell began teaching. (The nearest modern equivalent, although not a strict one, is the doctoral thesis.) Broad submitted his dissertation in the summer of 1911 and Russell was one of its readers. On 18 September he rendered this preliminary judgment on it:

Throughout the six years he taught at Cambridge this was a constant refrain in his letters, especially in those to Lady Ottoline. Although he often complained of being interrupted, he did ~ot let it curtail his output, for, as has already been mentioned, these were among his most fruitful years. Russell did not go to Cambridge with the expectation of having many pupils, for, as he well knew, the subjects he was to teach were both new and demanding. Only pupils who were ready to be scrupulous as to details were likely to be able to follow him, and, since most philosophy students at that time were unaccustomed to tightly-knit reasoning in symbolic form, it was unlikely that they would be prepared to follow an oral presentation of the system of Principia Mathematica and an exposition of the proofs on which it depended. But a few philosophy students did have a training in mathematics and some students of mathematics and science were interested in the foundations of their subjects: it was from students with these two backgrounds that Russell drew his serious pupils. Many others came to his lectures; but, as he lamented to Lady Ottoline, he was afraid that the great majority came only to hear his jokes. In the winter

I have finished my first reading of Broad: solid and sound and dull. He never makes either a mistake or an important discovery. His worst offences are his jokes, which have a kind of terrible flatness, like soda-water after the fizz has gone out of it. (#184)

XXl

I had tea with Shephard to meet a youth named Rawson, son of a lunatic Christian scientist who came to see me about two years ago. This youth is doing moral science and coming to my popular lectures, and Shephard thinks him intelligent; but he hadn't grasped the very first beginnings of what my lectures are trying to teach. I believe most of them only come for the jokes and for the thrill of an occasional paradox. It is dreadful. (#995)

In addition to his dissertation a candidate for a fellowship had to sit a three-hour exam, and Russell thought Broad's "much the best" (#206, 3 Oct. 1911). Six days later Russell reported that Broad had been elected a Fellow: The Fellowship Election is now over, quite satisfactorily to my mind. We elected Broad and Neville, the two men who stuck to my lectures to the end-Whitehead says I shall get to be known as a successful coach [the person whom all decent people in the University most despise] .... Neville (whose dissertation was not in my line) was more highly praised than I have ever heard a candidate praised. They say he attacked and completely solved a

xx ii

INTRODUCTION

problem which has baffled many of the best mathematicians for the last 100 years. Broad richly deserved election, and everybody quickly agreed about him. It is an anxious and responsible business; several of those not elected were apparently very good men. (#2I3b) A year later, when Broad held a junior teaching position in St. Andrews, Russell, in a letter of I5 March I912, offered his overall opinion of his ability: "he is very much abler than any of my present pupils except Wittgenstein. He is much the most reliable pupil I have had-practically certain to do a good deal of useful but not brilliant work" (#384). Broad continued to visit Russell whenever he was in Cambridge. On 18 June 1912 Russell recorded one such visit: Broad and a youth who is his pupil in Scotland came last night1 think it would be indiscreet to inquire into the nature of their relations. I always feel Broad essentially ignoble-he has a kind of cold flat monotonous intellectuality that makes me sick-yet apart from that essential quality of him he is good company. (#478) John Laird was also at Cambridge when Russell arrived. He sat for his degree at the end of that year, along with Karin Costelloe, who was Russell's niece by marriage. On 16 June 1911 Russell reported the results to Lucy Donnelly: You will have heard that Karin got a First with distinction, which is the best thing possible, and in many years is not got by anybody. I think she quite deserved it. She has great ability in philosophy. A certain Scotchman named Laird, who has been working at the subject about three times as long, did equally well, and was much annoyed at being no better than Karin. The exam as it stands tests ability rather than knowledge-he has much knowledge, and therefore thought there should be more testing of knowledge. In Edinburgh, where he was before, they require much more erudition, and he was praising this. But earlier in the evening he had said that until he came to Cambridge it had never occurred to him that philosophy had any connection with lifee.g. that one's philosophic views on the Deity had anything to do with whether one should go to church. I thought-and so did he on reflection-that the two were connected. It seems to me that people who have much to learn have no time to connect it with

INTRODUCTION

xxiii

life, and that for really able people, at the end of their time, it is good to have very little to learn. After Laird had left Cambridge to take up a teaching position, he too came back to visit Russell. This time his correspondent was Lady Ottoline Morrell: I am afraid most people grow less interesting when they get settled in life. Broad and Laird are both duller than when they were here. Laird came last night and we had a long argument on ethics-he was very muddle-headed, in the way that comes of no longer having any passion for clear thinking; intellectual laziness is coming on him. I told him he was muddled and he was a little vexed. He has been a whole year teaching, instead of learning. Broad will be a good teacher, for people who already have a great enthusiasm for philosophy, but for others he will be deadly. Laird will be deadly for everybody. They were the best people in philosophy in their respective years. (#505) This letter was written during the July of 1912. Given his praise of her in his earlier letter, Russell does not seem to be giving Karin Costelloe her due here. On 18 July he mentioned that Broad had given him news of Laird: I went a walk with Broad and had him to tea; he was just as usual. He says Laird (who was my pupil last year, and is now with him teaching at St. Andrews) is longing to revert to Presbyterianism and goes to chapel regularly, so I don't seem to have corrupted him very profoundly! Laird has just been appointed to a post at Halifax, Nova Scotia. He is a cautious Scot, innately respectable (like me!). (#502) Just a week later Laird came to pay him a farewell visit: I had a strenuous time with Laird last night-he is off to Canada and I shan't see him again. He was lapsing into stupidity, letting his mind grow sluggish. He said he found Logic too difficult and couldn't understand it. I told him he could if he would take the trouble, and that if he wouldn't take the trouble he had no business to teach philosophy. I made him realize all sorts of muddles he had got into, and tried to stir him up to use his faculties to the full. I put an incredible amount of energy into it, but I don't

xxiv

INTRODUCTION

know whether I produced any lasting effect.

INTRODUCTION

(#5I4, 26 July)

With Laird, at least, it seems that he failed to mould him in the way he had hoped to do: The importance of my work here is that I probably affect permanently the lives of the young men I come across, altering the direction of their work and thoughts for the rest of their days. Also .I keep alive the subject I have worked at which otherwise might die with Whitehead and me. It does seem to me worth a very considerable sacrifice. (#357) This expression of hope was written on 24 February I9I2. On II October 19II, when he was about to begin his second year of teaching, he wrote Lady Ottoline an amusing description of his probable audience: Mr. and Mrs. Mirrlees say they are coming to my lectures. He professes to have stumped me last year by the question how it is that half a sheep is not sheep, but mutton. Another man turned up saying he was coming, another normal mathematician. (G.H.) Hardy says he has told three others to come, two Scotch dwarfs and a Hindu. I think my class will afford material for the universal races congress. Hardy says he gave different accounts of me to different men: the man I have already seen was told I was very profound, the two dwarfs that I was full of jokes. But I gather they won't know it's a joke unless I say "this is an example of my bright humour". (#2I4) A week later, just before his first lecture, he had an unexpected visitor: I got home at 4.30. I had just read your letter and made my tea when Ogden (Secretary of Heretics) came to say Chesterton is speaking the night of my P(eople's) S(uffrage) F(ederation) meeting-this raised a lot of complicated problems, which we were in the middle of when an unknown German appeared, speaking very little English but refusing to speak German. He turned out to be a man who had learnt engineering at Charlottenburg, but during his course had acquired, by himself, a passion for the philosophy of mathematics, and has now come to Cambridge on purpose to hear me. This took till 5. 15; in the next few minutes I settled my business with Ogden, and then went off to my lecture, where I found my German duly established. I lectured very well, owing

xxv

to excitement and insufficient preparation. I am much interested by my German, and shall hope to see a lot of him. (#225) The next day a doubt creeps in: "My German friend threatens to be an infliction-he came back with me after my lecture and argued til dinnertime-obstinate and perverse, but I think not stupid" (#227). This tone dominates the next several mentions of Wittgenstein. "My German was very argumentative and tiresome. He wouldn't admit that it was certain there was not a rhinoceros in the room" (#238, 1 Nov.) And a day later: "My German engineer, I think, is a fool. He thinks nothing empirical is knowable-I asked him to admit that there was not a rhinocerous in the room, but he wouldn't" (#241). Then, on 2 November, he "was refusing to admit the existence of anything except asserted propositions" (#2478). Almost a fortnight later, on 13 November, Wittgenstein was still maintaining this position, which Russell felt obliged to tell him "was too large a theme" (#254). It is perhaps worth recording that shortly afterward, Russell remarked to Lady Ottoline, apropos of nothing at all: "Meanwhile I feel the real things are propositional functions" (#3I3, 18 March). By I5 March Russell despaired of teaching him: "My ferocious German (who is an Austrian I find) came and argued at me after my lecture. He is armour-plated against all assaults of reasoning,-it is really rather a waste of time talking with him" (#259). This unpromising beginning of their relationship must have troubled Wittgenstein too, for near the end of term, on 27 November, Russell reported that he was undecided about continuing his study of philosophy: My German (Wittgenstein) is hesitating between philosophy and aviation; he asked me today whether I thought he was utterly hopeless at philosophy, and I told him I didn't know but I thought not. I asked him to bring me something written to help me to judge. He has money, and is quite passionately interested in philosophy, but feels he ought not to give his life to it unless he is some good. I feel the responsibility rather, as I really don't know what to think of his ability. (#268) Two days later Russell called on Wittgenstein for the first time: "I am getting to like him; he is literary, very musical, pleasant-mannered (being an Austrian), and I think really intelligent" (#271). It was not until 23 January 19I2 that Wittgenstein responded to Russell's request for written work: Wittgenstein brought me some MS he had written in the Vacation, very good, much better than my English pupils do. I shall

xxvi

INTRODUCTION

certainly encourage him. Perhaps he will do great things. On the other hand I think it very likely he will get tired of philosophy. (#320) From this point onward the relationship between the two men begins to develop positively but with occasional hitches. On 26 January, "Wittgenstein came back to my rooms with me to propose a definition of logical form (as opposed to logical matter), to which I suggested various objections which made him very miserable" (#325). A month later, on 27 February, Russell reported another visit: This morning I had my lecture, then Wittgenstein, bringing some lovely roses-he is a dear man, and I like him very much, in spite of his being a bore. He also brought a very good original suggestion, which I think is right, on an important point in logic. He stayed while I shaved and ate my lunch, and only left me when I actually got on my bicycle. I hope my bedmaker is thinking "so that is whom Mr. Russell's flowers come from" .(#360) During this term Wittgenstein began attending G. E. Moore's lectures too. Moore thinks enormously highly of Wittgenstein's brains (Wittgenstein goes to Moore's lectures)-says he always feels Wittgenstein must be right when they disagree. He says during his lectures Wittgenstein always looks frightfully puzzled, but nobody else does. I am glad to be confirmed in my high opinion of Wittgenstein-the young men don't think much of him, or if they do it is only because Moore and I praise him. (#368, 5 March)

INTRODUCTION

xxvii

"No one could be more sincere than Wittgenstein," he wrote on IO March, "or more destitute of the false politeness that interferes with truth; but he lets his feelings and affections appear and it warms one's heart" (#375). As Russell got to know Wittgenstein better he noticed a profound similarity between teacher and pupil. He highlighted it by comparing Wittgenstein with his other German pupil: Blumenfelt (my other German, the rather stupid one) has just come to say Goodbye-he is returning to Germany. He was very shy, but managed to say very nice things about my lectures. He said he had loved them-that at first he hated the English people, and thought all the lectures poor, but mine had reconciled him to England in general. He was really sincere, not flattering. Wittgenstein on the other hand is rather a severe critic of my lectures-he says I make things seem too simple and easy, and encourage the dogmatic discipleship which I deplore. Wittgenstein is very excitable: he has more passion about philosophy than I have; his avalanches make mine seem mere snowballs. He has the pure intellectual passion in the highest degree; it makes me love him. His disposition is that of an artist, intuitive and moody. He says every morning he begins his work with hope, and every evening he ends in despair-he has just the sort of rage when he can't understand things that I have. (#385, I6 March)

very strongly-it is a very rare passion and one is glad to find it.

With the warming of their friendship Wittgenstein, as we have seen, started bringing Russell flowers, which doubly pleased Russell-the second way being that it might mislead his bedmaker into thinking that all the flowers he received, including those sent by Lady Ottoline, were from Wittgenstein. In Edwardian England it was still dangerous for a married man with a university post to be known as a philanderer. Russell's letter of 17 March I9I2 records a discussion with Wittgenstein which indicates that Russell was beginning to treat him as an intellectual peer:

He doesn't want to prove this or that, but to find out how things really are. He is very excitable and rather mad, but has excellent manners-though in argument he forgets about manners and simply says what he thinks. In spite of it all, though, something about him makes him a bore. In his flat moments, he still talks, slowly, stammering, and saying dull things. But at his best he is splendid. There is very much more in him than in any of my other pupils. (#373)

Wittgenstein came and stayed till after I2. We had a close equal passionate discussion of the most difficult point in mathematical philosophy. I think he has genius. In discussion with him I put out all my force and only just equal his. With all my other pupils I should squash them flat if I did so. He has suggested several new ideas which I think valuable. He is the ideal pupilhe gives passionate admiration with vehement and very intelligent

A few days later Russell explained what it was about Wittgenstein that pleased him: I like Wittgenstein more and more. He has the theoretical passion

INTRODUCTION

INTRODUCTION

dissent. He spoke with intense feeling about the beauty of the big book, said he found it like music. That is how I feel it, but few others seem to. Our parting was very affectionate on both sides. He said the happiest hours of his life had been passed in my room. He is not a flatterer, but a man of transparent and absolute sincerity. I have the most perfect intellectual sympathy with himthe same passion and vehemence, the same feeling that one must understand or die, the same sudden jokes breaking down the frightful tension of thought. He is far more terrible with Christians than I am. He had liked Farmer, the undergraduate monk, and was horrified to learn that he is a monk. Farmer came to tea with him, and Wittgenstein at once attacked him-as I imagine, with absolute fury. He made of course no impression. Yesterday he returned to the charge, not arguing but only preaching honesty. I wonder what will have come of it. He abominates ethics and morals generally; he is deliberately a creature of impulse, and thinks one should be. What he disliked about my last chapter was saying philosophy had value; he says people who like philosophy will pursue it, and others won't, and there's an end of it. His strongest impulse is philosophy. I wouldn't answer for his technical morals. When he left me I was strangely excited by him. I love him and feel he will solve the problems that I am too old to solveall kinds of vital problems that are raised by my work, but want a fresh mind and the vigour of youth. He is the young man one hopes for. But as is usual with such men, he is unstable, and may go to pieces. His vigour and life is such a comfort after the washed-out Cambridge type. His attitude justifies all I have hoped about my work. He will be up again next term. (#388)

treasure. I have got a number of new technical ideas from him, which I think are quite sound and important. I shan't feel the subject neglected by my abandoning it, as long as he takes it up. I thought he would have smashed all the furniture in my room today, he got so excited. He asked me how Whitehead and I were going to end our big book, and I said we should have no concluding remarks, but just stop with whatever formula happened to come last. He seemed surprised at first, and then saw that was right. It seems to me the beauty of the book would be spoilt if it contained a single word that could possibly be spared .... Yes, I think my daily round here is useful-Wittgenstein alone would have made it so. (#422)

XXVlll

The "last chapter" to which Russell refers is "The Value of Philosophy" in The Problems of Philosophy. Four days later he echoed this judgment of Wittgenstein: Yes, Wittgenstein has been a great event in my life-whatever may come of him. He has even the same similes as I have-a wall, parting him from the truth, which he must pull down somehow. After our last discussion, he said, "Well there's a bit of wall pulled down". (#397) By 23 April Russell realized that he was now learning from Wittgenstein: Wittgenstein brought me the most lovely roses today. He is a

xxix

This letter was written less than seven months after their first meeting, and it clearly establishes that they were now treating one another as colleagues. Others at Cambridge were also beginning to appreciate Wittgenstein's unique qualities. Russell arranged a tea party at which he introduced Wittgenstein to Lytton Strachey: Everyone has just begun to discover Wittgenstein; they all now realize that he has genius. He was very good at tea. He is the only man I have ever met with a real bias for philosophical scepticism; he is glad when it is proved that something can't be known. I told Hardy this, and Hardy said he himself would be glad to prove anything: "If I could prove by logic that you would die in five minutes, I should be sorry you were going to die, but my sorrow would be very much mitigated by pleasure in the proof." On reflection I found that I agreed with him. (#435, 2 May) Their transformed relationship led to new demands. In a letter of 7 May, Russell noted that after he had prepared his lecture, he took it round to Wittgenstein, "who likes to read it over beforehand", and "had half an hour's passionate discussion with him" (#440). His interaction with Wittgenstein began to alter his own philosophical outlook. On 21 May 1912 he confessed a worry to Lady Ottoline: It worries me rather having discovered that I have so little belief in philosophy. I did seriously mean to go back to it, but I found I really couldn't think it very valuable. This is partly due to Wittgenstein, who made me more of a sceptic; partly it is the result of a process which has been going on ever since I found you. It may be temporary, and I rather hope so; but I doubt it. The other

xxx

INTRODUCTION

INTRODUCTION

sort of writing, even if I can do it, ought hardly to fill all one's time, because inspiration is rare. (#459)

he had solved all the problems, because that time would never come. This produced a wild outburst-he has the artist's feeling that he will produce the perfect thing or nothing-I explained how he wouldn't get a degree or be able to teach unless he learnt to write imperfect things-this all made him more and more furious-at last he solemnly begged me not to give him up even if he disappointed me. I love his intransigeance; he makes me feel myself a puny compromiser. But I have such a strong protective feeling towards him that I find it hard to be as reckless for him as he is for himself, though I think he is quite right. (#566, 5 Sept. I912)

But the worry was not as important as it would have been had Wittgenstein not been there to carry on his work: Oddly enough, he makes me less anxious to live, because I feel he will do the work I should do, and do it better. He starts fresh at a point which I only reached when my intellectual spring was nearly exhausted. (#475, I June) His satisfaction that he had taught a pupil who could carry on his work was to be a consolation to him for several years. But getting Wittgenstein to write out his thoughts proved a difficult task. In September when Wittgenstein had returned from a summer in Vienna, he was Russell's house guest for a few days while he selected furnishings for his rooms in Trinity: Wittgenstein has gone off to Cambridge to measure his new College rooms (the ones Moore used to have), and as soon as he comes back he will go to buy furniture, because he can't stand the Cambridge shops. Tomorrow he starts for Iceland. It is a great delight having him. We talk about music, morals, and a host of things beside logic. He gives me such a delightful lazy feeling that I can leave a whole department of difficult thought to him, which used to depend on me alone. It makes it much easier for me to give up the technical work. Only his health seems to me very precarious-he gives one the feeling of a person whose life is very insecure. And I think he is growing deaf. (#564, 4 Sept.) One particularly striking example of leaving a problem to Wittgenstein is to be found in a letter Russell wrote to Lady Ottoline after his failure to answer the question he set himself in "What is Logic?" (4): "I feel very much inclined to leave it to Wittgenstein", which, as it happened, he substantially did. See the Headnote to that paper. The next day he involved Russell in a discussion of suitable furniture for his rooms: Wittgenstein is gone out to get furniture. He is very fussy, and bought nothing at all yesterday. He gave me a lecture on how furniture should be made-he dislikes all ornamentation that is not part of the construction, and can never find anything simple enough. Then I gave him sage advice, not to put off writing until

XXXI

Two days later he explained how Wittgenstein had altered the direction of his thinking: Also talking to Wittgenstein has made me feel that I must alter my lectures and put more into them. Hitherto they have contained chiefly what I have published, but I must put in more general remarks on method, and on what is and what is not possible in philosophy. (#569) Russell's interest in writing on philosophical method appears to stem from this urging, because it was only four days later that he briefly considered altering the topic of his presidential address to the Aristotelian Society from causality to philosophical method. For details see the Headnote to "On the Notion of Cause" (18). As Russell's romantic affair with Lady Ottoline deepened he allowed her to persuade him that he should try his hand at more popular forms of writing. Under her influence he wrote a novel, The Perplexities ofJohn Forstice, which has now been published in Volume 12 of the Collected Papers (1985), and Prisons, which has been lost, except for one chapter, "The Essence of Religion". Russell published it much to Wittgenstein's distress: "Here is Wittgenstein just arrived, frightfully pained by my Hibbert article which he evidently detests" (#597, 8 Oct. 1912). Three days later Russell gave her Wittgenstein's reasons for his detestation: Wittgenstein was really unhappy about my paper on religion. He felt I had been a traitor to the gospel of exactness, and wantonly used words vaguely; also that such things are too intimate for print. I minded very much, because I half agree with him. (#600) Throughout their subsequent relationship Wittgenstein was to be a very

INTRODUCTION

XXXll

severe critic of Russell's extra-philosophical writing. At this time G. E. Moore, too, earned Wittgenstein's disapproval:

INTRODUCTION

xxxiii

The following day there was much better news: Later. At this point Wittgenstein himself arrived, much more

Wittgenstein is going for Moore now. Moore has got a trick of repeating himself ad nauseam in writing and lecturing, and apparently he is also inclined to spend his time on unimportant questions. I found Wittgenstein in a state of fury last night, quite determined to tell Moore he must pull himself together. I was glad it wasn't me this time! He loves Moore but doesn't admire him as much as the others do. (#604, IS Oct.) Wittgenstein carried out his threat, and Russell informed Lady Ottoline that "Moore took it in very good part, and promised to amend if possible" (#610, 24 Oct.). During this period Wittgenstein fell sick, and for a time Russell did not know what the trouble was. But on 3I October he made a disturbing report on the state of Wittgenstein's health: Wittgenstein is on the verge of a nervous breakdown, not far removed from suicide, feeling himself a miserable creature, full of sin. Whatever he says he apologizes for having said. He has fits of dizziness and can't work-the Doctor says it is all nerves. He wanted to be treated morally, but I persisted in treating him physically-I told him to ride, to have biscuits by his bedside to eat when he lies awake, to have better meals and so on. I suppose genius always goes with excitable nerves-it is a very uncomfortable possession. He makes me terribly anxious, and I hate seeing his misery-it is so real, and I know it all so well. I can see it is almost beyond what any human being can be expected to bear. I don't know whether any outside misfortune has contributed to it or not. I had him to meet Keynes yesterday, but it was a failure. Wittgenstein was too ill to argue properly. It is funny how one person's presence will throw a new light on another person: Keynes seemed to me soft and woolly, not nearly as able as I have always thought him. This was not from anything Wittgenstein said, for he hardly talked-it was the mere effect of his being there. Keynes, like most people, will accept a view without accepting its consequences, which is what makes me call him soft.-1 also took Wittgenstein to see McTaggart, and that was more successful, though Wittgenstein found it hard to understand how McTaggart could believe such fantastic things. (#6I7)

cheerful, very full of some discoveries in his work, and no longer interested in his health. I hope it will last. (#6I8) Wittgenstein's health continued to fluctuate. In spite of his reluctance to adopt any of Russell's proposed remedies, Russell persisted in his attempts to help him: I had a passionate afternoon, provided by North (Whitehead) and Wittgenstein. I had arranged to walk with Wittgenstein, and felt bound to see North's race, so I took Wittgenstein to the river. North was beaten, not by much; he was rather done afterwards. The excitement and conventional importance of it was painful. North minded being beaten horribly, though he didn't show much. Wittgenstein was disgusted-said we might as well have looked on at a bull fight (I had that feeling myself), that all competition was of the devil, and so on. I was cross because North had been beaten, so I explained the necessity of competition with patient lucidity. At last we got on to other topics, and I thought it was all right, but he suddenly stood still and explained that the way we had spent the afternoon was so vile that we ought not to live, or at least he ought not, that nothing is tolerable except producing great works or enjoying those of others, that he has accomplished nothing and never will, etc.-all this with a force that nearly knocks one down. He makes me feel like a bleating lambkin. I soothed him down at last, and came home and soothed down North. You would think I had never felt the slightest excitement over anything, but was a mere bundle of common sense. (#629, 9 Nov. I9I2) From this time onward Russell was much involved in Wittgenstein's problems. His involvement exacted its price: "Wittgenstein has been wearing me out, and the effort to soothe him has left me no energy to soothe myself" (#634, I2 Nov. I9I2). And he reported that he had put Wittgenstein on notice: I told Wittgenstein yesterday that he thinks too much about himself, and if he begins again I shall refuse to listen unless I think he is quite desperate. He has talked it out now as much as is good for him. (#634)

xxxv

INTRODUCTION

INTRODUCTION

Despite this warning Russell still endured many more sessions with Wittgenstein of the sort reported in these letters. During the early months of 1913 Russell begins to think of Wittgenstein as his successor at Cambridge: "I think soon Wittgenstein will be able to take my place" (#678, 22 Jan. 1913). Russell's fondest hope at this time was to found a school of mathematical logic at Cambridge: "If once I had a school of mathematical logic established here, with Wittgenstein teaching, and people recognizing that it ought to be learnt, I should have no hesitation in leaving" (#693, 6 Feb.) But he is still critical of Wittgenstein, believing him to be "a very narrow specialist, and rather too much the champion of a party-that is, when judged by the highest standards" (#717, 6 March). To Russell's annoyance Wittgenstein had a way of monopolizing their philosophical topics of conversation:

all wrong, and that Wittgenstein will think me a dishonest scoundrel for going on with it. Well well-it is the younger generation knocking at the door-I must make room for him when I can, or I shall become an incubus. But at the moment I was rather cross. (#787, 27 May).

XXXlV

I find I no longer ever talk to him about my work, but only about his. When there are no clear arguments, but only inconclusive considerations to be balanced, and unsatisfactory points of view to be set against each other, he is no good; and he treats infant theories with a ferocity which they can only endure when they are grown up. The result is that I become completely reserved, even about work. (#753, 23 April) Russell made this ominous observation just a few days before he began writing Theory of Knowledge. Russell started Theory of Knowledge on 7 May 1913, setting himself a goal of ten pages of manuscript per day. On 13 May Wittgenstein came to tea and Russell told him what he was writing: "Wittgenstein was shocked to hear I am writing on Theory of Knowledge-he thinks it will be like the shilling book (The Problems of Philosophy), which he hates. He is a tyrant, if you li~e" (#775, 14 May). He had written 225 pages of the book by 26 May, and when Wittgenstein visited him the following day he made the fateful decision to allow him to read some of what he had written: Wittgenstein came to see me-we were both cross from the heat-I showed him a crucial part of what I have been writing. He said it was all wrong, not realizing the difficulties-that he had tried my view and knew it wouldn't work. I couldn't understand his objection-in fact he was very inarticulate-but I feel in my bones that he must be right, and that he has seen something I have missed. If I could see it too I shouldn't mind, but as it is, it is worrying, and has rather destroyed the pleasure in my writing-I can only go on with what I see, and yet I feel it is probably

Russell reported recovery from the attack the next day: I have recovered from the effect of Wittgenstein's criticisms, though I think in all likelihood they are just. But even if they are they won't destroy the value of the book. His criticisms have to do with problems I want to leave to him-which makes a complication. (#792) Still he maintained his rapid rate of composition, reaching page 300 on June (#793), but Wittgenstein's criticisms continued to worry him: The feeling of writing that must be done is like being in the middle of a mountain and having to tunnel one's way through before one can reach light and air. And of course I have only superficially and by an act of will got over Wittgenstein's attack-it has made the work a task rather than a joy. It is all tangled up with the difficulty of not stealing his ideas-there is really more merit in raising a good problem than in solving it. (#793) Three days later they quarreled: I had an awful time with Wittgenstein yesterday between tea and dinner. He came analyzing all that goes wrong between him and me, and I told him I thought it was only nerves on both sides and everything was all right at bottom. Then he said he never knew whether I was speaking the truth or being polite, so I got vexed and refused to say another word. He went on and on and on. I sat down at my table and took up my pen and began to look through a book, but still he went on. At last I said sharply "All you want is a little self-control". Then at last he went away with an air of high tragedy. He had asked me to a concert in the evening, but he didn't come, so I began to fear suicide. However, I found him in his room late (I left the concert, but didn't find him at first), told him I was sorry I had been cross, and then talked quietly about how he could improve. His faults are exactly mine-always analyzing, pulling feelings up by the roots, trying to get the exact truth of what one feels towards him. I see it is

1

xxxvi

INTRODUCTION

INTRODUCTION

very tiring and very deadening to one's affections. I think it must be characteristic of logicians-he is the only other one I have ever known intimately. (#798, 5 June)

Yes, I think Wittgenstein is making a very great mistake in going to Norway-I am sorry he should give up getting a degree here, which would have enabled him to teach later on, but what matters more is that he will have no distractions when his work goes wrong-it makes me very anxious. I did all I could to dissuade him, short of making myself really disagreeable. I hope he will soon get tired of it and come back. (#885, 5 Oct.)

For two more days Russell continued to write, reaching page 350 and the end of Part n, where the manuscript ends. Later he was to say that Wittgenstein's criticism was the cause of his failure to go on: All that has gone wrong with me lately comes from Wittgenstein's attack on my work-I have only just realized this. It was very difficult to be honest about it as it makes a large part of the book I meant to write impossible for years to come probably. I tried to believe it wasn't so bad as that-then I felt I hadn't made enough effort over my work and must concentrate more severely-some instinct associated this with a withdrawal from you. And the failure of honesty over my work-which was very slight and subtle, more an attitude than anything definite-spread poison in every direction. I am pure in heart again now, thanks to your divine gentleness and long-suffering .... Only yesterday I felt ready for suicide-tonight I feel gay and happy-certainly honesty is the best policy. (#811, 20 June) In later letters there are frequent references to "Wittgenstein's attack", showing that it left a lasting scar on Russell's mind. Towards the end of the summer he had happy news of Wittgenstein: I am writing under difficulties as Wittgenstein is here-he has done extraordinarily good work, and has I think practically solved the problems he was working at. You can hardly believe what a load this lifts off my spirits-it makes me feel almost young and gay. The oppression before was intolerable and affected my relation to you very much. (#856, 29 August) Russell expected Wittgenstein to continue as his pupil during the academic year 1913-14, so he was surprised when Wittgenstein returned to Cambridge and announced that he would not be staying: "he has decided he must be quite alone to work, so he is going to Norway, giving up Cambridge. I expect he will commit suicide towards the end of the winter, but it can't be helped. He has done admirable work" (#882, 2 Oct.). Two days later he again praised his work: "Wittgenstein stayed late last night and read me bits of the work he has done. I think it is as good as anything that has ever been done in logic" (#883). Understandably, Russell was distressed that Wittgenstein was leaving:

XXXVll

The next day, by way of apology for writing her a depressing letter, he suggests that worry over Wittgenstein was the cause of it: I think it was Wittgenstein-he wears me out nervously so that I long for nothing but escape from all serious thought and feeling. His decision to go to Norway was a blow and an anxiety, and at the same time he was explaining a number of very difficult logical ideas which I could only just understand by stretching my mind to the utmost. He was certainly the chief cause of my fatigue before. I feel him so terribly important and precarious that I go on making efforts when I should have given up with anybody else. For my own sake (and for yours) it is a godsend his going to Norway. It turns out that he can come back at any time to finish the residence for a degree, and he says he means to come back when he has got something written. He has promised to leave me a written statement of what he has already done before he starts for Norway. The more he talks about it, the more admirable I think it. He is certainly quite supreme. He and Whitehead and I are all in a fruitful vein, so I feel very happy about work. (#886, 6 Oct.) When he attempted to write down his thoughts on logic for Russell, Wittgenstein found he could not do it. So another way was tried: Yesterday Wittgenstein turned up as I finished with you, and was on my hands till near midnight, except a brief period when I had to deal with the prodigy (Norbert Wiener). You saw Wittgenstein's letter saying he wanted a means of preserving his work, and therefore wanted to tell me about it. I answered that I couldn't remember it that way, and he must write it down. Then his artistic conscience got in the way, and because he couldn't do it perfectly he couldn't_ do it at all. I tried one method after another: he spent Tuesday at Birmingham dictating extracts from his note-book to a German short-hand writer; then there were newer things, and things not sufficiently explained. He said he

xxxviii

INTRODUCTION

would make a statement of them, and sat down to do it. After much groaning he said he couldn't. I abused him roundly and we had a fine row. Then he said he would talk, and write down any of his remarks that I thought worth it, so we did that, and it answered fairly well. But we both got utterly exhausted, and it was slow. Today he is coming again, and (P. E. B.) Jourdain's secretary ... is coming to take down our conversation in short-hand. Mercifully Jourdain sent her this morning to borrow a book of mine so I grabbed her. It is early-closing day so no one can be got except as a favour. Tomorrow Wittgenstein goes to London, and Saturday to Norway .... All this fuss suits me to perfection and prevents me from feeling impatience, or indeed anything except the wish to drag Wittgenstein's thoughts out of him with pincers, however he may scream with the pain. (#89I, 9 Oct.) This method proved successful; the document produced, "Notes on Logic", was edited by Russell and used by him in his courses at Harvard University during the spring of I9I4. Wittgenstein did indeed go to Norway, and, while he was there, in February 19I4, he wrote Russell breaking off their relationship. On I7 February Russell wrote to Lady Ottoline: Since I began writing this, I have had a letter from Wittgenstein saying he and I are so dissimilar that it is useless to attempt friendship, and he will never write to me or see me again. I dare say his mood will change after a while. I find I don't care on his account, but only for the sake of Logic. And yet I believe I do really care too much to look at it. It is my fault-I have been too sharp with him. (#989) A few days later Russell sent him a conciliatory letter, but they did not meet again until after the First World War ended. When the war broke out, Wittgenstein returned to Austria and joined its army, thus becoming an enemy soldier. The thought of this greatly distressed Russell: It seems strange that of all the people in the war the one I care for much the most should be Wittgenstein who is an "enemy". I feel an absolute conviction that he will not survive-he is reckless and blind and ill. I can know nothing till the war is over. If he does survive, I think the war will have done him good. (#II48, I2 Nov. 1914)

Wittgenstein did survive; he spent the latter part of the war in a prisonerof-war camp where he completed the manuscript that was later published

INTRODUCTION

XXXlX

in England under the title Tractatus Logico-Philosophicus (I922). All their later interactions, including those concerned with finding a publisher for Wittgenstein's book, were strained and even sometimes difficult. Their best and most fruitful times together fell between 19u and 19I3. IV. THE PROBLEMS OF PHILOSOPHY

The Home University Library was a publishing venture designed to provide those without access to higher education with a set of cheap books covering the whole range of human knowledge. By selecting authors eminent in their fields, the editors and publishers hoped to ensure that, at the time of their publication at least, the books would embody the latest advances in their subjects. Gilbert Murray, a friend of Russell, was designated as one of the editors. He persuaded Russell to undertake the basic book in philosophy, but, because he had also engaged G. E. Moore to write the introduction to ethics and others to write on religion and politics, Russell was to confine his book to epistemology and metaphysics. Russell agreed to this proposition early in I9II, with delivery of the manuscript set for July of that year. The first mention of the book in the Ottoline letters occurs in a letter of 25 March 19n: "I have also undertaken to finish by July a popular book on philosophy, which I have not yet begun. Heaven knows how I shall manage, but I must do it as I have signed the contract" (#6). On his thirty-ninth birthday, 18 May, he was beginning to see his way: I find my thoughts are gradually coming round to the book I have to write then-I have a certain number of more or less vague ideas on the subject. But it will be a very difficult book to write. (#72) The "then" refers to "the fortnight I promised to spend with North (Whitehead) at Marlborough" in July. In an unnumbered letter, probably written on the 26th of May, he referred to the book's gestation: All this time ideas about the popular book I have to write are growing underground. I expect it will go easily when I get to work, and much the better for the time of not explicitly thinking about it. There is a great deal of all sorts going on in my mind, but I can't make out what. I feel though that something good will come of it, as soon as I have leisure to get it out. The pressure of other work delayed his start on the book until 3 June: "I

xl

INTRODUCTION

made a beginning with the popular book today but was very uninspired and only wrote five pages" (#99). On the morning of 5 June he resumed writing: "I dare say I shall get the popular book finished in July after all" (#102). But progress was slow, as this report, probably written on 22 June, testifies: Since then I have written more of my Shilling Shocker-I am getting into it now gradually, and I think I shall probably get it finished in July. I hate having to write to order-it is so much pleasanter to write as the spirit moves one. Still there are ways of moving the spirit. (#120) More often than not Russell, in his private correspondence, referred to the book as his "Shilling Shocker"; this nickname reflected both the projected selling price and the popular level of writing required to appeal to a wide public. On 24 June he reported that he had finally got the style for the book right: "I have made a new start with the Shilling Shocker, and feel it will go better now-I have at last got hold of the right style, which I hadn't before. Now I shall enjoy writing it" (#128). In another letter written on the same day, he indicated that a considerable manuscript had accumulated: I am getting on with my book. Would you like to see what exists, still rough, and going to be improved? Or shall I wait till it is in a more final shape? As I am not to deal with either religion or morals, I have had to confine myself to topics of which the interest is purely and exclusively i~tellectual. Thus there is not much scope for much that might otherwise be said. (#126) The next mention of the book, in a letter of IO July, is an optimistic one: "My writing on my Shilling Shocker goes very well, and I am getting well into it. I shall easily finish it in time I think" (#143). The next day the news is even better: I am glad the MS pleases you, and shall be very glad of your criticisms-the more the better-though I dare say I shall try to rebut them if I can. Yes please keep it till I come. I have written 2000 words this morning. I think the best so far. I began with two pages on Plato which delighted me as clear exposition. But after that it got rather difficult. (#144) By 12 July he was able to say that "quite half my book (is) done" (#145),

INTRODUCTION

xli

and at midnight he wrote to her again: It took me some time to get under way with this book, but now that I am under way it goes very well and I find my mind in very good order. I am always afraid of growing stupid but it hasn't begun yet, quite the contrary. It really is fortunate to have work to do that one believes in. Doing this book has given me a map of the theory of knowledge, which I hadn't before. From that point of view it will have been a great help to my own work. (#146)

Earlier in the day he had written his first progress report to Gilbert Murray: I am writing my book for your series, and have written more than half of it. I find, however, that one or two things are happening to it which you may not desire. In the first place, I find it deals almost entirely with theory of knowledge, only occasionally arriving at metaphysics through theory of knowledge. This seems difficult to avoid, owing to the exclusion of religion and ethics. In the second place, I find that, quite contrary to my intention, it is an exposition of my own views, not an impartial account of what is thought by various philosophers. I found it impossible to write interestingly or freshly or with conviction, unless I was trying to persuade the reader to agree with me. In the third place, I find that after the first few chapters it grows rather difficult. It remains quite easily intelligible, without trouble, to any educated man, however little he may know about philosophy; but it would be difficult for a shop-assistant unless he were unusually intelligent. I hardly know myself whether it is too difficult or not. If it is, I must re-write it. Don't bother to answer if you think it will be all right. But if you really wish stupid shop-assistants to be able to read it in armchairs, I must do it again. He proceeds to list the titles of the first eleven chapters of the book; a check of them against the published titles shows only slight modifications. He then concludes: "The chapters grow naturally out of each other, and but for doubt about difficulty I should be satisfied with the stuff." Murray generally approved of Russell's plan, although he did want reassurance that Russell had not introduced many purely technical terms and that the proposed title for the book was still appropriate. Russell was able to reassure him:

xiii

INTRODUCTION

INTRODUCTION

I think in view of what you say my stuff will be all right. I have used hardly any technical terms, and the very few that are used are introduced with a good deal of explanation. There is a fair amount of metaphysics, but there seems to me so little that is ascertainable in metaphysics that I felt I must make theory of knowledge the framework. However, the two mix: Platonic ideas, e.g., which I believe in and expound, are very much on the border-line. The title will do, I think.

Gilbert Murray about the book, also a letter from the firm. Read Gilbert's first. You will see he is quite satisfied. He hadn't yet had the two chapters I read yesterday. (#169a)

Even though the writing was progressing nicely, Russell still seemed to feel that the work on the book was interfering with his own personal development, especially as it related to her. On 14 July he informed Lady Ottoline that he had written "most of my little book". "It was a wrench bringing my mind on to it, and I hardly knew beforehand whether the effort would not, as you once said of yourself, tighten up my love inside, but it didn't" (#147). And three days later we find this outburst: "I wish I had finished this beastly book. I resent the way it compels me to turn my thoughts away from you and things that might interest you" (#151). A letter of 22 July (#155) reveals the fact that the well-known table in the first chapter was not Ottoline's, but one in his rooms at Trinity College. On 31 July he finished Chapter 12 on the nature of truth, and on 4 August he wrote Chapter 14 on the limits of philosophical knowledge. Presumably Chapter 13 on knowledge, error and probable opinion, which is not mentioned in his letters to her, was written between these two dates. This hypothesis is confirmed by the fact that the next day he notes that he has only one more chapter to write. On the 31st of July, just after returning from taking Chapter 12 to the typist, he was worried: "I have lost interest in the shilling shocker, and am in danger of finishing it in a perfunctory fashion, which I ought not to do" (#163). On 7 August he wrote to Murray: I am sending you today the greater part of my treatise on phi-

losophy for your shop-assistants; the rest will follow quickly. There will be three more chapters-two are now being typed, the third is not yet written. I think the total is about 46,000 words, so if you think any portion obscure from brevity there will be room for some expansion. I couldn't say much about metaphysics because so little of it seems to me true. Four days later he hoped to finish the book: I wrote two pages of my chapter last night, and have written four more now. I shall easily get it done today. I enclose a letter from

xii ii

Later that day he reported the completion of his famous chapter on the value of philosophy: I finished my last Chapter of the Shilling Shocker before tea, and sent it to be typed. It was not so good as it ought to be, but I didn't see how to improve it. I don't think it was bad, but it was rather too short. I took some sentences on Contemplation out of the Prisons MS, and the denunciation of those who make man the measure of all things. (#173)

During this period Russell was trying to please Lady Ottoline by writing in a more imaginative and inspirational vein than usual. Prisons was one such attempt. What remains of this aborted project has been printed in Volume 12 of the Collected Papers (1985). The soaring prose of "The Value of Philosophy" is our best indication of the sort of literary style he hoped to achieve in Prisons. The next day, I2 August, he replied to a letter from Murray. One paragraph deals with the philosophical content of the book: I did not mean to say that no physical objects were in any way

connected with dreams, but that the objects one would naturally infer from the sense-data in dreams don't exist-e.g. the Black Hole in your dream. I will add something to make the point clear. As regards the cat, the theories you suggest are of course possible, but seem to me less simple than the hypothesis of a real cat. This metaphysical cat makes its appearance in the chapter on the existence of matter where Russell argues that a cat cannot be regarded as a mere bundle of sense-data, because the cat can appear first in one part of a room and then in another, so it must have moved. Were it only a bundle of sense-data it would be necessary to suppose that it went out of existence in one part and sprang into existence in another part, an hypothesis Russell regards as less plausible than the one of a real cat, capable of moving about the room on its own. Russell wrote to Murray on 20 August informing him that the corrected manuscript had been sent to the publishers: I paid attention to the points you criticized, generally by making additions. Your point about what I know when xRy is true and

xliv

INTRODUCTION

INTRODUCTION

R is unknowable I forgot to go into because there was no note in the margin. The answer is in the principle: Any proposition which can be known or even understood must be composed entirely of constituents with which we are acquainted. Thus your case does not arise. But the working out of this principle is complicated. See forthcoming Aristotelian Society's Proceedings. I extracted the island in the western ocean, reluctantly. I never said Plato said so, but it seemed to me any mystical reader might naturally regard the ideas as living in Atlantis until they were corrupted, and I should have thought Plato intended this. What do you say? As for Hegel, I stick to my point, and have merely expanded the statement Hegel turns (as do your objections, I think) on confusing knowledge of things with knowledge of truths. Acquaintance with a thing does not (theoretically) involve any knowledge of truths about the thing, and in practice involves often very little such knowledge.

seems to me better worth doing. It is all puzzling and obscure. For so many years I have had absolutely no choice as to work that I have got out of the way of wondering what is best to do. I think really the important thing now is to make the ideas I already have intelligible. And I ought to try to get away from pedantry. My feelings have changed about all this; I did think the technical philosophy that remains for me to do very important indeed. (#286)

The article to which Russell makes reference is "Knowledge by Acquaintance and Knowledge by Description" (15), several paragraphs of which are also to be found in The Problems of Philosophy. It is also very likely that "On the Relations of Universals and Particulars" (16) and "On the Notion of Cause" (18) derive in part from the thinking about epistemological and metaphysical problems that preceded the writing of this book. Proofs for the book began to arrive on 2 November 1911, and on 9 November Russell complained to Murray about alterations the publisher was making to his manuscript: I don't know whether it is Perris or who it is, but I find they have altered my paragraphing in printing. In some cases it didn't seem to matter much, but in others it destroyed the style. I generally put quite a different sort of sentence at the end of a paragraph from any that I should put in the middle-I make the ones in the middle obviously incomplete. I hope Perris or whoever it is won't insist. On that day too he returned the last of the corrected proof sheets. Before the book was published, on 13 December, he revised his opinion of its worth: I have an uneasiness about philosophy altogether; what remains for me to do in philosophy (I mean in technical philosophy) does not seem of first-rate importance. The shilling shocker really

xiv

The book was published in late January 1912. "My shilling shocker is out", he wrote Lady Ottoline on 24 January, "I have seen it in the shops but have not got any copies of it. I will bring a copy on Saturday" (#323). By 27 January he had a copy of the book: "It seems all right on re-reading" (#326). As was noted earlier Wittgenstein did not like the chapter on the value of philosophy: "He dislikes the last chapter of my shilling shocker, I think because it vexes him that one should hint at philosophy having any end outside itself" (#387, 16 March). The next day he amplified on Wittgenstein's dislike: "What he disliked about my last chapter was saying philosophy has value; he says people who like philosophy will pursue it, and others won't, and there's an end of it" (#388). One of the people to whom Russell turned for advice on the literary worth of his popular writing was Goldworthy Lowes Dickinson. Earlier Russell had sent him a manuscript of one of his non-philosophical efforts, probably Prisons, and when The Problems of Philosophy was published he sent him a copy of it. On 5 March 1912 he proudly quoted from a letter Dickinson had written him: Goldie writes: "I have just finished your 'Problems of Philosophy'. I think it is a perfect masterpiece-in a way the most wonderful book I ever read-so temperate and so 'beautiful' in the mathematical (Platonic) sense. Your last chapter is particularly good. It's better (I think) than the MS you sent me. But I don't mean to be at all discouraging about that." (#366) Six days later he was still elated: "My mind has been very full of ideas, owing to having to do Bergson. That and Dickinson's letter made me feel perhaps I could really express a great deal of what is important by means of philosophical work" (#379). The book proved to be a best-seller. By 25 September 1912 13,000 copies of the British edition had been sold: I have had a letter from Perris saying they want a new edition of

xlvi

xlvii

INTRODUCTION

INTRODUCTION

my shilling shocker (only 2000 left out of 15000) and if I want to make any corrections I must do so by Monday. So I must go to Cambridge as there are several historical points on which McTaggart and others say I am mistaken, and I must look them upchiefly the interpretation of Berkeley. The time is shamefully short and I can't possibly get everything done in the time that wants doing. (#587)

(1903) and Principia Mathematica (1910-13), he believed he had, with Whitehead's assistance, established that much of mathematics was a branch of logic. While Principia was going through the press, he turned his attention to his long-postponed project. What he proposed to do was to use the logical tools that he and Whitehead had developed for mathematics in order to analyze the central concepts of classical physics, providing definitions of them in terms of logical constructions. These logical constructions would be formulated in the language of Principia plus any other terms which were required to capture the subject-matter of physics. The finished project would resemble Principia in its structure; there would be a set of axioms special to physics and theorems deduced from them using only logical rules. The central special term of physics, in his opinion, was "matter", and in 1912 he turned his attention to its analysis. In the Introduction to Volume 8 of the Collected Papers there is a full account of his many attempts to analyze "matter". The running account that he provided for Lady Ottoline of the frustrations he encountered along the way are fully documented there. In this volume further documentation of his attempts is reprinted. In addition to one completed, but unpublished, essay, "On Matter" (10), there are nine short working papers (11) exploring some of his ideas concerning the correct analysis of "matter". After thinking about matter for some months and trying out all of the ideas for its analysis that occurred to him, he concluded that certain epistemological problems required solution first:

Russell had only five days to submit the corrections. He finished them on Saturday, two days ahead of schedule: Yesterday after writing to you I worked continuously till I went to bed, except just during dinner. I began the day with a small triumph. McTaggart had told me confidently that my interpretation of Berkeley was wrong, and I had believed him, as he lectures on Berkeley and knows a lot of history of philosophy. But I looked up passages, and when he came yesterday morning to argue about it, I succeeded in persuading him I was right. The point is one on which Berkeley at different times held two incompatible views-McTaggart had got hold of one, and I of the other. Then we had a great argument about Hegel, and there too I made him admit that my interpretation was quite possible. All philosophers are very hard to understand exactly, because they are all in a muddle, and believe things that contradict each other; this is what is called depth-anything clear is regarded as shallow. -Then I quickly finished the corrections in the Shilling Shocker; I had thought I should have to re-write all the parts about Berkeley and Hegel, but that proved unnecessary. (#591) The "new and revised edition" of The Problems of Philosophy exhibits only very minor changes, nearly all of which, except typographical corrections, were restored to the text in the 1943 printing. V. THE PROBLEM OF MATTER

The problem of matter had occupied Russell's thoughts off and on since 1897, the year in which he published An Essay on the Foundations of Geometry. When that book was finished he decided that his next major task would be to provide philosophical foundations for physics. But almost immediately he realized that if the task were to be satisfactorily brought to completion, the philosophical foundations of mathematics must be secured, so his energies were diverted into this preliminary investigation. Sixteen years later, after the publication of The Principles of Mathematics

This morning I am working on theory of knowledge with a view to my lectures-it seems so easy to see what wants doing, and to do it with zest. I believe I shall have got a great part of that course done when you come back from Lausanne. I want it to be a book-,--for a long time I have planned a book on theory of knowledge, then I thought I could do Matter first, but now I see that even apart from having to lecture, theory of knowledge must come first-chiefly because of dreams. There are some subjects which it is considered bad form to mention in philosophy, and dreams is one of them, I suppose because almost everybody's philosophy is refuted by them. (#750, probably 22 April 1913) The lectures to which he refers are the Lowell lectures, which are discussed below in Section VIII. By 4 May he was "still sketching out my book on Theory of Knowledge-I have got the early part quite elaborately sketched, and the whole pretty fully in my head" (#764). Four days later he began writing the book.

xi viii

xlix

INTRODUCTION

INTRODUCTION

VI. THEORY OF KNOWLEDGE

published as a book. At first he toyed with "the search for insight" as his topic:

Although his attempt to solve the problem of matter was the immediate cause of his turning his attention to epistemological questions, it was the not the only one. It is clear from the surviving documents that Russell believed that the logical tools developed in Principia could, in capable hands, lead to a revolution in the way philosophy was done. In his letters to Lady Ottoline he usually refers to this projected new philosophy as "technical philosophy" to contrast it with the "popular" philosophy of The Problems of Philosophy. His work on definite descriptions is a paradigm of technical philosophy. His attempt to analyze "matter" would, if successful, create a new branch of technical philosophy. An early document of the new philosophy is "Analytic Realism" (14), one of two lectures-the other being "The Basis of Realism" (13)-he delivered in Paris in I9I I. Here for the first time he introduced the term "logical atomism". As the title suggests he espoused a realistic position both in theory of knowledge and in the theory of universals, and adopted, as his philosophical method, the analysis of complexes into their constituent simples. These simples are the logical atoms of the new philosophy. There seems to be no doubt that this new philosophy is modelled on the work he and Whitehead did in Principia. The structure of the logical constructions arrived at by analysis would be stated in the language of Principia, and the logical atoms on which the analysis rested would occupy positions in the formulas analogous to the places of variables in Principia. As we have seen, one unexpected benefit Russell derived from writing The Problems of Philosophy was to find at the end that he had "a map of the theory of knowledge" ( # 146) which he had not possessed before he began work on the book. Having discovered this map and feeling an urgent need to solve certain epistemological questions before he could get on with the problem of matter, the time seemed ripe for him to write a big book on the theory of knowledge. At some time during 1912 he committed himself to this project. His letters to Lady Ottoline, some of which have been quoted above, provide a running account of his thinking about the problems he would address in the book, which he called Theory of Knowledge, and the writing of it, up to the point where he abandoned it because of Wittgenstein's criticisms. Since the introduction to Volume 7 of the Collected Papers contains a full account of this aborted enterprise, it seems unnecessary to discuss it any further here. VII. OUR KNOWLEDGE OF THE EXTERNAL WORLD

Early in 1913 Russell was invited to deliver the Lowell Lectures in Boston the following spring. The lectures were to be delivered in person and also

I am very happy about that scheme of lectures. I don't know whether to call it "the search for wisdom" or "the search for insight." I think perhaps insight is better. I have been walking round the street thinking about it. I can think of the following heads: I.

II. III.

IV.

v. vr. VII.

Nature and importance of the problem Insight in politics: Lincoln and Mazzini Insight in human relations: who? Lamb Insight in poetry: Dante Insight in science: Galileo Insight in philosophy: Spinoza The nature of insight.

I want one more heading, to make eight lectures. Perhaps history? "Insight" doesn't seem quite the right word. But I can't think what would be right. ... I feel convinced I can do the lecturesthey will be what Prisons failed to be. (#737, 8 April) "The search for insight" seems to have been considered as his topic for the Lowell lectures largely to please Lady Ottoline. During this period he was constantly searching for writing projects which reflected her central interests, most of which bordered in one way or another on inspirational religion. In June he proposed to the Lowell Institute that he lecture on "the place of good and evil in the universe", but the Institute rejected this topic, "because of the terms of the trust lecturers are not allowed to question the authority of scripture" (#808, 14 June), so he decided on a more epistemological topic, which he referred to as "popular lectures on Scientific method" (#813, 21 June). Between the time he agreed to deliver the Lowell Lectures and the composition of them, he wrote Theory of Knowledge. The experience of having to abandon that book left him depressed about his ability to do original philosophical work. By 1 September 1913 he had recovered sufficiently to begin composing his lectures. "I have begun writing my popular lectures. The work does not interest me at all .... I will send you the lectures as they are written" (#859). Two days later the first lecture was nearly complete, and he judged it "poor stuff" (#735). The topic was now "on scientific method in philosophy", which is the subtitle of the published book. The second lecture was finished on 5 September, and six days later the

INTRODUCTION

INTRODUCTION

third was underway: "I am writing my third lecture. What bores me is that there is no new thought in these lectures, only old ideas that I am trying to make easy to understand" (#864). By 14 September his attitude has changed: "I make a point of writing ro pages a day, ... I have nearly finished my fourth lecture. I think I will re-write them all during the Xmas Vacation, but they are getting better and I am getting more into the mood" (#868). On 16 September he is in the middle of the fifth one, and he feels a need for jokes: "I am told one joke at least is de rigueur in an American lecture" (#869). Of this lecture, which is fourth in the published book, Russell remarked: "The last of the lectures I have sent you gives a brief resume of what I have arrived at about Matter" (#873). By 22 September he had "nearly finished the seventh": "I have at last got really interested in my lectures" (#874). The next day he had finished the seventh, but was dissatisfied with his ordering:

the composition of these two pieces in this passage from "How I Write" (1956b):

I find the order of the lectures is wrong-I will rearrange them when I have finished. From what you say, I suppose the third and fourth are the most difficult; if so, I had better put them at the end. I think I can build the whole course round the problem of our knowledge of the external world, and bring in Zeno in that connection, towards the end of the course. (#875) During the evening of 25 September he wrote that "I have finished my last lecture" (#876), just 25 days after he started. But he admitted, in the same letter, that he would have to re-write them. During the next weeks he occasionally referred to the lectures, mentioning on 1 October that "all except the seventh are too short" (#881). He went on to tell her that in Lecture yn (now III) he had "got a new idea about matter-the distinction and. relations of the three kinds of space". In an undated letter, which was probably written on 13 October, he reported that he has "three lectures into final shape, and nearly finished getting a fourth in order" (#892). The fifth lecture was put in the final form on 19 October, and by 2 November both the sixth, "the only erudite one of the lot" (#904), and the seventh were done. A week later the eighth and final one had been revised, and the whole set sent to the typist. In a letter postmarked 27 January 1914 he remarked that he "re-wrote part of one of my popular lectures" (#976). Kenneth Blackwell (1973) has argued, convincingly in my opinion, that this refers to published Lecture III, pages 87-93. This passage introduces a distinction which is also to be found in "The Relation of Sense-Data to Physics" (1914); it is the distinction of the two places associated with every aspect of a thing, namely, the place at which, and the place from which, the aspect appears. Blackwell advanced his suggestion to explain how Russell might have confused

Ii

I had undertaken to give the Lowell Lectures at Boston, and had chosen as my subject "Our Knowledge of the External World". Throughout I9I3 I thought about this topic. In term time in my rooms at Cambridge, in vacations in a quiet inn on the upper reaches of the Thames. I concentrated with such intensity that I sometimes forgot to breathe and emerged panting as from a trance. But all to no avail. To every theory that I could think of I could perceive fatal objections. At last, in despair, I went off to Rome for Christmas, hoping that a holiday would revive my flagging energy. I got back to Cambridge on the last day of 19I3, and although my difficulties were still completely unresolved I arranged, because the remaining time was short, to dictate as best I could to a stenographer. Next morning, as she came in at the door, I suddenly saw exactly what I had to say, and proceeded to dictate the whole book without a moment's hesitation. Pp. 195-

6. From his correspondence we know that he did dictate "The Relation of Sense-Data to Physics" and "Mysticism and Logic" very early in 1914. The first of these essays contains his earliest exposition of the theory of perspectives and perspective-space, according to his own account in My Philosophical Development (1959, 105). If later, as Blackwell has suggested, he incorporated the theory into his Lowell lectures, then the mix-up in his rather dramatic account of how that book came to be written is understandable. But understandable or not, we know from his letters to Lady Ottoline that it is mistaken; Our Knowledge of the External World was written during September, and revised during October and November, of 1913. VIII. CONTROVERSIES WITH OTHER PHILOSOPHERS

Sixteen of the papers (21-36) printed in this volume fall under this heading. Many of them are reviews. As we have seen, Russell emerged from his work on Principia with a developed realistic philosophy, which he was anxious to defend as superior to any of the prevailing philosophies. In his youth he had succumbed for a period to the spell of a kind of Hegelian idealism, which, with G.E. Moore's help, he had thrown off. The realism they adopted was in direct conflict with idealism, and they used it as a basis from which to criticize idealistic writings. Despite a deep philosophical disagreement with F. H. Bradley, Russell had a profound respect

INTRODUCTION

INTRODUCTION

for him and tended, when criticizing his views, to play up the agreement and play down the disagreement. But lesser idealists were treated more harshly. Russell's view of them was that they had no wish to be clear; that they confused obscurity with profundity; and that they were too optimistic. He read their books, not from a desire to know what they were saying, but because he was asked to review them. In this way he kept abreast of developments, such as they were, in the idealist camp. Besides rejecting the idealist's view that everything is mind or is minddependent, he rejected their logic virtually in toto. A major part of his case against the idealists is that they wrongly supposed that relations were "internal", so that if A is known by B its being known by B must form part of the "nature" of A. The precise character of this claim and, therefore, exactly what was meant by rejecting it and taking relations to be "external" rather than "internal" was much disputed. In "The Basis of Realism" (13) he relates his rejection of internal relations to the denial, first, that if A is related to B there must be some characteristic of A in virtue of which it has this relation to B, and, second, that it is impossible for precisely the same term to be a member of two different complexes. The rejection of the theory of internal relations is thus linked with a rejection of the Absolute Idealist view that to know anything we have to know everything in favour of the view that we build up our knowledge by coming to a complete understanding of particular situations, the path which Russell takes to be the route to genuine progress in philosophy. William James was also, in this sense, a pluralist and Russell often writes sympathetically of him. But in "Pragmatism" (21) and the other articles discussing James's philosophy, Russell is mainly intent on criticizing James's special version of pragmatism, to which Russell was completely hostile. Russell was opposed to pragmatism in any of its forms in so far as it asserted that the test of a proposition's truth lay in the degree to which it was profitable to accept it. But he was particularly critical of what James called "the will to believe", although "the right to believe" would be a less misleading term for it, which allowed a person to believe one of two competing propositions, when past and present experience failed to decide between them, on the ground that it was more emotionally satisfying than the other. Russell had only contempt for this doctrine. And he scorned James's contention that the true was what it paid to believe even on James's very broad interpretation of "paid"; his objection concerned linking truth to consequences. One should look for supporting evidence, he thought, and if you did not find it, then you should suspend belief:

whether the effects of thinking them infallible are on the whole good. Yet this question, of the truth of Roman Catholicism, is just the sort of question that pragmatists consider specially suitable to their method. (1910, 135)

Iii

It is far easier, it seems to me, to settle the plain question of fact: "Have Popes been always infallible?" than to settle the question

!iii

His critique of the pragmatic theory of truth, and the underlying pragmatic theory of meaning, led him to appreciate the attractiveness of the correspondence theory of truth. Pragmatists defended correspondence as one important test of truth, the other being coherence with well-established beliefs, with only James permitting the exercise of "the will to believe" when these two tests failed to decide a case. Russell came to think that only correspondence with facts was required to account for the truth of empirical propositions. Russell at first thought James's doctrine of neutral monism was also mistaken and he spent considerable time and energy between 1913 and 1921 attempting to prove he was right, but he was unable to make a convincing case. By degrees he came round to adopting it as his own position. But this did not happen until he wrote The Analysis of Mind (1921). The introduction to Volume 8 of the Collected Papers (Section III) provides a detailed account of his transition from rejection to acceptance of this doctrine. Russell knew and liked William James, and, as with Bradley, he treated him more kindly in print than he did some of his followers, F. C. S. Schiller for instance. Russell also liked Schiller but he did not think highly of his philosophical ability. All of the pragmatists, James included, offended him by giving human beings a central place in their philosophy: But the worship of my life, as you said, is Truth. That is the something greater than Man that seems to me most capable of giving greatness to Man. That is why I hate pragmatism-do read the last paragraph of my essay on Pragmatism in my book, where I have tried to express this. (#71, 17 May 1911) The passage he urged her to read is the last paragraph of Paper 21 in this volume. Russell often linked Bergson to the pragmatists, but it seems better to treat his opinion of Bergson and his work separately. At the time Russell wrote about his views Bergson enjoyed a wide following, both among philosophers and the public at large. To many he seemed very up-to-date with his use of evolutionary language and his interest in modern discoveries like the cinema. Russell, however, did not admire him. His basic criticism, made explicit in Scientific Method in Philosophy (1914), was that Bergson based his philosophy on the results of science, which Russell ar-

liv

INTRODUCTION

INTRODUCTION

gued was always a mistake. Science is continually changing; its results at any given moment are only tentative and subject to correction. Therefore, to select any moment in the history of a science-in Bergson's case biology-and erect a philosophy on its findings at that moment, is bound to yield a philosophy which will be out-of-date in the near future. Ad hoc additions will have to be made to it as the science upon which it is based changes, resulting in a philosophy whose deficiencies gradually become manifest. Only the method of science should be adopted by philosophers in Russell's view. The resulting philosophy will be piecemeal and tentative, but it will be soundly based. Russell also objected to the way in which Bergson presented his philosophy. Bergson depended too much on metaphor and too little on argument. It was this characteristic of his writing which, Russell suspected, appealed to a wide public. From a philosophical point of view it was not clear just what Bergson was saying, and, since there was almost no argument, it was just as unclear what truth there was to his points when they were literally interpreted. On a more substantial point, Russell examined at length Bergson's doctrine of space and time, along with correlative doctrines of memory and intuition, and accused him of relying on an outmoded concept of continuity which ignored the more recent mathematical analyses. Russell undertook to read Bergson because of his popularity among students at Cambridge. C. K. Ogden invited him to address "The Heretics" on Bergson's philosophy, and Russell agreed. He devoted much time to reading Bergson's work and the resulting paper was so long that the session had to be divided into two parts, so that the speaker (and his audience) could recoup their energies. After Russell spoke, and his paper was published, those who had felt uneasy about Bergson's philosophy were better equipped to criticize it. As the reader will have gathered by this time, Russell's opinion of most of his philosophic contemporaries was rather unfavourable. In a letter of 2 June I9I2 written at the end of a day on which he attended a joint meeting of the Mind Association and the Aristotelian Society, indulging in argument and "endless discussion; the whole time on the go, and enjoying it", he reflected on his fellow philosophers:

analyze or get things sharp and clear. They all lack passion. One must be mad to be able to think. These days my brain has been at its very best; cold, cutting intellect, probing down to the heart of things. (#476)

It is not easy for me to take a back seat gracefully, but I shall learn in time I dare say. Intellectual ambition is the chief thing that helps, and seeing a lot of philosophers stimulates that. And the hopeless muddle-headedness of almost all of them fills me with a passionate desire to make them think more clearly; as long as desires of that sort are uppermost I feel happy. Poor (A.D.) Lindsay is an excellent good soul, but helpless when he tries to

lv

On the morning after he read "On Matter" (10) to a Cambridge audience he put the blame for philosophic incompetence on bad education: All the philosophers I meet make me feel how very bad the education of philosophers is-they are mostly unaccustomed to things where one can be definitely right or wrong, and therefore precision is not part of their ideal. Making machines, like Wittgenstein, is a much better training-if a machine won't work it . ' is no use appealing to the reason against the understanding, or to the nobler parts of our nature, etc. etc. I should like to start philosophers on a different tack, and bring back the union of philosophy .and science that existed in the seventeenth century, as well as m Plato and Aristotle. But able men are sadly few-it strikes me with my pupils. None of my pupils this year will ever ~o an inch beyond what I tell them. One of them is a physicist, Just the sort of man who ought to be good by my theories but he is quite hopeless. (#608, 26 Oct. 1912) ' His subse.quent work on developing scientific method in philosophy stemmed m large part from his desire to improve the education of the young in philosophy. IX. "TECHNICAL" VERSUS "POPULAR" PHILOSOPHY

As we have seen above, Russell regarded The Problems of Philosophy as "popular philosophy", and at times he thought popular philosophy was "better worth doing" (#286) than any work he could do in technical philosophy. It is clear from his letters that he meant by "technical philosophy" the application of the techniques of symbolic logic to some of the traditional problems of philosophy. A lecture (14) he delivered in Paris in March 191 l outlines his program in technical philosophy: The philosophy which seems to me closest to the truth can be called "analytic realism". It is realist, because it claims that there are non-mental entities and that cognitive relations are external relations, which establish a direct link between the subject and a possibly non-mental object. It is analytic, because it claims that the existence of the complex depends on the existence of the sim-

!vi

INTRODUCTION

INTRODUCTION

pie, and not vice versa, and that the constituent of a complex, taken as a constituent, is absolutely identical with itself as it is when we do not consider its relations. The philosophy is therefore an atomic philosophy. (I33: I-9)

knowledge is a mixture of both. The analysis of a piece of actual knowedge into pure sense and pure apriori is often very difficult, but almost always very important: the pure a priori, like the pure metal, is infinitely more potent and beautiful than the ore from which it was extracted. As regards the mathematical element in science, Principia Mathematica does the extraction very elaborately. But there are a number of other more elusive a priori elements in knowledge-such problems as causality and matter involve them. It is these that I want to get hold of now-I am only quite at the beginning-it is a vast problem of analysis, wanting tools that one has to make oneself before getting to work. It is very hard, to begin with, to make out what science really asserts-for example, what the law of gravitation means. Neither science nor philosophy helps one there-mathematical logic is the only help. And when one thinks one has found out what it asserts, one can't state the result so as to be intelligible to any one who doesn't know mathematical logic. So one's audience must be small! (#6I6)

This position was later to become widely known as "logical atomism", and to its development Russell devoted much time and attention during the next several years, culminating in I9I8 in his lectures on logical atom1sm. In technical philosophy he believed that he had got hold of the right method and that the proper application of the method to certain philosophical problems would yield solutions which were partial and tentative but which, after further investigation by himself or others, would finally be established as true. In a letter of 29 October I9I2 he discussed the power of his method: The impulse to philosophy remains very strong in me-I feel that I really have got a method that gives more precision than there has ever been before, and more power of getting at the skeleton of the world-the framework that things are built on. It is difficult to get people to see it, because philosophers are not trained to precision, but probably I have many years work ahead of me, so I ought to accomplish a good deal. The work I have done since the big book was finished has some unity that I can't get hold of; if I could, I should probably see how to write a big book. But as yet I don't know what the central theme is. It is an odd blind impulse-I dread the slavery of another big book-it is really frightful-yet I would do anything to get hold of the central idea that would make a big book. (#6I5)

!vii

It is now clearer what he meant when he called Problems "popular": it could be understood not only by his fellow philosophers but also by intelligent shop-assistants. Technical philosophy, on the other hand, could only be understood by those with an excellent working grasp of mathematical logic, which at the time he was writing excluded nearly every philosopher and all laymen. But his popularly written philosophy could give his readers a taste at least of his technical work:

As it turned out his next "big" book was Theory of Knowledge, which he did not finish. The next day he spoke of applying his method to epistemological problems:

I am surprised to find how much of my philosophy comes into the Shilling Shocker-of course in technical writing all sorts of points would have to be more gone into, and there would have to be controversy; but the essentials are all there. (#243, 28 Oct. I9II)

I am enjoying being back whole-heartedly in the philosophical world. The time of half forgetting it has made things take their proper places-I see where each thing fits, what is important and central, and what is more or less accidental. The whole scheme of things is vastly clearer in my mind than it was a year ago, and I feel I can go ahead to new problems with the feeling of a certain territory conquered-The sort of thing that interests me now is this: Some of our knowledge comes from sense, some comes otherwise; what comes otherwise is called "a priori". Most actual

Here we have another difference between the two kinds of philosophical writing, namely, that technical philosophy unlike popular philosophy involves controversy with other philosophers. In popular philosophical writing it is not necessary to refute those with whom you disagree, you can simply ignore them. A third important difference between the two types of philosophy concerns their scope. As we have seen, technical philosophy is limited to those problems which are amenable to logical analysis, and for Russell this meant epistemological problems. In a letter of I3 December I9II, written

lix

INTRODUCTION

INTRODUCTION

while Problems was going through the press, he gave Lady Ottoline a rather clear indication of this difference of scope:

several years. "Mysticism and Logic", to cite a popular piece, includes discussions of a number of traditional problems; and Theory of Knowledge, to mention a technical one, attempts an answer to "the one great question". Indeed, he never really deviated from this way of classifying his philosophical writings, and in the volume on his philosophy in The Library of Living Philosophers series he rebuked a critic who cited his popularly written articles and books as evidence of his settled philosophy (Schilpp 1944, 730-1). The letters he wrote to Lady Ottoline make it abundantly clear that in expanding his writing scope he was attempting to please her. He wanted to involve her in his work, if this was at all possible, and since she had no head for mathematical logic, he cast about for other ways of sharing his thoughts with her. She had an abiding interest in religion and regarded herself as a religious person; Russell, of course, was just the opposite. But he attempted to write about religious questions in a way pleasing to her. His essay "The Essence of Religion", the only published part of the aborted book which was to be called Prisons, is perhaps the best example of his attempt to write according to a formula for which he had no natural talent. It is not entirely successful. It seems unlikely that his agreement to write The Problems of Philosophy was undertaken just to please her, although that consideration may have weighed in his decision. What seems more likely to have tipped the scales was his desire, to which we have already drawn attention, to make the philosophical consequences of his work in mathematical logic more widely known. He was concerned that if he did not call attention to them they would never become known, since readers of Principia Mathematica were never likely to be numerous. His decision to return to teaching was similarly motivated. And there was another factor operating. The success of his work on Principia Mathematica had established his reputation, and he knew it:

Iviii

I will try to write out what I think about philosophy; it will help clear up my own ideas. All the historic problems of philosophy seem to me either insoluble or soluble by methods which are not philosophical, but mathematical or scientific. The last word of philosophy on all of them seems to me to be that a priori any of the alternatives is possible; thus e.g. as to God: traditional philosophy proved him: I think some forms of God impossible, some possible, none necessary. As to immortality: philosophy can only say it may be true or it may be false; any more definite answer would have to come from psychical research. As to whether nothing exists except mind: philosophy, it seems to me, can only say that all the arguments adduced on either side are fallacious, and that there is absolutely no evidence either way. And I should say the same of optimism and pessimism. Except as a stimulus to the imagination, almost the only use of philosophy, I should say, is to combat errors induced by science and religion. Religion says all things work together for good; philosophy says this belief is groundless. Science leads people to think there is no absolute good and bad, but only evoked beliefs about good and bad, which are useful to gregarious animals in the struggle for existence; philosophy equally says this belief is groundless. All this is rather dismal. But as a stimulus to the imagination I think philosophy is important. But this use is not so much for the technical philosopher, but rather for the man who wants to see his own special pursuit connected with the cosmos; therefore it is popular rather than technical philosophy that fulfils this need. This is fundamentally why I think it is more useful to write popular than technical philosophy. There is one great question: Can human beings know anything, and if so what and how? This question is really the most essentially philosophical of all questions. But ultimately one has to come down to a sheer assertion that one does know this or thate.g. one's own existence-and then one can ask why one knows it, and whether anything else fulfils the same conditions. But what is important in this inquiry can, I think, be done quite popularly; the technical refinements add very little except controversy and long words. I was reinforced in this view by finding how much I could say on the question in the shilling shocker. (#286) We have here quite a good outline of much of his writing during the next

I don't know if it is too conceited, but I do feel myself in a way at home among the great philosophers. Spinoza and Leibniz on my mantelpiece seem like friends, I have conversations with them in which I explain how I am carrying on their work, and I can hardly resist the feeling that they hear and approve-sometimes it is all but a delusion, it grows so strong. It is one of the joys of work. An immense proportion of my work remains to be done. If I didn't think well of my work I should be almost ashamed of offering myself to you-for I don't think much of myself as a human being. (#185, 21 Nov. 1911) He had proved himself, and, in the tradition of his family, he felt obliged

Ix

INTRODUCTION

to society to use his talents in the broadest possible way. Popular philosophical writing, backed as it was by his growing reputation as a master of technical philosophy, was an obvious way to discharge this obligation. Lady Ottoline, then, was only the proximate cause of his decision; the remote, and deeper, cause was the fact that he was a Russell.

Acknowledgements

THIS VOLUME COULD not have been completed without the generous assistance of the University of Toronto, McMaster University and the Social Sciences and Humanities Research Council of Canada. The Council's award of a Major Editorial Grant, beginning in July I98o, has been of crucial significance for our plan to publish the Collected Papers. Without this grant the Russell Editorial Project could neither have obtained the extensive time freed from other duties needed for the editors, nor assembled the staff and equipment to produce this volume and the many volumes that are planned for the future. In addition the editor received an individual grant from s.s.H.R.C. for I986-87 and I989-92. Bernd Frohmann is the textual editor of this volume. The Textual Notes, the Bibliographical Index and the General Index were all compiled by him. In addition, he has supervised the final construction of the volume, a task which requires nearly unlimited patience and a practiced eye, for much can (and usually does) go wrong at that stage. Finally, I would especially like to thank him for calling to my attention certain matters in the Headnotes and Annotations requiring emendation or correction as he made his way through the volume. I have also been very well served by Michael Feindel, who gave the text a final, meticulous scrutiny. Two others who assisted on the textual side are Paul Gibson, for a preliminary draft of the Bibliographical Index, and Daniel Vaillancourt, for assistance with the Textual Notes for the papers in French. The Chronology is very largely the work of my assistant for the last three summers, Loren King. Sheila Turcon examined his work and made suggestions for its improvement. Loren King also rendered important assistance on the preparation of the General Index. To all of these people I give my heartfelt thanks; the volume simply could not have been completed without their help. On the research side I must thank Kenneth Blackwell, the archivist of the Bertrand Russell Archives, who has again helped me in important ways, most notably in determining the contents of the volume. I also greatly appreciate the work of Andre Vellino, for translations from the French, and Graeme Hunter, for translations from the German. Several students, both graduate and undergraduate, have assisted me during the gestation of the volume, all of who deserve my warmest thanks. In ad!xi

!xii

ACKNOWLEDGEMENTS

dition to his translations, Andre Vellino, helped with some of the more technical annotations; Peter Turney and Geoffrey Forguson turned in beautiful work on the annotations; Loren King joined me when the bulk of the annotations were written, but he assisted in tracking down some of the last holdouts and he also found documents in the Russell Archives which resulted in fuller Headnotes. Albert Lewis deserves thanks for his careful scrutiny of the mathematical text. Gerard Field prepared the diagrams for Paper 3 and Appendix IV. It is a pleasure to thank members of the Advisory Editorial Board for their assistance to the Project. John Passmore, who is also the General Editor of the Collected Papers, was especially helpful in the production of this volume. His wide learning is the source for a number of the more recondite annotations. I wish also to express my appreciation to the various persons in administrative positions in the University of Toronto and McMaster University whose attention to their duties made my work much easier. Camera-ready copy for Volume 6 of the Collected Papers was produced by Arlene Duncan. She deserves special thanks, not only for her expertise in the technical aspects of producing the actual volume, but also for her keen attention not only to consistency of design, but also to the text itself. McMaster's Printing Services, managed by Don Henwood, rendered important services to the production of the volume. For permission to quote from material in their copyright, I wish to thank T.E. Moore (for the letters ofG.E. Moore); the University of California, Los Angeles (for the letters of F. C. S. Schiller). Acknowledgement is made to the following publishers and journals which previously published some of the papers printed here, and to individuals and institutions for permission to quote from material in their possession: the Aristotelian Society (Papers 16, 18); Cambridge University Press (Paper 1 and Appendix vn); The Harry Ransom Humanities Research Center, The University of Texas at Austin (for Russell's letters to Lady Ottoline Morrell); Mathematical Gazette (Paper 7); The Monist, La Salle, Illinois 61301 (Papers 2, 7, 28); New Statesman and Society (Papers 9, 20, 22, 24, 32-36); Societe Frarn;aise de Philosophie (Appendix v); Societe Mathematique de France (Appendix IV); the University of California, Los Angeles (for Russell's letters to F. C. S. Schiller).

Chronology: Russell's Life and Writings, 1909-13

Tms VOL u ME DOES not contain all of Russell's writings for the period 190913. Readers are directed to Collected Papers, Vol. 12 for his more popular writings, and to Paper 9 of Collected Papers, Vol. 8 for one of a pair of related book reviews. In addition, readers interested in ·Russell's work on the theory of knowledge should consult the Volume 7 chronology. Life

Writings

Living with Alys at Bagley Wood, Oxford. c. 16-27 Feb. 1909 Apr. 1909 15-22 July 1909 30 July-15 Aug. 1909

21 written 6, 21 published. At St. Ives, Cornwall, with the Whiteheads. Takes walking tour of the Tyrol and the Italian Dolmites with C.P. Sanger.

c.Oct. 1909

18 Oct. 1909

6 Dec. 1909

A. N. Whitehead begins negotiations, with the Cambridge University Press, for publication of Principia Mathematica. Announces decision to join the People's Suffrage Federation. Participates in discussion following G. E. Moore's paper at the Aristotelian Society.

Feb.-May

19 published.

l9IO

May

l9IO

1 published (in French). !xiii

!xiv 27 May l9IO

July l9IO 29 July-c.27 Aug. 1910

3 Sept. 19!0 Oct. l9IO

3 Oct. l9IO

Receives word of his appointment to Trinity for a five-year lectureship in logic and the principies of mathematics.

Begins Paris lectures.

23 Mar. 1911

22 published. Tries to sell Bagley Wood; rents Van Bridge Cottage, Fernhurst, West Sussex for Alys. Moves into Nevile's Court, Trinity College, to take up his lectureship.

In lodgings at Cambridge with Alys. Elected president of the Aristotelian Society for l9II-l2. Informed by Whitehead of a fundamental error in Volume II of Principia Mathematica. Accepts an invitation to give three lectures in Paris, and begins affair with Lady Ottoline Morrell in London.

26 Mar. 1911 30 Mar. l9II May l9II 17 May l9II

Agrees to write a layman's introduction to philosophy for the Home University Library series, to be titled The Problems of Philosophy. 12, 19 published in Philosophical Essays. 20 published. Principia Mathematica, Vol. l published.

12 Nov. 1901 Dec. l9IO

Mar. l9II

13 Mar. l9II 22 Mar. l9II 7, 8, 31 published.

Nov. l9IO

20 Jan. l9II

6 Mar. l9II

Takes a walking tour, with C.P. Sanger, of Carinthia, Styria, and north-east Italy, ending in Venice.

IO Oct. l9IO

Jan.-Mar. l9II 3 July l9II

CHRONOLOGY

CHRONOLOGY

14 published.

29 May 1911

l-13 July l9II

24 July-23 Aug. l9II Aug. l9II c.20 Aug. l9II 23 Aug.-8 Sept. 191 l 9-13 Sept. l9II 16-19 Sept. l9II 30 Sept. l9II Oct. l9II

15 read to the Aristotelian Society. 13 published. 2 read to an audience at L'Ecole des hautes etudes sociales in Paris. 3 read to La Societe mathematique de France, also in Paris. 14 read to La Societe frarn;ais de Philosophie, as his final Paris lecture.

At Fernhurst, following Paris lectures. Returns to Cambridge. 2 published. Invited by Professor R.B. Perry to deliver the Lowell Lectures in Boston during the spring of 1914, while serving as a visiting Professor of Philosophy at Harvard. Terms of marital separation from Alys agreed upon. At Upper Wyche, near Great Malvern, Worcester, with North Whitehead. In lodgings at Ipsden, Oxfordshire, near Lady Ottoline. 3 published. Completes The Problems of Philosophy. In lodgings at Checkendon, Oxfordshire. Vacations with Lady Ottoline in Marienbad. In lodgings at Ipsden, Oxfordshire. Returns to Cambridge. Resigns from duties in the Cambridge branch of

!xv

lxvi

IO I8

28

30

Oct. Oct. Oct. Oct.

Nov.

I9II I9II

I9II

the People's Suffrage Federation. Cambridge term begins. First meets Ludwig Wittgenstein. In London to dine with Henri Bergson.

I9II

Nov. I9II Dec. I9II Dec. I9II

Leases fiat in Russell Chambers on Bury St., London.

I6

8

I9I2

I9I2-I3

c.Jan.

I6 I8 24

I9I2

Jan. Jan. Jan.

Feb.

I9I2

I9I2

I9I2

32

II

28

Feb. I9I2 Mar. I9I2 I5 Mar. I9I2 23 Mar. I9I2 mid-Marchearly April

published.

I9 24

I3 I7

I8 26

lla-lli probably written.

I9I2

Apr. Apr.

I9I2

May May

I9I2

May May

June

5-8

9 published. Takes a walking trip along the South coast, followed by a motor vacation with his brother, Frank, in the West Country. Returns to Cambridge in time for the new term. Cambridge term begins.

Principia Mathematica, Vol. published.

II

Plans made for a paper on matter

I9I2

(10). 10 completed. 10 read to the Philosophical Soci-

I9I2

ety of the University College of Cardiff. 24 published.

I9I2

I9I2

July

I9I2

Aug. I9I2 22-8 Aug.

His fortieth birthday. Wittgenstein disagrees with much of 10. In Lausanne and Paris with Lady Ottoline. Cambridge term ends.

28

published.

33

published.

Elected as president of the Aristotelian Society for a second year. At Fontainebleau with Lucy Donnelly.

I7

I9I2

published. read to the Heretics.

Cambridge term ends.

I9I2

8 June I9I2 July I9I2 I July I9I2

published. The Problems of Philosophy published.

Encouraged by George Santayana to accept Harvard's invitation.

IO

Apr.

27

Asked by E.W. Hobson to organize and to preside over the sessions of Section IV of the Fifth International Congress of Mathematicians at Cambridge. Wittgenstein gains admission to Trinity.

Feb.

I9I2

23

Cambridge term ends. Spends Christmas with the Whiteheads in Lockeridge, Wiltshire. Philosophical interests become concentrated upon the theory of knowledge and the problem of matter. Develops an interest in physics, particularly relativity theory. Alternately at his Cambridge rooms or his London fiat. Encourages Wittgenstein to pursue studies in philosophy. Cambridge term begins.

I9I2 I9I2

8

16 read to the Aristotelian Society as his first presidential address.

I9II

!xvii

CHRONOLOGY

CHRONOLOGY

Addresses, and presides over, the sessions of Section IV (Philosophical, Historical and Didactical Questions) of the Fifth Annual International Congress of Mathematicians at Cambridge.

!xviii

CHRONOLOGY

Sept. 1912 7 Sept. 1912 Oct. 1912 25 Oct. 1912

5 published. 34 published. 25 published, 4 probably written. Reads a revised version of 10 to the Cambridge Moral Sciences Club. 18 read to the Aristotelian Society as his second presidential address.

4 Nov. 1912

16 Nov. 1912

9-14 Dec. 1912 Dec. 1912

5 Dec. 1912 30 Dec. 1912c. 14 Jan. 1913 Jan. 1913 14 Jan. 1913 6 Feb. 1913

22 Feb. 1913 Mar. 1913

8 Mar. 1913 Apr. 1913 17 Apr. 1913 18 Apr. 1913 26 Apr. 1913 7 May 1913

May 1913

CHRONOLOGY

Agrees to a three-month U.S.A. visit, beginning March 1914, to deliver the Lowell Lectures and to lecture at Harvard. Takes a walking trip in the West Country. Spends Christmas with the Whiteheads in Lockeridge, Wiltshire. At the Beetle and Wedge, Moulsford, Berkshire.

6 June 1913

13 June 1913 14 June 1913 19 June 1913

4-c.28 July 1913

26 published. Works on the problem of matter.

19 July 1913

17 published. Cambridge term begins. Refuses an offer of a permanent position at Harvard.

22 July 1913

35 published. Takes a walking trip ending in Exmoor, Somerset. Cambridge term ends.

l-c.31 Aug. 1913 Principia Mathematica, Vol. published. 29 published.

III

Cambridge term begins. 33 published. Begins Chap. 1 of the analytic section of Theory of Knowledge. Wittgenstein severely criticizes Theory of

Knowledge, particularly its theory of judgement. President Lowell of Harvard finds Russell's topic for the Lowell Lectures, "the place of good and evil in the universe," inappropriate due to its implicit religious character. Cambridge term ends.

36 published. Remarks to Lady Ottoline that Wittgenstein's criticisms make Theory of Knowledge impossible. Vacations in Cornwall. Spends a fortnight in the Scilly Islands, then returns to London. Submits new proposal for the Lowell Lectures (the lectures will later be published as Our Know!edge of the External World). Wittgenstein apologizes for the effect of his criticisms upon Theory of Knowledge. Takes alpine walking tour with C.P. Sanger, beginning in Innsbruck and ending in Italy. Completes first draft of the Lowell Lectures.

25 Sept. 1913 26-9 Sept. 1913 10 Oct. 1913

!xix

Meets Norbert Wiener at Cambridge and plans his programme of work. Cambridge term begins.

Part

I

Logic and the Philosophy of Mathematics

1

The Theory of Logical Types [1910]

THIS PAPER w AS published for the first time, in French, in Revue de metaphysique et de morale, 18 (May 19rn): 263-301. It was published in English for the first time in the Introduction, Chaps. 2 and 3, of Principia Mathematica (19rn). There have been two reprints in more recent years: in French in Cahiers pour !'analyse, IO (1969), and in English in Essays in Analysis (1973). Since Russell had no role in these later reprints, their texts have no authority. The occasion for writing this paper was Henri Poincare's "La Logique de l'infini" (1909) in an earlier issue of the same journal. Poincare's paper is available in English in Mathematics and Science: Last Essays (1963). Russell's article is less a reply to Poincare than it is a fuller statement of his own position. Poincare's paper was but the latest in a series of articles which attacked the sort of mathematical logic that Russell was helping to develop. From the start Russell felt obliged to reply. Poincare sometimes ridiculed mathematical logic. In Science and Method (r9r4, 194) he states that "logistic is no longer barren, it engenders antinomies", an unmistakable reference to Russell's paradox. Henri Jules Poincare (1854-1912) made important discoveries in many branches of mathematics and science. Greatly gifted as a writer of clear and comprehensible prose, his popular articles on science and mathematics, of which "The Logic of Infinity" is one, enjoyed a wide audience. Russell greatly respected both his intellectual and his literary powers. For evidence in support of this claim, see the Headnote to his "Preface" to Poincare's Science and Method in Volume 8 (1986) of The Collected Papers of Bertrand Russell. The copy-text is the manuscript (RA 220.0112rn-F1-F2), which was emended by A. N. Whitehead (see the Textual Notes). It has been collated with both the first French and the English editions; the results are reported in the Textual Notes.

3

1 THE THEORY OF LOGICAL TYPES

M

POINCARE's INTERESTING article in a recent number of this Review' has explained, with his usual lucidity, what are his . reasons for not wholly accepting any of the theories recently put forward to explain the paradoxes of logic. As one of the authors concerned, I gratefully recognize that his article is not polemical in tone, and I freely admit that, on the points as to which he complains of my having given insufficient explanations, the article to which he was referring is no doubt too concise. As this article appeared in a mathematical journal, I was unwilling to devote more space to philosophical interpretations than rn appeared absolutely indispensable. From M. Poincare's criticisms, however, I see that certain obscurities resulted from the endeavour to be brief. These obscurities I shall try to remove in the following pages, of which the purpose is explanatory rather than controversial. I.

THE NATURE OF PROPOSITIONAL FUNCTIONS

It is agreed that the paradoxes to be avoided all result from a certain kind of vicious circle. The vicious circles in question all arise from supposing that a collection of objects may contain members which can only be defined by means of the collection as a whole. Thus, for example, the collection of propositions will be supposed to contain a proposition stating that 20 "all propositions are either true or false". It would seem, however, that such a statement could not be legitimate unless "all propositions" referred to some already definite collection, which it cannot do if new propositions are created by statements about "all propositions". We shall, therefore, have to say that statements about "all propositions" are meaningless. More generally, given any set of objects such that, if we suppose the set to have a total, it will contain members which presuppose this total, then such a set cannot have a total. By saying that a set has "no total", we mean, primarily, that no significant statement can be made about "all its members". Propositions, as the above illustration shows, must be a set 30 having no total. The same is true, as we shall shortly see, of propositional functions, even when these are restricted to such as can significantly have as argument a given object a. In such cases, it is necessary to break up our set into smaller sets, each of which is capable of a total. This is what the theory of types aims at effecting. The paradoxes of symbolic logic concern various sorts of objects: propositions, classes, cardinal and ordinal numbers, etc. By means of the theory (to be explained below) which reduces statements that are verbally concerned with classes and relations to statements that are concerned with propositional functions, the paradoxes are reduced to such as concern

propos1t10ns and propositional functions. The paradoxes that concern propositions are such as the "Epimenides", and are only indirectly relevant to mathematics. The paradoxes that more nearly concern the mathematician are all concerned with propositional functions. By a "propositional function" I mean something which contains a variable x, and expresses a proposition as soon as a value is assigned to x. That is to say, it differs from a proposition solely by the fact that it is ambiguous: it contains a variable of which the value is unassigned. It agrees with the ordinary functions of mathematics in the fact of containing an unassigned variable: where it differs is in the fact that the values of the function are propositions. Thus e.g. "x is a man" or "sin x = l" is a propositional function. We shall find that it is possible to incur a vicious-circle fallacy at the very outset, by admitting as possible arguments to a propositional function terms which presuppose the function. This form of the fallacy is very instructive, and its avoidance leads, as we shall see, to the hierarchy of types. The question as to the nature of a function 2 is by no means an easy one. It would seem, however, that the essential characteristic of a function is ambiguity. Take, for example, the law of identity in the form "A is A", which is the form in which it is usually enunciated. It is plain that, regarded psychologically, we have here a single judgment. But what are we to say of the object of the judgment? We are not judging that Socrates is Socrates, nor that Plato is Plato, nor any other of the definite judgments that are instances of the law of identity. Yet each of these judgments is, in a sense, within the scope of our judgment. We are in fact judging an ambiguous instance of the propositional function "A is A". We appear to have a single thought which does not have a definite object, but has as its object an undetermined one of the values of the function "A is A". It is this kind of ambiguity that constitutes the essence of a function. When we speak of "Ix). Ix). x will be defined as meaning "(x).,.., x". Thus, in the traditional language of formal logic, the negation of a universal affirmative is to be defined as the particular negative, and the negation of the particular affirmative is to be defined as the universal negative. Thus the meaning of negation for such propositions is different from the meaning of negation for elementary propositions. 40

14

IO

20

1 THE THEORY OF LOGICAL TYPES

An analogous explanation will apply to disjunction. Consider the statement "either p, or x always". We will denote the disjunction of two propositions p, q by "pvq". Then our statement is "p. v. (x). x". We will suppose that p is an elementary proposition, and that x is always an elementary proposition. We take the disjunction of two elementary propositions as a primitive idea, and we wish to define the disjunction "p. v. (x). x". This may be defined as "(x). pvx", i.e. "either pis true, or x is always true" is to mean "'p or x' is always true". Similarly we will define "p.v.(tI.x).x" as meaning "(tI.x).pV!x" is a function which contains no apparent variables, but contains the two real variables ¢!z and x. (It should be observed that when cf> is assigned, we may obtain a function whose values do involve individuals as apparent variables, for example if !x is (y). lf;(x, y). But so long as is variable, ¢!x contains no apparent variables.) Again, if a is a definite individual, ¢!a is a function of the one variable ¢!z. If a and b are definite individuals, "!a implies lf;!b" is a function of the two variables cf>!z, lf;!z, and so on. We are thus led to a whole set of new matrices,

IO

f(cf>!z), g(cf>!z, lf;!z), F(!z, x), and so on.

These matrices contain individuals and first-order functions as arguments, but (like all matrices) they contain no apparent variables. Any such matrix, if it contains more than one variable, gives rise to new functions of one variable by turning all its arguments except one into apparent variables. Thus we obtain the functions

20

(¢) .g(!z, lf;!z), which is a function of lf;!z (x).F(cf>!z, x), which is a function of cf>!z () .F(cf>!z, x), which is a function of x.

We will give the name of second-order matrices to such matrices as have first-order functions among their arguments, and have no arguments except first-order functions and individuals. (It is not necessary that they should have individuals among their arguments.) We will give the name of second-order functions to such as either are second-order matrices or are derived from such matrices by turning some of the arguments into apparent variables. It will be seen that either an individual or a first-order function may appear as argument to a second-order function. Secondorder functions are such as contain variables which are first-order functions, but contain no other variables except (possibly) individuals. We now have various new classes of functions at our command. In the first place, we have second-order functions which have one argument

30

20

ro

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

1 THE THEORY OF LOGICAL TYPES

which is a first-order function. We will denote a variable function of this kind by the notation/!( • ,...., x1'J(y, z), y1J(x, z). z1J(y, w). :::>. z1J(x, w), ....

We can then consider any relation iJ having these properties and we have a logic. However, a reduction to logic can only be obtained if we cease to demand with too much insistence that there exist objects which verify the hypotheses whose consequences we are considering. It will happen occasionally that such objects can be constructed a priori; for example, we can construct a priori a class having a finite number of arbitrary terms. (In fact, this construction is a priori only if we admit as a priori the axiom that there exists at least one object, or some equivalent axiom.) But most existence theorems (which, by the way, are not necessary for the truth of other theorems, but only for their importance) require data which are not purely logical. In pure mathematics, two existence axioms provide just about all the existence theorems one might want. These two axioms are: I. 2.

The axiom of infinity; The multiplicative axiom, otherwise known as Zermelo's axiom.

These two axioms cannot be proved using logic alone and, in my opinion, they are not intuitively obvious. However, before going on to examine the reasons one might have for affirming or denying their truth, I would first like to explain their nature and their consequences.

43

20

30

44

3 ON THE AXIOMS OF THE INFINITE AND TRANSFINITE

The axiom of infinity is formulated as follows: If n is any finite cardinal number, there is a set consisting of n individuals. Here the word individual contrasts with class, function, proposition, etc. In other words, an individual is a being in the actual world, as opposed to the beings in the logical world. To expand on this notion, I would have to explain the theory of logical types, which I do not wish to do; the lacunae which follow from this omission will no doubt be noticed by my audience. Let i,, be the class of all sets of n individuals. Then, given the axiom of infinity, the sequence . . . .

the cardinal number of terms in a sequence of type w 1 is greater than X0 ; following Cantor we will call it X1 • Thus we obtain a sequence of ordinal numbers and a sequence of cardinal numbers. The latter is X0 , XI> X2 , ••• , X,,, .... Assuming the axiom of infinity, the existence of all these numbers can be demonstrated. But, in my opinion, the existence of Xw, or any cardinal number greater than all the numbers X0 , Xi> ... , X,,, ... cannot be proved without another axiom. By assuming the axiom of infinity, we have established the existence of two progressions of cardinal numbers, namely

z0 ,

IO

l1'

z2 ,

••• ,

z,, ...

forms a progression, i.e. a sequence whose ordinal number is w (according to Cantor's notation). Thus, the axiom of infinity is a necessary and sufficient condition for the existence of progressions, i.e. for the existence of sequences of the form x1' x2 ,

20

45

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

••• ,

x,,, ... , ad inf.

This axiom is therefore a necessary and sufficient condition for the existence of the smallest transfinite ordinal number w, and of the smallest transfinite cardinal l'\0 • Given the existence of l'\ 0 , it follows that 2'''o exists, since this is the number of classes contained in a class having l'\ 0 terms. It also follows that

If we do not assume the multiplicative axiom, there is no proof that the terms of the first progression are either greater or smaller than the terms of the second progression (besides the term X0 ). Cantor had hoped to prove that 2'''o = X1' but neither he nor anyone else has succeeded in proving such an equation. In another order of ideas, the existence of dense series of different types can be proved from the axiom of infinity. For example, arrange the finite integers according to the following rule. Express the integers according to the dyadic scale so that they have the form

20

1

22"'0 , 22 2 ''0 and so on, each exist. It is easy to show that all these numbers are different, and that they form an ordered sequence of increasing magnitude. This then is a first infinite sequence of transfinite cardinals. Given the existence of w, we can also assert the existence of the other ordinal numbers constructed by changing the order of the terms in a progression. For example, we have the sequences

30

1, 2, 2, 2,

2, 3, 4, 4,

3, ... , n, ... , 4, ... , n+ 1, ... , 1, 5, ... , n+2, ... ' 1, 3, 6, ... , 2n, ... ' 1, 3, 5, ... , 2n-l, ....

Thus we obtain a certain set of ordinal numbers, namely the set of ordinal numbers of sequences whose terms are the finite integers. Cantor has shown that one can arrange these ordinal numbers in order of increasing magnitude. These numbers themselves form therefore a well-ordered sequence whose ordinal number is greater than any number occurring in the sequence. We shall call this ordinal number w 1 • It is easy to show that

i.e. the sum of the numbers 2" for every n belonging to a. Then, given "'

f3

the two numbers 2:2" and 2:2", put the first before the second if the smallest number which belongs to one of the two classes a or f3 without belonging to both classes, belongs to a. This yields a series of numbers where the numbers terminating with the most zeros come later; among the numbers terminating with the same number of zeros those which have the least number of ones before the zeros come later; among the numbers terminating with the same number of zeros and the same number of ones before the zeros, those having the most zeros before the ones will come later, and so on. This series can be constructed as follows: put 1 in the middle, 10 to the right, 11 to the left. We call lacunae not only the spaces between two numerals, but also the spaces to the right and to the left of all the numerals. Then fill the lacunae, starting from the right, first adding the numeral 0 without changing the order, then the numeral 1, omitting the numerals already obtained. In other words, if we abandon the dyadic scale, put 1 on the first row; below 1 put 2 and 3 in the right and left lacunae. Below the second row, double

30

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

3 ON THE AXIOMS OF THE INFINITE AND TRANSFINITE

the second row to fill the lacunae to the right of 1, and add 1 to the numbers belonging to the third row already obtained to fill the lacunae to the left of 1.

ample, that the House of Commons were composed exclusively of inhabitants of the districts they represent, then the House is a selection of the districts considered as classes of electors. Now, when the number of members belonging to xis finite, the arbitrary choice of representatives for the sets can always be made, so that there always are selections. It is evident that, in this case, the number of possible selections is the product of the number of sets belonging to X· The class of selections can always be used to define this product and hence we have a definition of product which extends to the case where the number of factors is infinite. It is for this reason that the theory of selections is intimately linked to the theory of multiplication. To begin with, we must give some precision to the idea of selection. When the number of members of x can be infinite, it is impossible to create a selection by arbitrary choices. It is therefore necessary to have a rule according to which the choices are made. (If, for example, sets belonging to x are sequences, each of which has a first term, the first term can always be the chosen one.) Now, in logic a rule of choice is represented by a one-one relation R such that, when a£ x, there is one and only one x which has the relation R to a, and this x is a member of a. That is to say, Risa one-one relation whose converse domain is x and which implies £,where£ is the relation which one term has to a class to which it belongs. A relation such as R will be called a selective relation of x which we will write as E'L'>X· In symbols,

46

2

3

15 31

IO

20

30

40

13

11

23 27

6

5

7

19

29 21 25 17

30

4

12

10

14

9

22 26

18

28

8

20 24 16

To obtain the nth row, start with zn- i from the extreme right and add zn 2 tO it; then add zn- 3 tO the numbers already Obtained While preserving the order; then add zn 4 to the numbers already obtained while preserving the order until you get to the odd numbers which fill the left side of the row. Continuing with this method obviously yields a dense sequence composed of finite integers. Therefore, assuming the axiom of infinity, there are dense sequences having l-\ 0 terms. The segments of such a sequence are continuous in Cantor's sense, that is to say, a sequence of the real number type. Therefore, assuming the axiom of infinity, we have proved the existence of the continuum. It follows that all kinds of Euclidean and non-Euclidean spaces can be proved to exist (in the mathematical sense). According to what we have just said, it is evident that the axiom of infinity alone suffices to prove most of the existence theorems needed in arithmetic. However, there is another existence axiom, namely the multiplicative axiom, which would be very useful if we could know it was true. The usefulness of this axiom concerns the theory of the transfinite rather than ordinary mathematics, whereas the axiom of infinity is necessary for the greater part of ordinary mathematics, the infinitesimal calculus for example. The multiplicative axiom has many equivalent forms. Zermelo, the first to have formulated it explicitly, gave it a form which may be expressed as follows: Given any set a, let x be the class of all non-empty sets belonging to a (including a itself). Then there exists a function f such that, if f3 is a member of x, f/3 is a member of {3. In other words, there exists a rule whereby one can choose a representative term from any continuous nonempty class in a. To understand this axiom and its usefulness for multiplication, mention must be made of the theory of selections. Let x be a class of mutually exclusive sets, then we call a selection from x, a class containing one and only one term from every set which is a member of X· One could call this term the representative of the set to which it belongs. Suppose, for ex-

~

E'L'>X = o~cls)n G'xnRl'E.

Now this notion can be generalized further by putting P'L">X =

(l~Cls)n

~

G'x· :J .Rl'P

(~j•~R~y, (~)-~R~-~y,

(3

Of.

47

ro

20

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

3 ON THE AXIOMS OF THE INFINITE AND TRANSFINITE

If we have RE P't:.x, R will resemble the relation in the diagram. The general notion is useful, but for the moment we will be chiefly concerned with c:'t:.X· Notice first of all that the class of selections may define the arithmetical product only when the members of x are mutually exclusive, whereas E't:.X is not subject to this restriction. I write N c' ex for the cardinal number of ex, where a is a class (a set); I write IINc' x for the product of the members of x, where xis a class of classes. We can therefore put

numbers. This proposition identifies the two definitions of the infinite. We can say that a class is infinite when it contains a part which can be put in a one-one relation with the whole class. According to this definition a number is infinite when it is not increased by the addition of one. Or else we can call any number which obeys complete induction starting from zero a finite whole number. That is to say, call a recursive property any property which belongs to n + 1, so long as it belongs to n, and call an inductive property any recursive property which belongs to zero. Then call an inductive number any number which possesses all inductive properties, i.e. any number for which the proofs using complete induction are valid. It is easy to see that the inductive numbers are the same as the natural numbers 0, 1, 2, .. ., 100, .. ., 1000, .. ., and that there are some numbers (assuming the axiom of infinity) which are not inductive numbers, for example, the number of inductive numbers. The so-called principle of complete induction then becomes a definition, namely the definition of inductive numbers. We can say, therefore, that the infinite numbers are the non-inductive numbers. Cantor always supposes that the non-inductive numbers are the same as the numbers which do not increase by the addition of one, but to prove this we need the multiplicative axiom, not, it is true, in all its generality, but only for products of X0 factors. 3. We define the product of two factors in the following manner: let ex and f3 be any classes, then the product of ex and f3 is the class of pairs (x, y) where x E a and y E f3 (to distinguish x, y from y, x). Most of the known properties of the product of these two factors can be proved without the multiplicative axiom. However, if we want to establish the relationship between addition and multiplication, we need the axiom, i.e. we need it to prove that the sum of µ, classes, each containing v terms, has µ, x v terms. The same can be said of the relationship between multiplication and exponentiation. 4. It follows from what we have just said that without the multiplicative axiom one cannot prove that the sum of X0 classes, each containing X0 members, has X0 x X0 members. We know that X0 x X0 = X0 , and it is usually inferred that the sum of X0 classes each having X0 members has X0 members. However, this inference is valid only if we assume the multiplicative axiom. 5. Using the methods of inference which we have just talked about, it is customary to prove that the limit of a progression of ordinal numbers of the second kind (i.e. those formed by sequences whose cardinal number is X0 ) is itself of the second kind. This also is valid only if we assume the multiplicative axiom. Thus a large part of the theory of transfinite ordinal numbers becomes questionable. The same can be said of the theorem that in any sequence, a term cannot be the limit of both a sequence of the type wand a sequence of the type w 1 • It follows that nearly all of the beautiful

48

IJNc'x IO

20

=

Nc\:'t:.X

Df.

Then when x is a finite class, even if the members of x are infinite classes, HNc'x will have the known properties of arithmetical products. But if x is an infinite class, we do not know that the class E't:.X is nonempty unless we assume the multiplicative axiom. It could be, then, that an arithmetical product is zero when none of its factors is zero. If we assume the multiplicative axiom, this becomes impossible; in fact, the proposition, the cardinal product can only be zero if one of its factors is zero, is equivalent to the multiplicative axiom. The multiplicative axiom is formulated as follows: Let x be a set of nonempty sets; then there is at least one class µ, which has one and only one term belonging to each set which is a member of X· Zermelo's axiom is formulated as follows: Let a be any class, and let x be the class of all non-empty classes contained in a; then there is at least one selective relation of X· In symbols, (a). 3! E't:,.Clex'ex.

30

It is easy to prove that these two axioms are equivalent. 1 Therefore, according to Zermelo's theorem, they are both equivalent to the proposition: Any class can be well-ordered. They are also equivalent to the proposition that, given a set x of non-empty sets, there always exist selective relations of x; as well as to the proposition that, if P is any relation and xis a class contained in the converse domain of P, then P't:.X is non-empty. The propositions which can only be proved with the help of the multiplicative axiom are numerous. Here are a few of them: I. Of two different cardinal numbers, one must be greater than the other. 2. The cardinal numbers which are increased by the addition of one are the same as the inductive numbers, i.e. the numbers we call the natural I

See Principia Mathematica, §88.

49

IO

20

30

40

so

IO

20

30

51

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

3 ON THE AXIOMS OF THE INFINITE AND TRANSFINITE

work by Hausdorff, Untersuchungen iiber Ordnungstypen, depends on the multiplicative axiom. 6. The multiplicative axiom is necessary to prove that the sum of µ, sets, each having v terms, has the same number of terms as the sum of v sets each havingµ, terms, except in the case whereµ, and v are both finite. To take the simplest case, consider a progression of pairs

is no reason to believe that the axiom is true. Peano, 3 after having proved the independence of the axiom, devotes only the following remarks to the discussion of its truth: "Now must we believe that the proposition is true or that it is false? Our opinion does not matter" (p. I48). In the same article Peano says that the question of obviousness is a psychological question, which does not concern logic. However, logic depends on the axioms of logic, and these axioms are assumed because they are evident, at least the obviousness of either the axioms or their consequences is the sole cause of our accepting the axioms. We may therefore ask if the multiplicative axiom is true, despite the fact that it does not follow from the other axioms. It must be admitted that there is not much to be said on this question; but it is possible to present some remarks which raise doubts concerning the axiom. As for the purported obviousness of this axiom, it seems to me that the imagination always provides a finite number of classes even when we wish to speak of an infinite number. Now, in the case of finite numbers, an axiom is not necessary, since the possibility of selections is easily proved. Therefore the apparent obviousness of the axiom tends to disappear under the influence of reflection. Moreover, when we speak of an infinite class, it must be generated by means of a property which all and only the memhers of the class possess. It is therefore impossible to generate the class by enumeration. Thus the multiplicative axiom must assert that, for any set of classes, there is always some property possessed by one and only one term in each class belonging to the set. But this is not at all evident, in my opinion. I am thus forced to the conclusion that the axiom ceases to be obvious the moment its significance is understood. It must be said, also, that Zermelo's theorem, namely, that if the theorem is true, then every class can be well-ordered, is reason to believe that the axiom is false; for it is hardly believable that every class can be wellordered. Many clever mathematicians have tried to find a well-ordered sequence of real numbers, but no one has succeeded in finding such a sequence. Such arguments do not have much weight, but they must be given a certain value. It is possible that a reductio ad absurdum showing the axiom is false will be found at a later date, but at the moment it seems to me to be merely doubtful. It might be true, but it lacks obviousness and its consequences are astonishing. Under these circumstances, it seems to me wise to abstain from using it, except in arguments which give the hope that it might yield an absurdity and thus resolve in the negative the question of the axiom's

and suppose it can be proved that the same terms can be arranged in a pair of progressions. Using the notation employed above, we do this by taking the sequence

But by adopting the notation xv, Yv for the vth pair, we have introduced the hypothesis that an order is given for each pair, so that within each pair each element can be identified as either the first or the second element. If we do not assume this hypothesis, it would be impossible to give a rule according to which a term in each pair can be selected simultaneously. Thus we would not know how to order the terms into two senes. What has just been said will become clearer with an example. There was once a millionaire who owned ~o pairs of boots. Can we prove that the number of boots he owned was even? Yes, for we could put all the left boots in one class and all the right boots in the other. But if this millionaire was eccentric enough to have pairs of similar boots so that he did not have a right boot and a left boot in every pair, then it would not be possible to divide the set of boots into two equal parts. It would then be impossible to show that the number of his boots was an even number, or that he had ~o boots, despite the fact that we have

Mathematicians used the multiplicative axiom unconsciously, not only in set theory but also in ordinary mathematics, until Zermelo explicitly formulated it as an axiom. 2 I myself had employed it unwittingly, but in I904 I discovered that it was an independent axiom. Many mathematicians, like Zermelo himself, claim that this axiom is as evident as the others and that it can be asserted without hesitation. Others say that there 2

"Beweis (a', R', b', C') results from (a, R, b, C) by substitution. Thus there is no issue this way. We must take form as a primitive. The values of a form are the complexes which have that form. A form is necessary if for every value of the variables in the form there is a value for the form and possible if the contradictory is not necessary. Then logic = the class of necessary and possible forms. Df. [Consider 3: !l] It won't do to take "being of the same form" as primitive, because that rules out non-existent complexes. It is possible for a form to be impossible, e.g. x ¥ x. A form is something·, though not a constituent of complexes having this 20 form. A form is not a mere symbol: a symbol composed entirely of variables symbolizes a form, but is not a form. A form is atomic if no part of it is a form. But what is "part" of a form? Are there simple and complex forms? A form is different from an expression containing both constituents and variables, e.g. "all men are mortal". This may be interpreted as a relation of man and mortal. But consider "x £a ..=> x.a • x £ mortal". This must be treated on the same principles as if it contained only variables. 30 In every proposition of logic, some expression containing only variables is said to be always true or sometimes true. The question "what is logic?" is the question what is meant by such propositions. In terms of complexes, x only exists when it is true. This makes it hard to explain (x). x. Consider (x): human(x). :::>. mortal(x). This seems to require that "Socrates is human. :::> • Socrates is mortal" should have a form with respect to Socrates, and that this form should be necessary. Then we state that, if x is any other entity, there is a complex having same relation to x as above has to Socrates. But this is again a general statement, and can't be the o 4 meaning of the original statement, or else we get an endless regress.

5

Reply to Koyre [1912]

Tms PAPER WAS published, in French, in Revue de metaphysique et de morale, 20 (Sept. 1912): 725-6, immediately after Koyre's own article. For the reader's convenience the French original and a translation of Koyre's "Sur !es Nombres de M. Russell", which was made by Andre Vellino, are printed in Appendix VIII. On l November l9II Russell wrote to Lady Ottoline Morrell: I enclose some French proofs of an article about me-the first two or three paragraphs express what always happens with me. When I first published my definition of number everybody found fault with it-then insensibly they all came to accept it. It is only the first few paragraphs that would interest you. It is an article I got in MS at Peppard-they didn't print it till they had asked me if I thought it worth printing, which was a compliment to my candour, as it is a criticism of me. (#238) A week earlier, on 26 October, he remarked that "I have written a French reply to a criticism-by what appears to be a Hungarian" (#234). Alexandre Koyre (1892-1964) was born in Russia but lived most of his life in France. He was educated at Giittingen where he studied under Husserl and Hilbert, and then in Paris with Bergson and others. At the time he wrote his article on Russell he was a student in Paris. From 1930 until his death he was Director of Studies at the Ecole Pratique des Hautes Etudes in Paris. During World War II he taught in the United States. In 1956 he joined the Institute for Advanced Study in Princeton, spending half of each year there. In addition to several books in the history of philosophy, he wrote important studies of Galileo and Kepler. The copy-text is an English translation of the French original; the translation was made by Andre Vellino.

57

5

M. Koyre deploys against the logical definition of cardinal numbers 1 would be definitive if we took no notice of l_ogical types., However, if we took no notice of logical types, the sets which M. Koyre does not want to call numbers would remain all the same and would be just as paradoxical as when they are given a name. It is not by refusing to talk about paradoxical sets that we solve the paradoxes. But, with the help of logical types, we can reply to the two claims by M. Koyre (p. 723). First, it is to be observed that there is an infinity of number twos or of any number. Let us define "two" to be "the class of all classes similar to the class whose only members are Socrates and Plato". These similar classes can only form a class if they are restricted to a single type. For example all the pairs of individuals provide one instance of the number two· all the pairs of classes of individuals provide a second instance of the nm~ber two; all the pairs of relations among individuals provide a third instance and so on. There is no class which can be made from the sum of two of these cl~sse~, since they are of different types. I have already explained the apphcat10n of the theory of types to numbers in my paper "Mathematical Logic As Based on the Theory of Types", 2 where I wrote (p. 257):

T

ro

HE ARGUMENTS

It is to be observed, however, that 0 and 1 and all the other cardinals, ... are ambiguous symbols, ... and have as many meanings as there are types. To begin with 0: the meaning of Odepends upon that of A, and the meaning of A is different according to the type of which it is the null-class. Thus there are as many O's as there are types; and the same applies to all the other cardinals. Nevertheless, if two classes a, {3, are of different types, we can speak of them as having the same cardinal, or of one as having a greater cardinal than the other, because a one-one relation may hold between the members of a and the members of {3, even when a and f3 are of different types.3

20

30

It follows from these remarks that a number does not contain itself as a s~co?~-order element; for example, call l(I) the number 1 which applies to mdividuals; then the number 1 to which l(I) belongs as a second-order element will not be l(I), i.e. it will not be a number which can apply to This definition is due_ rn Frege, not to me: see Die Gmndlagen der Arithmetik, Breslau, 1884, §68 (p. 79); Principles of Mathematics, p. 519. 2 American Journal of Mathematics, Vol. xxx, No. 3. 3 The development of arithmetic from this point of view can be found in the second volume of Principia Mathematica. I

58

REPLY TO M. KOY RE

59

classes of individuals; this will be a number which can be applied only to classes of classes of individuals. As for the proof that each. number is equivalent to the class of all objects, one would have to determine the type of the number and the type of the objects. It might then be found that the number (as a class) has the same number of objects as the class of all the objects, or that it has a smaller number, or that it has a greater number. That depends on the determination we give to the symbols of ambiguous type. M. Koyre's two arguments do not, therefore, succeed in destroying ~he logical definition of number, if we take into account the theory of logical types; and if we do not take types into account, the two ar_guments ~h?w that certain sets lead to paradoxes, as they do if we abstam from givmg these sets the name of "cardinal numbers". We can, and we must, therefore, accept all the detailed arguments of M. Koyre, but the conclusion which must be drawn from them is not that the definition of number must be abandoned, but that logic cannot do without the theory of types.

ro

6

Review of Reymond [1909]

Logique et mathematiques. Essai historique et critique sur le nombre injini. By

Arnold Reymond. Saint-Blaise: Foyer Solidariste, 1908. Pp. ix, 218. HIS BOOK CONSISTS, as its title indicates, of a critical history. It is divided into three parts, the first dealing with antiquity, the second with the discovery and philosophy of the infinitesimal calculus, the third with the modern treatment of infinity by means of symbolic logic. The first two parts, as is inevitable in so short a book, are necessarily somewhat sketchy, but in the main they tell what is of most importance. It is, however, somewhat surprising to find in the second part, where there is hardly a mention of any of the great philosophers except Leibniz, a ro whole chapter (one out of four) devoted to Renouvier and M. Evellin. M. Reymond justly criticizes Renouvier for supposing that the infinite could be replaced by the indefinite, since, as he points out, if a quantity can be subdivided indefinitely, it must contain an infinite number of parts. The views of M. Evellin, who regards space and time as finite and finitely divisible, are rejected, apparently on the ground that spatial and temporal continuity are data. M. Reymond has a theory that zero is not a number like other numbers, but a mere possibility, "une possibilite d'un certain genre, la possibilite numerique" (p. u8). His reasons for this view seem to rest on some con- 20 fusion. He says:

T

Tms REVIEW WAS published in Mind, 18 (April 1909): 299-301. N Arn~ld Reymond (1874-1958), a Swiss philosopher, taught in the University of euchatel from 1912 to 1925 when he accepted a professorship in the Uni · of ~ausanne. He retired in 1939, but continued to teach an introductory::~;~~ 1 UntI 944· I~ 1908 he was ~warded a doctorate from the University of Geneva; the book which Russell reviewed was his thesis for the degree. In addit' another bo k I ·. L p · · . 10n to h" h . o on og1c, es. rmc~pes de la logzque et la critique contemporaine (1932), w .IC. mcludes tw_o extensive discussions of Russell's work, he wrote widely on ~dehgl1_on and the history of science. His own philosophical position was a sort of 1 ea ism. The copy-text is that printed in Mind.

Some reflexions seem to justify this point of view. The expression 1 + 0 = 1 is very obscure from the logical point of view if zero enjoys exactly the same properties as other numbers. The unity in the first member of the equation is a given whole, which ought to be modified by the adjunction of zero, if this is a number like the others; this unity could not therefore be rigorously equal to that which appears in the second member (ibid.). From this passage (as well as from certain others, e.g., p. I72) it would appear that the author confuses arithmetical and logical addition, and does not distinguish between a number and a class having this number of members. "1 + O" means "the number of terms in a class consisting of two mutually exclusive parts, of which one has one term, while the other has none". The second part here is null, and therefore the first part is the whole of the supposed class. It is not easy to see any logical obscurity in this. The third part on "Logistic and Infinite Number", unfortunately shows a very inadequate knowledge of symbolic logic, and is full of elementary mistakes. The following are a few of these mistakes. On page I40, "p implies q" is said to be equivalent to "pis false, but q is true" (this may 60

61

30

40

62

IO

20

30

40

6 REVIEW OF REYMOND

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

be a misprint); on page 142, a wrong definition of "p or q" is given, according to which "p or q" would be meaningless unless p and q were both true or both false; on page 143, it is said that the product of two classes is equal to their sum, and on page 148, the product of a class of relations is spoken of where the sum is required. On pages 145 and 151, "logical constants" are spoken of where "constants" should occur. On page 146, the relation "greater or less" is said to be asymmetrical, whereas in fact it is symmetrical, since if x is greater or less than y, y is greater or less than x. Since M. Reymond proceeds to criticize various parts of the logical theory of number, it seems a pity that he has not first mastered it more thoroughly. M. Reymond's main objection to the logical theory of number is that the word "all" cannot be applied to classes of numbers in the same sense as to other classes, on the ground that numbers form a series of terms all differing inter se, and that the fact that they form a series is essential. Assuming the definition of "12" as a class of classes, namely the class of all dozens, he says (p. 153) that the terms of this class, qua dozens, are all alike, whereas the terms of the class "number" succeed each other in a definite order, and are all different one from another. It is very hard to see how this contention can be supposed to have any force. If the class of Apostles and the class of months are to be regarded as exactly alike qua consisting of twelve terms each, so are 1 and 2 exactly alike qua numbers. "The only property", he says, "common to all whole numbers is being obtained by a determinate law of succession" (ibid.). This is by no means true, since whole numbers have also, for example, the common property of being classes of similar classes. But even if it were true, it would not prevent numbers from forming a well-defined class. "The seed of Abraham" is defined by a certain law of succession, but that does not prevent it from being a well-defined class. A "law of succession" means a relation between any two consecutive terms; the field of this relation is then a class like any other, and the fact that the members (like those of any other class) differ inter se, offers no difficulty. M. Reymond criticizes at length the logical definitions of 0 and 1. His criticism of the definition of 0 (p. 17off.) seems to rest partly upon confusing a term with the class of which it is the only member, as when he says (p. 172): "The null-class is not a number; the arithmetical zero is therefore necessarily excluded from the null-class which therefore cannot serve to define it". The arithmetical zero is of course excluded from the null-class, since the null-class has no members, and therefore everything is excluded from it. But it is hard to see why, on this account, the nullclass should not serve to define zero; the definition of zero as the class whose only member is the null-class is untouched by M. Reymond's remark. The bulk of his argument against the definition of zero appears to

63

est on the contention that no statement can be found which is false for r b Yiet M Reymond himself contends that to every number every num er. · h. h 1 t 1 can be added. Hence the statement "n is a numbe~ to w ic ~an~~r be added" will be false for every number, and so will the even s~mp . "l t be added to n" Hence it would seem that his cntstatement, canno · icism cannot be maintained. " . 1d f The conclusion reached by M. Reymond is as follows: A nomma .e . .t' n either of whole number, or of infinite number, by means of !og~cal m1 IO ' 1 .s impossible and that principally because the pnnc1ple constants a one, i ' · 1· d'tions of . duction must be assumed as an indefinable, imp ymg con i . of m . b 1 · · " ( 1 99) This conexistence foreign to the relations envisaged y ogisuc p. . . clusion cannot, in my opinion, be regar~ed as rendered m ~~~ :fe~:~ probable by the preceding arguments, owmg to the large.numb 1 understandings which they contain, and to the assun_ipu~n t at genera statements about all the members of a class are imp?ssible .if the memb~r~ differ inter se in a manner which gives rise to a senes, as is the case wit the finite integers.

IO

7

Review of Carus [ r 909 ]

The Foundations of Mathematics: A Contribution to the Philosophy of Geometry. By Paul Carns. Chicago: Open Court, 1908. Pp. iv, 141. HIS BOOK IS a more or less popular exposition of a philosophy of geometry which is, in its main outlines, derived from Kant. The main title, if uncorrected by the sub-title, would be somewhat misleading, since the foundations of arithmetic and analysis are not discussed, but only the foundations of geometry. The author begins by a brief account of the development of non-Euclidean Geometry, which is followed by much longer chapters "on the philosophical basis of mathematics" and on "mathematics and metageometry". The historical chapter, though it does not profess to give more than a sketch, might with advantage have been enlarged by some account of projective geometry and the projective treatment of metrics. Dr. Carns speaks always as though non-Euclidean straight lines were not really straight, but were merely called straight out of wilfulness. The projective treatment shows, better than the metrical, wherein the straight lines of non-Euclidean spaces agree with those of Euclid, and ought therefore not to be omitted even in a mere outline. It would seem also that Dr. Carns regards a three-dimensional non-Euclidean space as necessarily contained in a four-dimensional Euclidean space, for he asks "what Riemann would call that something which lies outside of his spherical space", apparently not realizing that spherical space does not require anything outside it. The author's philosophical theory of geometry may be briefly summarized as follows. Geometry, like logic and arithmetic, is a priori, but it is not a priori in the same degree as logic and arithmetic. There is the a priori of being and the a priori of doing, and geometry belongs to the latter: it is derived from the contemplation of motion, and can be constructed from the "principles of reasoning and the privilege of moving about". We know a priori what are the possibilities of motion; thus, although there is nothing logically impossible about the assumption of four dimensions, yet "as soon as we make an a priori construction of the scope of our mobility, we find out the incompatibility of the whole scheme". The a priori is identical with the purely formal, which originates in our minds by abstraction; it is applicable to the objective world because the materials of formal thought are abstracted from the objective world. Most of the arguments in the book lead one to expect that Euclid will be declared to be certainly alone valid as against non-Euclidean geometry, yet this is not the conclusion drawn by the author. He says:

T

~HIS R~VIEW w ~s pu~l~shed

in The Mathematical Gazette, 5 (June-July 1909)· o3-4. _ams repnnted It m The Monist, 20 (Jan. l9ro): 64-5, as a part of his Ion. repCly to ~t. In a letter to P. E. B. Jourdain of 9 February l9IO Russell commente~ on ams s reply: Yes, Cams is amusing. I agree cordially, however, on one point namely when he ' , sa:s 'I _am not sure that I have always rightly interpreted his humor ' I like his solemnly proving that when he talks mathematics he knows what he is talking about and whether what h . . . I . ' e is saymg is true. a~ surpnsed that he thought my review of him favourable! (GrattanGumness 1977, 128- 9) Russell, of cour:e, _alludes to his own famous statement, first published in "Recent ~ork on th~ Prmciples of Mathematics" (1901) and later reprinted in his 1918 as Mathematics and the Metaphysicians"· "Th h . h . . . · us mat emattcs may be defined as t e subiec~ m ~hich we never know what we are talking about, nor whether what we are saymg is true" (Russell 1918, 75). th P~l ·C:~s (1852-1919) was born and educated in Germany; he emigrated to e mte tates as a young man. One of Carus's early books greatly impressed E. C. Hegeler, who invited him to become the editor of The Op C fi d db en ourt, a quarterly ;.un. e y Hegeler. Cams married Hegeler's daughter and went on to found The onzst and The Open Court Publishing Company both ofwhi"ch proved. · th d · . ' important m . e spr~a of philosophical and scientific ideas in the United States ff philosophical position was a kind of monism: he believed that all th" . is own b h mgs were one ecause t ey were all dependent on laws and these so-called laws of nature wer~ the~_selves depen~ent upon one law, which he called God. Despite this firmly held po~n10n, h~ published much original philosophy by others including Ru II whic~ conflicted wit? it. So~etimes, as in this case, he pub{ished his own ;;e I ' t_o ~omts ~~de by his contnbutors with which he disagreed. In addition to p:b~ hshmg ong~nal books, he reprinted many philosophical and scientific classics some of which h~ caused to be translated into English for the first time. ' The copy-text is the first printed version.

The result of our investigation is quite conservative. It re-establishes the apriority of mathematical space, yet in doing so it justifies the method of metaphysicians in their constructions of the 64 65

10

20

30

40

66

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

several non-Euclidean systems .... The question is not, "Is real space that of Euclid or of Riemann, of Lobatchewsky or Bolyai?" for real space is simply the juxtaposition of things, while our geometries are ideal schemes, mental constructions of models for space measurement. The real question is, "Which system is the most convenient to determine the juxtaposition of things?" (p.

8

Review of Mannoury [I 9 Io]

I2I)

10

20

Yet a few pages later he says: "The theorem of parallels is only a side issue of the implications of the straight line" (p. I29). It is not clear how these statements are reconciled, for the earlier statement seems to imply that there is no "theorem" of parallels at all. A few of the author's assertions are somewhat misleading. For example, he states, as a fact not open to controversy, that Euclid's axiom or postulate of parallels originally occurred first in the proof of the twenty-ninth proposition, not being mentioned either among the axioms or among the postulates (p. 2). On the other hand, Stiickel and Engel (Theorie der Parallellinien, p. 4) say that, following Heiberg, they do not regard the postulate of parallels as a later addition, which would seem to show that Dr. Carus's opinion is at least open to question. Again he says (p. 84): "While in spherical space several shortest lines are possible, in pseudospherical space we can draw one shortest line only." As regards spherical space, the more exact statement is that in general only one shortest line can be drawn between two given points, but when the two points are antipodes, an infinite number of shortest lines can be drawn between them. The book concludes with an epilogue, in which the existence and attributes of the Deity are deduced from the nature of mathematical truth.

THIS REVIEW WAS published in Mind, 19 (July 1910): 438-9. . . f A Gerrit Mannoury (1867-1956) taught mathematic~ in the Umversity ~ mterdam from 1903 to 1937; he was elected Professor m 19.17. He was particularly :nterested in the relationship of mathematics, on the one side, to psychology, and, on the other, to society. The copy-text is that printed in Mind.

67

8 REVIEW OF MANNOURY

Methodologisches und Philosophisches zur Elementar Mathematik. By G. Mannoury. Haarlem: P. Visser, I909. Pp. viii, 279. HIS BOOK IS an attempt to deal with the philosophical questions raised by modern mathematics from a standpoint which is half Hegelian and half Kantian. On fundamental questions the author refers with approval to Mr. Bradley and Mr. Joachim; on the axioms of Geometry he agrees in the main with M. Poincare. His reading is fairly extensive, though not quite sufficiently so to prevent occasional errors. For example, he gives a proof of the equivalence of the two definitions of ro finite cardinal numbers, namely (I) as those that are increased by the addition of 1, (2) as those that obey mathematical induction starting from 0. These two definitions can only be proved equivalent by the use of the "multiplicative axiom", which is not known to be true; but the use of this axiom in the proof in question is obviously unconscious. The book is divided into two parts, the first dealing with arithmetic, the second with geometry. Oddly enough, mathematical logic is dealt with in the second part, and its importance in connection with arithmetic is overlooked by the author. All mathematics, we are told, rests on the notions of unity and plurality, in the sense that, although these notions are 20 not simple, their genesis does not belong to mathematics. Certain difficulties are alleged to exist in the notions of plurality, and are accounted for on the ground that reality is one. The account of arithmetic appears in certain respects somewhat unsatisfactory. For example, the author admits that series depend upon asymmetrical relations, but nevertheless holds that all simple relations are symmetrical. He endeavours, by a somewhat elaborate method, to construct complex asymmetrical relations out of simple symmetrical relations; but his attempt does not appear successful, and it seems evident that every such attempt must fail. Again he assumes that in Mathematics "existence" means "freedom from contradic30 tion"-a view which, though widely held, appears untenable, if only because freedom from contradiction can never be proved except by first proving existence: it is impossible to perform all the deductions from a given hypothesis, and show that none of them involve a contradiction. The second part of the book, on geometry-apart from the chapter on mathematical logic-is chiefly concerned with the question whether, and in what sense, geometry is to be regarded as derived from, or as applicable to, facts of experience. The author contends (p. 126) that in so far as geometry includes certain facts of experience, it belongs to psychology, while in so far as it is mathematical, it is independent of experience and 40 of "fact". An examination of non-Euclidean geometry leads to the conclusion (p. 244) that experience shows that our rulers and compasses are Euclidean (so far as we have yet investigated them), but not that space is

T

68

.

69

mathematics Euclid and non-Euclid are equally justified.

Euc~1dea~: as f the place of rigid bodies in metrical geometry leads to the

A d1scuss10n o I · igid than to It that it is merely more convenient to treat P atmum _as r , :;:~\ guttapercha as rigid, though the a~thor, like ~- P_~ 1 ~~a~:' !~e~~~~ attempt to show why it is more convement, assummg I

tru~.merit of the book is the very full foot-notes, giving eno~gh qu~tat~ons from the authors referred to to enable the reader for himself.

to

appreciate thetr views

9 A MEDICAL LOGICIAN

9

A Medical Logician [1912]

Tms REVIEW WAS published in The Nation, IO (23 March 1912): 1,029-30. On 19 January 1912 Russell told Lady Ottoline Morrell: "An eminent Doctor named Mercier has written a fat book called The New Logic which the Nation has sent me to review" (#317). By 24 January he could report: "I have nearly finished Dr. Mercier on Logic-very creditable considering it is not his subject and he knows none of the recent work. It is pleasant reading and full of fun, but not really a contribution to the subject" (#323). The next day he wrote: "I have been reading my Dr. on Logic-I have only a few pages more" (#324). Russell "finished and posted" the review on the 26th of January (#326). The fine print under the author's name on the title page of his book supports Russell's judgment that he was eminent: Physician for Mental Diseases at Charing Cross Hospital; Examiner in Mental Diseases and Psychology in the University of London; Author of Criminal Responsibility; A Text-book of Insanity; Psychology, Normal and Morbid; Conduct and its Disorders; Crime and Insanity, &c, &c. Charles Arthur Mercier (1852-1919), M.D., was also a Fellow of both the Royal College of Physicians and the Royal College of Surgeons. This book was published in "The Modern Criminal Science Series". In a later book, On Causation with a Chapter on Belief (1916) Mercier offers an extensive critique (pp. 24-8) of Russell's views in "On the Notion of Cause" (18). In 1918 Mercier was the leader in an attempt to expel Russell from the Mind Association for his war-time activities. This interesting fact turned up during David Holdcroft's review of the minute book for the first eighty-seven years of the Association: But undoubtedly the most unusual item occurred at the 1918 A(nnual) G(eneral) M(eeting). The Agenda as usual gives nothing away, merely announcing under item four: 'New rule to be proposed by Dr Mercier.' His proposal was that 'no person who has been convicted of felony or of a serious offence against the Defence of the Realm Act be admitted or allowed to continue as a member of tlie Association'. Mercier clearly had

70

. mind· but he also ha\i a less obvious target, Balfour, _whom R usse 11 m ' · · 1h Ger he described in a letter to Ward as a 'man whose spmtua om~ is . ' He added for good measure, that he would rather associate with many. ' . h 1 · thieves than with such men. Mercier had tned to propo_se t e ru e m being able to attend himself because of ill-health, had t no t 1917, b u, H h a parently not been able to find an alternative proposer. owev_er, e :as able to attend in 1918, and his new rule was proposed but failed to carry, there being three for and eleven against. (Holdcroft 1987' 143) The minutes do not record the names of the voters. The copy-text is the printed version.

71

9 A MEDICAL LOGICIAN

A New Logic. By Charles Mercier. London: Heinemann,

422.

A

I912.

Pp. xxvii,

MAN WHO has achieved eminence in his own line must be possessed of notable courage and a certain love of adventure in order . . . to .desert t~e v~ntage.-ground of assured reputation and challenge criticism m a s~biec~ m "".hich he is unknown. There are not, perhaps, very many subiects m ':hich such an enterprise can be usefully undertaken. Among t~ese subi~cts, however, logic is certainly included, owing to the ~a~t that lt~ techmcal apparatus is small and fertile, and that the IO ~eal traimng required for logic (though possessed by very few logicians) is .compete~t fir~t-?and knowledge of scientific reasoning in at least one sci~nce. This trammg Dr. Mercier of course possesses; in addition, he has a vigorous common-sense, an interesting style, a lively sense of humour a~d a great belief in the importance of a sound logic. These qualities mak~ h.is book well wort~ readin~ by any educated man, not a professed logician, who takes an mterest m the subjects with which it deals. J?r. M.ercier's book is, in the main, an attack upon the traditional formal logic which has come down to us from Aristotle and the schoolmen. There are? broa?ly speaking, three kinds of logic now active; there is traditional 20 logic, umversally known to be pedantic nonsense, but still, for some in~cruta?le reason, cons~dered good enough for all young people who are so dl-a~vised as to submit to examination in logic; then there is the metaph~sic~l logic derived from Lotze, Sigwart, Bradley, and Bosanquet, whic? is tau~ht to advanced students by metaphysicians educated in the classi~s, and is by ~hem r~ga~ded as a correction to traditional logic; lastly, there is m~thematica.l logic, mvented by mathematicians as an analysis of ~athei:iatical reasomng, but not, as yet, giving any adequate account of mduct10n. Of these three ki?ds of logic, the third, which is recent, appears to be unknown to Dr. Mercier, except in early crude forms, probably those 3° of Jevons and Boole. The second, which he very justly regards as obscure ~ague, and ~~~less, is somewhat. inappropriately spoken of by him a~ mode~n logic , th?ugh ~he latest important work of this school (Mr. Bosanque~, s) was pubhs?~d 1888, and Dr. Mercier would hardly give the na~e modern medicme to a school whose most recent original contribut10n :V~s ~w~nty-~our years old. But in spite of many allusions to "moder? lo~i.c., lt is chiefly the traditional logic that Dr. Mercier pulverizes. ~is .crm.cisms are not in the main new; some of them were taught to Leibmz ~~ his you~h, two hundred and fifty years ago, and almost all will be famd~ar t? seno~s students. Nevertheless, he has performed a very useful 40 f~nct10n m se~tmg them forth vigorously, readably, wittily, and conclusively. Educat10n always lags behind the best knowledge, but in logic it has done so more than elsewhere, and it is nothing short of a scandal that

!?

72

73

young men and young women should still have their time wasted with Inversion, Obversion, "Barbara Celarent", and the whole farrago of r.ubbish seeing that all competent persons have long known it to be rubbish. D~. Mercier's original doctrines, especially in connection with induction show fresh first-hand thought, and are admirably illustrated from the ~easonings of science and daily life. His view, that induction always starts from what he calls a "problem", i.e. a proposition in which one term is unknown is illuminating and would seem to be new. But as a rule, ' . . although what is said constitutes a useful first answer to the questio~ m hand, it raises many new questions, sometimes ignored by Dr. Merci~r, sometimes referred for no apparent reason, to Psychology. He bases mduction upon the belief that what always has been, always will be; ~e apparently thinks that, though a few instances warrant .only an h~p~thes~s, very many instances will give absolute certainty. It might be s~id m c~it­ icism that induction never gives certainty, but only a gradually mcreasmg probability, and that the current statement of the pri~~iple of i~duction would be quantitative, introducing degrees of probability. It might also be said that the principle of induction itself, which we may state loosely in the form "what very often has been, probably will be in this instance", cannot in turn be proved inductively, but must be a logical principl~ of a different kind from any that Dr. Mercier recognizes. What he says is:

IO

20

Are we justified in concluding that since, in the experience of mankind a thing always has been, therefore it always will be? This problem does not belong to logic. It pertains t? Epistemology, and need not be considered here. Those .who wish to pursue the subject on these lines will find it treated m my book on Psychology. But psychology cannot tell us whether, in fact, the sun will rise tomorrow or bread be still nourishing when we eat our next meal, or any other f;ture event be as we should expect; yet it is such propositions that induction ought to render probable. It would be unjust to conclude without giving a specimen of Dr. Mercier's very excellent humour. One of the best specimens is a passage professing the profundity of his respect for Aristotle: "I suggest", he says, that if Aristotle is read aright, and understood in the sense he intended, it will be found that my doctrines are in harmony with his and are potentially contained in his 'Organon.' In case there is ;ny doctrine of mine that cannot be found in the 'Organon,' I submit that it was contained in those of his books that have been lost, and have not come down to us; and I defy my critics to prove

30

40

74

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

that it was not so contained. Moreover, if there is anything in Aristotle inconsistent with the doctrines I propound, then the passages in which these inconsistencies are found, are not to be attributed to Aristotle, but have been inserted by transcribers, either from carelessness or for their own nefarious purposes. This mode of argument is not, indeed, taught in Aristotelian logic, but it is freely employed by those who have been brought up in the Aristotelian atmosphere, and is an argument of great power and efficacy. It sometimes goes by the name of the Higher Criticism.

Part

II

The Problem of Matter

10

On Matter [1912]

THIS PAPER IS published here for the first time. Originally written for oral delivery to an audience in Cardiff on 17 May 1912, it was revised for reading to the Moral Science Club at Cambridge on 25 October 1912. It is assumed that the manuscript which exists is the revised version, although there is no way of proving that it is. Russell first mentioned the topic to Lady Ottoline Morrell on 23 April 1912: Wittgenstein brought me the most lovely roses today. He is a treasure. I have got a number of new technical ideas from him, which I think are quite sound and important .... I argued about Matter with him. He thinks it a trivial problem. He admits that if there is no Matter then no one exists but himself, but he says that doesn't hurt, since physics and astronomy, and all the other sciences could still be interpreted so as to be true. (#422) By the next night the project has become the writing of a paper: My intellect is amazingly clear these days-it sees into the heart of things in a white flash. I expect my paper on matter will be a model of cold passionless analysis, setting forth the most painful conclusions with utter disregard of human feelings. I haven't had enough courage hitherto about matter, I haven't been sceptical enough. I want to write a paper which my enemies will call "the bankruptcy of realism". There is nothing to compare to passion for giving one cold insight. Most of my best work has been done in the inspiration of remorse, but any passion will do if it is strong. Philosophy is a reluctant mistress-one can only reach her heart with the cold steel in the hand of passion. (#423) On 25 April he was hoping "to get a few moments for Matter" (#424). He began writing on 29 April: my whole morning was occupied with proofs, since then I have started writing on Matter and have reached p. 9. It will shock people, especially those who would like to agree with me-it is altogether too sceptical.

77

78

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

10 ON MATTER

Wittgenstein (who has just been here) is delighted, but no one else will be. (#427)

not dissatisfied-I thought there were many good ideas, still a little crude from newness, and too condensed, and none of them yet in a really clear or conclusive form, but I see the materials in it for something very important. Mr. William Pepperell Montague (the American Realist whom I met at dinner about a fortnight ago) was present, and provided comic relief by being an absolutely unspeakable fool. I itched to turn him into ridicule, but only yielded once to the temptation. He is pompous, pigheaded, and prolix; when one asks him for a definition, he makes a speech instead; he has apparently not even a wish to be clear. (#608)

During 2 May he "went on with my paper on Matter" ( #435); on the 7th the report is that he "went on with Matter, which I have nearly finished" (#440). But it was not until 13 May that he could write: I have just finished my paper on Matter-it is not as clear as it ought to be, but I don't think anybody will understand it. But I don't see how to improve it. I think the thought is all right, it is only the arrangement that is faulty. However, it will have to do. (#449) The day of its first delivery he was still dissatisfied with it: "I have been making additions and improvements in my paper on Matter, but it is still not fit to publish-however, it will do well enough to read" (#453). He gave the paper to Wittgenstein to read: "He thinks my paper on Matter the best thing I have done-but he has only read the beginning and end" (#460, 22 May 1912). Wittgenstein's approval, however, was short-lived. On 26 May Russell noted: "Wittgenstein doesn't like the rest of my 'Matter' paper, but only I think because of disagreement, not because of its being badly done" (#467). The paper recedes into the background until 27 September when he wrote: "I want to polish up the paper on Matter that I read at Cardiff, and get it published in the Monist" (#590). On 16 October he reported that he was reworking it: "I have been re-writing the paper on matter that I read at Cardiff in May, and I have got the problem very much clearer than it has ever been got before-I am very much pleased with what I have done" (#606). A letter of 24 October provides an additional reason for his return to it: "Tomorrow I read about Matter to the Moral Science Club" (#6ro). On the morning of the Club's meeting he confessed that he was not yet ready: I can't write properly because of a sense of hurry. I got hardly a moment yesterday for "Matter", which I have to read today-I don't know yet whether to say there is matter or there isn't! (#612) But a little later in the same letter he turns optimistic: "'Matter' opens out-I see vast problems growing out of it-it seems a central point from which to start one's journey. The concatenation of philosophical problems is always exciting" (#612). The morning after he read the paper to the Club he wrote: My paper on Matter last night was not a success, it was much too difficult. No one except Wittgenstein understood it at all. Moore was there, but hardly spoke; so was (G.H./ Hardy, and Desmond (MacCarthy/. Everybody had hoped for jokes and there were hardly any. I myself was

79

Later, in the same letter, he returned to the subject: My mind is very full of work, and there are hosts of things I seem to see my way to doing. Matter opens out a host of interesting problems. It is such a comfort to feel keen and have the zest for work, instead of having to force it out by continual efforts. (#608) Even after this disappointing reception he planned to work at the topic. On 27 October he told her: As soon as I have time, I want to go on with Matter. The subject is so fruitful that it might easily develop into a book. Just before I came to Lausanne, I got a central idea out of which a lot ought to come-I am pleased to find that Wittgenstein thinks just as well of that idea as I do. His health seems all right again, and he is full of good ideas. (#613) On 4 November he "got some time for Matter" (#621). At year's end, on 29 December, he was still thinking about the problem: The impulse to this work on Matter is extraordinarily strong, it quite possesses me, and drives me on like the lash of a slave-driver. The problem is one which nobody has considered or is aware of-I can't make people even see what I want to work at, except Whitehead and Wittgenstein, who feel its importance as much as I do. In fact Whitehead is partly the cause of my interest in it. I think it will be a long time before I get to the stage of writing anything. (#663) The last reference to the topic occurs on 5 March 1913 when he expressed the hope to "have some time for Matter" (#716). At about this time he concluded that epistemological questions must have priority over it, and began work on Theory of Knowledge (1984), which he also failed to complete. The manuscript (RA 220.ou360) is the copy-text.

10 ON MATTER

I

N WHAT FOLLOWS,

I shall endeavour to maintain three theses:

(r) That all the arguments hitherto alleged by philosophers against matter are fallacious; (2) That all the arguments hitherto alleged in favour of matter are fallacious; (3) That, although there may perhaps be reason to suppose that there is matter, yet we can have no means of finding out anything whatever as to its intrinsic nature.

IO

20

30

40

But before examining these theses, it will be necessary to determine what we mean when we speak of "matter", and this is by no means an easy task. It will also be desirable to show how the problem of matter is involved in any attempt to refute solipsism-a philosophy which, because of its unattractiveness, has never been treated with the respect which logically it deserves. The problem with which we are concerned is at once a metaphysical problem and a problem in the theory of knowledge; it is specially as a problem in theory of knowledge that I wish to consider it. All the knowledge we possess as to what exists rests upon two kinds of foundations: (r) immediate acquaintance, which assures us of the existence of our thoughts and feelings and sense-data, both those which we have at the moment and those which we remember; (2) general principles, according to which the existence of one thing can be inferred from that of another. The things with which we have immediate acquaintance are known to exist without the help of any general principles, but all other things, if we are to believe in their existence on any grounds other than blind prejudice, must be inferrible from things immediately known by the help of principles which must, in the last analysis, possess that very elusive quality called "self-evidence". The old-fashioned rationalist believed that some things-or at least one thing, namely God-could be known to exist by means of general principles alone, without appeal to acquaintance. The old-fashioned empiricist believed that all that science requires could be known to exist by means of acquaintance alone, without the need of general principles. The belief of the old-fashioned rationalist was embodied in the ontological argument; the belief of the old-fashioned empiricist was embodied in the notion that what is proved by induction is proved by experience alone, unaided by apriori logic. The ontological argument may perhaps by this time be left in the museum of historical curiosities. The empiricist's fallacy, however, still requires a word or two of refutation. The empiricist supposes that, in arguing that the sun will rise tomorrow because it has risen so often in the past, he is only appealing to what he calls "experience". But the slightest familiarity with logic would show him 80

81

that this is an error. Experience, by itself, will only prove itself; it will show that the sun has risen, not that it will rise. In order to argue from past to future, we must assume that the future will resemble the past in the relevant respects; and this cannot be proved by showing that it has been so in the past. Induction, therefore, requires an apriori logical principle of some kind, for which, by the nature of the case, no empirical evidence exists or ever can exist. Roughly, this principle is that when two properties are very often found together, and have never been found apart, if one of them is found in a new case the other will probably also be found; but it is not relevant to our present purpose to ask what the principle of induction is. It would be relevant to ask whether it is true; but for the present we may as well suppose it true, since otherwise all argumentation on the problem of matter is cut short at once. It must be admitted that, the more this principle is examined, the more it looks like a kind of cosmic red tape, unwarrantably assumed by those who find an insufficient amount of red tape here below. Nevertheless we will, for the present, assume it to be true, in order to concentrate on problems more specially connected with matter. The broad problem of which ours is a part may be stated as follows: "Can we know, either with certainty or with probability, of the existence of anything with which we have not direct acquaintance? And if so, how is such knowledge obtained?" Common sense believes that the things we see exist when we are not seeing them, it believes in other people's minds although it has no acquaintance with them, and it believes in such things as the inside of the earth, although, so far as we can discover, there is no reason to suppose that anybody has ever been acquainted with them. All such beliefs are challenged by our question. The common-sense belief in matter is a part of these beliefs. From the point of view of theory of knowledge, it is the most important part, since all knowledge of existence that goes beyond acquaintance must, apparently, be obtained, if at all, by way of knowledge of matter. Our knowledge that other people's minds exist, for example, seems plainly to depend upon our knowledge that their bodies exist. 1 In short, if we are to refute solipsism, it must be by first finding some reason to believe in matter. But before inquiring whether matter exists, we must first decide what we are going to mean by "matter". This question is by no means easy, and I am not sure that any very exact definition is possible. It might be thought that we could define matter as what is in space. But this at once raises difficulties as to what we mean by "space". There

IO

20

30

r I do not mean by their "bodies" something necessarily other than their minds; I mean 40 merely that existent, if any, which is inferrible from sense-data as existing independently of our perceptions.

82

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

is not, as in Kant's philosophy, a single given space containing all phenomena. There is a space of touch and a space of sight and a space of organic sensations; I do not know whether psychologists would consider that this exhausts the immediately given spaces. The single space of common sense is obtained by a rough correlation of these different spaces: it is not itself immediately given, but is a construction, an inference; moreover it cannot contain the immediate data, but only the "things" which common sense infers from the data and regards as the sources of correlated sensations of sight, touch, etc. These "things", however, are already matter, and are a IO highly disputable metaphysical construction. If the construction is legitimate, the space in which they are must be somehow connected with the given spaces and definable in terms of them. The same process, presumably, would enable us to define "things" in terms of sense-data. In fact, it would rather seem that the space of common sense is to be defined as the space in which matter is, than that matter is to be defined as that which is in common-sense space. Hence the attempt to define matter as that which is in space breaks down. In all speculations on space, it is important to distinguish clearly the abstract logical view from the view that starts from immediate experience. 20 Several spaces, as we saw, are given; each is a system of relations having certain logical properties. Hence the logician is led to give the name "space" to any system of relations having the same or similar logical properties. The geometer does not study given spaces, which in fact do not accurately fulfil his axioms; he studies any system of relations having that kind of logical resemblance to experienced spaces which is embodied in his axioms. It is easy to prove that, given any sufficiently numerous class of entities, there are relations between them which constitute a space in the abstract logical sense. Hence if "space" is understood in the abstract logical sense, we derive no substantial information about matter from 30 being told that it is in space. If, on the other hand, "space" is understood as experienced space, then matter, if it is to satisfy the demands of science and common sense, must not be in space at all, since it must be neutral as between the various immediately given spaces. It is plain that what is wanted is neither abstract space nor given space, but what we may call "physical" space, i.e. that space, whatever it may be, which physical science assumes and uses. This suggests that we might define "matter" as what is dealt with by physical science. It may be as well to remark that, from the standpoint of philosophy, aether (if it exists) and gross matter are both alike matter: 40 they both occupy the same space and are capable of moving in it. With this proviso, we may consider the definition of "matter" as that which has the properties discussed by physics. Take, for example, astronomy, and consider what the astronomer means when he speaks of the moon. He

10 ON MATTER

83

does not mean merely his sense-data when he looks at the moon: they do not describe an orbit round the earth, and there is no certainty that they exist when he is not looking at the moon. They are not spherical or mountainous or cold; in fact they are exceedingly different from the astronomer's moon. The difference is precisely the difference between sensedata and matter: the moon, as the astronomer means it, is a piece of matter, whereas the sense-data by which he comes to know about the moon are not matter. We may, then, by way of making a beginning, define "matter" as that which satisfies the hypotheses of physics. We may re-state this definition as follows. Physics may be studied, to IO begin with, as a piece of pure mathematics; the space and matter concerned in this study are variables, concerning which certain hypotheses are made; that is to say, they are not definite given entities, but merely anything having certain properties. But if physics is to be applied, as for example in calculating the motions of the moon, it must be possible to derive, from the sense~data which we have when we see the moon, some object, either inferred or constructed, which satisfies the hypotheses of abstract physics. Sense-data themselves do not satisfy these hypotheses; yet physics enables us to deal with sense-data, as for example when we infer that an eclipse will be visible. Thus the matter which physics deals 20 with, though not identical with sense-data, is connected with them and inferrible from them. Matter, therefore, may be defined as that class of objects, if any, which are inferrible from sense-data and satisfy the hypotheses of physics. The problem then arises whether there is any such class of objects, and, if there is, what is their relation to sense-data. This problem, however, is no longer one of definition, and we will therefore not consider it yet. It is important to realize that, whenever mathematics is applied to the actual world, there must, if the analysis is pushed far enough, be some substitution of actual particular sense-data for the variables of pure math- 30 ematics. Let us illustrate this by an elementary example. The proposition "If Socrates is a man, and all men are mortal, then Socrates is mortal" is a proposition in applied mathematics, obtained by giving special values to x and a and (3 in the proposition "If x is an a, and all a's are (3's, then xis a (3, whatever x and a and (3 may be", which is a proposition of pure mathematics. But "Socrates" and "man" and "mortal" are none of them sense-data, though they are all what we may call "empirical", that is, sense-data enter into their meaning. There are an indefinite number of different ways in which they may be defined in terms of sense-data, and these different ways are all legitimate. Let us choose one way. Let us de- 40 fine Socrates as "the entity called 'Socrates"'. Let us define "a man" as "any body closely resembling mine together with an associated consciousness also closely resembling mine". This, it will be seen, is a vague con-

84

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

ception, since the degree of resemblance required is left doubtful; but this vagueness is unavoidable on Darwinian grounds. Let us define the moment of a man's death as the earliest moment after which he has no consciousness. Then "Socrates is a man" becomes "The entity called 'Socrates' is a body closely resembling mine together with an associated consciousness closely resembling mine". "Socrates is mortal" becomes "There is an entity t such that no later entity is a conscious act of the entity called 'Socrates', and there are earlier entities which are such acts, and given any earlier entity t', there are such acts which are later than t' IO and earlier than t". In the above statements, the reduction to sense-data and a priori universals is not yet quite complete. The word "called" would demand elucidation; so perhaps would "earlier" and "later"; but above all the word "body", which occurs in defining "man", raises the very problem of matter which we are engaged in considering. The above definitions, however, incomplete as they are, suffice to indicate the process by which everything can be reduced to combinations of apriori universals with particulars immediately given. The processes applied above are just such as are required in order to travel from the sense-data called "seeing the moon" to what the astronomer means when he speaks of the moon. 20 We may now re-state the relation of physics, considered as pure mathematics, to our knowledge of the actual world. In physics as commonly set forth, the values of the variables are pieces of matter. But since matter is not among the things with which we have direct acquaintance, there is a further process philosophically necessary, connecting matter with sensedata, and introducing different variables from those of ordinary physics, namely variables among whose values actual sense-data are contained. This is in essence the same change as when, in place of "Socrates", we put "the entity which is called 'Socrates'", thus replacing the x of "x is a man" by an x whose values are names, which are things with which we 30 have direct acquaintance. Let us suppose that we have completed the process of substituting variables whose values are of the nature of sensedata for variables whose values are bits of matter. Then, and not till then, physics is capable of verification. The process of verification consists in discovering that sense-data satisfy the hypotheses of the hypotheticals which constitute pure physics as thus completed. Empirical verification, where physics is concerned, can only arise from sense-data; hence any two systems, or any two interpretations of the same symbols, which are equally verified by sense-data, are equally justified empirically. If we are to be able to choose between two such systems or interpretations, it must be on 40 the ground of some more or less self-evident a priori principle or principles. It follows from this that there are only two alternatives in regard to matter, if we are to have any reason to believe in matter. (r) It may happen that a piece of matter is a mere logical construction from certain

10 ON MATTER

85

sense-data, for example a combination of visual, tactile, and other sensedata associated together by some experienced relation. (2) It may happen that we know some a priori principle by which, from sense-data, we can infer the existence of entities of a sort with which we are not acquainted, but which we know to possess the kind of properties that physics assigns to matter, such as impenetrability and indestructibility and motion in a physical space given by the supposed a priori principle. At a later stage, we must attempt to decide between these alternatives. In interpreting and applying physics, we have to travel backward from matter as science conceives it to sense-data as we know them in immediate experience. In philosophizing on sense-data, we travel forward, from the immediately given towards some inferrible set of entities possessing, if possible, the properties which physics assigns to matter. It is time to consider more in detail the differences between the matter of physics and our sense-data, with a view to discovering how, if at all, a bridge between the two can be constructed. In the first place, our sense-data are only known to exist while our sensations last. The colour which I see when I look at an object is not known to be still there when I shut my eyes. So far as a priori considerations are concerned, there is no reason why colours should not exist unseen. Idealist arguments to the contrary all rest, I think, upon confusions, chiefly confusions between different meanings of "knowledge". It is argued that I cannot know of the existence of something I do not know. This sounds plausible, but is really a mere confusion. I know my great-grandfathers existed, although I have never known them. That is to say, one may "know" (in the sense of knowledge of truths) that there are entities with such and such properties, without "knowing" (in the sense of acquaintance) the particular entities which have these properties. As soon as this confusion is understood, it becomes obvious that the further idealistic contention, that nothing can exist which is not known to some mind, is equally fallacious. There is therefore no a priori reason why the objects which are at certain times sense-data, such as colours, should not continue to exist at times when they are not data to any mind. The arguments against naive realism, which simply identifies matter with collections of my sense-data, are not a priori arguments, but empirical arguments derived from the fact that sense-data apparently change in ways which we must attribute to ourselves rather than to the object. Also, if naive realism is accepted, we must admit other observers of the same object, and they, if they exist, do not see the object quite as we see it. For such reasons of detail, which we may here take for granted, naive realism probably cannot be accepted without modification, and we cannot assign to sense-data the properties which we wish matter to have. It is possible that the arguments against naive realism could be met by a theory which regards a piece of

IO

20

30

40

86

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

matter as consisting entirely of constituents of the nature of sense-data, by including everything that could be a sense-datum to any possible observer. I shall return to this point later. For the present, it is enough to observe that matter cannot be simply identified with sense-data. It should be observed that, in a fundamental discussion such as we are concerned with, we cannot bring in other observers besides ourselves, except in refuting theories which admit them. If matter is to be defined in terms of sense-data, it must be defined in terms of our own sense-data primarily, since it is only after we have decided on the existence of matter IO (in some sense) that we are in a position to know that other observers exist; for other observers, when once their existence has been questioned, can only be inferred from our sense-data by an inference more elaborate than that which leads us to matter. What we want matter to be is some object which is "associated", in a way to be further investigated, with the various sensations of sight, touch, etc., which common sense regards as giving knowledge of the "same thing"; and it is further necessary that matter should not exist only while we are having the sensations "associated" with it. This is the minimum statement of what we want, intentionally somewhat vague and tentative. 20 With this tentative statement in mind, we may now turn to the question whether there is any reason to suppose that matter, in this sense, exists. One argument in favour of the existence of matter, which formerly seemed to me very strong, is the inductive argument, which may be stated, in outline, as follows: Physical science, by supposing that there is matter, is able to frame theories which fit the facts in all verifiable respects, and combine in a system many facts which would otherwise remain isolated and chaotic. The appearance and disappearance of sense-data is primii facie irregular and capricious, but by supposing them caused by the interaction of matter and the observer, they can be brought under general 30 laws which are simple and render sense-data to some extent predictable. This argument, though it has some weight, no longer appears to me to give any very overwhelming probability in favour of matter. The argument has, however, several aspects, which must be considered separately. The argument from simplicity, to begin with, is merely teleological; and has absolutely no weight whatever. If it were known that the universe had been created for the purpose of delighting mathematicians, there would be some reason to suppose that, of two hypotheses which both fit the facts, the simpler is more likely to be true. As, however, there is no evidence that this is the purpose of the universe, there is no reason whatever to 40 expect the true laws of nature to be simple. There is, of course, a practical reason for adopting the simpler of two hypotheses in our scientific work, so long as both are equally applicable to the facts: it will be easier than the other, more likely to enable us to deduce results confirmed or confuted

10 ON MATTER

87

by observation, and giving a body of theory which can probably be translated into the language of the more complicated theory if subsequent facts should make this necessary. But this affords no reason whatever for supposing that the simpler theory is actually true. To take an illustration: The heavenly bodies may be mere bright patches moving about on the celestial sphere exactly as they appear to move; on the other hand they may be as the Copernican system supposes them to be. There is nothing in the facts to enable us to decide between these two hypotheses, but in practice we adopt the Copernican hypothesis because it gives a simpler statement of the motions of the heavenly bodies. IO It is worth while to analyze a little further the case of the heavenly bodies. If we refer them to polar coordinates with the earth as origin, their angular coordinates will be observable, but the radius vector will always be a pure inference. For any given planet, the radius vector is not a function of the angular coordinates, since these may recur with a different value of the radius vector. Thus the radius vector introduces an unobservable distinction between two cases which give the same sense-data. It is introduced, of course, because it enables us to state simple laws for the motions of the heavenly bodies; but what is called the verification of these laws applies only to the angular coordinates, since the radius vector is not 20 amenable to observation. It is obvious, therefore, that other values of the radius vector, with other laws of gravitation, would equally lead to all verifiable results, though the formulae would require some transformations before they became equally simple. From the point of view of the pure mathematician, such hypotheses are essentially indistinguishable inter se, but they are all distinguishable from the hypothesis of a single celestial sphere, since they introduce differences between configurations which that hypothesis would identify. A similar observation may be made as regards matter in general. There will be an infinite number of hypotheses, very different from the stand- 30 point of the metaphysician, but indistinguishable from the standpoint of the pure mathematician, since all give the same formulae, and the difference lies only in the meaning assigned to the symbols. If these different hypotheses agree, as many of them will, in all their verifiable consequences, it must be impossible, henceforth and forever, to decide between them on empirical grounds. This illustrates the second weakness in the inductive argument for the existence of matter. It is contended that, because a certain hypothesis fits all relevant known facts, therefore there is at least a probability that it is true. Such a contention is only valid if it is known that there are not likely 40 to be other hypotheses which also fit the facts. In our case, this knowledge is absent. The question to be answered, before the inductive argument can be considered valid, is this: Would it be possible, without assuming

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

10 ON MATTER

the existence of some non-sensible thing behind the sense-datum, to give to the symbols of dynamics a definition in terms of sense-data alone? If this is possible, then the whole of dynamics remains symbolically true even if there is no such thing as matter. The question whether it is so is a straightforward one, obviously capable of a demonstrable answer either affirmative or negative. Until this answer has been obtained, the inductive argument remains suspended, possibly valid, but very likely to be invalid. Can Dynamics be interpreted in terms of sense-data alone? (Of course I mean the sense-data of one person.) The answer, I think, must be negative. Assuming that Dynamics, suitably interpreted, is true, there must be a material world having a many-one relation to sense-data, not a oneone relation. Our previous example of the radius vector in astronomy illustrates this point: there is nothing in astronomical sense-data to correspond to the radius vector. The radius vector, therefore, is purely hypothetical: if there is to be any ground whatever for supposing it to have such and such values, the ground must lie, not in sense-data themselves, but in some kind of law by which inference is possible from sense-data to other things. When the condition of a system is a function of a certain number of quantities, these quantities are called the coordinates of the system. Now the Solar System, as dealt with by astronomy, has more coordinates than the system of sense-data from which it is derived. The white dot of light which represents a planet has only two coordinates, indicating its position on the celestial sphere; but the planet itself is supposed to have a third coordinate, namely its distance. We may so choose the coordinates of any physical system as that some of them shall have a one-one correspondence with sense-data, while others shall be entirely independent of sense-data, i.e. different values of them will not correspond to different sense-data. Those that correspond with sense-data may be called verifiable coordinates, while those that do not may be called hypothetical coordinates. Thus in the case of the planet, its angular coordinates are verifiable, while its distance from the earth is hypothetical. Thus stated in logical terms, the problem of matter is this: What ground have we for assuming hypothetical coordinates? In other words, what ground have we for supposing that the relation of the state of the world to our sense-data at any moment is manyone, not one-one? This is the problem of matter. Let us take another illustration from the sky. You may watch two clouds moving with different apparent velocities; presently one of them eclipses the other. You then infer that the one was nearer to you than the other. You do not believe that the one that disappears ceases to exist, and in fact you know that it will probably reappear when the lower cloud has passed. Seeing the two clouds approaching the same line of vision, you know that one of them will disappear, and you infer that, in spite of exactly similar

sense-data, there must be a difference between the case which ends in the disappearance of the one and the case which ends in the disappearance of the other. This difference you regard as arising from the hypothetical property of distance. It is of course by no means certain that visual distance is always hypothetical: with regard to small distances it may be immediately perceptible; but astronomical distances and even the distances of clouds are certainly not given in sense. The reason for inferring them seems to lie in the assumption of some kind of uniformity. If two states of affairs which are not sensibly different develop into states that are sensibly different, we argue that there must have been a non-sensible difference in the two original states. It is by no means easy to judge whether there is any validity in such an inference; obviously it cannot give certainty, and obviously it must be very difficult to state it validly at all. In order to decide, we should have to consider the nature of scientific laws, and whether there is any a priori law fulfilling the functions attributed by philosophers to the law of causality. If there is such a law, the inference from sense-data to matter may be based upon it; but if there is no such law, it is difficult to see how the inference can give more than a possible hypothesis. Before going further, it will be well to point out the connection of the above problem with the question whether there is any reason to assume any existents beyond sense-data. There are two distinct questions, which, though connected, must not be confused. (1) There is the general metaphysical question: Have we any ground for assuming the existence of anything except sense-data? (2) There is the more special question: Does physics in fact essentially require the existence of anything other than sense-data if it is to be true? The question we have been considering in the last few pages is the second of these. This, being a question of analysis of scientific procedure, is capable of a definite answer. If, as we have contended, the relation of the world assumed by physics to our sense-data is many-one, not one-one, then it must be impossible, however we may interpret our symbols, to preserve the truth of physics without assuming anything beyond sense-data. If, on the other hand, the relation had been one-one, it would have been possible to interpret our symbols in such a way that matter should be a purely logical function of sense-data. There would then have been no physical reason for assuming the existence of a matter distinct from sense-data. As it is, such a reason does seem to exist. A metaphysical or logical reason might exist, even if a physical one did not. As it is, analysis shows that there must be a physical reason, but it has not yet shown us what that reason is. This inquiry will bring together the physical and metaphysical questions which we distinguished above; for, when we have found the physical reason, we shall still have to consider whether it is valid.

88

10

20

30

40

89

10

20

30

40

90

IO

20

30

40

91

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

10 ON MATTER

The questions we have to consider now are: What are the reasons which lead Physics to assume that the relation of the whole of what exists at a given moment to our sense-data at that moment is many-one, not oneone? And are any of these reasons valid? The meaning of our question is the following: Why does science assume that in two cases where all that we can directly observe is exactly similar, there nevertheless are sometimes differences which we cannot observe? There is a certain difficulty in stating the question precisely, because as a matter of fact the whole of what is observable on two different occasions will never be exactly similar in all respects. There is an assumption that, in a complex sense-datum, every part severally corresponds with a part of the "real" or "physical" object. In the case of visual space, for example, we tend to assume that what we see in a given direction depends upon what there really is (whatever that may mean) in that direction. This assumption, however, is not merely gratuitous, but in many cases plainly false-in the case of mirrors and all other reflections, and also in cases of refraction. Thus there is a difficulty in securing that piecemeal point-topoint correspondence of visual appearances with physical objects without which our assumption of repetition in the observable data becomes unwarrantable. To clear our ideas, let us imagine an ideally simplified world, in which our only sense-data are two discs, one red and one yellow, which move backwards and forwards in what appears to be a straight line, with periodic changes of velocity. Let us suppose that when they reach the same line of sight, sometimes the red disc gradually disappears and reappears, and sometimes the yellow disc does so. How will science interpret such a series of events? One obvious interpretation will be to regard the two discs as the sensible representatives of two spheres, moving in ellipses about their common centre of gravity in the same plane with the observer. This interpretation, in common with any other that current science would be likely to entertain, assigns to the physical objects a property of distance to which nothing sensible corresponds. It may be said that the apparent size of the discs will vary with the distance. But if the difference between the greatest and least distances is very small compared to the mean distance, the difference of apparent size will be imperceptible. It is therefore not this that causes us to assume distance: what makes us assume distance is the difference between the case when the red disc is eclipsed and the case when the yellow disc is eclipsed. We may suppose the sensible antecedents of the two cases exactly similar, and yet the result different. Hence we infer differing antecedents which were not sensible. If this is the correct account, it would seem that, although the maxim "same cause same effect" is not very useful in its direct form, the inverse form "different effects, different causes" is of fundamental importance in

establishing the existence of matter. But we cannot yet regard ourselves as at the end of our labours. Our ideally simplifed world was merely an illustration, and we must be able to state our principle in terms which will apply to the actual world. In the real world, the total sense-data of one moment are practically certain to be different from the total sense-data of another moment. Hence, even assuming the truth of science, the relation of the total sense-data of one moment to the total state of the world at that moment is one-one, not one-many. Thus before our principle can be applied, we must be able to analyze the total sense-data, regarding some as corresponding to one object and some as corresponding to another. But this introduces a host of new difficulties, particularly as to what is meant by "corresponding". It is possible that all these difficulties can be satisfactorily dealt with, but I do not myself know how to deal with them. I must therefore for the present leave them entirely unsolved. In spite of these difficulties, I cannot help thinking that, with suitable care in interpretation, the principle "different effects, different causes" would be found to give a reason in favour of matter. It may be said that scientific laws which would lead, even in a purely hypothetical case, to a violation of this principle, cannot be true; and that, unless we assume matter, such hypothetical violations of our principle cannot possibly be avoided. It may also be suspected that the principle is merely a simplified form of some more general principle, of which there would be actual violations if sense-data constituted the whole universe. I cannot at present bring any arguments in support of this suspicion, but I think something may be done towards generalizing the principle "different effects, different causes". All statements that are apparently about causes and effects require a good deal of modification before they can be regarded as correct. To begin with, nothing short of two states of the whole material universe can be accepted as a true cause, if we accept science and insist that the same cause shall always produce the same effect. But this renders the conception of cause practically inapplicable. Hence we are thrown back on scientific laws as the working substitute for causality. Now the real purport of such rather vague principles as "same cause same effect" or the "uniformity of nature" seems to be that absolute time must not appear in any scientific law. Any law connecting events at different times must be independent of the actual date at which the events occur, though it may of course involve the lapse of time between the earlier and the later events. Suppose it were found, for example, that the constant of gravitation appeared to be a function of the time, and to be (let us say) slowly increasing. We should unhesitatingly attribute this, not to the mere change of date, but to some concealed process which led to different states of things at different dates. Why we assume this, I do not know. It might be thought to

IO

20

30

40

92

10

20

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

be on the ground that there is no such thing as absolute time, but this ground seems inadequate, since the absolute date of an event may be defined as the class of events simultaneous with the given event. However this may be, there seems to be some a priori principle which assures us that the date cannot enter into a scientific law. What this principle is, and whether it is true, are questions requiring more analysis than I have so far been able to effect. It is, however, not difficult to see that there might be a principle of this kind which would make the inference from sensedata to matter legitimate. It is possible that other more or less scientific grounds for inferring matter may also exist, but I have not succeeded in discovering any that seemed very cogent. Arguments from continuity might be urged: if, for example, you continue to look at a moving object, you find that it passes by a continuous path from one position to another, and this suggests that an object which has changed its position while you were not looking must also have pursued a continuous path, and must therefore have existed when it was not seen. But I doubt if there is any real force in considerations of this kind. It remains to consider arguments of a more general philosophical nature. It is plain that the realism of common sense is not based upon such considerations as we have been concerned with, and it is not impossible that the realism of common sense may contain some element of truth. II.

30

Matter, if it is to be known to exist at all, must be known through some a priori principle assuring us that our sensations in some way "correspond" with things which can exist without our sensations. In regard to touch and (to a less extent) sight, the unsophisticated adopt the standpoint of naive realism, with just so much correction as obvious facts force upon them; such obvious corrections-for example, the bent stick in the water-they make by means of the notion of "illusion". In regard to the other senses, it is doubtful whether even the most unsophisticated are naive realists. But, as Meinong pointed out,2 though it is easy to shake the belief that things are exactly as they appear to be, it is by no means easy to shake the belief that the appearance is the appearance of a thing. Tables and chairs, though they may not be exactly what is given in sight, 2

40

Ueber die Erfahrungsgrundlagen unseres Wissens, Berlin, 1906. Cf. p. 92: "Der Naive, von

1 J,,;.fc;_.~= JI a.

fC,,_ ol,;,k,_,v,

~A,_.:_._

C"-'A,

e~ .~Cr~

tQ f~·v=

,>e-,

L'!/id..J>.-J

ga,_,_,,

,0._

.:.c....

f. .o - Ja/J.-..._ ."k..9( ~ e.~{- d

c;.fl9

11. ~ fl.r fl.. a.cc-: ~ e-fr Gf2J! (~~ .,; ..,f,...;.. J2 jl.._ C.::..c'Lu,~ t._·r;,J d') ~ •

o..1)-

~• te..,._

{)jJfr_........

~ -~

Dh."l!Q a..

J.:(/;~7 ~ ~ .LJ. ,;., ~ .1!'4.J'~·aro

(72..e- [ ff:.._:.k. fl;, oJJ ""{~ /f.o4 (~ JL.J2R a.~~ ~ a r·~ ~ ~ ..... "")c,,,....... ~ J_..;,~ cJ-~, tf..."""'-J.frJ.s.~ "Yff~

°!..

f..-L

~ - ~d-a I teL Jll.,. c0...:..(/ ,,,,..___,._ r-~. tP PLi[-.,. .:_ ~ ,

1/ '-'-'~ ,;_

~ -'1- ~)( ic... ~,~

c~,J). ...,.;It £.....,...-~

0),. ....,_J fM ;.,, ,,:._~ . .:. .

f12-c".

,

. 9,~. ~ ~ (...:_ ~ ~~ '-"-"""- ~ fc--.-1.

~ ':

,,...µ. l:,c. ca.~~

&.

ff-...

~ ~ J..u.,h-..-~ f,f, ' c..~ ~.._ °'''-t:~ ~-, :: "l--

':.

o->C-

~



Oa_

a-..S>.

~

f

LU..

f, 1

ri'cl.·, ..

eA·- 0 ""'-"'cJ,-,_,,_ :· ~ 0 ~· 1 ~ wJ,,.:.......:i ~ ~- 1'~ e~Q( a,,._·J..,..J:- ... "~9-" ~ fo

s. . M.__ ' ~

er......)-

G,Jt.. ~ .:.. l~ ar.,

c.Ca...e 0

,.,.:_

re.. ae,· . '({..;.

Uh Matter

If:.. z 1 • Then however the particle moves, there must be, theoretically, functions f1> f 2 , f 3 , such that

207

It follows that, theoretically, the whole state of the material universe at time t must be capable of being exhibited as a function of t. Hence our universe will be deterministic in the sense defined above. But if this be true, no information is conveyed about the universe in stating that it is deterministic. It is true that the formulae involved may be of strictly infinite complexity, and therefore not practically capable of being written

IO

20

30

40

208

ro

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

down or apprehended. But except from the point of view of our knowledge, this might seem to be a detail; in itself, if the above considerations are sound, the material universe must be deterministic, must be subject to laws. This, however, is plainly not what was intended. The difference between this view and the view intended may be seen as follows. Given some formula which fits the facts hitherto-say the law of gravitation-there will be an infinite number of other formulae, not empirically distinguishable from it in the past, but diverging from it more and more in the future. Hence, even assuming that there are persistent laws, we shall have no reason for assuming that the law of the inverse square will hold in future: it may be some other hitherto indistinguishable law that will hold. We cannot say that every law which has held hitherto must hold in the future, because past facts which obey one law will also obey others, hitherto indistinguishable but diverging in future. Hence there must, at every moment, be laws hitherto unbroken which are now broken for the first time. What science does, in fact, is to select the simplest formula that will fit the facts. But this, quite obviously, is merely a methodological precept, not a law of Nature. If the simplest formula ceases, after a time, to be applicable, the simplest formula that remains applicable is selected, and science has no sense that an axiom has been falsified. We are thus left with the brute fact that, in many departments of science, quite simple laws have hitherto been found to hold. This fact cannot be regarded as having any a priori ground, nor can it be used to support inductively the opinion that the same laws will continue; for at every moment laws hitherto true are being falsified, though in the advanced sciences these laws are less simple than those that have remained true. Moreover it would be fallacious to argue inductively from the state of the advanced sciences to the future state of the others, for it may well be that the advanced sciences are advanced simply because, hitherto, their subject-matter has obeyed simple and easily-ascertainable laws, while the subject-matter of other sciences has not done so. The difficulty we have been considering seems to be met partly, if not wholly, by the principle that the time must not enter explicitly into our formulae. All mechanical laws exhibit acceleration as a function of configuration, not of configuration and time jointly; and this principle of the irrelevance of the time may be extended to all scientific laws. In fact we might interpret the "uniformity of nature" as meaning just this, that no scientific law involves the time as an argument, unless, of course, it is given in an integrated form, in which case lapse of time, though not absolute time, may appear in our formulae. Whether this consideration suffices to overcome our difficulty completely, I do not know; but in any case it does much to diminish it.

18 ON THE NOTION OF CAUSE

209

It will serve to illustrate what has been said if we apply it to the question of free will. (1) Determinism in regard to the will is the doctrine that our volitions belong to some deterministic system, i.e. are "determined" in the sense defined above. Whether this doctrine is true or false, is a mere question of fact: no a priori considerations (if our previous discussions have been correct) can exist on either side. On the one hand, there is no a priori category of causality, but merely certain observed uniformities. As a matter of fact, there are observed uniformities in regard to volitions; thus there is some empirical evidence that volitions are determined. But it rn would be very rash to maintain that the evidence is overwhelming, and it is quite possible that some volitions, as well as some other things, are not determined, except in the sense in which we found that everything must be determined. (2) But, on the other hand, the subjective sense of freedom, sometimes alleged against determinism, has no bearing on the question whatever. The view that it has a bearing rests upon the belief that causes compel their effects, or that nature enforces obedience to its laws as governments do. These are mere anthropomorphic superstitions, due to assimilation of causes with volitions and of natural laws with human edicts. We feel that 20 our will is not compelled, but that only means that it is not other than we choose it to be. It is one of the demerits of the traditional theory of causality that it has created an artificial opposition between determinism and the freedom of which we are introspectively conscious. (3) Besides the general question whether volitions are determined, there is the further question whether they are mechanically determined, i.e. whether they are part of what was above defined as a mechanical system. This is the question whether they form part of a system with purely material determinants, i.e. whether there are laws which, given certain material data, make all volitions functions of those data. Here again, there 30 is empirical evidence up to a point, but it is not conclusive in regard to all volitions. It is important to observe, however, that even if volitions are part of a mechanical system, this by no means implies any supremacy of matter over mind. It may well be that the same system which is susceptible of material determinants is also susceptible of mental determinants; thus a mechanical system may be determined by sets of volitions, as well as by sets of material facts. It would seem, therefore, that the reasons which make people dislike the view that volitions are mechanically determined are fallacious. (4) The notion of necessity, which is often associated with determinism, 40 is a confused notion not legitimately deducible from determinism. Three meanings are commonly confounded when necessity is spoken of:

210

ro

20

30

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

(a) An action is necessary when it will be performed however much the agent may wish to do otherwise. Determinism does not imply that actions are necessary in this sense. (/3) A propositional function is necessary when all its values are true. This sense is not relevant to our present discussion. ( y) A proposition is necessary with respect to a given constituent when it is the value, with that constituent as argument, of a necessary propositional function, in other words, when it remains true however that constituent may be varied. In this sense, in a deterministic system, the connection of a volition with its determinants is necessary, if the time at which the determinants occur be taken as the constituent to be varied, the timeinterval between the determinants and the volition being kept constant. But this sense of necessity is purely logical, and has no emotional importance. We may now sum up our discussion of causality. We found first that the law of causality, as usually stated by philosophers, is false, and is not employed in science. We then considered the nature of scientific laws, and found that, instead of stating that one event A is always followed by another event B, they stated functional relations between certain events at certain times, which we called determinants, and other events at earlier or later times or at the same time. We were unable to find any a priori category involved: the existence of scientific laws appeared as a purely empirical fact, not necessarily universal, except in a trivial and scientifically useless form. We found that a system with one set of determinants may very likely have other sets of a quite different kind, that, for example, a mechanically determined system may also be teleologically or volitionally determined. Finally we considered the problem of free will: here we found that the reasons for supposing volitions to be determined are strong but not conclusive, and we decided that even if volitions are mechanically determined, that is no reason for denying freedom in the sense revealed by introspection, or for supposing that mechanical events are not determined by volitions. The problem of free will versus determinism is therefore, if we were right, mainly illusory, but in part not yet capable of being decisively solved.

Part

IV

Ethics

19

The Elements of Ethics [1909]

THE PUBLISHING HISTORY of this paper is complicated. The essay is divided into six sections, of which four were published in journals before the entire essay was published in Philosophical Essays (19rn). Section IV, "Determinism and Morals", was published under that title in The Hibbert Journal, 7 (Oct. 1908): I 1321. Sections I and n, "The Subject Matter of Ethics" and "The Meaning of Good and Bad" were published in The New Quarterly, 3 (Feb. 19rn): 21-34, and Section III, "Right and Wrong'', in the same journal, 3 (May 19rn): 131-43. These three sections were collectively entitled "Ethics". Section III ends with "to be concluded'', but the May issue proved to be the last for The New Quarterly. The editor, Desmond MacCarthy, wrote to tell Russell on 9 July 19rn that he could expect proofs in a few days time. Presumably these proofs would have been of Sections v and VI. Russell included the entire essay, all six sections, in Philosophical Essays, under the title "The Elements of Ethics". The footnote which he added detailing its earlier appearances in print s,uggests that he had read the proofs of Sections v and VI, because he states that these sections were published in the September, 1910, edition of The New Quarterly. The Preface of the book is dated "July 19rn" which fits with the date of MacCarthy's letter. In 1952 Russell granted Wilfrid Sellars and John Hospers permission to reprint the paper in their Readings in Ethical Theory, but with a cautionary footnote: The author has requested that the following note be printed in conjunction with this selection: "'The Elements of Ethics' was written under the influence of Moore's Principia Ethica. There are some important points in which, not long after publishing it, I came to disagree with the theory that it advocates. I do not now think that 'good' is undefinable, and I think that whatever objectivity the concept may possess is political rather than logical. I was first led to this view by Santayana's criticisms of my work in his Winds of Doctrine, but have since found confirmation in many other directions. I am not, however, quite satisfied with any view of ethics that I have been able to arrive at, and that is why I have abstained from writing again on the subject." Russell's memory failed him here: he did write on ethics on several later occasions. The most important of them are mentioned later in this Headnote.

213

214

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

The first draft of this paper appears

to

have been written in April and May of

1905. A group of young men, led by Sydney Waterlow and George Trevelyan,

proposed to publish a "Manifesto" inspired by G. E. Moore's philosophical views. This manifesto was to be a book with chapters by different writers: Moore was to write on "Truth", Russell on "Ethics", etc. Paul Levy's Moore (1979, 251-9) provides as full an account of this aborted project as we are ever likely to have. On 2 May 1905 Russell wrote to Moore: We are going to have a meeting of the people concerned in Waterlow's project of a book on Saturday in Dickinson's rooms at 9 p.m. If you are in Cambridge, as I hear you may be, I hope you will come, though I shall be rather alarmed, as I am reading on Ethics, and the paper I have written is merely yours (Principia Ethica> boiled down, popularized, and with almost all the exactitude omitted. I have gone on the plan of not discussing points which everyone except a trained philosopher would regard as trivial or logic-chopping. Moore attended the meeting and was not favourably impressed with Russell's paper. A second discussion on the paper was arranged later in the month and Russell again urged Moore to attend. Moore borrowed Russell's paper and studied it carefully. On 23 October 1905 he returned it with these comments: I am sending back your MS on Ethics. I am sorry I have kept it so long; but I have been trying very hard to clear off other arrears of work, and always hoping to get them finished first. I am sending what comments I have to make, on separate sheets in pencil. When I began them, I intended to copy them out properly; but as my arrears are still not worked off and I am likely to be pressed for time, I send them as they are. I think you will be able to make them out just as easily from pencil as if I had copied them out: I have numbered each sheet in its proper order. It will appear from my comments that it is the four last Sections, which I think need the most extensive reconstruction. Of these four, I think Section VI is the least satisfactory: but I can't agree with you and Dickinson in thinking that there should be no such Section. After insisting that "right conduct" depends so much on good results, I think it would be absurd not to try to indicate what results are good: let alone the fact that I think it both important and easy to combat convincingly both your "moralist's fallacy" and Hedonism. I think your Section is unsatisfactory, not because any such section must be too dogmatic, but because you have laid too little emphasis on what are undoubtedly great goods and great evils, as compared with comparatively disputable points. E.g.

19 THE ELEMENTS OF ETHICS

215

I don't think there is the least need to mention the questions whether pleasure or knowledge are at all good per se, except simply to emphasize the point that pleasure per se is certainly not a great good; nor do I see any need to mention the question whether true and false beliefs affect the value of good or bad wholes. I don't know what Keynes' view about goods and organic unities is. But it seems to me plain that, in the form you mention it, it must be wrong. What seems to me plain is that it may be much better to have one and the same feeling towards one object than towards another. But, as Hawtrey explained the view to me, it seemed to me to be something which might be true: namely that all wholes of which any of the constituents are good (as distinguished from beautiful etc.) have a value equal to the sum of the values of their constituents. This, I think, may be true; but it can, I think, only be true, if we admit the very opposite of the view you mention: it can, I think, only be true, if we admit that no feelings are good in themselves; since, as I have said, I think there is, in the case of every feeling, a whole composed of that feeling and some object towards which it is felt, which has a different value from that of the feeling per se. If Keynes does mean to deny what I here say, and if you think there are good reasons for the denial; or if it is some quite different point he has raised, which you think important, I should be glad to know it. Russell replied on 25 October: Many thanks for the Ethics paper and the comments. I think I accept all the important criticisms you make-only a few minor points I don't agree about. As for right, it seems the whole thing is in a muddle, and I must think it out afresh with more kinds of right. The last section is unsatisfactory, partly because, thinking it ought not be there, I took very little trouble with it. But I am still of opinion that it would be better not to have such a section. I have forgotten all about Keynes's view beyond the note on my MS. Thus I can't say whether the view Hawtrey suggested to you is really Keynes's view or not. I am not able to deny what you say about the point, and I have so far forgotten what Keynes said that until I see him again I must wait. There is no further correspondence concerning this paper. The Waterlow project seems to have petered out. In An Outline of Philosophy (1927) Russell included a chapter on ethics. In it he mentions that he was led to abandon the position of Paper 19 by Santayana's criticisms. "I now think that good and bad are derivative from desire" (238). The

216

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

theory he goes on to espouse is a form of emotivism. It is summarized in a famous line: "The good life is one inspired by love and guided by knowledge" (243). Actually he was quoting himself here; his first statement of the line is in What I Believe (1925, 28). Russell also expounded his version of emotivism in Religion and Science (1935, 235-6): When a man says, "This is good in itself," he seems to be making a statement, just as much as if he said, "This is square" or, "This is sweet." I believe this to be a mistake. I think that what the man really means is: "I wish everybody to desire this," or rather, "Would that everybody desired this." If what he says is interpreted as a statement, it is merely an affirmation of his own personal wish; if, on the other hand, it is interpreted in a general way, it states nothing, but merely desires something. The wish, as an occurrence, is personal, but what it desires is universal. It is, I think, this curious interlocking of the particular and the universal which has caused so much confusion in ethics. His final statement of his ethical position is to be found in Human Society in Ethics and Politics (1954). The copy-text is the text as it appears in Philosophical Essays (1910). It has been collated with the earlier partial publications; the results of the collations are reported in the Textual Notes.

I.

THE SUBJECT-MATTER OF ETHICS

THE STUDY OF Ethics 1 is perhaps most commonly conceived as being concerned with the questions "What sort of actions ought men to perform?" and "What sort of actions ought men to avoid?" It is conceived, that is to say, as dealing with human conduct, and as deciding what is virtuous and what vicious among the kinds of conduct between which, in practice, people are called upon to choose. Owing to this view of the province of ethics, it is sometimes regarded as the practical study, to which all others may be opposed as theoretical; the good and the true are sometimes spoken of as independent kingdoms, the former belonging to ethics, while the latter belongs to the sciences. This view, however, is doubly defective. In the first place, it overlooks the fact that the object of ethics, by its own account, is to discover true propositions about virtuous and vicious conduct, and that these are just as much a part of truth as true propositions about oxygen or the multiplication table. The aim is, not practice, but propositions about practice; and propositions about practice are not themselves practical, any more than propositions about gases are gaseous. One might as well maintain that botany is vegetable or zoology animal. Thus the study of ethics is not something outside science and coordinate with it: it is merely one among sciences. 2. In the second place, the view in question unduly limits the province of ethics. When we are told that actions of certain kinds ought to be performed or avoided, as, for example, that we ought to speak the truth, or that we ought not to steal, we may always legitimately ask for a reason, and this reason will always be concerned, not only with the actions themselves, but also with the goodness or badness of the consequences likely to follow from such actions. We shall be told that truth-speaking generates mutual confidence, cements friendships, facilitates the dispatch of business, and hence increases the wealth of the society which practises it, and so on. If we ask why we should aim at increasing mutual confidence, or cementing friendships, we may be told that obviously these things are good, or that they lead to happiness, and happiness is good. If we still ask why, the plain man will probably feel irritation, and will reply that he does not know. His irritation is due to the conflict of two feelingsthe one, that whatever is true must have a reason; the other, that the l.

rn

20

30

r What follows is largely based on Mr. G. E. Moore's Principia Ethica, to which the reader is referred for fuller discussions. Sections I and II of the following essay are reprinted from the New Quarterly, February, 19rn; Section III from the New Quarterly, May, 19rn; Section IV from the Hibbert Journal, October, 1908; and Sections v and VI from the New 40 Quarterly, September, I9IO.

217

218

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

reason he has already given is so obvious that it is merely contentious to dem~nd a. reason for the reason. In the second of these feelings he may be nght; m the first, he is certainly wrong. In ordinary life, people only ask why, when they are unconvinced. If a reason is given which they do not doubt,. they. are satisfied. Hence, when they do ask why, they usually have a logical nght to expect an answer, and they come to think that a belief for which no reason can be given is an unreasonable belief. But in this they are mistaken, as they would soon discover if their habit of asking why were more persistent. 10 It is the business of the philosopher to ask for reasons as long as reasons can legitimately be demanded, and to register the propositions which give the most ultimate reasons that are attainable. Since a proposition can only be. i:iroved by means of other propositions, it is obvious that not all proposltlo~s can be proved, for proofs can only begin by assuming something. And s~nce the consequences have no more certainty than their premisses, the thmgs that are proved are no more certain than the things that are accepted merely because they are obvious, and are then made the basis of our proofs. Thus in the case of ethics, we must ask why such and such actions ought to be performed, and continue our backward inquiry for 20 r~asons until we reach the kind of proposition of which proof is impossible, because it is so simple or so obvious that nothing more fundamental can be found from which to deduce it. 3. Now when we ask for the reasons in favour of the actions which moralists recommend, these reasons are, usually, that the consequences of t~e actions are likely to be good, or if not wholly good, at least the best possible und~~ the circumstances. Hence all questions of conduct presuppose the decision as to what things other than conduct are good and what bad. What is called good conduct is conduct which is a means to other ~hings whic? are good on their own account; and hence the study of what 30 is good on Its own account is necessary before we can decide upon rules of conduct. And the study of what is good or bad on its own account must be included in ethics, which thus ceases to be concerned only with human conduct. The first step in ethics, therefore, is to be quite clear as to what we mean by good and bad. Only then can we return to conduct and ask how right conduct is related to the production of goods and th: avoidance of evils. In this, as in all philosophical inquiries, after a preliminary analysis of complex data we proceed again to build up complex things from their simpler constituents, starting from ideas which we understand though we 40 cannot define them, and from premisses which we know though we cannot prove them. !he appearance of dogmatism in this procedure is deceptive, for the premisses are such as ordinary reasoning unconsciously assumes,

19 THE ELEMENTS OF ETHICS

219

and there is less real dogmatism in believing them after a critical scrutiny than in employing theni implicitly without examination. II. THE MEANING OF GOOD AND BAD

4. Good and Bad, in the sense in which the words are here intended (which is, I believe, their usual sense), are ideas which everybody, or almost everybody, possesses. These ideas are apparently among those which form the simplest constituents of our more complex ideas, and are therefore incapable of being analyzed or built up out of other simpler ideas. When people ask "What do you mean by Good?" the answer must consist, not in a verbal definition such as could be given if one were asked "What do you mean by Pentagon?" but in such a characterization as shall call up the appropriate idea to the mind of the questioner. This characterization may, and probably will, itself contain the idea of good, which would be a fault in a definition, but is harmless when our purpose is merely to stimulate the imagination to the production of the idea which is intended. It is in this way that children are taught the names of colours: they are shown (say) a red book, and told that that is red; and for fear they should think red means book, they are shown also a red flower, a red ball, and so on, and told that these are all red. Thus the idea of redness is conveyed to their minds, although it is quite impossible to analyze redness or to find constituents which compose it. In the case of good, the process is more difficult, both because goodness is not perceived by the senses, like redness, and because there is less agreement as to the things that are good than as to the things that are red. This is perhaps one reason that has led people to think that the notion of good could be analyzed into some other notion, such as pleasure or object of desire. A second reason, probably more potent, is the common confusion that makes people think they cannot understand an idea unless they can define it-forgetting that ideas are defined by other ideas, which must be already understood if the definition is to convey any meaning. When people begin to philosophize, they seem to make a point of forgetting everything familiar and ordinary; otherwise their acquaintance with redness or any other colour might show them how an idea can be intelligible where definition, in the sense of analysis, is impossible. 5. To explain what we mean by Good and Bad, we may say that a thing is good when on its own account it ought to exist, and bad when on its own account it ought not to exist. If it seems to be in our power to cause a thing to exist or not to exist, we ought to try to make it exist if it is good, and not exist if it is bad. When a thing is good, it is fitting that we should feel pleasure in its existence; when it is bad, it is fitting that we

10

20

30

40

220

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

19 THE ELEMENTS OF ETHICS

should feel pain in its existence. But all such characterizations really presuppose the notions of good and bad, and are therefore useful only as means of calling up the right ideas, not as logical definitions. It might be thought that good could be defined as the quality of whatever we ought to try to produce. This would merely put ought in the place of good as our ultimate undefined notion; but as a matter of fact the good is much wider than what we ought to try to produce. There is no reason to doubt that some of the lost tragedies of Aeschylus were good, but we ought not to try to re-write them, because we should certainly fail. What IO we ought to do, in fact, is limited by our powers and opportunities, whereas the good is subject to no such limitation. And our knowledge of goods is confined to the things we have experienced or can imagine; but presumably there are many goods of which we human beings have absolutely no knowledge, because they do not come within the very restricted range of our thoughts and feelings. Such goods are still goods, although human conduct can have no reference to them. Thus the notion of good is wider and more fundamental than any notion concerned with conduct; we use the notion of good in explaining what right conduct is, but we do not use the notion of right conduct in explaining what good is. 20 6. A fairly plausible view is that good means the same as desired, so that when we say a thing is good we mean that it is desired. Thus anything is good which we either hope to acquire or fear to lose. Yet it is commonly admitted that there are bad desires; and when people speak of bad desires, they seem to mean desires for what is bad. For example, when one man desires another man's pain, it is obvious that what is desired is not good but bad. But the supporter of the view that good means desired will say that nothing is good or bad in itself, but is good for one person and perhaps bad for another. This must happen, he will say, in every case of a conflict of desires; if I desire your suffering, then your suffering is good 30 for me, though it is bad for you. But the sense of good and bad which is needed in ethics is not in this way personal; and it is quite essential, in the study of ethics, to realize that there is an impersonal sense. In this sense, when a thing is good, it ought to exist on its own account, T\Ot on account of its consequences, nor yet of who is going to enjoy it. We cannot maintain that for me a thing ought to exist on its own account, while for you it ought not; that would merely mean that one of us is mistaken, since in fact everything either ought to exist or ought not. Thus the fact that one man's desire may be another man's aversion proves that good, in the sense relevant to ethics, does not mean the same as desired, since every40 thing is in itself either good or not good, and cannot be at once good for me and bad for you. This could only mean that its effects on me were good, and on you bad; but here good and bad are again impersonal.

221

7 There is another line of argument, more subtle but more instructive, by ~hich we can refute those who say that good means desii:ed, or who propose any other idea, such as pleasure, as th~ actual meaning of good. This line of argument will not prove that the thmgs that are g?od ~re not the same as the things that are desired; but it will prove t~at, if this were the case, it could not be proved by appealing to the meaning of the word " d" So far it might be thought that such an argument could only goo . ' . . . h. I have a purely logical importance. But m f~ct this 1~, not ~?· Many et ica theories have been based upon the contention that good n:eans s~-an~so, and people have accepted consequences of this contention which, if they had relied upon inspection untrammell~d by false,~heor~,' they w~uld almost certainly have rejected. Whoever believes that g~od m~ans . desired" will try to explain away the cases where it se~ms as if what is desired is bad; but if he no longer holds this theory, he will be able to allow ~ree play to his unbiased ethical perceptions, and will thus escape errors mto which he would otherwise have fallen. . The argument in question is this: If any o?e affirms that t?e go~d is the desired we consider what he says, and either assent or dissent, but in any case'our assent or dissent is decided by considering wh~t the good and the desired really are. When, on the contrary, s~me o.ne g~ves a definition of the meaning of a word, our state of mind is qmte different. If we are told "a pentagon is a figure which has five sides", ':"'e do not consider what we know about pentagons, and then agree or disagree; we ~ccept this as the meaning of the word, and we know that we are gettmg information, not about pentagons, but merely about t h. e ~ord". pentagon " · What we are told is the sort of thing that we expect d1ct1onanes to tell us. But when we are told that the good is the desired, we feel at on~e that '!'e are being told something of philosophic~! iU:p?rtan~e, somethmg which has ethical consequences, something which it is qmte beyond the scope of a dictionary to tell us. The reason of this is, that ~e already.know what we mean by the good, and what we mean by the desired; and if these two meanings always applied to the same objects, that would not be a. ~erb~l definition but an important truth. The analogue of such a propos1t10n is not the above definition of a pentagon but rather: "A pentagon (defined as above) is a figure which has five angles." Whenever a proposed de~­ nition sets us thinking whether it is true in fact, and not whether tha~ is how the word is used, there is reason to suspec: .that .we ar~ not dealing with a definition, but with a significant proposltlon, m ~h1ch the .word professedly defined has a meaning already known to us, either as s1m~le or as defined in some other way. By applying thi~ :est, we shall easily convince ourselves that all hitherto suggested defimuons of the good are significant, not merely verbal, propositions; and that therefore, though

IO

20

30

40

222

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

they may be true in fact, they do not give the meaning of the word "good". The importance of this result is that so many ethical theories depend upon the denial of it. Some have contended that "good" means "desired", others that "good" means "pleasure", others again that it means "conformity to Nature" or "obedience to the will of God". The mere fact that so many different and incompatible definitions have been proposed is evidence against any of them being really definitions; there have never been two incompatible definitions of the word "pentagon". None of the above are really definitions; they are all to be understood as substantial affirro mations concerning the things that are good. All of them are, in my opinion, mistaken in fact as well as in form, but I shall not here undertake to refute them severally. 8. It is important to realize that when we say a thing is good in itself, and not merely as a means, we attribute to the thing a property which it either has or does not have, quite independently of our opinion on the subject, or of our wishes or other people's. Most men are inclined to agree with Hamlet: "There is nothing good or bad but thinking makes it so." It is supposed that ethical preferences are a mere matter of taste, and that if X thinks A is a good thing, and Y thinks it is a bad thing, all we can 20 say is that A is good for X and bad for Y. This view is rendered plausible by the divergence of opinion as to what is good and bad, and by the difficulty of finding arguments to persuade people who differ from us in such a question. But difficulty in discovering the truth does not prove that there is no truth to be discovered. If X says A is good, and Y says A is bad, one of them must be mistaken, though it may be impossible to discover which. If this were not the case, there would be no difference of opinion between them. If, in asserting that A is good, X meant merely to assert that A had a certain relation to himself, say of pleasing his taste in some way; and if Y, in saying that A is not good, meant merely to deny that A 30 had a like relation to himself: then there would be no subject of debate between them. It would be absurd, if X said "I am eating a pigeon-pie", for Y to answer "that is false: I am eating nothing". But this is no more absurd than a dispute as to what is good, if, when we say A is good, we mean merely to affirm a relation of A to ourselves. When Christians assert that God is good, they do not mean merely that the contemplation of God rouses certain emotions in them: they may admit that this contemplation rouses no such emotion in the devils who believe and tremble, but the absence of such emotions is one of the things that make devils bad. As a matter of fact we consider some tastes better than others: we do not hold 40 merely that some tastes are ours and other tastes are other people's. We do not even always consider our own tastes the best: we may prefer bridge to poetry, but think it better to prefer poetry to bridge. And when Christians affirm that a world created by a good God must be a good world,

19 THE ELEMENTS OF ETHICS

223

they do not mean that it must be to their taste, for often it is by no means to their taste, but they use its goodness to argue that it ought to be to their taste. And they do not mean merely that it is to God's taste: for that would have been equally the case if God had not been good. Thus good and bad are qualities which belong to objects independently of our opinions, just as much as round and square do; and when two people differ as to whether a thing is good, only one of them can be right, though it may be very hard to know which is right. 9. One very important consequence of the indefinability of good must be emphasized, namely the fact that knowledge as to what things exist, have existed, or will exist, can throw absolutely no light upon the question as to what things are good. There might, as far as mere logic goes, be some general proposition to the effect "whatever exists, is good", or "whatever exists, is bad", or "what will exist is better (or worse) than what does exist". But no such general proposition can be proved by considering the meaning of "good", and no such general proposition can be arrived at empirically from experience, since we do not know the whole of what does exist, nor yet of what has existed or will exist. We cannot therefore arrive at such a general proposition, unless it is itself self-evident, or follows from some self-evident proposition, which must (to warrant the consequence) be of the same general kind. But as a matter of fact, there is, so far as I can discover, no self-evident proposition as to the goodness or badness of all that exists or has existed or will exist. It follows that, from the fact that the existent world is of such and such a nature, nothing can be inferred as to what things are good or bad. IO. The belief that the world is wholly good has, nevertheless, been widely held. It has been held either because, as a part of revealed religion, the world has been supposed created by a good and omnipotent God, or because, on metaphysical grounds, it was thought possible to prove that the sum-total of existent things must be good. With the former line of argument we are not here concerned; the latter must be briefly dealt with. The belief that, without assuming any ethical premiss, we can prove that the world is good, or indeed any other result containing the notion of good, logically involves the belief that the notion of good is complex and capable of definition. If when we say that a thing is good we mean (for example) that it has three other simpler properties, then, by proving that a thing has those three properties we prove that it is good, and thus we get a conclusion involving the notion of good, although our premisses did not involve it. But if good is a simple notion, no such inference will be possible; unless our premisses contain the notion of good, our conclusion cannot contain it. The case is analogous to the case of elements and compounds in chemistry. By combining elements or compounds we can get a new compound, but no chemical operation will give an element

10

20

30

40

224

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

w~i~h was not present in the beginning. So, if good is simple, no propos_ltl~ns not containing this notion can have consequences which do con-

tam 1t.

As a matter of fact, those who have endeavoured to prove that the world as a whole is good have usually adopted the view that all evil consists wholly in the absence of something, and that nothing positive is evil. This th~y have usually supported by defining good as meaning the same as real. Spmo~a says 2 : "B~ reality and perfection I mean the same thing"; and hence ~t follows, with much less trouble than metaphysicians have usually ro taken m the proof, that the real is perfect. This is the view in "Abt Vogler": "The evil is null, is nought, is silence implying sound." Whenever it is said that all evil is limitation, the same doctrine is involved; what is meant is that evil never consists in the existence of something which can be called bad, but only in the non-existence of something. Hence everything that does exist must be good, and the sum-total of existence, since it exists most, must be the best of all. And this view is set forth as resulting from the meaning of evil. The notion that non-existence is what is meant by evil is refuted exactly as the previous definitions of good were refuted. And the belief that, as a 20 matter of fact, nothing that exists is evil, is one which no one would advocate except a metaphysician defending a theory. Pain and hatred and envy and cruelty are surely things that exist, and are not merely the abs~nce_ of their opposites; but the theory should hold that they are indistmgmshable from the blank unconsciousness of an oyster. Indeed, it would seem that this whole theory has been advanced solely because of the u~conscious bias in favour of optimism, and that its opposite is logically JUst as tenable. We might urge that evil consists in existence and good in non-existence; that therefore the sum-total of existence i~ the wo_rst thing there is, and that only non-existence is good. Indeed, Bud30 dh1sm does seem to maintain some such view. It is plain that this view is false; but logically it is no more absurd than its opposite. I I. We cannot, then, infer any results as to what is good or bad from a stu?y of the things that exist. This conclusion needs chiefly, at the present time, to be applied against evolutionary ethics. The phrase "survival of the fittest" seems to have given rise to the belief that those who survive ar~ the fittest in some ethical sense, and that the course of evolution gives ev1de~ce that the later type is better than the earlier. On this basis, a wo~sh1p of force is easily set up, and the mitigation of struggle by civilization comes to be deprecated. It is thought that what fights most suc40 cessfully is most admirable, and that what does not help in fighting is worthless. Such a view 1s wholly destitute of logical foundation. The 2

Ethics, Pt.

II,

Df.

VI.

19 THE ELEMENTS OF ETHICS

225

course of nature, as we have seen, is irrelevant in deciding as to what is good or bad. A priori, it would be as probable that evolution should go from bad to worse, as that it should go from good to better. What makes the view plausible is the fact that the lower animals existed earlier than the higher, and that among men the civilized races are able to defeat and often exterminate the uncivilized. But here the ethical preference of the higher to the lower animals, and of the exterminators to the exterminated, is not based upon evolution, but exists independently, and unconsciously intrudes into our judgment of the evolutionary process. If evolutionary ethics were sound, we ought to be entirely indifferent as to what the course of evolution may be, since whatever it is is thereby proved to be the best. Yet if it should turn out that the negro or the Chinaman was able to oust the European, we should cease to have any admiration of evolution; for as a matter of fact our preference of the European to the negro is wholly independent of the European's greater prowess with the Maxim gun. Broadly, the fact that a thing is unavoidable affords no evidence that it is not an evil; and the fact that a thing is impossible affords no evidence that it is not a good. It is doubtless foolish, in practice, to fret over the inevitable; but it is false, in theory, to let the actual world dictate our standard of good and evil. It is evident that among the things that exist some are good, some bad, and that we know too little of the universe to have any right to an opinion as to whether the good or the bad preponderates, or as to whether either is likely in the future to gain on the other. Optimism and pessimism alike are general theories as to the universe which there is no reason whatever for accepting; what we know of the world tends to suggest that the good and the evil are fairly balanced, but it is of course possible that what we do not know is very much better or very much worse than what we do know. Complete suspense of judgment in this matter is therefore the only rational attitude. III. RIGHT AND WRONG 12. The ideas of right and wrong conduct are, as we have seen, those with which ethics is generally supposed to be most concerned. This view, which is unduly narrow, is fostered by the use of the one word good, both for the sort of conduct which is right, and for the sort of things which ought to exist on account of their intrinsic value. This double use of the word good is very confusing, and tends greatly to obscure the distinction of ends and means. I shall therefore speak of right actions, not of good actions, confining the word good to the sense explained in Section II. The word "right" is very ambiguous, and it is by no means easy to distinguish the various meanings which it has in common parlance. Owing to the variety of these meanings, adherence to any one necessarily involves

ro

20

30

40

226

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

us in apparent paradoxes when we use it in a context which suggests one of the other meanings. This is the usual result of precision of language; but so long as the paradoxes are merely verbal, they do not give rise to more than verbal objections. In judging of conduct we find at the outset two widely divergent methods, of which one is advocated by some moralists, the other by others, while both are practised by those who have no ethical theory. One of these methods, which is that advocated by utilitarians, judges the rightness of an act by relation to the goodness or badness of its consequences. The other method, advocated by intuitionists, judges by the approval or disapproval of the moral sense or conscience. I believe that it is necessary to combine both theories in order to get a complete account of right and wrong. There is, I think, one sense in which a man does right when he does what will probably have the best consequences, and another in which he does right when he follows the dictates of his conscience, whatever the probable consequences may be. (There are many other senses which we may give to the word right, but these two seem to be the most important.) Let us begin by considering the second of these senses. I3. The question we have to ask ourselves is: What do we mean by the dictates of the moral sense? If these are to afford a definition of right conduct, we cannot say that they consist in judging that such and such acts are right, for that would make our definition circular. We shall have to say that the moral sense consists in a certain specific emotion of approval towards an act, and that an act is to be called right when the agent, at the moment of action, feels this emotion of approval towards the action which he decides to perform. There is certainly a sense in which a man ought to perform any act which he approves, and to abstain from any act which he disapproves; and it seems also undeniable that there are emotions which may be called approval and disapproval. Thus this theory, whether adequate or not, must be allowed to contain a part of the truth. It is, however, fairly evident that there are other meanings of right conduct, and that, though there is an emotion of approval, there is also a judgment of approval, which may or may not be true. For we certainly hold that a man who has done an action which his conscience approved may have been mistaken, and that in some sense his conscience ought not to have approved his action. But this would be impossible if nothing were involved except an emotion. To be mistaken implies a judgment; and thus we must admit that there is such a thing as a judgment of approval. If this were not the case we could not reason with a man as to what is right; what he approves would be necessarily right for him to do, and there could be no argument against his approval. We do in fact hold that when one man approves of a certain act, while another disapproves, one of them is mistaken, which would not be the case with a mere emotion. If one man likes

19 THE ELEMENTS OF ETHICS

227

oysters and another dislikes them, we do not say that either of them is mistaken. . Thus there is a judgment of approval, 3 and this must consist of a JUd~ment that an act is, in a new sense, right. The judgment of approval is not merely the judgment that we feel the emotion of approval, for then another who disapproved would not necessarily hold our judgment of approval to be mistaken. Thus in order to give a ~eaning to the judgment of approval, it is necessary to admit a sense of right ot~e~ th~n approved. In this sense when we approve an act we judge that it is right, and we may be mist~ken in so judging. This new sense i~ objective, in the sense IO that it does not depend upon the opinions and feelings of the a~ent ..Thus a man who obeys the dictates of his conscience is not always actmg rightly in the objective sense. When a man does what his conscience a~proves, he does what he believes to be objectively right, but not necessarily what is objectively right. We need, therefore, some other criterion than the moral sense for judging what is objectively right. . It is in defining objective rightness that the consequences of an ac14 tion become relevant. Some moralists, it is true, deny the dep~nde~ce upon consequences; but that is to be attributed, I think, to confus10n with the subjective sense. When people argue as to whether such and such an 20 action is right, they always adduce the consequences which it .has or i:nay be expected to have. A statesman who ha~ to de~ide what i.s the :ight policy, or a teacher who has to decide what is the.rig~t e~ucauon, will be expected to consider what policy or what educatio? is hkely to have the best results. Whenever a question is at all complicated, and cannot be settled by following some simple rule, such as "th?u shalt not steal'.''. or "thou shalt not bear false witness", it is at once evident that the decis10n cannot be made except by consideration of consequences. But even when the decision can be made by a simple precept, such as not to lie or not to steal, the justification of the precept is found only by 30 consideration of consequences. A code such as the Decalogue, it must be admitted, can hardly be true without exception if the goodness or ba~ness of consequences is what determines the rightness wrongness of actions; for in so complex a world it is unlikely that obedience to the Dec~lo?ue will always produce better consequences than disobedience. Yet 1t is a

o:

3

The judgment of approval does not always coincide .with the emotion o~ approval. For example, when a man has been led by his reason to re1ect a moral ~ode which he formerly held, it will commonly happen, at least for a time, that his emot10n of approval follows the old code, though his judgment has abandoned it. Thus he may have been brought up, like Mohammed's first disciples, to believe it a duty to av~nge the murder of relations 40 by murdering the murderer or his relations; and he may contmue.tofeel approval of such vengeance after he has ceased to judge it approvingly. The emotion of approval will not be again in question in what follows.

228

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

suspicious circumstance that breaches of those of the Ten Commandments which people still hold it a duty to obey do, as a matter of fact, have bad consequences in the vast majority of instances, and would not be considered wrong in a case in which it was fairly certain that their consequences would be good. This latter fact is concealed by a question-begging addition of moral overtones to words. Thus, e.g. "thou shalt do no murder" wo~ld be an important precept if it were interpreted, as Tolstoy interprets it, to mean "thou shalt not take human life". But it is not so interpreted; on the contrary, some taking of human life is called "justiro fiable homicide". Thus murder comes to mean "unjustifiable homicide"· and it is a mere tautology to say, "Thou shalt do no unjustifiable homi~ cide." That this should be announced from Sinai would be as fruitless as !Jamlet's report of the ghost's message: "There's ne'er a villain, dwelling m all Denmark, but he's an arrant knave." As a matter of fact, people do make a certain classification of homicides, and decide that certain kinds are justifiable and certain others unjustifiable. But there are many doubtful cases: tyrannicide, capital punishment, killing in war, killing in selfdefence, killing in defence of others, are some of these. And if a decision is sought, it is sought usually by considering whether the consequences 20 of actions belonging to these classes are on the whole good or bad. Thus the importance of precepts such as the Ten Commandments lies in the fact that they give simple rules, obedience to which will in almost all cases have better consequences than disobedience; and the justification of the rules is not wholly independent of consequences. 15. In common language the received code of moral rules is usually presupposed, and an action is only called immoral when it infringes one of these rules. Whatever does not infringe them is regarded as permissible, so that on most of the occasions of life no one course of action is marked out as alone right. If a man adopts a course of action which, though not 30 contrary to the received code, will probably have bad consequences he is called unwise rather than immoral. Now, according to the distincti~n we hav~ made between objective and subjective rightness, a man may well act m a ~ay which objectively wrong without doing what is subjectively wrong, i.e. what ~1s conscience disapproves. An act (roughly speaking, I shall return to this point presently) is immoral when a man's conscience disapproves it, but is judged only unwise or injudicious when his conscience approves it, although we judge that it will probably have bad consequences .. Now the usual moral code is supposed, in common language, t~ be a_dm1~ted by every man's conscience, so that when he infringes it, 40 his act10n is not merely injudicious, but immoral; on the other hand, where the ~od~ is silent, we regard an unfortunate action as objectively but not subiectlvely wrong, i.e. as injudicious, but not immoral. The acceptance of a moral code has the great advantage that, in so far as its rules

!s

19 THE ELEMENTS OF ETHICS

229

are objectively right, it tends to harmonize objective and subjective rightness. Thus it tends to cover all frequent cases, leaving only the rarer ones to the individual judgment of the agent. Hence when new sorts of cases become common, the moral code soon comes to deal with them; thus each profession has its own code concerning cases common in the profession, though not outside it. But the moral code is never itself ultimate; it is based upon an estimate of probable consequences, and is essentially a method of leading men's judgment to approve what is objectively right and disapprove what is objectively wrong. And when once a fairly correct code is accepted, the exceptions to it become very much fewer than they would otherwise be, because one of the consequences of admitting exceptions is to weaken the code, and this consequence is usually bad enough to outweigh the good resulting from admitting such and such an exception. This argument, however, works in the opposite direction with a grossly incorrect code; and it is to be observed that most conventional codes embody some degree of unwarrantable selfishness, individual, professional, or national, and are thus in certain respects worthy of detestation. 16. What is objectively right, then, is in some way dependent on consequences. The most natural supposition to start from would be that the objectively right act, under any circumstances, is the one which will have the best consequences. We will define this as the most fortunate act. The most fortunate act, then, is the one which will produce the greatest excess of good over evil, or the least excess of evil over good (for there may be situations in which every possible act will have consequences that are on the whole bad). But we cannot maintain that the most fortunate act is always the one which is objectively right, in the sense that it is what a wise man will hold that he ought to do. For it may happen that the act which will in fact prove the most fortunate is likely, according to all the evidence at our disposal, to be less fortunate than some other. In such a case, it will be, at least in one sense, objectively wrong to go against the evidence, in spite of the actual good result of our doing so. There have certainly been some men who have done so much harm that it would have been fortunate for the world if their nurses had killed them in infancy. But if their nurses had done so their action would not have been objectively right, because the probability was that it would not have the best effects. Hence it would seem we must take account of probability in judging of objective rightness; let us then consider whether we can say that the objectively right act is the one which will probably be most fortunate. I shall define this as the wisest act. The wisest act, then, is that one which, when account is taken of all available data, gives us the greatest expectation of good on the balance, or the least expectation of evil on the balance. There is, of course, a dif-

ro

20

30

40

230

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

fi~u~ty a~ to :Vhat ar~ to be considered available data; but broadly we can distmgmsh, i_n any g~ven state of knowledge, things capable of being foreseen from thmgs which are unpredictable. I suppose account to be taken of the general body of current knowledge, in fact the sort of consideration which people expect when they ask legal or medical advice. There is no doubt this brings us nearer to what is objectively right than we were when :ve :vere considering the actually most fortunate act. For one thing, it JUs~ifie~ the unavoidable limitation to not very distant consequences, which is almost always necessary if a practical decision is to be reached. IO For the _likeliho?d ~f error in calculating distant consequences is so great tha~ thelf ~ont~ibution to the probable good or evil is very small, though thelf contribution to the actual good or evil is likely to be much greater than that of the nearer consequences. And it seems evident that what it is quite impossible to know cannot be relevant in judging as to what conduct is ri~ht. If, as is possible, a cataclysm is going to destroy life on this planet this day week, many acts otherwise useful will prove to have been wasted labour, for example, the preparation of next year's Nautical Almanac; but since we have no reason to expect such a cataclysm, the rightness or wrongness of acts is plainly to be estimated without regard to it. 20 I7. One apparent objection at once suggests itself to the definition. Very few acts are of sufficient importance to justify such elaborate and careful consideration as is required for forming an opinion as to whether they are the wisest. Indeed, the least important decisions are often those which it would be hardest to make on purely reasonable grounds. A man who debates on each day which of two ways of taking exercise is likely to prove most beneficial is considered absurd; the question is at once difficult and unimportant, and is therefore not worth spending time over. But although it is true that unimportant decisions ought not to be made with ~xc~ssive care, there is danger of confusion if this is regarded as an ob30 1ect10n to our definition of objective rightness. For the act which, in the case supposed, is objectively wrong is the act of deliberation not the act decided upon as the result of deliberation. And the deliber~tion is condemned by our definition, for it is very unlikely that there is no more beneficial way of spending time than in debating trivial points of conduct. Th~s, although the wisest act is the one which, after complete investi~at10n_, a~pears li~ely to give the most fortunate results, yet the complete ~nvestlgation reqmred to show that it is the wisest act is only itself wise m ~he case of ~ery important decisions. This is only an elaborate way of sa~mg that a wise man will not waste time on unimportant details. Hence 40 this apparent objection can be answered. . I 8. One further addition is required for the definition of the objectively nght act, namely that it must be possible. Among the acts whose consequences are to be considered we must not include such as are either physi-

19 THE ELEMENTS OF ETHICS

231

cally impossible to perform or impossible for the agent to think of. This last condition introduces difficulties connected with Determinism, which are discussed in Section IV. Ignoring these difficulties, we may say that the objectively right act is that one which, of all that are possible, will probably have the best consequences. I9. We must now return to the consideration of subjective rightness, with a view to distinguishing conduct which is merely mistaken from conduct which is immoral or blameworthy. We here require a new sense of ought, which it is by no means easy to define. In the objective sense, a man ought to do what is objectively right. But in the subjective sense, which we have now to examine, he sometimes ought to do what is objectively wrong. For example, we saw that it is often objectively right to give less consideration to an unimportant question of conduct than would be required for forming a trustworthy judgment as to what is objectively right. Now it seems plain that if we have given to such a question the amount and kind of consideration which is objectively right, and we then do what appears to us objectively right, our action is, in some sense, subjectively right, although it may be objectively wrong. Our action could certainly not be called a sin, and might even be highly virtuous, in spite of its objective wrongness. It is these notions of what is sinful and what is virtuous that we have now to consider. 20. The first suggestion that naturally occurs is that an act is subjectively right when it is judged by the agent to be objectively right, and subjectively wrong when it is judged to be objectively wrong. I do not mean that it is subjectively right when the agent judges that it is the act which, of all that are possible, will probably have the best results; for the agent may not accept the above account of objective rightness. I mean merely that it is the one towards which he has the judgment of approval. A man may judge an act to be right without judging that its consequences will be probably the best possible; I only contend that, when he truly judges it to be right, then its consequences will probably be the best possible. But his judgment as to what is objectively right may err, not only by a wrong estimate of probable consequences, or by failing to think of an act which he might have thought of, but also by a wrong theory as to what constitutes objective rightness. In other words, the definition I gave of objective rightness is not meant as an analysis of the meaning of the word, but as a mark which in fact attaches to all objectively right actions and to no others. We are to consider then the suggestion that an act is moral when the agent approves it and immoral when he disapproves it; using moral to mean subjectively right and immoral to mean subjectively wrong. This suggestion, it is plain, will not stand without much modification. In the first place, we often hold it immoral to approve some things and disap-

IO

20

30

40

232

ro

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

prove others, unless there are special circumstances to excuse such approval or disapproval. In the second place, unreflecting acts, in which there is no judgment either of approval or disapproval, are often moral or immoral. For both these reasons the suggested definition must be regarded as inadequate. 2I. The doctrine that an act is never immoral when the agent thinks it right has the drawback (or the advantage) that it excuses almost all the acts which would be commonly condemned. Very few people deliberately do what, at the moment, they believe to be wrong; usually they first argue themselves into a belief that what they wish to do is right. They decide that it is their duty to teach so-and-so a lesson, that their rights have been so grossly infringed that if they take no revenge there will be an encouragement to injustice, that without a moderate indulgence in pleasure a character cannot develop in the best way, and so on and so on. Yet we do not cease to blame them on that account. Of course it may be said that a belief produced by a course of self-deception is not a genuine belief, and that the people who invent such excuses for themselves know all the while that the truth is the other way. Up to a point this is no doubt true, though I doubt if it is always true. There are, however, other cases of mistaken judgment as to what is right, where the judgment is certainly genuine, and yet we blame the agent. These are cases of thoughtlessness, where a man remembers consequences to himself, but forgets consequences to others. In such a case he may judge correctly and honestly on all the data that he remembers, yet if he were a better man he would remember more data. Most of the actions commonly condemned as selfish probably come under this head. Hence we must admit that an act may be immoral, even if the agent quite genuinely judges that it is right. Unreflecting acts, again, in which there is no judgment as to right or wrong, are often praised or blamed. Acts of generosity, for example, are more admired when they are impulsive than when they result from reflection. I cannot think of any act which is more blamed when it is impulsive than when it is deliberate; but certainly many impulsive acts are blamed-for example, such as spring from an impulse of malice or cruelty. 22. In all these cases where reflection is absent, and also in the case of inadequate reflection, it may be said that blame does not belong properly to the act, but rather to the character revealed by the act, or, if to some acts, then to those previous deliberate acts by which the character has been produced which has resulted in the present act. The cases of self-deception would then be dismissed on the ground that the self-deceiver never really believes what he wishes to believe. We could then retain our original definition, that a moral act is one which the agent judges to be right, while an immoral one is one which he judges to be wrong. But I do not think this would accord with what most people really mean. I rather think that

19 THE ELEMENTS OF ETHICS

233

moral act should be defined as one which the agent would .have judged

~o be right if he had considered the question candidly ~nd "'.nh du~ care;

if, that is to say, he had examined t~e dat~ before hi~ with a view to discovering what was right, and not with a view to provmg such-a~d-such a course to be right. If an act is unimportant, and at the same tu~e n~t obviously less right than some obvious alternative, we shall consider it neither moral nor immoral; for in such a case the act .~oes not deserve careful consideration. The amount of care which a decision deserves depends upon its importance and difficulty; in the case of a statesma~ advocating a new policy, for example, years of delibe~ation may. sometn~es 10 be necessary to excuse him from the charge of levny. But with less i.mportant acts, it is usually right to decide even when furthe~ reflectl~n might show the present decision to be erroneous. 1:hus there. is a certam amount of reflection appropriate to various acts, while some nght acts ~re best when they spring from impulse (though these are such as r~fl~ction would approve). We may therefore say that an act is moral.when 1t is one which the agent would judge to be right after an appropriate amount of candid thought, or, in the case of acts which are best ."'.hen they are unreflecting after the amount and kind of thought reqmsne to form a first opinion. An act is immoral when the agent would judge it to b~ wrong 2 0 after an appropriate amount of reflection. It is neither moral nor immoral when it is unimportant and a small amount of reflection would not suffice to show whether it was right or wrong. 23 . We may now sum up our discussion of right and wrong. When .a man asks himself: "What ought I to do?" he is asking what conduct is right in an objective sense. He cannot mean: "What ou~~t a p~rso~ to do who holds my views as to what a person ought t~ do? .for his views as to what a person ought to do are what will constitute his a.nswer to the question "What ought I to do?" But the onlooker, who thmks that t~e man has answered this question wrongly, may nevertheless hold ~hat? m 30 acting upon his answer, the man was acting rightly in a s~cond, sub1ecuve, sense. This second sort of right action we call moral action. We held that an action is moral when the agent would judge it to be right a~ter an appropriate amount of candid thought, or after. a small amount. m the case of acts which are best when they are unreflectmg; the ap?ropnate amount of thought being dependent upon the difficulty and the importance o~ the decision. And we held that an action is right when, of all that are possible, it is the one which will probably have the best results .. There ar~ many other meanings of right, but these seem to be the meanmgs reqmred .for answering the questions: "What ought I to do?" and "What acts are im- 40 moral?"

234

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

19 THE ELEMENTS OF ETHICS

IV. DETERMINISM AND MORALS

As the outcome of our discussions in the previous section, I shall assume the following definitions. The objectively right action, in any circumstances, is that action which, of all that are possible, gives us, when account is taken of all available data, the greatest expectation of probable good effects, or the least expectation of probable bad effects. The subjectively right or moral action is that one which will be judged by the agent to be objectively right if he devotes to the question an appropriate amount of candid thought, or, in the case of actions that ought to be impulsive, a small amount. The appropriate amount of thought depends upon the importance of the action and the difficulty of the decision. An act is neither moral nor immoral when it is unimportant, and a small amount of reflection would not suffice to show whether it was right or wrong. After these preliminaries, we can pass to the consideration of our main topic. 25. The principle of causality-that every event is determined by previous events, and can (theoretically) be predicted when enough previous events are known-appears to apply just as much to human actions as to other events. It cannot be said that its application to human actions, or to any other phenomena, is wholly beyond doubt; but a doubt extending to the principle of causality must be so fundamental as to involve all science, all everyday knowledge, and everything, or almost everything, that we believe about the actual world. If causality is doubted, morals collapse, since a right action, as we have seen, is one of which the probable effects are the best possible, so that estimates of right and wrong necessarily presuppose that our actions can have effects, and therefore that the law of causality holds. In favour of the view that human actions alone are not the effects of causes, there appears to be no ground whatever except the sense of spontaneity. But the sense of spontaneity only affirms that we can do as we choose, and choose as we please, which no determinist denies; it cannot affirm that our choice is independent of all motives, 4 and indeed introspection tends rather to show the opposite. It is said by the advocates of free-will5 that determinism destroys morals, since it shows that all our actions are inevitable, and that therefore they deserve neither praise nor blame. Let us consider how far, if at all, this is the case. 26. The part of ethics which is concerned, not with conduct, but with the meaning of good and bad, and the things that are intrinsically good and bad, is plainly quite independent of free-will. Causality belongs to the description of the existing world, and we saw that no inference can be drawn from what exists to what is good. Whether, then, causality holds

24. The importance to ethics of the free-will question is a subject upon which there has existed almost as much diversity of opinion as on the freewill question itself. It has been urged by advocates of free-will that its denial involves the denial of merit and demerit, and that, with the denial of these, ethics collapses. It has been urged on the other side that, unless we can foresee, at least partially, the consequences of our actions, it is impossible to know what course we ought to take under any given circumstances; and that if other people's actions cannot be in any degree predicted, the foresight required for rational action becomes impossible. I do not propose, in the following discussion, to go into the free-will controversy itself. The grounds in favour of determinism appear to me overwhelming, and I shall content myself with a brief indication of these grounds. The question I am concerned with is not the free-will question itself, but the question how, if at all, morals are affected by assuming determinism. In considering this question, as in most of the other problems of ethics, the moralist who has not had a philosophical training appears to me to go astray, and become involved in needless complications, through supposing that right and wrong in conduct are the ultimate conceptions of ethics, rather than good and bad, in the effects of conduct and in other things. The words good and bad are used both for the sort of conduct which is right or wrong, and for the sort of effects to be expected from right and wrong conduct, respectively. We speak of a good picture, a good dinner, and so on, as well as of a good action. But there is a great difference between these two meanings of good. Roughly speaking, a good action is one of which the probable effects are good in the other sense. It is confusing to have two meanings for one word, and we therefore agreed in the previous section to speak of a right action rather than a good action. In order to decide whether an action is right, it is necessary, as we have seen, to consider its probable effects. If the probable effects are, on the whole, better than those of any other action which is possible under the circumstances, then the action is right. The things that are good are things which, on their own account, and apart from any consideration of their effects, we ought to wish to see in existence: they are such things as, we may suppose, might make the world appear to the Creator worth creating. I do not wish to deny that right conduct is among the things that are good on their own account; but if it is so, it depends for its intrinsic goodness upon the goodness of those other things which it aims at producing, such as love or happiness. Thus the rightness of conduct is not the fundamental conception upon which ethics is built up. This fundamental conception is intrinsic goodness or badness.

235

IO

20

30

4 A motive means merely a cause of volition. 5 I use free-will to mean the doctrine that not all volitions are determined by causes, which 40 is the denial of determinism. Free-will is often used in senses compatible with determinism, but I am not concerned to affirm or deny it in such senses.

236

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

always, sometimes, or never is a question wholly irrelevant in the consideration of intrinsic goods and evils. But when we come to conduct and the notion of ought, we cannot be sure that determinism makes no difference. For we saw that the objectively right action may be defined as that one which, of all that are possible under the circumstances, will probably on the whole have the best consequences. The action which is objectively right must therefore be in some sense possible. But if determinism is true, there is a sense in which no action is possible except the one actually performed. Hence, if the two senses of possibility are the same, the action actually performed is always objectively right; for it is the only possible action, and therefore there is no other possible action which would have had better results. There is here, I think, a real difficulty. But let us consider the various kinds of possibility which may be meant. In order that an act may be a possible act, it must be physically possible to perform, it must be possible to think of, and it must be possible to choose if we think of it. Physical possibility, to begin with, is obviously necessary. There are circumstances under which I might do a great deal of good by running from Oxford to London in five minutes. But I should not be called unwise, or guilty of an objectively wrong act, for omitting to do so. We may define an act as physically possible when it will occur if I will it. Acts for which this condition fails are not to be taken account of in estimating rightness or wrongness. 27. To judge whether an act is possible to think of is more difficult, but we certainly take account of it in judging what a man ought to do. There is no physical impossibility about employing one's spare moments in writing lyric poems better than any yet written, and this would certainly be a more useful employment than most people find for their spare moments. But we do not blame people for not writing lyric poems unless, like FitzGerald, they are people that we feel could have written them. And not only we do not blame them, but we feel that their action may be objectively as well as subjectively right if it is the wisest that they could have thought of. But what they could have thought of is not the same as what they did think of. Suppose a man in a fire or a shipwreck becomes so panic-stricken that he never for a moment thinks of the help that is due to other people, we do not on that account hold that he does right in only thinking of himself. Hence in some sense (though it is not quite clear what this sense is), some of the courses of action which a man does not think of are regarded as possible for him to think of, though others are admittedly impossible. There is thus a sense in which it must be possible to think of an action, if we are to hold that it is objectively wrong not to perform the action. There is also, if determinism is true, a sense in which it is not possible to think of any action except those which we do think of. But it is ques-

19 THE ELEMENTS OF ETHICS

237

tionable whether these two senses of possibility are the same. A man who finds th~t his house is .on fire may run out of it in a panic without thinking of w~rnmg th~ other mmate,s; but we feel, rightly or wrongly, that it was possible for him to think of warning them in a sense in which it is not possible for a prosaic person to think of a lyric poem. It may be that we are wrong .in feeling this difference, and that what really distinguishes the two. cases is dep~ndence up~n past decisions. That is to say, we may recogmze that no different ch01ce among alternatives thought of at any time would have turned an ordinary man into a good lyric poet; but that most m~n, by suitably choosing among alternatives actually thought of, can acqmre t.he sort of cha~acter which will lead them to remember their neighbours ma fire. And if a man engages in some useful occupation of which a natural effect is to destroy his nerve, we may conceivably hold that this excuses his panic in an emergency. In such a point it would seem that our judgment may really be dependent on the view w~ take as to the existence of free-will; for the believer in free-will cannot allow any such excuse. If we try to state the difference we feel between the case of the lyric poems and the case of the fire, it seems to come to this: that we do not hold an act objectively wrong when it would have required what we recognize as a special aptitude in order to think of a better act and when we believe that the agent did not possess this aptitude. But this distinction seems to imply that there is not such a thing as a special aptitude for this or that virtue; a view which cannot, I think, be maintained. An aptitude for generosity or for kindness may be as much a natural gift as an aptitude f?r poetry; and an aptitude for poetry may be as much improved by practice as an aptitude for kindness or generosity. Thus it would seem that there is no sense in which it is possible to think of some actions which in fact we do not think of, but impossible to think of others, except the sense that the ones we regard as possible would have been thought of if a different choice among alternatives actually thought of had been made on some previous occasion. We shall then modify our previous definition of the objectively right action by saying that it is the probably most beneficial among those that occur to the agent at the moment of choice. But we shall hold that, in certain cases, the fact that a more beneficial alternative does not occur to him is evidence of a wrong choice on some previous occasion. 28. But since occasions of choice do often arise, and since there certainly is a sense in which it is possible to choose any one of a number of different actions which we think of, we can still distinguish some actions as ~ight and some as wrong. Our previous definitions of objectively right act10~s and of i:ioral a~tions still hold, with the modification that, among physically possible actions, only those which we actual/,y think of are to be ~e~arded ~s possible. When several alternative actions present themselves, it is certam that we can both do which we choose, and choose which we

IO

20

30

40

238

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

19 THE ELEMENTS OF ETHICS

will. In this sense all the alternatives are possible. What determinism maintains is, that our will to choose this or that alternative is the effect of antecedents; but this does not prevent our will from being itself a cause of other effects. And the sense in which different decisions are possible seems sufficient to distinguish some actions as right and some as wrong, some as moral and some as immoral. Connected with this is another sense in which, when we deliberate, either decision is possible. The fact that we judge one course objectively right may be the cause of our choosing this course: thus, before we have decided as to which course we think right, either is possible in the sense that either will result from our decision as to which we think right. This sense of possibility is important to the moralist, and illustrates the fact that determinism does not make moral deliberation futile. 29. Determinism does not, therefore, destroy the distinction of right and wrong; and we saw before that it does not destroy the distinction of good and bad: we shall still be able to regard some people as better than others, and some actions as more right than others. But it is said that praise and blame and responsibility are destroyed by determinism. When a madman commits what in a sane man we should call a crime, we do not blame him, partly because he probably cannot judge rightly as to consequences, but partly also because we feel that he could not have done otherwise: if all men are really in the position of the madman, it would seem that all ought to escape blame. But I think the question of choice really decides as to praise and blame. The madman, we believe (excluding the case of wrong judgment as to consequences), did not choose between different courses, but was impelled by a blind impulse. The sane man who (say) commits a murder has, on the contrary, either at the time of the murder or at some earlier time, chosen the worst of two or more alternatives that occurred to him; and it is for this we blame him. It is true that the two cases merge into each other, and the madman may be blamed if he has become mad in consequence of vicious self-indulgence. But it is right that the two cases should not be too sharply distinguished, for we know how hard it often is in practice to decide whether people are what is called "responsible for their actions". It is sufficient that there is a distinction, and that it can be applied easily in most cases, though there are marginal cases which present difficulties. We apply praise or blame, then, and we attribute responsibility, where a man, having to exercise choice, has chosen wrongly; and this sense of praise or blame is not destroyed by determinism. 30. Determinism, then, does not in any way interfere with morals. It is worth noticing that free-will, on the contrary, would interfere most seriously, if anybody really believed in it. People never do, as a matter of fact, believe that any one else's actions are not determined by motives,

however much they may think themselves free. Bradshaw consists entirely of predictions as to the actions of engine-drivers; but no one doubts Bradshaw on the ground that the volitions of engine-drivers are not governed by motives. If we really believed that other people's actions did not have causes, we could never try to influence other people's actions; for such influence can only result if we know, more or less, what causes will produce the actions we desire. If we could never try to influence other people's actions, no man could try to get elected to Parliament, or ask a woman to marry him: argument, exhortation, and command would become mere idle breath. Thus almost all the actions with which morality is concerned would become irrational, rational action would be wholly precluded from trying to influence people's volitions, and right and wrong would be interfered with in a way in which determinism certainly does not interfere with them. Most morality absolutely depends upon the assumption that volitions have causes, and nothing in morals is destroyed by this assumption. Most people, it is true, do not hold the free-will doctrine in so extreme a form as that against which we have been arguing. They would hold that most of a man's actions have causes, but that some few, say one per cent, are uncaused spontaneous assertions of will. If this view is taken, unless we can mark off the one per cent of volitions which are uncaused, every inference as to human actions is infected with what we may call one per cent of doubt. This, it must be admitted, would not matter much in practice, because, on other grounds, there will usually be at least one per cent of doubt in predictions as to human actions. But from the standpoint of theory there is a wide difference: the sort of doubt that must be admitted in any case is a sort which is capable of indefinite diminution, while the sort derived from the possible intervention of free-will is absolute and ultimate. In so far, therefore, as the possibility of uncaused volitions comes in, all the consequences above pointed out follow; and in so far as it does not come in, determinism holds. Thus one per cent of free-will has one per cent of the objectionableness of absolute free-will, and has also only one per cent of the ethical consequences. In fact, however, no one really holds that right acts are uncaused. It would be a monstrous paradox to say that a man's decision ought not to be influenced by his belief as to what is his duty; yet, if he allows himself to decide on an act because he believes it to be his duty, his decision has a motive, i.e. a cause, and is not free in the only sense in which the determinist must deny freedom. It would seem, therefore, that the objections to determinism are mainly attributable to misunderstanding of its purport. Hence, finally, it is not determinism but free-will that has subversive consequences. There is therefore no reason to regret that the grounds in favour of determinism are overwhelmingly strong.

239

IO

20

30

40

240

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

19 THE ELEMENTS OF ETHICS

V. EGOISM

attainment will give me more or less pleasure, and its non-attainment will give me more or less pain. Hence it is inferred that I desire it on account of the pleasure it would give me, and not on its own account. But this is to put the cart before the horse. The pleasure we get from things usually depends upon our having had a desire which they satisfy; the pleasures of eating and drinking, for example, depend upon hunger and thirst. Or take, again, the pleasure people get from the victory of their own party in a contest. Other people would derive just the same pleasure from the victory of the opposite party; in each case the pleasure depends for its existence upon the desire, and would not exist if the desire had not existed. Thus we cannot say that people only desire pleasure. They desire all kinds of things, and pleasures come from desires much oftener than desires from imagined pleasures. Thus the mere fact that a man will derive some pleasure from achieving his object is no reason for saying that his desire is self-centred. 33. Such arguments are necessary for the refutation of those who hold it to be obvious a priori that every man must always pursue his own good exclusively. But, as is often the case with refutations of a priori theories, there is an air of logic-chopping about a discussion as to whether desire or the pleasure expected from its satisfaction ought to have priority. Let us leave these questions, and consider whether, as a matter of fact, people's actions can be explained on the egoistic hypothesis. The most obvious instances to the contrary are, of course, cases of self-sacrificeof men to their country, for example, or of parents to children. But these instances are so obvious that the egoistic theory is ready with an answer. It will maintain that, in such cases, the people who make the sacrifice would not be happy if they did not make it, that they desire the applause of men or of their own consciences, that they find in the moment of sacrifice an exaltation which realizes their highest self, etc. etc. etc. Let us examine these arguments. It is said that the people in question would not be happy if they did not make the sacrifice. This is often false in fact, but we may let that pass. Why would they not be happy? Either because others would think less well of them, or because they themselves would feel pangs of conscience, or because they genuinely desired the object to be attained by their sacrifice and could not be happy without it. In the last case they have admittedly a desire not centred in self; the supposed effect upon their happiness is due to the desire, and would not otherwise exist, so that the effect upon happiness cannot be brought into account for the desire. But if people may have desires for things that lie outside their ego, then such desires, like others, may determine action, and it is possible to pursue an object which is not "my" good in any sense except that I desire and pursue it. Thus, in all cases of self-sacrifice, those who hold the egoistic theory will have to maintain that the outside end secured by the self-

31. We have next to consider an objection to the view that objective rightness consists in probably having the best consequences on the whole. The objection I mean is that of egoism: that a man's first duty is to himself, and that to secure his own good is more imperative than to secure other people's. Extensions of this view are, that a man should prefer the interest of his family to that of strangers, of his countrymen to that of foreigners, or of his friends to that of his enemies. All these views have in common the belief that, quite apart from practicability, the ends which one man IO ought to pursue are different from those which another man ought to pursue. Egoism has several different meanings. It may mean that every man is pyschologically bound to pursue his own good exclusively; it ma~ mea~ that every man will achieve the best result on the whole by pursumg his own good; it may mean that his own good is the only thing a man ought to think good; and it may mean, lastly, that there is no such thing as the general good at all, but only individual goods, and that each man is only concerned with what is good for himself. These meanings all presuppose that we know what is meant by "my good"; but this is not an easy con2o ception to define clearly. I shall therefore begin by considering what it is capable of meaning. 32. "My good" is a phrase capable of many different meanings. It may mean any good that I desire, whether this has any further special relation to me or not. Or, again, it may mean my pleasure, or any state of mind in me which is good. Or it may include honour and respect from others, or anything which is a good and has some relation to me in virtue of which it can be considered mine. The two meanings with which we shall be concerned are: (1) any good I desire, (2) any good having to me some relation other than that I desire it, which it does not have to others, of the kind o 3 which makes it mine, as my pleasure, my reputation, my learning, my virtue, etc. The theory that every man is psychologically bound to pursue his own good exclusively is, I think, inconsistent with known facts of human nature, unless "my good" is taken in the sense of "something which I desire" and even then I do not necessarily pursue what I desire most ' . . strongly. The important point is, that what I desire has not necessanly any such other relation to me as would make it my good in the second of the above senses. This is the point which must now occupy us. If "my good" means a good which is mine in some other sense than o that I desire it, then I think it can be shown that my good is by no means 4 the only object of my actions. There is a common confusion in people's thoughts on this subject, namely the following: If I desire anything, its

241

IO

20

30

40

242

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

sacrifice is not desired. When a soldier sacrifices his life he does not desire the victory of his country, and so on. This is already sufficiently preposterous, and sufficiently contrary to plain fact. But it is not enough. Assuming that this is the case, let us suppose that self-sacrifice is dictated, not by desire for any outside end, but by fear of the disapproval of others. If this were so there would be no self-sacrifice if no one would know of its non-performance. A man who saw another drowning would not try to save him if he was sure that no one would see him not jumping into the water. This also is plainly contrary to fact. It may be said that the desire for approval, as well as the fear of disapproval, ought to be taken into account; and a man can always make sure of approval by judicious boasting. But men have made sacrifices universally disapproved, for example, in maintaining unpopular opinions; and very many have made sacrifices of which an essential part was that they should not be mentioned. Hence the defender of psychological egoism is driven back on the approval of conscience as the motive to an act of self-sacrifice. But it is really impossible to believe that all who deny themselves are so destitute of rational foresight as this theory implies. The pangs of conscience are to most people a very endurable pain, and practice in wrong-doing rapidly diminishes them. And if the act of self-denial involves the loss of life, the rapture of self-approbation, which the virtuous man is supposed to be seeking, must in any case be very brief. I conclude that the psychology of egoism is only produced by the exigencies of a wrong theory, and is not in accordance with the facts of observable human nature. Thus when we consider human actions and desires apart from preconceived theories, it is obvious that most of them are objective and have no direct reference to self. If "my good" means an object belonging to me in the sense of being a state of my mind, or a whole of which a state of my mind is a part, or what others think about me, then it is false that I can only desire or pursue my good. The only sense in which it is true is when "my good" is taken to mean "what I desire"; but what I desire need not have any other connection with myself, except that I desire it. Thus there is no truth in the doctrine that men do, as a matter of fact, only desire or pursue objects specially related to themselves in any way except as objects desired or pursued. 34. The next form of egoism to be considered is the doctrine that every man will best serve the general good by pursuing his own. There is a comfortable eighteenth-century flavour about this doctrine-it suggests a good income, a good digestion, and an enviable limitation of sympathy. We may admit at once that in a well-ordered world it would be true and ' even that, as society becomes better organized, it becomes progressively truer, since rewards will more and more be attached to useful actions. And in so far as a man's own good is more in his control than other people's,

19 THE ELEMENTS OF ETHICS

243

his actions will rightly concern themselves more with it than with other people's. For the same reason he will be more concerned with the good of his family than with that of people with whom he has less to do, and more with the good of his own country than with that of foreign countries. But the scope of such considerations is strictly limited, and every one can easily find in his own experience cases where the general good has been served by what at any rate appears to be a self-sacrifice. If such cases are to be explained away, it is necessary to alter the conception of "my own good" in a way which destroys the significance of the doctrine we are considering. It may be said, for example, that the greatest of goods is a virtuous life. It will then follow that whoever lives a virtuous life secures for himself the greatest of goods. But if the doctrine means to assert as . ' lt usually does, that self-centred desires, if they are prudent and enlightened, will suffice to produce the most useful conduct, then a refutation may be obtained either from common experience or from any shining example of public merit. The reformer is almost always a man who has strong desires for objects quite unconnected with himself' and . ' mdeed this is a characteristic of all who are not petty-minded. I think the doctrine depends for its plausibility, like psychological egoism, upon regarding every object which I desire as my good, and supposing that it must be mine in some other sense than that I desire it. 35. The doctrine that my good is the only thing that I ought to think good can only be logically maintained by those who hold that I ought to believe what is false. For if I am right in thinking that my good is the only good, then every one else is mistaken unless he admits that my good, not his, is the only good. But this is an admission which I can scarcely hope that others will be willing to make. But what is really intended is, as a rule, to deny that there is any such thing as the general good at all. This doctrine cannot be logically refuted, unless by discovering in those who maintain it some opinion which implies the opposite. If a man were to maintain that there is no such thing as colour, for example, we should be unable to disprove his position, provided he was careful to think out its implications. As a matter of fact, however, everybody does hold opinions which imply a general good. Everybody judges that some sorts of communities are better than others· ' and most people who affirm that when they say a thing is good they mean merely that they desire it, would admit that it is better two people's desires should be satisfied than only one person's. In some such way people fail to carry out the doctrine that there is no such concept as good; and if there is such a concept, then what is good is not good for me or for you, but is simply good. The denial that there is such a thing as good in an impersonal sense is only possible, therefore, to those who are content to have no ethics at all.

IO

20

3o

40

244

10

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

36. It is possible to hold that, although there is such a thing as the general good, and although this is not always best served by pursuing my own good, yet it is always right to pursue my own good exclusively. This doctrine is not now often held as regards individuals; but in international politics it is commonly held as regards nations. Many Englishmen and many Germans would admit that it is right for an English statesman to pursue exclusively the good of England, and a German the good of Germany, even if that good is to be attained by greater injury to the other. It is difficult to see what grounds there can be for such a view. If good is to be pursued at all, it can hardly be relevant who is going to enjoy the good. It would be as reasonable for a man on Sundays to think only of his welfare on future Sundays, and on Mondays to think only of Mondays. The doctrine, in fact, seems to have no merit except that it justifies acts otherwise unjustifiable. It is, indeed, so evident that it is better to secure a greater good for A than a lesser good for B, that it is hard to find any still more evident principle by which to prove this. And if A happens to be some one else, and B to be myself, that cannot affect the question, since it is irrelevant to the general maxim who A and B may be. If no form of egoism is valid, it follows that an act which ought to be performed may involve a self-sacrifice not compensated by any personal good acquired by means of such an act. So unwilling, however, are people to admit self-sacrifice as an ultimate duty that they will often defend theological dogmas on the ground that such dogmas reconcile self-interest with duty. Such reconciliations, it should be observed, are in any case merely external; they do not show that duty means the pursuit of one's own interest, but only that the acts which it dictates are those that further one's own interest. Thus when it is pretended that there are logical grounds making such reconciliations imperative, we must reply that the logical purpose aimed at could only be secured by showing that duty means the same as self-interest. It is sometimes said that the two maxims, "You ought to aim at producing the greatest possible good" and "You ought to pursue your own interest", are equally evident; and each is supposed to be true in all possible circumstances and in all possible worlds. But if that were the case, a world where self-interest and the general good might conflict ought not only to be non-existent, but inconceivable; yet so far is it from being inconceivable that many people conceive it to be exemplified in the actual world. Hence the view that honesty is the best policy may be a comfort to the reluctant saint, but cannot be a solution to the perplexed logician. The notion, therefore, that a good God or a future life can be logically inferred to remove the apparent conflict of self-interest and the general good is quite unwarrantable. If there were a logical puzzle, it could only be removed by showing that self-interest and the general good mean the same thing, not by showing that they coincide in fact. But if the above

19 THE ELEMENTS OF ETHICS

245

discussion has been sound, there is no logical puzzle: we ought to pursue the general good, and when this conflicts with self-interest, self-interest ought to give way. VI. METHODS OF ESTIMATING GOODS AND EVILS

37. In order to complete our account of ethics, it would be natural to give a list of the principal goods and evils of which we have experience. I shall, however, not attempt to give such a list, since I hold that the reader is probably quite as capable as I am of judging what things are good and what bad. All that I propose to do in this section is to examine the view that we can never know what is good and what bad, and to suggest rn methods to be employed and fallacies to be avoided in considering intrinsic goodness or badness. There is a widespread ethical scepticism, which is based upon observation of men's differences in regard to ethical questions. It is said that A thinks one thing good, and B thinks another, and there is no possible way in which either can persuade the other that he is wrong. Hence, it is concluded, the whole thing is really only a matter of taste, and it is a waste of time to ask which is right when two people differ in a judgment of value. It would be absurd to deny that, as compared with physical science, 20 ethics does suffer from a measure of the defect which such sceptics allege. It must be admitted that ultimately the judgment "this thing is good" or "that thing is bad" must be an immediate judgment, which results merely from considering the thing appraised, and cannot be proved by any argument that would appeal to a man who had passed an opposite immediate judgment. I think it must also be admitted that, even after every possible precaution against error has been taken, people's immediate judgments of value do still differ more or less. But such immediate differences seem to me to be the exception: most of the actual differences are of a kind which argument might lessen, since usually the opinion held is either one of 30 which the opposite is demonstrable or one which is falsely believed to be itself demonstrable. This second alternative embraces all false beliefs held because they flow from a false theory; and such beliefs, though often the direct contraries of what immediate inspection would lead to, are apt to be a complete bar to inspection. This is a very familiar phenomenon. Sydney Smith, believed to be always witty, says "pass the mustard", and the whole table is convulsed with laughter. Much wrong judgment in ethics is of this nature. 38. In regard to the things that are good or bad, in themselves, and not merely on account of their effects, there are two opposite errors of this 40 sort to be avoided-the one the error of the philosopher, the other that of

246

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

19 THE ELEMENTS OF ETHICS

the moralist. The philosopher, bent on the construction of a system, is inclined to simplify the facts unduly, to give them a symmetry which is fictitious, and to twist them into a form in which they can all be deduced from one or two general principles. The moralist, on the other hand, being primarily concerned with conduct, tends to becomes absorbed in means, to value the actions men ought to perform more than the ends which such actions serve. This latter error-for in theorizing it is an error-is so forced upon us by the exigencies of practice that we may easily come to feel the ultimate ends of life far less important than the proximate and intermediate purposes which we consciously endeavour to realize. And hence most of what they value in this world would have to be omitted by many moralists from any imagined heaven, because there such things as self-denial and effort and courage and pity could find no place. The philosopher's error is less common than the moralist's, because the love of system and of the intellectual satisfaction of a deductive edifice is rarer than the love of virtue. But among writers on ethics the philosopher's error occurs oftener than the other, because such writers are almost always among the few men who have the love of system. Kant has the bad eminence of combining both errors in the highest possible degree, since he holds that there is nothing good except the virtuous will-a view which simplifies the good as much as any philosopher could wish, and mistakes means for ends as completely as any moralist could enjoin. 39. The moralist's fallacy illustrates another important point. The immediate judgments which are required in ethics concern intrinsic goods and evils, not right and wrong conduct. I do not wish to deny that people have immediate judgments of right and wrong, nor yet that in action it is usually moral to follow such judgments. What I mean is that such judgments are not among those which ethics must accept without proof, provided that (whether by the suggestions of such judgments or otherwise) we have accepted some such general connection of right action with good consequences as was advocated in Section III. For then, if we know what is good and bad, we can discover what is right or wrong; hence in regard to right and wrong it is unnecessary to rely upon immediate inspectiona method which must be allowed some scope, but should be allowed as little as possible. I think when attention is clearly confined to good and bad, as opposed to right and wrong, the amount of disagreement between different people is seen to be much less than might at first be thought. Right and wrong, since they depend upon consequences, will vary as men's circumstances vary, and will be largely affected, in particular, by men's beliefs about right and wrong, since many acts will in all likelihood have a worse effect if they are generally believed to be wrong than if they are generally be-

lieved to be right, while with some acts the opposite is the case. (For example, a man who, in exceptional circumstances, acts contrary to a received and generally true moral rule, is more likely to be right if he will be thought to be wrong, for then his action will have less tendency to weaken the authority of the rule.) Thus differences as regards rules of right action are not a ground for scepticism, provided the different rules are held in different societies. Yet such differences are in practice a very powerful solvent of ethical beliefs. 40. Some differences as to what is good in itself must, however, be acknowledged even when all possible care has been taken to consider the question by itself. For example, retributive punishment, as opposed to deterrent or reformative punishment, was almost universally considered good until a recent time; yet in our own day it is very generally condemned. Hell can only be justified if retributive punishment is good; and the decay of a belief in hell appears to be mainly due to a change of feeling on this point. But even where there seems to be a difference as to ends, this difference is often due to some theory on one side or on both, and not to immediate inspection. Thus in the case of hell, people may reason, consciously or unconsciously, that revelation shows that God created hell, and that therefore retributive punishment must be good; and this argument doubtless influences many who would otherwise hold retributive punishment to be bad. Where there is such an influence we do not have a genuine difference in an immediate judgment as to intrinsic good or bad; and in fact such differences are, I believe, very rare indeed. 4I. A source of apparent differences is that some things which in isolation are bad or indifferent are essential ingredients in what is good as a whole, and some things which are good or indifferent are essential ingredients in what is bad as a whole. In such cases we judge differently according as we are considering a thing in isolation or as an ingredient in some larger whole. To judge whether a thing is in itself good, we have to ask ourselves whether we should value it if it existed otherwise than as an ingredient in some whole which we value. But to judge whether a thing ought to exist, we have to consider whether it is a part of some whole which we value so much that we prefer the existence of the whole with its possibly bad part to the existence of neither. Thus compassion is a good of which some one's misfortune is an essential part; envy is an evil of which some one's good is an essential part. Hence the position of some optimists, that all the evil in the world is necessary to constitute the best possible whole, is not logically absurd, though there is, so far as I know, no evidence in its favour. Similarly the view that all the good is an unavoidable ingredient in the worst possible whole is not logically absurd;

247

IO

20

30

40

248

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

but this view, not being agreeable, has found no advocates. Even where none of the parts of a good whole are bad, or of a bad whole good, it often happens that the value of a complex whole cannot be measured by adding together the values of its parts; the whole is often better or worse than the sum of the values of its parts. In all aesthetic pleasures, for example, it is important that the object admired should really be beautiful: in the admiration of what is ugly there is something ridiculous, or even sometimes repulsive, although, apart from the object, there may be no difference in the value of the emotion per se. And yet, apart from the admiration it may produce, a beautiful object, if it is inanimate, appears to be neither good nor bad. Thus in themselves an ugly object and the emotion it excites in a person of bad taste may be respectively just as good as a beautiful object and the emotion it excites in a person of good taste; yet we consider the enjoyment of what is beautiful to be better, as a whole, than an exactly similar enjoyment of what is ugly. If we did not we should be foolish not to encourage bad taste, since ugly objects are much easier to produce than beautiful ones. In like manner, we consider it better to love a good person than a bad one. Titania's love for Bottom may be as lyric as Juliet's for Romeo; yet Titania is laughed at. Thus many goods must be estimated as wholes, not piecemeal; and exactly the same applies to evils. In such cases the wholes may be called organic unities. 42. Many theorists who have some simple account of the sole good have also, probably without having recognized them as such, immediate judgments of value inconsistent with their theory, from which it appears that their theory is not really derived from immediate judgments of value. Thus those who have held that virtue is the sole good have generally also held that in heaven it will be rewarded by happiness. Yet a reward must be a good; thus they plainly feel that happiness also is a good. If virtue were the sole good it would be logically compelled to be its own reward. A similar argument can be brought against those who hold that the sole good is pleasure (or happiness, as some prefer to call it). This doctrine is regarded as self-evident by many, both philosophers and plain men. But although the general principle may at first sight seem obvious, many of its applications are highly paradoxical. To live in a fool's paradise is commonly considered a misfortune; yet in a world which allows no paradise of any other kind a fool's paradise is surely the happiest habitation. All hedonists are at great pains to prove that what are called the higher pleasures are really the more pleasurable. But plainly their anxiety to prove this arises from an uneasy instinct that such pleasures are higher, even if they are not more pleasurable. The bias which appears in hedonist arguments on this point is otherwise quite inexplicable. Although they hold that, "quantity of pleasure being equal, pushpin is as good as po-

19 THE ELEMENTS OF ETHICS

249

~try", the~ are careful to argue that quantity of pleasure is not equal, but is greater m the case of poetry-a proposition which seems highly disputa?le, and chiefly commended by its edifying nature. Any one would adm~t that the pleasure of poetry is a greater good than the pleasure of bathmg on a hot day; but few people could say honestly that it is as int~nse. And even states of mind which, as a whole, are painful, may be ~ighly good. Love of the dead may easily be the best thing in a life; yet i~ cannot but be full of pain. And conversely, we condemn pleasure de~ived ~rom the love of what is bad; even if we admit that the pleasure in itself is a good, we consider the whole state of mind bad. If two bitter IO enemies lived in different countries, and each falsely believed that the other wa_s undergoing tortures, each might feel pleasure; yet we should not consider such a state of things good. We should even think it much worse than a state in which each derived pain from the belief that the other was in torture. It may, of course, be said that this is due to the fact that hatred in general causes more pain than pleasure, and hence is cond_emned bro~dly on hedonistic grounds, without sufficient regard to possible exceptions. But the possibility of exceptions to the principle that hatred is bad can hardly be seriously maintained, except by a theorist in difficulties. 20 Thus while we may admit that all pleasure, in itself, is probably more or less good, we must hold that pleasures are not good in proportion to their intensity, and that many states of mind, although pleasure is an element in ~hem, are bad as a whole, and may even be worse than they would be if the pleasure were absent. And this result has been reached by appealing to ethical judgments with which almost every one would agree. I conclude, therefore, from all that has been adduced in this section, that although some ultimate ethical differences must be admitted between different people, by far the greater part of the commonly obse~ved _differences are due either to asking the wrong question (as, e.g. by 3o mistakmg means for ends), or to the influence of a hasty theory in falsifying immediate judgments. There is reason to hope, therefore, that a very large measure of agreement on ethical questions may be expected to result from clearer thinking; and this is probably the chief benefit to be ultimately derived from the study of ethics. 43· We may now sum up our whole discussion of ethics. The most fundamental notions in ethics, we agreed, are the notions of intrinsic good and evil. These are wholly independent of other notions, and the goodness or badness of a thing cannot be inferred from any of its other qualities such as its existence or non-existence. Hence what actually occurs has n~ 40 bearing on what ought to occur, and what ought to occur has no bearing on what does occur. The next pair of notions with which we were concerned were those of objective right and wrong. The objectively right act

250

ro

20

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

is the act which a man will hold that he ought to perform when he is not mistaken. This, we decided, is that one, of all the acts that are possible, which will probably produce the best results. Thus in judging what actions are right we need to know what results are good. When a man is mistaken as to what is objectively right, he may nevertheless act in a way which is subjectively right; thus we need a new pair of notions, which we called moral and immoral. A moral act is virtuous and deserves praise; an immoral act is sinful and deserves blame. A moral act, we decided, is one which the agent would have judged right after an appropriate amount of candid reflection, 6 where the appropriate amount of reflection depends upon the difficulty and importance of his decision. We then considered the bearing of determinism on morals, which we found to consist in a limitation of the acts which are possible under any circumstances. If determinism is true, there is a sense in which no act is possible except the one which in fact occurs; but there is another sense, which is the one relevant to ethics, in which any act is possible which is contemplated during deliberation (provided it is physically possible, i.e. will be performed if we will to perform it). We then discussed various forms of egoism, and decided that all of them are false. Finally, we considered some mistakes which are liable to be made in attempting to form an immediate judgment as to the goodness or badness of a thing, and we decided that, when these mistakes are avoided, people probably differ very little in their judgments of intrinsic value. The making of such judgments we did not undertake; for if the reader agrees, he could make them himself, and if he disagrees without falling into any of the possible confusions, there is no way of altering his opinion.

20

Spinoza [1910]

THIS REVIEW w AS published in The Nation, 8 (12 Nov. 19ro): 278, 280. W. Hale White, one of the translators, wrote to Russell after the review appeared: I cannot help thanking you for your paper on Spinoza in the Nation of last week. I began to read him fifty years ago and I have not for any length of time ceased to read him. I have read also a good deal about him, but I have not found, even in bulky volumes, anything more instructive than what you have said in two or three columns. I know but little of metaphysics as a science. I study Spinoza because he teaches me how to live and you in this article make me better understand in what way he excels in teaching it. The two sentences "he sees no end to strife except by persuading men to choose as their end things which all may enjoy in common", and "it seems that in the knowledge of what is necessary we place ourselves" in harmony "with what is greatest in the universe" are worth much lengthy exposition. Thank you also for the kind words about the translation. Russell retained the book for his library. An examination of it reveals many vertical lines, but only three very brief verbal comments and one correction. The correction is to Definition IV, where he marks out "if": "By attribute, I understand that which the intellect perceives of substance, as if constituting its essence." All other translations checked agree that "as if" is incorrect. The other comment worth noting here concerns this passage from the scholium to Proposition LXXII: If it be asked whether, if a man by breach of faith could escape from the danger of instant death, reason does not counsel him, for the preservation of his being, to break faith; I reply in the same way, that if reason gives such counsel, she gives it to all men, and reason therefore generally counsels men to make no agreements for uniting their strength and possessing laws in common except deceitfully, that is to say, to have in reality no common laws, which is absurd.

6 Or after a small amount in the case of acts which ought to be impulsive.

Against this passage Russell wrote "Kant". The printed version is the copy-text. 251

20 SPINOZA

Ethic, Demonstrated in Geometrical Order and Divided into Five Parts, which Treat (I) of God; (2) of the Nature and Origin of the Mind; (3) of the Nature and Origin of the Affects; (4) of Human Bondage, or of the Strength of the Affects; (5) of the Power of the Intellect, or of Human Liberty. By Benedictus de Spinoza. Translated by W. Hale White and Amelia Hutchison Stirling. 4th ed. London, New York and Toronto: Oxford University Press, 1910.

T

HE WORK OF the popularizer, though sometimes depreciated by professional students, is a very useful and necessary work, and in few cases more useful or more necessary than in the case of 10 Spinoza. For, although Spinoza is often so difficult that even the best philosophers cannot be sure of having understood him, the essence of his doctrine is capable of being interesting and profitable to many who cannot devote themselves to metaphysics. Although the task of interpretation has been admirably performed for the technical reader by Mr. Joachim, and for a wider class by Sir Frederick Pollock, there must be many who will be grateful for the chance of reading the Ethics itself, without having to work their way through Spinoza's Latin. The translation, though not wholly devoid of errors, seems in the main accurate and careful. It is, perhaps, a pity that the translators have chosen the manufactured word 20 "affect" to translate affectus rather than the word "emotion", which is used by Sir Frederick Pollock and Mr. Joachim. But their choice has, at least, the merit of avoiding misleading associations, and of making the reader aware that Spinoza's meaning cannot be accurately rendered by any existing word in its common signification. There is a useful preface, giving the main facts of Spinoza's life, together with some account of his other works, of his relations to other philosophers, and of his influence on subsequent writers. In this preface, Mr. Hale White rightly emphasizes, what is sometimes lost sight of, that Spinoza's purpose, as his title indicates, was ethical, and that he only introduced metaphysics in so far as seemed 30 essential for his ethical doctrine. There is also a careful comparison of the Ethics with the Short Treatise on God, Man, and Man's Well-Being, a sort of first draft of the Ethics, which was not discovered and printed until 1862, and was not even known to have existed until an abstract of it was found in I 85 I attached to a manuscript of the life of Spinoza by his friend Colerus. The unassisted reader who opens the Ethics casually is likely to be completely misled as to Spinoza's purpose. In the first book he will find only pantheism; in the second he will find antiquated physiology, with a suggestion of materialism; in the third he will be tempted to regard Spinoza 40 as a pedantic La Rochefoucauld, retaining the cynicism without the wit. It is only in the fourth and fifth books that Spinoza's purpose becomes

252

253

obvious; but the casual reader is hardly likely to persevere until he reaches them. Spinoza, more than any other modern philosopher, writes always with a strong sense of the importance of philosophy in the conduct of life, and with a firm belief in the power of reason to improve men's conduct and purify their desires. Like many men of great independence of mind, he feels the need of something great enough to justify him in submitting to its authority. Like all who contemplate human life without sharing its baser passions, he is oppressed by the endless strife produced by conflicting aims and unrestrained ambitions. Believing, as he does, that self-preservation is the very essence of everything that exists, he sees no end to strife except by persuading men to choose as their ends things which all may enjoy in common. Contempt and moral condemnation stand in the way of toleration; he therefore sets out to prove that what men do they do from a necessity of their nature. As well might one condemn a triangle for not having made the effort to increase its angles beyond two right angles as condemn men for being what their nature makes them. His theory of the emotions, in which, by his geometrical method, he demonstrates that men must act in ways which it is common to condemn, contains much admirable psychology; but it was not this that made him value his theory: what he valued was the conclusion that moral condemnation is foolish. It is for this reason partly that Spinoza inveighs against free-will, and finds pleasure in showing the necessity of everything. But there is also another reason: what is transitory, though it may be tolerated, cannot be worshipped; but the proof of its necessity connects it with the Divine nature, and thereby removes its pitifulness. To a certain type of mind there is something sublime about necessity: it seems that in the knowledge of what is necessary we place ourselves in harmony with what is greatest in the universe. This constitutes, to those who feel it, a great part of the value of mathematical demonstration; even Spinoza's geometrical method, which has been almost universally condemned, will be held appropriate by those who know the "intellectual love of God". "He who loves God", Spinoza says, "cannot strive that God should love him in return." Goethe, in a passage of characteristic sentimentality, misquotes this proposition in singling it out for special praise; he quotes it as, "Who loves God truly must not expect God to love him in return", and regards it as an example of "Entsagen sollst du, sollst entsagen." If Goethe had understood Spinoza's religion, he would not have made this mistake. Spinoza, here and elsewhere, is not inculcating resignation; he himself loved what he judged to be best, and lived, so far as one can discover, without effort in the way which he held to be conformable to reason. There seems to have been in him, what his philosophy was in-

10

20

30

40

254

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

tended to produce in others, an absence of bad desires; hence, his nature is harmonious and gentle, free from the cruelty of asceticism, or the monkishness of the cloister, or the moralistic priggery of Goethe's praises. "He who clearly and distinctly understands himself and his emotions", Spinoza says, "loves God, and loves Him better the better he understands himself and his emotions." It is through the love of God that we are freed from bondage to the passions, and that our minds become in some degree eternal. "God loves Himself with an infinite, intellectual love", and "the intellectual love of the mind towards God is the very love with which He ro loves Himself." Hence, though immortality in the ordinary sense is an error, the mind is nevertheless eternal in so far as it consists in the intellectual love of God. To represent such a philosophy as one of renunciation is surely to miss the whole of the mystic joy which it is intended to produce, and to misunderstand the reconciliation of the individual with the whole, which is the purpose of so much elaborate argument. Spinoza's ethical views are inextricably intertwined with his metaphysics, and it may be doubted whether his metaphysics is as good as is supposed by followers of Hegel. But the general attitude towards life and the world which he inculcates does not depend for its validity upon a system 20 of metaphysics. He believes that all human ills are to be cured by knowledge and understanding; that only ignorance of what is best makes men think their interests conflicting, since the highest good is knowledge, which can be shared by all. But knowledge, as he conceives it, is not mere knowledge as it comes to most people; it is "intellectual love", something coloured by emotion through and through. This conception is the key to all his valuations. "It is knowledge", he says, "which is the cause of love, so that when we learn to know God in this way, we must necessarily unite ourselves to Him, for He cannot be known, nor can He reveal Himself save as that 30 which is supremely glorious and good." And owing to Spinoza's pantheism, love of God, for him, included love of humanity. The love of humanity is a background to all his thoughts, and prevents the coldness which his intellectualism might otherwise engender. It was through the union of the love of truth and the love of humanity, combined with an entire absence of self-seeking, that he achieved a nobility, both in life and in speculation, which has not been equalled by his predecessors or successors in the realm of philosophy.

Part v Critique of Pragmatism

21

Pragmatism [1909]

THIS PAPER w AS PUBLISHED, unsigned, in The Edinburgh Review, 209 (April 1909): 363-88. It was reprinted by Russell in Philosophical Essays (1910). In 1966 Amelie Rorty included it (with omissions) in Pragmatic Philosophy: An Anthology. We know that Russell was sent proofs for its original publication, because, on 21 March 1909, his kinsman, Hugh Elliot, wrote to him: I have been deeply interested in your admirable essay on Pragmatism, of which the proof has, I hope, reached you. Your views are in entire accordance with my own; and although I have no special knowledge of Philosophy, I am able to follow with ease every part of your paper. With reference to a title, do you think "The New Philosophy" would be suitable? or do you think it would be better to indicate more directly the name of the philosophy. Elliot was, at the time, the editor of the journal. William James (1842-1910) was the most famous of the pragmatists. Russell had earlier reviewed his Pragmatism (1907) under the provocative title, "Transatlantic 'Truth"' (1908). To Margaret Llewelyn Davies, in a letter of 16 January 1908, he maintained that he had given the review a "plain title" which the editor of The Albany Review had seen fit to change. James replied to Russell in "Two English Critics", first published in The Meaning of Truth (1909). The present review is Russell's second salvo, delivered while James was preparing his reply to the first. Despite what he says about James's pragmatic views, Russell remained fond of James as a n}an and appreciative of his importance as a philosopher: The death of William James, which occurred when the printing of this book was already far advanced, makes me wish to express, what in the course of controversial writings does not adequately appear, the profound respect and personal esteem which I felt for him, as did all who knew him, and my deep sense of the public and private loss occasioned by his death. For readers trained in philosophy, no such assurance was required; but for those unaccustomed to the tone of a subject in which

257

258

259

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

21 PRAGMATISM

agreement is necessarily rarer than esteem, it seemed desirable to record what to others would be a matter of course.

realized in rebus, and having only this quality in common, that they pay." Russell offers "i.e. kinds of truth" as a gloss on "truths". His last note comes on page 226 where he asserts that James's "reality itself or truth itself" "should not be coupled". Schiller's Humanism shows evidence (Russell's usual vertical lines in the margins) of having been read from the beginning to page 61 and then again from page 167 to the end. the only verbal comment is found beside this sentence of Schiller's: "Naturalism is valid enough and useful as a method of tracing the connexions that permeate reality from the lowest to the highest level: but when taken as the last word of philosophy it subjects the human to the arbitrament of its inferior" (xxiv). "You appeal to a scholastic maxim here" is Russell's judgment. Russell's comments in Schiller's Studies in Humanism are both more numerous and more scathing. Two examples may be cited. Russell accuses Schiller of "pure sophistry" in this passage:

This "Postscript", dated October 1910, was added to the Preface of Philosophical Essays (reproduced in Appendix n). For a more extended tribute to James, see Paper 22 below. For biographical information on F. C. S. Schiller, and Russell's relationship with him, see the Headnote to Paper 24. Russell sent Schiller the article in proof, and on 25 March 1909 Schiller wrote him a long letter in reply. That letter is printed in Appendix I. John Dewey (1859-1952) was, after James, the next most famous pragmatist. Russell met him for the first time in 1914 at Harvard University. In a letter to Lady Ottoline Morrell on 22 March he offered his first impression of the man: Dewey (the third pragmatist, with James and Schiller) has been here. I met him at lunch yesterday and then had a walk with him. To my surprise I liked him very much. He has a large slow-moving mind, very empirical and candid, with something of the impassivity and impartiality of a natural force. He and Perry and I had a long argument about "!"-Dewey saw a point I was making but Perry didn't-he is a good man but not a very clever one, as the country gentlemen said of Dizzy. (#1008) Years later when they met again in China Russell developed an antipathy toward Dewey which was to last to the end of Russell's life. From China, on 21 February 1921, Russell wrote to Lady Ottoline: The Deweys, who are here, and who got in trouble in America during the war for their liberalism, are as bad as anybody-American imperialists, hating England as (Leo) Maxse used to hate Germany, and unwilling to face any unpleasant facts. In 1914, I liked Dewey better than any other academic American; now I can't stand him. (#1583) Despite this estrangement Dewey was to prove both kind and helpful when Russell was embroiled in the fracas which his appointment as Professor of Philosophy in the College of the City of New York in 1940 stirred up. Of the six books reviewed, only three remain in Russell's library: James's Pragmatism and the two books by Schiller. An examination of Pragmatism reveals many vertical lines in the margins but only three verbal comments. Against James's assertion (77), '"What would be better for us to believe'! This sounds very like a definition of truth", Russell notes "Not at all". At page 218 James says: "Our account of truth is an account of truths in the plural, of processes of leading,

I find it hard, therefore, to understand why a religious assumption, such as, e.g. the existence of a "God," should require a different and austerer mode of proof, or why the theologian should be debarred from a procedure which is always reputable, and sometimes heroic, in a man of sCience. (362) .. )

And Russell poses a question in the margin beside this passage: Whether a reality is called "independent" of our knowing, and said to be merely "discovered" when it is known, or not, seems to depend essentially on whether it is aware of being known; or rather on how far, and in what ways, it is aware of being known. His question: "Is a murderer only discovered when he doesn't know he is?" The copy-text is the earliest printed version.

21 PRAGMATISM

The Will to Believe, and Other Essays in Popular Philosophy. By William James. London: Longmans, 1897. Pp. xvii, 332. Pragmatism: A New Name for Some Old Ways of Thinking. By William James. London: Longmans, 1907. Pp. xiii, 309. Humanism: Philosophical Essays. By F. C. S. Schiller. London: Macmillan, 1903. Pp. xxvii, 297. Studies in Humanism. By F.C.S. Schiller. London: Macmillan, 1907. Pp. xvii, 492. Studies in Logical Theory. By John Dewey, with the Co-Operation of Members and Fellows of the Department of Philosophy. (The Decennial Publications, 2nd ser., Vol. II.) Chicago: The University of Chicago Press, 1903. Pp. xiii, 388.

ro

Essays Philosophical and Psychological in Honor of William James, Professor in Harvard University. By His Colleagues at Columbia University. London: Longmans, 1908. Pp. viii, 6rn.

T

20

30

HE APPEARANCE IN

the world of a genuinely new philosophy is

~t al~ times an event of very great importance. More particularly

is this the case when the new philosophy embodies the prevailing temper of the age better than any of its older rivals; for in that case it is likely to establish itself in popular favour, to colour the thoughts of the educated and half-educated public, and to strengthen those elements in the mental atmosphere to which it owes its success. It would be a mistake to suppose that new philosophies are always adapted to the age in which they appear; but when they are not, they fail to win wide acceptance whatever their other merits may be. Spinoza, for example, deserved success as well as Leibniz; yet his works were almost wholly neglected until more than a century after his death, because the political and intellectual milieu was not one in which they could thrive. Leibniz, on the contrary, gave scope to the love of calculation which men derived from the discoveries of his time, and represented the world as a hierarchy of systems, each exactly like the Holy Roman Empire; his system, therefore, ruled the German mind until the ferment which preceded the French Revolution set men's thoughts running in new channels. The philosophy which is called Pragmatism or Humanism 1 is genuinely I

These two names arc distinguished by William James and Dr. Schiller in various ways at various times. For our purposes, it is unnecessary to consider these distinctions.

new, and is singularly well-adapted to the predominant intellectual temper of our time. As regards its adaptation to the age, we shall have more to say when we have considered what it is. As regards novelt~, its authors show a modesty which, in our opinion, is somewhat excessive. Pragmatism: A New Name for Some Old Ways of Thinking, William James calls his book; and Dr. Schiller constantly asserts that his doctrines are ~hose of Protagoras. As for Protagoras, we know sufficiently little about him to be able to read into him almost any doctrine we please; and the appeal to him may be regarded as mainly due to the desire to produce an ancestry which has acquired respectability by the lapse of time. With regard to ro more modern precursors, it must be admitted that many philosophersas chief among whom we may mention Nietzsche-have paved the w~y for the new doctrines. Nevertheless, the cardinal point in the pragmatist philosophy, namely its theory of truth, is so new.' and so necessary ~o the rest of the philosophy, even to those parts which had been prev10usly maintained by others, that its inventors cannot be regarded as merely developing the thoughts of less explicit predecessors. . The name "pragmatism" was first invented by Mr. C. S. P~ir~e, as long ago as l 878. It was applied by him to the doctrine that the s1gm~cance of a thought lies in the actions to which it leads. In order to estimate ~he 20 difference between two different beliefs about the same matter, he mamtained, we ought to consider what difference in conduct ':ould result according as we adopted the one belief or the other. If no di~ference w~uld result the two beliefs are not effectively different. Mr. Peirce's doctrine, howe~er remained sterile until it was taken up twenty years later by William Ja~es who while retaining the word "pragmatism'', gave it a more sweeping signifi;ance. The full-fledged philosophy is to be_ attri~ute~ to him and Dr. Schiller jointly. Professor Dewey, of Columbia Umversity, is also to be reckoned among the founders of pragmatism. His wri~ings are more technical and less popular than those of Jam es and Dr. Schiller, 30 but on certain points his exposition is perhaps preferable to theirs.~ . As an introduction to pragmatism, it is interesting to read Wilham James's essay on "The Will to Believe'', first published in 1896, and reprinted in book form in the following year. In this essay, though t~e :vord "pragmatism" does not appear, we find much that is ~haract~nstic of James's later views. The thesis he is advocating is that, i? certam cases, it is right to believe whole-heartedly in one of two alternatives~ even when there is no evidence as to which of them is true. These cases anse, he says, when we are compelled to choose between two hypotheses, each of which seems to us possible, and when it makes a great difference which we 40 2

Cf. especially an article on "The Experimental Theory of Knowledge", Mind, n.s., no. 59 (July 1906).

260

261

262

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

ch~o.se. The instances he has in mind are chiefly questions of morals and r~hgi~n. I~ a rr_ioral perplexity we are compelled to come to some decision, smce mact10n is as much a decision as action. In regard to religion also we must act as though it were true or as though it were false· \~e ar~ therefore practically compelled to choose. His contention is that' in such cases, it would be foolish to refuse to have faith merely on th; ground that we .do not find conclusive evidence on either side of the question. To quote his own words:

IO

Ou: passional nature not only lawfully may, but must, decide an opt10n bet~een propositions, whenever it is a genuine option that cannot by its natu.re be decided on intellectual grounds; for to say, ~nder such circumstances, "Do not decide, but leave the quest10n open," is itself a passional decision-just like deciding yes or no-and is attended with the same risk of losing the truth.

He proceeds to justify himself against the charge of insufficient regard for trut~, not, as he would do now, by contending that, in the absence of other evidence, the answer which gives the greatest emotional satisfaction is tru~, but on a variety of grounds tending to show that there are no s~fficient moral .arguments .against thinking it true. He points out, to begin 2 0 with, that emotions and wishes, though often unable to alter our beliefs :V.hen thes~ ?ave become established, nevertheless play a great part in inltlally decidmg what our beliefs are to be. He points out next that our duty in the matter of opinion has two branches: (I) We must know the t~uth; (2) we must ~void error. These two precepts, he says, have very differe~t results_. If, m cases where evidence is lacking, we abstain wholly from either belief, we are sure of not incurring error, but on the other hand we ar~ sure of not knowing truth. If, however, we decide for one of the alternatives, we have an even chance of knowing the truth. It follows that those who urge us to abstain from belief in the absence of evidence 30 consider it more important to avoid error than to believe truth. This "horror of being duped" he represents as a somewhat contemptible form of co:vardice; "our errors", he says, "are surely not such awfully solemn thmgs. ~n a world .wh.ere we are so certain to incur them in spite of all our caut10n, a ccrtam lightness of heart seems healthier than this excessive nervousness on thei~ behalf." The legitimate conclusion from this argument :vould be that, m ~uch cas~s as William James has in mind, we ought to bel~cve both alternat~vcs; for i? that case we are sure of "knowing" the ~ruth i~ ~~e matter. If lt "".e~e said that to believe both is a psychological impossibility, we would rciom that, on the contrary, it is often done, and 40 tha~ those who cannot yet do it need only practise the "will to believe" until they have learnt to believe that the law of contradiction is false-a

21 PRAGMATISM

263

feat which is by no means as difficult as it is often supposed to be. William James proceeds to point out that, in the case of religion, the choice between believing and disbelieving possesses all the characteristics of the options which, according to him, ought to be decided by the emotions. He tacitly assumes that there is no evidence for or against religion, and he points out that by refusing either to believe or to disbelieve we lose the benefits of religion just as much as by deciding to disbelieve. Scepticism, then, is not avoidance of option; it is option of a certain particular kind of risk. Better risk loss of truth than chance of error-that is your faith-vetoer's exact position. He is actively playing his stake as much as the believer is; he is backing the field against the religious hypothesis, just as the believer is backing the religious hypothesis against the field .... It is not intellect against all passions, then; it is only intellect with one passion laying down its law. And by what, forsooth, is the supreme wisdom of this passion warranted? Dupery for dupery, what proof is there that dupery through hope is so much worse than dupcry through fear?

IO

The conclusion is that, although there is no evidence in favour of religion, we ought nevertheless to believe it if we find satisfaction in so d~ng.

w This essay on the will to believe is important, because it has been widely read and much criticized, both adversely and favourably, and because it affords a good introduction to the pragmatist temper of mind. Some practice in the will to believe is an almost indispensable preliminary to the acceptance of pragmatism; and conversely pragmatism, when once accepted, is found to give the full justification of the will to believe. We shall, therefore, before proceeding to pragmatism proper, consider briefly what there is to be said, on a common-sense basis, against the doctrines so persuasively set forth in this essay. We may observe, to begin with, the agnostic hypothesis upon which the 30 whole argument rests. The hypothesis is, that no evidence for or against religion is at present known. Pragmatists pose as the friends of religion (except in Italy), and many religious people have accepted them as allies. It is therefore worth while to emphasize this underlying hypothesis, and to point out the very questionable wisdom of accepting it as the basis of a defence of orthodoxy. With the truth or falsehood of this hypothesis, however, we need not concern ourselves in this discussion; the question for us is whether, granting the hypothesis, we can accept the results which William James derives from it. Let us observe, in the first place, a confusion which runs through the -to whole pragmatist account of knowledge, namely the confusion between

264

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

acti~g on. an hypothesis and believing it. In the cases which William James has m mmd, the option between rival hypotheses is, he says a "forced" option; i.e. it is not avoidable: ' If I say, "Either accept this truth or go without it," I put on you a forced option, for there is no standing place outside of the alternative.

This statement appears to us to be contrary to many of the plainest facts of daily life. If, in walking along a country road, I come to a fork where the.re is no signpost and no passer-by, I have, from the point of view of 10 act10n, a "forced" option. I must take one road or other if I am to have any chance of reaching my destination; and I may have no evidence whatever as to which is the right road. I then act on one or other of the two possible hypoth~ses, .until I find some one of whom I can ask the way. But I do n~t b.elzev~ eithe~ hypothesis. My action is either right or wrong, bu~ my belief is neither, smce I do not entertain either of the two possible beliefs. The pragmatist assumption that I believe the road I have chosen to b~ the right one is erroneous. To infer belief from action, in the crude way mvolved in the assumption that we must "either accept this truth or go without it", is t?.i~nore the plain fact that our actions are constantly 20 based upon pr?bab1li~1es, and that, in all such cases, we neither accept a trut~ nor go without It, but entertain it as an hypothesis. This applies, in particular, to the working hypotheses of science. A man of science who considers it worth while to devise experimental tests of an hypothesis, and to construct elaborate theories which use the hypothesis, is not on that account to be regarded as believing the hypothesis. Pragmatists tell us that in such cases, the initial unverified belief is a necessary condition for th~ subsequent established theory, and by so doing they make out a case for th~ usefulness of believing before we have evidence. This is, however, a m~staken anal_Ysis of.the state of mind of a man who is testing an hypoth30 ~sis. All t~at is reqmred, and all that occurs among careful investigators, IS the belief tha~ the ~ypothes~s has a greater or smaller chance of being tru~, and for this belief there is probably sufficient evidence. The actual b.ehef .that the hypothesis is true, when it occurs, is apt to be a hindrance, smce it retards the abandonment of false hypotheses when the evidence goes ~gainst the1:1, a~d if the belief is general, it makes people regard exp~nmental venfication as unnecessary. The Aristotelians who opposed Galileo and refu~ed to give ~eight to his experiments had faithfully obeyed the precepts revived by William James. ~he matter is, however, more complicated in such cases as religious 40 beliefs, where the chief benefit is derived from the emotional satisfaction of the belief itself, not from the useful actions to which it directly prompts.

21 PRAGMATISM

265

But here, too, the antithesis of "accepting" or "going without" is far too crude; we may regard the belief as more or less probable, entertain a greater or less degree of hope that it may be true, and derive, accordingly, a greater or less proportion of the comfort we should derive from complete belief. In practice, to adopt the pragmatist's test, the effect of partial belief is very different from that of complete belief. Complete belief, if the issue is sufficiently momentous, will justify persecution-assuming, as history warrants us in doing, that the blood of Protestant martyrs is the seed of the Catholic Church. An incomplete belief, on the contrary, will not warrant the infliction of an indubitable evil for the sake of a gain which may possibly be illusory. This affords a pragmatic argument against conceding full belief in such cases as those with which William James is concerned. But if, as he assumes, there is a genuine possibility of the truth of an hypothesis, it is in accordance with all the strictest tenets of scientific veracity that we should bear the hypothesis in mind, and allow to it whatever influence over our emotions and actions corresponds to the degree of its probability. We will next examine the argument that, in doubtful cases, the precept "we must know the truth" should lead us to believe one hypothesis at a venture, since, if we believe neither, we certainly do not know the truth. This argument rests upon an ambiguity in the word "know". At first sight, it might be thought that if we believe what is in fact true, we must have knowledge. But this is not the sense in which the word is commonly used. Suppose, to take a trivial instance, that a man believed that the late Prime Minister's name began with a B, but believed this because he thought Mr. Balfour was the late Prime Minister. What he believes is in fact true, yet no one would say that he "knew" that the late Prime Minister's name began with a B. In this case, the true belief is based upon a false reason. But the case is similar when the true belief is based upon no reason (except, indeed, in the case of immediate data such as the facts of perception). Thus if, in the case of an option which we have no rational means of deciding, we believe one alternative at a venture, we cannot be said to know, even if, by good luck, we have chosen the alternative which in fact is true. In such cases, we cannot know the truth, though we may by chance believe it. Hence the precept "we must know the truth", which James invokes, is irrelevant to the issue. The usual antitheses of belief and disbelief, what is known and what is unknown, are not adequate to meet the situation. The true precept of veracity, which includes both the pursuit of truth and the avoidance of error, is this: "We ought to give to every proposition which we consider as nearly as possible that degree of eredence which is warranted by the probability it acquires from the evidence known to us." The further questions what propositions to consider, and how much trouble to take to acquire knowledge of the evidence, depend

ro

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

21 PRAGMATISM

of course upon our circumstances and the importance of the issue. But to go about the world believing everything in the hope that thereby we shall believe as much truth as possible is like practising polygamy in the hope that among so many we shall find some one who will make us happy. Another interesting point to observe in James's doctrine is the immense multiplicity of differing beliefs which it simultaneously justifies in different people. This arises from the condition that the option must be what he calls a "living" option, that is, it must be one in which either alternative seems to us possible.

The pragmatic theory of truth takes credit to itself-rightly, as we think-for a due consideration of error. Most theories as to the nature of truth have tacitly assumed to begin with that all our beliefs are true, and have arrived at results incompatible with the existence of error. They have then had to add a postscript explaining that what we call error is really partial truth. If we think it is Tuesday when it is really Wednesday, we are at least right in thinking that "it" is a day of the week. If we think America was discovered in 1066, we are at least right in thinking that something important happened in that year. If we think Charles I died in his bed, we are at least so far right that, in view of the many people who do die in their beds, he probably had the potentiality of dying in his bed. And so on. Dr. Schiller rightly points to the Theaetetus as showing the difficulties to which a theory of knowledge is reduced by neglecting to take due account of error from the beginning; and among more recent books, Mr. Joachim's The Nature of Truth is used to point the same moral. Pragmatism, then, emphasizes from the start the fact that some of our beliefs turn out to be mistaken, and that the proper business of a theory of truth is to show how truth and falsehood are distinguished. This might seem, to those not sophisticated by philosophy, to be an obvious truism; but in fact philosophy has always regarded it as its business to prove (as far as possible) that everything is true, rather than to distinguish between truth and falsehood. Similarly in ethics philosophers have not sought to distinguish between the good and the bad, so much as to prove that everything is good. If little truth has been attained in philosophy, the reason is chiefly that few philosophers have wished to attain truth. Whether pragmatists are superior in this respect we shall not venture to pronounce; but at any rate the peculiarity of their bias makes them willing to admit facts which other philosophers find inconvenient, and among such facts is the prevalence of error. In order to discover the difference between truth and falsehood, pragmatism sets about a Socratic inductive inquiry as to the things we call "true" and "false". These words, to begin with, are applied to beliefs, and are applied only when a question has arisen. Concerning the ordinary facts of perception, we do not ask questions until we have become philosophers; we do not apply either of the words "true" and "false" to such unquestioned matters. But when once the question has arisen concerning some actual belief, "Is it a true or a false belief?" how do we in fact decide the question? The answer of pragmatism is that if the belief furthers the purpose which led us to ask the question, it is regarded as a "true" belief; if it fails to further the purpose it is regarded as a "false" belief. This, therefore, according to pragmatism, is the meaning of the words "true" and "false". "True" means "furthering the purpose which led to the question". Or, more explicitly: When, in pursuing any purpose, a belief is

266

IO

20

30

40

If I say to you, "Be a theosophist or be a Mohammedan," it is probably a dead option, because for you neither hypothesis is likely to be alive. But if I say, "Be an agnostic or be a Christian," it is otherwise: trained as you are, each hypothesis makes some appeal, however small, to your belief.

He points out that to different people different options are living. It follows that the beliefs which, on his principles, different men ought to adopt, are different, since the three conditions for adopting a belief without evidence are that the option should be living, forced, and momentous. One gathers (perhaps wrongly) from his instances that a Frenchman ought to believe in Catholicism, an American in the Monroe Doctrine, and an Arab in the Mahdi (he wrote before the battle of Omdurman). It seems odd that, in view of this outcome, he should maintain that acceptance of his doctrine would diminish persecution; for an essential part of each of the above three creeds is that people who think otherwise must be taught their place. To sum up our criticism of "The Will to Believe": It ignores the distinction between believing and entertaining an hypothesis, and wrongly assumes that if we do not completely believe an hypothesis, we must either completely disbelieve it or wholly suspend judgment. Hence it is able to represent the option "Either accept this truth or go without it" as one from which there is no escape, whereas all experiment, both in science and in daily life, implies a state of mind which accepts neither alternative. He assumes that we may be said to "know" a truth when we believe it at a venture, without reasons, and that therefore, in order to maximize our knowledge, we have only to maximize our beliefs. And his doctrines lead to the conclusion that different people ought to have incompatible beliefs. These objections, we shall find, may also be urged against full-fledged pragmatism. But we must now approach somewhat more difficult topics than those which have concerned us hitherto, since pragmatism cannot be understood without examining its doctrine as to the nature of truth. To this doctrine, therefore, we will now turn our attention.

267

IO

20

30

40

10

21 PRAGMATISM

entertained which is relevant to the purpose, the belief is "true" if it furthers the achievement of the purpose, and "false" if it does not do so. 3 A few quotations will serve to amplify and elucidate the above brief statement. After explaining recent changes in the methodology of science, James says:

reality-presumably on the ground that an elasticity is not an actually existing thing. The question is, then, what sort of agreement with reality is possible in such cases? "The great assumption of the intellectualists", he says, "is that truth means essentially an inert static relation." An intellectualist, by the way, is anyone who is not a pragmatist. He proceeds:

Riding now on the front of this wave of scientific logic, Messrs. Schiller and Dewey appear with their pragmatistic account of what truth everywhere signifies. Everywhere, these teachers say, "truth" in our ideas and beliefs means the same thing that it means in science. It means, they say, nothing but this, that ideas (which themselves are but parts of our experience) become true just in so far as they help us to get into satisfactory relations with other parts of our experience. 4

Pragmatism, on the other hand, asks its usual question. "Grant an idea or belief to be true," it says, "what concrete difference will its being true make in any one's actual life? How will the truth be realised? What experiences will be different from those which would obtain if the belief were false? What, in short, is the truth's cash-value in experiential terms?" The moment pragmatism asks this question it sees the answer: True ideas are those that we can assimilate, validate, corroborate and verify. False ideas are those that we cannot .... The truth of an idea is not a stagnant property inherent in it. Truth happens to an idea. It becomes true, is made true by events. Its verity is in fact an event, a process: the process namely of its verifying itself, its veri-fication. Its validity is the process of its valid-ation. 6

Again:

20

30

269

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

268

I am well aware how odd it must seem to some of you to hear me say that an idea is "true" so long as to believe it is profitable to our lives. That it is good, for as much as it profits, you will gladly admit. ... But is it not a strange misuse of the word "truth," you will say, to call ideas also "true" for this reason? ... You touch here upon the very central point of Messrs. Schiller's, Dewey's and my own doctrine of truth .... Let me now only say this, that truth is one species of good, and not, as is usually supposed, a category distinct from good, and co-ordinate with it. The true is the name of whatever proves itself to be good in the way of belief, and good, too, for definite assignable reasons. 5 The sixth of William James's lectures on pragmatism is concerned wholly with the notion of truth. He begins by assenting to the dictionary definition that "truth" means the "agreement" of our ideas with "reality". But, as he justly observes, this definition does not take us very far, unless we know what we mean by "agreement" and what we mean by "reality". The pragmatist holds that different sorts of "agreement" and different sorts of "reality" are concerned in different cases. The popular notion that a true idea must copy its reality, holds good, he says, of sensible things, but goes wrong as soon as we come to abstractions. The idea of the elasticity of a spring, for example, cannot, according to him, be a copy of a 3 Cf Schiller, Studies in Humanism, p. 154. 4 Pragmatism, pp. 57-8. 5 Pragmatism, pp. 75-6.

Recurring to the definition of "truth" as "agreement with reality", James sums up by distinguishing three kinds of reality, (1) concrete facts, (2) "abstract kinds of things and relations perceived intuitively between them", (3) truths already in our possession. "Agreement" he defines as follows:

ro

20

To "agree" in the widest sense with a reality can only mean to be guided either straight up to it or into its surroundings, or to be put into such working touch with it as to handle either it or something connected with it better than if we disagreed. (Pp. 212-13) Two further quotations will complete the material required for understanding James's account of truth. "The true,'' to put it very briefly, is only the expedient in the way of our thinking, just as "the right" is only the expedient in the way of our behaving. Expedient in almost any fashion; and expedient in the long run and on the whole of course. (P. 222)

6 Ibid., pp.

200-r.

30

270

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

Our account of truth is an account of truths in the plural, of processes of leading, realised in rebus, and having only this quality in common, that they pay. (P. 2I8) Before proceeding further, it will be as well to clear up a misunderstanding, from which the pragmatists themselves appear not to be exempt. When it is said that truth is "one species of good", it is natural to suppose that ethical considerations are involved, and that logic will become dependent upon ethics. This view is, in fact, adopted in Dr. Schiller's essay 7 on "the ethical basis of metaphysics". But a closer examination shows that ro pragmatists mean by the word "good" whatever satisfies desire. 8 So far as we know, they have nowhere justified this use of the word, but that is not our present concern. What concerns us at present is to observe that, in virtue of this definition, only psychological considerations are relevant where, to judge from the language, ethical considerations might seem to be involved. In order to judge whether a belief is true, it is only necessary to discover whether it tends to the satisfaction of desire. 9 The nature of the desire to be satisfied is only relevant in so far as it may involve conflict with other desires. Thus psychology is paramount, not only over logic and the theory of knowledge, but also over ethics. In order to discover what 20 is good, we have only to inquire how people are to get what they want; and "true" beliefs are those which help in this process. This is the pragmatist theory of truth; and its consequences, as might be supposed, are far- reaching. Before considering the metaphysic which Dr. Schiller has deduced from the pragmatist theory of truth, let us examine the grounds upon which that theory is based. Most philosophies are determined by their initial questions, and by the facts which habitually fill the imagination of the philosopher. The initial question of pragmatism is: What characteristics of beliefs do in fact lead men to regard some as true, others as false? The 30 answer to this question-so pragmatism assumes-will give us the mean-

40

7 The first essay in his Humanism. 8 Schiller, Studies in Humanism, p. 152: "Good and bad also (in their wider and primary sense) have reference to purpose. 'Good' is what conduces to, 'bad' what thwarts, a purpose." 9 Schiller, Studies in Humanism, p. 154: "In all actual knowing the question whether an assertion is 'true' or 'false' is decided uniformly and very simply. It is decided, that is, by its consequences, by its bearing on the interest which prompted to the assertion, by its relation to the purpose which put the question. To add to this that the consequences must be good is superfluous. For if and so far as an assertion satisfies or forwards the purpose of the inquiry to which it owes its being, it is so far 'true'; if and so far as it thwarts or baffles it, it is unworkable, unserviceable, 'false."'

21 PRAGMATISM

271

ing of truth and falsehood. The facts which fill the imaginations of pragmatists are psychical facts: where others might think of the starry heavens, pragmatists think of the perception of the starry heavens; where others might think of God, pragmatists think of the belief in God, and so on. In discussing the sciences, they never think, like scientific specialists, about the facts upon which scientific theories are based: they think about the theories themselves. Thus their initial question and their habitual imaginative background are both psychological. In order to arrive at an external world, they have to prove that the belief in an external world has the marks which (according to them) distinguish a true belief. Hence they ro infer that there is an external world. And a similar process is necessary as regards all other facts which transcend the Ego. One of the approaches to pragmatism is through the consideration of induction and scientific method. The old inductive philosophy, as exemplified in Mill's logic, conceived the nature and scope of induction far too narrowly, and pragmatism deserves credit for having remedied this defect. Induction, though it cannot give complete certainty, underlies all the sciences, even pure mathematics. In any science, we have a collection of facts bound together (as far as possible) by general laws. The facts appear, in the formal exposition, as deductions from the laws; this, at least, holds 20 for the most advanced sciences, such as mathematics and physics. But in reality the laws are inductions from the facts. We cannot say that this or that fact proves this or that law: the whole body of facts proves (or, rather, renders probable) the whole body of laws. It might be thought that, in an experimentum crucis, a single fact establishes a single law; but this is only the case so long as the other laws of the science are taken for granted. If other facts should lead us to doubt the other laws, the interpretation of our experimentum crucis might be wholly changed. Thus the justification of a science is that it fits all the known facts, and that no alternative system of hypotheses is known which fits the facts equally well. We may therefore 30 say truly that scientific theories are adopted simply because they work, i.e. because their consequences are satisfactory. Thus it would appear as though a right analysis of scientific induction led us straight to the pragmatic test of truth. Certain objections to this conclusion, however, at once suggest themselves. In the first place, scientific induction assumes certain data, the "facts" with which our theories have to agree. That the heavenly bodies have the apparent positions, in the sky, which we perceive them to have, is not proved by astronomy, but is assumed as the datum upon which astronomy proceeds. It would seem, therefore, that there are truths of fact 40 which are prior to the whole inductive procedure, and that these truths of fact must be "true" in some other sense than that the consequences of supposing them true are satisfactory. To this argument pragmatists reply

272

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

that what really is "fact" is neither true nor false, but prior to the whole antithesis of truth and falsehood. Day follows day, and its contents are simply added. The new contents themselves arc not true, they simply come and are. Truth is what we say about them, and when we say that they have come, truth is satisfied by the plain additive formula. 10 Pragmatists contend, therefore, that the mere recognition of facts is the simplest case of the application of their formula. If all "truth" were of this simple nature, the pragmatist doctrine would be unnecessary, though Io there would be nothing to show that it was false. But the "truths" which do ~ot consist in the mere recognition of facts cannot, according to pragmatism, be explained in this simple way; hence we are forced to adopt a theory of truth not derived from the exclusive consideration of this simplest case. For the moment let us allow this answer to pass. We shall return to the subject of "facts" in connection with Dr. Schiller's doctrine of the making of reality. A more serious objection to the argument from the procedure of the sciences is derived from the ambiguity of the conception of "working". What science requires of a working hypothesis is that it shall work theoreti20 cally, i.e. that all its verifiable consequences shall be true, and none false. The. law of gravitation enables us to calculate the motions of the heavenly b~dies: so far as these motions can be observed, they are found to agree with our calculations. It is true that the heavenly bodies have such and such apparent positions at such and such times, and the law of gravitation agrees with this truth. This is what we mean when we say that the law "works". We do not mean that it gives us emotional satisfaction that it satisfies our aspirations, that it is a help in navigation, or that it fa~ilitates a virtuous life. Any or all of these may be true, but they are irrelevant; if they were all false, we should still say that the law "works", because it 30 agrees with observed facts. Thus the kind of "working" which science desiderates is a very different thing from the kind which pragmatism considers to be the essence of truth. To this, as to our previous objection, pragmatists reply that the "truth" concerned is a particular species of "truth", and that scientific working is a particular species of their general conception of working. Our purpose, they say, in asking the question to which the law of gravitation is an answer, is to be able to calculate the motions of the heavenly bodies. The law of gravitation furthers this purpose, and is therefore true in the pragmatic sense. This answer shows that the procedure of science, so far, has 40 not been shown to contradict pragmatism; but it does not show that the IO

James, Pragmatism, p.

62.

21 PRAGMATISM

273

procedure of science positively supports pragmatism. Where, as in science, our purpose is to discover truth, an answer which furthers our purpose will be true. But from this truism it cannot be inferred (as pragmatists pretend) that if we had had some quite different purpose, an answer which furthered it would still have been true. Another objection to the argument from "working hypotheses" is that by men of science these are explicitly contrasted with established truths. An hypothesis, as experience shows, may explain all known relevant facts admirably, and yet may at any moment be rendered inadequate by new facts. For this reason, prudent men give only a very provisional assent to Io a working hypothesis. Thus the cases from which pragmatism endeavours to discover the nature of truth are the very cases in which we have least assurance that truth is present at all. This is certainly a curious and not very hopeful mode of procedure. It may be said, however, that what leads us to feel doubtful about a working hypothesis is merely that it has not yet been shown to work over a sufficiently wide field: the more it works, the more we believe in it. But to this again it may be rejoined that the more it works the less probability is there that any other hypothesis would also work. To pursue this topic, however, would require a discussion of the laws of probability, for which this is not the place. 20 From what has been said it results that the utmost that pragmatism can derive from science is that the scientific conception of working is not incompatible with the pragmatist conception, since the scientific working may be regarded as a species of the pragmatic working. It is, however, a species whose differentia adds just those elements which other philosophies declare to be necessary to truth, while pragmatism declares them to be unnecessary. The essential novelty of pragmatism is that it admits, as a ground of belief, any kind of satisfaction to be derived from entertaining the belief, not merely the theoretic satisfaction which is sought by science. For this contention no support whatever is to be found in science. Let us 30 see whether any support is to be found elsewhere. Pragmatists are never weary of inveighing against those who say that our beliefs ought not to be influenced by considerations which in fact do influence them. They point triumphantly to the influence of desire upon belief, and boast that their theory alone is based upon a true psychological account of how belief arises. With this account we have no quarrel; what we deny is its relevance to the question: What is meant by "truth" and "falsehood"? At first sight it might seem a perfectly proper inductive proceeding to inquire what properties a belief must have in order that we may call it true, and to infer that those properties constitute the meaning of .;o "truth". There is, however, a fallacy in this method of inquiry; and this fallacy, in our opinion, is at the bottom of the whole pragmatist philosophy.

274

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

21 PRAGMATISM

There is, in the first place, an ambiguity in the word "meaning". We may say "that cloud means rain", or we may say "pluie means rain". It is obvious that these two senses of "meaning" are wholly different. What they have in common is that in each case we have one thing which points to another. The cloud is a sign that rain is coming; the word pluie is a sign which signifies rain. But beyond this, the two senses of "meaning" have little in common. In the first sense, one thing "means" another when the existence (past, present, or future) of the other can be inferred from the one, i.e. when there is a causal connection between them. In the second sense, "meaning" is confined to symbols, i.e. to words, and whatever other ways may be employed for communicating our thoughts. It is this second sense of "meaning" which we expect a dictionary to give us. When we ask "what does such and such a word mean?" what we want to know is "what is in the mind of a person using the word?" A confusion of the two senses of "meaning" is not uncommon in philosophy; and, if we are not mistaken, pragmatism has confused them in its inquiry as to the "meaning" of truth. It has discovered something which has a causal connection with our beliefs that things are true, and which, therefore, in the first sense of "meaning", may be taken to be what these beliefs "mean". It has then supposed that this is what is "meant", in the second sense, by "truth", i.e. what we have in mind (or should have in mind?) when we use the word "truth". This confusion between the two senses of "meaning" seems to be necessarily involved in the method adopted by pragmatists, namely the method which inquires into the causes of our judging things to be true, in the hope of thereby discovering what "truth" means. Let us grant to the pragmatists, in order to avoid disputes concerning what is unimportant, that what causes people to judge that a belief, about which a doubt has arisen, is true, is the fact that this belief is found to further the purposes which led us to inquire into its truth. Then to judge that a belief is true "means" that this belief furthers our purposes, in the sense in which the cloud "means" rain, i.e. there is a causal connection between them. But truth is not the same thing as furthering our purposes any more than the cloud is the same thing as rain. When we say that a belief is true the thought we wish to convey is not the same thought as when we say that the belief furthers our purposes; thus "true" does not mean "furthering our purposes" in the sense in which "pluie" means rain. Thus pragmatism docs not answer the question: What is in our minds when we judge that a certain belief is true? We find pragmatists, when pressed, willing to admit this fact. Thus Dr. Schiller says: 11

In a sense, therefore, the predications of "good" and "bad," "true" and "false," etc., may take rank with the experiences of "sweet," "red," "loud," "hard," etc., as ultimate facts which need be analysed no further.

l l

Studies in Humanism, p. 144.

275

To which he adds, in a footnote, The purport of this remark is to confute the notion, which seems dimly to underlie some intellectualist criticisms, that the specific character of the truth-predication is ignored in pragmatist quarters. This fundamental meaning of "truth" is treated by Dr. Schiller as unimportant because it does not enable us to distinguish the cases in which we have rightly predicated truth from those in which we have done so wrongly. The pragmatist test, he maintains, enables us to distinguish the truly true from the falsely true. An untested predication of truth he calls "truth as claim"; a predication which is subsequent to the application of the pragmatist test he calls "truth validated". The distinction between the two is treated at length in his essay on "the ambiguity of truth" . 12 This "ambiguity" appears to us to be wholly non-existent. The distinction involved is the distinction between what is true and what is thought to be true. The reader who will, throughout this essay on the ambiguity of truth, substitute "butter" for "truth" and "margarine" for "falsehood", will find that the point involved is one which has no special relevance to the nature of truth. There is "butter as claim", i.e. whatever the grocer calls butter; this, we will suppose, includes margarine. There is "butter validated", which is butter that, after the usual tests, has been found not to be margarine. But there is no ambiguity in the word "butter". When the grocer, pointing to the margarine, says "this is butter" he means by "butter" precisely what the customer means when he says "this is not butter". To argue from the grocer's language that "butter" has two meanings, one of which includes margarine, while the other does not, would be obviously absurd. Similarly when the rash man, without applying any tests, affirms "this belief is true", while the prudent man, after applying suitable tests, judges "this belief is not true", the two men mean the same thing by the word "true", only one of them applies it wrongly. Thus Dr. Schiller's reasons for regarding "the specific character of the truth-predication" as unimportant are not valid. We must now return to the two senses of "meaning", and show how they are relevant to our problem. It is evident that, in the sense in which 12

Studies in Humanism, pp.

141-62.

IO

20

30

276

ro

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

the meaning of a word is "what is in our minds when we use the word", the meaning of the word "truth" is just that "specific character of the truth-predication" which, as Dr. Schiller confesses, is something quite other than "furthering our purposes". His contention is that the beliefs of which we can predicate truth truly are those which further our purposes. And his reason for saying this is that the beliefs which further our purposes are those which we persist in calling true after reflection. But that only proves that these are the beliefs which we continue to think true, not that these are the beliefs which are true. Owing, however, to confusion of the two senses of "meaning", he is led to argue that usefulness gives the meaning of truth, and that therefore when a belief is useful it must be true. All that really follows, if we grant the whole of the psychological argument, is that beliefs which are found to be useful will continue to be thought to be true. This is an entirely different proposition, and one which, by itself, throws no light whatever either upon the nature of truth or upon what beliefs are in fact true. It may well be that beliefs which fulfil certain purposes are true, while beliefs which fulfil others are not true; or, again, that there is no connection whatever between truth and usefulness. Dr. Schiller's argument (and William James's, for the two are practically identical on this point) involves a variety of the very assumption which he criticizes in others, namely the assumption that all our beliefs are true. In pragmatism the assumption is that the beliefs which we persist in holding must be true. It is then pointed out how very unreasonable our grounds often are for persisting in a belief, and this fact, instead of being used to throw doubt on the belief, is used to discredit reasonableness. Thus we are brought back to the standpoint of "The Will to Believe", and we find that the precepts of that essay really underlie the whole pragmatist theory of truth. But the superstructure is so vast that pragmatists appear to be no longer aware of the foundations upon which their edifice is reared. We may now restate the pragmatist theory of truth in bald outline, giving due prominence to presuppositions of which pragmatists themselves are perhaps not fully conscious. Their major premiss is: Beliefs which persist after a doubt has been raised are true. Their minor premiss is: Beliefs which are found to be serviceable persist after a doubt has been raised. Hence it follows that such beliefs are true. The pragmatist then turns round and exhorts us to cherish such beliefs, on the ground that they are true. But if his psychology was right the exhortation is needless, since, by his minor premiss, we certainly shall cherish such beliefs. His major premiss should be: "Beliefs which we cherish after you have raised a doubt are true." But those who have raised the doubt can hardly be expected to be much impressed by this premiss. The argument is a form of the old refutation of an opponent by the contention that the whole

21 PRAGMATISM

277

human race thinks as you do, which is a somewhat unsuccessful weapon against a human being who does not think as you do. It is now time to turn our atte1,1tion to the metaphysic which Dr. Schiller has based upon the pragmatist theory of truth. Pragmatism as such professes to be only a method; the metaphysical doctrine which Dr. Schiller derives from it he calls Humanism. In regard to metaphysics, pragmatism professes to be a kind of universal provider, willing and able to suit all tastes. As William James puts it: Against rationalism as a pretension and a method pragmatism is fully armed and militant. But, at the outset, at least, it stands for no particular results. It has no dogmas, and no doctrines save its method. As the young Italian pragmatist Papini has well said, it lies in the midst of our theories, like a corridor in a hotel. Innumerable chambers open out of it. In one you may find a man writing an atheistic volume; in the next some one on his knees praying for faith and strength; in a third a chemist investigating a body's properties. In a fourth a system of idealistic metaphysics is being excogitated; in a fifth the impossibility of metaphysics is being shown. But they all own the corridor and all must pass through it if they want a practicable way of getting into or out of their respective rooms. 13 In spite of this catholicity, however, we agree with Dr. Schiller in thinking that his metaphysic is the one which naturally results from pragmatism. It will be remembered that, in considering induction, we pointed to the dependence of inductive verification upon an appeal to "facts". Humanism, as a metaphysic, results from the application of the pragmatic method to the question: What is a "fact"? This subject has been treated by Dr. Schiller in his essay on "the making of reality" . 14 The main purpose of humanist metaphysics is to emphasize the primacy of the Will. The Will, it is true, requires a datum of "fact" to which to apply its operations, but this datum is itself the product of previous volitions, and although we cannot quite deny some original vAYJ which has been moulded by will, yet this is remote and unimportant, and has been transformed into genuine reality by the agency of human beings and other beings more or less resembling them. Nothing that can be known, nothing that can properly be called "real", is independent of the knower. There is no such thing as "mere" knowing, in which we passively apprehend 13 Pragmatism, p. 54. 14 Studies in Humanism, pp. 421-51.

IO

20

30

278

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

21 PRAGMATISM

the nature of a merely "given" object. All knowing is bound up with doing, and everything that we know has been in some degree altered by our agency. This, Dr. Schiller says, is obvious in the case of our acquaintances, who plainly are more or less affected by the fact that we arc acquainted with them. When we say that something is "independent" of our knowing, we mean, according to him, that the thing is not aware that we know it. But, as a matter of fact, everything we know, even a stone, is aware of us in its own way. To the charge that this is Hylozoism, Dr. Schiller replies by admitting it. The grounds for these opinions are not set forth quite so clearly as could be wished, but we may gather them from a complimentary allusion to Hegel's dialectic at the beginning of the Essay. Imagine some "fact" in regard to which we entertain a belief. The belief leads to action, and the action alters the "fact". If it alters it into harmony with our wishes the belief is proved to have been what pragmatists call "true", since it has proved successful in action. In this case, since the belief in the fact is true, it follows that the fact is real. Thus the belief has made the fact. But if the outcome of the belief is a "fact" which, though in harmony with the wishes which originally led us to concern ourselves with the matter, is in conflict with others of our wishes, the belief is not "true" as regards these other wishes; hence we shall have to change our belief, and take fresh action on the new belief, and so bring the "fact" into harmony with these new wishes. In this way, so long as we have any unsatisfied wishes, we are led on in a cycle of beliefs and actions, the beliefs becoming gradually "truer", and the "facts" with which the beliefs are concerned becoming gradually more "real" as greater harmony is established between the "facts" and our wishes. The motive power of this whole development is the pragmatic definition of truth. For if we believe A to be a fact, that belief is true if it is successful as a means to satisfying our wishes; hence so long as our wishes are not completely satisfied, the belief that A is a fact is not completely true, and therefore A is not completely a fact. Thus complete truth and complete reality go hand in hand, and both are only to be found at the end of the road which leads to the complete satisfaction of all our wishes. The similarity of the above process to the Hegelian dialectic is emphasized by Dr. Schiller; with his inveterate love of a pun, he has christened his process "trialectic". He does not seem, however, to have observed that his process, like Hegel's, introduces a distinction between appearance and reality; that appearance embraces the whole of the world as we know it, and that it is only to reality that the pragmatic test of truth applies. The "facts" which he can accept as real must be such as not to thwart our purposes; the "facts" which appear are very often such as to thwart our purposes. If a fact is such as to thwart our purposes, the pragmatist test

of truth is not fully applicable to it; for by believing that it will thwart our purposes, we do not prevent it from doing so, and our belief, though possibly preferable pragmatically to any other, does not secure the satisfaction of our desires. If, on the other hand, we believe that the fact is not such as to thwart our purposes, we believe what, ex hypothesi, is not the case. Hence it follows that such facts cannot be real. Since many apparent facts thwart our purposes, we are led to distinguish between real and apparent facts. Hence it is not here on earth that pragmatism applies, but only in Dr. Schiller's heaven, just as it is only in Mr. Bradley's heaven that Mr. Bradley's metaphysic applies. The whole doctrine, therefore, reduces itself to the proposition that it would be heavenly to live in a world where one's philosophy was true, and this is a proposition which we have no desire to controvert. The distinction between appearance and reality is one which Dr. Schiller is never weary of attacking; indeed, a very large proportion of his writings is directed against it. His complete reality, he holds, is being progressively realized, and is not, like the Absolute, something wholly unconnected with our actual world of appearance. But his only reason for supposing that his complete reality is being progressively realized is a tacit assumption of cooperation among the agents composing the universe. He assumes, that is, that the various desires which (according to him) form the motive power of all that occurs in the universe are not such as to counteract each other: the world's activities are not to be conceived as a tug-of-war. For this view there is, we fancy, no argument except the pragmatic argument, that it is pleasant and cannot be conclusively disproved. Thus the whole humanist metaphysic rests upon the pragmatic theory of truth, and falls with that theory. Moreover, it introduces, in a slightly modified form, the old distinction of appearance and reality, of which the difficulties have been admirably set forth by Dr. Schiller himself. Since the distinction and, therefore, the difficulties, result inevitably from the pragmatic theory of truth, they afford a new argument against that theory; for they show that the theory is applicable, not to our actual world, but to an ideal world where all the hopes of pragmatists have been realized. Although, for the reasons alleged above, we do not ourselves accept the pragmatist philosophy, we nevertheless believe that it is likely to achieve widespread popularity, because it embodies some of the main intellectual and political tendencies of our time. This aspect of pragmatism deserves consideration, since the influence of a doctrine (as pragmatists have very prudently pointed out) is by no means proportional to its intellectual value. On the intellectual side, pragmatism embodies scepticism, evolution, and the new insight into the nature and scope of scientific induction. On the political side, it embodies democracy, the increased belief in human

279

IO

20

30

40

280

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

21 PRAGMATISM

power which has come from the progress of mechanical invention, and the Bismarckian belief in force. The scepticism embodied in pragmatism is that which says "Since all beliefs are absurd, we may as well believe what is most convenient." This is by no means a new contention; in England it has been popularized by Mr. Balfour's Foundations of Belief and Notes on Insular Free Trade. Scepticism is of the very essence of the pragmatic philosophy: nothing is certain, everything is liable to revision, and the attainment of any truth in which we can rest securely is impossible. It is, therefore, not worth while to trouble our heads about what really is true; what is thought to be true is all that need concern us. Instead of the old distinction between true and false, we adopt the more useful distinction between what we persist in thinking true, and what merely seems true at first sight. Later on, the old meanings of true and false may slip back unnoticed, and we may come to think that what is true in the pragmatic sense is true in the old sense also; this happens especially in regard to religion. But on pragmatist principles, there is no reason to regret this; for the "true" is what it is useful to believe, and therefore it is useful to believe what pragmatism declares to be true. Scepticism, therefore, though necessary at the start, must be banished later on if we are to get the full benefits of pragmatism. In this there is no great psychological difficulty, since, as Hume confessed, the sceptical attitude is one not easily maintained in practice. The philosophy of evolution has also had its share in generating the pragmatic tone of mind. It has led people to regard everything as fluid and in process of development, everything as passing by imperceptible gradations into everything else. Some biologists, it is true, have begun to regard development as discontinuous, proceeding by the sudden appearance of freaks; but philosophers and the general public have not been influenced by this change. Hence it has come to be felt that all sharp antitheses, such as that of true and false, must be blurred, and all finality must be avoided. We must always build a road by which everything can pass into everything else at a leisurely pace and with small steps. Instead of "the true" we shall have "the more true", or "the most true up to date". And between different claimants for truth, we must provide a struggle for existence, leading to the survival of the strongest. All this is admirably effected by the pragmatic theory of truth. M. Bergson, whom pragmatists claim as an ally, may be regarded as embodying this tendency. The influence of modern theories of scientific induction has probably been more restricted, in point of numbers, than the influence of scepticism or of evolution, but the men influenced have been important by their scientific eminence. We may take as their protagonist M. Poincare, who, while not extending the pragmatist doctrine to particular facts, has dealt in a thoroughly pragmatic spirit with the general hypotheses of logic,

mathematics, and physics, showing that what leads to the acceptance of a scientific hypothesis is its convenience. Such general assumptions as causality, the existence of an external world, etc., cannot be supported by Mill's canons of induction, but require a far more comprehensive treatment of the whole organized body of accepted scientific doctrine. It is in such treatment that the pragmatic method is seen at its best; and among men of science, its apparent success in this direction has doubtless contributed greatly to its acceptance. The influence of democracy in promoting pragmatism is visible in almost every page of William James's writing. There is an impatience of authority, an unwillingness to condemn wide-spread prejudices, a tendency to decide philosophical questions by putting them to the vote, which contrast curiously with the usual dictatorial tone of philosophic writings. Dr. Schiller at one time set to work to elucidate the question of a future life by taking a poll. 15 William James claims for the pragmatist temper "the open air and possibilities of nature, as against dogma, artificiality, and the pretence of finality in truth". A thing which simply is true, whether you like it or not, is to him as hateful as a Russian autocracy; he feels that he is escaping from a prison, made not by stone walls but by "hard facts", when he has humanized truth, and made it, like the police force in a democracy, the servant of the people instead of their master. The democratic temper pervades even the religion of the pragmatists: they have the religion they have chosen, and the traditional reverence is changed into satisfaction with their own handiwork. "The prince of darkness", James says, "may be a gentleman, as we are told he is; but whatever the God of earth and heaven is, he can surely be no gentleman." 16 He is rather, we should say, conceived by pragmatists as an elected president, to whom we give a respect which is really a tribute to the wisdom of our own choice. A government in which we have no voice is repugnant to the democratic temper. William James carries up to heaven the revolt of his New England ancestors: the Power to which he can yield respect must be a George Washington rather than a George III. Closely connected with this democratic spirit is the belief in human power, which is one of the dominant notes of pragmatism. By the progress of mechanical invention, the possibilities of our command over nature have been shown to be much greater than they were formerly supposed to be, and no definite limits can be set to them. Hence has arisen-especially in America, where the economic conditions are favourable, and

281

IO

20

30

15 See his essay on "The Desire for Immortality" (Humanism, pp. 228-49). We do not, of course, suggest that he would have considered the result of the poll decisive, even if the 40 electorate had been larger. 16 Pragmatism, p. 72.

282

ro

20

30

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

21 PRAGMATISM

the chief concern of most people is with those matters in which recent advances have been greatest-a general feeling that by energy and hope all obstacles can be overcome, and that it is a mark of laziness or pusillanimity to admit that anything is impossible. The habit of mind which believes that there are no essential impossibilities has been fostered by the doctrine of evolution, with its literary corollary of the U ebermensch. Hence have arisen a self-confidence and a pride of life which in many ways remind one of the Renaissance, and establish some affinity between historical humanism and its modern namesake. For the modern humanism is essentially the philosophy which is appropriate, as Dr. Schiller himself has said, to "the young, the strong, the virile" . 17 The inventor, the financier, the advertiser, the successful man of action generally, can find in pragmatism an expression of their instinctive view of the world. Such men, both for good and evil, expect the world to be malleable to their wishes, and in a greater or less degree find their expectation justified by success. Hence arises a disbelief in those "hard facts" which pragmatists tend to deny, and a confidence of victory in contests with the outer world, whether these contests be cognitive or more directly practical. An Italian pragmatist has expressed this confidence in victory as follows:

The worship of force, as we find it in Nietzsche, is not to be found in the same form in William James, who, though he lauds the will and the life of action, does not wish action to be bellicose. Nevertheless, the excessive individualism of the pragmatic theory of truth is inherently connected with the appeal to force. If there is a non-human truth, which one man may know, while another does not, there is a standard outside the disputants, to which, we may urge, the dispute ought to be submitted; hence a pacific and judicial settlement of disputes is at least theoretically possible. If, on the contrary, the only way of discovering which of the disputants is in the right is to wait and see which of them is successful, there is no longer any principle except force by which the issue can be decided. It is true, of course, that in a private dispute the public opinion of the community, especially as embodied in the law, will usually compel a peaceful decision. But this public opinion is formed (at least in theory) upon an objective estimate of the rights and wrongs of the case; in place of this, if pragmatism were the accepted creed, public opinion would have to be guided by the interests of the community. To this there would be no objection if, as would commonly be done, the maintenance of justice could be taken as one of the ends which it is in the interest of the community to pursue. But in a pragmatist community, this would be impossible, since justice is derivative from the interests of the community, and not an independent constituent of those interests. In international matters, owing to the fact that the disputants are often strong enough to be independent of outside control, these considerations become more important. If the pragmatist urges that always and everywhere the only ultimate arbiter in a dispute must be force, the reply is that, although this is true at the actual moment of the battle, it is yet not true in a wider sense, since it ignores the motives which generate the force on either side. The hopes of international peace, like the achievement of internal peace, depend upon the creation of an effective force of public opinion formed upon an estimate of the rights and wrongs of disputes. Thus it would be misleading to say that the dispute is decided by force, without adding that force is dependent upon justice. But the possibility of such a public opinion depends upon the possibility of a standard of justice which is a cause, not an effect, of the wishes of the community; and such a standard of justice seems incompatible with the pragmatist philosophy. This philosophy, therefore, although it begins with liberty and toleration, develops, by inherent necessity, into the appeal to force and the arbitrament of the big battalions. By this development it becomes equally adapted to democracy at home and to imperialism abroad. Thus here again, it is more delicately adjusted to the requirements of the time than any other philosophy which has hitherto been invented.

Dio e perfetto perche e onnipossente. Sostituiamo dunque al misticismo della rinunzia, dell' Imitazione di Cristo, ii misticismo della conquista, dell' Imitazione di Dio. 18 Other pragmatists have been less explicit than this modern Thomas a Kempis, but he has correctly expressed the spirit of their philosophy. From the confidence of victory in contests, it is an easy passage to the love of contest. For this, pragmatism provides full scope. The many different "truths as claim" must fight it out among themselves, and the victor will become "truth validated". Dr. Schiller on one occasion implicitly confesses that, with his theory of truth, persecution can actually make a doctrine true which would otherwise be false, since it can make a doctrine "useful to our lives" . 19 In the absence of any standard of truth other than success, it seems evident that the familiar methods of the struggle for existence must be applied to the elucidation of difficult questions, and that ironclads and Maxim guns must be the ultimate arbiters of metaphysical truth. 17 Humanism, p. viii.

40

18 Leonardo, April 1905, L:Imitazione d'iddio, p. 64. 19 Humanism, Pp. 59-60: "Delicate questions may arise out of the fact that not only does what works receive social recognition, but also that what receives social recognition for this very reason largely works."

283

ro

20

30

40

284

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

To sum up: Pragmatism appeals to the temper of mind which finds on the surface of this planet the whole of its imaginative material; which feels confident of progress, and unaware of non-human limitations to human power; which loves battle, with all the attendant risks, because it has no rc~l doubt that it will achieve victory; which desires religion, as it desires railways and electric light, as a comfort and a help in the affairs of this world, not as providing non-human objects to satisfy the hunger for perfection and for something to be worshipped without reserve. But for those who feel that life on this planet would be a life in prison, if it were not ro for the windows into a greater world beyond; for those to whom a belief in man's omnipotence seems arrogant, who desire rather the Stoic freedom that comes of mastery over the passions than the Napoleonic domination that secs the kingdoms of this world at its feet-in a word, to men who do not find Man an adequate object of their worship, the pragmatist's world will seem narrow and petty, robbing life of all that gives it value and making Man himself smaller by depriving the universe which he con~ templates of all its splendour.

22

The Philosophy of William James [1910]

THIS PAPER WAS published in The Nation, 7 (3 Sept. 19rn): 793-4. It was reprinted in The Living Age, Boston, o.s. 267 (I Oct. 19rn): 52-5. The occasion for writing this paper was the death, on 27 August 19rn, of James. Although no correspondence proving as much survives, it seems likely that the editor of The Nation invited Russell to pay tribute to James. The copy-text is the earlier printed version.

285

22 THE PHILOSOPHY OF WILLIAM JAMES HE GREAT LOSS which philosophy has sustained by the death of William James will be felt as also a personal loss by all who knew him. He was one of the most eminent, and probably the most widely known, of contemporary philosophers. It was by his work on psychology that he first achieved fame, but his later years were devoted almost entirely to the advocacy of the philosophy known as pragmatism. The high value of his work on psychology is universally admitted, but his work on pragmatism is still the subject of acute controversy. The same qualities of mind appear in both, but those who do not accept his general philosophy would contend that he was at his best where he had concrete facts to deal with, and that in the somewhat rarefied atmosphere of metaphysics his powers failed to find full scope. Abstract argumentation appeared to him futile, and subjects which require it were therefore uncongenial to him. His mind worked by flashes of brilliant insight, by an aptitude for fresh and untrammelled observation, and by a singular freedom from reigning academic prejudices. Unlike most professors, he saw facts first, instead of first seeing theories and then searching out facts to confirm or refute them. In psychology, this resulted in a very concrete presentation of mental phenomena, illustrated with a wealth of everyday observation which, however it might disconcert the learned, made his Principles of Psychology by far the most delightful and readable book on its subject. For example, in illustrating the proposition that our self-esteem depends not simply upon our success, but upon the ratio of our success to our pretensions, and can therefore be increased by diminishing our pretensions, he observes:

T

IO

20

30

Many Bostonians, crede experto (and inhabitants of other cities, too, I fear), would be happier women and men to-day, if they could once for all abandon the notion of keeping up a Musical Self, and without shame let people hear them call a symphony a nuisance. How pleasant is the day when we give up striving to be young-or slender! Thank God! we say, those illusions are gone.

In his study of the mind, William James refuses to begin, as psychologists are apt to do, with sensations, because a mere sensation is an abstraction which never really occurs. He begins instead with "the stream of thought", taking thought as we find it, with all kinds of thinking mixed in a vague continuum. This leads him to a long and very able discussion of "the consciousness of self", in which he reduces the Self to the passing thought, with its memory of other thoughts and its consciousness of the body. "All attempts", he says, to explain our phenomenally given thoughts as products of deep286

287

er-lying entities (whether the latter be named "Soul", "Transcendental Ego", "Ideas", or "Elementary Units of Consciousness") are metaphysical. This book consequently rejects both the associationist and the spiritualist theories; and in this strictly positivistic point of view consists the only feature of it for which I feel tempted to claim originality. "Of course", he continues, "this point of view is anything but ultimate." This concession, however, belongs to the period before he had discovered pragmatism; indeed, the attempt to make such a point of view ultimate constitutes an essential element in the pragmatist's philosophy. William James's best account of his later views is Pragmatism, a New Name for Some Old Ways of Thinking (Longmans, 1907). The essence of pragmatism is its theory as to what constitutes "truth". Some men believe one thing and some another, and metaphysical disputes seem interminable. Pragmatism contends that it has discovered a way of deciding such disputes. Examine the consequences of the two beliefs; if one has good consequences, while the other has bad ones, the one with good consequences is "true", at least until a third view with still better consequences has been discovered. And not only is this a test of truth; it is the actual meaning of truth. Hence all truth is experimental, and is at best true up to data; no truth is immutable or permanently certain. "Pragmatism" (we are told) represents a perfectly familiar attitude in philosophy, the empiricist attitude, but it represents it, as it seems to me, both in a more radical and in a less objectionable form than it has ever yet assumed. A pragmatist turns his back resolutely and once for all upon a lot of inveterate habits dear to professional philosophers. He turns away from abstraction and insufficiency, from bad a priori reasons, from fixed principles, closed systems, and pretended absolutes and origins. He turns towards concreteness and adequacy, towards facts, towards action and towards power. This is the philosophy of those who love the battle better than the victory. Felicity, Hobbes tells us, consists in prospering, not in having prospered, and so knowledge is to consist in discovering, not in having discovered. James is never tired of inveighing against the great static systems of metaphysics: he would have us be for ever building our house and never living in it. A Western energy seems to be required in order to find full satisfaction in this philosophy: those whose energy sometimes flags will sigh for moments of contemplation to temper the monotony of unceasing action.

IO

20

30

288

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

But pragmatism has another aspect, also very important; I mean, its religious aspect. In the philosophy of religion, it is allied with Modernism. While unable to defend the rigid and unchanging dogma of Catholic orthodoxy, it is anxious to urge that any religion, so long as it is useful in promoting happiness or virtue, is to be considered "true" in the times and places of its utility. The classic philosophies have debated endlessly the old a priori arguments for and against free will or the existence of God, or the immortality of the soul: pragmatism, instead, will inquire into the effects of such beliefs on emotion and conduct. "On pragmatic prinro ciples", James says in his chapter on "Pragmatism and Religion", we can not reject any hypothesis if consequences useful to life flow from it. Universal conceptions ... have, indeed, no meaning and no reality if they have no use. But if they have any use they have that amount of meaning. And the meaning will be true if the use squares well with life's other uses. To most people, such a utilitarian basis will seem inadequate, but it has at least the merit of producing a sympathetic attitude towards whatever faith men really live by. This attitude is seen at its best in one of James's most interesting books, Varieties of Religious Experience. This book is filled 20 with the pragmatist temper. Religious men of all types are described, and an attempt is made to show wherein their religion was useful to them. Even the types naturally most repugnant to the author are treated with kindly insight, and we are invited to admire the happiness which they derived from their various beliefs. But to say that this suffices to prove that their beliefs were in any way true will remain a paradox to many, in spite of all the charm and persuasiveness of William James's exposition. Unlike many pragmatists, William James had a realistic temper of mind, and even a certain tendency towards materialism. The plain facts of common sense were the starting-point of his philosophy, and, though 30 he was willing to submit them to a certain amount of interpretation, he could never have tolerated a philosophy which, like Hegelian idealism, regards them all as appearances differing toto caelo from Reality. His materialistic tendencies are apparent in various parts of his Psychology, particularly in the much-discussed theory of the emotions, according to which we are pleased because we smile, or sad because we weep; in his own words: "The bodily changes follow directly the perception of the existing fact, and ... our feeling of the same changes as they occur is the emotion." But this materialistic tendency, which belonged to the scientific side of his intellect, was held in check by a strong religious impulse, and by a 40 vehement rebellion against the intellectual tyranny of hard-and-fast theories as to how things must be. He was determined to vindicate human

22 THE PHILOSOPHY OF WILLIAM JAMES

289

liberty in the domain of truth as in other domains; and it was largely this feeling that made pragmatism attractive to him. Every philosophy has its merits and its demerits. Among the unquestionable merits of pragmatism, apart from the useful work it has done in challenging sleepy philosophic orthodoxies, are its inductive temper, its reliance on experiment, its readiness to question even what seems most certain. These merits have recommended it to many men of science; but in philosophy, it may be doubted whether pragmatism is in reality less dogmatic than the systems it seeks to replace. To take it as certain that not even the multiplication table contains final and infallible truth is to adopt a dogma which, in philosophy, may be just as great a bar to openmindedness as any other dogma. Indeed, to have a philosophy at all is necessarily to abandon something of that experimental and hypothetical temper which pragmatists praise. But while we may doubt whether pragmatism is either so true or so useful as its advocates believe, we cannot but admire the urbanity, the large tolerance, and the humanity with which, both in public and in private, it was invariably advocated by William James.

ro

23

Review of James's Memories and Studies [1911]

Memories and Studies. By William James. New York, London and Bombay: Longmans, Green, 191 I. Pp. 41 I. H Is Boo K Is a collection o~ addresses ~~d short art~cles on several occasions. Some, on Amencan celebnttes of varymg degrees of fame, are of chiefly local or topical interest. Two on psychical research give William James's final verdict in this matter, based (in part at least) on the maxim that "when a man's pursuit gradually makes his face shine and grow handsome, you may be sure it is a worthy one". This suggests that mankind should be annually inspected by a Government official, selected as an expert in beauty, and having power to send the ugliest ro ten per cent to prison. It is to be feared that chimney-sweeps, coal-miners and stokers must be incredible villains. There is an amusing essay on Herbert Spencer's autobiography, and a very interesting account of the psychological effects of the San Francisco earthquake. There is an article called "A Pluralistic Mystic", which is really in praise of the worship of Dionysus in his nitrous-oxide incarnation (he has taken advantage of modern science and is quite abreast of the age). The most delightful part of the book consists of an essay and a speech on war-especially the speech, called "Remarks at the Peace Banquet". Since William James got away alive, one must suppose the other diners 20 practised what they preached. He told the assembled pacificists that "the plain truth is that people want war. They want it anyhow; for itself, and apart from each and every possible consequence. It is the final bouquet of life's fireworks." "War is human nature at its uttermost. ... It is a sacrament. Society would rot without the mystical blood-payment." "Let the soldiers dream of killing, as the old maids dream of marrying." The companion essay, "On the Moral Equivalent of War", suggests peaceful methods of securing the same moral values. William James did well in insisting on the urgency of this problem; men's energies need an enemy to fight, but all progress demands that the enemy should not be human. 30

T

Tms REVIEW w AS published in The Cambridge Review, 33 (16 Nov. 1911): 118. It was reprinted in Volume 12 of The Collected Papers of Bertrand Russell (1985); it is included in the present volume in order to bring all of Russell's writings on William James during these years together in one place. On II November 1911 Russell wrote to Lady Ottoline Morrell: "This morning I have been writing my review of W. James. Quite short, as the book is unimportant" (#252). In an earlier letter, on 17 July 1911, he noted that, during his visit to Peppard, "I can finish W. James's posthumous book, which I am reading. It is pleasant reading, like everything of his" (#150). This may refer to the book under review or it may refer to Some Problems of Philosophy, which was published in England on 3 June 191 I. The copy-text is the printed version.

290

291

24 PRAGMATISM AND LOGIC

24

Pragmatism and Logic [1912]

Tms REVIEW WAS published in The Nation, II (I8 May I9I2): 258-9. It appeared on Russell's fortieth birthday. Russell first mentioned this book to Lady Ottoline Morrell on 23 February I9I2: I am in a state of fury because Schiller has sent me a book on Formal Logic which he has had the impertinence to write. He neither knows nor respects the subject, and of course writes offensive rot. I am already thinking of all the jokes I will make about the book if I have to review it. I don't really dislike Schiller. I am the only human being who doesn't-because though he is a bounder and a vulgarizer of everything he touches, he is alive, adventurous and good-natured. So I don't feel venomous about him as I do about Bergson. If I could love Bergson I should have achieved a really Christian spirit-but I am miles away from it as yet. I also prefer Schiller to the conventional Oxford philosophers, such as J.A. Smith. I think a man has to be a Pope and a necromancer before I really hate him; Schiller is not solemn enough for that. He fails to be a humbug because he hasn't the self-control to pretend not to be one. As you can see I am in a dry intellectual mood today-very useful for getting on with my work, if only it will last". (#356) John Alexander Smith (I863-I939), who was Professor of Moral and Metaphysical Philosophy in Oxford from I9IO to I935, was an idealist of the Italian sort. A few days later, on 3 March, he remarked: "Schiller's book on Formal Logic has come to me to review in the Nation. I wonder" (#373). In several later letters he notes that he has books to review, but there is no further mention of Schiller's book. Despite his state of fury, Russell's acknowledgement, in a letter of 23 February I9I2, of the gift of the book was cordial: Many thanks for your kindness is sending me your book on formal logic. I agree with you that traditional formal logic is rot, but beyond that I suppose I shan't agree with much. I am however very glad to have the book, and I shall be interested to see how pragmatism deals with the subject.

When he had read the book he wrote again, on II April I9I2: I have just finished your Formal Logic and am quite surprised to find how much of it I agree with. Everything that you say in favour of empiricism, doubt, adventure, has my entire sympathy. It is only the "humanism" that I don't go with. I can't bow down to Man and worship him. Of course except from the point of view of examinations you are flogging a dead horse; active students of logic have long ago abandoned all the views you criticize, or most of them. As to the doubtfulness of axioms, I commend to your notice p. 62 of Principia Mathematica Vol. I. You suggest as a reductio ad absurdum that if logic is not to be psychological it must not deal with inference. I have long held this view, and regard inference as outside the scope of logic, strictly understood. It is a monstrous scandal that the old formal logic is still taught, and I am grateful to any one who boldly calls it nonsense, as you do. The reference to Principia Mathematica (1925, 59-60) is to the section entitled "Reasons for Accepting the Axiom of Reducibility". Before arguing the specific case, Russell makes a general point about all axioms: That the axiom of reducibility is self-evident is a proposition which can hardly be maintained. But in fact self-evidence is never more than a part of the reason for accepting an axiom, and is never indispensable. The reason for accepting an axiom, as for accepting any other proposition, is always largely inductive, namely that many propositions which are nearly indubitable can be deduced from it, and that no equally plausible way is known by which these propositions could be true if the axiom were false, and nothing which is probably false can be deduced from it. If the axiom is apparently self-evident, that only means, practically, that it is nearly indubitable; for things have been thought to be self-evident and have yet turned out to be false. And if the axiom itself is nearly indubitable, that merely adds to the inductive evidence derived from the fact that its consequences are nearly indubitable: it does not provide new evidence of a radically different kind. Infallibility is never attainable, and therefore some element of doubt should always attach to every axiom and to all its consequences. In formal logic, the element of doubt is less than in most sciences, but it is not absent, as appears from the fact that the paradoxes followed from premisses which were not previously known to require limitations. He goes on to argue that the inductive evidence for the axiom of reducibility is very strong.

292

293

294

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

Schiller replied to this letter on 15 April 1912: I was somewhat prepared for your approval of my criticism of the old 'Logic' by your sympathetic handling of Mercier, but none the less glad to hear this from you directly. As a fact I consider that the recognition of the omnipresence of risk in every thought is far more fundamental and revolutionary, than anything else pragmatism has engendered up to date, except perhaps the discovery of Relevance. But I am sure you are wrong in thinking that the old stuff is dead; it is still dominant: e.g. all the essential errors of Formalism may be found in Joseph's Logic, (which you have not perhaps studied as carefully as I) and the ordinary Oxford Logic Lecture contains certain passages which have never been published, but been copied faithfully from generation to generation ever since, I should guess, the early Middle Ages. One of my objects in writing was to render it impossible that this should ever be published. Or in other words that Cook Wilson should ever write a Logic! I know all this by the only way in which it is possible to get at such 'secret doctrines' viz. by examining in Greats. I don't understand wherein you think me to have yielded to the weakness of worshipping Man (or anything!). I will of course look up the reference to Principia Mathematica as soon as I get back to my books, though I cannot promise to understand you, as you always seem to me to abstract wholly from the problems which interest me, application &c. If you admit (as I knew) that 'Logic' does not deal with Inference, do you agree also that it should not deal with Judgment? For the grounds for regarding both as 'merely human' seem to be the same. Anyhow I am not sanguine about superseding the traditional logic, however variously it is shown to be nonsense. Your 'symbolic logic' is still less capable of being made a 'pass' subject and an alternative to mathematics than my psychologic, and I sometimes think that most of what is taught everywhere is nonsense, and the more easily taught because it is. Cf. Classics and Theology. After the review appeared Schiller wrote to Russell again: this letter, which is printed in Appendix I, is a lengthy reply to Russell's review. Schiller was generally pleased with the review, but took exception to a number of Russell's judgments. Ferdinand Canning Scott Schiller (1864-1937), a Fellow of Corpus Christi College, Oxford, was the chief exponent of pragmatism in England. Beginning in 1926 he spent part of each academic year teaching in the University of Southern California. His extensive correspondence with Russell has been preserved. Despite all his reservations about Schiller as a person and about the quality of his philosophical work, Russell was prepared to support him for various professorial positions. On 23 January 19rn he gave this testimonial:

24 PRAGMATISM AND LOGIC

295

I learn that Dr. Schiller is a candidate for the Waynflete Professorship of Moral and Metaphysical Philosophy. He is, as is generally recognized, the leading representative in Europe of the philosophy known as Pragmatism or Humanism, and it would be difficult to apportion between him and Professor William James their respective shares in the creation of this philosophy. Although I do not personally agree with many of Dr. Schiller's opinions, I yet consider that he has shown a high degree of originality, and has produced a genuinely new system which is likely to have many adherents, and to remain one of the recognized types of philosophical theory. The cardinal point in Dr. Schiller's philosophy is his doctrine as to the nature of truth, and in this his views are, in my opinion, both new and important. Moreover he has applied these views to a large number of philosophical questions, and has thus produced a unified and internally coherent body of thought. His reputation stands very high both on the Continent and in America, and he is certainly among the most eminent of living British philosophers. Schiller was not elected to the position. Russell was to continue as his referee for another decade or more. The book remains in Russell's library. In addition to many vertical lines it contains a number of verbal comments, none of them complimentary. His draft of his remark in the review about new wine in old bottles is in the margin on page x: Schiller is new wine in old bottles. N. W. "Everything in Aristotle is rubbish." O.B. "All truth is in Aristotle." Hence pragmatist theory of truth. "Rot" was his judgment on "Of all the discoveries which man has made by dint of sheer reflection the Syllogism is assuredly the greatest" (187). Our final example concerns this passage: For example, it may be generally true that "all men love good stories," and undeniable that "Smith is a man"; yet the inference that "therefore Smith loves this good story," may be falsified in this particular case by the fact that the story is told about him, and that, therefore, he hates it. Now technically this result may be ascribed to an ambiguity of the middle term. Smith is in general a "man" and, therefore, loves good stories, but he is not a "man" for the purpose of this particular conclusion; and so the sense of "man" in the two premisses is not the same, and this vitiates the argument. Russell's comment is a devastating "Oh Oh". The copy-text is the printed version.

24 PRAGMATISM AND LOGIC

Formal Logic: A Scientific and Social Problem. By F. C. S. Schiller, London: Macmillan, I9I2. Pp. xviii, 423.

T

HIS WORK IS called by the author a "critical text-book"; it aims at ~iving as m~ch kno"'.led~e of traditional formal logic as is reqmred for passmg exammat10ns at Oxford together with as much pragmatism as may persuade the student that all formal logic is worthless. An exposition of a subject by a man who does not believe in it has always certain drawbacks; it is hardly to be expected that he will go out of his way to state a case better than it has been stated by its advocates, and IO therefore their more superficial errors tend to be treated as fundamental. Dr. Schiller, in common with all (or nearly all) teachers of formal logic at Oxford, and the vast majority elsewhere, is ignorant of the work that has been done on the subject during the last sixty-eight years, since Boole published his Laws of Thought. The subject he is criticizing is not the subject treated of by modern writers, but the subject which has filtered down from Aristotle to the schoolmen, and thence to the examinationroom. If mankind entertained any respect either for truth or for the young, this subject would long ago have been eliminated from every curriculum. Since it remains, attacks on it are to be welcomed, and, in spite 20 of being in the sheep's clothing of a text-book, Dr. Schiller's attack cannot fail to be useful. Formal logic, he justly observes, could never have escaped detection so long if it had not been ranked among the "literary" subjects. But his attack on it is itself a "literary" attack. There is a remarkable contrast between Dr. Schiller and William James, considering the close agreement in their opinions. William James's outlook was scientific, and much coloured by physiology and biology. Dr. Schiller's outlook is literary; his ancestors are the philosophical sceptics, Protagoras, Sextus Empiricus, and their successors. It is true that he speaks of "science" with respect, 30 but he uses it essentially as a stick to beat his literary colleagues. For this reason, his criticisms are apt to be not very fruitful, even when they are just. His philosophy really results from putting new wine into old bottles. "Everything in Aristotle is rubbish"; this is the new wine. "All truth is in Aristotle"; this is the old bottles. By putting these two premisses together, we obtain a syllogism whose conclusion is the pragmatist theory of truth. "Of all the discoveries", says Dr. Schiller, "which man has made by dint of sheer reflection, the syllogism is assuredly the greatest" (p. I87). This is an illustration of the old bottles; the following pages, which argue 40 that the syllogism is utterly worthless, are an illustration of the new wine. The present work aims, of course, at proving that the fundamental principles of logic are neither a priori intuitions nor generalizations from experience, but postulates-that is to say, working assumptions-which we make because they have been found to work, and retain so long as they

296

297

continue to work. This view, though it undoubtedly has a certain truth, raises new problems which Dr. Schiller seems dimly to perceive, though he does not discuss them. If a postulate works, even partially, this fact shows us something as to the nature of the real world; the fact that it works is no longer a mere postulate. "Postulates," he says (p. 301), have, of course, to make good their claims to be applicable to the actual course of nature. But this is not a question that can be settled by making postulates, however strenuously; it depends on experience, and, as we urged against the rationalist theory, must always continue to do so. For even if they hold good up to date, it is always conceivable, though never presumable, that nature may become more recalcitrant to our postulations. In this passage, the via media between empiricism and rationalism appears to be abandoned, and pure empiricism is recommended on the question as to how far we are to believe our postulates. This raises the problem-unanswerable on a purely empiricist basis-as to why there should be even the slightest presumption that the future will in any way resemble the past. A change, we are told, "is always conceivable, though not presumable". But why it is not presumable we are not told. There are two elements in Dr. Schiller's philosophy, both present in this book, but important to disentangle. There is, on the one hand, the insistence on risk and adventure, on the uncertainty of all our beliefs, and the impossibility of removing all reason for doubt. This element is very valuable, and calculated to keep thought fresh and living. In regard to formal logic, where the medieval tradition still fetters the minds of many teachers, and where, to a superficial reflection, there may seem to be very little room for doubt, it is particularly desirable to insist upon the chance of error. In so far as Dr. Schiller's book does this, it will be judged useful by all open-minded people. The other element in his philosophy is subjectivism, or what he calls humanism, the doctrine that man is the measure of all things. The value of this element is much more open to question. It leads him to be content with postulation as the source of all our fundamental principles, with the tradition that logic is concerned with thought, and with a theory of truth which does not demand that a true belief shall in any way correspond with fact. Much of the attractiveness of pragmatism, to some scientific minds, is undoubtedly due to the former element; they accept the subjectivism, because they believe it inseparably bound up with the other. If, however, this belief is erroneous, it becomes highly important to separate the two elements; and, whatever may be thought of the second, the value of the first is such that Dr. Schiller's attack on formal logic is likely to be useful to many who would not otherwise have escaped from the traditional nonsense.

IO

20

30

40

25 REVIEW OF WILLIAM JAMES

25

Review of James's Essays in Radical Empiricism [1912]

299

means that, as the actual tissue of experience is constituted, the idea is capable of leading into a chain of other experiences on my part that go from next to next and terminate at last in vivid sense-perceptions of a jumping, barking, hairy body. These are the real dog, the dog's full presence, for my common sense. (198) "Then a bone you hold in your hand is knowledge of the dog" is Russell's wry marginal comment. The copy-text is the printed version.

THIS REVIEW WAS published in Mind, n.s. 21 (Oct. 1912): 571-5. On 28 July l9I2 Russell wrote to Lady Ottoline Morrell: "Yesterday I did a review of W. James for Mind (a more serious matter than reviewing for the Nation)" (#517). Although one would never guess it from this remark, the philosophical doctrines which James advanced in this book were destined to become central to Russell's own thinking. In attempting to refute James's neutral monism, Russell gradually persuaded himself that it was true, or nearly true. See the Introduction to Collected Papers 8 (xx-xxiii). Russell added the book to his library. There are the usual vertical lines in it, as well as several verbal comments. On page 25 James writes: Consciousness connotes a kind of external relation, and does not denote a special stuff or way of being. The peculiarity of our experiences, that they not only are, but are known, which their 'conscious' quality is invoked to explain, is better explained by their relations-these relations themselves being experiences-to one another. Russell underlines the last "experiences" and comments: "I wish James would explain this word 'experiences'." But he found nothing to criticize in this passage: Let the case be what it may in others, I am as confident as I am of anything that, in myself, the stream of thinking (which I recognize emphatically as a phenomenon) is only a careless name for what, when scrutinized, reveals itself to consist chiefly of the stream of my breathing. The 'I think' which Kant said must be able to accompany all my objects, is the 'I breathe' which actually does accompany them. (36-7) It is marked by four vertical lines and the word "Excellent!" He disagrees with James on this point: "No one denies that they (thought and thing) have some categories in common. Their relations in time are identical" (29). "Not when 'things' are universals", Russell comments. Finally, Russell offers a gloss on this passage:

To call my present idea of my dog, for example, cognitive of the real dog 298

25 REVIEW OF WILLIAM JAMES

Essays in Radical Empiricism. By William James. London, New York and Bombay: Longmans, Green, I9I2. Pp. xvi, 283. HIS BOOK CONSISTS ofa collection of reprints, with a preface by the editor, Prof. Perry. Most of the essays in the volume were collected by William James himself in an envelope with the title which Prof. Perry has given to the book. Some of them were subsequently reprinted, in whole or in part, in The Meaning of Truth and in A Pluralistic Universe, but they are rightly included here in order to give as complete a view as possible of the doctrine called "Radical Empiricism". IO "Radical Empiricism" is not the same thing as pragmatism or humanism: in William James's mind the two were connected, but he admitted that one might hold the one without the other. Prof. Perry, in his Preface, quotes the summary from the Preface to The Meaning of Truth. According to this summary, Radical Empiricism consists of (1) a postulate, (2) a statement of fact, and (3) a generalized conclusion. The postulate is that only things definable in terms drawn from experience shall be debatable among philosophers. The statement of fact is that relations between things are just as much matters of direct particular experience as the things themselves. The generalized conclusion is that the parts of experience hold 20 together from next to next by relations that are themselves parts of experience. The postulate is not intended to rule out the possibility that there may be "transempirical" objects, but only methodologically to exclude the consideration of them from philosophy (p. 241). The statement of fact and the generalized conclusion distinguish James's empiricism from the more atomic traditional empiricism of Hume and his followers; they may, I should suppose, be now accepted as indubitable. With regard to the postulate, it is evident that its truth or falsehood must turn on the meaning given to that very slippery word, "experience". On this subject one could wish that James had been more explicit. He seems to regard 30 "experience" as something simple, with which we are all acquainted. I cannot help thinking that, in the first two essays (which are the most important in the book), the failure to analyze "experience" has concealed important difficulties in the views advocated. The first essay, called "Does 'Consciousness' Exist?" contends that the dualism of thought and things is an error: that the very self-same entity is at once a table and the perception of a table. The distinction between the mental and the physical, James says, is a distinction of context and function. When I perceive a table, the table is physical when regarded as part of the chain of physical causation, but mental when regarded as phys40 ically inefficacious and merely one of my events. There is no stuff out of which thoughts, as opposed to matter, are made; pure experience is the only stuff of the world; what distinguishes consciousness is a certainfunc-

T

300

301

tion, namely the function of knowing, which is a relation between different parts of pure experience. Locke's use of the word "idea", James says, began the tendency to obliterate the traditional dualism, and Berkeley continued it. It would be a mistake to name the doctrine materialism, or to name it idealism; both these names operate within the distinction of mental and physical, whereas with James there is an absolute identification of the mental and the physical, giving to each the characteristics of the other, making alternately the impression of materialism and of idealism, according to the context. "What possible meaning", James asks, "has it to say that, when we IO think of a foot-rule or a square yard, extension is not attributable to our thought?" (p. 30). Again: "Mental fire is what won't burn real sticks; mental water is what won't necessarily (though of course it may) put out even a mental fire" (p. 33). When it comes to accounting for our apparent immediate awareness of thinking, James is ready with an answer. "The stream of thinking (which I recognize emphatically as a phenomenon) is only a careless name for what, when scrutinized, reveals itself to consist chiefly of the steam of my breathing .... Breath, which was ever the original of 'spirit,' ... is, I am persuaded, the essence out of which philosophers have constructed the entity known to them as consciousness" (p. 20 37). The second essay, called "A World of Pure Experience", further describes the identification of the mental and the physical, and explains what, on this theory, is meant by "knowing". There is nothing that should be called "knowing", we are told elsewhere (p. 204), in "absolutely terminal experiences", if such experiences could ever be reached. Knowing always involves two experiences (which may, however, be the same experience in different contexts), and definitely felt transitions from the one which is said to know to the one which is said to be known. The knower and the known, we are told (p. 53), (1) may be the same piece of experi- 30 ence in different contexts, (2) may be two pieces of actual experience of the same subject, or (3) may be respectively an actual experience and a merely possible experience to which conjunctive transitions would lead if prolonged. This theory, with its rejection of the traditional dualism, is certainly profoundly original and profoundly interesting. Its truth or falsehood, so far as I can see, does not stand or fall with that of pragmatism, though its truth would make pragmatism much more plausible. A really new theory in philosophy comes always with a shock, and I must confess that, after many efforts, I cannot conceive the facts as William James does. His 40 hypothesis, in any case, deserves to be elaborated; and until it has grown more familiar, it would be rash to feel too sure that it is false. But, since the view has many advocates among those who owe their philosophic edu-

302

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

cation to Harvard, it rnay be worth while to point out sorne objections which ought to be definitely answered, though they appear to have been entirely absent frorn Jarnes's thoughts. Let us consider, for the sake of argument, the occurrence of a perception-say of a table. It is possible to take the view that the fact is constituted by three elements, act, content and object; or we rnay eliminate the content, and have only two elements, act and object; or we rnay also eliminate the act, and have only one element, the object. The first of these three views is advocated by Meinong, the second is the one I should adro vocate, the third is J arnes's view. (When I speak of the "object", I do not rnean the "physical" object, if any, which science infers from what is given; I rnean the coloured surface which is seen.) Jarnes's arguments for his position, if I have not misunderstood thern, seern chiefly effective in eliminating the "content". He seems concerned to show that the object is given without any "mental" intermediary. But if this be granted-as I think it should be-it still does not follow that there is nothing distinctively mental. To be "given", to be "experienced", is not the sarne thing as to "be". To be "given" or "experienced" seems to imply a subject, to be in fact constituted by a relation to a cognizing act. Tables and chairs 20 and other bits of matter might, it would seern, exist and have relations without there being any "experience" or "knowledge". James will not regard irnrnediate perception as knowledge, though he does apparently regard it as "experience"; his reason being, so far as can be discovered, that he thinks knowledge rnust be round-about. Thus he does not regard "experience" as in its nature a relation, of which one terrn rnust be what we call "mental". The processes of leading, which he regards as constituting knowledge, seern to include cases which cannot be called knowledge, and to exclude cases which rnust be so called. He says:

30

To call rny present idea of rny dog, for example, cognitive of the real dog means that, as the actual tissue of experience is constituted, the idea is capable of leading into a chain of other experiences on rny part that go frorn next to next and terminate at last in vivid sense-perceptions of a jumping, barking, hairy body. (p. 198)

But merely holding a bone in rny hand, without thinking of the dog, rnay have the sarne effect; and it seems paradoxical to maintain that I do not know the dog when it is actually present, which I gather James would have to maintain. There are, of course, various empirical reasons of detail for supposing 40 that what is irnrnediately presented in sense is not the "physical" object which physics deals with. But these reasons, which rnay or rnay not be

25 REVIEW OF WILLIAM JAMES

303

valid, are concerned with a far less fundamental problem than that which J arnes is discussing. In considering his problem, it is simpler to concede the view of naive realism, that the actual object dealt with by physics is irnrnediately given. This, so far as I can see, is all that James attempts to prove. But this by no means proves that "experience" is not a relation, involving something over and above the object, something to which the object is given, by which it is experienced. It is this something, if anything, that is mental; and it is this something which James ought to have disproved. On grounds of the purest ernpiricisrn, frorn rnere inspection of experience, I for rny part should hold it obvious that perception is in its ro intrinsic nature a fact of relation, involving an act as well as an object. For this reason, I cannot accept J arnes' s view, in spite of its very attractive simplification of the world. The third essay, on "The Thing and Its Relations", argues against the current idealist view that no one thing can stand in two relations. "It really seems 'weird"' (he says) "to have to argue ... for the notion that it is one sheet of paper ... which is both under rny pen and on the table while I write .... Yet I sornetirnes suspect the absolutists of sincerity" (p. rn5n.). The fourth essay applies the result of the third to show that it is logically possible for two minds to know the sarne thing-a question also briefly 20 discussed in the second essay, where a cursory consideration of the empirical difficulties leads to the conclusion (on grounds which seern not very decisive) that two minds can perceive at least the sarne space (p. 84), though probably not the sarne objects in space. The fifth essay is on "The Place of Affectional Facts in a World of Pure Experience"; in this essay the James-Lange theory of the emotions is used to prove that they need not be regarded as purely mental. The sixth essay on "The Experience of Activity", which has already appeared in A Pluralistic Universe, is very interesting; there is a beautifully clear account of the psychological experience of activity, with a rejection of any supposed 30 metaphysical principle of activity. The seventh essay, on "The Essence of Hurnanisrn" and the eleventh, on "Hurnanisrn and Truth Once More", belong rnore to the familiar field of the pragmatist controversy. The eighth, in French, on "La Notion de conscience", adds little to the first two. The ninth is called "Is Radical Ernpiricisrn Solipsistic?" (the answer being in the negative); the tenth refutes "Mr. Pitkin's Refutation of 'Radical Ernpiricisrn' ". The last essay in the book, called "Absolutism and Ernpiricisrn", is rnuch earlier in date than the others, having appeared in Mind in 1884. It explains Jarnes's emotional reasons for disliking Absolutism, and is ex- 40 quisitely witty. I will not spoil the sale of the book by quoting the story of the two curates at a funeral, or by' explaining why Hegelianisrn is like a sea-side boarding-house.

304

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

26

The book, like everything by its author, is interesting and delightful to read, full of brilliant flashes, humour, and quick insight, but impatient of plodding careful argument or analysis. I cannot believe that empiricism, however radical, requires that we should deny the difference between mind and matter; but otherwise most of the book is such as any modern empiricist must agree with.

Review of Boutroux [1912]

REVIEW w AS published, unsigned, in The Cambridge Review, 34 (5 Dec. 1912): 176. It is known to be by Russell because, on 28 October 1912, he wrote

Tms

to Lady Ottoline Morrell: "This morning I wrote a review of a worthless little book by Emile Boutroux on W. James" (#614). His judgment of the book's worth is repeated in the review. Emile Boutroux (1845-1921) was a French philosopher who "belonged in the tradition of French spiritualism with its roots in Schelling" (Perry 1935, 2: 561). In 1908 he met William James for the first time. Despite basic philosophical differences, "their friendship ripened quickly into love, and formed one of the most significant episodes of the last two years of James's life" (ibid.). The copy-text is the printed version.

305

William James. By Emile Boutroux. Translated by Archibald and Barbara Henderson. London: Longmans, Green, 1912. Pp. vii, 126. T IS JUST possible that this book was worth writing in good French; it is quite certain that it was not worth translating into indifferent American. M. Boutroux seems to have thought that the French nation could not be expected to appreciate William James in his native uncouthness; in place of the transatlantic wildness, the roughness, the homely humour, the simple democratic friendliness, which make his writings delightful, we find here a smooth surface, a polished blandness, a style in which tradition and literary convention make all real thought impossible. The result is a portrait bearing about as little resemblance to the original as Voltaire's intelligent Huron bore to the fierce savage of the backwoods. After a brief biographical notice, the rest of the book is concerned with William James's contributions to psychology and philosophy, summed up in a set of neat dogmas, and finally reduced to a form in which they can be placed in the academic tradition without the risk of altering any man's lectures. There are, we are told, two common methods of introspection, of which the first perceives "the multiple without unity", while the second perceives "the one without the multiple". But "the true introspection is the living synthesis, the intimate fusion, the concrete unity of these two methods. It has for its object the actual, the immediate datum of consciousness." The book abounds in statements which may be true or may be false, but are merely recommended by the assumption of omniscience which is implicit in the style. For example: "It requires courage to say it: the Galileo or the Lavoisier of psychology, the man who shall unveil the truly fundamental principle, if he is ever to appear, will be a metaphysician." Or again: "The more we force ourselves to see things in a natural way, and not to use our eyes like a rude microscope or telescope, the more we see that beings are one with their relations." This dogmatic manner is hardly avoidable, perhaps, in a short account of another man's philosophy; but if so, it may be doubted whether such short accounts can serve a useful purpose. The translators have not always realized that error cannot be avoided by looking up each successive word in the dictionary; for example, they often translate "l'etre", by "the being" when they should translate it by "being".

I

ro

20

30

306

Part

VI

Critique of the Philosophy of Bergson

27

The Professor's Guide to Laughter [1912]

THE REVIEW w AS published first in The Cambridge Review, 33 (18 Jan. 1912): 193-4. It was reprinted in Volume 12 of The Collected Papers of Bertrand Russell (1985). The reason for including it in this volume too is the reader's convenience: all of Russell's writings on Bergson can now be found in one place. Russell wrote his review on IO January 1912. That evening he wrote to Lady Ottoline Morrell: I spent the afternoon writing a review of Bergson's Laughter (enclosed). I showed it to Lucy Silcox (who came to tea)-she thought it unkind; certainly love did not inspire it. I think the book nearly worthless, and as he is such a distinguished man it seems hardly necessary to be kind. But I should like to know what you think. (#312) He does not mention the review again in his correspondence with her. Russell thought the thesis of this book contradicted Bergson's philosophical position. In a 1926 paper, "Behaviourism", he made the point humourously: Bergson, writing for a French public, holds a threat over those whose acts refute him, which is even more terrible than moral condemnation; I mean, the threat of ridicule. He shows that human beings never behave mechanically, and then, in his book on laughter, he argues that what makes us laugh is to see a person behaving mechanically-that is, you are ridiculous when you do something that shows Bergson's philosophy to be false, and only then. (1988, 70-1) This was one of three examples which he adduced in support of a general law: It may be laid down that every ethical system is based upon a certain non sequitur. The philosopher first invents a false theory as to the nature of things, and then deduces that wicked actions are those which show the theory is false. (Ibid, 70) Russell kept the book for his library. In addition to his usual vertical lines and a list of page numbers at the rear, he made six verbal comments in the margins. 309

310

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

"Quite untrue" is his judgment on Bergson's "we begin, then, to become imitable only when we cease to be ourselves" (33). "As we hinted at the outset of this study, it would be idle to attempt to derive every comic effect from one simple formula" (36) draws a "hear hear" from Russell. He disagrees strongly with this passage: Comedy depicts characters we have already come across and shall meet with again. It takes note of similarities. It aims at placing types before our eyes. It even creates new types, if necessary. In this respect it forms a contrast to all the other arts. "All rot'', he writes, "Balzac e.g. gets tragedy from types". A final example is found beside this remark: But in Don Quixote, on the contrary, there is one group of memories in command of all the rest and dominating the character itself: thus it is reality that now has to bow to imagination, its only function being to supply fancy with a body. Once the illusion has been created, Don Quixote develops it logically enough in all its consequences; he proceeds with the certainty and precision of a somnambulist who is acting his dream. (184) "Why is Macbeth's dagger not comic?" is Russell's question. Russell's library also contains a copy of the first French edition of the book, but it is not marked in any way. The copy-text is the printed version.

Laughter, an Essay on the Meaning of the Comic. By Henri Bergson. Authorized translation by Cloudesley Brereton and Fred Rothwell. London: Macmillan, I9I I. Pp. vii, 200. T HAS LONG been recognized by publishers that everybody desires to be a perfect lady or gentleman (as the case may be); to this fact we owe the constant stream of etiquette-books. But if there is one thing which people desire even more, it is to have a faultless sense of humour. Yet, so far as I know, there is no book called "Jokes without Tears, by Mr. McQuedy". This extraordinary lacuna has now been filled. Those to whom laughter has hitherto been an unintelligible vagary, in which one must join, though one could never tell when it would break out, need only study M. Bergson's book to acquire the finest flower of Parisian wit. By observing a very simple formula, they will know infallibly what is funny and what is not; if they sometimes surprise their unlearned friends, they have only to mention their authority in order to silence every doubt. "The attitudes, gestures and movements of the human body", says M. Bergson, "are laughable in exact proportion as that body reminds us of a mere machine." When an elderly gentleman slips on a piece of orange peel and falls, we laugh, because his body follows the laws of dynamics instead of a human purpose. When a man falls from a saffolding and breaks his neck on the pavement, we presumably laugh even more, since the movement is even more completely mechanical. When the clown makes a bad joke for the first time, we keep our countenance, but at the fifth repetition we smile, and at the tenth we roar with laughter, because we begin to feel him a mere automaton. We laugh at Moliere's misers, misanthropists and hypocrites, because they are mere types mechanically dominated by a master impulse. Presumably we laugh at Balzac's characters for the same reason; and presumably we never smile at Falstaff, because he is individual throughout. "Was it for me to kill the heir-apparent? Should I turn upon the true prince?" is not the remark of an automaton; therefore it is tragedy, not comedy as was hitherto supposed. M. Bergson's general philosophy is largely a protest against the attempt to bind down living things by the fetters of cast-iron formulas. Yet his Latin instinct for order has proved too strong for him, and he has attempted to imprison laughter, the most living of all living things, within a formula which, in spite of his amazing ingenuity, cannot be made to apply to more than a tiny fraction of the jokes that will occur to any reader. Lamb's answer to the man who asked about the prospects of the turnip crop, "I believe it depends upon the crop of legs of mutton", comes under a formula exactly the opposite of M. Bergson's: it depends upon assuming something like human purpose in what is really mechanical. No doubt

I

311

ro

20

30

40

312

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

surprise is an element in laughter; and it is chiefly this element, I think, which makes M. Bergson's instances amusing. Charles Lamb's humour suggests another criticism. M. Bergson considers that absence of feeling is characteristic of laughter; sympathy, he says, makes the whole world gloomy, whereas the comic makes its appeal to intelligence, pure and simple. This is certainly not true of Lamb. In quite a different way, Hamlet's wit is full of passion; in quiet times, he would have made much fewer jokes. The truth seems to be that the comic differs with the individual, the ro country, and the age. Latin wit is different from Teutonic humour; the laughter of the Parisian is different from the laughter of the Londoner. For this reason, it would seem to be impossible to find any such formula as M. Bergson seeks. Every formula treats what is living as if it were mechanical, and is therefore by his own rule itself a fitting object of laughter.

28

The Philosophy of Bergson [1912]

THIS PAPER WAS published in The Monist, 22 (July 1912): 321-47. Open Court, the publisher of the journal, also published the paper as a (repaginated) pamphlet at the same time. It was reprinted, with a reply by H. Wildon Carr (Appendix IX) and Russell's rejoinder to Carr (Paper 30), as a pamphlet by The Heretics in Cambridge in 1914. Russell included a revised version of it in A History of Western Philosophy (1945); it was further shortened in the 1961 edition of this book. In addition to the printed versions, a manuscript of Part I exists. The paper was originally written to fulfil a speaking obligation, set for 11 March 1912, to The Heretics, a group of young people at Cambridge whose leading light was C. K. Ogden. On 12 October 1911 Russell mentioned to Lady Ottoline Morrell that he had met Ogden, the secretary of the club: The Heretics are a society who discuss more or less religious questions, and Ogden is a very energetic undergraduate, who gets hold of Shaw and Chesterton and even my brother to come and speak. He wanted a paper out of me, and I more or less undertook to write about Bergson next term. (#216) Russell was already reading Bergson when this interview took place, because he went on to state: "In the intervals of my morning I read Bergson on Nothing, which he thinks is nothing. This led me into a number of reflections on negationa very puzzling topic" (#216). Ten days earlier he had reported that "I am started on Bergson but I shan't get much time for him" (#205, 2 Oct. 1911). The next day he offered her some advice on what she could do about his vices: I think the first vice you should take in hand is the vice of writing to you instead of reading Bergson, whom I have brought in my pocket. I have read a good deal of him and I can see why people like him, also, I think, how he comes to believe such stuff. But it is stuff. His God is deduced from the eyes of shell-fish. He is the antithesis to me; he universalizes the particular soul under the name of elan vital and loves instinct. Ughl (#206) By 9 October he had read more: 313

314

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

28 THE PHILOSOPHY OF BERGSON

I have had a quiet morning with Bergson-he begins to interest me more, in something the same way as heraldry and theology do, from the queerness and fantasticality of the stuff. He thinks the intellect a wicked imp whose practice is to show us everything as space and matter; what reveals the real truth is activity, especially artistic creation. But it won't do to stop and think, or else his philosophy will no longer seem true. This is the only remark I agree with so far. (#213)

away from it as yet" (#356). He spent the morning of the 25th reading Bergson: "His temperament repels me, but I am beginning to have some imaginative realization of him-it takes me ages to overcome the instinctive hostility" (#358). On the 27th he explained further:

Theology and heraldry remained fascinating to Russell. In 1929 when asked what he liked, he replied: "I like the sea, logic, theology, heraldry. The first two because they are inhuman, the last two because they are ridiculous" (Russell 1929, 73). On 12 October, the day he mentioned the paper for The Heretics for the first time, he wrote to Lady Ottoline again, after his visitors had left. Ogden and another student were among those who had attended his "evening": the two of them obviously had a great admiration for Bergson, which I did my best to dispel. One of his chapters ends with a wonderful peroration in which he compares human life to a cavalry charge-he describes the whole human race careering so madly that they pass all obstacles, "perhaps even Death". I think after a cavalry charge lasting seventy years one would be sorry to pass that obstacle successfully. It is characteristic of him to choose for his ideal a state of things in which reason is entirely dormant. I begin, however, to feel why people like him. He has a very strong imagination, and has conceived the world through and through according to his scheme. For those who like Life and Action and Movement, he is admirable. He hates Plato, because the Ideas are static, and he wants everything to be dynamic, like the cavalry charge. When he is better, he conceives life to be essentially like artistic creation, in which a more or less blind impulse urges one towards something, without one's knowing what beforehand; then, when it is created, it is seen to be what was wanted. (#216a) A week later, on 19 October, he made a further report: "I am still reading Bergson-on Time and Free Will this time. It is better than Creative Evolution but not very good" (#227). Bergson falls out of the correspondence until 5 February 1912 when he wrote: "I have been doing my duty reading Bergson-he does bore me so. But I have to write a paper on him, worse luck" (#333). Eight days later Bergson was on his mind again: "I shall be busier the last half of term than I have been the first half because of the big meeting on March l and work connected with it, and my pape; on Bergson on March rr" (#340). On the 18th he reported that he was "reading Bergson" (#347). By the 23rd he was feeling "venomous" about Bergson: "If I could love Bergson I should have achieved a really Christian spirit-but I am miles

315

No, speaking seriously I don't hate Bergson. If I didn't have to read him I shouldn't be tempted to; when I met him I liked him. But I think he is a man of low ideals, action without contemplation. And I hate his dogmatic, pontifical style. But I quite admit I ought to feel more charitably towards him. (#360) As late as 3 March he had not begun writing: I shan't probably work any more at Bergson after March l l, when I read my paper. If he were sympathetic to me, the paper would be less trouble, but also less use to me. As it is, I have finished the dull part of the job, which is reading him, and have come to the pleasant part. I have read most of him twice, and I am now making a careful abstract-when that is done, I shall probably have my paper in my head and only have to write it out. I mean to begin with a really sympathetic account, before criticizing him; but all the phrases that occur to me are turning him into ridicule. However, I do now really feel his philosophy. There are philosophies of action, feeling and knowledge. Those of feeling are optimisms and pessimisms. Those of knowledge are contemplatives, like Plato and most of the great philosophers. Bergson and the pragmatists are philosophers of action. "Act, act in the living present, let the dead past bury its dead". Ugh! (#365) The next day the abstract had been made: My abstract of Bergson is made, and now I must begin the actual writing. I think his philosophy is the outcome of some mystic illumination which I don't quite understand-I haven't got inside his temperament. I think many years ago he must have fallen in love and come to the conclusion the intellect was a poor thing: probably nothing has happened to him since, but he still looks back to his one moment of life, and his thought is still in the grip of instinct. His philosophy is a glorification of instinct as against reason. (attached to #372) On 5 March he remarked: "Now darling I must return to Bergson. I take him as moral discipline-a lesson in sympathy-as hard to me as your moneygrubbers. He describes certain wasps who lay their eggs in other insects, whom they sting very scientifically so as to paralyse them; he says they do it by 'sympathy'; that is

316

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

28 THE PHILOSOPHY OF BERGSON

the sort I have!" (#366). In another letter written later the same day he said: "I have made a start writing my Bergson paper-it seems as if it would go all right" (#368).

needs something solid at the back of it" (#387). On the 23rd of March he responded to Lady Ottoline's reaction to the paper:

Before going to bed on 6 March he noted: "I wrote 20 pages of my Bergson paper today" (#359). In a letter which was begun very late in the evening of 7 March and finished only at l: 30 the next morning he felt pressed:

I should love to explain Part n of my Bergson paper to you if you would really like it-there are few things I should like better, I am glad you liked the writing. Glad you spotted the joke. Darling I am so glad you like my letters. Yes, I have been very full of new life-it all comes out of you and love-love gives life, and with an effort one can turn the life to other things. I have been much more full of new life than before. I have had less conflict of intellect and feeling than usual.-In philosophical writing, one aims at a beauty of style which is not that of ordinary writing-clearness and force are the chief things-it wants to be very chiselled so as to have exactness. I thought my Bergson paper had succeeded in this, and I am glad you think so. (#399)

Tomorrow morning I must finish my paper. I have rather enjoyed writing it; I think it begins well but tails off. Till it is done, it fills my mind very much-it has required a great deal of thought. I fought Bergson in Paris at Easter-I remember sitting in my hotel bedroom struggling to concentrate my mind on him, but it would wander off to a wholly irrelevant topic. The elan vital had hold of me too much for me to wish to read about it. (#378) By the morning of 8 March he could report that "I have just finished Bergson and have at last a free mind" (#372). That afternoon he wrote to her again: When my Bergson paper comes back from being typed, I will send you the MS. If your head is up to it I should very much like you to read it. I am not sure you will like it, but it is almost all very easy, so it would be no great labour to you-and I do very much want to share things of that sort with you as much as I possibly can. There is one joke in it that delights me, but I expect it would be less amusing to people less accustomed to the things it deals with. I have lived in the Bergson world so much these days that the Coal Strike and all the rest of the every-day things have hardly touched me. How easy critical work is-one only has to go through a mechanical process and the result emerges. It is really vastly inferior to original work. The difficult thing is imposing form on chaos. In criticism, form is given-it is mind on mind, not mind on outer night. And in criticism one doesn't have to settle what to talk about. (#373) Just before he left to deliver his lecture he dropped her a note: "The whole world seems to be coming tonight-Cornford will be in the chair-Browne has been telephoning to know where it is. I shall be glad when it is over" (#379). The next morning he offered his opinion of its worth: "I think the Bergson paper is very good indeed"; and he went on to describe its reception: "It had a great success last night-the place was packed and they seemed to enjoy it .... Not a soul rose to defend Bergson at the end, so there was no discussion. McTaggart spoke a few graceful words and we all went away" (#379). On 15 March he told her that Wittgenstein "liked the peroration of my Bergson article, but doesn't like that sort of thing unadulterated as in the free man's worship. He says in his judgment it

317

In an earlier letter he had let her in on the secret of the joke: "The joke I liked was near the end, about St. Sebastian. You would understand it, but I don't think it would be as ridiculous to you as to a person long accustomed to Zeno" (#391, 18 March). Alan Wood, Russell's first biographer, had the advantage of frequent access to his subject. We learn from him that Russell regarded this paper as an important milestone in his development: One event worth mentioning here is Russell's celebrated lecture on Bergson, to the Cambridge society called "The Heretics". Bergson's mystical philosophy of evolution was then enjoying a tremendous vogue, which Russell set out to demolish; there was an eager audience to hear him, and everyone had a sense of a great occasion. The lecture can be found reprinted in Russell's History of western Philosophy; to enjoy its savour, the reader must imagine it delivered in Russell's dry, precise and ironic voice, and punctuated by the laughter and applause which greeted his sallies. It was an event of some importance in Russell's life, helping to re-establish him as one of the leading figures in Cambridge; and especially because it was his first big success as a public speaker. (Wood 1957, 89) Later in his book Wood notes that the paper "was divided into two parts simply because, when addressing the Heretics in l9II, there had been a break half way through so that the lecturer could recover his breath, and the audience could pause and think" (ibid., 197). Wood's account of the way in which Russell regarded this paper goes a long way toward explaining his great expenditure of effort on it.. Russell was sufficiently interested in Bergson's cinematographic representat10n of the world to go to the cinema for the first time in his life to determine what

318

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

28 THE PHILOSOPHY OF BERGSON

truth there was to it. On 31 July 1912 he told Lady Ottoline that he had spent the earlier part of the evening discussing metaphysics with a friend: "At last we went out to a cinematograph to see if it bore out Bergson's philosophy, which it did" (#520).

Bertrand Russell, Bergson told me, has never forgiven me for the refutation that I made one day, orally, of his completely materialistic presentation of Platonic forms. He avenged himself by saying: For Bergson, evolution has culminated, on the one hand in intelligence, which reaches its most complete development in the mathematician and, on the other hand, in instinct, which reaches its peak in bees, ants, and Bergson. (197)

Russell's letters to Lucy Donnelly during this period nicely complement those to Lady Ottoline. On 28 October 191 l he wrote of his first meeting with Bergson: Tonight I am in London, having come up to meet Bergson at dinner. He is giving lectures in London which are reported in the daily papers-all England has gone mad about him for some reason. It was an amusing dinner. Our host was Wildon Carr, a humble stockbroker who happens to be Secretary of the Aristotelian Society-a man rather like the host in one of Peacock's novels, but milder. He had Bergson on his right and Shaw on his left. I sat between Bergson and Younghusband (the Thibet man), who cares much more about philosophy than about soldiering. I had heard of him from McTaggart, but had never met him before-I liked him very much indeed-simple, sincere, and massive. The only other guests you would know about were Zangwill and Wallas. Bergson's philosophy, though it shows constructive imagination, seems to me wholly devoid of argument and quite gratuitous; he never thinks about fundamentals, but just invents pretty fairy-tales. Personally, he is urbane, gentle, rather feeble physically, with an extraordinarily clever mouth, suggesting the adjective "fin" (I don't know any English equivalent). He is too set to be able to understand or answer objections to his views. Shaw made an amusing speech explaining how glad he was that Bergson had adopted his (Shaw's) views, and expounding how Bergson thought we came to have eyes. Bergson said it wasn't quite that way, but Shaw set him right, and said Bergson evidently didn't understand his own philosophy. Everybody congratulated themselves and each other on their possession of freedom and on their escape from the barren scientific dogmas of the sixties. I still believe in these dogmas, so I felt out of it. When people laughed during Shaw's speech he said "I don't mean to make a comic speech, and I don't know why you laugh, unless because religion is such an essentially laughable subject". They seemed to me like naughty children when they think (mistakenly) that the governess is awayboasting of their power over matter, when matter might kill them at any moment. A year after the paper had been published, on 3 July 1913, he confessed, again to Lucy Donnelly, to second thoughts about it: "My paper on Bergson was unduly flippant, and I am now mildly ashamed of some of the jokes." From Entretiens avec Bergson (1959) by Jacques Chevalier we learn that Bergson, on 30 May 1933, attributed Russell's attack upon him to an earlier incident:

319

Presumably this "refutation" occurred during the discussion period following Russell's delivery of "On the Relations of Universals and Particulars" (16) on 30 October l9II to the Aristotelian Society. The minutes record that Bergson did join the discussion, but there is no record of the points he made. We do know, however, that Bergson was never present at any other time when Russell read a paper. Henri Bergson (1859-1941) was Professor of Philosophy in the College de France from 1900 until 1921, when he was forced to retire for health reasons. In later years he devoted most of his time and efforts to promoting world peace and cooperation among nations. In addition to Laughter and Introduction to Metaphysics, which are described in the Headnotes to Papers 27 and 29 respectively, Russell's library contains the following books by Bergson: Creative Evolution, The Creative Mind, Essai sur les donniies immediates de la conscience, !:Evolution creatrice, Matiere et memoire, Matter and Memory, La Perception du changement, and Time and Free Will. All of them, except The Creative Mind, which was not published until 1946, show evidence that Russell studied them carefully. In addition to the usual vertical lines, all of the books he read contain verbal comments, nearly all of them very negative. Three examples from Creative Evolution will give their flavour. Against this remark in the French original, "instinct perfected is a faculty of using and even of constructing organized instruments; intelligence perfected is the f acuity of making and using unorganized instruments" (152; Eng. trans., 147), Russell writes "use of a quill is instinct; of a steel pen, reason". In the English translation, Russell makes an addition to this sentence, "It is as if a vague and formless being, whom we may call, as we will, man or superman, had sought to realize himself, and had succeeded only by abandoning a part of himself on the way" (281). He inserts a caret after "superman" and adds "or Bergson" in the margin. In the same book, Russell greets "life has a tendency to accumulate in a reservoir, as do especially the green parts of vegetables" (260) with "Cheers". The copy-text for Part l is the manuscript (RA 220.orr340); for Part II it is the version printed in The Monist. Collations with the other printed versions of the paper have been made; the results are recorded in the Textual Notes.

11

T

IO

20

30

HE C~ASSIFICA TION OF philosophies is effected, as a rule, either by their methods or by their results: "empirical" and "a priori" is a classification by methods, "realist" and "idealist" is a classification by results. An attempt to classify Bergson's philosophy in either of these ways is hardly likely to be successful, since it cuts across all the recognized divisions. But there is another way of classifying philosophies, less precise, but perhaps more helpful to the non-philosophical; in this way, the principle of division is according to the predominant desire which has led the philosopher to philosophize. Thus we shall have philosophies of feeling, inspired by the love of happiness; theoretical philosophies, inspired by the love of knowledge; and practical philosophies, inspired by the love of action. Among philosophies of feeling we shall place all those which are primarily optimistic or pessimistic, all those that offer schemes of salvation or try to prove that salvation is impossible; to this class belong most religious philosophies. Among theoretical philosophies we shall place most of the great systems; for though the desire for knowledge is rare, it has been the source of most of what is best in philosophy. Practical philosophies, on the other hand, will be those which regard action as the supreme good, considering happiness an effect and knowledge a mere instrument of successful activity. Philosophies of this type would have been common among Western Europeans, if philosophers had been average men; as it is, they have been rare until recent times, in fact their chief representatives are the pragmatists and M. Bergson. In the rise of this type of philosophy we may see, as M. Bergson himself does, the revolt of the modern man of action against the authority of Greece, and more particularly of Plato; or we may connect it, as Dr. Schiller apparently would, with Imperialism and the motor-car. The modern world calls for such a philosophy, and the success which it has achieved is therefore not surprising. M. Bergson's philosophy, unlike most of the systems of the past, is dualistic: the world, for him, is divided into two disparate portions, on the one hand Life, on the other Matter, or rather that inert something which the intellect views as Matter. The whole universe is the clash and conflict of two opposite motions: Life, which climbs upward, and Matter, which falls downward. Life is one great force, one vast vital impulse, given once for all from the beginning of the world, meeting the resistance of matter, struggling to break a way through matter, learning gradually to I

40

The abbreviations of the titles of the works of M. Bergson referred to are: C. E., Creative Evolution; M. and M., Matter and Memory; T. and F. W., Time and Free Will. The references are to the English translations of M. Bergson's books.

320

28 THE PHILOSOPHY OF BERGSON

321

use matter by means of organization; divided by the obstacles it encounters into diverging currents, like the wind at the street-corner; partly subdued by matter through the very adaptations which matter forces upon it; yet retaining always its capacity for free activity, struggling always to find new outlets, seeking always for greater liberty of movement amid the opposing walls of matter. Evolution is not primarily explicable by adaptation to environment; adaptation explains only the turns and twists of evolution, like the windings of a road approaching a town through hilly country. But this simile is not quite adequate: there is no town, no definite goal, at the end of the road along which evolution travels. Mechanism and IO teleology suffer from the same defect: both suppose that there is no essential novelty in the world. Mechanism regards the future as implicit in the past, since it believes the future to be calculable; teleology also, since it believes that the end to be achieved can be known in advance, denies that any essential novelty is contained in the result. As against both these views, though with more sympathy for teleology than for mechanism, M. Bergson maintains that evolution is truly creative, like the work of an artist: an impulse to action, an undefined want, exists beforehand, but until the want is satisfied it is impossible to know the nature of what will satisfy it. For example, we may suppose some vague desire in sightless animals 20 to be able to be aware of objects before they were in contact with them; this led to efforts which finally resulted in the creation of eyes. Sight satisfied the desire, but could not have been imagined beforehand. For this reason, evolution is unpredictable, and determinism cannot refute the advocates of free will. This broad outline is filled in by an account of the actual development of life on the earth. The first division of the current was into plants and animals: plants aimed at storing up energy in a reservoir, animals aimed at using energy for sudden and rapid movements. "The same impetus", he says, "that has led the animal to give itself nerves and nerve centres 30 must have ended, in the plant, in the chlorophyllian function" (C. E., p. 120). But among animals, at a later stage, a new bifurcation appeared: instinct and intellect became more or less separated. They are never wholly without each other, but in the main intellect is the misfortune of man, while instinct is seen at its best in ants, bees, and Bergson. The division between intellect and instinct is fundamental in his philosophy, much of which is a kind of Sandford and Merton, with instinct as the good boy and intellect as the bad boy. Instinct at its best is called intuition. "By intuition", he says, "I mean instinct that has become disinterested, selfconscious, capable of reflecting upon its object and of enlarging it indef- 40 initely" (C. E., p. 186). The account of the doings of intellect is not always easy to follow, but if we are to understand Bergson we must do our best. Intelligence or intellect, "as it leaves the hands of nature, has for its

322

10

20

30

40

28 THE PHILOSOPHY OF BERGSON

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

chief object the inorganic solid" (C. E., p. 162); it can only form a clear idea of the discontinuous aqd the immobile (pp. 163-4); its concepts are outside each other like objects in space, and have the same stability (p. 169). The intellect separates in space and fixes in time; it is not made to think evolution, but represents becoming as a series of states (p. 171). "The intellect is characterized by a natural inability to understand life" (p. 174); geometry and logic, which are its typical products, are strictly applicable to solid bodies, but elsewhere reasoning must be checked by common sense, which, as Bergson truly says, is a very different thing (p. 170). Solid bodies, it would seem, are something which mind has created on purpose to apply intellect to them, much as it has created chess-boards in order to play chess on them. The genesis of intellect and the genesis of material bodies, we are told, are correlative: both have been developed by reciprocal adaptation (p. 196). "An identical process must have cut out matter and the intellect, at the same time, from a stuff that contained both" (p. 210). This conception of the simultaneous growth of matter and intellect is ingenious, and deserves to be understood. Broadly, I think, what is meant is this: Intellect is the power of seeing things as separate one from another, and matter is that which is separated into distinct things. In reality there are no separate solid things, only an endless stream of becoming, in which nothing becomes and there is nothing that this nothing becomes. But becoming may be a movement up or a movement down: when it is a movement up it is called life, when it is a movement down it is what, as misapprehended by the intellect, is called matter. I suppose the universe is shaped like a cone, with the Absolute at the vertex, for the movement up brings things together, while the movement down separates them, or at least seems to do so. In order that the upward motion of mind may be able to thread its way through the downward motion of the falling bodies which hail upon it, it must be able to cut out paths between them; thus as intelligence was formed, outlines and paths appeared (p. 199), and the primitive flux was cut up into separate bodies. The intellect may be compared to a carver, but it has the peculiarity of imagining that the chicken was always the separate pieces into which the carving-knife divides it. "The intellect", Bergson says, "always. behaves as if it were fascinated by the contemplation of inert matter. It is life looking outward, putting itself outside itself, adopting the ways of inorganized nature in principle, in order to direct them in fact" (p. 170). If we may be allowed to add another image to the many by which Bergson's philosophy is illustrated, we may say that the universe is a vast funicular railway, in which life is the train that goes up, and matter is the train that goes down. The intellect consists in watching the descending train as it passes the ascending train in which we are. The obviously nobler faculty, which concentrates its at-

323

tention on our own train, is instinct or intuition. It is possible to leap from one train to the other; this happens when we become the victims of automatic habit, and is the essence of the comic. Or we can divide ourselves into parts, one part going up and one down; then only the part going down is comic. But intellect is not itself a descending motion, it is merely an observation of the descending motion by the ascending motion. Intellect, which separates things, is, according to Bergson, a kind of dream: it is not active, as all our life ought to be, but purely contemplative. When we dream, he says, our self is scattered, our past is broken into fragments (p. 212), 2 things which really interpenetrate each other are seen 10 as separate solid units: the extra-spatial degrades itself into spatiality (p. 218), which is nothing but separateness. Thus all intellect, since it separates, tends to geometry; and logic, which deals with concepts that lie wholly outside each other, is really an outcome of geometry, following the direction of materiality (pp. 222-4). Both deduction and induction require spatial intuition behind them (p. 225); "the movement at the end of which is spatiality lays down along its course the faculty of induction, as well as that of deduction, in fact, intellectuality entire". It creates them in mind, and also the order in things which the intellect finds there (p. 228). Thus logic and mathematics do not represent a positive spiritual effort (p. 224), 20 but a mere somnambulism, in which the will is suspended, and the mind is no longer active. Incapacity for mathematics is therefore a sign of grace-fortunately a very common one. As intellect is connected with space, so instinct or intuition is connected with time. It is one of the noteworthy features of Bergson's philosophy that, unlike most writers, he regards time and space as profoundly dissimilar. Space, the characteristic of matter, arises from a dissection of the flux which is really illusory, useful, up to a certain point, in practice, but utterly misleading in theory. Time, on the contrary, is the essential characteristic of life or mind. "Wherever anything lives", he says, "there is, 30 open somewhere, a register in which time is being inscribed" (C. E., p. 17). But the time here spoken of is not mathematical time, the homogeneous assemblage of mutually external instants. Mathematical time, according to Bergson, is really a form of space; the time which is of the essence of life is what he calls duration. This conception of duration is fundamental in his philosophy; it appears already in his earliest book Time and Free Will, and it is necessary to understand it if we are to have any comprehension of his system. It is, however, a very difficult conception. I do not fully understand it myself, and therefore I cannot hope to explain it with all the lucidity which it doubtless deserves. 40 2

It is noteworthy that elsewhere Bergson speaks of dreams as giving us duration more pure than in waking life (T. and F W, p. 126).

324

10

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

"Pure duration", we are told, "is the form which our conscious states assume when our ego lets itself live, when it refrains from separating its present state from its former states" (T. and F. W, p. mo). It forms the past and the present into one organic whole, where there is mutual penetration, succession without distinction (ibid.). "Within our ego, there is succession without mutual externality; outside the ego, in pure space, there is mutual externality without succession" (p. rn8). "Questions relating to subject and object, to their distinction and their union, should be put in terms of time rather than of space" (M. and M., p. 77). In the duration in which we see ourselves acting, there are dissociated elements; but in the duration in which we act, our states melt into each other (M. and M., p. 243). Pure duration is what is most removed from externality and least penetrated with externality, a duration in which the past is big with a present absolutely new. But then our will is strained to the utmost; we have to gather up the past which is slipping away, and thrust it whole and undivided into the present. At such moments we truly possess ourselves, but such moments are rare (C. E., pp. 2IO-II). Duration is the very stuff of reality, which is perpetual becoming, never something made (C. E., p. 287). It is above all in memory that duration exhibits itself, for in memory the ~ast survives in the present. Thus the theory of memory becomes of great importance in Bergson's philosophy; Matter and Memory is concerned to show the relation of mind and matter, of which both are affirmed to be real (p. vii), by an analysis of memory, which is "just the intersection of mind and matter" (p. xii). There are, to begin with, two radically different things, both of which are commonly called memory; the clear distinction between these two is one of the best things in Bergson. "The past survives", he says, "under two distinct forms: first, in motor mechanisms; secondly, in independent recollections" (M. and M., p. 87). For example, a man is said to remember a poem if he can repeat it by heart, that is to say, if he has acquired a certain habit or mechanism enabling him to repeat a former action. But he might, at least theoretically, be able to repeat the poem without any recollection of the previous occasions on which he has read it· thus there . ' is no consciousness of past events involved in this sort of memory. The second sort, which alone really deserves to be called memory, is exhibited in .recollections of separate occasions when he has read the poem, each umque and with a date. Here there can be no question of habit, since each event only occurred once, and had to make its impression immediately. It is suggested that in some way everything that has happened to us is remembered, but as a rule only what is useful comes into consciousness. Apparent failures of memory, it is argued, are not really failures of the mental part of memory, but of the motor mechanism for bringing memory

28 THE PHILOSOPHY OF BERGSON

325

into action. This view is supported by a discussion of brain physiology and the facts of amnesia, from which it is held to result that true memory is not a function of the brain (M. and M., p. 315). The past must be acted by matter, imagined by mind (M. and M., p. 298). Memory is n~t an emanation of matter; indeed the contrary would be nearer the truth if we mean matter as grasped in concrete perception, which always occupies a certain duration (M. and M., p. 237). "Memory must be, in principle, a power absolutely independent of matter. If, then, spiri~ is a reality,. it ~s here, in the phenomena of memory, that we may come mto touch with It IO experimentally" (M. and M., p. 81). . At the opposite end from pure memory Bergson places pure percept10n, in regard to which he adopts an ultra-realist position. "In pure perception", he says, "we are actually placed outside ourselves, we touch the reality of the object in an immediate intuition" (p. 84). So complete.ly does he identify perception with its object that he almost refuses to call it mental at all. "Pure perception", he says, "which is the lowest degree of mind-mind without memory-is really part of matter, as we understand matter" (M. and M., p. 297). Pure perception is constituted by dawning action, its actuality lies in its activity (M. and M., p. 74). It is in this way that the brain becomes relevant to perception, for the brain is not an in- 20 strument of representation, but an instrument of action (M. and M., p. 83). The function of the brain is to limit our mental life to what is practically useful. But for the brain, one gathers, everything would be perceived but in fact we only perceive what interests us (cf. M. and M., P· 34). "The body, always turned towards action, has for its essential function to limit, with a view to action, the life of the spirit" (M. and M., P· 233). It is, in fact, an instrument of choice. We must now return to the subject of instinct or intuition, as opposed to intellect. It was necessary first to give some account of duration and memory' since Bergson's theories of duration and memory are . . presup. . . 30 posed in his account of intuition. In man, as he now exists, mtmuon is the fringe or penumbra of intellect: it has been thrust out of the cen:re by being less useful in action than intellect, but it .has deeper uses "".h1ch make it desirable to bring it back into greater prommence. Bergson WIShes to make intellect "turn inwards on itself, and awaken the potentialities of intuition which still slumber within it" (C. E., p. 192). The relation between instinct and intellect is compared to that between sight and touch. Intellect, we are told, will not give knowledge of things at a di~tanc~; indeed the function of science is said to be to explain all perceptions m terms of touch. "Instinct alone", he says, "is knowledge at a distance. It 40 has the same relation to intelligence that vision has to touch" (C. E., p. 177). We may observe in passing that, as appears in many passages, Bergson is a strong visualizer, whose thought is always conducted by means of

326

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

visual images. Many things which he declares to be necessities of all thought are, I believe, characteristic of visualizers, and would not be true of those who think by means of auditory images. He always exalts the sense of sight at the expense of the other senses, and his views on space would seem to be largely determined by this fact. I shall return to this question at a later stage. The essential characteristic of intuition is that it does not divide the world into separate things, as the intellect does; although Bergson does not use these words, we might describe it as synthetic rather than analytic. IO It apprehends a multiplicity, but a multiplicity of interpenetrating processes, not of spatially external bodies. There are in truth no things: "things and states are only views, taken by our mind, of becoming. There are no things, there are only actions" (C. E., p. 26I). This view of the world, which appears difficult and unnatural to intellect, is easy and natural to intuition. Memory affords an instance of what is meant, for in memor~ the past lives on into the present and interpenetrates it. Apart from mmd, the world would be perpetually dying and being born again; ~he past would have no reality, and therefore there would be no past. It is memory, with its correlative desire, that makes the past and the future 20 real and therefore creates true duration and true time. Intuition alone can understand this mingling of past and future: to the intellect they remain external, spatially external as it were, to one another. Under the guidance of intuition, we perceive that "form is only a snapshot view of a transition" (C. E., p. 319), and the philosopher "will see the material world melt back into a simple flux" (C. E., p. 390). Closely connected with the merits of intuition is Bergson's doctrine of freedom and his praise of action. "In reality", he says, "a living being is a centre of action. It represents a certain sum of contingency entering into the world, that is to say, a certain quantity of possible action" (C. E., p. 30 276)_. The. arguments against free will depend partly upon assuming that the mten~ny of psychical states is a quantity, capable, at least in theory, of numerical measurement; this view Bergson undertakes to refute in the first chapter of Time and Free Will. Partly the determinist depends, we are told, upon a confusion between true duration and mathematical time which Bergson regards as really a form of space. Partly, again, the deter~ minist rests his case upon the unwarranted assumption that, when the state of the brain is given, the state of the mind is theoretically determinate. Bergson is willing to admit that the converse is true, that is to say, that the state of brain is determinate when the state of mind is given, but 40 he regards the mind as more differentiated than the brain, and therefore holds that many different states of mind may correspond to one state of brain. He concludes that real freedom is possible; "we are free when our acts spring from our whole personality, when they express it, when they

28 THE PHILOSOPHY OF BERGSON

327

have that indefinable resemblance to it which one sometimes finds between the artist and his work" (T. and F. W., p. 172). In the above outline, I have in the main endeavoured merely to state Bergson's views, without giving the reasons adduced by him in favour of their truth. This is easier than it would be with most philosophers, since as a rule he does not give reasons for his opinions, but relies on their inherent attractiveness, and on the charm of an excellent style. Like the advertisers of Oxo, he relies upon picturesque and varied statement, and an apparent explanation of many obscure facts. Analogies and similes, especially, form a very large part of the whole process by which he recommends his views to the reader. The number of similes for life to be found in his works exceeds the number in any poet known to me. Life, he says, is like a shell bursting into fragments which are again shells (C. E., p. 103). It is like a sheaf (ibid., p. 104). Initially, it was "a tendency to accumulate in a reservoir, as do especially the green parts of vegetables" (ibid., p. 260). But the reservoir is to be filled with boiling water from which steam is issuing; "jets must be gushing out unceasingly, of which each, falling back, is a world" (ibid., p. 261). Again, "life appears in its entirety as an immense wave which, starting from a centre, spreads outwards, and which on almost the whole of its circumference is stopped and converted into oscillation: at one single point the obstacle has been forced, the impulsion has passed freely" (ibid., p. 280). Then there is the great climax in which life is compared to a cavalry charge. All organized beings, from the humblest to the highest, from the first origins of life to the time in which we are, and in all places as in all times, do but evidence a single impulsion, the inverse of the movement of matter, and in itself indivisible. All the living hold together, and all yield to the same tremendous push. The animal takes its stand on the plant, man bestrides animality, and the whole of humanity, in space and in time, is one immense army galloping beside and before and behind each of us in an overwhelming charge able to beat down every resistance and to clear many obstacles, perhaps even death. (C. E., pp. 285-6) But a cool critic, who feels himself a mere spectator, perhaps an unsympathetic spectator, of the charge in which man is mounted upon animality, may be inclined to think that calm and careful thought is hardly compatible with this form of exercise. When he is told that thought is a mere means of action, the mere impulse to avoid obstacles in the field, he may feel that such a view is becoming in a cavalry officer, but not in a philosopher, whose business, after all, is with thought; he may feel that in the passion and noise of violent motion there is no room for the fainter

IO

20

30

40

328

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

music of reason, no leisure for the distinterested contemplation in which gre~tn~ss is_ sought, not by turbulence, but by the greatness of the universe which is mirrored. In that case, he may be tempted to ask whether there are any reasons for accepting such a restless view of the world. And if he asks this question, he will find, if I am not mistaken, that there is no rea_s?n whatever for accepting this view, either in the universe or in the wntmgs of M. Bergson. II

ro

20

The two foundations of Bergson's philosophy, in so far as it is more than an i~aginati~re and poetic view of the world, are his doctrines of space ~nd time. His doctrine of space is required for his condemnation of the mtellect, and if he fails in his condemnation of the intellect the intellect will succeed in its condemnation of him, for between the t~o it is war to the knife. His doctrine of time is necessary for his vindication of freedom for hi_s escape from what William James called a "block universe", for hi~ d?ctnne of a perpetual flux in which there is nothing that flows, and for his whole account of the relations between mind and matter. It will be well, therefore, i? criticism, to concentrate on these two doctrines. If they are true, such rumor errors and inconsistencies as no philosopher escapes wo~ld n?t g.reatly .matter, while if they are false, nothing remains except an imagmative epic, to be judged on esthetic rather than on intellectual grounds. I shall begin with the theory of space, as being the simpler of the two.

Berg~on's theory of space occurs fully and explicitly in his Time and F'_ree Will, and therefore belongs to the oldest parts of his philosophy. In his first chapter, he contends that greater and less imply space, since he regards the greater as essentially that which contains the less. He offers no argur:ients whatever, either good or bad, in favour of this view; he merely ~xclaims, as though he were giving an obvious reductio ad absurdum: "As 30 if one could still speak of magnititude where there is neither multiplicity nor sp~ce!" (p. 9): The obvious cases to the contrary, such as pleasure and pam, afford him much difficulty, yet he never doubts or re-examines the dogma with which he starts. In his next chapter, he maintains the same thesis as regards number. "As soon as we wish to picture number to ourselves", he says, "and not ~erely figures or words, we are compelled to have recourse to an extended ~mage" ~~· 78), and "every clear idea of number implies a visual image m space (p. 79). These two sentences suffice to show, as I shall try to prove: that Bergson does not know what number is, and has himself no 40 clear id~a of it. This is shown also by his definition: "Number may be defined m ~eneral as a collection of units, or, speaking more exactly, as the synthesis of the one and the many" (p. 75).

28 THE PHILOSOPHY OF BERGSON

329

In discussing these statements, I must ask the reader's patience for a moment while I call attention to some distinctions which may at first appear pedantic, but are really vital. There are three entirely different things which are confused by Bergson in the above statements, namely: (1) number, the general concept applicable to the various particular numbers; (2) the various particular numbers; (3) the various collections to which the various particular numbers are applicable. It is this last that is defined by Bergson when he says that number is a collection of units. The twelve apostles, the twelve tribes of Israel, the twelve months, the twelve signs of the zodiac, are all collections of units, yet no one of them is the number ro 12, still less is it number in general, as by the above definition it ought to be. The number 12, obviously, is something which all these collections have in common, but which they do not have in common with other collections, such as cricket elevens. Hence the number 12 is neither a collection of twelve terms, nor is it something which all collections have in common; and number in general is a property of 12 or 11 or any other number, but not of the various collections that have twelve terms or eleven terms. Hence when, following Bergson's advice, we "have recourse to an extended image" and picture, say, twelve dots such as are obtained by 20 throwing double sixes at dice, we have still not obtained a picture of the number 12. The number 12, in fact, is something more abstract than any picture. Before we can be said to have any understanding of the number 12, we must know what different collections of twelve units have in common, and this is something which cannot be pictured because it is abstract. Bergson only succeeds in making his theory of number plausible by confusing a particular collection with the number of its terms, and this again with number in general. The confusion is the same as if we confused a particular young man with youth, and youth with the general concept "period of human life", 30 and were then to argue that because a young man has two legs, youth must have two legs, and the general concept "period of human life" must have two legs. The confusion is important because, as soon as it is perceived, the theory that number or particular numbers can be pictured in space is seen to be untenable. This not only disproves Bergson's theory as to number, but also his more general theory that all abstract ideas and all logic are derived from space; for the abstract 12, the common property of all dozens as opposed to any particular dozen, though it is never present to his mind, is obviously conceivable and obviously incapable of being pictured in space. 40 But apart from the question of numbers, shall we admit Bergson's contention that every plurality of separate units involves space? Some of the cases that appear to contradict this view are considered by him, for

330

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

example successive sounds. When we hear the steps of a passer-by in the street, he says, we visualize his successive positions; when we hear the strokes of a bell, we either picture it swinging backwards and forwards, or we range the successive sounds in an ideal space (T. and F. W, p. 86). But these are mere autobiographical observations of a visualizer, and illustrate the. remark we made before, that Bergson's views depend upon the predommance of the sense of sight in him. There is no logical necessity to range the strokes of a clock in an imaginary space: most people, I imagine, count them without any spatial auxiliary. Yet no reason is alleged ro by Bergson for the view that space is necessary. He assumes this as obvious, and proceeds at once to apply it to the case of times. Where there seem to be different times outside each other, he says, the times are pict~red as s~read. out in space; in real time, such as is given by memory, different times mterpenetrate each other, and cannot be counted because they are not separate. The view that all separateness implies space is now supposed established, and is used deductively to prove that space is involved wherever there is obviously separateness, however little other reason there may be for suspecting such a thing. Thus abstract ideas, for example, obviously 20 exclude each other: whiteness is different from blackness, health is different from sickness, folly is different from wisdom. Hence all abstract ideas involve space; and therefore logic, which uses abstract ideas, is an offshoot of geometry, and the whole of the intellect depends upon a supposed habit of picturing things side by side in space. This conclusion, upon which Bergson's whole condemnation of the intellect rests, is based, so far as can be discovered, entirely upon a personal idiosyncrasy mistaken f?r a necessity of thought, I mean the idiosyncrasy of visualizing success10ns as spread out on a line. The instance of numbers shows that, if Bergson were in the right, we could never have attained to the abstract ideas 30 which are supposed to be thus impregnated with space; and conversely, the fact that we can understand abstract ideas (as opposed to particular ~hings w~ich exemplify them) seems sufficient to prove that he is wrong m regardmg the intellect as impregnated with space. One of the bad effects of an anti-intellectual philosophy, such as that of Bergson, is that it thrives upon the errors and confusions of the intellect. Hence it is led to prefer bad thinking to good, to declare every momentary difficulty insoluble, and to regard every foolish mistake as revealing the bankruptcy of intellect and the triumph of intuition. There are in Bergson's works many allusions to mathematics and science, and 40 to a careless reader these allusions may seem to strengthen his philosophy greatly. As regards science, especially biology and physiology, I am not competent to criticize his interpretations. But as regards mathematics he has deliberately preferred traditional errors in interpretation to the ~ore

28 THE PHILOSOPHY OF BERGSON

331

modern views which have prevailed among mathematicians for the last half century. In this matter, he has followed the example of most philosophers. In the eighteenth and early nineteenth centuries, the infinitesimal calculus, though well developed as a method, was supported, as regards its foundations, by many fallacies and much confused thinking. Hegel and his followers seized upon these fallacies and confusions, to support them in their attempt to prove all mathematics self-contradictory. Thence the Hegelian account of these matters passed into the current thought of philosophers, where it has remained long after the mathematicians have removed all the difficulties upon which the philosophers rely. And so long as the main object of philosophers is to show that nothing can be learned by patience and detailed thinking, but that we ought rather to worship the prejudices of the ignorant under the title of "reason" if we are Hegelians, or of "intuition" if we are Bergsonians, so long philosophers will take care to remain ignorant of what mathematicians have done to remove the errors by which Hegel profited. Apart from the question of number, which we have already considered, the chief point at which Bergson touches mathematics is his rejection of what he calls the "cinematographic" representation of the world. Mathematics conceives change, even continuous change, as constituted by a series of states; Bergson, on the contrary, contends that no series of states can represent what is continuous, and that in change a thing is never in any state at all. This view that change is constituted by a series of changing states he calls cinematographic; this view, he says, is natural to the intellect, but is radically vicious. True change can only be explained by true duration; it involves an interpenetration of past and present, not a mathematical succession of static states. This is what is called a "dynamic" instead of a "static" view of the world. The question is important, and in spite of its difficulty we cannot pass it by. Bergson's position is illustrated-and what is to be said in criticism may also be aptly illustrated-by Zeno's argument of the arrow. Zeno argues that, since the arrow at each moment simply is where it is, therefore the arrow in its flight is always at rest. At first sight, this argument may not appear a very powerful one. Of course, it will be said, the arrow is where it is at one moment, but at another moment it is somewhere else, and this is just what constitutes motion. Certain difficulties, it is true, arise out of the continuity of motion, if we insist upon assuming that motion is also discontinuous. These difficulties, thus obtained, have long been part of the stock-in-trade of philosophers. But if, with the mathematicians, we avoid the assumption that motion is also discontinuous, we shall not fall into the philosopher's difficulties. A cinematograph in which there are an infinite number of films, and in which there is never a next film because an infinite number come between any two, will perfectly represent a con-

ro

20

30

40

332

IO

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

28 THE PHILOSOPHY OF BERGSON

tinuous motion. Wherein, then, lies the force of Zeno's argument? Zeno belonged to the Eleatic school, whose object was to prove that there could be no such thing as change. The natural view to take of the world is that there are things which change; for example, there is an arrow which is now here, now there. By bisection of this view, philosophers have developed two paradoxes. The Eleatics said that there were things but no changes; Heraclitus and Bergson said that there were changes but no things. The Eleatics said there was an arrow, but no flight; Heraclitus and Bergson said there was a flight but no arrow. Each party conducted its argument by refutation of the other party. How ridiculous to say there is no arrow! say the "static" party. How ridiculous to say there is no flight! ·say the "dynamic" party. The unfortunate man who stands in the middle and maintains that there is both the arrow and its flight is assumed by the disputants to deny both; he is therefore pierced, like St. Sebastian, by the arrow from one side and by its flight from the other. But we have still not discovered wherein lies the force of Zeno's argument. Zeno assumes, tacitly, the essence of the Bergsonian theory of change. That is to say, he assumes that when a thing is in a process of continuous change, even if it is only change of position, there must be in the thing some internal state of change. The thing must, at each instant, be intrinsically different from what it would be if it were not changing. He then points out that at each instant the arrow simply is where it is, just as it would be if it were at rest. Hence he concludes that there can be no such thing as a state of motion, and therefore, adhering to the view that a state of motion is essential to motion, he infers that there can be no motion and that the arrow is always at rest. Zeno's argument, therefore, though it does not touch the mathematical account of change, does, prima facie, refute a view of change which is not unlike M. Bergson's. How, then, does M. Bergson meet Zeno's argument? He meets it by denying that the arrow is ever anywhere. After stating Zeno's argument, he replies: "Yes, if we suppose that the arrow can ever be in a point of its course. Yes again, if the arrow, which is moving, ever coincides with a position, which is motionless. But the arrow never is in any point of its course" (C. E., p. 325). This reply to Zeno, or a closely similar one concerning Achilles and the Tortoise, occurs in all his three books. Bergson's view, plainly, is paradoxical; whether it is possible, is a question which demands a discussion of his view of duration. His only argument in its favour is the statement that the mathematical view of change "implies the absurd proposition that movement is made of immobilities" (C. E., p. 325). But the apparent absurdity of this view is merely due to the verbal form in which he has stated it, and vanishes as soon as we realize that motion implies relations. A friendship, for example, is made out of people who are friends, but not out of friendships;

a genealogy is made out of men, but not out of genealogies. So a motion is made out of what is moving, but not out of motions. It expresses the fact that a thing may be in different places at different times, and that the places may still be different however near together the times may be. Bergson's argument against the mathematical view of motion, therefore, reduces itself, in the last analysis, to a mere play upon words. And with this conclusion we may pass on to a criticism of his theory of duration. Bergson's theory of duration is bound up with his theory of memory. According to this theory, things remembered survive in memory, and thus interpenetrate present things: past and present are not mutually external, but are mingled in the unity of consciousness. Action, he says, is what constitutes being; but mathematical time is a mere passive receptacle, which does nothing and therefore is nothing (C. E., p. 4I). The past, he says, is that which acts no longer, and the present is that which is acting (M. and M., p. 74). But in this statement, as indeed throughout his account of duration, Bergson is unconsciously assuming the ordinary mathematical time; without this, his statements are unmeaning. What is meant by saying "the past is essentially that which acts no longer" (his italics), except that the past is that of which the action is past? The words "no longer" are words expressive of the past; to a person who did not have the ordinary notion of the past as something outside the present, these words would have no meaning. Thus his definition is circular. What he says is, in effect, "the past is that of which the action is in the past". As a definition, this cannot be regarded as a happy effort. And the same applies to the present. The present, we are told, is "that which is acting" (his italics). 3 But the word "is" introduces just that idea of the present which was to be defined. The present is that which is acting as opposed to that which was acting or will be acting. That is to say, the present is that whose action is in the present, not in the past or in the future. Again the definition is circular. An earlier passage on the same page will illustrate the fallacy further. "That which constitutes our pure perception", he says, "is our dawning action .... The actuality of our perception thus lies in its activity, in the movements which prolong it, and not in its greater intensity: the past is only idea, the present is ideo-motor" (ibid.). This passage makes it quite clear that, when Bergson speaks of the past, he does not mean the past, but our present memory of the past. The past when it existed was just as active as the present is now; if Bergson's account were correct, the present moment ought to be the only one in the whole history of the world containing any activity.

333

IO

20

30

3 Similarly in Matter and Memory (p. 193) he says it is a question whether the past has 40 ceased to exist, or has only ceased to be useful. The present, he says, is not that which is, but that which is being made. The words I have italicized here really involve the usual view of time.

334

10

20

30

40

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

In earlier times there were other perceptions, just as active, just as actual in their day, as our present perception; the past, in its day, was by no means only idea, but was in its intrinsic character just what the present is now. This real past, however, Bergson simply forgets; what he speaks of is the present idea of the past. The real past does not mingle with the present. Our memory of the past does of course mingle with the present, since it is part of it; but that is a very different thing. The whole of Bergson's theory of duration and time rests throughout on the elementary confusion between the present occurrence of a recollection and the past occurrence which is recollected. But for the fact that time is so familiar to us, the vicious circle involved in his attempt to deduce the past as what is no longer active would be obvious at once. As it is, what Bergson gives is an account of the difference between perception and recollection-both present facts-and what he believes himself to have given is an account of the difference between the present and the past. As soon as this confusion is realized, his theory of time is seen to be simply a theory which omits time altogether. The confusion between present remembering and the past event remembered, which seems to be at the bottom of Bergson's theory of time, is an instance of a more general confusion which, if I am not mistaken, vitiates a great deal of his thought, and indeed a great deal of the thought of most modern philosophers-I mean the confusion between an act of knowing and that which is known. In memory, the act of knowing is in the present, whereas what is known is in the past; thus by confusing them the distinction between past and present is blurred. In perception, the act of knowing is mental, whereas what is known is (at least in one sense) physical or material; thus by confusing the two, the distinction between mind and matter is blurred. This enables Bergson to say, as we saw, that "pure perception, which is the lowest degree of mind ... is really part of matter". The act of perceiving is mind, while that which is perceived is (in one sense) matter; thus when these two are confused, the above statement becomes intelligible. Throughout Matter and Memory, this confusion between the act of knowing and the object known is indispensable. It is enshrined in the use of the word "image", which is explained at the very beginning of the book. 4 He there states that, apart from philosophical theories, everything that we know consists of "images", which indeed constitute the whole universe. He says: "I call matter the aggregate of images, and perception of matter these same images referred to the eventual action of one partic4 Bergson's use of the word "image" is made clearer by a very penetrating analysis of Berkeley in a recent article, "L'Intuition philosophique" (Revue de metaphysique et de morale, Nov. 19II). This article displays very distinctly the profound influence of Berkeley on Bergson's thought. Bergson's "image" is practically Berkeley's "idea".

28 THE PHILOSOPHY OF BERGSON

335

ular image, my body" (M. and M., p. 8). It will be observed that matter and the perception of matter, according to him, consist of the very same things. The brain, he says, is like the rest of the material universe, and is therefore an image if the universe is an image (p. 9). Since the brain, which nobody sees, is not, in the ordinary sense, an image, we are not surprised at his saying that an image can be without being perceived (p. 27); but he explains later on that, as regards images, the difference between being and being consciously perceived is only one of degree (p. 30). This is perhaps explained by another passage in which he says: "What can be a non-perceived material object, an image not imag- ro ined, unless it is a kind of unconscious mental state?" (p. 183). Finally (pp. 304-5) he says: "That every reality has a kinship, an analogy, in short a relation with consciousness-this is what we concede to idealism by the very fact that we term things 'images'." Nevertheless he attempts to allay our initial doubt by saying that he is beginning at a point before any of the assumptions of philosophers have been introduced. "We will assume", he says, "for the moment that we know nothing of theories of matter and theories of spirit, nothing of the discussions as to the reality or ideality of the external world. Here I am in the presence of images" (p. 1). And in the new Introduction which he wrote for the English edition he says: "By 20 'image' we mean a certain existence which is more than that which the idealist calls a representation, but less than that which the realist calls a thing-an existence placed halfway between the 'thing' and the 'representation'" (pp. vii-viii). The distinction which Bergson has in mind in the above is not, I think, the distinction between the imaging as a mental occurrence and the thing imaged as an object. He is thinking of the distinction between the thing as it is and the thing as it appears, neither of which belongs to the subject. The distinction between subject and object, between the mind which thinks and remembers and has images on the one hand, and the objects 30 thought about, remembered, or imaged-this distinction, so far as I can see, is wholly absent from his philosophy. Its absence is his real debt to idealism; and a very unfortunate debt it is. In the case of "images", as we have just seen, it enables him first to speak of images as neutral between mind and matter, then to assert that the brain is an image in spite of the fact that it has never been imaged, then to suggest that matter and the perception of matter are the same thing, but that a non-perceived image (such as the brain) is an unconscious mental state; while finally, the use of the word "image", though involving no metaphysical theories whatever, nevertheless implies that every reality has "a kinship, an analogy, in 40 short a relation" with consciousness. All these confusions are due to the initial confusion of subject and object. The subject-a thought or an image or a memory-is a present fact

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

28 THE PHILOSOPHY OF BERGSON

in me; the object may be the law of gravitation or my friend Jones or the old Campanile of Venice. The subject is mental and is here and now. Therefore, if subject and object are one, the object is mental and is here and now; my friend Jones, though he believes himself to be in South America and to exist on his own account, is really in my head and exists in virtue of my thinking about him; St. Mark's Campanile, in spite of its great size and the fact that it ceased to exist ten years ago, still exists, and is to be found complete inside me. These statements are no travesty of Bergson's theories of space and time; they are merely an attempt to show what is the actual concrete meaning of those theories. The confusion of subject and object is not peculiar to Bergson, but is common to many idealists and many materialists. Many idealists say that the object is really the subject, and many materialists say that the subject is really the object. They agree in thinking these two statements very different, while yet holding that subject and object are not different. In this respect, we may admit, Bergson has merit, for he is as ready impartially to identify subject with object as to identify object with subject. As soon as this identification is rejected, his whole system collapses: first his theories of space and time, then his belief in real contingency, then his condemnation of intellect, then his account of the relations of mind and matter, and last of all his whole view that the universe contains no things, but only actions, movements, changes, from nothing to nothing, in an endless alternation of up and down. Of course a large part of Bergson's philosophy, probably the part to which most of its popularity is due, does not depend upon argument, and cannot be upset by argument. His imaginative picture of the world, regarded as a poetic effort, is in the main not capable of either proof or disproof. Shakespeare says life's but a walking shadow, Shelley says it is like a dome of many-coloured glass, Bergson says it is a shell which bursts into parts that are again shells. If you like Bergson's image better, it is just as legitimate. The good which Bergson hopes to see realized in the world is action for the sake of action. All pure contemplation he calls "dreaming", and condemns by a whole series of uncomplimentary epithets: static, Platonic, mathematical, logical, intellectual. Those who desire some prevision of the end which action is to achieve are told that an end foreseen would be nothing new, because desire, like memory, is identified with its object. Thus we are condemned, in action, to be the blind slaves of instinct: the life-force pushes us on from behind, restlessly and unceasingly. There is no room in this philosophy for the moment of contemplative insight when, rising above the animal life, we become conscious of the greater ends that redeem man from the life of the brutes. Those to whom activity without purpose seems a sufficient good will find in Bergson's books a pleasing

picture of the universe. But those to whom action, if it is to be of any value, must be inspired by some vision, by some imaginative foreshadowing of a world less painful, less unjust, less full of strife than the world of our every-day life, those, in a word, whose action is built on contemplation, will find in this philosophy nothing of what they seek, and will not regret that there is no reason to think it true.

336

ro

20

30

40

337

29 METAPHYSICS AND INTUITION

29

Metaphysics and Intuition [1913]

the fact that we insist on passing from space to movement, from the trajectory to the flight, from immobile positions to mobility, and on passing from one to the other by way of addition. (45) "Bergson always simply forgets relations" is Russell's comment. The copy-text is the earliest printed version.

THIS REVIEW WAS published in The Cambridge Review, 34 (17 April 1913): 3767. It was reprinted in The Cambridge Mind (1970), edited by Eric Hornberger, et al. On 4 March 1913 Russell told Lady Ottoline Morrell that he had "read for review a little book consisting of a translation of Bergson's Introduction a la metaphysique" (#715). The next day he reported:

In the morning I wrote my review of Bergson-not one joke, however mild-this in consequence of my disgust the other day. I do not feel with him, as I did with James, that he is a man of first-rate ability whom I happen to disagree with-there is something fundamentally rhetorical and false about his writing-it never has the simplicity of real thought. Without the style it would be nothing. The review and innumerable letters took up my morning. (#716) Neither the book nor the review is mentioned again. Russell added the book to his library. It contains many vertical lines in pencil as well as a half-dozen or so verbal annotations, the most scathing of which is "ROT [True in history but not in logic]" against the second sentence in this passage: This inversion has never been practised in a methodical manner; but a profoundly considered history of human thought would show that we owe to it all that is greatest in the sciences, as well as all that is permanent in metaphysics. The most powerful of the methods of investigation at the disposal of the human mind, the infinitesimal calculus, originated from this very inversion. (59-60) The "inversion" of which Bergson writes is the deliberate cultivation of "intellectual sympathy which we call intuition" as a substitute for the usual mental operations. This passage draws an annotation from Russell: The difficulties to which the problem of movement has given rise from the earliest antiquity have originated in this way. They result always from

338

339

29 METAPHYSICS AND INTUITION

An Introduction to Metaphysics. By Henri Bergson. Authorized translation by T.E. Hulme. London: Macmillan, 1913. Pp. vi, 79.

T

HIS LITTLE BOOK is a translation, revised by M. Bergson, of an article which appeared in the Revue de metaphysique et de morale in 1903, before Creative Evolution, but after Matter and Memory. The translation is excellent: so far as I have verified it, I have found only two points in which it differs from the original, namely, on p. 5 the translation has "extraordinarily simple" where the original has merely "simple", and on p. 77 it has "infinitely simple" where the original has "simplicity it10 self". The work well deserves translation, since it affords an admirable "Introduction to [M. Bergson's] Metaphysics". The one topic dealt with is the nature of "intuition", and the way in which it differs from "intellect". In the first paragraph we are told that there are "two profoundly different ways of knowing a thing. The first implies that we move round an object; the second that we enter into it." The difference between the two lies chiefly in the fact that the first involves analysis, while the second does not. Analysis, according to M. Bergson, consists essentially in the mention of qualities which the object analyzed appears to share with other objects, whereas, since everything is 20 in truth unique, description by qualities which may be shared is description by what is external and not really characteristic. Although it is, of course, true that every object is, in a certain sense, unique, the use which M. Bergson makes of this fact to discredit analysis does not seem justified. Like most enemies of analysis, and like the older empiricist supporters of analysis, he invariably assumes that in analyzing no account is to be taken of the relations of the parts. "Suppose I am shown", he says, "the letters which make up a poem I am ignorant of. If the letters were parts of the poem, I could attempt to reconstitute the poem with them by trying the different possible arrange30 ments, as a child does with the pieces of a Chinese puzzle. But I should never for a moment think of attempting such a thing in this case, because the letters are not component parts, but only partial expressions, which is quite a different thing." But as an argument against analysis, this is wholly fallacious. A complete analysis would mention also the order of the letters, and then the difficulty would disappear. The letters A, R, T, for example, do not constitute a complete analysis of the word "art"; they may equally form the word "rat" or the word "tar". But if we are told further that A is to come before R, and R before T, the analysis is complete and the word is no 40 longer doubtful. Exactly the same answer applies to M. Bergson's objections to the analysis of motion into a series of positions in a certain temporal order. He assumes always that the analysis retains only the positions,

340

341

and ignores the time-relations which give the order. Hence he is led to regard motion as indivisible and ultimate, and moving things as fictions. "All reality, therefore, is tendency, if we agree to mean by tendency an incipient change of direction." He regards his view as the only alternative to the opinion that rest is more real than motion, that "there is more in the immutable than in the moving"-an opinion which he attributes to Plato, and regards as infecting, more or less, every philosophy except his own. It is natural to the intellect, he says, to seek out the likenesses among processes, to ignore what is unique and therefore unpredictable, and to divide movements embodying a single impulse into parts which are purely ro fictitious. If we wish to become philosophers, we must resist this vice of the intellect. "To philosophize", he says, "is to invert the habitual direction of the work of thought." But if a philosopher is to persuade us of so momentous a thesis, he must first persuade us that he knows what is relevant in "the work of thought". M. Bergson does not persuade us of this. The nature of analysis, the mathematical theory of motion, the difference between predication and division into parts, are matters upon which it is vitally necessary to him to understand the best that his opponents can say. So far from such an understanding, however, we find, in his accounts of the views he is combating, a mere jumble of antiquated ideas acquired in 20 youth and never revised. It may be that something could be said for his philosophy even by one who knew what is to be said against it; but such knowledge is closed to M. Bergson by the habit of rhetoric and intellectual impatience from which few philosophers have been exempt.

30 MR. WILDON CARR'S DEFENCE OF BERGSON

30

Mr. Wildon Carr's Defence of Bergson [1913]

THIS REPLY w AS published in The Cambridge Magazine, 2 (26 April 1913): 490, 492. It was written in response to "On Mr. Russell's Reasons for Supposing That Bergson's Philosophy Is Not True" by H. Wildon Carr, published in the same journal on 12 April. For the convenience of the reader Carr's article is reprinted in Appendix IX, and the letter he sent to Russell after Russell's rejoinder appeared is printed in the Headnote to that Appendix. Both Carr's article and Russell's reply were reprinted, along with Russell's original essay on Bergson (28), as a pamphlet by The Heretics in 1914. Russell wrote his reply to Carr's criticism on 17 April 1913, a very busy day, as he told Lady Ottoline Morrell the next morning: Yesterday I got through a good deal-I wrote a reply to a criticism of my article on Bergson which Wildon Carr had published in the Cambridge Magazine-I read half of a book I have to review for the Nation, I had a profitable discussion with Moore, which led me to write out a sketch of Theory of Knowledge, much better than anything I had done before in that line; then I had my lecture, then I had my evening, ending with Broad (the man whose dissertation I was reading at Ipsden), who is also working on matter, which we argued about. By that time it was one o'clock, but I was still quite fresh. I find I am quite extraordinarily fit. (#747) Herbert Wildon Carr (1857-1931) was a stockbroker who made a sufficient fortune to retire, while still relatively young, in order to pursue his first love, philosophy. For many years he was the Honorary Secretary of the Aristotelian Society. After his retirement he was elected Professor of Philosophy in the University of London, where he served until he was appointed to a similar post in the University of Southern California. His own philosophical position was a sort of monadism. On 17 July 1957, in response to a query from Sverre Lyngstad as to whether Carr had influenced Virginia Woolf; Russell had this to say about him: I do not think that Wildon Carr had any influence of importance on anybody. He was originally in the Stock exchange where he made a comfortable fortune. Philosophy was his hobby and he used to invite selected 342

343

groups of philosophers to extremely pleasant meals. We described him to each other as the Maecenas of British philosophy but he was no more capable of philosophy than Maecenas was of poetry. It is true that after a time he endowed a Professorship of Philosophy on the understanding that he should be the first holder of the chair and in the University of Southern California he seems to have been taken seriously. He was an ardent admirer of Bergson ... I think it probable that Virginia Woolf, at some time or another met Wildon Carr, but I am quite certain that he cannot have influenced her since no one in philosophical circles took him seriously. He was an amiable and kindly old gentleman but not a philosophical force. My sister in law, Elizabeth (author of Elizabeth in her German Garden) used to call him "Woolly Lamb" a nick-name which perfectly suited him. Carr's patronage of philosophers extended to those in prison. When Russell was a prisoner in 1918 Carr made himself extremely useful by supplying Russell with the books he required to carry out his writing projects and seeing that his manuscripts were typed and sent off to journal editors. The copy-text is the version printed in The Cambridge Magazine. The two versions have been collated; the results are reported in the Textual Notes.

30 MR. WILDON CARR'S DEFENCE OF BERGSON N THE NUMBER of the Cambridge Magazine for April 12 Mr. Wildon Carr published a reply to certain criticisms of Bergson which I had published in The Monist, and the editor has invited me to make a rejoinder to Mr. Carr's reply. At the outset, it seems necessary to clear up a misconception of my purpose. I did not attempt to prove that "Bergson's philosophy is not true", if we mean by his philosophy the conclusions at which he arrives rather than the reasons which he gives for them. The conclusion of the first part of my paper, quoted by Mr. Carr, is that "there is no reason whatever for accepting this view"; the conclusion of the second part is almost verbally the same, namely "that there is no reason to think it true". These phrases were intentional. M. Bergson's philosophy, like all other ambitious systems, is supported by arguments which I believe to be fallacious, but it does not follow that it is in fact false. I hold that much less can be known about the universe as a whole than many philosophers are inclined to suppose; I should not therefore assert dogmatically that the universe is other than it is said to be in this or that system, unless the account in question appeared self-contradictory. What I do maintain is that, in view of the mistakes in Bergson's reasoning, his conclusions remain mere imaginative possibilities, to be placed alongside of the thousand other possibilities invented by cosmic poets. Mr. Carr, however, in spite of an apparent concession in his first paragraph, proceeds to defend Bergson's arguments; and we must therefore proceed to examine his defence. In supposing Bergson to be a visualizer, it appears I was mistaken; but the important point remains, that his speculation is dominated by the sense of sight to a remarkable extent, and that this seems connected with the importance which he assigns to space. Mr. Carr next considers the distinction which I emphasize between (1) the general concept Number, (2) the particular numbers, (3) the various collections to which numbers are applicable. He says I allow "that Bergson's doctrine that intellectual apprehension compels us to have recourse to an extended image is true of meaning (3)". This is a misunderstanding; the view in question is examined and rejected in the paragraph on p. 329 beginning "But apart from the question of numbers, shall we admit Bergson's contention that every plurality of separate units involves space?" Hence the inferences drawn by Mr. Carr from my supposed concession fall to the ground. The next question raised, as to the order in which we come to know the above three meanings, appears to me logically irrelevant, and it is only under protest that I am willing to consider it. He asks: "Can we apprehend the number 12 if we have never had acquaintance with any particular instance of 12 units?" He supposes that my answer must be in the negative, because I say that the number 12 is "something which all these collections

I

IO

20

30

40

344

345

of 12 units have in common", and he supposes that this is a definition of 12. It is not a definition, and does not have the form of a definition. And I certainly hold that we might apprehend the number 12 without having acquaintance with any particular dozen. I have not, so far as I know, ever been acquainted with a collection of 34,361 units, yet I apprehend the number 34,361. But it is impossible to pursue this topic without raising the whole question of our acquaintance with universals. With regard to Zeno, Mr. Carr says that I remove one paradox only to leave him with a greater. I admit that this impression is partly my fault, and that I have not always been sufficiently careful to display my slavish IO adherence to common sense. But in the main the impression-which Mr. Carr shares with many philosophers who have tried to understand the mathematical theory of infinity and continuity-is due to the almost unconscious drawing of fallacious inferences. For instance, if I say "no part of Tristram Shandy's biography would remain permanently unwritten", I am supposed to imply that some day the biography will be finished, which is by no means implied, and in the circumstances supposed is plainly false. This applies to Mr. Carr's doubt whether, on my principles, the tortoise cannot overtake Achilles. I say that, if they go on for ever, every place reached by Achilles will ultimately be reached by the tortoise; 20 and at first sight this seems inconsistent with the statement that, after Achilles has passed the tortoise, the distance between them will continually increase. But this apparent inconsistency disappears as soon as the matter is understood. With regard to the phrase "from nothing to nothing", Mr. Carr says I ignore Bergson's doctrine that "nothing" is a pseudo-idea. This is a misunderstanding. I hold just as strongly as Bergson (though for different reasons) that "nothing" is a pseudo-idea, I used the phrase, as it ordinarily would be used, as an abbreviation for the phrase "not from anything and not to anything". 30 With regard to the confusion of subject and object with which I charge Bergson, Mr. Carr says that as regards intuition "so far from the identity of subject and object being a confusion, it is of the very essence of the doctrine." It was precisely my contention that it was of the essence of the doctrine; but I fail to see how this proves that it is not a confusion. It seems to me that only one who has never clearly distinguished subject and object can accept Bergson's "intuition". In the case of memory, this seems particularly evident, since it becomes necessary for Bergson to identify remembering with what is remembered, and therefore to say that whatever is remembered still endures. To say that such identification is of the 4o essence of his doctrine is no defence; the only valid defence would be to show that remembering is in fact identical with what is remembered.

346

10

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

In conclusion, I must admit that there is an element of question-begging in all refutations of Bergson. When we have shown that this or that doctrine is self-contradictory, we have only shown that it does not appeal to the intellect; if the intellect is in fact misleading, as Bergson contends, it is useless to employ it against him. It is true that Bergson continually employs it in his own defence, by advancing arguments which plainly are intended to be intellectually satisfying. But this perhaps is a concession to the unconverted: when his philosophy has triumphed, it is to be supposed that argument will cease, and intellect will be lulled to sleep on the heaving sea of intuition. But until that consummation the protests of intellect will continue.

Part

VII

Critique of Idealism

31

Some Explanations in Reply to Mr. Bradley [1910]

THIS PAPER WAS published in Mind, 19 (July 1910): 373-8. It was written in response to certain criticisms which Bradley had published in the April number of the same journal. For the reader's convenience, the two parts of that paperpages l59n-16on and 178-85--which contain these criticisms are reprinted in Appendix III. In the January 1911 issue of Mind Bradley published "Reply to Mr. Russell's Explanations"; it is also reproduced in Appendix III. On 20 February 1910 Bradley wrote to Russell on the "etiquette as to Testimonials" and also to tell him that his paper criticizing him was about to appear: I was in any case writing to you about an article of mine to appear in The Mind in April. I have appended to this some critical remarks on some of your ideas. I have done this with great reluctance because I feel sure I only partially understand them. I perhaps understand them just enough to misunderstand them. I have very little capacity for following a great deal of what you say-even when it is not mathematical. There, of course, I understand nothing at all. The reason I publish this is because no one, or almost no one (I don't forget Joachim) has raised the doubts and difficulties which have occurred to me-though I am sure that many have felt them. I hope that even if what I say is all wrong, it therefore may do good in a sense that it may lead to certain issues being made clearer. And, if so, I don't mind being mistaken. I think also that I am right in assuming that you can bear being criticised mistakenly. I wish that I were thirty years younger or had had your book thirty or forty years ago. Don't trouble to answer this which (as to the article) is premature. I expect to return in a few weeks. Russell wrote a detailed answer to his criticisms on 9 April: I waited to answer your letter of some time back, and to thank you for the offprints you kindly sent me, until I had seen your article in this month's Mind. And now to begin with I wish to thank you for your very great courtesy, and to express the encouragement which I derive from your words of praise. With regard to the matter of your criticisms, I cannot of course yet give a considered answer, but on various points (some of them trivial) I 349

350

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

31 SOME EXPLANATIONS IN REPLY TO MR. BRADLEY

should like to make some preliminary remarks. p. l59n (351: 37). I don't think I can have ever meant to say that existence exists (I have not got my book here). What I wanted to urge was that there are true judgments about existence, and that these are not existential, because existence does not exist. (I should no longer be prepared to lay much stress upon this argument as it stands.) p. 179 (389: 19-37). Yes, I am a strict pluralist, but I do not consider pluralism incompatible with the existence of complex entities. I consider that in every case where two simples have a relation, there is a complex entity consisting of the two simples so related. Where I differ (I think) from you is in holding that, in this case, the simples are themselves constituents of the complex, and that the fact of their entering into the complex does not involve any complexity in their natures. I do not affirm (though I hesitate to deny) that the universe forms a single complex. p. 180 (390: ro-28). I use the word "implication" in a special technical sense which does not carry with it the consequences you indicate. I say that one proposition "implies" another whenever the first is false or the second true (not excluding both). I do not pretend that this is the usual meaning of the word, but it is a relation for which I need a name, and no other name occurred to me. As for "such that", this applies to a subject which is "variable", and the conception of the variable is the conception of something standing midway between particular and universal; I do not pretend to have solved all the difficulties in this conception, but at any rate I have discussed them at length. p. 180 (390: 32-391: 1 l). I doubt whether the question of the possibility of a term being related to itself is as real as you suggest. The word "relation" has a more general meaning with me than is usual in philosophy. All that I am really concerned to affirm is that a term may occur twice in a proposition. For example, 2 + 2 = 4 asserts (with my meaning of "relation") that 2 has to 2 that relation which x has toy whenever x + y = 4. I think probably you attribute a greater significance to this statement than I intend. p. 181 (391: 12-393: 6). With regard to the conception of "class", I admitted in my Principles that I had not yet found a satisfactory theory, but I believe that now I have found a satisfactory theory, enabling me to interpret propositions verbally concerned with classes without assuming that there are classes. A class is many, and therefore not one; but I accept the principle "ens et unum convertuntur". By wholly denying the reality of classes, the dilemma is avoided. I have explained this view in the current number of the Revue de Metaphysique et de Morale, of which (with your permission) I will send you an offprint as soon as I get one. It appears to be implied in consistent pluralism, that, though there are many things, there is nothing which is many. These two statements now appear to me to be reconcileable, and I therefore accept both.

As regards possibility, I have thought a good deal about the subject, and my view is that, possibility and necessity are not fundamental notions; when used as such, they appear to me erroneous. I should express what I mean by saying that they attach to propositional functions, not to propositions. I should define a propositional function x as possible when there is at least one x of which it is true, and as necessary when it is true of every x for which it is significant (i.e. true or false). p. 183 (393: 7-24). I readily admit that negation requires more attention than I have given it, though I have given it more than appears in my writings. It is a difficult subject, and I have been unwilling to say much about it as yet. I do not, however, wholly admit the element of negation in the meaning of such words as "some" and "any". This is because I do not regard what is implied as necessarily part of the mean-

351

ing. p. 184 (393= 25-394: 17). On the subject of the quantitative (not the numerical) zero, I do not feel sure that the views in my Principles can be defended. The whole subject of quantity lies outside pure mathematics, and Part III of my book is really a digression, not bound up with the rest. If I had to re-write it, I should, I think, alter the account of zero, but in a way which would make no essential or fundamental change. It seems to me that the question of external relations is the one which dominates all others. Much of what you say appears to me to assume that there are no external relations; and I think (though I may be mistaken) that when you have explicitly argued against external relations, you have argued against something not quite the same as what their advocates mean. This again is connected, I believe, with the question of necessity; it is urged that if relations are external, terms which have a certain relation might have failed to have it. This statement, in view of my rejection of possibility as a fundamental notion, would need much interpretation before I could accept it. In consequence of the view expressed in your letter, I wrote a testimonial for Joachim. I very much hope he will get the post. There is some prospect of my getting drawn into politics, but if that fails, I shall endeavour to write something for Mind on the points you raise. The reference top. 159n. is to this passage: At the same time the very form of predication prevents any judgment from being perfectly true (Appearance, p. 544). Subject to this condition the above doctrine to my mind holds good. There is an objection, raised by Mr. Russell (Principles of Math., p. 450), that on this view you cannot say that "Reality is real" or that "Existence exists''. No truth (I have

352

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

31 SOME EXPLANATIONS IN REPLY TO MR. BRADLEY

just stated) can upon my view be perfectly true, but, apart from that, I should find it easier to deal with this objection if I were told the sense in which any one ever could want to say that Reality is real. To affirm that Reality has the character of reality, I presume, is harmless, while to suggest that Reality is a member of a class "real," to my mind is monstrous. And it would be of course wrong to call it "real," in some sense which would restrict it. With regard to "Existence exists," once more, until I know exactly what that means, I can hardly reply. What I can say is this, that to place "Existence" itself within the sphere of existence would be clearly indefensible.

In what you say about a term being related to itself, as well as elsewhere, I perceive that we are at cross-purposes as to axioms. When I speak of an axiom, I do not mean that we argue downwards from the general axiom to the particular case of it. On the contrary, I should say that the particular case is usually more evident and more worthy of credence than the general axiom, and that the general axiom is arrived at by starting from the consideration of particular cases. The general axiom is a compendious form for a number of self-evident propositions which arc its instances (it is not merely this, but that does not matter at present). Your account of why you think diversity essential to a relation is just what I should mean by saying this is an axiom to you. (a) You examine various special cases, in which it appears to you self-evident that diversity is essential. (b) It also appears to you self-evident that this result did not depend upon their being those cases. (c) Hence you are led to the general statement. This is precisely the process by which logical axioms are reached. The step (b) is apt to be ignored, but it is important, and has the same degree of generality as (c). What you call "the result of an ideal experiment" seems closely akin to what I call "self-evidence". I have been endeavouring to state what I mean by saying relations are external in the Journal of Philosophy, Psychology and Scientific Method. I have no copy at present, but if I get one I will venture to send it to you.

The "above doctrine" to which Bradley refers is this: "All judgment (I have argued elsewhere) predicates its idea of the ultimate Reality." In a letter written on I I April Bradley promises to "attend to what you have written in your letter" and expresses "some alarm at the prospect of you being occupied with politics, if that means that you will have no time for philosophy", because "no one else will do your work in philosophy so far as human probability goes". After Russell's reply was published in Mind Bradley wrote again, on 23 September 19rn: Many thanks for the off-print of your paper. I unfortunately did not see the number of Mind until about a fortnight ago, having been out of England. Otherwise I would have written one or two further remarks. But I will try to do this for the next number. You will see from this that I should like to have, if possible, some further explanation on certain points where I still fail to realize your meaning. It is unfortunate that I did not have Mind sent on or I could have done this already. On 2 March I9II Russell wrote to Bradley sketching the thoughts which had occurred to him after studying Bradley's "Reply to Mr. Russell's Explanations": I am much ashamed of not having written sooner about your article in the January Mind. I have been hesitating as to whether to write an answer, but on the whole I came to the conclusion that the questions at issue are too fundamental to be treated shortly, and that I must wait until I could treat them systematically. I had realized that a not had dropped out on P. 74, I, but it was kind of you to write. With regard to unities, I have nothing short to say. The subject is difficult (in any philosophy, I should say), and I do not pretend to have solved all its problems. With regard to implication, I should say that disjunctions are facts, and deducibilities likewise. I was meaning merely to urge that such things merely are so, and have no special property of necessity.

353

The article to which Russell refers at the end of his letter is "The Basis of Realism" (13). The dropped "not" on page 74 makes Bradley's sentence (395: 29-30) read: "What I of course am forced to assume here is that I have not correctly performed my experiment." Bradley replied on 15 March 1911: Many thanks for your letter. I was afraid that the subjects were too large for compendious treatment. I don't know if you think it worth while to insert a statement in Mind to that effect. I should hardly think that it is necessary however. I am glad to find that I was wrong as to axioms and really there seems to be little, if any, difference of opinion between us as to the nature of an axiom. I have lately written a long article in which I have touched briefly on this subject. I will leave it as it was, as in any case it helps to explain my position. Russell did not continue the controversy publicly. Bradley's "long article" is entitled "On Some Aspects of Truth"; it was published in Mind in July of the same year. See his Essays on Truth and Reality (r9r4, 311-14), where the article is reprinted, for the passage on axioms. The printed version is the copy-text.

31 SOME EXPLANATIONS IN REPLY TO MR. BRADLEY R. BRADLEY' s VERY courteous examination in the April Mind of some of the views advocated in my Principles of Mathematics calls for some explanations, not by way of polemic, but by way of elucidation. I shall not attempt, in what follows, to give the grounds for my views, since that would require a volume, but only to state, as clearly as I can, what they are and what they are not, how far I admit the justice of Mr. Bradley's criticisms, and how far I believe that they can be answered. I hold, as Mr. Bradley states, that mathematical truth is wholly and ultimately true, and that the contradictions with which it has appeared to be infected can all be removed by patience in distinguishing and defining. These contradictions are broadly of two kinds (not sharply distinguishable), namely those which are specifically mathematical, such as the traditional difficulties concerning infinity, and those which, though relevant to mathematics, belong properly to general philosophy. It is especially the second kind of difficulties with which Mr. Bradley is concerned. It has seemed to me that these difficulties were all connected with a certain doctrine as to the nature of relations which, though widely held, has been, so far as my knowledge goes, more explicitly and effectively advocated by Mr. Bradley than by any other philosopher. I shall not here repeat my reasons for rejecting this view, but shall content myself with trying to state my own view. Mr. Bradley finds an inconsistency in my simultaneous advocacy of a strict pluralism and of "unities which are complex and which cannot be analyzed into terms and relations". It would seem that everything here turns upon the sense in which such unities cannot be analyzed. I do not admit that, in any strict sense, unities are incapable of analysis; on the contrary, I hold that they are the only objects that can be analyzed. What I admit is that no enumeration of their constituents will reconstitute them, since any such enumeration gives us a plurality, not a unity. But I do not admit that they are not composed of their constituents; and what is more to the purpose, I do not admit that their constituents cannot be considered truly unless we remember that they are their constituents. The view which I reject holds (if I understand it aright) that the fact that an object x has a certain relation R to an object y implies complexity in x and y, i.e., it implies something in the "natures" of x and y in virtue of which they are related by the relation R. It seems to be held that otherwise all relations would be purely fortuitous, and might just as well have been other than they are, and this, it is thought, would be intolerable. This opinion seems to rest upon some law of sufficient reason, some desire to show that every truth is "necessary". I am inclined to think that a large part of my disagreement with Mr. Bradley turns on a disagreement as to the notion of "necessity". I do not myself admit necessity and pos-

M IO

20

30

40

354

355

sibility as fundamental notions: it appears to me thc1t fundamentally truths are merely true in fact, and that the search for a "sufficient reason" is mistaken. I can see many ways of defining necessity which will account for its common uses: we may call a proposition necessary when it follows from a proposition known to be true, or when it can be known without empirical evidence, or when what is affirmed would be equally true of any other subject. And whatever meaning of "necessity" we adopt, we obtain, of course, a corresponding meaning of "possibility": a proposition is possible when its contradictory is not necessary. But none of the above meanings of necessity and possibility justify the traditional doctrines as to mo- IO dality, or the objection which philosophers are apt to feel to a "mere fact". I do not mean to deny that one fact is often deducible from another; but such deducibility is in turn a fact, i.e. it has no modal property of necessity not possessed by the facts which it relates. To return to relations: I maintain that there are such facts as that x has the relation R toy, and that such facts are not in general reducible to, or inferable from, a fact about x only and a fact about y only: they do not imply that x and y have any complexity, or any intrinsic property distinguishing them from a z and a w which do not have the relation R. This is what I mean when I say that relations are external. But I maintain also- 20 and it is here that Mr. Bradley sees an inconsistency-that whenever we have two terms x and y related by a relation R, we have also a complex, which we may call "xRy", consisting of the two terms so related. This is the simplest example of what I call a "complex" or a "unity". What is called analysis consists in the discovery of the constituents of a complex. A complex differs from the mere aggregate of its constituents, since it is one, not many, and the relation which is one of its constituents enters into it as an actually relating relation, and not merely as one member of an aggregate. I confess I am at a loss to see how this is inconsistent with the above account of relations, and I suspect that the meaning which I attach 30 to the word "external" is different from Mr. Bradley's meaning; in fact he seems to mean by an "external" relation a relation which does not relate. The word "implication" occurs very frequently in my writings, and I am afraid I did not state with sufficient emphasis that, for technical purposes, I was using this word in a technical sense different from its usual sense. For the purposes of symbolic logic, it is convenient to define "p implies q" in the widest manner which will enable us, when we know that p "implies" q, and that pis true, to infer that q is true. The widest meaning of "p implies q" which will secure this is "either p is false or q is true", 40 where the alternatives are to be taken as not mutually exclusive. 1 With r I have stated this with, I think, the necessary clearness, in the American Journal of Mathematics, Vol. xxx, p. 245: "'p implies q' is to mean 'pis false or q is true'. I do not mean

356

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

this meaning of "implication", it seems plain that the consequences of which Mr. Bradley speaks (p. I8o) are not involved. On the question of implication, there is however a substantial as well as a verbal difference between Mr. Bradley's views and mine. I admit that in inference, we use a relation somewhat different from that which I hav~ defined as implication. When we infer, the premiss and the implication are known to us first, and are the means by which we come to know the conclusion. This requires that premiss and conclusion should be so related as to enable us to perceive the implication, and this in turn requires some IO formal relation between premiss and conclusion such as those considered in the rules of formal logic. But this formal relation is only required in order that we may perceive the implication, and what we perceive when we perceive the implication is that either the premiss is false or the conclusion is true. Thus implication as above defined is still the fundamental logical concept, and what is further required for inference is psychological, namely such conditions as shall enable us to perceive the implication without knowing first whether the conclusion is true or the premiss false. !he view advocated by Mr. Bradley, that what can be inferred is always m some sense already contained in the premiss, is one which I cannot 20 accept. I shall return to this point in connection with negation. The next question raised is the question whether a term can be related to itself. Here the question is, I think, partly-though only partly-solved by pointing out the sense in which I use the word "relation". For my purposes, any proposition of which x and y are constituents asserts a relation of x toy, and the proposition which results from replacing x and y by z and w asserts the same relation between z and w. Thus in order to be able to assert that a term may be related to itself, it is only necessary to show that a term may occur twice in one proposition. For example " x +y = 4" asserts a re1ation . between x and y, and since 2 + 2 = 4 this' 30 relation is one which 2 has to itself. Mr. Bradley would, no doubt, 'deny ~hat the self-same term, without any diversity, could occur twice, either m the same proposition or in two different propositions. This question is connected with the fundamental question as to the nature of relations. Mr. Bradley invites me to say that a term is diverse from itself, on the ground that "it is all one to the term what its relations are" (p. I8I). There is here an appeal to the law of sufficient reason, and an assumption that, if there is nothing in the nature of a term to compel it to have one relation to itself rather than another, it must have both relations to itself or neither.

31 SOME EXPLANATIONS IN REPLY TO MR. BRADLEY

What was said above about necessity seems to cover this case: I should say that it is a fact that every term is identical with itself and not diverse from itself. As for the assertion that "diversity is required for a relation" (p. 18I), I can only say that no reason is alleged for holding this view, and that for my part I see no reason to hold it. As regards what Mr. Bradley says about the idea of "class", I find myself very largely in agreement with him. The theory of classes which I set forth in my Principles was avowedly unsatisfactory. 2 I did not, at that time, see any way of stating the elementary propositions of Arithmetic without employing the notion of "class". I have, however, since that time discovered that it is possible to give an interpretation to all propositions which verbally employ classes, without assuming that there really are such things as classes at all. 3 Apart from other contradictions, the fact that a class, if there is such a thing, must be both one and many constitutes a difficulty. That it is meaningless (as Mr. Bradley contends) to regard a class as being or not being a member of itself, must be assumed for the avoidance of a more mathematical contradiction; but I cannot see that this could be meaningless if there were such things as classes. The theory that there are no such things as classes avoids at once the difficulties raised by Mr. Bradley and the difficulties with which I endeavour to contend in the Principles. The general contention that classes are a mere fac;on de parler has, of course, been often advanced, but it has not been accompanied by an exact account of what this manner of speaking really means, or by an interpretation of arithmetic in accordance with this contention; and such an accompaniment was essential before a philosophy of mathematics could dispense with classes. On the subject of zero quantity (as opposed to the number 0), I am no longer prepared wholly to defend the view which Mr. Bradley has criticized, but the correction which it requires, while avoiding the difficulties (I can hardly admit that they are contradictions) emphasized by him, introduces no change of philosophic importance. The question of the definition of the quantitative zero now seems to me one of mainly technical interest, not one of interest to philosophy. The subject of negation, on the contrary, is of the highest interest to philosophy, and I fully admit that it calls for more notice than I have given it. I do not think, however, that negation offers any special difficulty in the logic which I advocate; indeed, I am led to hesitate, not by being

to affir~ t~at. 'implies' cannot have any other meaning, but only that this meaning is the one which 1.t 1s m~st convenient to give to 'implies' in symbolic logic." In the Principles, I defin~d d1sJunct1on m terms .of implication, rather than vice versa; but this is merely a quest10n of taste and convenience.

IO

20

30

"In the case of classes, I must confess, I have failed to perceive any concept fulfilling the conditions requisite for the notion of class. And the contradiction discussed in chapter x proves that something is amiss, but what this is I have hitherto failed to discover" 40 (Principles, pp. v-vi). 3 I have explained briefly how this is to be done in the American Journal of Mathematics, Zoe. cit.

2

40

357

358

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

unable to think of any theory of negation which would be tenable, but by the fact that two different theories appear prima facie equally tenable. In a negative judgment, we may place the negation either in the act, or in that which is judged. In the first case, there will be no such thing as believing a negative proposition; there will merely be disbelief in positive propositions. In the second case, there will be both positive and negative propositions, and negation will, in the case of negative judgments, be part of the content of that which we believe. There are no doubt arguments to be found which would decide between these two theories, or perhaps ro in favour of some mediating theory. Meanwhile, I cannot as yet find any fundamental objection to either view. Mr. Bradley's contention that such notions as "a man" or "any man" contain an element of negation is one which I cannot admit. "'A man"', he says, "appears to assert one instance of man and to deny more than one man" (p. I83). It is of course undeniable that "a man" implies the denial of more than one man; but it does not follow that this denial is part of its content. Such a view involves the assumption-implicit in many such arguments-that all inference is essentially analytic, that whatever can be inferred from a proposition is necessarily part of that proposition. 20 This view appears to me to be erroneous, and to be connected with the theory of relations upon which most of my disagreements with Mr. Bradley depend. Exactly similar remarks apply to "any", "every", "all", and "some". In all these cases, the negation appears to me to be merely implied, and to be no part of the content. It is, of course, highly probable that there are difficulties in my position which I have failed to appreciate; meanwhile, the chief hope of philosophical progress seems to lie in the endeavour to discover clearly the exact points of difference between divergent views. For example, it appears selfevident to Mr. Bradley that a relation implies diverse terms, whereas to 30 me this appears by no means self-evident. Such a state of things is eminently unsatisfactory, and seems to lead to a deadlock. In favour of the premisses from which I start, there is, however, a kind of inductive argument: they allow much more truth to science and common sense than is allowed by the opposite premisses, and they do not require us to "condemn, almost without a hearing, the great mass of phenomena". I should not lay stress upon this argument, but for the fact that, where there is a dispute as to fundamentals, more strictly philosophical arguments become impossible. The progress of philosophy seems to demand that, like science, it should learn to practise induction, to test its premisses by the 40 conclusions to which they lead, and not merely by their apparent selfevidence. To reject such a test is to assume-what none but a philosopher would assume-that metaphysical theories have a greater degree of certainty than the facts of science and of daily life.

32

The Philosophy of Theism [1912]

Tms REVIEW w AS published in The Nation, IO (IO Feb. 1912): 788. It is unsigned, but both internal and external evidence support its attribution to Russell. The writing style and its argumentation closely resemble that of other reviews by him which are contemporary with this one. His pocket diary shows that he was paid £2.2.0 by The Nation in March, and a search of the magazine for February reveals no other likely candidate for Russell's authorship. Additional support is found in his letters to Lady Ottoline Morrell. On 12 December l9II he wrote: I have been reading poor old Ward, he is dull and antiquated. Very few philosophers are much good. I think they are inferior to men of science, partly because there are no such definite tests of achievement. (#284) The next day he told her: "I have read a good deal of Ward, and ordered two more fenders-otherwise nothing has happened to me" (#286). Evidently he is finding it a chore to read the book. On 22 December he wrote: "I have finished my review of Ward, which is a comfort. It is very dull because I didn't want to say all I really thought" (#290). He mentioned the book again on 24 January 1912: Earlier in the day I went to see the James Wards-I didn't talk about his book. He was reading Bosanquet's Gifford lectures, and thought as ill of them as I do of his. Philosophers are a censorious crew. He spent his time looking up all the passages about me in Bosanquet. (#323) Russell's review of Bosanquet's book is printed in this volume as Paper 34. A day later Russell visited McTaggart: "He had been reading Ward and thought as ill of him as I do" (#324). James Ward (1843-1925) was one of Russell's teachers at Cambridge. In 1880 he had been appointed lecturer; in 1897 he was elected to a new chair in logic and mental philosophy, a position he held at his death. In philosophy he defended a form of pluralistic idealism derived from that of Rudolf Hermann Lotze under whom he had studied. His work in psychology was important in turning British psychology away from the associationism of Alexander Bain. It was Ward who

359

360

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

suggested to Russell the topic of non-Euclidean geometry for his Fellowship dissertation. In Portraits from Memory (1956) Russell paid a public tribute to Ward's memory:

For James Ward ... I had a profound respect and a considerable affection. He was my chief teacher in philosophy and, although afterwards I came to disagree with him, I have remained grateful to him, not only for instruction, but for much kindness. (61-2) And in his Autobiography (1967) he noted Ward's importance in his development as a student of the foundations of mathematics: During my fourth year I read most of the great philosophers as well as masses of books on the philosophy of mathematics. James Ward was always giving me fresh books on this subject, and each time I returned them, saying that they were very bad books. I remember his disappointment, and his painstaking endeavours to find some book that would satisfy me. In the end, but after I had become a Fellow, I got from him two small books, neither of which he had read or supposed of any value. They were Georg Cantor's Mannichfaltigkeitslehre and Frege's Begriffsschrift. These two books at last gave me the gist of what I wanted, but in the case of Frege I possessed the book for years before I could make out what it meant. Indeed, I did not understand it until I had myself independently discovered most of what it contained. (68) He goes on to mention that Ward was one of two examiners of his Fellowship dissertation; Whitehead was the other. Whitehead "criticized it rather severely" but Ward "praised it to the skies". Russell was elected a Fellow of Trinity College. The copy-text is the printed version.

The Realm of Ends, or, Pluralism and Theism: The Gifford Lectures Delivered in the University of St. Andrews in the Years 1907-10. By James Ward. Cambridge: At the University Press, 1911. Pp. xv, 490. HESE LECTURES, Professor Ward tells us, are intended to serve as a sequel to his earlier course on "Naturalism and _Agnostic~s~". They are a sequel in the sense that they assume certam propos1t1ons which it was the aim of the earlier course to establish, but they demand less technical knowledge than the earlier course, and deal more dire~tly with those parts of philosophy which are of most interest to non-techmcal readers. They should, therefore, appeal to a wider circle, and they ~ay ro be confidently recommended to all who wish to see the case for theism stated by an author who has studied all the object~ons t~~t may be urge_d by pantheists, materialists, and agnostics. The mam position adv~cated is very near to orthodoxy: apart from the denial of eternal pun~shment (which can hardly now be regarded as heretic~!), :he only c?ns1derab~e departure from the modern Christian view consists m the demal of God s foreknowledge of men's free choices. Accepting from "Naturalism and Agnosticism" the view that t?e. only ultimate reality is Mind, Professor Ward begins fro~ the pl:1ral_1suc assumption that there are many minds, not, as pantheists .~amtam, ~nly 20 one mind with many manifestations. Against various cnt1cs, especially Lotze he maintains that pluralism is not self-contradictory; but he holds that the systematic unity which (according to him) exists i? the univer.se, is more satisfactorily accounted for by theism than by a philosophy which accepts the co-ordination of the many as a brute fa~t. T~e rela:ion of God to the world is an ancient difficulty of philosophic theism, smce, when the dependence of the world on God is empha.sized, there. is a tendency to regard the world as in God, and so to fall mto pantheism. Professor Ward contends that, though the many depend on God for their existen~e, they depend on each other for their experie~ce; they wer~ created with 30 free will and thus their choices originate genumely new senes, not chosen or even foreseen when they were created. Thus evolution is not the mere unfolding of a plan already pre-determined in all its details: not everythi~g is decreed in advance, though not every possibility is left open. In this anxiety to preserve real initiative for human .beings, th~se lectures .s~ow the same tendency as M. Bergson's Creative Evolution, and Wilham J ames's protests against what he called a "block univers~". . The above view, besides preserving genuine freedom, is useful m dealing with the problem of evil-the most serious pro?Iem, _as has always been recognized, that theism has to face. Moral evil, which Dr. Ward 40 regards as the worst evil, appears as an inevitable co~sequence of the creation of beings endowed with the power of free will. It therefore only

T

361

362

rn

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

remains to be proved or assumed that the world was worth creating in spite of the evil that it contains. The conclusion reached at the end of the discussion of evil is that, while no theoretic proof of the existence of God is possible, there is also no theoretic disproof. This result, if the discussion were to end at this point, would lead only to agnosticism. But in the concluding lectures there is an appeal to what are called "moral arguments", in virtue of which it is contended that, if belief in God and a future life is not theoretically impossible, our moral ideals justify us in accepting it. "We cannot, it has been said, 'argue from the reality of desires to the truth of dogmas.' A good deal depends surely upon the rationality of the desires." And, again: Either the world is not rational, or man does not stand alone, and this life is not all. But it cannot be rational to conclude that the world is not rational, least of all when an alternative is open to us that leaves room for its rationality-the alternative of postulating God and a future life.

20

30

This passage reminds us of pragmatism, and it is not surprising to find William James's Will To Believe mentioned with praise. But Professor Ward does not accept the pragmatist's account of the nature of truth, and has not, therefore, the pragmatist's grounds for believing in the truth of the postulates which our moral ideals are supposed to require. It is perhaps regrettable that the fundamental principles upon which the argument rests are scarcely mentioned, and nowhere defended in the course of the book. It is assumed-for example in the passage quoted above-that we cannot refuse to decide whether the world is rational or not, even though the evidence is admittedly inadequate. Philosophical ideas, we are told, "are justified in proportion as they enable us to conceive this whole [of experience] as a complete and systematic unity". To one who regards experience as incomplete, unsystematic, and not in any vital sense a unity, the arguments contained in these lectures will make little appeal. In philosophy it is the premisses that are important and difficult, and it is on them that discussion should be concentrated. Those who dissent from Professor Ward's conclusions will probably also dissent from his premisses, and will, therefore, fail to be convinced. Those, however, who accept the premisses will find in these lectures a singularly careful, candid, and serious defence of what is most vital in the Christian faith.

33

Hegel and Common Sense [1912]

THIS REVIEW w AS published, unsigned, in The Nation, II (17 August 1912): 739-40. There is both internal and external evidence to support the claim that Russell is its author. His pocket diary records a payment of £2.2.0 from The Nation in September 1912, and there are no other likely candidates amongst the August reviews. On 12 July 1912 he wrote to Lady Ottoline Morrell that among other tasks: I have two books to review. It all seems rather futile compared to writing, but I suppose one couldn't always be writing. (#497) The internal evidence is strong. In addition to its style, there is the remark that Hegel's argument sometimes depends upon puns, a charge which Russell repeated in Our Knowledge of the External World (1914, 39n-4on). At that place too he refers to the book under review. In My Own Philosophy (1972, 20) Russell again made the same charge against Hegel. Henry Stewart Macran (1867-1937) was a Fellow of Trinity College (18921937), Professor of Moral Philosophy (1901-1934), and Professor of the History of Philosophy (1934-1937), all in the University of Dublin. In addition to this translation he also published Hegel's Logic of World and Idea (1929), a translation of the second and third sections of the subjective logic. The copy-text is the printed version.

363

33 HEGEL AND COMMON SENSE

Hegel's Doctrine of Formal Logic: Being a Translation of the First Section of the Subjective Logic. By G. W. F. Hegel. Translated by H. S. Macran. Oxford: At the Clarendon Press, I9I2. Pp. 3I5. HIS BOOK CONSISTS of a translation of the first section of the second part of Hegel's Greater Logic, together with a rather full introduction and some notes. It is to be regretted that Mr. Macran did not translate the whole work, since it forms one continuous argument, difficult enough to follow when we begin at the beginning, and almost impossible when we begin in the middle. It is a disgrace to English phi10 losophy that the task of translating the Greater Logic should still remain to be done. It is to be hoped that Mr. Macran, who has made a valuable beginning, will be persuaded to complete the work of which the present volume can only be regarded as an installment. The translation, like all translations, is not wholly free from inaccuracies; but in the main it is excellent. Where (as sometimes occurs) Hegel's argument depends upon a pun, one could wish for a note to explain the joke; but presumably Mr. Macran thought that the difficulties of Hegel were great enough without the attempt to understand his humour. There is, probably, no other of the great philosophers about whom 20 opinion is so divided as it is about Hegel. Many regard him as the discoverer of immensely important truths, as, in fact, the greatest (at least in achievement) of all philosophers. Others find in him nothing but confusions of thought, alternating with sheer nonsense. Whichever of these parties may be in the right, it seems plain that the grounds for his conclusions are chiefly to be found in his logic, and that there is very little to be said for those who adopt his results without the dialectic method by which he seeks to establish them. Mr. Macran is not among these; he is fully aware of the paramount importance of the Logic, and insists that Hegel's is "an altogether definite, sober, and methodical attempt to solve 30 the riddle of the universe". Whether he is equally right in rejecting the view that Hegel is a mystic, may be doubted. William James treated him as a man dominated by the mystic feeling of the oneness of opposites; and some certainly among his followers would regard his final view of the universe as mystical. Mr. Macran's introduction is largely concerned to persuade us that Hegel was really quite a sensible man, who did not maintain anything seriously contrary to common sense. This view seems unjust. Hegel's system is a vast effort of imagination, which has certainly the merit of suggesting an interesting view of the world, and possibly the merit of suggesting a true view, but emphatically not the merit (or de40 merit) of re-stating, in complicated terms, what ordinary sensible people believe.

T

364

365

Hegel's is the last of the great systems built on the belief that the nature of the universe can be discovered by a priori reasoning. All such systems derive their apparent force from one or more simple mistakes-generally so simple as to be hardly credible. The portion of Hegel's dialectic translated by Mr. Macran seems to rest throughout on a confusion between two meanin~s of the word is. If we say "Hildebrand is Gregory VII", or "Na~oleo~ 1s the greatest general of modern times", the word is expresses identity: the same person, in each case, is designated by either of two phrases. But if we say "Socrates is mortal", the is does not express identity: Socrates is not the same thing as mortality. Hegel, however, insists that it must express identity. Since Socrates and mortality are obviously different, he says that what is expressed is "identity-in-difference". This conception, arrived at in this way, is perhaps the main distinction of the Hegelian philosophy. Being obviously self-contradictory, it affords an admirable basis for the dialectic movement, which is propelled by contradiction. Those who cannot believe that any philosopher could commit such a simple mistake may consult pages I39, I8I, I86. For example, the argument on page I39 is essentially the following: "Socrates is a singular, mortal is a universal; therefore, in saying that Socrates is mortal, we say that the singular is a universal, which is self-contradictory." This argument turns simply on the ambiguity of is. Hegel's treatment of formal logic, like his treatment of everything else, is divided into three parts: the concept, the judgment, and the syllogism. Each passes into the next by its inherent dialectic, and within each of them there are many forms connected by dialectical transitions. For example, there is the syllogism of existence, the syllogism of reflection, and the syllogism of necessity; and this last, again, may be categorical, hypothetical, or disjunctive. As we advance we get nearer to perfect truth. "All men are mortal, and Socrates is a man; therefore, Socrates is mortal", is not so true as "If Socrates is a man, he is mortal; now he is a man, therefore he is mortal"; and this in turn is not so true as "Socrates is either a man or an immortal; now he is not an immortal, therefore he is a man." It is difficult not to regard these distinctions in degree of truth as purely fantastic; and the argument, where it is intelligible, seems always based on mere confusions. Nevertheless, Hegel's philosophy has a great fascination from its sweep and scope; and even if it is wholly false, it deserves to be studied as enriching the imaginative possibilities of thought.

10

20

30

34 THE PHILOSOPHY OF GOOD TASTE

34

The Philosophy of Good Taste [1912]

THIS REVIEW w AS published, unsigned, in The Nation, II (7 Sept. 1912): 840, 842. Russell's authorship is established by an appeal to both internal and external evidence. Internally, both the style and the content are distinctively his. In addition to the fact that he received a payment of £2.2.0 from The Nation in October 1912 and there is no other review in September which he could have written, there is the evidence from his letters to Lady Ottoline Morrell. On 25 April 1912 he wrote: "The whole way down in the train I read Bosanquet, who is unspeakably disgusting-Hegel's rotting carcase infecting the air" (#424). He returned to the subject on 23 May 1912: "I am making use of my quiet dayreading Bosanquet, who is atrocious, and thinking" (#462). On May 25th he made a longer report, some of which is echoed in his review: I haven't much time now as I waited till after the second post-reading the wretch Bosanquet. His view is that all truth is in Plato and Aristotle, who must however be interpreted in the light of Hegel, who in turn must be interpreted in the light of T. H. Green. On this basis, which one may take for granted, one may build up any elaborate system without bothering about the foundations. The root principle is that everything ought to be vague and hazy, because sharp outlines are crude. The smug dogmatism of it makes me sick-it is all bland and superior. And it bears no resemblance whatever to either Plato or Aristotle. (#466) By 28 May he could report: "This morning I finished old Bosanquet" (#467b); and the next morning he remarked: "This morning I must write my review of Bosanquet. I think it is the worst book of philosophy I ever read" (#469). That same night he reported it done: "This morning I wrote my review of Bosanquet, which took me till lunch" (#470). In none of these remarks is the title of Bosanquet's book mentioned, but a letter of 7 June establishes that Russell had been reading the book under review: Bosanquet says love and logic are practically the same thing; I found I considered this instinctively a blasphemy against logic. Some things seem removed from the horror of human life: flowers, the song of birds, the sea and the stars. Logic to me belongs with those things, and I feel it 366

367

degraded by any contact with life. Love, unfortunately, belongs with life, and therefore is not shining and clear and strong, but turbid, painful, transitory, not a thing one can worship. I make a mess of human things always, because I try to think them like the eternal things; with the eternal things I am all right, my ways of feeling are what they call for. But I have not the strength to forego all human things, which is what I ought to do. (#482) The remark that triggered this revelatory discussion was Bosanquet's claim that "it is the strict and fundamental truth that love is the mainspring of logic" (1912, 341). Bernard Bosanquet (1848-1923) was, at the time this book was published, a leader of the neo-Hegelian idealism then ascendant in Britain. Oxford-trained, he taught there for only a decade (1871-1881) before moving to London, where he pursued a career of writing and discreet social activism. Except for a five-year period (1903-1908) as Professor of Moral Philosophy in the University of St. Andrews, he held no other academic appointments. Although he and F.H. Bradley are often mentioned as the leaders of British neo-Hegelianism, they did not agree on all issues. Indeed Bosanquet's first philosophical book, Knowledge and Reality (1885), bears the subtitle, A Criticism of Mr. F.H. Bradley's "Principles of Logic". It is hardly surprising, given his opinion of the book's worth, that Russell did not include it in his library. The copy-text is the printed version.

34 THE PHILOSOPHY OF GOOD TASTE

The Principle of Individuality and Value: The Gifford Lectures for 191 l, Delivered in Edinburgh University. By B. Bosanquet. London: Macmillan, 1912. Pp. xxxvii, 409. T Is AN interesting speculation whether great men do more good by their discoveries than harm by the respect which they inspire. We all know the widow who suppresses originality in her children with the remark, "Your dear father would never have said that." Very similar is the attitude of many students of philosophy towards the philosophers whose authority they acknowledge. Dr. Bosanquet, in all questions of importance, assumes that the truth is to be found in Plato and Aristotlenot, it is true, in what they said, but in what they meant, which is to be discovered by a study of Hegel, who, in turn, is to be interpreted in the light of T. H. Green. By means of this highly selective view of the history of philosophy, he arrives at what he calls the philosophic tradition, which enables him to state his main premisses as established by the consensus of the ages. It is true that Locke, Berkeley, Hume, Kant, and the vast majority of contemporary philosophers reject these premisses, but they are all condemned as outside the main current of philosophy. "I do not conceal my belief", he says, "that in the main the work has been done, and that what is now needed is to recall and concentrate the modern mind out of its distraction, rather than to invent wholly new theoretical conceptions" (pp. v-vi). This recall from distraction is not to be effected, as in other studies, by an endeavour to ascertain the facts, but by acquiring a certain frame of mind: our attitude to life should be "sane and central" (p. 4), we should attempt "the largest and bravest attitude of soul" (p. 5), and we should remember that "bad taste is bad logic, and bad logic is bad taste" (p. 7). Most people, no doubt, judge philosophy as they judge cookery, by whether it tickles their palate; among such people there will be some who will think Dr. Bosanquet's dish agreeable, and finding a good taste will infer a good logic. But those who, in philosophy, desire clear conceptions, penetrating analyses, and some attempt to prove the truth of the positions advocated, will look in vain in this book for anything of what they seek. Dr. Bosanquet's general position is that of a monism derived from Hegel. The only thing that is completely real is the Whole or Absolute, and this is proved by means of the law of contradiction as developed in the Dialectic. "Every true proposition", he says, "is so in the last resort because its contradictory is not conceivable in harmony with the whole of experience; in other words, is not merely a contradiction of fact, but a self-contradiction" (p. 51). It should follow that, in theory, absolutely every fact should be demonstrable by pure logic. It is difficult to believe that, for example, every detail of the course of history is deducible from

I

10

20

30

40

368

369

the law of contradiction; but then history, we are told, is incapable of any considerable degree of "being or trueness" (p. 79). The same habit of picking and choosing among facts appears in many parts of the book: some of our experiences are more true or real than others; man's nature is not just what he is, but is something in process of being communicated to him (p. 259). Broadly, we may say that a fact is "real" when it confirms Dr. Bosanquet's philosophy, but if it refutes it, it is not quite "real", however undeniably it may be a fact. What is real, we are told, must be a "concrete universal", which is not a general concept, but a union of particular and general; in fact, the same thing as a true individual. "The true embodiment of the logical universal takes the shape of a world whose members are worlds" (p. 37). One would naturally suppose that if the members of a world are worlds, their members in turn must be worlds, and so on, ad infinitum. This view, however, is rejected, on the ground that Dr. Bosanquet cannot understand the modern mathematical infinite (p. 38n.), though no attempt is made to show how the infinite series is to be avoided. Individuality, it appears, is what has ultimate value, and the Absolute is a perfect individual. Finite selves are not perfect individuals and are not fully real; indeed, it would seem that they are no more real than matter. Dr. Bosanquet definitely rejects panpsychism: he sees no reason to believe that whatever appears as matter has really some form of mental life. He also sees no reason to believe that separate selves are immortal (p. 268). In a lecture on Teleology, he explains that it is a mistake to regard the purpose as something to be achieved in the future; the purpose of the whole, he says, simply is the whole (p. 162). It follows that, though his philosophy must be classed as an optimism, it affords no ground for supposing that the world will get better, or even that it will not get worse "Finiteness, pain, and evil", he says, "are essential features of Reality, and belong to an aspect of it which leaves its marks even on perfection" (pp. 240-1). Apparently men are to be exhorted to bear misfortunes patiently, on the ground that at any rate the Absolute profits by them. Something great and precious, he says, can be made of pain and sin (p. 241), but not, apparently, by the sufferer or the sinner. The Whole is perfect already, so there can be no ground to wish for any diminution of what appear to us to be evils. (Dr. Bosanquet does not, of course, draw this consequence, but it seems unavoidable.) In a lecture on "Freedom and Initiative", we are told that it is useless to try to prove that acts are not determinate, given the antecedent conditions: the act is necessary, but the agent is not (p. 354-5). This view, Dr. Bosanquet says, is derived primarily from T. H. Green, but is really the view of the philosophical tradition from Plato downwards (p. 355n.). It seems, however, as if an agent's freedom would be somewhat illusory

10

20

30

40

370

ro

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

if his act were not free. Of course it is said that the agent determines his own act, but that only slightly antedates the compulsion, since the agent himself is the product of heredity, environment, and other circumstances which lie outside himself. Dr. Bosanquet's book rests upon a certain view of logic-the view, namely, that when things are interrelated, the whole composed of them is more real than they are. This view leads obviously and quickly to contradictions, but by invoking the Hegelian dialectic the contradictions are made to minister to the development of the philosophy, instead of being used, as they would be in any other study, to show that the foundation is unsound. The assumption of a consensus as to fundamentals, which underlies the book, can only be maintained by one who ignores most of the work of philosophers in most parts of the civilized world; and so long as this is the case, it would seem that the time has not come for building elaborate systems on very dubious premisses.

35

The Twilight of the Absolute [1913]

Tms REVIEW WAS published in The Nation, 12 (22 Feb. 1913), 864. Although unsigned, it can confidently be attributed to Russell because on 3 February he wrote to Lady Ottoline Morrell: "Wrote review of Bosanquet today" ( #688). And on 26 February he mentioned it again: "Yes, if there was a review of Bosanquet in the Nation it was by me-I didn't see it myself. (Did you think very ill of it?)" (#710). For biographical information on Bosanquet see the Headnote to Paper 34. In his youth Russell read Bosanquet's Logic, or the Morphology of Knowledge (1888) and admired it. But by this time he had come to regard him as a malignant force in philosophy. The printed version is the copy-text.

371

35 THE TWILIGHT OF THE ABSOLUTE

The Value and Destiny of the Individual: The Gifford Lectures for 1912, Delivered in Edinburgh University. By B. Bosanquet. London: Macmillan, 1913. Pp. xxxii, 331. HIS BOOK, WHICH is a continuation of Dr. Bosanquet's previous Gifford Lectures on The Principle of Individuality and Value, endeavours to sum up the religious value of modern English idealism. It must be confessed that the Absolute no longer shines with the lustre which it displayed in the days of Hegel. In those days it interpreted the course of history, deduced (wrongly) the number of the planets, and proved a priori the transcendent excellence of the Prussian State. But it is to humbler functions that Dr. Bosanquet calls this awful Power. What was negative in Hegel remains: time and space are unreal; the sciences give only partial truths, infected with contradiction, and not metaphysically valid; whatever is fragmentary is insubstantial and a mere fragment of the whole. But the positive satisfaction of human hopes which is promised is very meagre. God and immortality are rejected, pain and evil are admitted to be as real as anything else that we know; even the Whole feels "pain and conflict, and the sense of an overwhelming burden". There is no good reason to expect the future to be better than the past, though perhaps we may hope that we may in time acquire a better grasp of the whole, and better standards of values than those we have at present. And, in the end, Dr. Bosanquet confesses that "the atmosphere of our pilgrimage has necessarily been sombre". For these admissions Dr. Bosanquet deserves all praise. But it may be asked whether, for so meagre an outcome, it is worth while to preserve the whole apparatus of Hegelian logic. The grounds in favour of the opinions contained in this book are mainly not to be found within it, but are to be sought ultimately in Hegel's "Logic". The school to which Dr. Bosanquet belongs takes, rightly or wrongly, a very highly selective view of what is important to philosophy. Contemporary work in other countries is almost wholly ignored; Locke, Berkeley, and Hume are recognized only to be refuted; the whole world of science, in spite of the frequent occurrence of the blessed word "evolution", has never enlarged the imaginations or vivified the conceptions of these writers. Plato and Aristotle severely bowdlerized, and interpreted by a process of torture, Hegel, Lotze, Sigwart, and T. H. Green-these, apart from living British writers, are almost the sole authors sympathetically studied. In view of the fact that no philosophy can claim anything approaching to universal assent, and that, in particular, the philosophy advocated in the present work is accepted by only a very small proportion of philosophers, it seems at best doubtful whether the elaboration of deductions from such very disputable premisses can have any great value. The business of philosophy for the

T

IO

20

30

40

372

373

present, it would seem, is rather to seek for some new method by which, as in science, it should be possible to overcome differences of temperament, and arrive at results, however few and however modest, which should be universally recognized by all competent students. But if this is to be done, the attempt to construct great ambitious systems must be abandoned, and the limits of our present capacity for knowledge must be more freely recognized. Essentially independent of Dr. Bosanquet's metaphysic, there is in his work an ethic and an attitude towards human destiny which-at least to the present reviewer-appears both profoundly true and profoundly important. Man, as we know him, Dr. Bosanquet says, is a compound of finite and infinite, of separate selfhood wrongly claiming independence, and consciousness of the whole struggling after greater comprehensiveness and harmony. The feelings which underlie mysticism appear curiously constant in different ages and countries; but the intellectual super-structures to which they give rise are extraordinarily various. Of these superstructures, Hegelianism is one of the most surprising. Overlaid by a long academic tradition, clothed in logical forms alien from its spirit and shocking to many logicians, the old vital mystic impulse may still be discerned in these lectures. Accept the world, live in the whole, adapt your wishes and demands to what the whole permits, and you will achieve a certain wisdom, and a liberation from much evil that would seem otherwise inevitable-this is, in effect, what Dr. Bosanquet's philosophy becomes when the apparatus of metaphysic is cleared away. And it is thus possible to agree with much of what he urges, even when the doctrines from which it seems to flow are utterly rejected. The book consists of three parts: the moulding of souls, the hazards and hardships of finite selfhood, and the stability and security of finite selfhood. The title of the last part is, perhaps, somewhat misleading. All that is contended is that what we ought to wish preserved in our self will be preserved; but what we ought to wish preserved is not really the particular self, but the objects to which a good life is devoted. And even these, in so far as they are not wholly good (being infected with finiteness), will not be realized exactly; but whatever it is that gives them value is eternally possessed by the whole. What is practically operative in this attitude, omitting a somewhat illogical hope for the future, seems little more than willing acquiescence; and this is possible without the dubious help of a gradually fading Absolute.

IO

20

30

36

Philosophy Made Orthodox [1913]

THIS REVIEW WAS published, unsigned, in The Nation, 13 (14 June 1913): 431. Both internal and external evidence support its attribution to Russell. The writing style and the opinions expressed are both unmistakably his. His pocket diary for 1913 records a payment of £1.1.0 from The Nation in July and there are no other suitable candidates amongst the June reviews to justify the payment. In addition, there are his remarks to Lady Ottoline Morrell. He wrote her twice on 18 April 1913. In his morning letter he said: "Yesterday ... I read half of a book I have to review for the Nation" (#747); that evening he told her: "I finished my book and wrote the review of it-the book was short and worthless" (#746). As the reader will see, this conclusion is repeated, albeit less bluntly, in the review. Frank Byron Jevons (1858-1936) joined the teaching staff of the University of Durham in 1882 as a tutor in classics. In 19IO he was elected Professor of Philosophy, a position he held until his retirement in 1930. From 1896 until 1923 he served as Principal of Bishop Hatfield's Hall in the University of Durham. He is best known for his work on the history of religion. The copy-text is the printed version.

374

Personality. By F.B. Jevons. London: Methuen, 1913. Pp. ix, 171. HIS LITTLE BOOK, we are told in the preface, forms the matter of four lectures given at Oxford in the Vacation Term for Biblical Study. Their purpose is to defend what may be called the Orthodox conception of personality-human and divine-against the attacks of psychologists and philosophers, by showing that these attacks are based upon confusions and mistakes, and that the most strict and profound thought agrees at all points with the preconceptions of the ignorant. The book is a contribution to apologetics, not to philosophy, for it is plain that the conclusions to be reached are fixed in advance, and determine the nature 10 of the discussion instead of being determined by it. The first chapter, on Personality and Impersonality, discusses the extrusion of personal agents from the scientific view of the physical world, and the pre-animistic stage in human development, when savages believed that things were endowed with some vague power not conceived as personal. The chief argument used is that "impersonal" is merely the negation of "personal", and therefore the savage cannot have conceived of impersonal power until he had the conception of personality. The second chapter discusses the attacks on the Self by Hume and James. These it dismisses by means of the verbal inconsistencies which 20 grammar forces on those who deny the Self. "When I enter most intimately into what I call myself," says Hume, "I always stumble on some particular perception or other." To this Dr. Jevons retorts by asking what is meant when it is said, "/ enter" or "I stumble", thus arriving at the conclusion that the Self is assumed by the very words in which Hume intends to deny it. Such short and easy arguments, however, never really succeed in proving what they mean to prove; it is always possible to evade them by more explicit and cumbrous phrases. Hume does not, of course, mean to deny that the word "I" can be significantly used; he means merely to deny that it denotes a single simple entity. Whether he was right or 30 wrong remains a very doubtful question; but he was certainly not so stupid as Dr. Jevons's refutation implies. Dr. Jevons's discussion of the analysis of the Self in James's Psychology is similar, and is open to similar objections. The third chapter deals with Bergson's theory that there are only changes, and no things or persons that undergo change. The refutation urged against this theory depends partly upon Kant's contention that there can be no change without a permanent subject of change-a contention which, though it has been widely accepted, reduces itself on analysis to a mere verbal definition. But the more important part of the refu- 40 tation consists of an appeal to introspection:

T

375

376

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

Persistence in change-change which is never complete changeis the characteristic and essence of our consciousness. Here if nowhere else-or, rather, here as everywhere else-existence is neither change alone, nor unchanging sameness, but sameness in change. It is an identity which does not exclude change: a change which does not exclude identity.

ro

20

Here, as before, the view advanced may be true, but it can hardly be established by such summary methods as those of Dr. Jevons. The last chapter, on Personality and Individuality, is positive rather than critical. The word "individual" is taken to mean a closed, self-contained system, and therefore, though persons are admitted, "individuals" are denied. Although he quotes with approval Dr. Bosanquet's saying, "it is the strict and fundamental truth that love is the mainspring of logic", he is obliged to part company with Dr. Bosanquet as regards the Absolute, since the Absolute is too pantheistic for the conclusions which he intends to reach. It is not very clear where his logic differs from Dr. Bosanquet's, or why it fails to lead to a similar monism. The principle of unity between persons, he says, is love; but unity is never completely attained, except between the Persons of the Trinity; for us unity is an object of striving, "to be gained only by that love which is the impulse towards unity with one's neighbor and one's God". The book may be recommended to those who already agree with its conclusions, but it is not well adapted to bring conviction to those who come to it with opinions other than its own.

Appendixes

a )

Appendix

I

F. C. S. Schiller's Replies to Papers 21 and 24 [1909-12]

SCHILLER'S COMMENTS ON "Pragmatism" (21) are contained in a letter which he sent Russell from the Hotel Hassler in Rome. The letter, dated 25 March 1909, makes it clear that Russell had paid him the courtesy of sending him proofs of his article. There is some evidence that Russell revised the article in the light of Schiller's comments, since John Dewey's University affiliation is given as "Columbia" in the published article, and the various misprints that Schiller noted are corrected. But he did not alter his statement that Peirce introduced the word "pragmatism" into philosophy in 1878, nor did he alter his discussion of Schiller's study of immortality along the lines Schiller suggests. Schiller's reactions to "Pragmatism and Logic" (24) are to be found in his letter of 19 May 1912. A "new two volume Irish Jesuit Logic" refers to The Science of Logic (1912) by Peter Coffey, who was at the time Professor of Logic and Metaphysics in Maynooth College (now St. Patrick's College), County Kildare, Ireland. The others mentioned in the letter are Charles Arthur Mercier, whose only book on logic Russell reviewed (9); Alfred Sidgwick (1850-1943), who published five books on logic between 1883 and 1914, and John Henry Muirhead (1855-1940), an idealist philosopher and editor of many important series of books.

379

APPENDIX I F.C.S. SCHILLER'S REPLIES

I.I

F.C.S. Schiller's Reply to Paper 21. Hotel Hassler Rome

25 March I909.

DEAR RUSSELL ,-Many thanks for the proof of your article, which I was sorry not to receive in person, but which reached me safely this morning. I read it of course with the utmost interest, and heartily congratulate you on it. It is most interesting and brilliant as literature. It is a triumph to have wholly omitted the Absolute. You have concentrated your attention on essentials, and have not omitted the psychological aspects of the matter (by the way on p. I5 1[2 s.f. (280: 21) you haven't noticed "pyschological"). And on the whole I don't complain of your criticisms. They are at any rate animated by a spirit of fairness to which we have not been accustomed in similar performances. I append seriatim a few reflections. P. 3 (263: 41-264: I); "confusion between acting on a hypothesis and believing it". It may be a wrong theory, but it is certainly not a confusion, to hold that action is a clue to belief. Belief is surely one of the most ambiguous and multiform of psychical facts, and there has hitherto been no adequate theory of it. Our clue certainly covers most of (the) facts. On p. 4 (264) init. you don't allow for shades of belief, or the growth of its intensity as verification accrues. I will admit that in forced options a choice may at first be random, but it doesn't follow that it remains unreasonable. You assume that "the will to believe" must be final ab initio and thus misrepresent the whole Jacobean psychology of religion, which rests on the experimental verification of beliefs adopted hypothetically. Moreover this is just what happens in science. You can't prove e.g. the conservation of energy unless you've assumed it first. Again your objection here doesn't seem consistent with that on p. ro (273: 6-20). Here you assume that when we "believe" we must commit ourselves finally, on p. IO you censure us for never getting beyond hypothetical assent. P. 4, 5 and I7, 18 (264: 7; 266: 23; 283: 4-5; 283: 36-9). Intolerance and coercion, etc. Surely quite an untenable charge as I have shown in my "Infallibility and Toleration" article, Hibbert Journal, October 1908. You implicitly admit this when you say p. 5 (266: I9-2l) that different people ought to adopt different beliefs. This is the genuine doctrine and the only provision I can see for assuring that the most adequate and satisfactory theory will ever be tried. It systematizes toleration whereas if you hold that only I can be right and every one thinks he is the one you directly invite a fight and the probable suppression of truth. As has constantly happened. As for what would happen if Roman Catholicism or Mohammedanism became pragmatist in its theory of belief, the answer is that (they) would have ipso facto to have to cease to be intolerant. Wherefore I argued it probably would not become pragmatist. Finally note that (p. 5 s.f. (266: 36)) ought to have in-

380

381

compatible beliefs does not = different. For Pragmatism "different" beliefs are no longer incompatible. P. 8 1[1 I. 15 (270: 13); "only" is unjust. As for the meaning of "good" (l. IO (261: IO)) I have not discussed it at length, but I have clearly indicated that it and bad refer to (furthering and thwarting) a purpose (which purpose also is conveniently called "good"). Hence the "good" of Logic is Truth, of Conduct Virtue, of Art Beauty, of Feeling Pleasure, and the ultimate problem of life is how to harmonize them in "The Good". Thus there are four species of "Good" in the genus "The Good" and this is all the quotation on p. I I (275: I-4) refers to. I wasn't thinking of what you think "the fundamental meaning of truth" (275: IO) at all. N.B. that "purpose" does not = "desire". P. 9 (271-2). Excellent and I wonder how long it will take "the orthodox" logicians to see this! 1[3. But it is true that to find the moon going in its calculated orbit does give the astronomer emotional satisfaction and that when it doesn't he is distressed. Therefore the pragmatic formula holds of this case also. You can say, if you (wish) that he should judge no other but this intrinsic satisfaction to be relevant, and qua astronomer perhaps it should not. But as he is also a man he has the problem (noted above) of correlating all his satisfactions, and in case of a conflict between his scientific and (say) his religious "satisfactions" can't abandon either. Pragmatism emphasizes this problem but doesn't decide it for each of us; only it provides a common denominator for the conflicting claims. N.B. I deny that "working theoretically" can ultimately be conceived without a reference to "practice". P. I I (275). Meaning of Truth . You don't say what you think it is, and it is part of our case that there is no other than the pragmatic. We have not made this point very prominent so far, because we have hoped that someone on the other side would come forward with a coherent account. But I am growing confident that this won't happen because it can't. Admitting then your distinction as to the two "meanings" in the abstract, we have in this case only one, viz. the genetic/causal meaning of truth. As no one can tell what truth means per se, we are entitled to say it means what it means in relation to action, when the illusion that it means something other has been uprooted from the mind. P. II, I2 (275:20-36). Ambiguity of Truth. I should say that the word "butter" was ambiguous de facto and in usage. If you ask for "butter" you may always get "margarine". The grocer has a double meaning of "butter" in one of which it includes margarine in the other not, and his interest is in confusing the two. But the illustration fails because ex hypothesi the grocer knows which is which. We never know when our "truth" is finally purified from "error". Hence we never get past "what is thought to be true" to "what is true". P. I2. 1[2 (276: 2-3); "the specific character of truth-predication" only means "the habit of distinguishing cognitive satisfactions by the epithet true". But that is not "a meaning of truth" or an alternative theory of truth.

383

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX I F .C.S. SCHILLER'S REPLIES

P. 12. ~2 (275: 37-276: 29). Suppose a belief persisted during the whole existence of intelligence in the world which was concluded by a cosmic catastrophe. Would there be any sense in denying that this belief was true? P. 12. 113 (276: 35). You attribute to us the assertion that (some? all?) "serviceable beliefs are true" whereas what we say is that "all 'true' beliefs are (in some sense) serviceable". The "exhortation" is not superfluous because so many have been prone to believe (at least half) that vitally necessary beliefs are false, and that noxious beliefs might be true. Hence devil-worships, etc. Anyhow, pragmatism will make a difference when new beliefs are propounded for our acceptance.-P. 14 l. l (probably 279: l). "For" for "Far". 111 (279: 16, 19). You've substituted "complete" for "absolute". P. 14. 112 (279: 26-33). You can state my doctrine in terms of the distinction of appearance and reality, but that is not to show that I must state it thus. In fact I don't for reasons alleged. "Facts" I can wish to alter I don't regard as ultimate and absolute, i.e. I reserve the right to try to alter them. But I need not therefore call them "apparent" or "illusory"; nay the fact that I call 'cm "facts" at all proves that I consider it better to recognize them. Pro tanto the pragmatic test does apply to them, though doubtless heaven would feel more "real". P. 14. ~3 (279: 34-40). Admitting that abstractly the constituents of the universe may be conceived as persisting in conflict, I note that empirically conflict brings with it so much suffering even to the most successful, that it is "reasonable" to aim at harmony. With a growth of intelligence therefore the elimination of conflicts may gradually be achieved. P. 15 (280: 3-22). Scepticism. Pragmatism is the only cure for Scepticism. It does not say "O is certain" but "though 0 is absolutely certain, everything must be certain enough to act on, and even 'sceptics' must acknowledge such certainty in practice". "Liable to revision" = "capable of improvement"; "rest securely" = "be lazy". These differences you may think are only in the emphasis, but they seem to me to make an enormous social difference. However the relations of pragmatism and scepticism is a topic I have marked out for treatment anon. P. 16 (281: 14-15). You have surely omitted a very relevant fact that my inquiry was not into whether "immortality" was a fact or even whether most people thought it was, but as to what was their state of sentiment. It had hitherto been taken for granted (uncritically) that of course all men must desire immortality and various sage arguments as to men's natural bias, etc., had been based on this assumed fact. It was therefore thoroughly scientific to try to get definite statistics on the matter, and I will add that a thorough statistical study of the existing nature of moral sentiments seems to me to be at present the chief desideratum of ethics. I am sure that the "facts" now moralized about are mainly traditional and suspect that they will be found to be largely unreal. P. 17 (282: 34). Ironclads, maxim guns, autos de fC, etc., have been the historic methods of propagating "Truth", to which the belief in its absoluteness has led. You can hardly render the nascent pragmatism responsible for them. Besides as

I've pointed out it really insists on toleration, free inquiry and experiment and as a condition of the plurality of truth which must precede even its ultimate unification. So far as I can see it is the only philosophy which can really believe in these things. I will finish with two more corrections. P. 18 (283:39) line 9 is wrong? "becomes" or "has become"? P. 2 (261: 18). The name Pragmatism, though invented, was not used by Peirce in 1877-78. It was first published by James in his California address. Also Dewey is no longer at Chicago but now at Columbia. Please consider yourself under no obligation to answer this lengthy screed and believe me-Ever sincerely yours, F.C.S. SCHILLER.

382

1.2

F. C. S. Schiller's Reply to Paper 24. Corpus Christi College Oxford

19 May 1912

DEAR RussELL,-1 was of course greatly interested to read your review of my Logic in this week's Nation, as I had looked forward to it eagerly. And in the main I am glad to say I can welcome it. You have been the first reviewer to discern that though the word pragmatism does not occur in the book, the criticism is systematic and proceeds on definite principles. (You might put it in a nutshell as "it is shown that throughout Formal Logic is caught in the dilemma 'either verbalism or psychology'.") But I must gently demur to your restricting the extension of "modern writers" to the symbolists, (who after all have not yet produced a systematic criticism of the old stuff such as Mercier, Sidgwick and I have attempted). Surely in point of fact most of the recent Logics are still of the old Formal type. Only in this week's Athenaeum, the reviewer (a grossly ignorant and stupid person of the Oxford type) much prefers a new two volume Irish Jesuit Logic, on which all the good Catholics in that country will be brought up for evermore, to Mercier's and mine, and that after giving himself utterly away by accusing Mercier of psychologism, and then "refuting" me by manifest but unconscious psychologism! And Muirhead in the Birmingham Post clinches his argument by urging that all this stuff has been taught for 2,000 years and therefore must continue to be! If you had read these reviews and the "Back to Aristotle" cry in full force in the British Medical Journal review of Mercier, and were familiar with "Logic as she is taught" all the world over (is Cambridge an exception, or is she saved by not teaching Logic?), I cannot think that you would speak so confidently of the prospect of effecting any alteration in the academic meaning of "Logic". To the charge that I am the victim of a "literary" education, I must of course plead guilty. I can only urge that I try to counteract the bias this causes, and have

384

385

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX I F .C .S. SCHILLER'S REPLIES

at any rate corrected it by the study of one philosophically very relevant science, viz. psychology. Also I have been accustomed to think in biological terms for a very long time, and have been a Darwinian for quite forty years. And my conception of Pragmatism is that it is essentially the philosophic generalization of scientific method as it is-and not as it has been misdescribed by philosophers. The objection that I do not state the case for Formal Logic better than its advocates, seems unduly severe: how could I singlehanded be expected to improve on the labours of eighty generations of formal logicians? But I have taken the case as it was put by the strongest and most renowned advocates and answered thatthough I have not mentioned names. And the errors I have attacked are hardly "superficial", the notion of valid inference, the independence of "form", the conception of self-evidence and cause, the abstraction from psychology and meaning, the reliance on verbalism, the laws of thought, the definition and delimitation of the "science". Regarding the syllogism you hardly represent me fairly. When I called the syllogism a great discovery made by sheer reflection I was saying nothing that the history of Logic has not proved. To what other single truth can you point which has lasted 2,000 years and still evokes so much enthusiasm? But a thing may be a great discovery without being completely right, and my criticism was not intended to show that the Syllogism is "utterly worthless", but how it must be understood and handled in order not to be either a tautology or a petitio, either meaningless or "invalid". You are mistaken also in supposing that I tried to prove that "the fundamental principles of logic ... are postulates"-! wasn't mentioning the principles of real logic at all (except incidentally and by way of contrast): they are not to be found in Formal Logics. I was investigating only whether, and if so how, a meaning could be attached to the principles of Formal Logic. Hence it was not necessary to explain the logical nature of postulates completely and systematically, the more so as I had already done so elsewhere. You must have forgotten this, if you think that I have changed my view or should ever have scrupled to agree that "a postulate which works is no longer a mere postulate" (of course that is why it becomes an "axiom"). The answer as to why we presume that the future will resemble the past is given on p. 295 (5): it is a methodological necessity, if we want to predict. As to the charge of "subjectivism" we shall probably never agree: I have often told you that it is not subjectivism but psychology, and that you can't abstract from the psychological side in all known objects, simply because there is a knowing process involved in getting to them. But that need no more render philosophy subjectivistic than astronomy becomes so by recognizing and evaluating the "personal equation". Likewise I venture to predict that you will never be able to conceive the difference between "truth" and "error" until you consent to consider the process of discriminating them. And this I say after reading the interesting passage in your Principia Mathematica p. 62 to which you so kindly referred me. I am afraid your restatements of the "self-evidence" doctrine don't lessen its difficulties, cf. a view

which involves the notions of the "nearly indubitable" and the "equally-plausibleknown" can hardly reach that "exactness" which I understand you to aim at. But I do not want to be led into a criticism even of the fundamentals of your magnum opus, and will conclude therefore by again thanking you for your review, the signing of which is all the more valuable because the Times, Spectator and the other papers controlled by the idealist ring are still persisting in their silly ostrich policy of ignoring all departures from what they treat as "orthodoxy" .-Ever sincerely yours, F.C.S. SCHILLER.

Appendix

II

Preface to Philosophical Essays [1910]

THE PREFACE WAS published for the first time in I9l0. The book was Russell's first collection of essays, and except for "On the Nature of Truth and Falsehood" (12) and Sections v and VI of "The Elements of Ethics" (19), all of them had been published previously in journals. When he heard of the death of William James, which occurred on 26 August I9IO, Russell must have written the publisher to ask whether, in his opinion, some at least of his more polemical remarks in "William James's Conception of Truth" required alteration. In I September C. J. Longman replied: I now return the article on William James' Conception of Truth, which we have received from the printers. On looking at it, it does not appear to me that there is much to which exception can be taken, and it may be on reconsidering the matter that you might see your way to make such alterations as would obviate the objection you feel. The matter, however, is one entirely for your decision, and if you prefer to omit it we are quite content. Russell annotated the letter: "[Concerns changes suggested by James's death]". The essay was not dropped from the book, nor was it altered. Instead, Russell added a postscript to his Preface, to explain to those readers unfamiliar with philosophical controversy that nothing personal was intended by his polemics. The copy-text is the printed version.

386

11 Preface to Philosophical Essays (1910). THE FOLLOWING ESSAYS, with the exception of the last, are reprints, with some alterations, of articles which have appeared in various periodicals. The first three essays are concerned with ethical subjects, while the last four are concerned with the nature of truth. I include among the ethical essays the one on "The Study of Mathematics", because this essay is concerned rather with the value of mathematics than with an attempt to state what mathematics is. Of the four essays which are concerned with Truth, two deal with Pragmatism, whose chief novelty is a new definition of "truth". One deals with the conception of truth advocated by those philosophers who are more or less affiliated to Hegel, while the last endeavours to set forth briefly, without technicalities, the view of truth which commends itself to the author. All the essays, with the possible exception of the one on "The Monistic Theory of Truth", are designed to appeal to those who take an interest in philosophical questions without having had a professional training in philosophy. I have to thank the editor of The New Quarterly for permission to reprint "The Study of Mathematics" and Sections I, II, III, v and VI of the essay on "The Elements of Ethics", and for Section IV I have to thank the editor of the Hibbert Journal. My acknowledgments are also due to the editors of The Independent Review, The Edinburgh Review, The Albany Review, and the Proceedings of the Aristotelian Society, for permission to reprint the essays II, IV, v and VI respectively. In the sixth essay as originally printed, there was a third section, which is now replaced by the seventh essay. OXFORD July I9IO Postscript.-The death of William James, which occurred when the printing of this book was already far advanced, makes me wish to express, what in the course of controversial writings does not adequately appear, the profound respect and personal esteem which I felt for him, as did all who knew him, and my deep sense of the public and private loss occasioned by his death. For readers trained in philosophy, no such assurance was required; but for those unaccustomed to the tone of a subject in which agreement is necessarily rarer than esteem, it seemed desirable to record what to others would be a matter of course. October I9l0

387

Appendix III

F. H. Bradley's Criticism of Russell and His Reply to Russell [1910-II]

THE FIRST OF these two critical pieces was published in Mind, n.s. 19 (April l9IO): 153-85; the second appeared in Mind, n.s. 20 (Jan. l9II): 74-6. Russell's reply to the first of them is reprinted in this volume as Paper 31. The Headnote to Paper 31 provides an extended account of the philosophical interaction of Bradley and Russell.

388

III. l Excerpt from F. H. Bradley's "On Appearance, Error and

Contradiction". Mind, 19 (April 1910): 153-85. I HAVE NOW to remark on some of the fundamental ideas used by Mr. Russell, and must endeavour to show that these ideas contain inconsistency. It is a task to which in one sense I am quite unequal. I am incompetent utterly to sit in judgment on Mr. Russell's great work (Principles of Mathematics). But, if the mathematical part is as good as the part which is philosophical, I am sure that he has produced a book of singular merit. To confine myself here to a one-sided criticism of ideas which I can only partially comprehend, is ungrateful to me, and I could not do it if I did not feel myself in a sense compelled to say something. I understand Mr. Russell to hold that mathematical truth is true perfectly and in the end, since the principles as well as the inferences are wholly valid. The fundamental ideas, I understand, are throughout self-consistent. If there were an exception the extent of its influence would raise a question at once of the most formidable kind, and the main doctrine obviously would be imperilled. But this is a point on which, through my own incapacity, I have been unable to appreciate Mr. Russell's decision. I must therefore, passing this by, go on to inquire as to the consistency of some leading ideas. I encounter at the outset a great difficulty. Mr. Russell's main position has remained to myself incomprehensible. On the one side I am led to think that he defends a strict pluralism, for which nothing is admissible beyond simple terms and external relations. On the other side Mr. Russell seems to assert emphatically, and to use throughout, ideas which such a pluralism surely must repudiate. He throughout stands upon unities which are complex and which cannot be analysed into terms and relations. These two positions to my mind are irreconcilable, since the second, as I understand it, contradicts the first flatly. If there are such unities, and, still more, if such unities are fundamental, then pluralism surely is in principle abandoned as false. Mr. Russell, I cannot doubt, is prepared here with an answer, but I have been unable to discover in what this answer consists. To urge that these unities are indefinable would to myself be merely irrelevant. If they had no meaning they could serve no purpose, and the question is with regard to their meaning. If that is not consistent with itself or with Mr. Russell's main doctrine, then that meaning is not admissible as true, unless it is taken subject to an unknown condition. But, if so taken, that meaning, I would urge, is not ultimate truth. For a certain purpose, obviously, one can swallow whole what one is unable to analyse; but I cannot see how, with this, we have rid ourselves of the question as to ultimate truth. On my own position here I need npt dwell. For me immediate experience gives us a unity and unities of one and many, which unities are not completely analysable or intelligible, and which unities are self-contradictory unless you take them as subject to an unknown condition. Such a form of unity seems to me to be in principle the refutation of pluralism, and on the other side it more or less vitiates 389

390

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

the absolute claim of all truths (I cannot stop here to make the required qualification) including those of mathematics. Now what is Mr. Russell's attitude towards a position of this kind? On the one hand I understand him to reject it most decidedly. On the other hand, wherever anything like "implication" or "unity" is involved (and how much have we left where these are excluded?), Mr. Russell seems to myself to embrace a conclusion which in principle I find it hard to distinguish from my own. And, it being clear to me that there is something here which I have failed to comprehend, I must leave this fundamental issue and go on to consider some difficulties more in detail. The notion of "implication,"' I understand Mr. Russell to say, is necessary for mathematics; and let us consider very briefly what this notion involves. It seems to mean (if it means anything) that something is both itself and more than itself. There is a difference here which is both affirmed and denied; for of course that anything should imply merely itself is meaningless. But how can anything be at once itself and in any sense not-itself? Mr. Russell leaves us here, so far as I have seen, without any assistance. But with this we are face to face with the familiar problem of the one and many, the universal and particular. We are driven back to the immediate experience where the whole is in the parts and where, through the whole, the parts are in one another. But such an immediate experience seems in the first place (I would repeat) to contradict pluralism, and in the second place it offers by itself no theoretical solution. The same difficulty appears in "such that". If this phrase does not mean that a particular is also a universal, and with a certain consequence, it surely has no meaning at all. But how to justify this necessary inconsistency Mr. Russell does not tell us. Among other fundamental troubles of the same kind I would mention the ideas of "occupation" and of "magnitude of". Certainly Mr. Russell asserts here the existence of a relation, but this assertion to my mind seems obviously opposed to fact, and once more I find an unjustified recourse to the inconsistency of immediate experience. I will enter now on some instances of a somewhat different kind, where however the difficulty remains at bottom the same. I will not repeat what in a former article I have urged with regard to the word "And" (Mind, no. 72, p. 497, note). Its relevancy and its importance in this connexion however are obvious. But, leaving this, I will touch briefly on the subject of relation and identity. Mr. Russell, I understand, defends and builds on such an idea as the relation of a term to itself. This idea to my mind is unmeaning or else self-contradictory. To my mind a relation must imply terms, and terms which are distinct and therefore different from one another; and our only ground for thinking otherwise in any case is our failure to apprehend the diversity which has really been introduced. Mr. Russell in particular uses and justifies the abstract identity of a term with itself. He does not, I In connexion with "implication" the axioms given by Mr. Russell (p. 16) demand the

attention of logicians. But want of space makes it impossible for me to offer here any criticism.

APPENDIX III BRADLEY'S CRITICISM OF RUSSELL

391

I think, say the same thing here with regard to difference. But, if difference is a relation (and, if it is not a relation, its nature seems puzzling), and, if again all relations are external,-then the difference of a term from itself seems as justifiable as its identity with itself. For, ex hyp., it is all one to the term what its relations are. But, however that may be, Mr. Russell defends identity between a term and itself. And this idea surely contradicts itself, since (to repeat this) diversity is required for relation, and Mr. Russell would not admit that the idea can be at once the same with itself and different from itself. He attempts to justify his doctrine here by producing a number of examples (p. 96). But I can see no meaning in any one of these unless diversity is introduced, and I will lower down say something more with regard to one instance. I will proceed now to remark more in detail on the inconsistency of such an idea as "class". We have here no fresh difficulty in principle, any more than if we examined, for example, such a word as "instance". It is still the old problem of the universal, and of the one in the many, and the dilemmas which everywhere arise change their particular shape but not their radical essence. Mr. Russell however has attached great importance to the problem raised specially by the word "class". I regret that my incapacity for following abstract arguments has prevented me in great part from understanding the position which he has here taken up. But I will venture briefly to exhibit some of the puzzles and inconsistencies from which I cannot find that he delivers us. I will first remark that no class can be related merely to itself. We have seen above that everywhere relation without diversity is meaningless. In the next place no class can consist only of one member. Such an idea is a fiction which contradicts itself. It ceases to do this only when you introduce plurality in the form of possible members. Where these are excluded, as in the idea of the Universe, you can no longer speak of a class. The Universe obviously is no class nor any member of a class of Universes. And in any case, with the introduction of possibility into the idea of class, difficulties would arise, which, as I understand it, on Mr. Russell's view would be fatal. The idea of possibility, I may perhaps add, seems to call for an attention on his part which it appears hardly to have received. The account on page 476 seems scarcely adequate, and the idea, I submit, must be dealt with in any satisfactory account of Continuity and Infinity. After this necessary preface I will set out briefly the inherent inconsistency of "class". (a) The class is many. It is its members. There is no entity external to and other than the members. The class is a collection. And it is not a mere possible collection, nor is it a collection of mere possibles. Either of these alternatives would ruin the idea of class, as could be shown, if required. The class is an actual collection of actuals. But it is a collection which is not collected by itself (that idea would seem meaningless), nor is it again collected by anything from the outsidefor, if so, it would have to contain this other agency. It is a collection, since it is taken together; but it is a collection collected by nothing-an idea which seems either senseless or self-contradictory.

392

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX III BRADLEY'S CRITICISM OF RUSSELL

(b) The class is One, but the One is not something else outside the members. The members even seem to be members because of what each is internally. And this apparent quality in each cannot be a relation to something outside the class. The One clearly is something within the members. If there are two qualities they must be taken in one, or else we have forthwith two classes. And (to return to the idea of a collection) two collections, differing only as collections and not differing at all in their contents, seem certainly not the idea which we seek in a class. On the other hand a quality merely internal to each member seems to leave the class without any unity at all. The unity therefore, not being external, must be taken itself as a member of the class. And, since this once more seems senseless, the class appears to be dissolved.

each of them is so different from "being" that our assertion "is is" may be significant. And then I might go on to urge, of "what" and "that," that each is included in the class of the other, and that each is a part of the other and so perhaps even of itself. And in short I might develop all those monstrous results which follow when an inconsistent idea like "class" is taken as true, not for a limited purpose, but absolutely. I will end by some remarks on the subject of negation. It seems to me that negation is a topic which, on a general view like Mr. Russell's, causes difficulty, and calls for more notice than (so far as I can find) it has received. Mr. Russell's doctrine of zero to myself appears to be philosophically untenable; and in various other ideas negation is present in a way which seems to me to call for explanation. I will take the last point first in connexion with such ideas as "a" and "any". (i.) "A man" appears to assert one instance of man and to deny more than one man. (ii.) "Any man" seems to affirm that there is a man, and to assert also the existence of other men actual or possible. 2 It denies, with regard to these others, any difference-in a certain respect. "Any" therefore contains negation in its essence in the form of "it does not matter who or what". (iii.) "Every man" and "all men" (I will not here discuss the difference between these) contain the denial of "man" outside of certain limits; while (iv.) "some man or men" means a man or several men, together with a negation as to my further knowledge. It conveys that "I know, or need know, no more about it than that". Now I do not suggest that the negation in these terms is a matter with which Mr. Russell is not perfectly familiar. I am urging merely that I do not understand the place which in his general system of ideas negation is to occupy. To come now to the account of zero, this idea, unless I have failed to understand it, seems to contain an open self-contradiction. It would seem that "no pleasure" has the same relation to pleasure as the various magnitudes of pleasure have, though it has also, of course, the special relation of negation (p. 186). The "also" here to my mind involves a self-contradiction. To my mind "no pleasure" excludes pleasure, and by consequence the required relation; and how this consequence is avoided by Mr. Russell I have been unable to see. On the alleged positive relation I have already remarked, and the difficulties attaching themselves to Mr. Russell's idea of a kind of magnitude to myself seem insuperable. Every magnitude has "a certain specific relation to the something of which it is the magnitude. This relation is very peculiar, and appears to be incapable of further definition." I must repeat with regard to this relation that to my mind it is a sheer fiction, as is also the relation alleged to exist in "occupation". The fact is a complex not consisting of or reducible to terms in relation. But, however that may be, the proposal to unite this relation by an "also" to the relation of negation I can only understand as a demand to bring together simply two elements which exclude each other. And with regard to "indefinable," what troubles me is not that I insist on defining

To save ourselves from ruin we may construct a new class which is wider, and which includes within itself, as members, both the members of the old class and their unity. But since the principle of inconsistency is left, any such expedient is useless. We are forced once more to dissolve our class and to seek refuge in a still wider class. And, when we have reached our widest class of all, our bankruptcy is visibly exposed. We are then compelled openly to make the class as one a single member of itself as many. And with this we end in what is meaningless or else plainly is in contradiction with itself. The discussion of these inconsistencies (the reader is perhaps aware) might be pursued almost ad libitum. Since the class cannot fall outside the several members, each member by itself will be the class, and will even be the whole class. And from this will follow results which are obviously ruinous. For instance, the member itself will become many, and will be internally dissipated. But the reader, if so inclined, can develop these consequences for himself, as well as the puzzles which arise in connexion with the ideas of "a collection" and of what is "actual" and "possible". I have, I hope, said enough to show that the idea of class is inconsistent ultimately, and that every region, where it is employed, must be more or less infected with self-contradiction. How Mr. Russell would avoid this conclusion I regret to say I have been unable to understand. He apparently defends the idea of a class being a member of itselfan idea which to myself contains a glaring self-contradiction. And, as we have seen, he advocates the doctrine that a term can be related to itself-a view which for the same reason I am forced to reject. In every instance adduced, such, for example, as "Predicability is predicable," I find (I would repeat) a distinction and difference, or else I find nothing. The reader will permit me perhaps to illustrate and explain this statement by the instance of "being". I do not reject as meaningless such a judgment as "being is" or "is is". I only insist that, in order to have a meaning, I must introduce distinction and diversity. I might, for instance, mean by such an assertion that only or merely being is and that anything else must be denied. I might wish to convey that after all, or whatever else it is, being still is. I might in the end mean that in "being" itself is the distinction and diversity of "what" and "that", and might imply that either of these thus "is," and yet that

2

393

"Any" tends to drift away from this assertion, but so tends to drift away from itself.

394

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX III BRADLEY'S CRITICISM OF RUSSELL

everything. What troubles me is that, if an indefinable is meaningless, to me it is nothing, and that here the meaning which I must give to zero (if I am not to leave it meaningless) seems inconsistent with itself. It is intolerable to my mind to speak of "no pleasure" as being a decreased lot of pleasure, or, when pleasure is once more added, to speak of pleasure as being increased. On the other hand, since to me there is no such thing as bare nothing, and since all negation rests on a positive basis, you can rightly speak of diminution when you descend from pleasure to no pleasure, and, when you pass the other way, you can rightly speak of increase. But what is this positive something which has here become less or more, and has become less or more by pleasure? To call this something "pleasure," even where pleasure is specifically excluded, surely involves self-contradiction. And the same remark applies to any attempt to begin with less than something, and to increase this until it becomes something, or to descend by degrees of diminution from something to nothing. If such ideas are useful, then of course they must be used, but in the end they do not hold together. But I hasten to add that I think it probable that on the subject of zero I have wholly failed to understand Mr. Russell. These pages have been written, I would repeat, with great reluctance and with a sense of compulsion. I have felt myself coming forward, or rather driven, to speak on matters where on one side I am quite ignorant, and where this ignorance is only too likely to have led me into fatal error. And I have criticised a writer whose work as a whole I am unable to appreciate, and in connexion with whom I can say nothing on some of those merits which I am sure are very great, but which are really beyond me. And, even where mere metaphysics or mere logic is concerned, I have had to confine myself here to dissent. I regret this, for I do not think, amongst those present writers on philosophy whom I know, there is any one who, as compared with Mr. Russell, calls for more or even for as much attention. For any student of first principles that attention seems to me to be not merely advisable but imperative. The problem of the general nature of order and series has been too much neglected, and yet surely it is a problem which seems infinitely promising. Not only has this inquiry been brought to the front by Mr. Russell, but he has, at the lowest estimate, supplied matter for its solution which no one can neglect. And to have done this by itself, even if he had done nothing beyond, is to have helped our philosophy in a way which, I hope and believe, will become more and more manifest.

to explain somewhat further; for in the main I am left still unable to understand. If, however, Mr. Russell should feel that within convenient limits there is no more to be done, such a position, so far as I am concerned, would call for no justification. 1. In the first place, my difficulty as to "unities" remains. Is there anything, I ask, in a unity beside its "constituents", i.e. the terms and the relation, and, if there is anything more, in what does this "more" consist? Mr. Russell tells us that we have got merely an enumeration or merely an aggregate. Even with merely so much I should still have to ask how even so much is possible. But, since we seem to have something beyond either, the puzzle grows worse. If I remember right, Prof. Stout some years ago stated the problem as attaching essentially to the fact of "relatedness". What is the difference between a relation which relates in fact and one which does not so relate? And if we accept a strict pluralism, where, I urge, have we any room for this difference? 2. In the next place, as to "implication" my troubles continue. If we have nothing but facts, I see no room for implication, and if we have anything more or less than facts, I cannot understand what this is. By all means banish possibility as real, but where among facts does implication fall? Is a disjunction with its "Either-or" an actual fact? Are "conditions" facts? Is "deducibility" a fact? With regard to facts I thought our attitude was one of "It is" or (perhaps also) "It is not". I do not in the least understand the position of "either-or" or of "can be" or "may be". 3. I urged against the possibility of a term being related to itself the fact that relation implies diversity, and I should like to explain my reason for holding to this fact. I do not proceed here by arguing downwards from some assumption or axiom. I proceed on the contrary by way of actual experiment. With any relation remove diversity (this is my experience), and the relation is destroyed. You have (I find) no relation left unless you also leave that diversity which you may have failed to notice. What I of course am forced to assume here is that I have correctly performed my experiment. If Mr. Russell on the other side says that he can perceive a relation where there is absolutely no diversity about the terms, I do not see how we are to argue about our difference. 4. With regard to diversity, externality and mere fact, the assumptions (I do not call them such) which I make are as follows. I assume first that, where I get the unmeaning or the self-destructive, I have not got even the possible. And I assume that what is is, in the sense that, so far as I have truth and reality, I have not got something which is true and real merely because of something else. This second assumption, if it is to be called one, bears on the question of externality and mere fact in a way which I will explain. (a) But, first, with regard to diversity Mr. Russell maintains, as I understand, that our only reason for denying the relation of diversity between a term and its own self is that this relation is not a fact. Whether Mr. Russell means more than that the relation has not yet been found, I am unable to judge. To myself on the

m.2 "Reply to Mr. Russell's Explanations". Mind, 20 (Jan. 1911):

74-6. THE EXPLANATIONS OFFERED by Mr. Russell in the July number of Mind have been read, I am sure, with interest by many readers. I unfortunately did not see the number at the proper time, but still I hope it is not too late to ask Mr. Russell

395

396

397

VOL . 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX III BRADLEY'S CRITICISM OF RUSSELL

other hand the above relation is not possible. To myself it either is meaningless or self-destructive. In making an ideal experiment I either have no diversity, or else the terms are different; and, when I suppress the difference, the relation is destroyed. I therefore deny this possibility, and I go on further to argue that any premisses from which such a possibility follows are false. (b) With regard to externality and mere fact I should first explain that, in my opinion, these are things which are not and which can not be observed. To have bare A in bare external relation to B is not possible in any observation or experiment. The supposed fact is really an inference reached by vicious abstraction. We saw above how "unities" and "implications," without which Mr. Russell apparently cannot move a step, involve always a something more which on his view seems inexplicable. And the same thing holds good with regard to any alleged perception of mere conjunction. To myself the mere fact in which something seems to qualify A from the outside, is never really the whole fact. There is always here a condition left outside of what you take as the fact. Your statement is therefore true not of A itself but of A qualified by x. And hence the opposite of your statement is also true. On the other hand to say something about A which in no sense qualifies A, remains to my mind meaningless. In other words, no "and" which is purely external is thinkable. This is once more the point to which Mr. Russell is invited to address himself. The above is the ground of objection to externality and to mere fact. You want, that is, to say something about something, and not about something else, particularly when the something else is unknown. The demand for "intrinsic" relations I take to be an expression of this want, but I agree that here once more complete satisfaction is impossible. There is of course with me no question of any "axiom". Naturally I realise that in this way doubt may be thrown upon every possible conclusion, however certainly it seems to follow in ideal experiment. How are we anywhere to save ourselves from doubt arising from the presence of the possibility of an unknown condition? Have we not with every result a counter-possibility? This question in its turn leads to the inquiry whether the alleged counter-possibility is everywhere really possible. But I must not here digress into a defence of what I have argued elsewhere. 5. I have stated the main principle on which objection is taken to absolute externality and bare conjunction. I would go on to add that I am still in doubt as to the sense in which according to Mr. Russell relations are external. The terms are to contribute nothing, and so much I understand. But I still do not know whether Mr. Russell takes the relations apart from any terms to be thinkable. To be consistent he should, in my opinion, hold this view, but I cannot say that he does so. If all that is meant is that this or that term contributes no more than any other term, clearly, from so much, absolute externality and pluralism do not follow. On the other hand, a relation apart from terms is to me unmeaning or self-destructive, and is an idea produced by an indefensible abstraction.

6. I will end by noticing briefly Mr. Russell's contention that on his view we are less in conflict with science and with common sense. This is an argument which I am very far from undervaluing. In fact the doctrine which I hold I hold largely because it seems to me to remain, more than others, in harmony with life as a whole. I am speaking of course only of views which aim at theoretical consistency, and not of those where inconsistency and self-contradiction are of minor importance. But I could not on this ground compare the conclusions advocated by myself with those taught by Mr. Russell, because on the most important point I do not know what his conclusion is. To myself the things which matter most in life arc not to be resolved into terms with relations between them. And I am ignorant as to what on this point Mr. Russell may really hold. The question is in a word as to experiences which, to a greater or less extent, arc non-relational. Obviously, when I do not know whether and how far Mr. Russell denies the existence of such facts, or in what sense he admits them, it is not in my power to judge as to how far his views are in harmony with science and common sense, if I use these terms, that is, in anything like a wide meaning. This is a point on which some explanation by Mr. Russell would be welcome, I am sure, to others as well as to myself. We return here to the doubt as to "unity" with which we began. We have again on our hands the whole question as to sensible fact and as to all that is covered by the word feeling. I should perhaps add that, so far as I can judge, Mr. Russell's view as to the inviolability of "facts" would make indefensible the constructions in and by which the entire body of history and of natural science consists.

Appendix

IV

Sur les axiomes de l'infini et du transfini [191 l]

PAPER, an English translation of which is printed as Paper 3, was published in Societe mathematique de France, Comptes rendues des seances, Paris, no. 2 (r9II): 22-35. THIS

IV

"Surles axiomes de l'infini et du transfini". Bulletin de la Societe Mathimatique de France, 39 (1911): 488-50I.

LA MATHEMATIQUE PURE, on le sait, peut s'cxprimer et se demontrer entierement en termes d'idees ct d'axiomes de la logique. Cette these est evidente pour quiconque considerc la nature de la deduction. Quand on deduit, on dit: x possCde telle propriete, done x possCde tellc autre propriete. Or, si l'on pent savoir ceci, il est certain que lcs deux prop:-ietes doivent avoir un lien formel, qui permct de transformer les proprietes en variables, et d'affirmcr: dcux proprietes quclconques ayant tel ou tel lien formel sont tclles que l'unc d'ellcs implique l'autrc. On a la unc proposition de logiquc. En effet, toute proposition mathematique devient une proposition de logique en transformant en variables un nombre suffisant des constantes que peut contcnir la proposition. Je donnerai un excmplc geometrique de cc procede. La Geometric des espaces infinis (qui comprend ccllc d'Euclide) pcut sc developper commc theorie de la relation entre parmi lcs points. On definira lcs points: tout tcrmc y qui est entre deux termes x ct z. Mcttons {J pour entre. On definira la ligne (x, z): les tcrmes y tels que yt>(x, z) ou xlJ(y, z) ou zt>(x, y) et x et z. On aura des axiomes, tels que yt>(x, z). :::>: - x1'1(y, z), yi'J(x, z). z1'1(y, w). :::>. z!J(x, w), .... On pcut alors considerer une relation i'J quelconque ayant ces proprietes, et l'on a de la logique. Cependant on n'obtient la reduction a la logique qu'en cessant de demander avec trop d'insistance s'il existe des objcts qui verifient lcs hypotheses dont on considere les consequences. Il arrivera parfois qu'on pourra construire de tels objets a priori; par excmple, on peut construire a priori une classe ayant un nombre fini quelconque de termes. (A vrai dire, cette construction n'est a priori qu'en admettant comme a priori l'axiome qu'il existe au moins un objct, ou quelque equivalent.) Mais la plupart des thforemes d'existence (qui, du restc, ne sont pas necessaires pour la verite des autrcs theoremes, mais seulement pour leur importance) ont bcsoin de donnees qui ne sont pas purement logiqucs. Dans la Mathematiquc pure, deux axiomcs d'existence donnent a pcu pres tous les thforemes d'cxistence qu'on pent desirer. Ces deux axiomes sont: 1° 2°

L'axiome de l'infini; L'axiome multiplicatif, autremcnt dit, l'axiome de Zermelo.

Ces deux axiomes nc pcuvent se demontrer par la logique, et, a mon avis, ils n'ont pas d'evidence intuitive. Cependant je voudrais d'abord expliqucr lcur nature ct leurs consequences, avant d'cxaminer les raisons qu'on pourrait avoir pour admettre OU nicr lcur verite. 398

399

400

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX IV AXIOMES DE L'INFINI ET DU TRANSFINI

J;axiome de l'infini s'enonce comme suit: Si n est un nombre cardinal fini quelconque, il ya des ensembles consistant en n individus. lei le mot individu s'oppose a classe, fonction, proposition, etc.; en d'autres termes, individu signifie etre du monde actuel, par opposition aux etres de la logique. Pour amplifier cette notion, ii faudrait expliquer la theorie des types logiques, ce que je ne desire pas faire; Jes lacunes qui resultent de cc silence seront sans doute visiblcs a mes auditeurs. Appelons i,, la classe de tous Jes ensembles de n individus. Alors, etant donne l'axiome de l'infini, la suite

dinaux. Cette derniere est X0 , X1' .. ., Xm .... On peut demontrer !'existence de tous ces nombres, en admettant l'axiome de l'infini. Mais, a mon avis, on ne peut pas demontrer, sans un nouvel axiome, !'existence de Xw, ou d'un nombre cardinal quelconque plus grand que tous Jes nombres X0 , X., .. ., Xn, .... En admettant l'axiome de l'infini, nous avons done demontre !'existence de deux progressions de nombres cardinaux, a savoir

forme une progression, c'est-a-dire une suite dont le nombre ordinal est w (selon le langage de Cantor). On trouve ainsi que l'axiome de l'infini est la condition neccssaire et suffisante pour !'existence des progressions, c'cst-a-dire pour !'existence des suites de la forme x 1, x,, .. ., x,,, .. ., ad inf. Cct axiome est done la condition necessaire et suffisante pour !'existence du plus petit nombre ordinal transfini w, et du plus petit nombre cardinal transfini l'\ 0 • Etant donnee !'existence de l'\0 , on aura aussi !'existence de 2'''o, puisque ce nombre est le nombre des classes contenues dans une classe ayant l'\ 0 termes. On \'

n, n+ l, n+2, 2n,

"' 2No, 22"~ ' ... ' ''o'

Si !'on n'admet pas l'axiome multiplicatif, ii n'existe aucune demonstration que Jes termes de la premiere progression sont ou plus grands ou plus petits que Jes termes de la seconde progression (a part le terme X0 ). Cantor a espere pouvoir demontrer que 2No = X1' mais ni Jui ni aucun autre n'a reussi a demontrer une telle equation. II y a un autre ordre d'idees qui, en partant de l'axiome de l'infini, demontre !'existence de suites Cls) n Ox.::). Rl'P

Df.

Si l'on a R EP'6 x, R ressemblera a la relation dans le diagramme.

(~" ~' (~~~R-~y,

404

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

APPENDIX IV AXIOMES DE L'INFINI ET DU TRANSFINI

La notion generale est utile, mais pour le moment nous nous occuperons principalement de E'Ll.X· On remarquera d'abord que la classe des selections ne pourra definir le produit arithmetique que quand les membres de x s'excluent l'un l'autre, tandis que E'Ll.X n'est pas sujet a cette restriction. J'ecris Ne' a pour le nombre cardinal de a, ou a est une classe (un ensemble); j'ecris IINc'x pour le produit des nombres des membres de x, ou x est une classe de classes. On peut done mettre

ii ne s'augmente pas par l'additio.n de l'unite. Ou bien on peut dire qu'on appellera nombre entier fini tout nombre qui obeit a !'induction complete a partir de zero. C'est-a-dire: appelons propriete recurrente toute propriete qui appartient an+ 1 du moment qu'elle appartient a n, et appelons propriete inductive toute propriete recurrente qui appartient a zero. Nous appellerons alors nombre inductif tout nombre qui possede toutes les proprietes inductives, c'est-a-dire tout nombre pour lequel Jes demonstrations par le moyen de !'induction complete sont valables. 11 est facile de voir que Jes nombres inductifs sont les memes que Jes nombres naturels 0, 1, 2, ... , 100, ... , 1000, ... ,et qu'il ya des nombres (en admettant l'axiome de l'infini) qui ne sont pas des nombres inductifs, par exemple le nombre des nombres inductifs. Le soi-disant principe de !'induction complete devient done une definition, a savoir la definition des nombres inductifs. On peut alors dire que Jes nombres infinis sont Jes nombres non inductifs. Cantor suppose toujours que Jes nombres non inductifs sont Jes memes que Jes nombres qui ne s'augmentent pas par !'addition de !'unite; mais, pour demontrer ceci, on a besoin de l'axiome multiplicatif, non pas, ii est vrai, dans toute sa generalite, mais seulement pour Jes produits de Xo facteurs. 3° On definit le produit de deux facteurs de la maniere suivante: soient a et f3 deux classes quelconques; alors le produit de a et f3 sera la classe des couples (x, y) dont x Ea ety £ f3 (en distinguant x, y de y, x). On demontre alors sans l'axiome multiplicatif la plupart des proprietes connues du produit de deux facteurs. Cependant, si l'on veut etablir le rapport entre !'addition et la multiplication, on a besoin de l'axiome; c'est-a-dire, on en a besoin pour prouver que la somme de µ, classes, dont chacune a v termes, possede µ, x v termes. 11 en est de meme pour le rapport entre la multiplication et !'exponentiation. 4° De ce que nous venons de dire ii s'ensuit qu'on ne peut prouver, sans l'axiome multiplicatif, que la somme de X0 classes ayant chacune X0 membres possede X0 x X0 membres. On sait que X0 x X0 = X0 , et ii est usuel d'en deduire que la somme de X0 classes ayant chacune X0 membres a X0 membres. Cependant cette deduction n'est bonne que si !'on admet l'axiome multiplicatif. 5° Par le moyen de la deduction dont nous venons de parler, on a coutume de prouver que la limite d'une progression de nombres ordinaux de la deuxieme espece (c'est-a-dire formee par des suites dont le nombre cardinal est X0 ) est ellememe de la deuxieme espece. Ceci aussi n'est valable que si !'on admet l'axiome multiplicatif; done une tres grande 'partie de la theorie des nombres ordinaux transfinis devient douteuse. II en est de meme du theoreme que, dans une suite quelconque, un terme ne peut etre a'Ia fois la limite d'une suite du type wet d'une suite du type wl" II s'ensuit que presque toute la belle oeuvre de Hausdorff, Untersuchungen uber Ordnungstypen, depend de l'axiome multiplicatif. 6° L'axiome multiplicatif est necessaire pour demontrer que la somme de µ, ensembles ayant chacun v termes a le meme nombre de termes que la somme de v ensembles ayant chacun µ, termes, excepte dans le cas ou µ, et v sont tous deux finies. Pour prendre le cas le plus simple, supposons une progression de couples

rINc'x

=

Nc'E'Ll.X Df.

Alors quand X est une classe finie, meme si Jes membres de x sont des classes infinies, HNc'x aura Jes proprietes connues des produits arithmetiques. Mais six est une classe infinie, on ne sait pas que la classe E'Ll.X n'est pas nulle, si !'on n'admet pas l'axiome multiplicatif. II pourra done advenir qu'un produit arithmetique soit zero quand aucun de ses facteurs n'est zero. Si !'on admet l'axiome multiplicatif, ceci devient impossible; en effet, la proposition: un produit cardinal ne peut etre zero que si un de ses facteurs est zero, est equivalente a l'axiome multiplicatif. L'axiome multiplicatif s'enonce comme suit: Soit x un ensemble d'ensembles non nuls; alors il y a au moins une classe µ, qui possede un terme, et un seul, dans chaque ensemble qui est membre de X· L'axiome de Zermelo s'enonce comme suit: Soit a une classe quelconque, et soit X la classe de toutes les classes non nulles contenues dans a; alors il y a au moins une relation selective de X· En symboles, (a). 1:1! E'Ll.Cl ex' a. II est facile de prouver que ces deux axiomes sont equivalents.' Done, d'apres le theoreme de Zermelo, ils sont tous deux equivalents a la proposition: Toute classe peut etre bien ordonnee. Ils sont aussi equivalents a la proposition que, etant donne un ensemble x d'ensembles non nuls, ii existe toujours des relations se!ectives de x; et a la proposition que, si p est une relation quelconque, et X une classe contenue dans le domaine converse de P, alors P'ax n'est pas nu!. Les propositions qu'on ne peut demontrer que par le moyen de l'axiome multiplicatif sont tres nombreuses. En voici quelques-unes: 1° De deux nombres cardinaux differents, l'un doit etre plus grand que l'autre. 2° Les nombres cardinaux qui s'augmentent par !'addition de l'unite sont les memes que !es nombres inductifs, c'est-a-dire Jes nombres qu'on appelle naturels. Cette proposition identifie Jes deux definitions de l'infini. On peut dire qu'une classe est infinie quand elle contient une partie qu'on peut mettre en relation biunivoque avec la classe entiere; d'apres cette definition, un nombre est infini quand I

Voir Principia Mathematica, §88.

405

406

VOL. 6 LOGICAL AND PHILOSOPHICAL PAPERS, 1909-13

et supposons qu'on ait a demontrer qu'on peut arranger les memes termes clans un couple de progressions. Avec la notation que nous venons d'employer, ceci se fait en prenant la suite

Mais en adoptant la notation xv, Yv pour le vieme couple, nous avons introduit l'hypothese qu'un ordre est donne pour chaque couple, de sorte qu'on peut distinguer un premier et un deuxieme membre du couple. Si l'on ne fait pas cette hypothese, il sera impossible de donner une regle d'apres laquelle on choisira simultanement un terme clans chaque couple. Done on ne saura ranger les termes en deux progressions. Ce que nous venons de dire deviendra plus evident en prenant un exemple. 11 y avait une fois un millionnaire qui possedait X0 paires de bottines. Peut-on demontrer que le nombre des bottines qu'il possedait erait un nombre pair? Oui, car on pourra mettre clans une classe toutes les bottines gauches, et clans l'autre toutes les bottines droites. Mais si ce millionnaire avait l'excentricite d'avoir Jes deux bottines semblables, de sorte qu'il n'y avait pas une bottine droite et une bottine gauche clans chaque paire, alors ii devient impossible d'effectuer la division de l'ensemble des bottines en deux parties egales. 11 est done impossible de montrer que le nombre de ses bottines erait un nombre pair, ou qu'il y avait X0 bottines, malgre le fait qu'on a

Les mathematiciens ont employe l'axiome multiplicatif inconsciemment, non pas seulement clans la theorie des ensembles, mais aussi clans la Mathematique ordinaire, jusqu'a ce que Zermelo l'ait enonce explicitement comme axiome. 2 Moimeme je l'ai employe autrefois sans le savoir; mais, en 1904, j'ai fait la decouverte qu'il y avait la un axiome independant. Beaucoup de mathematiciens, comme Zermelo lui-meme, affirment que cet axiome est aussi evident que Jes autres, et qu'on peut l'affirmer sans hesitation. D'autres disent qu'il n'y a aucune raison de croire que l'axiome soit vrai. Peano, 3 apres avoir demontre l'independance de l'axiome, consacre a la discussion de sa verite seulement la remarque suivante: «Maintenant devons-nous croire que la proposition est vraie, ou qu'elle est fausse? Notre opinion est indifferente» (p. 148). M. Peano