Ex Libris 
 C. K. OGD 
 
 THE LIBRARY 
 
 OF 
 
 THE UNIVERSITY 
 
 OF CALIFORNIA 
 
 LOS ANGELES 
 
 w^:t/ 
 
 AD.CMASON 
 
 liiiiiiiiiiiiiiiiiiyMigjri
 
 /<rv'^<fi^j \ jvv-.-v.yftv
 
 LETTERS ON REASONING
 
 BY THE SAME AUTHOR. 
 
 ESSAYS TOWARDS A CRITICAL METHOD. 
 
 NEW ESSAYS TOWARDS A CRITICAL METHOD. 
 
 MONTAIGNE AND SHAKSPERE. 
 
 BUCKLE AND HIS CRITICS : a Sociological Study. 
 
 THE SAXON AND THE CELT : a Sociological Study. 
 
 MODERN HUMANISTS : Studies of Carlyle, Mill, 
 Emerson, Arnold, Ruskin, and Spencer, with an Epi- 
 logue on Social Reconstruction. 
 
 THE FALLACY OF SAVING : a Study in Economics. 
 
 THE EIGHT HOURS QUESTION: a Study in 
 Economics. 
 
 THE DYNAMICS OF RELIGION: an Essay in 
 English Culture-History. (By " M. W. Wiseman.") 
 
 A SHORT HISTORY OF FREETHOUGHT, Ancient 
 and Modern. 
 
 PATRIOTISM AND EMPIRE. 
 
 STUDIES IN RELIGIOUS FALLACY. 
 
 AN INTRODUCTION TO ENGLISH POLITICS. 
 
 WRECKING THE EMPIRE. 
 
 CHRISTIANITY AND MYTHOLOGY. 
 
 A SHORT HISTORY OF CHRISTIANITY. 
 
 PAGAN CHRISTS. 
 
 CRITICISMS. 2 vols. 
 
 TENNYSON AND BROWNING AS TEACHERS. 
 
 ESSAYS IN ETHICS. 
 
 ESSAYS IN SOCIOLOGY. 2 vols.
 
 LETTERS ON REASONING 
 
 BY 
 
 JOHN M. ROBERTSON 
 
 SECOND EDITION, REVISED, WITH ADDITIONS 
 
 [issued for the rationalist press association, limited] 
 
 London 
 
 WATTS & CO. 
 
 17, JOHNSON'S COURT, FLEET STREET, E.G. 
 
 1905
 
 First Edition, September. 1902
 
 Be 
 
 CONTENTS 
 
 Preface to Second Edition Pp. xv-xvi. 
 
 PROLEGOMENA 
 
 The drawbacks of conventional logic— These the motive for the Logique 
 de Port Royal — Arnauld's experience— Value of his treatise — 
 Condillac's new departure — Both protests implicit in Descartes 
 and in Bacon — Leonardo — Roger Bacon — Ultimate conventionali- 
 sation of new methods — Bain on that of Bacon — Arnauld and 
 Condillac no longer efificient — Admissions as to scholastic logic — 
 Whately, G. Bentham, and Mill— Bain— Uebervveg— Bain on the 
 Aristotelian treatment of fallacies— Minto on Aristotle's aim — 
 Need for exercise in reasoning before resort to abstract forms — 
 Lesson from the cases of Mill and Jevons— Jevons's protest against 
 the discrediting of logic — His miscarriages — Mr. Alfred Sidgwick's 
 warnings — Justification of a new experiment — Personal experience 
 —Drawbacks of academic conditions— Need to tread real grounds 
 of debate, and for precaution in treading them — Limitations of 
 Descartes, Arnauld, and Bacon— Grant need for new approaches, 
 and we may grant the claims of the technicists — Reply to 
 Ueberweg — The protest of Kant— Syllogistic figures justifiable, 
 but not as practical exercise in reasoning — Academic drawbacks 
 seen in the case of Venn— Exclusion of vital problems — Abstract 
 philosophy should come after formation of many judgments. 
 
 Pp. xvii — XXX. 
 
 LETTER L 
 
 Motive in writing — The rational and sesthestic life to be alike culti- 
 vated — Study of history — Importance of right reasoning in all 
 matters — The education of women — Fallacious objections to the 
 use of "reason" and "logic" — Such fallacies blundering 
 appeals to reason — The errors of reason to be corrected by more 
 reasoning — Bad and good reasoning — Common toleration of the 
 former — Relation of consistent thought to consistent action — Con- 
 sistency the basis of right action — The relations of logical percep. 
 tion and moral impulse — Intellectual inconsistency is intellectual 
 
 93'73S3
 
 vi CONTENTS 
 
 immorality — The study of technical logic — Relation of logic to the 
 processes of reasoning as that of grammar to language — Fallacious 
 notions of "logic" — Macaulay's notion — Analysis of his proposi- 
 tion on the subject — Logical compromise and the pursuit of the 
 ideal — Logic and emotion — and imagination — Effect of special 
 studies on the general reasoning powers — Comment of Professor 
 Jevons — Sample of a formal fallacy — Professor Minto on strict 
 deductive logic — All-round mental hygiene— To be sought by both 
 sexes Pp. 1-16 
 
 LETTER II. 
 
 Errors arising from false testimony — The need to doubt — The fallacy 
 of fearing to doubt — Doubt in the economy of Nature — The 
 fallacy of imputing " unbelief" — Our primary bent to credulity — 
 Professor Minto's acceptance of a false generalisation— Race 
 characteristics — The manner of false generalisations — Similar accept- 
 ance of error by Professor Clifford — Despite his censure of uncritical 
 belief — The need for vigilance — Prejudices of learning and pre- 
 judices of ignorance — All beliefs open to revision — Doubt therefore 
 to be welcomed — This not the way to live in hesitation but to 
 escape the most painful hesitations — The gain from doubt — Severe 
 intellectual standards to be applied to ourselves first— Thought 
 and life — Ignorant inaction and instructed activity — The pleasures 
 of thought Pp. 17-29- 
 
 LETTER III. 
 
 Snares of prejudice — The play of " will " — In believers and unbelievers 
 — Carnal inclination primary — Difficulty of (a) candour and (d) 
 sincerity — At bottom one difficulty — Honesty the best policy in 
 reasoning — The bane of self-contradiction — ^Ueberweg's avowal — 
 Modes of uncandid argument — The meaning of "sincerity" — An 
 etymological hint — Insincerity the moral aspect of inconsistency 
 — Fallacious notion of consistency as = constancy — Error and 
 culpability — Rarity of intellectual honesty — Perverting power 
 of malice — The physiological bias to misrepresentation — Primary 
 maxims for disputants — Abstract logic and concrete judgment — 
 Importance of information and study — Problem of the historicity of 
 Jesus — Progressive modification of confidence in historic narratives 
 — Infirmities of reasoning in men of science — Drawbacks of one- 
 sided culture — Case of Dr. James CroU — Effect of health and other 
 circumstances on departments of men's judgment — Croll and 
 Edwards — Knowledge a matrix for opinion — The general moral. 
 
 Pp. 30-46.
 
 CONTENTS vii 
 
 LETTER IV. 
 
 The right to liave an opinion — Influence of authority — First-hand and 
 second-hand opinion — Provisional assent — The dogmatism of ignor- 
 ance — Authority in the field of special knowledge — Morals and 
 religion not provinces of special knowledge— The claim for authority 
 in religion — Argument of Cardinal Newman — Normal Collapse of 
 the authoritarian position — The meaning of believing only what wc 
 understand — We can believe only the propositions that we under- 
 stand — The doctrine of Trinity in unity — Its origin — Sacerdotalism 
 and error — Anomalies of the method of authority — Authority in 
 ethics — The impossibility of imposing wisdom by authority — 
 Unwisdom to be modified only by the eftbrt of thinking — Authority 
 in matters of taste — Admiring to order — The mean between 
 obstinacy and flightiness Pp. 47-60. 
 
 LETTER V. 
 
 Errors of logical oversight — Case of transformation of terms — Sample 
 from Professor Jevons : disregard of a posited definition — The 
 probable explanation : defect of verbal faculty — ^Jevons's verbal 
 miscarriages — Transition from verbal error to real fallacy — Trans- 
 formation of terms under stress of prejudice — Sample from 
 Professor Robertson Smith — The temptations of the platform — 
 The snare of temper — Professor Momerie's self-contradiction as to 
 the opinion of the multitude — Hegel — His self-contradictions — 
 The theological motive — Varying use of the terms God and Reason 
 — The illicit transition from the process of reasoning to the process 
 of Nature — Hegel's self-contradictory treatment of the problem of 
 Evil — His personal equation — The Will-to-comprehend and the 
 Will-to-live — The problem of Evil insoluble on theistic lines — 
 Optimistic sophisms — The projection of self into the cosmos — 
 Moral dangers of the theistic method — Theological cruelty — 
 Curiosity to be cherished even if not to be satisfied. 
 
 Pp. 61-76. 
 LETTER VI. 
 
 Fallacies of verbal confusion — Opinions about opinion-making — Case of 
 Mr. W. E. H. Lecky — His merits and defects as a historian — 
 False developments from a true proposition — The conditions of 
 intellectual change — Belief one of the intellectual conditions — 
 Fallacy of supposing that great changes of belief occur without 
 argument — Influence of the philosophers — Mr. Lecky's self-con- 
 tradictions — His notion of logic — Distinction between no-judgment 
 and non-independent judgment— Mr. Lecky re-contradicts himself 
 — His noti(m of the "collective wisdom of mankind" — The laws
 
 viii CONTENTS 
 
 of average opinion-making — Use of the word Rationalism — Its 
 meanings — Montaigne — Misconception of his character and develop- 
 ment — His and Charron's risks and popularity — The belief in 
 witchcraft — Shakspere's supposed acquiescence in it — How the 
 belief was discredited — Mr. Lecky's theory of silent scepticism — 
 Resulting manipulation of historical facts— Practical applications of 
 the formula of "silent change" — The formula of "ideas in the 
 air " — Confronted by facts — True process of intellectual change — 
 The case of abolition of torture — Beccaria's movement — Mr. 
 Lecky's efforts to save his formula — Resulting historical misstate- 
 ments — Torture in England — False distinction between emotional 
 and intellectual movements — Emotion accrues to all judgments — 
 The opposition to Beccaria as Utopian — Effects of such opposition 
 — The phenomena of intellectual conservatism^Astrology — 
 Gabriel Naude — Mr. Lecky's admissions and further contradictions 
 as to the influence of Descartes — The general lesson of such error : 
 need for self-criticism ... ... ... Pp. 77-101. 
 
 LETTER Vn. 
 
 The logical problem of "chance" — Professor Minto's treatment of 
 chance as one of many " causes " — Meaning of the term " chance " 
 — Illustrations — We drop the term where we have divined manner 
 of causation — Games of chance — " Probability logic " and "laws of 
 chance "—Argument of Mr. R. A. Proctor — His fallacy — Practical 
 nullity of the numerical " law of chance " — Dr. Venn's Treatise — 
 His rebuttal of the thesis that "probability" means merely our 
 degree of expectation — His self-contradiction — Formal nugatoriness 
 of the mathematical calculus — The statistical estimate of chances in 
 terms of wide experience — This a different conception — The con- 
 fusion of chance with non-causation — Hume's contradictions in 
 terms — Futility of the formula "there is no such thing as" — 
 Hegel's lapse into fallacy — Inconsistency of his disciples — The 
 labour of exact thinking ... ... ... Pp. 102-114. 
 
 LETTER VIII. 
 
 The logical problem of "design" — Transition from the fallacy "not 
 by chance" to the fallacy " designed by an Infinite Person" — Fallacy 
 of the watch argument — Its negation of design in the desert to 
 prove design in the watch — Logical correctness but practical 
 inutility of the proposition that the order of Nature is a matter of 
 chance — Confusion of the argument from a bag of letter-blocks — 
 Similar procedure yields phenomena such as are recognised as
 
 CONTENTS ix 
 
 part of natural order — Theistic argument from the " machine" to 
 Nature— The totality of things is as such necessarily incommensur- 
 able—The sympathetic fallacy exploded as to phenomena, insisted 
 on as to the totality— There can be no rational ascription of single 
 mode to the totality of things — Terms of mode, to be significant, 
 must be comparative or contrastive — Professor Jevons's use of the 
 term "thing" — Infinity a negative predication: otherwise non- 
 significant— Nullity of the formula "Good Infinite" — The formal 
 fallacies of idealism— Its propositions finally verbalisms, signifying 
 only changes in definitions of terms — Progression of fallacy up to 
 theism— Momerie on " purpose " and " chance " — His metaphysical 
 inconsequence — Mill's self-contradictions as to the external world 
 — Hamilton on primary and secondary qualities of matter— The 
 idealist verbalism as to thought maintaining objects — Mill and 
 Hamilton on the metaphysics of the Infinite — Infinite Goodness a 
 countersense — Mill and Mansel — Momerie on Infinite Personality 
 — The outcome of argument on The Absolute — The making of an 
 Absolute from Self— Dr. W. Smith's exposition of Fichte- 
 —Nullity of the teaching— Professor Bradley's grounds of re- 
 jection — Summary of the debate on design in terms of logical first 
 principles — Theological insincerity as to causation— Causation not 
 a modal conception but part of the reasoned conception of sheer 
 existence — The theologian's self-stultification in passing from 
 "infinite design" to "individual free-will" — Such intellectual 
 perversity an aspect of the evil in the order of things — To be 
 conceived in terms not merely of insincerity but of difficulty of 
 problem — Above all in terms of imperfection of the instrument ot 
 language — The admissions of Jonathan Edwards — Language as 
 such peculiarly hard to reform — Conduct, however, in somewhat 
 similar case through force of early education — Rarity of faculty for 
 independent and morally consistent thought — Need for recognition 
 of this in order to the maintenance of charity— Approach to the 
 problem of Free-Will — Need to prepare for its difficulty. 
 
 Pp. 1 1 5- 1 38. 
 LETTER IX. 
 
 The problem of " Free-Will " — Sketch of the historical development — 
 Early notion of Fate — Theory of the influence of the stars — Mono- 
 theistic tendencies — The polytheistic ethic the more considerate — 
 Interchange of theories between races — Inconsistencies of all — 
 Special contradictions of the specialised sacrificial cults — The 
 Christian dilemma and compromise — Freedom and the need for 
 grace to pray for grace — Men's consciousness of their own moral 
 variability — Demoralising bias of ritual and sacrificial principle —
 
 CONTENTS 
 
 Common-sense reactions — The theistic complication — Protestant 
 and Catholic tendencies — Pelagius — Augustine — Gottschalk — 
 Luther and Calvin — Insoluble confusion arising from the premisses 
 on both sides : Omnipotence and All-Goodness — Fresh compro- 
 mises, with fresh strifes — Calvinism and Arminianism — Reasons 
 for the sectarian attitudes — Rationalistic objection to both, 
 resulting in new theological transformations — The logic of the 
 problem — Fallacy of ascriptions of relative mode to an absolute 
 and infinite series — Volitions neither free nor unfree, since 
 admittedly not alternatively either— Attempted compromise by 
 Dr. Shadworth Hodgson — Its fallacy — Mill and Green right, but 
 without theoretic demcastration — Fundamental confusion between 
 freedom and non-causation — "Non-free" not equivalent with 
 "caused," nor "uncaused" with "free" — " Uncausedness " a 
 chimera — The argumentation of Jonathan Edwards — His high 
 capacity — His problem — The theological crux — His measure of 
 success — Definition of virtue and vice — ^His scripturalism — His 
 collapse over the origin of Sin — Calvinism and Arminianism in the 
 same dilemma — No solution on theistic lines — Edwards's intel- 
 lectual circumstances — His ignorance of Spinoza — Spinoza's short- 
 coming — His great ethical advance — Immoral implications of 
 theism — Its bias to cruelty — The theological demurrer to deter- 
 minism — Recognised by Professor Fowler and Professor Sidgwick 
 — Their logical oversight — Sidgwick's argument that we "cannot 
 use determinist conceptions " — Its irrelevance — Gravitation analo- 
 gous to determination — -The moral gain from acceptance of deter- 
 minism — Purification of theological ethic— Dr. Martineau on 
 moral dynamics — The last theological defence — Responsibility — 
 the rational conception — Theological cruelty in punishment — 
 Corruption set up not by determinism but by theology — Praise 
 and blame unaffected by determinism save as regards the rational 
 control of blame — Blame bottomed on the law of self-preservation 
 ■ — The need to rationalise its application — Mr. Balfour's thesis an 
 obvious insincerity — Determinism cannot make us indifferent to 
 moral any more than to physical conditions— Solution of the 
 problem as to what constitutes true goodness — The theorem that 
 evil is non-ens — "The point of view of the Infinite" — Insincerity 
 of the theological defence — Risks of insincerity on the side of 
 rationalism — The impulse to revenge or retribution — Moral element 
 in the theological denunciation of iniquity — Rational ethic begins 
 where the theological ends — The rationalist's better chance of 
 checking his insincerities — His consolation in the knowledge of 
 men's improvableness ... ... Pp. 139-169.
 
 CONTENTS xi 
 
 LETTER X. 
 
 The moral problem of instinct — Instinct as determination— Some 
 instincts primary, some superinduced - Prejudice or hereditary 
 belief as instinct — Instinct either good or bad— Instinct modifiable 
 in terms of instinct — Egoism and altruism — Conflicts of opinion 
 as feeling— Struggle of ideals for survival— Morality progresses by 
 way of variation— Feelings modifialtle by argument— Progress in 
 morals extension of sympathy — Desire for Vjetterment a condition 
 of the argument — The new anti-reciprocity doctrine in national 
 morals — Its restriction to the relations of communities — The argu- 
 ment as to ancient slavery — Empirical treatment of it — Nature of 
 the ethical problem — Relative degrees of moral apathy in ancient 
 and modern times— Aristotle's attitude to slavery — His self-contra- 
 dictions—Fallacy of the formula justifying slavery — The moral and 
 the sociological analyses — Confusion over terms " right and wrong " 
 —The purpose of the pro-slavery thesis— Criticism of it— Eugene 
 Aram's private formula — Dostoyevsky's handling of it — The new 
 formula pseudo-Utilitarianism — Need for a test — The merits of the 
 Intuitionist attitude— The developed instinct of reciprocity- 
 Dilemma set up by conflicting intuitionisms — Process of growth of 
 the instinct of reciprocity — Possibility of expanding it — Fallacies of 
 the anti-sympathetic doctrine — The historic analysis — Fallacies as 
 to civilisation— Results of repudiating the law of reciprocity — 
 Failure of the repudiators to study utility — Moral convictions may 
 fail to stand rational tests — Proper course then is to test all — The 
 problem as to the occasional permissibility of falsehood — Kant's 
 absolutism — Criticism of it — His gross inconsistency — The rational 
 compromise— Rational criticism of social deceit— Misrepresenta- 
 tions of rationalist principle by intuitionists — Their unconscious 
 reversion to utilitarianism — Martineau ... Pp. 170-194. 
 
 LETTER XI. 
 
 The problem of moral ends — Hedonism and its gainsayers — The term 
 and the opposition alike faulty — Locke on " pleasure " and Pope on 
 Epicureanism — Carlyle on happiness and Ijlessedness — The theo- 
 logical embitterment of the dispute — Rectification of terms — 
 Pleasure as an end and as an aim — Reconciliation of part of the 
 strife — The clash of ideals and intuitions — Definition of egoism 
 and altruism — Common standing-ground — Ideals are tastes — 
 Green's argument and surrender — Source of his inconsistency in bis 
 theistic presuppositions — His Spiritual Principle a rhetorical name 
 for his bias — His argument in a circle — His account of the 
 "categorical imperative" — His argument for an eternal mind — Its
 
 xii CONTENTS 
 
 nullity — His compromise as to religious practice — Its equivocal 
 justification — The practical explanation — Danger and inutility of 
 such opportunism — Cases of defensible make-believe — Instances to 
 prove the general inutility — Stoning the prophets — The risks of the 
 ideal being wrong — Spirituality consists in facing them after testing 
 it — Unreality of much so-called spirituality — Contrast between 
 tested and untested ideals — Each of us the present consummation 
 of the cosmic drift Pp. 195-216. 
 
 LETTER XII. 
 
 The problem of "reality" — Special pressure of the difficulty from 
 language — ^Danger of metaphors — Professor Bradley's use of them — 
 Need for verbal analysis instead of fresh complication — The Hindu 
 countersense, " All is Illusion" — Approach to be made by way of 
 elementary physics — Metaphysics of colour — Existence and percep- 
 tion — Resort to "God" to solve verbal dilemmas — Attempts to 
 save the God-idea as a reality — Their end in making all fallacies 
 aspects of truth -But subjective ideas still to be reckoned with — All 
 perceptions " real " in some degree — Mirage an actual phenomenon 
 — Regression to detected illusions — Relativity of moral realities- 
 Relapse into predication of imaginary existence to be avoided — 
 Subjective reality to be tested, and relativity of tests to be admitted 
 — Samples of avoidable error — Safety to be found in openness to 
 every aspect of reality— The mystery of the universe — Some 
 knowledges useless, but brooding profitable — Return to the formula 
 of "existence only in consciousness" — Analysis thereof — Nugatory 
 use of "only" and "in" — Rational form of predication as to 
 colour — "Exists only " and "cannot exist" equally fallacious — 
 Futility of attempt to formulate relations in terms of infinite 
 perception — History of the dispute as to matter and substance — 
 Berkeley — The God-idea, sought as an abstraction of unchange- 
 ableness, ends in being an abstraction of Change — Rational use of 
 the term "Appearance" — Not equivalent to " illusion " or " un- 
 reality"— Formula of the measure of reality — The quasi-scientific 
 formula of knowledge as "shorthand" for reahty. 
 
 Pp. 217-236. 
 LETTER XIII. 
 
 Return to the starting point : how is the reasoning faculty best to be 
 developed ? — At least the effort to reason out important issues may 
 help us to study logic more critically — Symbolic logic — Its new 
 developments — Issues still left open— A sample case, to be tried by 
 the tests of consistency — The ethics of suicide — How is the act to 
 be judged ? — The analogy of positively anti-social acts — Conditions
 
 CONTENTS xiii 
 
 of rational denunciation — The theory of cowardice — Analogy of 
 dipsomania and kleptomania — But does the charge apply ? — 
 Resulting anomalies — Suicide to be dispassionately discussed — 
 An act sui generis — Its different aspects in different cases — The 
 duty of living to help others — Inaptitude of the despondent for the 
 task — Suicide pathologically and statistically considered — The right 
 to suicide arguable in certain circumstances — Futility of denuncia- 
 tion — Inconsistency and theological origin of present practice — 
 Symbolic logic leaves such processes of argument still to be made — 
 The healthy line of growth — Moral logic is mainly in terms of life : 
 the logic of propositions of existence is in terms of the facts of per- 
 ception and conception — The question as to the best gymnastic for 
 the reason — Value of science — Special discipline and point of view 
 for the moral sciences— Advantages of the study of the physical 
 sciences — The per contra — The problem of society — Inadequacy of 
 the procedure of many men of science on this side — A pseudo- 
 scientific view of ethics — Darwin's mishandling of the problem of 
 natural procedure — Exposure of the fallacy — Spencer's implication 
 in it — The test of consistency applied — Rational meaning of 
 " following Nature " — All action natural : argument to be founded 
 on results of attempts, not on absence of attempts — Wrong methods 
 as natural as right — Nature cannot yield a code of conduct 
 by mere example — The fallacy of studying the "cosmic plan" — 
 Mr. Rhodes — Reason as well founded in Nature as egoism — 
 Claims for supernaturalist religion — Story of Lykurgus — Vogue of 
 the ideal of brute force — Its popularity due to ignorance of history 
 — Knowledge corrective — The case of Major Macpherson — His 
 reclamation of the Khonds — His ideal and his methods — His 
 success a triumph of reason and of reasoning — His painstaking 
 research before acting — His moral and mental superiority to mere 
 conquerors — Rarity of his ideal to-day — His avowed debt toGuizot 
 — Preference to-day for rule of thumb and the light of instinct — 
 The difference between the ideals is the difference between good 
 and bad reasoning. This the thesis of these letters. 
 
 Pp. 237-256. 
 
 Index ... p. 257
 
 PREFACE TO THE SECOND 
 EDITION 
 
 The comparatively rapid sale of the first edition of this work has 
 encouraged me to think that it in part meets the needs which suggested 
 it. Of the many friendly criticisms passed upon it, some suggested a 
 more gradual appeal to the reader's thinking powers, a leading-up from 
 small logical difficulties to the greater. Such a treatment, I can easily 
 see, would have advantages. It would, however, involve a general 
 recasting of the book ; and I feel it could be better undertaken in a 
 separate volume, which I have not at present the leisure to write. 
 
 It seems well to explain here, once for all, that the book is not 
 planned for juvenile readers. The fact of its being addressed to my 
 children has led several reviewers — who, it is interesting to know, 
 cannot have read the first paragraph of the first letter — to observe, not 
 unamiably, that they must be clever children who can understand it. 
 Here the ambiguity of the term "children," although expressly 
 guarded against, has served to illustrate the facility and frequency of 
 wrong inference. As the opening paragraph states, the book was 
 planned to be read by my children when they are grown up : mean- 
 time it is meant to help others if it can. 
 
 To the more relevant criticisms passed upon it, I have not seen 
 reason to give way in any important matter. One of the canons 
 frequently founded-upon in it was impeached by one reviewer as a 
 fallacy ; but as he vouchsafed no argument I am unable to catch his 
 point of view. Several journals, as was to be expected, complained 
 that I had dealt with religious opinions. Had I supported or founded- 
 on the opinions in question they would not have so complained. So 
 far from admitting that the discussion of such opinions is a fault of 
 method, I have somewhat expanded some of the passages in question. 
 In the belief that such handling of living questions is a good way of 
 vitalising discipline in reasoning, I am more and more encouraged by 
 the movement of academic opinion. Since the issue of the first edition 
 I have had the advantage of reading the instructive work of Mr. Alfred 
 Sidgwick on The Use of Words in Reasoning (1901), wherein is
 
 xvi PREFACE TO THE SPXONI) EDITION 
 
 temperately maintained and illustrated its author's earlier contention on 
 the inefficacy of formal logic as a means of developing the reasoning 
 powers. Although I cannot wholly acquiesce in his conclusion that 
 " mistakes in reasoning are nothing but mistakes in the facts from 
 which the reasoning proceeds," I can cordially assent to his judgment 
 that "our modern errors have a pedigree; and Logic will remain an 
 almost useless study so long as we forget that it is in the subject-matter 
 of reasoning, not in any abstract ' reasoning-process,' that all effective 
 error is concealed." The more need to grapple, in such a treatise as 
 the present, with some of the more vital errors known to be current. I 
 might even claim, I think, the ratification of a sentence of Professor 
 Bradley, who has written • — 
 
 " For myself, though I have not hesitated to point out the falsity and immorality 
 of some Christian doctrines (where this seemed necessary), I cannot approve of the 
 widespread practice of treating them as devoid even of existence " — 1 
 
 though I baulk at the phrasing of the last clause. Unfortunately, 
 while an academic writer such as Green may turn his function to the 
 account of his religious beliefs, academic writers who do not agree 
 with him are debarred from confuting him in treatises meant for the 
 use of students. One can but hope that ere long the education of the 
 reasoning faculty will be generally gone about by way of a real 
 gymnastic in at least non-religious discussion of real problems. But 
 as regards religious problems we are bound to say, with Mr. 
 Sidgwick (p. 39), that "In the general ignorance of Logic which 
 prevails, and which is fostered by the traditional teaching system, it 
 is not difficult to make people accept a circular truism as a deep 
 philosophical truth." I have taken some pains to expose one such 
 process. 
 
 To the criticism, passed by one reviewer, that in the Prolegomena I 
 have unwarrantably ignored the recasting of logic by such writers as 
 Sigwart and Mr. Bosanquet, it should suffice to answer that mv remarks 
 implied no undervaluing of the very able works of those writers, which 
 are rather systems of epistemology than treatises aiming at training 
 young students in the practice of reasoning. Mr. Bosanquet ex- 
 pressly gives to his the sub-title of "The Morphology of Knowledge." 
 I readily admit that he does his work much more searchingly and 
 instructively than did the older analysts before Mill and Bain, and that 
 Sigwart outgoes those writers ; but neither comes nearer than they 
 to the humble object aimed at in the present volume. 
 
 ' Appearance ami Reality, 3rd ed. App. p. 558. 
 March, 1905.
 
 PROLEGOMENA 
 
 The aim of the following letters is sufficiently explained in 
 the first ; but it may be well to avow to the critical reader, 
 other than those there addressed, that since the book was 
 first thought of, a good many years ago, and even since it 
 began to take form, I have met with quite a number of 
 recent writings which similarly urge the need for a cultiva- 
 tion of the reasoning faculties on other lines than that of 
 conventional logic. Indeed, to say nothing of immemorial 
 gibes at logical forms and futilities, such suggestions have 
 been chronic ever since the issue of the Logiqiie de Port 
 Royal (1662). 
 
 Thus the first English translators of that work (1685) 
 declare in their preface that " the common treatises of Logic 
 are almost without number, and while every author strives 
 to add something of his own, sometimes little to the pur- 
 pose, sometimes altogether from the matter, the art is 
 become not only obscure and tedious, but in a great 
 measure impertinent and useless." And Arnauld himself, 
 in his second prefatory discourse, justifies his treatise by 
 the very true plea that " experience shows that, of a 
 thousand young men who learn logic, there are not ten 
 who know anything of it six months after they have 
 finished their course. Now, the real cause of this common 
 oblivion or negligence seems to be that the matters treated 
 of in logic being in themselves so abstract and remote
 
 xviii PROLEGOMENA 
 
 from usage, they are further bound up with unattractive 
 examples, never otherwise discussed, so that the mind, 
 grasping them only with difficulty, has no way of holding 
 them, and easily loses the ideas it had acquired because 
 they are never renewed by practice." 
 
 Of the merit of Arnauld's Art de penser there can, I 
 think, be no question ; for though he partly conformed 
 to the old methods by setting out with a formal classi- 
 fication of ideas, he goes about his task with such a 
 vivacious hold on current interests, and follows up his 
 analysis with so much of practical illustration and counsel, 
 that the treatise must have been nearly as stimulating as it 
 was popular. 
 
 A century and more later, nevertheless, Condillac begins 
 his Logique (1781) with the remark that as men were 
 mechanicians before they sought to be such, so they have 
 thought before they inquired how we think ; and he pro- 
 poses to frame a logic which shall "resemble none of 
 those hitherto made," in that it begins by showing how 
 " nature teaches us to analyse." Here, in effect, a sub- 
 jective psychology is substituted for scholastic logic, not 
 quite dissimilarly to the opening procedure of Arnauld. 
 Perhaps these two modes of vitalising the reasoning con- 
 sciousness are the essentials of every such attempt ; and 
 perhaps both are implicit in the Discours de la Methode 
 of Descartes, who also had revolted from the scholastic 
 logic, and doubtless gave the impulse that led Arnauld to 
 his performance. Nay, they are implicit in the Novum 
 Orgattum of Bacon, who in turn is only the great type, not 
 the beginner, of the revolt from the verbalism of the older 
 schools. Leonardo da Vinci had said in little what he 
 said at large. Finally, as Professor Minto reminds us, the
 
 PROLEGOMENA xix 
 
 regression to experience was argued for by Roger Bacon 
 centuries before. 
 
 Yet probably every one who is much concerned in a 
 practical way about right thinking will admit that, after all 
 that has been accomplished in mental and physical science 
 since Bacon and Descartes, their protests and warnings and 
 suggestions need urging to-day as much as ever. As Pro- 
 fessor Bain puts it, the warnings of Bacon, " instead of being 
 laid to heart and followed tip by fresh examples, became a 
 matter of mere parrot repetition." From the higher 
 standards of our scientific method we normally lapse 
 even as Bacon and Descartes did from theirs: above all, 
 in our every-day handling of the problems of the moral 
 sciences — history, politics, sociology — and the common 
 conflicts of literary and personal -opinion, we are about 
 as far from good scientific method as was Bacon in his 
 physics. 
 
 It would seem to be in the nature of the case, then, that 
 such attempts as Arnauld's have to be renewed, especially 
 as the scholastic or diagrammatic logic always holds its 
 ground in education, however frequently it be recast. The 
 practical argumentation of Arnauld is so thoroughly coloured 
 by the thought and the opinions of his time that it must be 
 more of a bewilderment than of a guide to a young reader 
 to-day : its merit for its time can be appreciated only 
 through a comparative historical study. Condillac, in turn, 
 is in a manner repeating, though quite in his own way, the 
 processes of Descartes and Bacon : he is formulating a 
 method and outlining a psychology ; and in so far as his 
 method and results have been embodied in later scientific 
 procedure he is out of immediate touch with the needs of 
 to-day, which, indeed, does not read him. 
 
 B 2
 
 XX PROLEGOMENA 
 
 That the scholastic logic all the while has been a bad 
 educational instrument is practically admitted by nearly 
 every notable author of a logical treatise. Whately, who 
 was credited by Hamilton and Mill with reviving the study 
 of logic in England, did so, if at all, by making it speak in a 
 measure the language of real life, while retaining many of 
 the formulas and processes of the " abhorred pedantic 
 sanhedrim." George Bentham praised him for "divesting 
 the science of that useless jargon, of those unmeaning puer- 
 ilities, with which it had been loaded by the schoolmen." ^ 
 Bentham in turn saw " much remaining to be done," and 
 made what are held to be valuable improvements (whether 
 new or rediscovered), such as the " quantification of the 
 predicate "—a small mercy for which the thanks have been 
 suggestively cordial. With these technical improvements 
 we are not here concerned : they are part of the anatomic 
 study which is so commonly recognised to fail in developing 
 or refining men's reasoning habits. Bentham's definition 
 of Logic, indeed, was " the branch of art-and-science which 
 has for its object the advantageous application of the human 
 mind to the study of any other branch of art-and-science " ; ^ 
 but he only overhauled the old machinery, however intelli- 
 gently. John Stuart Mill handsomely prefixed to his bril- 
 liant System of Logic, as motto, the claim of his chosen foe, 
 Hamilton, that " to the schoolmen the vulgar languages are 
 principally indebted for what precision and analytic subtlety 
 they possess," as well as the less sweeping claim made for 
 scholasticism by Condorcet. Those claims may be just, 
 though Condillac says the exact opposite ; but nonetheless 
 Mill sought to vivify his handling of scholastic logic by 
 
 ^ Out Hue of a New System of Logic, 1827, pref. 
 - Id., p. 14.
 
 PROLEGOMENA xxi 
 
 founding it in psychology and quadrating it with the pro- 
 cesses of science. 
 
 Professor Bain in his turn has done more systematically 
 and at many points more cogently what Mill undertook 
 to do, the result of their (one might say) joint work being 
 an expansion of so-called logic far beyond the strait frame- 
 work of the traditional processes. He who studies their 
 books has gained something incomparably more momentous 
 than the old drill in the anatomy of the obvious types of 
 proposition : he has been kept through the greater part of 
 the course in contact with living thought and illuminating 
 knowledge. And some similar tribute is due to the work 
 of Ueberweg, who, gathering up the mass of German logical 
 activities, makes of Logic an integral part of philosophy and 
 a discipline in philosophic criticism. Whether these and 
 other such services wholly ward off the paralysis so apt to be 
 set up by the ancient machinery of " All animals are mortal " 
 and " Some X is Y" and "Any A is some C," or whether 
 the service suffered by the retention of those benumbing 
 implements, I will not venture to inquire, in face of such 
 displays of mental power through technique. But it is 
 perhaps not unwarrantable, and it is certainly not on my 
 part disrespectful, to suggest that in the new logic, on its 
 own implied assumptions, the educative and, as it were, 
 muscle-making training given by the surveys of Induction or 
 " Applied Logic " might more fitly come before than after 
 the partly metaphysical and wholly formal training of 
 Deduction. Professor Bain seems to endorse what he cites 
 when he says, " there is a general conviction that the 
 utility of purely Formal Logic is but small ; and that the 
 rules of Induction should be exemplified even in the most 
 limited course of logical discipline." Ought not then the
 
 xxii PROLEGOMENA 
 
 admittedly more helpful and educative exercise to come 
 first? 
 
 Professor Bain, with his usual sureness of method, points 
 out the incongruity of the old habit, dating from Aristotle, 
 of putting a general collection of fallacies at the end of a 
 treatise on logic. The very fallacies so treated by the older 
 logicians, as it happened, were aloof from useful thought : 
 as George Bentham put it, these logicians, after defining a 
 fallacy as " an argument which appears to prove a question 
 which is not fair or true," proceed to " give as examples of 
 fallacies, arguments (or rather puns) which can never, in 
 this sense, appear to be correct." ^ And even if they did, 
 their relegation to an appendix goes far to deprive them of 
 educational value. As Bain urges, there is no propriety in 
 treating fallacies, in a logical treatise, separately from the 
 rules of which they are transgressions. But the inveteracy 
 of that anomalous plan is perhaps fitly to be set down 
 simply to the common consciousness of the need for a 
 more practical training than rule-framing supplies. As 
 Professor Minto so clearly impresses on us, the primary 
 aim of Aristotle was to guard against fallacy, against bad 
 reasoning : that is in fact the main end of logic. And 
 Professor Bain, while demurring to the customary ragbag 
 of miscellaneous fallacies, virtually excuses it on the score 
 of " the narrowness of the field of logical precepts, from 
 Aristotle down to the present time," proceeding for his 
 own part to justify the separate handling of " the fallacious 
 tendencies of the human mind, and the Fallacies of Con- 
 fusion." 
 
 Now, in the opinion of some of us, these are the 
 commonest drawbacks to all processes of reasoning, and 
 
 ^ Work cited, p. 223.
 
 PROLEGOMENA xxiii 
 
 they notoriously survive, in most minds, the discipline of 
 codified logic, to-day as in Arnauld's time. Ought they not 
 then to be brought home to the student, by a real dialectic 
 experience, before he is set to the often tedious and 
 frequently stupefying work of analysing idle propositions by 
 way of arriving at the fieshless and bloodless norms of 
 logical thought? Ought not, in fact, a free exercise in 
 intelligent reasoning to precede the logic of the schools ? 
 
 One reason for thinking so is that we are still forced to 
 note, even in the case of Mill, that the old scholastic course 
 of culture was not wholly fortunate in its intellectual results. 
 He is not indeed to be blamed for having taken, in his 
 manifold performance, so many intellectual risks ; but the 
 educational fact remains that he is after all capable of 
 seriously bad reasoning when he handles the living 
 problems of economics and philosophy. The late Mr. 
 Jevons, again, has like Mill exposed some of the blunders 
 of Hamilton in the very technique to which he devoted 
 himself; and Mr. Jevons in turn seems to some of us not 
 merely to make formal logic at times more than usually 
 nugatory — as in his celebrated " conversion " of the 
 proposition "it rains" by "something which is letting rain 
 fall, is the atmosphere " ^ — but to burden it with bad philo- 
 sophy, as in the propositions '•'■ thing existing has its contrary 
 in thing not existing'^ ; and "even things the widest noun 
 in the language, has a contrary in that which is not a 
 thing^ 2 Here the terms thing and contrary are formally 
 unfitted for any practical use. 
 
 At the close of his interesting and able little work. The 
 Substitution of Similars (1869), Mr. Jevons complains 
 
 ^ Studies in Deductive Logic, 2nd ed., p. 40. 
 " Pure Logic, 1864, p. 65.
 
 xxiv PROLEGOMENA 
 
 that " Logic is under the ban of metaphysics. It is falsely 
 supposed to lead to no useful works — to be mere specula- 
 tion ; and accordingly there is no journal, and no society 
 whatever, devoted to its study. Hardly can a paper on a 
 logical subject be edged into the proceedings of any learned 
 society except under false pretences." And he concludes 
 by protesting " how absurd it is to cultivate the branches 
 of the tree of knowledge, and neglect the root — which root is 
 undoubtedly to be found in a true comprehension of logical 
 method." The public is probably as indifferent to-day as 
 in 1869 to the special cultivation of a logic which leaves 
 a specialist capable of seeing the root of knowledge 
 " undoubtedly '' in a true comprehension of logical method ; 
 and Mr. Jevons's " logical abacus " has failed to take rank 
 as an educational requisite, perhaps because, among other 
 things, the power to use it did not save him from stumbling 
 scandalously over his accepted definition of wealth in his 
 little Primer of Political Economy. Coherence in such 
 matters is rather more important than the capacity to treat 
 " it rains " as a convertible proposition. 
 
 Apart, however, from such grounds for doubt as to the 
 educational effects of scholastic logic, the old verdict on it as 
 a partially vain employment of time is repeated in our own 
 day by such an experienced teacher as Mr. Alfred Sidgwick, 
 with many of whose remarks on the subject I find myself in 
 hearty agreement. These, for instance : — 
 
 " Beginners in Logic arc even more apt than those who have never 
 liegun the study to commit the unfairness, in actual argument, of 
 overlooking . . . troublesome considerations. The first view we get 
 of Logic is generally much too abstract, too hard and fast in its con- 
 ception of the reasoning process. The forms of argument that we learn 
 from our text-books are far too simple for direct application to actual 
 pieces of reasoning- except of the indisputable kind— but this fact is
 
 PROLEGOMENA xxv 
 
 rather hidden from us by the traditional doctrine, and has to be 
 rediscovered when we get beyond that study. There seems no reason, 
 however, why we should not at once begin with the recognition of it, 
 especially since it is already seen with some clearness by many whose 
 only Logic is provided by common sense." 
 
 "Whatever study best corresponds to the old name of Logic, or 
 provides the best training in grammar, or simplifies most the art of 
 setting questions which shall test a student's industry, the subject that 
 is usually taught as Logic gives little or no help in regard to the actual 
 difficulties of reasoning, or as to the sources of the more plausible kinds 
 of error and verbal confusion. As a historical study it may have great 
 value — for the few who have time to pursue it as a part of the general 
 history of philosophy. As a mental exercise it may or may not be as 
 good as a game of chess. But for the purpose here kept in view — of 
 gaining some insight into the distinction between sound and unsound 
 inference — it is an open question whether the good or the harm it does 
 is greater. Taken as a whole, it is an accumulation of odds and ends 
 that have survived from various outgrown philosophies ; and, so far 
 as it does deal with the distinction between sound and unsound 
 inferences, it introduces an artificial simplification of the difficulties in 
 too unyielding a form."^ 
 
 Even a partial acquiescence in these strictures will be 
 a sufficient admission of the need for other than the 
 scholastic fashions of cultivating the reasoning faculties 
 of young people. And it would seem to follow from our 
 slight survey of the matter that the attempts made in other 
 directions may profitably vary in their procedure. What 
 is possible in the way of classifying and analysing the 
 methods of scientific proof, in the present state of know- 
 ledge, has been accomplished by Mill and Professor Bain : 
 at least I cannot make the faintest pretence of competing 
 with them. The " Inductive Logic " of Professor Bain is 
 an admirable store of ordered and tested judgment ; and 
 the very fact that it so far exceeds in bulk as well as in 
 general interest his volume on Deduction is a support to 
 such pleading as Mr. Sidgwick's. But inasmuch as all 
 
 ^ The Process of Argument, 1893, pp. 50-51, 74-75.
 
 xxvi PROLEGOMENA 
 
 such treatises as theirs, being framed more or less for 
 academic purposes, must aim at codification, and codifica- 
 tion is an authoritative synopsis of reasoned truth rather 
 than a help or stimulus to be reasonable, I am fain to think 
 that my old wish to set up such a stimulus on non-scholastic 
 lines is still worth acting on. 
 
 I had never been able to feel either that the youthful 
 study of Formal Logic greatly sharpened my own wits, or 
 that their craft in it made my masters impeccable reasoners. 
 When, therefore, the idea latterly came to me to put my 
 effort in a specially concrete form, by turning it to the 
 prospective needs of some of those nearest to me, I decided 
 the more readily to make the experiment in that manner 
 because of the contemplation of what had been done. I 
 may thus best gather what profit is possible from comparing 
 notes with contemporaries. 
 
 Some of these perhaps may share, finally, my feeling that 
 there is room for a small book which, seeking to do some- 
 thing for rational culture in a non-academic way, escapes the 
 burden so unavoidably laid on academic treatises, of keeping 
 away from some of the most universally important grounds 
 of debate among men. My special purpose leads me, and 
 my general purpose entitles me, to tread these grounds, not 
 aggressively or perpetually, but with a due sense of their 
 breadth and lasting interest. It is not to be denied that the 
 first deUberate departures from abstract logic gave ground 
 for a return to it, in that they ran to discussions where the 
 innovators logically miscarried. As soon as Descartes and 
 Arnauld broke away from the old methods they began 
 arguing about God, and the argument is not now profitable. 
 Bacon, on the other hand, began arguing about physical 
 science, where he had no practical competence, and where
 
 PROLEGOMENA xxvii 
 
 he was no more " logical " than Aristotle. But even those 
 miscarriages were probably more educative than the correct 
 trifling of the contemporary syllogists. There was probably 
 no way by which in the seventeenth century men could 
 better be provoked to critical thought than by Cartesian 
 theism and Baconian groping in induction. The pioneer, 
 the innovator, is sure to err ; but his effort may widen the 
 field of intellectual action. 
 
 And where, as in the present modest compilation, there 
 is no pretence of methodic innovation, but simply an 
 attempt to apply to the province of general reasoning 
 methods already tried with success in school education, 
 there is at least ground for hope that the effort will not be 
 wholly vain. Teachers have before now undertaken to 
 begin teaching botany by taking their pupils on a ramble 
 in the fields. I have essayed something similar. Grant me 
 that you do not best set about training a chemist by putting 
 the youth to books with never an experiment — that work in 
 the laboratory must precede mastery of the text-books — and 
 I will grant the claim of Ueberweg and Drobisch that " it 
 belongs to Logic to lay down completely in an exhaustive 
 division the different relations possible," and that "it abso- 
 lutely belongs to the strictly scientific demands to develop 
 completely the possible forms of inference." But Ueber- 
 weg's further claim : 
 
 "If it be justly consideretl that something valuable has been done 
 when natural science, by empirical collection of the discoverable species 
 to any one genus, has reached complete cognition, how much higher 
 must the gain be when we succeed in reducing the former possible to a 
 vmiversal principle, and in proving with mathematical accuracy the 
 completeness of the enumeration?" 
 
 calls forth just such a rejoinder as would be made by a 
 thoughtful botanist to one who insisted on the enumeration
 
 xxviii PROLEGOMENA 
 
 and classification of species as the best training in botany. 
 To revert to the analogy of the chemist, there must be 
 actual handling. 
 
 If it be answered that such handling is really given in a 
 good logical treatise, we come back to the old crux. Partly 
 by reason of tradition, partly because of the very purpose of 
 reducing to types the forms of argument, and partly by 
 reason of the need to keep academic treatises free of 
 current disputes about belief and action (a constraint 
 which, it is to be feared, keeps academic treatises on ethics 
 similarly " safe "), the fact remains that, save in the inductive 
 parts of logicians of the school of Mill and Bain, the forms 
 presented to the student as types of argument are in gene- 
 ral the mere unidentifiable and uninteresting skeletons of 
 beliefs and errors, or rather artificial skeletons framed to 
 exhibit typical structures and exclude their variations. If 
 Kant, whom Ueberweg convicts clearly enough of miscon- 
 ceiving the nature of the syllogistic figures, had made this 
 the gist of his protest against them, he would have been 
 unanswerable. They have their justification, but not as a 
 means of teaching us to reason soundly on the complex 
 problems of our daily intellectual life. Kant's revolt is at 
 bottom the chronic historic revolt— already glanced at — of 
 the living human intellect from formal logic considered as 
 a drill in the use of its limbs. 
 
 Some minds, it may be, positively prefer symbols and 
 diagrams to the lore they purport to represent ; but to 
 many more, I am satisfied, they are irksome and unattrac- 
 tive ; and even among those who find symbols helpful to 
 concrete thinking, many find nothing to appeal to them 
 in the formulas of the logicians. A final reason for pre- 
 ferring another method is supplied to me on the one hand
 
 PROLEGOMENA xxix 
 
 by the criticism passed on some of the mathematical 
 logicians by Dr. Venn, in his notable work on The Logic of 
 Chance, and on the other hand by his own practice. In 
 his preface, remarking on the common opinion that 
 Probability-logic is but a very ingenious apparatus for 
 setting and solving mathematical puzzles, he says : — 
 
 " II must be adnulled Ihat some ground has been given for such an 
 opinion. The examples commonly selected by writers on the subject, 
 though very well adapted to illustrate its rules, are for the most part of 
 a special and peculiar character, such as those relating to dice and cards. 
 When they have searched for illustrations drawn from the practical 
 business of life, they have very generally, but unfortunately, hit upon 
 just the sort of instances which, as I shall endeavour to show hereafter, 
 are among the very worst that could be chosen for the purpose. It is 
 scarcely possible for any unprejudiced person to read what has been 
 written about the credibility of witnesses by eminent writers, without 
 experiencing an invincible distrust of the principles which they adopt. 
 To say that the rules of evidence sometimes given by such writers are 
 broken in practice would scarcely be correct ; for the rules are of such 
 a kind as generally to defy any attempt to appeal to them in practice. " 
 
 So much for the worth of a large body of academic logic. 
 Though, as will appear in the following Letters, I make 
 some minor objections to even Dr. Venn's handling of the 
 problem of "Chance," I share what I believe to be the general 
 opinion that his book is one of the most original and 
 perspicacious logical treatises of our time, and that it has 
 done much to take away the reproach of trifling from 
 Probability-logic. It seems to me an important contribution 
 to psychological philosophy. Yet I must confess that when 
 I read its discussions of real problems of belief, such as 
 those in the acute chapters on testimony and the credibility 
 of extraordinary stories, I am struck by its inutility as an 
 educational instrument applied to unformed minds. Dr. 
 Venn was not afraid (albeit he had to put himself under the 
 academic segis of Probability) to handle such a question as
 
 XXX PROLECiOMENA 
 
 that of belief in Miracles : he even introduces at that point 
 reflections which do not belong to the apparatus of Proba- 
 bility ; but by reason of his apparatus he does not contrive 
 (as it seems to me) to bring home the nature of the problem 
 to minds that have not repeatedly looked all round it already. 
 He is in the grip of the academic machine. The procedure 
 of abstraction keeps the thought elusive : there is no help 
 to vivid realisation of what is meant ; and I can conceive of 
 students going through that part of Dr. Venn's magistral 
 treatise without catching his point of view. It is the old 
 trouble of putting young men to the formation of a strictly 
 abstract philosophy — the occupation of maturity — before 
 they have attained to any store of self-made judgments on 
 the problems of common beUef. 
 
 So much by way of explanation and apology to the patient 
 reader for putting in print a work schemed as this is. The 
 practical test, as regards the present small performance, will 
 be its effect on his judgment. Solvitur ambulando.
 
 LETTERS ON REASONING 
 
 LETTER I. 
 
 My Dear Children, 
 
 I sometimes wonder whether, when you grow 
 up, you will care enough about any of my studies to wish to 
 read what I wTOte about them. It is not at all likely that 
 either of you will care enough about all of them to read all 
 my books ; and I can neither guess what choice either of 
 you may make, nor decide how I would like to have you 
 choose. But when I think of you growing up and forming 
 your minds, perhaps after I am dead, there comes to me a 
 wish that at that stage you should have at least one book of 
 mine which I may ask you to read for your own sake and 
 for mine, whatever your mental tastes may be. If I should 
 be gone when you become old enough to understand these 
 letters, you will read them none the less willingly because, 
 while written for you, they may have served also for other 
 young people. 
 
 There is a fair chance that you may both turn out less 
 fond of reading than I was ; and I have no great wish that 
 you should grow up to write books. There are so many 
 already ; and it is so hard to write a good one. ^Vhen I 
 look into the earlier of my owti, I always wince over some- 
 thing, and wish I could rewrite them ; and in the margins 
 of the later and larger I often make additions, to improve
 
 2 LETTERS ON REASONING Letter L 
 
 them. There are many other ways of using the mind and 
 the eyesight that are at least quite as pleasure-giving and 
 quite as useful. What I would counsel you to do is to live 
 your lives cordially and joyously, never shunning serious 
 matters because they are serious, but living, so to speak, 
 much in the sunshine. Partake freely of great music and 
 great art ; think about them all you can ; and read — so I 
 advise, whatever may be your bias — plenty of poetry and 
 good fiction, taking what guidance you can get from people 
 and writers who seem to you intelligent, but always trying 
 to judge for yourselves what you have read. 
 
 There is, however, little need thus to counsel you to range 
 freely in imaginative literature : we all seem to take to 
 that very readily, whether or not we read critically. More 
 necessary is it to urge people to read a good deal of history ; 
 and this I trust you will both do, as I cannot see how 
 mankind is ever to grow collectively wiser until it has 
 learned from the errors and successes of past generations 
 how to escape their failures and provide for a steadier 
 progress. But I am not going to make these letters a 
 guide-book to your studies in general. Such a guide-book, 
 even if better done than I can do it, might be profitably 
 superseded in ten years' time by a new manual, telling of 
 new books. What I want to do is to leave you some 
 suggestions which I think may be of use to you whatever 
 your other studies or pursuits may be. 
 
 I am not concerned to think of you as experts in any one 
 science (though it is well worth while to be that), or in 
 philology, or in archaeology, or even in music, or Greek, or 
 Shakespeare. A clever logician. Professor de Morgan, has 
 well said that it is good to know everything of something, 
 and something of everything ; and though I agree with him 
 (it being understood, of course, that such counsels of 
 perfection can never be fulfilled), I shall not try to choose 
 your something for you. But whether you lean towards the 
 second or towards the first of those splendid impossibles, I
 
 Letter I. LETTERS ON REASONING 3 
 
 trust you will both be good reasoners. I want you, Guenn, 
 as well as you, Guy, to regard this as specially well worth 
 your while. If I read your little " bumps " aright, you will 
 not be equally devoted to study ; but in so far as our 
 reasoning powers can be improved by exercise and dis- 
 cipline, there is no good reason why women should not 
 undergo it as well as men, and as successfully. 
 
 When you grow up there will still, I fear, be people who 
 think that a woman should not be reasonable : indeed, you 
 are both likely to hear it said that " reason " is a " cold " or 
 otherwise repellent faculty, which needs checking rather 
 than encouraging. Such sayings are their own sufficient 
 refutation. If you will weigh them with me, you will have 
 taken one practical step in reasoning. 
 
 Only the other day I heard an educated man argue that 
 much harm had been done in politics by the cultivation of 
 and the appeal to reason. When it was suggested to him 
 that his own remark was meant to be a piece of reasoning, 
 an appeal to reason, he sought to make his case clearer by 
 saying he meant " logic " ; and when asked whether he did 
 not want to be logical he said he had in view " formal logic," 
 and that by formal logic he understood an entire disregard 
 of feeling and emotion. This was all a verbal confusion on 
 his own part. " Formal logic " is a well-understood term 
 for a quasi-mathematical or algebraic way of handling 
 questions of abstract logic, which never arise in political 
 discussion at all. This was not what was in his mind. 
 What he ought to have said, to express his own thought, 
 was that harm may be done in politics by assuming, what 
 is not true, that men in the mass are able to appreciate and 
 to apply a highly reasonable system of government. A 
 thinker who frames a theory of government on grounds 
 of abstract justice is certainly "appealing to reason"; but it 
 is not "reason" or good reasoning that makes him overlook 
 the incapacity of many men to live justly by the light of 
 reason. In forgetting to take such a fact into account he 
 
 c
 
 4 LETTERS ON REASONING Letter I. 
 
 has failed to reason with due care. To realise his over- 
 sight is an act of reason. Unfortunately some, in vaguely 
 realising the oversight, fall into the worse absurdity of 
 saying that we had better reason less. 
 
 That very remark is a blundering appeal to reason. All 
 argument, every attempt to influence opinion or conduct by 
 presenting a "because," is a process of reasoning. In 
 studying religious questions you are likely to read, or to 
 hear, the kind of protest I have just glanced at ; and if you 
 have not learned to think clearly you may be confused by 
 it. Great writers — and many who are not great — have given 
 forth the bad sophism that since our reason is fallible, since 
 we are liable to make mistakes, we should cease to reason. 
 As if that very formula were not simply a self-confounding 
 appeal to reason, an attempt to persuade by a " because." 
 If we were really convinced that our judgment— which is 
 only another way of naming our reason — is quite untrust- 
 worthy, we should have reached that very conclusion (which, 
 if convinced, we should trust) by reasoning; and if we 
 should then propose either to stop judging altogether or to 
 accept thenceforth whatever any particular teacher might 
 tell us, we should only stultify ourselves ; for to decide to 
 stop judging is to judge that we ought to stop, and to accept 
 another's judgment is to judge that it is acceptable. 
 
 I deal with this verbal problem at the outset of these 
 letters because it is typical of many of the confusions that 
 will meet you even in argumentative literature, to say 
 nothing of ordinary table-talk. One of the most pretentious 
 of the serious books published in my time— Mr. Benjamin 
 Kidd's Soda/ Evolution— \^ a systematic application of the 
 absurd species of argument we have been considering. He 
 says to his readers, in effect : " I reason thus : reasoning 
 ( = reason) is a dangerous guide : let us then reasonably 
 decide not to be reasonable, and so by reasoning escape 
 the evil consequences of all reasoning." And his book 
 has had a great sale, for the " reason " that it flatters the
 
 Letter I. LETTERS ON REASONING 5 
 
 folly of many religious people who like to hear " reason " 
 disparaged. 
 
 Seeing that such books still gratify many educated people, 
 I cannot hope that when you are growing to maturity their 
 ideas will not come in your way. What I wish is that you 
 shall be able to judge them intelligently, and to dismiss 
 them with decision. I want you therefore to grasp first the 
 truth that a// attempts to persuade are processes of reasoning. 
 Some, we say, are good or "logical" or "valid"; by which 
 we mean that on analysis their parts or stages are con- 
 sistent. A good reasoner is one who does not contradict 
 himself in the course of his argument, and who further 
 takes intelligent account of all the important facts of the 
 case he is dealing with. A " bad reasoner " is one who, in 
 seeking to prove or to convince, takes up (like Mr. Kidd) 
 contradictory positions, whether or not he has the main 
 facts of his case before him. It is probable that the 
 difference between a very bad and a very good reasoner 
 is on a par with that between an ugly and a beautiful 
 person, or that between a " colour-blind " person and one of 
 normal vision — something irreparable, something fundamen- 
 tally unalterable. Happily, however, there are countless 
 degrees between extreme ugliness and perfect loveliness, 
 and utter uncomeliness is as rare as flawless beauty. And 
 as the plain face may become lovely in virtue of goodness 
 looking from the eyes, or be pleasing through its perfect 
 health, so a mind little gifted with quickness or clearness of 
 insight may through candour and careful exercise become 
 competent for all the normal tasks of judgment. 
 
 To fail of such competence, and yet to put forth judg- 
 ments freely in daily intercourse, is to be a nuisance to 
 more intelligent people, though the incompetent pass muster 
 well enough in each other's society — until, that is, they 
 dispute over some question which stirs their passions. Then 
 they are apt to disesteem each other with virulence. In 
 ordinary society, there is ready and severe criticism of 
 
 c z
 
 6 LETTERS ON REASONING Letter L 
 
 slovenliness in dress, unpleasant habits, ill-breeding in the 
 matter of small talk or offensive gossip, or an uneducated 
 pronunciation ; but for incoherent reasoning as such there 
 is the tolerance of a common laxity. I want you nevertheless 
 to take as much care about your opinions as about your 
 clothes and your bodies and your manners, and so to be fit 
 for the true " good society," that of cultured and thoughtful 
 men and women. 
 
 It is, of course, part of the law of mutual forbearance that 
 we should bear \vith bad reasoning when it meets us in 
 intercourse, especially if it come from unaffectedly unwise 
 persons. But if we think long and seriously about human 
 affairs we can hardly fail to see that endless evil comes of 
 the failure of most people even to aim at consistency in 
 thought and action. Consistency in thought is the gist of 
 right thinking, of good reasoning ; and consistency in action 
 is the gist of right conduct. All reasoning, all morality 
 works down to that test ; and all deflection from reason- 
 ableness and righteousness may be expressed in terms of 
 inconsistency. Now, conduct depends finally upon thought, 
 and it seems clear that, though a habit of reasoning soundly 
 on non-moral problems will not necessarily develop scrupu- 
 lous thinking on moral problems, the habit of aiming at 
 consistency in one's reasoning in general will help one to 
 check inconsistency in one's practice when one is concerned 
 to do so. 
 
 On the other hand, it is hard to conceive that a habitually 
 slovenly or inconsistent reasoner can be highly consistent in 
 conduct, though he keep strictly within the bounds of 
 criminal law. It is quite true that a not very clear-headed 
 person may have strong sympathetic or unselfish instincts, 
 while a clear-headed one may not ; but if the former is 
 habitually consistent in normal conduct it must be because 
 on that side he is clear-headed, or at least anxiously reflective 
 and it is historically certain that normally conscientious 
 people may, for lack of the habit of consistent reasoning,
 
 Letter I. LETTERS ON REASONING 7 
 
 act with gross inconsistency in a new relation, where 
 moral habit does not guide them. The naturally selfish or 
 unsympathetic person, again, cannot be made worse by the 
 habit of scrupulous reasoning, and he may be very much 
 improved by it on the moral side. I take it, then, that to 
 be a good reasoner is a duty only less pressing than that of 
 being a good citizen, a good man or woman ; and my ideal 
 for you is that you should so regard the intellectual life. 
 
 All honourable people, I suppose, realise that to act 
 rightly we must habitually seek consistency : the law of 
 " doing as we would be done by " is the normal summary 
 of the moral code. Yet even of those who avowedly 
 accept that code, many are lax in applying it ; and this I 
 suspect to be explicable in terms of the fact that it is com- 
 paratively rare to realise at all the bearing of the principle 
 of consisteiicy on processes of thought. You see I say 
 "explicable in terms of" rather than "due to," because 
 I do not wish to prejudge here the question whether it is a 
 moral impulse that moves us to reason more scientifically 
 or a logical perception of inconsistency that moves us to 
 be more honest. That is a very interesting problem, 
 whether you consider it as one of metaphysics or as one 
 of psychology ; and I hope you will be able to take an 
 interest in metaphysics and psychology, were it only to fit 
 you to defend yourselves against bad metaphysicians. But 
 I do not want to load these letters with special problems : 
 I want to help you, if I can, to develop your general powers 
 of judgment by way of the exercise of reasoning. And as 
 I have a conviction that, in respect of certain natural 
 tendencies which hinder us, we all have more or less 
 difficulty in realising the need for consistency in our 
 reasoning processes, I wish to stress the importance of 
 the connection between the so-called " intellectual " and 
 the moral processes. 
 
 To me it serves as a measure of both the moral and 
 the intellectual shortcomings of mankind, thus far, that
 
 8 LETTERS ON REASONING Letter L 
 
 intellectual consistency is so little regarded. I should put 
 it that such inconsistency is intellectual immorality ; and I 
 can hope for you nothing better than that you may find 
 yourselves at middle life in a world in which such immorality 
 is more clearly realised and more seriously resisted than it 
 is in my day. In that case you will live among better and 
 wiser people, generally speaking, than your predecessors. 
 
 You see, then, that in begging you to be thoughtful and 
 careful reasoners I am urging on you not a mere intellectual 
 gymnastic, much less a habit of quibbling and wanton 
 disputation, but a vital self-culture that shall evolve and 
 elevate your personalities. I am not prescribing, you will 
 find, a study of logic in the technical sense of that term. I 
 do indeed hope that, whether before or after you read these 
 letters, you will study some treatises on logic. The works 
 of such men as John Mill, Professor Bain, Professor Minto, 
 and Mr. Bradley are full of instructive thought as well as of 
 gymnastic discipline. But logic is the name we give to the 
 body of analysis of the process of reasoning in general, and 
 is rather a general formulation of the rules of all proof and 
 reasoned belief than a training in correct reasoning. 
 Technical logic is to actual reasoning very much what 
 grammar is to language : that is to say, its rules are 
 abstracted from study of the processes of careful and tested 
 reasoning, as grammar is primarily a statement of rules seen 
 to prevail in the speech of educated people. And as the 
 way we learn to speak and write correctly is just by speaking 
 as educated people speak and write, with formulated 
 grammar to remind us and to fix our memories, so the way 
 we learn to reason correctly is by following or checking 
 arguments, with logic to fix for us in condensed forms our 
 recognition of the nature of all processes of argument. 
 
 If you should learn any foreign language from books, you 
 will find that even to get by heart the rules of its grammar 
 is a very imperfect way of mastering grammatical speech, as 
 compared with the method of following actual speech and
 
 Letter I. LETTERS ON REASONING 9 
 
 written style. So I believe it to be with logic, in the 
 technical sense. Of course we often use the term logic to 
 signify simply consistent reasoning, and the term logical to 
 signify " coherent, consistent, reasonable." In that sense I 
 want you to be logical for the reasons I have already given. 
 Those who disparage logicality in this sense are either 
 thoughtful but confused people who fall into fallacy like the 
 reasoner I mentioned above, or people ill-fitted for thinking. 
 To despise or to fear logicality of mind is as foolish or 
 mistaken as to despise or fear truthfulness. 
 
 But even in the technical sense of the term, remember, 
 logic is not an ignoring or stunting of the processes or 
 forces of emotion and imagination. Many common phrases 
 on the subject are bad fallacies. When Macaulay in his 
 History remarks that " Logic admits of no compromise : the 
 essence of politics is compromise : it is therefore not strange 
 that some of the most important and most useful political 
 instruments in the world should be among the most illogical 
 compositions that ever were penned," he is creating a con- 
 fusion similar to that I have discussed above. To say that 
 logic does not admit of compromise is about as useful as to 
 say that it does not admit of bicycling or making love. 
 Compromise is an action or an agreement to act. Logic, 
 as I have said, is the analysis and correct statement of the 
 processes of proof — proof of propositions. Now, in a legal 
 document, such as a deed of sale, a declaration of the 
 change or establishment of a dynasty, or a will, we are not 
 dealing with a process of proof: we are setting forth an 
 agreement of certain persons to act in a given way, or a 
 stipulation that certain persons, specially appointed, shall 
 do certain things. If, accordingly, a testator disposes of his 
 property in a way that we think foolish, his will is not there- 
 fore to be described as illogical. If it secures its end, it is 
 practically " logical." 
 
 So with a parliamentary declaration. What Macaulay 
 had in view was the declaration of the English Parliament
 
 lo LETTERS ON REASONING Letter I. 
 
 declaring the throne vacant after the revolution of 1688 : a 
 statement embodying the various grounds on which different 
 politicians were willing to make a change in the monarchy. 
 Now, the " logical " course in such a matter was just to set 
 forth these grounds in such a way as to secure the great 
 majority of votes : the question was not one of proving a 
 truth, but one of getting men to pledge themselves to a 
 certain course, to enact a certain law. The framers of the 
 resolution would have been "illogical" indeed if they had 
 so framed it that it was offensive to the majority. They 
 were " logical " in the sense that they carried out their 
 practical purpose in a prudent way. Macaulay is talking 
 at random when he says : " They cared little whether their 
 major agreed with their conclusion, if the major secured 
 two hundred votes, and the conclusion two hundred more." 
 There was no major, no minor, and no syllogistic conclusion 
 in the rnatter ; they were not maintaining a proposition in 
 the logical sense of the phrase ; they were drawing up a 
 " motion to be carried." For them, the sole " proposition 
 to be proved " was this : " We shall best carry our point — 
 the crowning of William — by moving a resolution formally 
 expressing the reasons for which the largest number of 
 members of the legislature are willing to crown him." Of 
 such a proposition the desired " proof" could be given only 
 by the actual voting. The business in hand was not to 
 prove a truth, but to please a number of men, and forecast 
 their action. 
 
 To call the document " illogical " is merely to show mis- 
 conception of what " logic " means. If you will read the 
 arguments of Lord Somers on the wording of the resolution, 
 you will see that he was a very clear and logical — that is, a 
 consistent — reasoner; and his reasonings may be reduced 
 to a series of syllogisms by anyone who cares to take the 
 trouble. Macaulay himself glimpsed the nature of the 
 case when, after citing the phrases incriminating the deposed 
 king, he wrote ; " Such words are to be considered, not as
 
 Letter I. LETTERS ON REASONING II 
 
 words, but as deeds." Quite so — though it would be better 
 to read " not as arguments " : they are not an attempt to 
 prove an assertion ; they are the avowal of an agreement 
 as to a line of action ; and it is a gratuitous sophism 
 to say of them, " they are rational though they may be 
 contradictory." They are not contradictory. It is open to 
 anyone, certainly, to argue that they proceed upon, and are 
 an attempt to conserve in theory, a superstitious practice — 
 the practice of hereditary monarchy. The current theory 
 of the British monarchy is no doubt illogical in the sense 
 of being inconsistent and incoherent. It is easy to show 
 that men hold by monarchy either because they are them- 
 selves unreasonable or because they distrust the capacity of 
 their fellows to act justly and reasonably in politics beyond 
 a certain point. It would not be hard to prove that those 
 men are either illogical or insincere who profess to believe 
 in the indefeasible right of a community to choose its rulers 
 and also in the indefeasible right of monarchs to rule through 
 hereditary descent. Broadly speaking, the English nation 
 in 1688 was mainly made up of muddle-headed people. It 
 is so now. But it does not follow that the formula in 
 question is a contradiction in terms, or even that it contra- 
 dicted the (unreasonable) standing principle that the blame 
 for the king's errors lies on his counsellors, or even that it 
 was framed by muddle-headed or superstitious men. It 
 may have been framed by men who were themselves above 
 the monarchic superstition, and capable of living judiciously 
 under a republican system, but convinced of the need for 
 humouring and managing most of their superstitious fellow- 
 citizens by means of arbitrary compromises between monarchy 
 and self-government. 
 
 Such a procedure, so far as it goes, is in terms of "reason" 
 — reason on a fairly low plane. It remains perfectly open 
 to us, after saying all this, to try to bring the majority to a 
 nobler course, to appeal to them to live by " reason " in the 
 sense of learning to do away with monarchy as a degrading
 
 12 LETTERS ON REASONING Letter I. 
 
 institution, suited only to ignoble or superstitious minds. 
 In making such an appeal, we should make all the use we 
 can of logic : that is to say, we should show how incon- 
 sistent with the behef in democracy is the belief that 
 monarchy is in itself a desirable thing. But a large part 
 of our procedure — perhaps the most effective — would consist 
 in showing our fellows that their monarchism is on a different 
 moral plane from their democratism — in a different spiritual 
 key, so to speak, from their statement of their other political 
 ideals. And this also would be reasoning. That is to say, 
 our reason, instead of failing to take account of emotions, 
 prejudices, ideals, would be occupied with them, would 
 proceed upon them, would be at work in correcting their 
 inconsistencies. And inasmuch as we reasoned consistently 
 we should be reasoning, in the practical sense of the term, 
 logically. 
 
 On the other hand, the phenomena or forces of emotion 
 and imagination may as well as any other forces be subject- 
 matter of logical propositions ; and if, in any argument 
 which claims to trace and explain a process of social or 
 personal causation, the actual play of emotion and imagina- 
 tion in all such processes be overlooked, the argument is so 
 far fallacious. Further, a logical mind may easily be more 
 emotional and imaginative than an illogical. The love of 
 truth and rectitude is an emotion ; the recognition of a 
 good argument is a solidly pleasurable feeling ; and it is an 
 observable fact that the habit of self-criticism, to the end of 
 attaining consistency, can expand imagination by widening 
 sympathy. People who cannot think in an orderly way 
 are fain to suppose themselves more imaginative than 
 others ; but though some kinds of imagination may 
 flourish in the inverse ratio of the usage of consistent 
 reasoning, some of the highest kinds of all actually arise 
 from it. 
 
 It is a bad mistake to credit great "imagination" to 
 peasants who explain puzzling facts by stories of fairies and
 
 Letter I. LETTERS ON REASONING 13 
 
 fiends, and to deny it to the mathematic mind of Newton, 
 " voyaging through strange seas of thought alone." Never, 
 I beg you, let yourselves be browbeaten by people who tell 
 you that to cultivate your reason is to lose the faculty for 
 enjoying poetry, music, or any other art. The truth lies the 
 other way. Exercise of the reason may indeed raise you 
 above some kinds of enjoyment that appealed to your 
 untrained mind, but it will rather enlarge your faculty for 
 enjoying greater art, by widening, so to speak, the range of 
 the vibration of your feelings. 
 
 This, of course, holds true of many forms of special study 
 as well as of the habitual use of the reflective reason, because 
 every study involves processes of reasoning. But I have 
 often observed that the study of a special subject or science 
 may leave a man awkward in the use of his reason on other 
 subjects, even those which more or less obviously concern 
 all of us. And it is because, in particular, I do not think 
 manuals of logic the best exercises for developing the 
 reasoning powers that I have set about writing these letters. 
 The late Mr. Jevons tells, in the preface to his Studies in 
 Deductive Logic, that he has " often been astonished at the 
 way in which even well-trained students," prepared only by 
 mathematics, " break do\\-n before a simple logical problem. 
 A man who is very ready at integration begins to hesitate 
 and flounder when he is asked such a simple question as 
 the following : ' If all triangles are plane figures, what 
 information, if any, does this proposition give us concerning 
 things which are not triangles?'" I readily accept Mr. 
 Jevons's testimony as to the insufticiency of mathematics to 
 make us good reasoners in other fields of thought ; but I 
 doubt whether the kind of inefficiency of which he has 
 given a sample would not be more easily cured by general 
 exercise in reasoning followed by a study of logic, than by 
 making logical exercises a primary means of training the 
 reason. An athlete, I suspect, can better be trained by 
 free exercise in play, followed by special gymnastic, than
 
 14 LETTERS ON REASONING Letter I. 
 
 by taking him with undeveloped physique and setting him 
 at once to special exercises. 
 
 In my copy of Mr. Jevons's Studies, bought at second- 
 hand, some previous reader had written in pencil, under 
 the passage above quoted, these words : " Some plane figures 
 are not triangles." That is his answer to the question which 
 Mr. Jevons says has puzzled clever mathematical students. 
 Now, you Avill see that this answer is a blunder. The 
 proposition cited in the question above quoted gives no 
 information as to things which are not triangles. No doubt, 
 if you heard such a proposition you might, without any 
 further information about plane figures, surmise that things 
 not triangles could also be plane figures ; but you do not 
 get that fact as information from the proposition under 
 notice. The pencilled answer I have quoted happens to 
 be a true proposition \ but it is not here a properly drawn 
 logical inference. 
 
 Technical or deductive logic, you \vill observe, deals thus 
 strictly with the content of propositions, taking no account 
 of merely possible implications ; and as Mr. Minto remarks, 
 "nothing has contributed more to bring upon it the 
 reproach of quibbling." I do not for a moment join in 
 that reproach. The resort to quibbling should bring 
 reproach not on the exactitude of strict logic, but on the 
 person who turns it to the account of quibbling. But, 
 inasmuch as the exercises of technical logic involve much 
 artificial verbal distinguishment, reduced to symbolic 
 unreality, I doubt whether they prepare our minds to draw 
 the material distinctions which are of the essence of right 
 thinking in regard to the living problems of belief and 
 action. On the other hand, the free use, so to speak, of 
 your limbs in living reasoning will prepare you as well as 
 may be for the small feats of technical logic, which discipline 
 you can undergo all the more easily when so prepared. 
 
 To return, finally, to the question of the value of good 
 habits of reasoning. I would have you realise very clearly
 
 Letter I. LETTERS ON REASONING 15 
 
 that all discussion, all criticism, whether wise or unwise, is 
 reasoning. The blunderers who warn you against reason 
 are simply bad or temporarily confused reasoners. There 
 is no getting away from reasoning, save by way of insanity ; 
 and insanity itself simulates the process of reasoning. The 
 common use of the word " irrational " is thus lax, though 
 convenient. We may indeed fitly describe as irrationalists 
 the people who cry down reason ; but their fallacy is none 
 the less an attempt to reason. Still more clearly do those 
 appeal to reason who warn you that " man does not live 
 by reason alone," or that close devotion to processes of 
 reasoning may weaken the emotional and imaginative side 
 of your nature. 
 
 Such warnings, as they stand, are simply counsels of 
 mental hygiene. No one ever disputed that in order to 
 live a full and rounded life we should cultivate our aesthetic 
 faculties, the life of the eye and ear and the conscience 
 and the imagination. Travel and living observation and 
 friendly intercourse are all capital elements in a cultured 
 life ; and poetry is as it were the flower of all literature, all 
 experience. But the givers of such counsels, when they 
 begin to disparage reason, are themselves doing on that side 
 what they warn you against doing on the other. Too often 
 they have cultivated only the aesthetic faculties ; and we 
 find them disposing of the deepest intellectual problems by 
 purely aesthetic tests, applauding a fallacious theory or a 
 false narrative merely because it pleases their imagination 
 or comforts their feelings. Disparaging reason, they are 
 really perverting and corrupting it in their own case. Such 
 fallacy-mongers are sometimes very amiable and charming 
 people ; but in the intellectual life, as I have already put it, 
 they are nuisances, by reason of the ignorant confidence 
 with which they darken counsel. 
 
 It would mortify me to believe that you will grow up 
 members of that fallacious tribe. You, my dear Guenn, 
 must not let yourself feel that because at any time social
 
 16 LETTERS ON REASONING Letter L 
 
 usage or your immediate circle does not urge you to use 
 your brains scrupulously you do well to be satisfied with 
 hand-to-mouth thinking. I hope that you will never be 
 swayed by the judgments of either men or women who do 
 not like that a woman should be wise or instructed. Your 
 tact, I trust, will always save you from being pedantic, or 
 from obtruding unseasonably the process of reasoning. 
 But you, my dear Guy, owe just the same consideration to 
 the laws of pleasant intercourse : tact and adaptability 
 should not be left to women any more than sound and 
 scrupulous reasoning should be left to men. It pleases me 
 to hope that you will so grow up as to be comrades both 
 socially and intellectually ; that if either of you lags 
 mentally the other will encourage and help, and that if 
 either tends to be lacking in tact or amenity the other will 
 affectionately admonish.
 
 LETTER II. 
 
 In the previous letter I remarked that a careful process of 
 reasoning may be left practically useless or misleading 
 because of the reasoner's failure to note some essential 
 facts or data. It would not, perhaps, be too much to say 
 that the majority of men's errors of opinion proceed rather 
 from wrong or insufficient information than from fallacy of 
 inference. When I reflect on the nature of great masses of 
 the didactic literature of the past, now admittedly super- 
 seded, it seems to me to wear a certain uniform air of 
 confidence in error ; and while the confidence has to be 
 set down as the besetting sin of humanity, the error comes 
 at least as often from wrong testimony, traditional or other- 
 wise, as from illicit reasoning. Education of the reason, 
 then, is likely to involve no less a vigilant scrutiny of 
 testimony than a scrupulous regard to inference. 
 
 I cannot, indeed, too strongly press upon you the 
 importance of being slow to believe where your decision 
 will have any width of bearing. Here again you will be 
 met by loud protests from a certain order of minds. Doubt, 
 you will often be told, is something " cold " and " dark." I 
 do not much value argument by metaphor ; but it might 
 perhaps benefit such metaphor-mongers to ask themselves 
 whether in the order of nature we are not daily refreshed 
 by darkness and yearly braced by cold. It would not be 
 unjust, I think, to say of them that their fear of the " cold " 
 is like that of the nerveless bather who recoils from the 
 plunge, and their fear of the " dark " like that felt by or 
 instilled into many children — I say " instilled into," because 
 in your own case your mother and I have thus far succeeded
 
 i8 LETTERS ON REASONING Letter IL 
 
 in preventing its being suggested to you, and neither of you 
 has ever shown the slightest sign of feeling it. In the old 
 fashion I would say, May this be a good omen. 
 
 The censure of doubt is in fact only an aspect of the 
 censure of reason; for doubt is the beginning of reason. 
 In other words, it is the beginning of wisdom. If I were 
 bent on meeting a priori speculations with a priori specula- 
 tions, I could make some play, I think, mth the formula 
 that, since the whole of intellectual evolution as we know it 
 is through doubt, the " purpose " of the process seems to be 
 to evoke doubt. Consider the myriads of animals that 
 must have perished for lack of doubt about the fitness of 
 certain things for food, the safeness of certain places, the 
 possible nearness of dangerous enemies. Most animals, I 
 should say, are now doubters upon instinct, their species 
 having survived by dint of doubting. The trout passing the 
 hook ; the horse swerving from a strange object ; the sheep, 
 even in captivity, scurrying from one who approaches them ; 
 the mouse darting towards its hole at a sound ; the dog 
 barking at a strange footstep ; the deer flying at the scent 
 of some possible enemy far away — do they not all exemplify 
 an " economy of doubt^" in nature ? 
 
 As for man, every step he has made in civilisation has 
 been taken in virtue of either doubt or the doubt-involving 
 substitution of a new belief for an old ; and every such step, 
 depend upon it, has been resisted by experienced people 
 who denounced criticism as their type to-day denounces 
 doubt and reason. It has always been common for theo- 
 logians to denounce, not merely wrong belief, but " unbelief." 
 We shall consider in another letter the temper in which they 
 speak : let us here note their formal inconsistency— their 
 want of logic. They themselves are just as much "un- 
 behevers " as anybody else. Their creed took its rise by 
 way of disbelief in another creed ; their first teachers were 
 unbelievers. Right and wrong beliefs alike involve dis- 
 belief; every new religion negates in whole or in part
 
 Letter II. LETTERS ON REASONING 19 
 
 previous religions ; and the process from one belief to 
 another which negates it is either an insane spasm of 
 emotion or a process of doubt. 
 
 Each believer in turn disbelieves the doctrine which 
 contradicts him ; and just as the Christians spoke of the 
 "unbelieving Moslems," so the Moslems spoke of the 
 "infidel Christians." To-day the religionist who denounces 
 " unbelief " is an unbeliever in the Naturalism which rejects 
 his faith ; and in every generation the so-called " believers " 
 have been the most obstinate unbelievers in new truth. 
 AVhen they are on the way to a reluctant assimilation of 
 some of the truths which they and their predecessors had 
 denied, they are doubters. 
 
 It is implied in all this that we need to be on our guard 
 against errors both of credulity and of incredulity; but 
 inasmuch as uncritical belief to start with commits us in 
 a measure to disbelief in new lore Avhich contradicts that 
 already held, it is on the side of over-ready confidence that 
 the stress of danger lies. I have advised you to repel the 
 self-stultifying counsel " Do not trust to reason," but I urge 
 you strongly " Do not lightly trust to statements of fact." 
 Guard against the risks of wrong reasoning, not by reason- 
 ing less, but by reasoning more. That is, demand evidence 
 or argument not merely for new doctrine, but for old. The 
 kind of thinker who denounces "unbelief" is himself 
 extremely hard of belief in the new doctrine ; but the 
 intelligently rational course is to be equally critical of all. 
 As Professor Bain insists, there is in our nature a primary 
 and powerful bent to credulity ; but with the growth of 
 reason there is evolved an equally powerful tendency to 
 reject whatever clearly contradicts the belief in possession. 
 The first belief blocks the way ; and to cure this evil we 
 had need acquire, first, the habit of inquiring before we 
 believe to begin with ; and secondly a willingness to 
 reconsider, on challenge, even a belief taken up after in- 
 vestigation. iVnd such habits are hard to set up. Mankind 
 
 D
 
 20 LETTERS ON REASONING Letter II. 
 
 has learned to doubt even as it has learned to follow the 
 sequences of nature, to control forces, to heal diseases, to 
 manage social affliirs. The faculty for doubt develops with 
 the range of knowledge ; and it is still the general rule that 
 where we have imperfect knowledge we tend to believe too 
 readily unless our previous beliefs clearly negate the pro- 
 position in question. 
 
 Let me give you some instances. Professor Minto, one 
 of the logicians I have praised above, early won my admira- 
 tion by the sagacity with which in his writings he detected 
 current errors of statement or narrative on various subjects 
 which he had to investigate. His book on Logic : Dedndive 
 and Inductive is, I think, the best introduction you could 
 have to technical logic, so lucid is it, so fresh and alive in 
 its handling. But now and then, though as I have said he 
 had a gift for investigation, even he is unduly compliant 
 in his acceptance of statements of historic fact. Thus, 
 writing about " the coincidence that has been remarked 
 between race and different forms of Christianity in Europe," 
 he says : — 
 
 " If the distribution of systems were entirely independent of race, it 
 might be said that you would expect one system to coincide equally 
 often with different races in proportion to the positive number of their 
 communities. But the Greek system is found almost solely among 
 Slavonic peoples, the Roman among the Celtic, and the Protestant 
 among Teutonic. The coincidence is greater than chance will account 
 for. Is the explanation then to be found in some special adaptability 
 of the religious system to the character of the people? This may be 
 the right explanation, but we have not proved it by merely discounting 
 chance. To prove this we must show that there was no other cause at 
 work ; that character was the only operative condition in the choice of 
 system ; that political combinations, for example, had nothing to do 
 with it. The presumption from extra-casual coincidence is only that 
 there is a special cause : in determining what that is, we must conform 
 to the ordinary conditions of explanation."^ 
 
 Here, in the very act of ostensibly warning us against a 
 too hasty inference, Mr. Minto has actually made that 
 1 Work cited, pp. 359, 360.
 
 Letter ri. LETTERS ON REASONING 21 
 
 inference, and his warning is consequently a contradiction 
 in terms. And the whole miscarriage, I think, has arisen 
 through his too ready acceptance of a wrong statement of 
 fact. The importaiit point for us at present is this mistaken 
 acceptance ; but if you will take the trouble to follow with 
 me the miscarriage in the argument you will perhaps realise 
 more vividly how important the factual error is. 
 
 I have italicised, you see, the sentence about the co- 
 incidence being greater than chance will account for. Now, 
 that is, to begin with, an oddly loose phrase for Mr. Minto. 
 It is strictly meaningless to say that chance can " account 
 for " anything. Chance, as I shall try to show at length in 
 another letter, is the name we give to untraced or Jintrace- 
 able causation of events or coincidences. It will here suffice, 
 however, to note that while Mr. Minto in effect says : 
 " These coincidences of phenomena are so nearly invariable 
 that there must be a causal connection between them " — 
 that is, between race and racial creed — he yet goes on to 
 say, in effect, that " political combinations " might con- 
 ceivably have something " to do with " the distribution of 
 creeds on racial lines. Then, unless the name " political 
 combinations " includes the idea of " race character " (iia 
 which case both phrases are reduced to insignificance), it 
 follows that there may be no causal connection whatever 
 between race and creed. 
 
 The confusion is complete ; and I think we can trace it 
 all to the acceptance of the error of fact. Religion and 
 race do «(?/ coincide, as is alleged, i. The Greek system 
 is not confined to Slavonic peoples : the Greeks even now 
 are certainly not all ;^Slavonic ; and some " Slavonic " popu- 
 lations either never were or have ceased to be Greek 
 Christians. 2. The Roman system flourishes to-day in 
 great sections of the German-speaking peoples, who pass 
 for " Teutonic " ; also in various Teutonic cantons of 
 Switzerland, including the oldest ; also among Flemings, 
 who equally rank as Teutonic. 3. The Protestant system 
 
 D 2
 
 22 LETTERS ON REASONING Letter IL 
 
 flourishes among the " Celtic " peoples of the Scotch High- 
 lands, Wales, and Cornwall ; it prevails in several French- 
 speaking cantons of Switzerland ; and it did live vigorously 
 for generations in France. If the people of France and 
 French-speaking Switzerland are not to be reckoned as 
 " Celtic," then many of them are — in the terms of the 
 division subsumed — Roman Catholic Teutons. 
 
 I do not here concern myself with the question as to 
 what races are properly to be called Celtic or Teutonic, and 
 why : on that problem, as well as on the racial distribution 
 of Catholicism and Protestantism, I have written at length 
 in other books. I simply proceed on Mr. Minto's own 
 terminology and classification. Taking these for granted, 
 we find that he is astray in his facts : there is a very frequent 
 crossing of the lines of race and creed. And it was pro- 
 bably a suspicion of his own error that led him to go on to 
 suggest that political combinations may have determined 
 the distribution of church systems. This, I have else- 
 where undertaken to show, is really the case. Yet, not 
 liaving eradicated his historical error, he leaves standing, 
 in his sentence about chance, the theorem usually bracketed 
 with that error. (I say bracketed with, because I suspect 
 that the theorem partly preceded the formulated error ; men 
 reasoning thus : " I, a Teuton, am Protestant ; he, a Celt, 
 is Catholic ; doubtless all Teutons are Protestants, and all 
 Celts Catholics ; indeed, such is the fact.") Finally, he is 
 committed to suggesting that what he affirms to be an 
 evidently causal coincidence between race and church- 
 system may not be such a coincidence at all. Had he with 
 sufficient energy doubted the pseudo-historical statement, 
 the train of confused reasoning would not have arisen.^ 
 
 Now, if such a man as Mr. Minto could thus miscarry, 
 you and I may be pretty sure that we can. Seeking to be 
 
 ' I ought to note that Trofessor Minto hail completed his book on 
 I,()gic only a short lijiie before his death, and after years of worsening 
 ill-health.
 
 Letter II. LETTERS 0\ REASONING 23 
 
 on our guard, \vc may ask ourselves whether he was 
 originally misled by force of a survival of either religious 
 or racial bias ; whether, that is, he held himself for a 
 " Teuton," and reckoned Protestantism a superior form of 
 church polity to Catholicism — as he possibly might, even 
 after ceasing to be a churchman at all. On that hypothesis 
 we shall presumably do well to ask ourselves, when we are 
 disposed to accept without anxious investigation any im- 
 portant generalisation, whether it flatters in any way what- 
 ever our self-esteem. 
 
 I recall a similar case of historical error on the part of 
 another very able man — the late Professor W. K. Clifford. 
 That brilliant thinker has asserted that " the Teutonic con- 
 science protested in the Reformation " against the reference 
 of right and wrong to the arbitrary will of deity as a 
 standard.^ In this passage the Professor is seeking to build 
 up a rational moral standard which shall be more scrupulous 
 than others; and in his essay on "The Ethics of Belief" 
 he becomes still more stringent in his prescriptions for the 
 intellectual life. Again and again he insists on the harm 
 done by accepting current assertions without strict examina- 
 tion, on the danger to society of such careless credulity, on 
 the sacred obligation of inquiry before belief. ^Vell, in the 
 passage above cited he has without any inquiry echoed 
 and endorsed a vulgar error ; an error, too, which specially 
 merits his own censure in that it ministers to racial pride 
 and malice. There was nothing specially " Teutonic " in 
 the revolt of "conscience" against false morals at the 
 Reformation : such a revolt had been made a hundred 
 times before by non-Teutons ; and if we closely study the 
 history of the Reformation we shall find that it took rather 
 more " conscientious " lines in France than in Germany ; 
 the German movement being, by the admission of Luther, 
 in many respects far from conscientious. Nay, Luther 
 
 ^ Essay on " Right and Wrong," in Lectures and Essays, ed. 1886, 
 P- 335-
 
 24 LETTERS ON REASONING Letter IL 
 
 himself actually framed a theological doctrine of morals 
 which realises the very conception that Professor Clifford 
 denounces; and the "Teutonic" Locke, in ascribing the 
 nature of right and wrong to divine volition, did the same 
 thing with a difference. 
 
 Thus our moralist has been careless where he incul- 
 cates care ; and his carelessness certainly justifies his own 
 warning, for by endorsing that falsism about the " Teutonic 
 conscience " he encourages those of his readers who 
 suppose themselves Teutons to believe that they have a 
 better sort of conscience than members of other races, and 
 to look on other races, accordingly, through a perverting 
 conceit, which easily evolves a pernicious ill-will. Once 
 more, if such an acute and original thinker can fall into 
 such inconsistency, all through an unthinking acceptance of 
 a familiar formula, you and I may so err. Let me urge on 
 you then to form the habit, which has only very gradually 
 become normal with me, of asking yourself, over every such 
 use of a far-reaching generalisation, whether it stands for 
 tested knowledge, whether it is not merely a shibboleth of 
 national or racial vanity, or at least an imperfect formula- 
 tion of a complex mass of facts, caught at by most people 
 as a labour-saving expression. 
 
 Science progresses by the scrutiny of such imperfect 
 formulas at the hands of men who have the scrutinising 
 instinct or habit, or who happen in a special case to 
 have been luckily sceptical. But I have noticed that men 
 trained in the physical sciences are very prone to accept 
 without question the imperfect formulas of the moral 
 sciences ; and you see how (Clifford, even in an original 
 investigation of ethics, thoughtlessly accepted a claptrap 
 formula from history. 
 
 My words imply, however, that even in their own fields 
 many specialists are fixed in false formulas. The claptrap 
 historical formula came from historians ; the new theory 
 in physics or biology has to fight tlie experts in possession,
 
 Letter II. LETTERS ON REASONING 25 
 
 who stand by the old. This is one of the most familiar 
 phenomena in the history of culture ; and yet men are 
 surprisingly slow to act warily on the knowledge. I have 
 long had the suspicion that if for a single year the specialists 
 in every field of thought would systematically doubt and 
 challenge every one of their current generalisations, taking 
 nothing for granted, there would be stirring times in the 
 way of new truth. 
 
 In my own intellectual pilgrimage every special investiga- 
 tion, I think I may say, has led me to revise some of the 
 conclusions I have found current even among " experts." 
 In such various fields as political economy, mythology, 
 Christian origins, the philosophy of history, psychology, the 
 study of Shakespeare, the analysis of verse, the criticism of 
 literature and men of letters, I seem to myself to find some 
 errors accepted, some truths missed, some traditional judg- 
 ments blocking the way. It was not that I set out expect- 
 ing such an experience : I simply wanted, I think, to under- 
 stand, to realise how things had gone : the general doubt 
 emerges from the experience. And when I reflect how un- 
 leisured has been much of my own research, how much has 
 been made by the midnight lamp, how ill-equipped have 
 often been my expeditions, I can easily conceive that even 
 if I have not been wholly astray in my innovating conclu- 
 sions, my formulas will soon need recasting, my survey fresh 
 adjustments. If either of you should ever follow any of my 
 trails, I beg you to aim at correcting my views ; it ought to 
 be no hard matter, in your day, if you take fair pains. 
 
 In any case, remember that, whatever views you set out 
 with, whatever beliefs you have in stock in youth, you pro- 
 bably hold some of them over-confidently ; some of them 
 will probably turn out to be wrong. Try, therefore, to 
 look critically at every fact or set of data on which you 
 found an argument or a conviction. I mean, of course, in 
 the way of serious study and important action. One must 
 guard against overdoing doubt in normal intercourse :
 
 26 LETTERS ON REASONING Letter IL 
 
 table-talk will not bear the strain of either the Socratic or 
 the scientific method ; and it is unamiable, not to say 
 clownish, habitually to obtrude distrust. But it is well to 
 have comrades who do not dislike doubt, whether by way 
 of jest or of earnest. As Emerson says, the truly well-bred 
 people (or let us say the well-educated people) are those 
 who are not easily shocked — shocked, that is, by serious 
 challenges to their beliefs. 
 
 It may seem to you, on reading thus far, as if I were 
 urging you to pass your lives in a state of hesitation. 
 Rather I am seeking to save you, by a discipline in doubt, 
 from the most painful kind of hesitations, those which 
 come to us when we are forced in our own despite to ask 
 whether we have not on some great issue been working in- 
 justice, whether we have not been long on the wrong side. 
 If such a doubt is ever forced upon you, I trust you will 
 find it painful. Many people, I confess, seem not to be 
 made unhappy by it. They forgive themselves for injustice 
 as readily as they commit it, as easily as they have credited 
 untruth. I hope you will not be of that order of moralists. 
 I can conceive no more painful intellectual experience than 
 the discovery that we have stoned a true prophet, resisted 
 a true doctrine, backed an injustice. People are always 
 doing these things for lack of due doubt ; and there is 
 small profit in the discovery if it do not warn us against 
 fresh blunders. 
 
 Religious people, I notice, speak much of the pain caused 
 to them and their like by attacks on their beliefs ; they say 
 little of any pain they have felt on finding that they had 
 denounced the bringer of a new truth. Now, every chal- 
 lenge to an opinion which we hold warmly or emotionally 
 is likely to be at least annoying. I can remember well how 
 in my teens I was hurt or angered by some criticisms of 
 my literary heroes which I learned later to regard as sub- 
 stantially just. I therefore advise you, when you are so 
 hurt or angered, to think a long while before you condemn
 
 Letter II. LETTERS ON REASONING 27 
 
 the critic. The effort will reward you. You will not 
 always find, after careful study, that you have been wrong ; 
 and the opinion that thus bears an anxious reconsideration 
 is more helpful, more sustaining, than one resentfully re- 
 afifirmed without hesitation. 
 
 This is the gain from doubt. Where you have doubted 
 carefully, you can have a higher confidence ; and I have no 
 fear that my counsel will unfit you for useful action. You 
 will learn that the majority are often and easily wrong ; and 
 when you have realised how they came to err, when you 
 have rigorously tested your own predilection, you can be 
 firmer and more persistent than they. To many of them, 
 doubt and change of opinion will come from the continuous 
 pressure of unexpected fact ; and if they change brazenly 
 or ignobly, making light of their old obstinacy, they do but 
 reveal their unworthiness. When you have entitled your- 
 selves to condemn them, you will not be more apt than 
 they to shilly-shally. Rather your confidence will outlast 
 theirs ; you will have the satisfaction — a high one as our 
 satisfactions go — of knowing in the end that you have been 
 " in the right with two or three," and that you have not 
 flinched at unpopularity. But when all is said, it is well 
 to remember that high standards of criticism are for our- 
 selves first and last ; that it is more profitable to scan our 
 own slips than to denounce those of other people. Act 
 on Clifford's ethical doctrine by all means and in all serious 
 cases ; but be more concerned to try yourselves by it than 
 to fulminate it against others. 
 
 You are much more likely to fall below it than to carry 
 it too far. I would not have you very cautious in every 
 matter of action — in tliu choosing of your clothes, of a 
 hotel, of a holiday route. Perhaps I am temperamentally 
 given to rapid decision in these minor matters : I fear I have 
 been so in some greater things ; but I reason that in separate 
 or isolated issues it is probably more economical to decide 
 quickly than to weigh lengthily a number of small considera-
 
 28 LETTERS ON REASONING Letter II. 
 
 tions. And I would not have you unadventurous for your- 
 selves. Where I counsel you to doubt and ponder is in 
 the framing of your beliefs, of your doctrines ; in taking up 
 a political cause, which may mean the helping or hindering 
 of many lives ; in deciding on the rights or wrongs of a 
 war, which is so apt to mean incalculable iniquity and 
 cruelty ; in crediting or discrediting a new teaching, or 
 even a new literary school or method. ^Vhen you have 
 vigilantly made up your mind, you will act, I trust, energeti- 
 cally. Action is life. A shrewd statesman of two hundred 
 years ago said that " If men would think more they would 
 act less : the greatest part of the business of the world is 
 the effect of not thinking"; but he had in view the blind 
 activities, the blundering activities. You will find, I hope, 
 that thinking will simply lead you to apply your energies 
 better, and to act in ways that ignorance would not have 
 dreamt of, while abandoning the activities of crude self- 
 assertion in which heady ignorance delights. 
 
 If we found that the stupid and ignorant people were 
 always active, and the wiser people usually inactive, we 
 might endorse the old statesman's maxim unreservedly. 
 But I see multitudes of ignorant and bigoted people doing 
 as little as they can help, and many thoughtful and instructed 
 people scheming and acting strenuously. I knew pretty 
 intimately a man famous as an iconoclast, a denier of current 
 dogmas, an " unbeliever " in current sanctities, and his 
 energy of action was to that of most of his gainsayers as 
 the strength of the arm to that of the litde finger. And for 
 myself I have always found thinking the best stimulant to 
 action, the best diet for confidence. 
 
 Much of what I am saying will be a set of truisms for 
 you when you come to middle age ; but I have little fear 
 that it will be superfluous for you in youth. You will find, 
 I hope, more help to right reasoning in your early environ- 
 ment than I did, and a commoner habit of conscientious 
 thinking than I sec around me to-day : if you do not, the
 
 Letter II. LETTERS ON REASONING 29 
 
 world will he no better or happier than it is now. It 
 is not (juitc inconceivable to me, however, that it may be 
 worse. In any case, you will have your lives to live, and I 
 have reason to believe that you can live them all the better 
 by forming habits of mental circumspection and thorough- 
 ness, even if such habits involve a fuller realisation of the 
 lack of them around you. Thought itself is a pleasure, 
 yielding activity, provided it be done by a healthy brain. 
 
 On some important matters, such as falling in love and 
 marrying, I shall not attempt to give you advice at all ; 
 because I do not believe that if I did you would act on it. 
 ]iut in the way of thought and action in general, I am fain 
 to hope that I may be able to help you somewhat.
 
 LETTER III. 
 
 In the two preceding letters I have urged two counsels on 
 you : — (i) to have no fear of reason, such fear being its own 
 contradiction ; but to guard against errors of reasoning by 
 reasoning more thoroughly tlian before; and {2) to be at 
 once slow of belief and slow of disbelief; disbelief being in 
 effect merely the negative aspect of the belief that you 
 already know the facts, or that the facts cannot be as now 
 asserted. But, unless you are very fortunately organised, 
 you will be slow to realise how hard it is to act on the 
 second of those precepts, even when you acquiesce in both. 
 The difficulty is indeed part of the explanation of that 
 despair of reason against which the first precept is directed : 
 we are all so prone to let prejudice shut the door on judg- 
 ment, so slow to bring the first-born, bias, under the control 
 of the second-born, reflection, that when we sadly scan each 
 other's aberrations the cause of reason is apt to seem 
 desperate. 
 
 Let me, in passing, offer you a consolation. The element 
 of bias in us is also the element of energy ; and it is trans^ 
 mutable ; so that when broken to harness, so to speak, it is 
 as useful a part of character as reason itself I speak, you 
 see, in metaphors, for the purpose of the moment. More 
 scientifically speaking, bias is simply the first thought ; 
 reason the second thought. The difficulty is to ensure that 
 the second thought shall always have its turn. 
 
 The older writers in support of religion against " unbelief" 
 used to protest that those who opposed them were following 
 will rather than reason ; and to a certain extent they 
 probably spoke truly. Only, the charge was much more
 
 Letter III. LETTERS ON REASONING 31 
 
 widely and radically true against themselves. All of us 
 tend to argue — that is, when we are disputing with others — 
 on lines of will. It belongs to our nature that we should 
 all wish to show good grounds for the belief we now hold, 
 that we should wish to turn out right. But it stands to 
 reason that most men must take more trouble to get out of 
 a belief in which they were brought up than they took to 
 get into it. To have held a belief, and then to have doubted 
 it, is to be potentially a stage further on the path to right 
 judgment than is one who stays all along where he was put. 
 The latter may, after all, be right : but he should clearly be 
 slow to charge on others a following of their carnal inclina- 
 tion. The first carnal inclination is always to think we 
 are right to begin with — that we had not believed in an 
 abstn-dity. The usc-and-wont man may make out a fair 
 case if he can show that the innovator is either adopting a 
 belief which will ostensibly give free scope to his passions, 
 or professing one which may win him worldly advantages ; 
 and it was common to charge the former tendency on 
 " unbelievers," so-called. But as it is historically certain 
 that large bodies of men have claimed a very free scope for 
 their passions on grounds of actual religious belief — belief 
 in the very religion of the use-and-wont man — and as it is 
 tolerably obvious that gain is to be had rather by professing 
 orthodoxy than by opposing it, such arguments are now 
 abandoned by all the more thoughtful of the orthodox. 
 They are valid in their type, but certainly not specially so 
 against minorities. What I am about to say is, so to speak, 
 sauce for botli goose and gander ; it bears on all forms and 
 fields of opinion. 
 
 Wc all incur two difficulties in arguing for the opinions 
 we have formed, and against those which clash with them— 
 the difficulty of being candid and the difficulty of being 
 sincere, taking those terms in definitions which give them 
 a moral significance. The two difficulties are at bottom 
 one, and the virtues in question are so likewise. I am
 
 32 LETTERS ON REASONING Letter IIL 
 
 wont to say that if we were all perfectly honest we should all 
 be better reasoners, so much bad reasoning being a matter 
 of unreadiness to be fair. In other words, we have so 
 many motives to opinion, so many incitements to wish that 
 this or that proposition may be true or untrue, this or that 
 man right or wrong, that we are apt to prejudge as often as 
 to reason. And — such is the paradox of judgment — the 
 play of variety of motive to belief has often the effect of 
 making us hold or maintain opinions which contradict each 
 other. If you Avill think well of it — so, at least, it has often 
 struck me — there is something very perplexing and dis- 
 quieting in the spectacle of self-contradiction. ^Ve nearly 
 all tend naturally either to assent to any earnest utterance 
 which does not clash with our previous opinions, or at least 
 to see in it a " sincere " doctrine which deserves a respect- 
 ful attention. But it often happens to us, after such an 
 assent or appreciation, to find the same teacher laying down 
 with equal earnestness a contradictory doctrine ; and if we 
 ponder such things deeply they set up a certain distrust of 
 human nature. As, however, there is no getting away from 
 human nature, we shall do well to adjust ourselves to this 
 one of its infirmities as we do to those of the physical life. 
 
 Ueberweg, the eminent German logician, has gravely and 
 well said that " Absolutely to shun contradiction is a task 
 demanding so harmonious and thorough a construction of 
 thought, and at the same time such a purity and freedom of 
 intention, that to fulfil it remains an ideal which is ever to 
 be reached proximately only. Not merely gaps in our in- 
 vestigations, but every kind of ethical narrow-mindedness, 
 the tenacity of religious, political, and social prejudices, 
 lead to contradiction." And again : — " Nothing is com- 
 moner in difficult problems than a half and one-sided 
 apprehension of thoughts strange to us, confounding it with 
 our own opinion, and then combathig this chimaera. The 
 opinion disputed classed under some abstract category or 
 other, which looks suspicious to common judgment or
 
 Letter III. LETTERS ON REASONING 33 
 
 prejudice, or else an introduction branding it as heretical, is 
 prefixed to a garbled statement, in order to prevent the im- 
 pression which the thought itself even in this form might 
 
 make The contest is transferred to a dififerent province, 
 
 and, by its construction of suspected consequences, polemic, 
 which ought to serve for the common investigation of truth, 
 is degraded to be an instrument for making attacks on in- 
 dividuals. The experience of all times shows that these 
 perversities are not solely produced by a specially dull and 
 narrow power of thought, and by a specially weak and 
 degenerate will. It is a rare power and structure of 
 thought and conscience which can keep itself entirely free 
 from them." ^ 
 
 And he concludes : — " To overcome narrowness, and to 
 enter fully within the circle of an opponent's thoughts and 
 into the motives for his doctrine — which is a very different 
 thing from the languid toleration of indifferentism — pre- 
 supposes a height of intellectual and ethical character which 
 is not innate either in individuals or in the race of man, 
 but must be acquired by a long and earnest struggle in 
 development. Yet this is the only path which leads man 
 to truth. His judgment only emancipates who has shown 
 himself to be docile." 
 
 It is a serious matter, you see ; and I want to impress on 
 you with a little detail the symptoms of the common 
 frailty — or, let us say, the aspects of the paths on which 
 we all tend to go astray. 
 
 First, as to candour : the mere constitutional reluctance 
 to admit that we are wrong sets us (i) upon resisting our 
 gainsayer before we weigh his case ; and (2) upon distorting 
 his case when we find it at all difificult to rebut. Further, 
 dislike of a person may incline us (3) to oppose any of his 
 opinions ; or, again, dislike of some of a man's doctrines, 
 after investigation, may move us (4) to negate a new 
 
 ' System of Logic and History of Logical Doctrines, Eng. Ir. 1S71, 
 pp. 248, 527-8.
 
 34 LETTERS ON REASONING Letter IIL 
 
 doctrine of his without investigation. Yet again, when 
 we have framed a theory and used it for some time, sheer 
 self-love is apt to make us (5) strain the facts in its favour 
 and give too little weight to the arguments against it. 
 Finally, general sympathy with or antipathy to a cause, or 
 party, or school of opinion, may easily make us prejudge 
 either for or against it on a particular issue. Here we have 
 seven common causes of bad reasoning or unreasoned 
 judgment. Do not then be surprised if you should find 
 what you suspect to be cases in point in these letters ! 
 I am very sure that if I should escape the snares it will be 
 only in virtue of being somewhat anxiously on the look-out 
 for them. They are the more insidious because the act of 
 succumbing to them has for us the semblance of loyalty to 
 the right, of service to truth and righteousness. But I think 
 you will agree with me, on reflection, that they all represent 
 lapses from candour. 
 
 I define candour, you see, as more than mere avowal of 
 what you see. In that sense, a deeply prejudiced person is 
 candid in respect that his prejudice really prevents his 
 properly seeing the other side. So to define the word is to 
 make it insignificant, or at least practically useless, as it 
 will then mean only " not fraudulent or malicious." All 
 considerate usage makes it suggest the taking of some 
 trouble to be fair, the attainment of a somewhat difficult 
 attitude, or else a great natural gift for fairness. It is 
 common, however, when a prejudiced person is charged 
 with being uncandid, to hear it said for him that he is 
 "sincere." Now, I propose that we should so define sin- 
 cerity as to make it, like candour, mean more than mere 
 abstinence from calculated wrong-doing. In the loose sense 
 I allude to, any man who is in a passion is sincere ; and 
 any unfair or inconsistent person may be so. I can see 
 no use in the word if it is not to be more stringently 
 defined. 
 
 According to some etymologists, the Latin word sincerus,
 
 Letter III. LETTERS ON REASONING 35 
 
 from which ours comes, originally meant sine cera, " without 
 wax," and was applied to statues to signify that they were 
 not, as often happened, flawed or broken in the stone and 
 puttied-up with wax to pass muster. That has always 
 seemed to me a funny and questionable etymology ; and I 
 see that some scholars surmise the original form of the 
 term to have meant " having (or consisting of) one wax,'' 
 which is at first sight not much more satisfying. But either 
 etymology, even if wrong, may serve to suggest the proper 
 and useful force of the term. Sincerity, to be a deeply 
 significant term, should mean not merely the state of 
 believing what you say at any one moment — a state normal 
 to many foolish or thoughtless or malevolent people — but 
 the state of attained or perfected consistency. 
 
 Let me illustrate my meaning. Suppose that a certain 
 political leader, rightly resenting a breach of law among the 
 party opposed to him, insists that the act ought to be 
 severely punished ; and that nevertheless, when an exactly 
 similar act is committed among his own party, he makes 
 light of the offence and finds excuses for it, then he is 
 clearly " inconsistent." We so describe his conduct in its 
 intellectual or logical aspect. I argue that, in its moral 
 aspect, we should call him insincere. It matters not, I say, 
 whether at the moment of self-stultification he did or did 
 not remember how he had formerly denounced the kind of 
 conduct he now palliates. 
 
 If indeed he remembered his former verdict, repented of 
 it, and avowed his repentance, he would be neither insincere 
 nor inconsistent ; he would simply be changing his opinion 
 and avowing so. Inconsistency means not change of prin- 
 ciple, but failure to conform to the principle you still profess 
 to hold. Consistency is not the same thing as constancy- 
 But if you profess to act on a given principle, yet under the 
 sway of partisan or other interest either forget it or set it 
 aside, you are not merely, I say, inconsistent, you are 
 insincere. You do not thoroughly and deeply hold the 
 
 E
 
 36 LETTERS ON REASONING Letter IIL 
 
 principle you profess ; it is not a part of your habitual 
 code ; it is a pretext you adopt at convenience, or merely 
 to gratify a particular aversion. Some forms of inconsistency 
 are indeed so insignificant on the moral side that it would 
 be doing a violence to our term to say they proved insin- 
 cerity ; but where inconsistency of conduct plus doctrine 
 amounts to committing serious unfairness, the deflection 
 from intellectual consistency is rightly to be stigmatised 
 on the moral side as insincerity. If not — if sincerity be 
 compatible with meting out different measures to men and 
 causes in terms of your mood or your partiality — what is 
 sincerity worth ? We mean the term to be complimentary : 
 is it really so when so laxly used ? 
 
 It may be argued against me, I can see, that insofar as I 
 make the terms inconsistency and insincerity mean the 
 same thing I am reducing the utility of one of them ; 
 but I would answer that I am making them apply to 
 different aspects of the thing — the intellectual or logical, and 
 the moral. I am seeking, in short, to introduce at this 
 point a moral element into the characterisation of error. 
 This raises the old question whether " error " is ever to be 
 regarded as "guilty." Either (it may be said) a man knows 
 that he is perverting the truth, in which case he is not erring 
 but lying, or he does not know, in which case he is to be 
 compassionated, not condemned. 
 
 I answer that the problem is nearly on a par with that 
 raised in the expression "culpable neghgence." We all 
 admit that we may be " culpably careless "; in other words, 
 we agree that it is desirable to keep current the opinion that 
 a carelessness which endangers life or limb is seriously 
 " WTong." (Whether it ought to be "punished " as it some- 
 times is in the case of an engine-driver or pointsman is 
 another matter : I am strongly inclined to think such 
 " punishments " even more uselessly cruel than others.) 
 It seems clear that social stigma has an educative effect 
 as applied to injurious negligence no less than to injurious
 
 Letter III. LKTTERS ON REASONING 
 
 37 
 
 aggression. Now, if carelessly inconsistent reasoning and 
 conduct — inconsistency of one with the other, or of parts 
 of either with other parts — involves demonstrably unjust 
 criticism, to say nothing of far more extensive and grievous 
 evil (as in the promotion of an unjust war), the application 
 of the term insincerity to the moral aspect of such incon- 
 sistency is as well warranted as the expression "culpable 
 negligence " usually is. 
 
 Broadly speaking, we all tend to be more or less incon- 
 sistent, and we all tend to be more or less insincere. Let 
 us, however, make the term a test for our own thinking, as 
 I am seeking to do now, before we use it to impeach others ; 
 even as we do well to apply Clifford's stringent law against 
 undue belief more readily to our own thought than to that 
 of others. If I find that, seeking to carry a certain point, 
 I once advanced a certain proposition in one connection ; 
 and that, forgetting this, but not proceeding on any new 
 knowledge, I advanced at another time a contrary proposi- 
 tion in another connection to carry another point, I ought 
 to admit that I have been insincere, and it is my duty to 
 recant one of my propositions if recantation be feasible. 
 I had presumably been arguing from hand to mouth, saying 
 one thing or another without thorough or weighed convic- 
 tion, playing fast and loose with truth and demonstration. 
 
 You will now see better what I mean when I say it is 
 hard to be candid and sincere in argument. On a multitude 
 of issues we frame propositions as we go along, on the 
 most various motives. Mere objection to being strongly 
 importuned by a new doctrine may tempt us to use against 
 it an argument which, when faithfully applied, counters 
 some other opinion of our own, and is therefore either 
 fallacious or a demonstration that we hold by a fiillac)-. 
 At one point or other, then, we have been insincere. Some- 
 one, again, utters a dictum which disparages one of our 
 heroes or makes light of one of our theories or conclusions, 
 and in our resentment we so describe the anno)'ing proposi- 
 
 E 2
 
 38 LETTERS ON REASONING Letter IIL 
 
 tion as to make it appear other than it is. When we so act 
 we deserve severe terms at the hands of any moralist who 
 is himself more scrupulous. We have been uncandid. To 
 discover afterwards that we have in this fashion penned an 
 injustice or a misrepresentation should be to any one of us 
 a painful and humbling experience. Yet there is a shame- 
 lessness about such things on the part of many that seems 
 to me almost more odious than shamelessness about acts 
 of fraud; and it is because I would fain think of you as 
 incapable of it that I thus repeatedly press upon you the 
 moral aspects of bad reasoning. 
 
 Intellectual honesty, I am satisfied, is much rarer than 
 pecuniary honesty, probably because the stress of social 
 evolution has run so much more to regulating the physical 
 and commercial than the intellectual relations of men. 
 No doubt it follows that intellectual deformity analogous 
 to kleptomania is relatively frequent — that many men really 
 cannot see straight in discussion, even when they are used 
 to it. But new perception of risks is for each of us a new 
 element of " determination " ; and if you realise that I am 
 urging these things on you seriously for your good I may be 
 saving you from some slips. 
 
 I would ask you to take note, then, (i) that malice is 
 very apt to make us positively misread an opponent's 
 meaning. Of this I have had some amusing experiences 
 in reviews of my own books. In one, for instance, I 
 chanced to speak of a certain book as published in " 1790, 
 and later." The type was large and clear ; but one hostile 
 critic actually cited the words^as "1790, or later," and 
 pointed to them as proof that I had an absurdly happy- 
 go-lucky way of gues.sing at the date of a book when I 
 might ascertain it by inquiry. The fact was that the book 
 had been published in 1790, and expanded in a later 
 edition, hence the manner of my reference. Now, that 
 critic cannot have been wilfully lying ; the perversion was 
 too imbecile, too exposable, to be deliberately made. But,
 
 Letter III. LETTERS ON REASONING 39 
 
 as his review showed, he was very angry, and his niaHce 
 positively affected his vision. Less gross cases are frequent 
 in reviewing. 
 
 Even witliout malice, however, one is very apt (2) to mis- 
 state an argument one does not like. I have seen this 
 done many a time by men not hasty or passionate in 
 temper, and much disposed to lay stress on the importance 
 of ethics. They simply lacked precision of perception on 
 the side of dialectics, and had never realised the need for 
 moral discipline in such matters. A proposition that jarred 
 on their feelings spontaneously took for them a more repul- 
 sive form ; like some nervous artists, they instinctively 
 exaggerated the feature they did not like, seeing it through 
 a medium of disturbed lesthesis. Such a tendency must 
 be hard to guard against ; and if either of you should chance 
 to be organised in that way my present preaching may avail 
 you little. Still, I will lay down a few prescriptions, some 
 of which might be useful. 
 
 1. In such matters we ought to ask ourselves, as it were 
 by rule, whether we are doing as we would be done by. 
 
 2. Even if we are, so to speak, in a " state of war " with 
 any writer or disputant, and feel that he ought to be dis- 
 credited, a patient analysis of his argument is the best 
 preparation for an effective reply. If he is quite wrong, 
 we shall realise this more fully and clearly after repeated 
 reflection ; if he is not quite wrong, we are much likelier 
 when patient to guard against indiscriminate or blundering 
 denial of what he says. The clearest perception, as a rule, 
 will always j'ield the tersest and most clear-cut rebuttal. 
 
 In other words, conscientiousness is a great help to right 
 reasoning. Perhaps, as I have said, most of the bad argu- 
 ment in the world is the result of sheer unreadiness to give 
 full weight to an opposing argument. 
 
 3. Make it a point, if possible, to argue against yourself 
 before you undertake to maintain any position. That is, 
 try to conceive at every step how an opponent might answer
 
 40 LETTERS ON REASONING Letter IIL 
 
 you. I say " if possible," as I am not sure to what extent 
 this habit may be acquirable or may be a special idiosyn- 
 crasy. In any case, I can assure you that it is well worth 
 your while to try to form it. It has sometimes happened 
 to me to convert myself completely from a particular view 
 by checking it when I set about propounding it. 
 
 But with all this, remember, it is of supreme and constant 
 importance to 'know the subject under discussion. Some 
 logicians, including Kant, have sought to keep logic as it 
 were independent of "things, insisting that it has to deal 
 with laws of thought which stand in their own right. They 
 mean that the principles on which we draw a logical con- 
 clusion are the same for all men and in all cases, and that 
 by them we can judge whether a given argument is sound 
 even if it be on a subject we have not studied — or rather, 
 that every student, no matter what his subject, must, if he 
 would reason soundly, come to those principles, which we 
 find on reflection in our own minds. But even as regards 
 the universal principles, it is certain that only after long 
 use of the judging faculty on things, on the problems of 
 daily life, did men attain to the power of stating those 
 principles, and holding at all steadily by them in a com- 
 plicated argument. And since, as I have said, many if not 
 most errors result from imperfect or false information, or 
 prejudice, rather than from wrong processes of reasoning on 
 facts, it follows that improvement depends very much on 
 further search for facts. 
 
 It is not merely that study will inform you of the error 
 of a given premiss, as in the case of that statement about 
 the religion of Teutons and Celts ; but that it will help you 
 in a general way to a judgment on probabilities. Having 
 detected one general error, you will be on your guard against 
 similar errors ; and having reached some general truths 
 about, say, the processes of human affairs or natural pheno- 
 mena, you will be helped (though here you must be on your 
 guard against new error) to look for sequences in a new set
 
 Letter III. LETTERS ON REASONING 41 
 
 of facts. Above all, the more you know the better are you 
 fitted to give provisional assent to or dissent from a new 
 proposition. 
 
 Take for, instance, the problem of the historic actuality 
 of the Gospel Jesus. To any one brought up as a believing 
 Christian, and even to most people brought up in or con- 
 verted to what we call rationalism or naturalism, the sug- 
 gestion that " no such person as that Jesus " ever existed is 
 apt to seem merely preposterous. It seemed so to me when 
 I heard it at the age of twenty or more. It is not that even 
 when we set aside all the miracles there is anything very 
 lifelike in the narrative that is left. An average novel is 
 much more lifelike ; but as to the novel we assume from 
 the start that it is a fiction, whereas we had all assumed 
 from the start that the gospels told, with whatever fabulous 
 addenda, of the life of a real personage ; and the first shock 
 of a challenge to that view is too disturbing to permit of a 
 prompt seizure of the essentials of the problem. 
 
 It is precisely the least prepared believer who is most 
 confident. When the student proceeds to weigh the argu- 
 ments he becomes gradually less surprised at the challenge, 
 even though he adheres to his first belief. He is forced to 
 acknowledge that a narrative which abounds in myths of 
 action — stories of unbelievable events — -is open to suspicion 
 in its narratives of mere teaching, or narratives of events 
 which, like those in novels, might really have happened. 
 When he clearly realises that the gospels are a patchwork, 
 frequently eked out and interpolated, he admits that it is 
 difficult to settle what parts are primordial, and to what 
 extent those parts are trustworthy. When further he learns 
 that even some parts supposed to be primordial, such as the 
 Sermon on the Mount, are really compilations from earlier 
 and contemporary Jewish literature, he cannot but concede 
 that his right to his old certainty is much shaken. And 
 when, finally, he faces the problem as to the apparently 
 complete ignorance of the Jesuine teaching on the part of
 
 42 LETTERS ON REASONING Letter IIL 
 
 the author or authors of the Pauline epistles, which are 
 ostensibly the earHest documents, he must, if he be candid, 
 grant that there are clear and fair reasons for questioning 
 the historic actuality of the Gospel Jesus. He may for 
 various reasons continue to believe in and affirm it, but 
 he will neither put aside the doubt with the confident 
 contempt of the ignorant man nor think to settle the 
 question once for all by mere a priori argument such as 
 the claim that the story cannot be an aggregation of 
 fictions, or that " there must have been a real founder." 
 Neither will he say, as an unprepared person is apt to say 
 offhand, that " if we believe in the existence of Julius 
 Caesar we must believe in the existence of Jesus Christ ; we 
 have the same kinds of evidence in the two cases." He 
 will have realised, after working at the problem, that the 
 two cases are profoundly different ; and that just as we may 
 reasonably be sure of the historic actuality of Charlemagne 
 and reject as a fable the story of William Tell, though the 
 latter is the later, so we may be sure of the actuality of 
 Socrates and at least doubt the actuality of the Gospel 
 Jesus. 
 
 A priori or abstract reasoning, in fine, can settle only 
 abstract problems : all problems of concrete existence and 
 occurrence are conditioned by testimony, and they are to 
 be solved only after careful reasoning on the nature and 
 value of the testimony concerned. Remember that just as 
 people now scoff at doubts of the actuality of Jesus, they 
 scoffed not very long ago at doubts of the historic actuality 
 of Adam and Eve and the siege of Troy; and, a little 
 earlier, at doubts concerning King Arthur and the miracles 
 of the saints. Our opinions are constantly liable to be 
 coloured or determined by our presuppositions in general ; 
 and just as men in the Middle Ages had no difficulty in 
 crediting stories which we see to be absurd, because their 
 education began with such stories, so even an able man who 
 to-day is educated in a religious creed will be a bad reasoner
 
 Letter III. LETTERS ON REASONING 43 
 
 on that creed unless he takes patient pains to test it. 
 There have been many cases of acute men of science — 
 such as Faraday, Kelvin, and CroU — who applied their 
 minds with originality and success to problems of natural 
 science, and remained astonishingly credulous and un- 
 critical on questions of religious history and dogma. To 
 the study of the latter they had no strong natural bent ; 
 and as they failed to make up for lack of bent by special 
 study, they remained ignorant thinkers on one set of sub- 
 jects, while they were experts in others. When you come 
 to understand how given men are thus variously gifted and 
 deficient, educated and uneducated, you will be saved, 
 perhaps, from a good deal of perplexity and discourage- 
 ment. 
 
 Let me give you an interesting illustration. The late Dr. 
 James Croll, who began life with little schooling and no 
 money-advantages, gradually qualified himself, despite many 
 breakdowns and hindrances of health, to deal authorita- 
 tively with such problems as those of glacial action, ocean 
 currents, and planetary history. I am not entitled to a 
 definite personal opinion on the value of his results in these 
 matters, but I find it highly rated by competent men of 
 science. On the other hand I find him to have retained in 
 religion an attitude so unintelligent that I should have pro- 
 nounced it foolish if I had not, from a study of his life, 
 learned to see in it the result primarily of education and 
 secondarily of the sheer curtailment of his mental life by 
 pain and disease, and the absorption of what brain energy 
 he had in other problems. In his latter years he could use 
 his brain thoroughly for only an hour or two at a time ; and 
 it is impossible to read his autobiographic sketch without 
 warm sympathy and admiration for the patient courage 
 with which he bore up under a series of desperate in- 
 firmities, doing what he could to the end. Some of his 
 moral qualities were of the rarest kind. 
 
 By nature the man was an acute, candid, and scrupulous
 
 44 LETTERS ON REASONING Letier III. 
 
 reasoner ; and while his bent was to physical science he 
 greatly preferred > those sciences in which ratiocination 
 counted for most, and the mastery and enumeration of 
 details least. Thus he " positively shrank " from chemistry 
 and geology, though he was led to pay heed to the latter 
 science partly by way of money-earning avocation and 
 partly by the need for facts in connection with his specula- 
 tions on cosmic time. Such, too, was his thoroughness 
 that while engaged in his early manhood as a tea-dealer 
 he gave all his spare time for at least a year and a half to 
 mastering the debate on Freewill, as brought before him 
 by the treatise of that very remarkable thinker Jonathan 
 Edwards — who, by the way, was one of the ancestors of 
 your mother. More than five times did Croll read the 
 book, always unable to find a flaw in the reasoning, yet 
 long unable to agree to its conclusions. Only when the 
 strength of the argument was finally established for him by 
 the failure of all attempts to refute it did he submit. Now, 
 this tells of an uncommonly honest and sure though slow 
 mind. Only men with good reasoning powers are thus 
 impressed by Jonathan Edwards. Yet Croll, while reasoning 
 thus deeply and logically on one aspect of theology, never 
 proceeded to sift the quasi-historical premisses on which 
 Edwards as a Christian proceeded, and remained to the 
 end of his life satisfied with his earlier creed that " salvation 
 was entirely of free grace." As he says : — 
 
 " Simple trust in Christ's vicarious death gave me complete peace of 
 mind and true happiness, a peace which the world can neither give nor 
 take away. The agnostic will smile at my experience. How different 
 would he feel if he experienced this blessed peace himself. 
 
 " ' The love of Jesus, what it is 
 
 None but his loved ones know.' " ^ 
 
 Here we have the complete renunciation of the procedure 
 by which Croll attained his scientific results and even his 
 
 ^ Autobiographical Sketch of James Croll, 1896, pp. 17-23.
 
 Letter III. LETTERS ON REASONING 45 
 
 acquiescence in the argument of Edwards. He wilfully 
 assumes that " the agnostic " can never have such peace of 
 mind as his — a childish delusion. He has never asked 
 whether his special satisfaction is not exactly like that of 
 religionists of other creeds, including men in the savage 
 state of culture, whose way of holding and acting on the 
 belief in an atoning sacrifice would have been to him 
 morally horrible, though philosophically on the same plane 
 of belief as his. Now, had CroU been able to give his 
 attention and thought to history and anthropology as he 
 gave them to matters of natural science ; had he begun his 
 serious religious reading by a study of the article on " Atone- 
 ment " in the old Penny Cyclopcedia, and carried it on through 
 Tylor and Darwin and Robertson Smith and Clifford, Strauss 
 and Baur and Renan and Havet, he would, I think, have 
 become incapable of his early belief in " salvation by atone- 
 ment." He would have transcended it as men transcend 
 the ideals of their childhood. The whole problem would 
 have assumed for him a different aspect ; and what was 
 latterly for him an emotional need would have become as 
 remote from him as the inebriate's craving for alcohol or 
 the invalid's for morphia. He would have seen too, without 
 difficulty, that the close and acute argumentation of Jonathan 
 Edwards on Freewill has no valid connection whatever with 
 the dogmas of Christianity, which in fact serve to degrade 
 his philosophic work into a miserable ministry to super- 
 stition. 
 
 The moral is that our knowledge makes a matrix, a 
 " climate," in which our reasonings are well or ill nourished, 
 developed or stunted. Not all men, indeed, suffer as CroU 
 did from the lack of relevant culture. Men far inferior to 
 him in power of speculation on physical problems, and 
 even in power of connected ratiocination, might have a 
 natural power of judgment which with no special culture would 
 withhold or withdraw them from his more grotesque religious 
 beliefs. As I have elsewhere claimed to show, there are in
 
 46 LETTERS ON REASONING Letter IIL 
 
 all Stages of civilisation some men naturally averse to the 
 currenc absurdities of religious creed, varying towards good 
 sense as others vary towards other forms of faculty. But 
 for the average run of us special culture makes nearly all 
 the difference between acquiescence in and release from 
 common delusions. 
 
 All that I have said, then, as to the need for slowness 
 alike of belief and of disbelief points not to a disregard 
 for testimony and lore, but to a zealous gathering of it. 
 For though a little-read man with good judgment is pre- 
 ferable to a widely-read man with none, it remains true 
 that knowledge is the soil in which judgment waxes, and 
 that every process of reasoning tends to be deepened and 
 refined as it is based on a widened knowledge of the sum of 
 things. " With all thy getting, get understanding," said the 
 wisdom-loving Hebrew. I am trying to help you on that 
 course ; but you will best help yourselves, perhaps, by taking 
 pains to know the facts on any issue on which you seek to 
 form a judgment.
 
 LETTER IV. 
 
 You will remember that, in speaking of the theories of Dr. 
 CroU, I said I had no right to a definite opinion on their 
 value. It probably occurred to you in reading that some 
 such avowal has often to be made by sensible people, and 
 that it is not always easy to decide where it is rightly called 
 for. I have been arguing, in effect, that it ought often to 
 be made where it is not, as when people pronounce without 
 any special study on such a problem as the historic actuality 
 of the Gospel Jesus. But after we have recognised the 
 unwisdom of such pronouncements, it is a serious matter 
 for every one of us to decide how far he shall be content 
 either to remain neutral on current questions or to rely 
 on the judgment of specialists. One of the standing per- 
 plexities of life is this of " the influence of authority in 
 matters of opinion " ; and you are likely to be met by 
 claims on the subject which will give you trouble. This 
 is another aspect of the discipline of opinion, above con- 
 sidered. 
 
 It is clear that we cannot investigate all subjects for 
 ourselves, or even all the details of every subject of which 
 we make a special study. In biology, for instance, even 
 the biologist takes the observations of many investigators 
 for granted, provided that they seem to him competently 
 done, and in no way inconsistent with his previous know- 
 ledge. The non-specialist, again, must either keep his mind 
 pretty much a blank on many great branches of science or 
 accept as trustworthy the results set forth by men who seem 
 to him to know their business and to think clearly. For 
 such cases it is a sufficient common-sense rule to regard as 
 provisionally true or probable the teachings which pass
 
 48 LETTERS ON REASONING Letter IV. 
 
 current as authoritative among scientific specialists, simply 
 taking care to remember that here we " walk by faith," and 
 never joining in any resistance to any other teaching without 
 taking great pains to understand the merits of the dispute. 
 If we do not take such trouble we act unworthily and 
 blamably in lending our voices to cry down any innovator. 
 
 It is to be observed, however, that many people, without 
 making any independent investigation whatever, will give 
 the most emphatic endorsement to some scientific or other 
 propositions, and treat with angry contempt those who do 
 not accept them. It is the curse of " faith " commonly 
 so called that it has inspired multitudes during many 
 centuries thus to be most zealous precisely where they 
 have thought least for themselves. But the tendency is 
 not confined to the field of religion. The food-value of 
 many substances commonly eaten, for instance, is very 
 variously estimated ; and some of us persist in eating some 
 things (such as ordinary bread) which are often declared to 
 be innutritious in comparison with others equally available. 
 We do this either because we doubt the alleged innutritious- 
 ness, not finding proof of it in our own experience, or 
 because we dislike the foods urged on us in preference. 
 In this, many of us are no doubt regrettably careless. But 
 on the other hand we have cause to distrust the judgment 
 of many who, having no skill or practice whatever in 
 chemical or physiological analysis, not only adopt a precise 
 doctrine of food values, but deride those who do not do 
 likewise. Such persons exhibit the primary bias to easy 
 belief and disbelief, in the form of a complete submission 
 to a given authority, which they seek to impose on others 
 in a spirit of fanaticism. Even if they happen to be right, 
 they are so " by chance," in a dangerously uncritical way. 
 Of course a man who carefully observes the effects of given 
 foods on himself is well entitled to urge his conclusions on 
 others. But even he need remember that it is hazardous to 
 make a general induction of the kind from one or a few
 
 Letter IV. LETTERS ON REASONING 49 
 
 cases. Given foods would seem to suit different persons in 
 very different degrees. 
 
 So far the course is pretty clear. Where "authority" 
 clearly turns upon special knowledge, in a field which we 
 have not the time to explore for ourselves, we all accept 
 authority as a working principle. I take from experts the 
 latest theory of the chemical structure of sugar, as I take 
 from an explorer his account of a country that I do not 
 expect to visit. They may err : therefore I must not be 
 their partisan against those who appear competently to 
 criticise them ; but under that reservation I take the risk of 
 being misled. In the case of Dr. CroU's theory of ocean 
 currents I remain neutral, because it does not rest upon any 
 specialist knowledge that I lack, and fails to convince me, 
 yet without leaving me at all sure that it is wrong. To 
 come to a more definite opinion would cost me more time 
 and study than I can give to the matter. In such a case, 
 one may be content to be neutral. But there are certain 
 provinces of opinion in which it behoves us all to think and 
 decide for ourselves if we would be valid personalities, or 
 fulfil worthily our duties as citizens. To take on authority 
 our notions of right and wrong, of what is just or expedient 
 in politics, of what is true in religion or religious history, 
 would be to become contemptible, or at least intellectually 
 null. 
 
 To the greater part of this last proposition most men will 
 assent ; though I have latterly seen educated men, including 
 at least one of professorial status, avow that they form their 
 opinion on the justice of an act of international aggression 
 by simple acceptance of the views of one alleged " expert." 
 But while few will justify this course, which means a renun- 
 ciation of a main part of the duties of a citizen in such a 
 State as ours, there are many who, going with my claim so 
 far, will draw the line at religion, and argue that there we 
 ought to bow to authority. Let us see what this position 
 involves.
 
 50 
 
 LETTERS ON REASONING Letter IV, 
 
 The first and simplest argument of a religious person as 
 against one who rejects his beliefs, or one who gives them 
 different dogmatic form, is {a) that man must bow to a 
 revelation from " God," or (b) that in religious matters the 
 proper safeguard against the fallibility of individual opinion 
 is the authority of the head of the historic Church. The 
 rational answer to the first assertion is that it assumes the 
 very point in dispute : that most religions have claimed to 
 be supernatural revelations ; that it must lie with each one 
 of us to decide for himself whether any is ; and that the 
 man who alleges "revelation" is finally doing so on the 
 strength of his personal opinion. To the second formula 
 the answer is, again, that we actually use our private 
 judgment when we reason that the Church or its head is 
 the proper authority on religious truth. Either we are 
 repeating a mere rote lesson, or we are passing an individual 
 judgment. The man who puts his opinions for shaping in 
 the hands of a given authority is either not reasoning at all, 
 or is proceeding on his own judgment exactly as when he 
 chooses a doctor. And it turns out in practice that among 
 Roman Catholics, who nominally stand for the principle of 
 authority in religion, the argument finally takes the form 
 that, just as some men have a special genius for medicine 
 or war, so some have a special genius for religious truth, and 
 that this genius is found collectively among religious people 
 and supremely in the organised Church, which embodies both 
 the special gift and the maximum of educative experience. 
 Such at least seems to be the argument of the late Cardinal 
 Newman, in his Grammar of Assent^ and elsewhere. 
 
 But here again there is either an appeal to reason or 
 there is not ; and in the end Cardinal Newman does not 
 rest on his mere argument that we should recognise the gift 
 of religious people for religion, but actually appeals to us, on 
 the most hackneyed lines of Protestant argument, to accept 
 the Christian religion as something visibly supernatural in 
 'Ch. ix., "The Illative Sense."
 
 Letter IV. LETTERS ON REASONING 51 
 
 the manner of its propagation and acceptance. He had 
 previously argued that in all fields there are special gifts of 
 perception ; and he had protested against Paley's under- 
 taking to prove the truth of Christianity to anybody on 
 common-sense lines, warmly disparaging " the conduct of 
 those who resolve to treat the Almighty with dispassionate- 
 ness, a judicial temper, clear-headedness, and candour." 
 Here, as usual, he has " begged the question "; he is angry 
 with those who need to be converted to his creed. And 
 yet in the end, tacitly admitting the futility of such language, 
 which appeals only to those who need not be appealed to, 
 he falls back on the usual orthodox methods of refuting 
 Gibbon's thesis that Christianity arose in a natural, sequent, 
 and intelligible fashion. 
 
 This, I think, is the invariable outcome of Roman 
 Catholic reasoning in the hands of the ablest Catholics. 
 Newman indeed does not strike me as at all a coherent 
 reasoner in comparison with some others of his Church ; 
 but in point of the power of seeing by flashes deep into a 
 problem he is entitled to be ranked high. To follow his 
 Grami7iar of Assent closely is to realise that on any line of 
 reasoning he soon reached the knowledge that he believed 
 because he wanted to. For this course he seeks to find a 
 sanction alternately in the ordinary procedure of human 
 nature — as if the bald fact that men commonly reach their 
 beliefs in a certain way were a justification of them — and in 
 the suggestion that some men's thought on a given theme 
 has a superiority against which there is no use in arguing. 
 Apparently, however, he always realises at this stage that on 
 that view every creed and every iconoclast is free to make 
 the same claim, and the argument for authority dies away 
 in a bad demonstration. 
 
 There remains a possible difficulty on one score : namely, 
 that many persons avowedly believe on " authority " certain 
 propositions which they do not understand. Those pro- 
 positions come to them, they say, from sources which they 
 
 F
 
 52 LETTERS ON REASONING Letter IV. 
 
 have found to be otherwise trustworthy ; from a Church 
 and a scripture and a set of teachers whom they find to be 
 profoundly wise where they can check them by reason ; and 
 when such a consensus of authority lays down a mystical 
 doctrine, unintelligible to ordinary judgment, it is to be 
 accepted (so runs the argument) as a divine mystery. Over 
 this question you may be met by the kind of argument con- 
 tained in the familiar story about the college dignitary who, 
 when a young man said in his presence, " I believe nothing 
 that I cannot understand," answered, "Then you will have 
 the shortest creed of any man I know." In that argument, 
 as commonly used, there is a bad verbal confusion. People 
 who use it do not seem to me to have settled what they 
 mean by " what I understand." 
 
 In all likelihood (taking the story to be true) the young 
 man meant, " I do not believe propositions which I cannot 
 understand " ; and if that was his meaning he was speaking 
 reasonably. The academic dignitary, on the other hand, 
 seems to have understood him to mean, " I do not believe 
 in the occurrence of processes whose nature I do not under- 
 stand." So to interpret him was not very candid or 
 sensible ; we may suppose that the dignitary was provoked 
 by an air of bumptiousness in his utterance, and resentfully 
 wanted to snub him. If, however, the dignitary really resented 
 and condemned the avowal " I do not believe propositions 
 which I cannot understand," his resentment was that of an 
 "irrational " person, as the phrase goes. And as many of 
 the religious people of whom I have spoken do notoriously 
 profess to believe propositions they do not understand, we 
 shall do well to examine their position. 
 
 It is clear, to begin with, that we must often give our 
 assent to statements of the occurrence of processes whose 
 nature we do not understand. Take for instance the law 
 of gravitation. It would be straining the use of the word 
 to say that we " understand " the universal process or 
 phenomenon of gravity ; though we may rightly say we
 
 Letter IV. LETTERS ON REASONING 53 
 
 understand given phenomena in terms of the law of gravita- 
 tion, which we take for granted. To explain gravity itself, 
 many hypotheses have been and are being made, in terms 
 of molecular motion ; and it may be that one day we shall 
 " understand " gravitation in terms of some one or more of 
 those hypotheses. Meantime, we assent to the proposition 
 which formulates for us the subsistence of gravitation ; and 
 we do this because we understand the proposition and 
 recognise the evidence in support of it. So with many 
 other matters. We can understand propositions which 
 affirm the normal occurrence of " mysterious " processes — 
 that is, processes which we cannot yet analyse in terms of 
 other processes — and when we are satisfied with the evidence 
 we say we believe those propositions. 
 
 It is quite another matter, however, to say we can or 
 should assent to propositions which we do not understand. 
 So far was the young man of the story from being wrong in 
 repudiating belief in such propositions, he might, I should 
 say, have ventured to assert that no man believes in 
 propositions which he does not understand. How can 
 there be " belief " on such lines? If I say "Abracadabra 
 sings green," it matters not whether any one professes 
 assent ; there is no belief, simply because there is nothing 
 understood. And if such a phrase be offered to me as a 
 divine truth by men who find it in a " sacred " book and 
 say they are ecstatically sure of its divinity, it is all the 
 same. Belief is assent to an understood proposition, 
 whether or not formally expressed. Try, if you like, to 
 think of yourself as believing that greenness cures toothache, 
 and that pleasure makes boots ! 
 
 Let us apply this test to a religious doctrine over which 
 men frequently profess to give a devout assent without 
 understanding it. The Christian creed affirms that Deity 
 consists of three " persons " which are perfectly distinct yet 
 perfectly one ; not three Gods, but one God : yet not one 
 person but three ; one being Father, the second Son, and 
 
 F 2
 
 54 LETTERS ON REASONING Letter IV. 
 
 the third Holy Spirit. Now, it is quite possible so to para- 
 phrase the bare formula "Trinity in Unity "as to make it 
 intelligible : you have only to say that " person " means in 
 the strict Latin sense persona, a character or function, and 
 that one Deity is " three persons " in the sense that he 
 " plays three parts " or has " three character aspects." But 
 such a rationalisation is not orthodox Christian doctrine ; 
 every intelligible compromise of the kind has been branded 
 as antitrinitarian heresy ; and the Trinitarian creeds continue 
 to insist on the personality of the three " persons " in the 
 normal sense of the term. The creed is in fact avowedly 
 an affirmation of the unintelligible : the Christian is called 
 upon to accept it as an incomprehensible proposition ; and 
 the orthodox Christian does so accept it, avowing that he 
 " believes " the doctrine as being given by supernatural 
 revelation. The more thoughtful believers, on challenge, 
 will say that they are satisfied on reasonable grounds (i) of 
 the supreme wisdom of the Scriptures in general, (2) of the 
 spiritual genius of multitudes of the men and women who 
 have accepted the dogma ; or, if they be Roman Catholics, 
 they may put it (3) that they are by historical study satisfied 
 of the fulfilment of the Gospel promise of divine-indwelling 
 to " the Church " ; and that on that score they believe what 
 the Church officially teaches. 
 
 You will readily see, I think, the rational answer to such 
 avowals. In the first place, the assent given to the dogma 
 is merely nominal : it is not belief ; it is a make-believe. 
 The proposition that three literal persons are literally one — 
 that three separate Almighties are but one Almighty — is 
 believed by nobody, let him say what he will. The 
 " believing " Christian is either playing the part of a parrot 
 or is spending his life in alternation between the two 
 " heresies " of Tritheism and Sabellianism. And the final 
 solution of this strange dispute is to be found in realising 
 why men came to set up such an astonishing shibboleth, 
 and to maintain it. Broadly speaking, it was in this wise.
 
 Letter IV. LETTERS ON REASONING 55 
 
 Long before Christianity, priesthoods found their advantage 
 in grouping as husband and wife and child, or in some other 
 relation, deities who had been separately worshipped ; or in 
 distinguishing among a multitude of deities sets who had 
 long been reputed to be so related. The beginnings of the 
 idea probably lie in the remotest ages of human culture, 
 when Gods and Goddesses of Sun and Earth, River and 
 Field, were figured in terms of human personalities and 
 relationships. As a result of all that primeval guesswork, 
 Triads were common in the Babylonian and Egyptian 
 world before our era. 
 
 In the earliest Christian documents the Triad idea is not 
 present ; it arose, like most of the rites of the cult, by way 
 of assimilation of convenient doctrines from other systems ; 
 men trained in Egyptian and Syrian mysticisms turning the 
 formulas of these to the uses of the new system. We need 
 not here ask whether they were " dishonest " or merely 
 " confused." In our strict sense of the term they were 
 both ; they could not be " sincere " because their intellectual 
 processes were so undisciplined, so lax, so incompetent- 
 Once set up, however, the trinitarian formula became a 
 stumbling-block for the more intelligent theologians ; and 
 many of these sought to rationalise it in some such fashion 
 as I have above indicated. But to do this was to put in 
 jeopardy one or other of the elements of the faith on 
 which its prestige appeared to rest. If " the Son " were 
 defined as a mere " phase " of the Deity, the gospel story 
 in general and the doctrines of the divine sacrifice and the 
 eucharist were resolved into mere avowed metaphors; the 
 hold of the priesthood on the hopes and fears of the 
 multitude would be gone ; and with the faith would vanish 
 the revenue. If, on the other hand, the separateness of 
 " the Son " from the Father were alone insisted on, the 
 monotheistic basis, emphasised in the Old Testament, would 
 be upset, and Christianity would be only one school of 
 polytheism competing with others. The insoluble dilemma
 
 56 LETTERS ON REASONING Letter IV. 
 
 was met by an unintelligible formula ; the Church affirmed 
 both sides of a contradiction ; the religious habit sufficed 
 to make the little-reasoning majority acquiesce ; and there 
 the dogma stands to-day, a shibboleth fit for savages, the 
 intellectual scandal and demoralisation of the Christian 
 system. 
 
 I have gone thus briefly into a historic matter in order to 
 bring home to you on another line the truth of the proposi- 
 tion that men's errors and absurdities are the results more 
 often of overruling motives of interest and prejudice than 
 of simple fallacy committed in good faith. To guess 
 wrong is human, so to speak ; to fix a wrong guess in 
 an inconceivable dogma is sacerdotal. But sacerdotalism 
 is merely the organisation of error on the lines of a 
 particular set of interests : other interests pervert us in 
 other ways. 
 
 It is hardly necessary to point out, further, the fallacy 
 involved in saying : " I find the Scriptures in general marvel- 
 lously wise ; the Scriptures lay down the doctrine of the 
 Trinity, therefore I accept it ; the Church, too, is collectively 
 wise above any individual, I therefore accept the Church's 
 interpretation." It might suffice to answer (i) that " the 
 Scriptures " are demonstrably the work of many hands ; 
 but (2) even if they were not, the general wisdom of a 
 scripture would be no reason why we should go through the 
 form of accepting from it a proposition to which we can 
 attach no meaning. In point of fact, however, almost no 
 man now believes every intelligible statement made in the 
 Christian scriptures ; there is therefore a plain " insincerity " 
 in the official maintenance of an unintelligible proposition 
 on scrii)tural grounds. Many believers, I have noticed, are 
 ready to reduce to allegory or trope plain Biblical teachings 
 which happen to be inconvenient, yet persist in giving a 
 literal force to dogmas that convey no idea whatever. Such 
 are the fruits of the principle of "authority." 
 
 Is there then any reasonable ground for admitting " autho-
 
 Letter IV. LETTERS ON REASONING 57 
 
 rity " in matters of religion ? The answer is, I think, that 
 insofar as our opinions in this field turn upon special know- 
 ledge, such as that of experts in Hebrew and Assyriology, we 
 have to depend on their research, remembering always that 
 one expert may contradict another ; but that as regards the 
 great problems of belief in a definable deity and a " moral 
 government of the universe," as well as with regard to the 
 moral soundness of religious precepts and dogmas, we 
 must either think matters out for ourselves or confessedly 
 remain on the intellectual plane of Catholic peasants. The 
 man who tells me that I am spiritually blind, and that he 
 or another has spiritual vision, merely invites me to tell 
 him that he is given over to a strong delusion, and that 
 what he calls special insight is in my opinion special 
 blindness — an inability to reason intelligently from the 
 facts which lie before us all. 
 
 Certainly there are great differences in men's faculty for 
 philosophic thought and for moral judgment : we all make 
 such discriminations among those we know and among the 
 writers we read. And when we find ourselves indisposed 
 to agree on a point of philosophy or ethics with one whom 
 we have usually felt to be highly competent, we naturally 
 hesitate long before deciding. If on a question of moral 
 theory or practice I find myself diverging from Spinoza, or 
 Fichte, or Sidgwick, or Spencer — much more if I am 
 diverging from all four — I shall think twice, or thrice, or 
 twenty times before I make up my mind. But in all such 
 cases, observe, we are finally relying on our own judgment, 
 even if we see fit to yield to the " authority " ; and, whereas 
 we do that to the extent of taking a risk when we choose to 
 follow one physician's rule rather than another's, we there 
 do the best we can for ourselves, we not being physicians ; 
 while to " take a risk " by way of accepting a moral principle, 
 or a philosophic principle which has moral consequences, 
 merely because we think highly of the man who lays it 
 down, is not to do the best we can for ourselves, but to
 
 58 LETTERS ON REASONING Letter IV. 
 
 surrender our moral dignity and prepare the atrophy of our 
 moral sense. 
 
 It is not enough to remember this for ourselves : we 
 should constantly remember it for others. A thoughtful 
 and well-informed man may naturally be impatient of the 
 " self-opinionated " opposition of the ill-informed ; but if he 
 be morally as well as technically wise he will always 
 remember that there is no moral or educational value in 
 the mere assent of an ill-informed person on a matter of 
 general judgment, as distinguished from one of special 
 knowledge ; and that it is better for the less wise man to 
 think wrongly for himself than to be right by the mere 
 chance of his obedience to a teacher who may happen to 
 be right. One is sometimes tempted to think that a great 
 many people could not do better than follow unquestioningly 
 the guidance of some one wiser than themselves. But 
 rightly to choose a guide, to begin with, is a matter that 
 would call for careful judgment ; and if the chooser should 
 first make a wise choice and afterwards be in the position 
 of following the guide when the guide is wrong, he will in 
 the terms of the case have lost some of his power of judging 
 — that is to say, he will have sunk to a lower moral and 
 intellectual level. 
 
 Let us, then, never seek a mere docile assent, even from 
 the unwise. They ought, indeed, like the wise, to practise 
 a candid and careful attention ; we may rightly ask them 
 for that if we in turn accord it ; but we should not better 
 them or ourselves if we could induce them to accept 
 opinions on "authority." The authority who persuades 
 them to do so is in a fair way to acquire all the vices of the 
 historic priest. Being accustomed to deference, he is 
 easily enraged by contradiction, and thus himself loses 
 the precious power of learning new truth. He then figures 
 as a possessor of power without the qualifications for its 
 safe use. Better that the unwise should err in opinion 
 than that they should be occasionally and unwittingly right
 
 Letter IV. LETTERS ON REASONING 59 
 
 on the authority of one thus sure to be at times seriously 
 wrong. The only way by which they can be made really 
 wiser is that of thinking for themselves, tentatively following 
 indeed what guidance appeals to them, but never without 
 weighing for themselves the opinion in hand, if it be not 
 one that only expert knowledge can test. 
 
 Perhaps this strain of thought may strike you as plati- 
 tudinous. If so, so much the better — so much the better, 
 that is, if you have already seen the point for yourselves. 
 It was powerfully put long ago, on somewhat different lines, 
 by John Stuart Mill in his essay On Liberty ; and, in another 
 aspect, it was put less vividly but still well a generation 
 before him, by Samuel Bailey in his Essays on the Formation 
 and Publication of Opifiions. But I feel sure that the matter 
 needs to be brought home afresh to each new generation : 
 the perception of such truth is not inherited. 
 ■ To conclude : — You will sometimes find yourselves in 
 the position of having formed an opinion which some one 
 with greater authority — that is, ostensibly fuller knowledge 
 of the subject — contravenes. If you are quite sure of his 
 fuller knowledge, you will at once, I hope, suspend your 
 judgment till you have studied the matter further. If, after 
 all the study you can give it, you are still at a loss to 
 acquiesce, you will best preserve your integrity and your 
 power of future judgment by either reserving the question 
 as doubtful or clearly working out your reasons for rejecting 
 the " authoritative " opinion. If the matter be one of 
 what we call " taste," that is, a question of judgment on 
 the sesthetic merit of a picture, or a poem, or a novel, or 
 a piece of music, the proper course remains the same. If 
 you cannot really admire a picture or composition that is 
 highly praised by people accustomed to judge of such 
 things, you will never, I trust, proceed to affect to do so. 
 You will be content, I hope, to admit your difficulty and 
 await the probable effect of further experience on your 
 taste. To admire to order, to admire in advance the things
 
 6o LETTERS ON REASONING Letter IV. 
 
 praised by eminent critics, is not very good for any of us. 
 We are all, I fear, rather apt to do it in youth ; but even 
 then we instinctively object to it when we see it done by 
 others, feeling that such foregone or fashion-following 
 opinion is a sign of either weakness or immaturity of 
 character. 
 
 But in saying all this I do not forget or cancel my 
 warnings to you against letting self-will keep you wrong. 
 We all develop in taste, as in thought, up to a certain point 
 at least ; and you must have matured much faster than I 
 did if you are not often startled on a return to a book or 
 a picture that you spontaneously admired a few years ago. 
 Such experience ought to be educative. Some people, I 
 think, flaunt their changes rather gratuitously, making one 
 feel that they admire lightly and lightly alter ; but it is better 
 to do that than to brazen out our early declarations when 
 we have at heart faltered in them. There is a decent mean. 
 We can turn our change to account by thinking out what 
 it is and why it is that we once admired and no longer 
 admire ; and we may profitably restrain the emphasis of our 
 new preferences, deciding to wait and see how the new 
 impression wears before we undertake to do critical battle 
 for it. 
 
 In advising you thus I am perhaps trying to put a middle- 
 aged head on young shoulders. Youth does not take long 
 views and " hedge " on its opinions ; at least the youth that 
 was I did not spontaneously do so. But I may save you 
 now and then from some stress of unreasonable zeal, or 
 help you here and there to begin to know yourselves — a 
 study that is usually begun rather late in life.
 
 LETTER V. 
 
 Thus far I have been counselling you as to the moral 
 conditions of right reasoning, on the ground that many, 
 if not most, perversities or fallacies of judgment arise out of 
 faulty states of temper or moral attitudes, or else from lack 
 of the due atmosphere of knowledge, rather than from 
 honest logical oversight or confusion. By implication, 
 however, we recognise that such honest oversights and 
 confusions are common ; and whether a propounded fallacy 
 be the result of prejudice or of honest error, it may easily 
 happen that an unprejudiced inquirer shall be led astray 
 by it. Let us then look around us for cases of actual 
 fallacy or of bad reasoning, noting how they arise, and how 
 haply we may guard against similar lapses. 
 
 First, we have to note the common case of obliviously 
 using terms in different senses. This may happen at times 
 even to a trained reasoner. I find an instance in the 
 procedure of Professor Jevons, the eminent logician, in 
 his little Primer of Political Economy. After defining 
 wealth as consisting in "What is (i) transferable,"(2) limited 
 in supply, (3) useful," he goes on to make this observation 
 
 for the benefit of young readers: "Wealth is far from 
 
 being the only good thing ; nevertheless it is good, because 
 // saves us from too severe labour, from the fear of actual 
 want, and enables us to buy such pleasant things and 
 services as are transferable." Here, you will see, the 
 Professor unwittingly passes from the use of the term 
 wealth in the special sense of his definition — the generality 
 of exchangeable commodities — to the use of it in the 
 ordinary sense, that of individual " riches," the possession
 
 62 LETTERS ON REASONING Letter V. 
 
 or command of a relatively large share of economic " wealth." 
 Taken in the first sense, it reduces to nullity the assertion 
 that "it" saves us from too severe labour. The poorest 
 man has some of "it," and the richest has only a portion 
 of "it." The Professor's "it" is not "wealth" as he had 
 defined it, but riches — command of a large quantity 
 of wealth as defined. So again, it is nugatory to say that 
 wealth as defined "enables us to buy such pleasant things 
 
 as are transferable ": the proposition amounts to this, 
 
 that wealth enables us to buy wealth. 
 
 Still further does the Professor play false to his definition 
 when he proceeds to say that " in a diving-bell or a deep 
 mine," air, which is usually costless because there is plenty 
 for all, "becomes limited in supply, and then may be 
 
 considered a part of wealth Even in the Metropolitan 
 
 Railway tunnel a little more fresh air would be truly 
 valuable." The Professor has here forgotten the force of 
 the term "transferable" in his own definition. It there 
 meant " conveyable from one person to another "; or, more 
 strictly, "from seller to buyer." Now, there can be no 
 such transfer, no such conveyance, in the mine, or the 
 diving-bell, or the tunnel. The railway passenger cannot 
 buy air for use in the tunnel. The miner or the diver, if 
 working on his own account, might indeed arrange with 
 some one to pump air down to him ; but even then he 
 would not be buying the air, he would be buying the services 
 of the pumper. Air in all these cases, then, is " valuable " 
 in a sense quite other than that covered by the given defini- 
 tion of " wealth ", and to assert its " value " is as irrelevant to 
 the purpose as to say that pain may be a valuable discipline.^ 
 Here the excessive blundering was probably a result 
 ^ It may have occurred to you that the term "useful" is an inade- 
 quate one for the comprehensive definition of what constitutes wealth. 
 Diamonds, for instance, rank as wealth, but are not "useful" in the 
 ordinary sense. " Useful," however, is here taken in the broad sense 
 of " ministering to desire," and such usage is permissible. It is essential 
 only that given definitions be adhered to.
 
 I>etter V. LETTERS ON REASONING 63 
 
 of sheer failure of attention through fatigue. Professor 
 Jevons was a very hard worker ; and he presumably wrote 
 his Primer when his brain was tired. It would be quite 
 unwarrantable to suppose that he was not concerned to be 
 careful in writing for young readers ; but perhaps the 
 unwonted need to "talk down" may have further relaxed 
 his intellectual processes. I incline to think, however, that 
 in his case there was some natural lack of the capacity to 
 use words accurately and easily. He has told that he 
 found it impossible to learn German, for want of linguistic 
 gift ; and though one rarely finds the specially good linguists 
 to be deep thinkers, it may be that an abnormal shortcoming 
 in the power to learn a foreign language is a phase of some 
 congenital defect of verbal faculty in general. 
 
 This, at least, seems the only excuse for the lapses from 
 logical statement which occur even in Mr. Jevons's most 
 careful performances. For instance, in his essay on " The 
 Substitution of Similars " he has this sentence : — 
 
 " These laws may seem truisms, and they were ridiculed as such by 
 Locke ; but, since they describe the very nature of identity in its three 
 aspects, they must be assumed as true, consciously or unconsciously ; 
 and if we can build a system of inference upon them, their self-evidence 
 is surely in our favour." ^ 
 
 Here the " but " is absurd : a truism is a self-evident propo- 
 sition, that is the meaning of the word ; and from such a 
 proposition no logical " inference " can be drawn. Aristotle 
 framed the formula, "A is either B or not B," in order to 
 bar the triflers who were capable of saying that it might be 
 both. But that admission is merely a condition precedent 
 for all rational discussion ; it cannot yield an " inference," 
 since it leaves all real propositions untouched. 
 
 Let me now give you, from the same work of Professor 
 Jevons, a paragraph constituting an entirely fallacious argu- 
 ment on a question of practical politics. When you have 
 
 ^ Essay cited, pp. 46-47.
 
 64 LETTERS ON REASONING Letter V. 
 
 analysed it, you will be partly experienced in one mode 
 of confused reasoning : — 
 
 " At the present day the Government is called upon to take charge 
 
 of the telegraphs and railways, because great benefit has resulted from 
 their [its] management of the post-office. It is implied in this demand 
 that the telegraphs and railways resemble or are even identical with the 
 post-office in those points which render Government control beneficial. 
 
 The whole question turns, of course, upon the degree and particular 
 
 nature of the similarity. Granting that there is sufficient analogy 
 between the telegraph and the post-office to render the Government 
 purchase of the former desirable, we must not favour so gigantic an 
 enterprise as the purchase of the railways until it is clearly made out 
 that their successful management depends upon principles of economy 
 exactly similar to the case of the post-office."^ 
 
 You have already noted, I hope, the illicit procedure. 
 To begin with, it was not implied in the appeal in question 
 that the main points of management of railways were 
 " identical " with those of the post-office. The argument 
 was : " We have done the one ; we may do the other." To 
 say, then, that we " must not favour " the appeal until the 
 principles of successful management are proved to be 
 " exactly similar " is to affirm a plain noti-sequitur. We 
 may fitly favour the appeal for many reasons. All that 
 properly follows is : " We must not ground our demand for 
 railway nationalisation on the mere success of the post-office 
 until we have shown that the enterprises would have sub- 
 stantially the same economic or financial conditions." Here 
 again we seem to be dealing with a want of faculty for 
 statement; and I incline to think that Mr, Jevons's love 
 for symbolic forms of argument, and for his "logical 
 abacus," was correlative with this shortcoming. Symbols, 
 it is conceivable, may be specially helpful to a thinker who 
 is awkward in the use of words ; and I have more than once 
 noticed skill in or proclivity to the use of them in men who 
 were either normally or chronically slow of speech, or slow 
 at sentence formation and syntax. 
 
 1 Work cited, p. 72.
 
 Letter V. LETTERS ON REASONING 65 
 
 Far be it from me, however, to suggest that the glib- 
 tongued men are usually accurate reasoners. What I mean 
 is that Jevons had a large part of the outfit or structure of 
 a good reasoner, but miscarried because of one congenital 
 defect. Many a fluent man has neither the gift of coherence 
 nor the concern to acquire it ; and if a slow-tongued 
 professor of logic yields you samples of spurious argument, 
 eloquent politicians will probably yield you a good many 
 more. 
 
 Another case of covert transformation of terms, to the 
 detriment of the argument, occurs in the work of that 
 very gifted scholar and anthropologist, the late Professor 
 Robertson Smith, on The Religion of the Semites. On 
 page 55 he writes : — 
 
 "It is not with a vague fear of unknown powers, but with a loving 
 reverence for known Gods who are Ivnit to their worshippers by strong 
 bonds of kinship, that religion in the only true sense of the ivoi-d begins. 
 Religion in this sense is not the child of terror ; and the difference 
 between it and the savage's dread of unseen foes is as absolute and 
 fundamental in the earliest as in the latest stages of development." 
 
 In the next chapter (p. 88), apparently forgetting altogether 
 that he had written as above, he writes : — 
 
 " The immediate inference from all this is that the origins of Semitic, 
 and indeed of all antique religion, go back to a stage of human thought 
 in which the question of the nature of the Gods, as distinguished from 
 other beings, did not even arise in any precise form, because no one series 
 of existences was strictly differentiated from another.'''' 
 
 That this inference is his permanent or normal position we 
 may gather from pp. 27, 88, and other passages. Thus the 
 passage first above cited is a flat contradiction of his general 
 teaching. The second passage, however, is in turn contra- 
 dicted by one on p. 129, where it is asserted that "the 
 transformation of certain groups of hostile demons into 
 friendly and kindred powers " is "already effected, by means 
 of totemism, in the most primitive societies of savages " — 
 there is contradiction, that is, unless the writer means
 
 66 LETTERS ON REASONING Letter V. 
 
 " savages now existing." But he proceeds to assert that 
 " there is no record of a stage in human society in which 
 each community of men did not claim kindred and alliance 
 with some group or species of the living powers of nature." 
 
 The inconsistency in the latter case is one that might 
 arise through mere over-readiness to generalise emphatically. 
 The generaliser, like the general liar, had need have a good 
 memory. But the glaring contradiction of the two first- 
 cited passages is to be traced, I think, back to the old 
 moral infirmity which we have already considered. Pro- 
 fessor Robertson Smith was a professional theologian first 
 and an anthropologist afterwards ; and his outburst about 
 " religion in the only true sense of the word " appears to 
 have been an expression of his carnal resentment of the 
 attitude of rationalists who rejected his theology altogether. 
 In this state of temper he was capable of giving as the 
 " only true sense " of religion one which he himself usually 
 rejected, as he afiirms it only in this passage. After the 
 passage cited from p. 55 he categorically asserts that " It is 
 only in times of social dissolution, as in the last age of the 
 
 small Semitic states that magical superstitions based on 
 
 mere terror invade the sphere of tribal or national 
 
 religion." In the passage cited from p. 129 he asserts on 
 the contrary that such superstitions are immeasurably early. 
 
 Of such an exhibition we are entitled to speak strongly. 
 It proves that the writer's prejudice made him " insincere " 
 and a<af/^<7<r unscrupulous ; and if we studied only such aspects 
 of his work we should be led to think him an extremely 
 untrustworthy and indeed incompetent reasoner. But this 
 he was not. Where his theological prepossessions did not 
 come into play he was an acute and original thinker ; and the 
 first lesson of his lapses is that which I have already urged 
 on you — that we must be constantly on the watch against our 
 emotional leanings if we would be faithful and just reasoners. 
 But there is further to be kept in view the risk that we may 
 without prejudice fall into other men's errors if we read
 
 Letter V. LETTERS ON REASONING 67 
 
 them without constant critical attention. Be not such 
 readers, I beg of you. One book critically read yields 
 more education than twenty skimmed inattentively. Those 
 who so read Robertson Smith as to be satisfied equally 
 with his original and scientific thinking on the psychology 
 of ancient sacrificial rites, and with his fulminations of 
 pulpit claptrap, have substantially failed to profit by him. 
 His lectures on Semitic religion, remember, were originally 
 delivered to audiences of church-going people — that is to 
 say, in an atmosphere stimulative and evocative of claptrap. 
 It takes an uncommon degree of intellectual scrupulosity to 
 save a man from ever saying on a popular platform things 
 that are below the standard of critical taste which he 
 recognises at his desk. 
 
 Even at the desk, however, a professed expert may 
 blunder grievously under the spur of temper. The late 
 Professor Momerie, who held the ch^ir of Logic and Meta- 
 physics in King's College, London, published a little book 
 On Belief in God, which was praised as " wise " and "philo- 
 sophic " by certain reviewers of his own way of thinking. 
 At the outset he founds on the (alleged) universality of the 
 " desire for God " as affording a strong presumption, though 
 not a proof, of the existence of a Supernatural Person. The ab- 
 sence of " the religious sentiments," he asserts, " is exceptional 
 and rare." Yet in the same chapter, growing visibly angry 
 over the suggestion that advance in culture will raise men 
 above theism as they have been raised above polytheism, 
 he writes : " Agnostic opinions, I admit, are rapidly gaining 
 acceptance among the multitude ; but this is just because 
 agnosticism is essentially shallow and superficial, and therefore 
 very easily mastered." Not only does he thus contradict 
 his opening proposition : he expressly disparages the think- 
 ing of the multitude as being shallow and superficial, in the 
 act of maintaining the thesis that an opinion of the mul- 
 titude is likely to turn out sound. Here we have a flagrant 
 insincerity, plainly attributable to temper, on the part of 
 
 G
 
 68 LETTERS ON REASONING Letter V. 
 
 one whose special business it was to transcend the snares of 
 temper. And other educated men, sharing his animosities, 
 fail to detect his self-contradiction even when it is their 
 special business to do so. 
 
 I seem always to return, thus far, to my original warnings 
 against the moral snares in the path of the reasoner. Can 
 it be, I wonder, that fallacies arise even more often through 
 temper and prejudice than I had supposed? Howsoever 
 that may be, the next sample of inconsistent reasoning that 
 occurs to me is in Hegel ; and I fancy we shall have to 
 revert even in his case to the moral mode of explanation. 
 Arguing for his proposition that all History is the process 
 of the universal Reason, he writes : — 
 
 ' ' It was for a while the fashion to profess admiration for the wisdom 
 
 of God as displayed in animals, plants, and isolated occurrences* 
 
 Why not also in Universal History ? But Divine Wisdom — i.e.. Reason, 
 is one and the same in the great as in the little " ^ 
 
 Yet only a few lines above he had written : " God wishes 
 no narrow-hearted souls or empty heads for his children ; 
 but those whose spirit is of itself indeed poor, but rich in 
 the knowledge of Him." I have never met with a Hegelian 
 who would undertake to reconcile this foolish sentence with 
 such a proposition as that cited above. In that, Hegel 
 affirms the universal immanence of what he calls " God." 
 In this he speaks of his God as a vaunting soldier might 
 speak of his General, or as any vulgar person might wrangle 
 with any other concerning his own merits or procedure. 
 But Hegel was certainly not a mere braggart or vulgarian : 
 he was a man of singular originality and depth of thought, 
 however often he may " maunder," as one of his disciples 
 says he does. For the explanation of an imbecility on his 
 part, then, we must look either in his temper or in some 
 failure of due knowledge. 
 
 In many cases he was absurd because he arrogantly 
 insisted that he could know facts in nature a priori : there 
 ^ Philosophy of History, Eng. trans., p. i6.
 
 Letter V. LETTERS ON REASONING 69 
 
 temper and ignorance combined to make him miscarry 
 extravagantly. Here the trouble is simply temper. He has 
 set out to rebut " the doctrine that it is impossible to know 
 God " ; and on this errand he passes from one form of 
 insolence to another. First, he asserts that if we make the 
 affirmation he challenges we " have the convenient licence 
 of wandering as far as we list in the direction of our own 
 fancies " — a suggestion that the men he opposes are seeking 
 a philosophy which shall leave no restraints on their actions. 
 This is the method of the lowest order of theologian : it 
 would be difficult to reason more ignobly, more contemp- 
 tibly. To attribute your opponent's alleged but unproved 
 error to a criminal motive which is peculiar to him, is to 
 make an end of the conditions of argument between you 
 and him. There is all the difference of two stages of 
 civilisation between such a fashion of initially ascribing your 
 adversary's views to bad character, and the process of first 
 showing him to be in error and then surmising that he erred 
 through eniotional bias. 
 
 By way of fulminating the more effectually against his 
 philosophic foes, Hegel proceeds in the passage in question 
 to profess that, while their philosophy is heterodox, he is 
 orthodox. " In the Christian religion," he writes, " God 
 has revealed himself — that is, he has given us to understand 
 what He is ; so that He is no longer a concealed or secret 
 existence." There are many grounds for believing that this 
 was not even an honest expression of Hegel's belief. And 
 it is after this that he explodes as above noted against the 
 " narrow-hearted souls or empty heads " ; an utterance 
 opposed alike to Christian doctrine, Hegelian philosophy, 
 common-sense, and good manners, its aim being simply to 
 asperse those who held another view of things than the 
 Hegelian. 
 
 When we turn from the outbreak of temper to the doctrine 
 it aims at vindicating, we are presumably on higher ground. 
 Inconsistency in argument, as I have urged, is a proof of 
 
 G 2
 
 70 LETTERS ON REASONING Letter V. 
 
 " insincerity " ; but when we come to a main thesis, whether 
 or not it be consistent with the other positions of its framer, 
 we have to reckon with it as the expression of a deliberated 
 thought. And Hegel's formula of History is a main item 
 in his philosophy, which laudably strove to reach a unified 
 and all-sufficing notion of the universe. Here we have an 
 exercise on the higher levels of reasoning. 
 
 Let us ask ourselves, first, what we are to understand by 
 Hegel's reiterated theorem, or theorems, that Reason rules 
 the universe, " That Reason is the Sovereign of the World ; 
 that the history of the world, therefore, presents us with a 
 rational process " ; that " Reason is the substance of the 
 Universe " ; that " it reveals itself in the world, and that in 
 that world nothing else is revealed but this and its honour 
 and glory " ; that, finally, the history of the world " has 
 constituted the rational necessary course of the World 
 Spirit." What is here really meant ? Hegel goes on to 
 declare that the laws of cosmic movement " are Reason, 
 implicit in the phenomena in question " : but, of course, 
 adds that the cosmic bodies are not conscious. Soon he 
 proceeds to say that " Reason is thought conditioning itself 
 with perfect freedom " ; and his reconciliation of this with 
 the notion of Reason as implicit in phenomena — in the 
 movements of unconscious bodies— seems to consist in 
 saying that there is a fundamental distinction between an 
 abstract principle and its " determinate application and 
 concrete development." But in the same page he says that 
 " Divine Providence is Wisdom, endowed with an infinite 
 Power, which realises its aim — viz., the absolute rational 
 design of the world " ; and in the next, after some incon- 
 clusive reflection, he tells us that " our earnest endeavour 
 must be directed to the recognition of the ways of Providence, 
 the means it uses, and the historical phenomena in which 
 it manifests itself." It is after this that he breaks out as 
 aforesaid about God's wishing no narrow-hearted souls or 
 empty heads for his children.
 
 Letter V. LETTERS ON REASONING 71 
 
 Here,' despite the dialectic advantage derived by Hegel 
 from sheer laxity of expression — the advantage, that is, of 
 keeping us in hesitation as to his meaning— it becomes 
 finally clear that he has committed in his whole argument 
 thus far the self-contradiction reached in that explosion 
 about God's wishes. 
 
 He makes cosmic Reason alternately the process of events 
 and a force directing events (" Reason directs the world " is 
 one of his phrases : "Reason is the substance of the Universe," 
 is another) ; and after laying it down that every event is 
 " providential " he tells us to recognise the " ways of 
 Providence," a proposition which either limits those ways 
 to special orders of phenomena or amounts to the mere 
 nullity that we ought to recognise what we perceive. This 
 contradiction he seeks, as it were, to brazen out when he 
 goes on to say that he is undertaking " a justification of 
 
 the ways of God so that the ill that is found in the 
 
 world may be comprehended, and the thinking Spirit recon- 
 ciled with the fact of the existence of evil." And how does 
 he propose to effect the reconciliation ? In the only way 
 possible, he says ; that way being " by recognising the 
 positive existence, in which that negative element is a 
 subordinate and vanquished nullity." Try to imagine a 
 " vanquished nullity " ; and then ask yourselves why there 
 was any need to trouble about reconciling the discord of 
 a vanquished nullity with " the thinking spirit." " Evil," 
 says the philosopher, " has not been able permanently to 
 assert a competing position." Has not been able ! This, 
 after we have learned that Reason directs the whole world ; 
 that all acts are rational acts and divine acts ; that in the 
 world " nothing is revealed but reason and its honour and 
 glory." What then is this Evil which is a nullity, and cannot 
 permanently compete, and yet is an expression of Reason, 
 and nevertheless needs to be reconciled with Spirit ? Does 
 the philosopher believe what he has been saying ? 
 
 At this critical point he avows that "an adequate
 
 72 LETTERS ON REASONING Letter V. 
 
 definition of Reason is the first desideratum wnthout 
 
 such a definition we get no further than mere words." It 
 is most true ; and the avowal reminds us that we are in 
 presence of a strenuous effort to think the universe into an 
 order, to comprehend it, to explain it. Terms have been 
 transformed, definitions outraged, far more seriously than 
 we saw to be the case in the cited passages of Jevons 
 and Robertson Smith ; but the psychology of the process 
 seems to be different. Hegel is vehemently struggling all 
 the while with his problem : his incoherences of phrase are 
 the expression of his difficulty. But the more clearly we 
 realise his absorption, the intensity of his desire, the 
 pressure of his conviction that it all must be explicable, the 
 more clearly, I think, do we feel that his quest is vain. He 
 is in the toils of an insoluble enigma, and his desire to 
 solve it can no more avail than the Will-to-Live can make 
 us immortal. The passion to solve the enigma, we begin 
 to see, is one of the lures of life. Hegel would not give up 
 the quest : here he is on all fours with the theosophist of 
 all ages ; and, where others affect to have revelations, he 
 doggedly insists with himself to the last that his intuition 
 must take form in a reasoned demonstration. 
 
 Inevitably the further demonstration is a repeated move- 
 ment in a circle. " What," he asks, " is the ultimate design 
 of the World ? " and he works out a fresh formula : " The 
 final cause of the World at large is the consciousness of its 
 own freedom on the part of Spirit." The next question is : 
 By what means is this consciousness reached ? and at once 
 we are launched on a theorem which, instead of presenting 
 us with a pervading Reason, gives us the " World-Spirit " 
 as proceeding through " Unreason," evil, ruin, corruption, 
 decay. Still he insists on finding a rational purpose ; and 
 once more he readjusts his struggling thought, avowing 
 that " Principle, Plan of Existence, Law, is a hidden, 
 undeveloped essence, which as such, however true in itself, 
 is not completely real. It becomes real, that is to say,
 
 Letter V. LETTERS ON REASONING 73 
 
 through Will, human passion, desire, action ; and it is 
 admitted that " the history of mankind does not begin 
 with a conscious aim of any kind." Yet it is in the history 
 of mankind alone that Spirit, in this theory, can realise 
 itself; so that the "plan " of absolute "Reason " is simply 
 the slow emergence of an ideal in evolving man ; and 
 its absolute " Freedom " is nothing more than the human 
 consciousness of aspiration and of personality. And now, 
 after the express undertaking to show Evil as a vanquished 
 nullity, we have this smashing confession : — " The History 
 of the World is not the theatre of happiness. Periods 
 of happiness are blank pages in it, for they are 
 periods of harmony, periods when the antithesis is in 
 abeyance." 
 
 Just here, however, Hegel grasps a conception which 
 seems to him to yield him his justification. Even as wind 
 and water lend themselves to the making of iron and the 
 working of wood which form a house that will keep out 
 wind and water ; and even as the gravity of stones, their 
 tendency to fall, makes possible the wall which rises high 
 and remains firm, so do the passions of men " build up the 
 edifice of human society, thus fortifying a position for Right 
 and Order against themselves." But here, in the very flight 
 of verbal triumph, the collapse of the thesis is plain ; for 
 the upbuilding alleged is in terms of the case conditioned 
 by the hostile forces \ the house is there only because of 
 the forces of •■downfall, which never cease their attack; the 
 fortress of Right and Order must in the terms of the case 
 be for ever besieged ; and as the assault must be capable 
 of partial or chronic success if the alleged process is not to 
 be a farce, the Evil is no vanquished nullity, but an eternal 
 element in the " scheme " of things. 
 
 We are left with the familiar theistic sophism that Evil 
 is right because it is the condition of the existence of Good ; 
 a proposition maintained by the very mouths which most 
 wrathfully denounce Evil. They do but brazen out their
 
 74 LETTERS ON REASONING Letter V. 
 
 insoluble problem. Hegel must do as much throughout 
 his philosophy of history ; must condemn, indict, expose ; 
 must, in short, exhibit the World-Spirit — which was to be 
 proven Good, Reason, Freedom realising itself — as the mere 
 totality of human action, whereof the imperfection is always 
 driving the theosophist to frame schemes in which the Order 
 that was to be vindicated and explained is to vanish and 
 be superseded by a perfect one that in the terms of the 
 whole discussion will be unthinkable, because its Good 
 will not be made recognisable by Evil ! 
 
 I shall not attempt further to expound and criticise Hegel 
 for you : I am trying to help you to reason, not to supply 
 you with a rounded philosophy. But at least you will 
 catch the drift of my exposition — that Hegel is caught in 
 his own net ; that where some men commit fallacy by 
 force of commonplace prejudice, or the oversight of fatigue, 
 or inadequate reasoning faculty, he commits it in virtue of 
 an intense intuition that the solution of the infinite problem 
 is within his reach, and that he can somehow beat out that 
 solution in words. In the face of such an ardent aspiration 
 the ignoble insolences which we first considered are partially 
 excusable : we see, at least, that they are not the gist of 
 Hegel's exposition. But do not forget that to such 
 insolences the argument comes when it attempts to meet 
 the rebuttals of its procedure. 
 
 An accomplished Hegelian, to whom I once observed 
 that Hegel's pronouncements on practical issues — such as 
 those of politics and current religion — were always those of 
 quite unphilosophic men, and never in accord with his 
 philosophy, answered [in terms of one of Hegel's own 
 avowals, that in the all-thinking philosopher himself there 
 is a simple " John Smith," a man of common passions, who 
 has his say in the philosopher's name. It is verily so ; but 
 I have never been able to understand how the avowal could 
 content a Hegelian. In any case, it corroborates the posi- 
 tion I have so frequently taken up in these letters, that
 
 Letter V. LETTERS ON REASONING 75 
 
 emotion dictates, if not most, certainly far too many of our 
 reasonings. And though so intense a preoccupation as 
 Hegel's seems almost to exclude the notion that in his case 
 there can have been any of that " insincerity " on which 
 I have so often insisted, there is no escape at least from 
 the reflection that we ourselves will do well to put the 
 challenge of that conception to even the most intense of 
 our own presuppositions if we find that our reasonings in 
 support of them are not at every joint impenetrable to the 
 sharpest pressures of the test of consistency. 
 
 I need hardly point out to you in so many words that 
 Hegel's miscarriage is reducible to an identification of his 
 own mental processes with the infinite process of the 
 Universe. There is this justification for his calling the 
 infinite process " reason," that it becomes significant for us 
 only as a process of causation, a perpetual " therefore," an 
 endless chain of " becauses." Such a chain is our reason. 
 But we merely addle our heads with words when we jump 
 to the notion that, even as our reason is a perpetual tracing 
 of " therefores," so the causation of things in all Nature 
 amounts to the sequence of therefores in an infinite Reason. 
 Reason is too weak a process of bead-stringing, when all 
 is said, to be without puerility likened to an inconceivable 
 totality of which we can but say negatively that it is no 
 mere sequence, but infinite in all directions, infinite qua 
 co-existences as well as qua sequences — that is to say, not 
 totally a " sequence " at all, since sequence is but an 
 expression of relativity. 
 
 Hegel, indeed, as we have seen, strikes down his own 
 truth in his effort to transcend the absolute. What we have 
 to learn, as the total lesson of Nature, is that every human 
 experience, like every state of natural objects, is a result, of 
 which it is our business to know the order, item by item. 
 Error and evil, delusion and crime and madness, are to be 
 so envisaged, if we would rationally face life : we must 
 understand them in order to correct or exclude them in
 
 76 LETTERS ON REASONING Letter V. 
 
 future. But if we begin to define them as aspects of an 
 Infinite Reason, parts of the plan of a Divine Wisdom, 
 we shall either paralyse our every test, and stultify our 
 corrective purpose, or else relapse into that theological 
 irrationalism which yields the maximum of cruelty in 
 human affairs, since it reconciles men to the deliberate 
 infliction of misery by way of punishing an evil that in 
 the terms of the argument is willed and wrought by Divine 
 Goodness. Of that negation of reason Hegel, I fear, was 
 not incapable. 
 
 We have repeatedly come, in this last discussion, to the 
 terminology of theism : Hegel's doctrine of Reason in 
 history being, in fact, a way of approaching and handling 
 the theistic problem. That problem is so likely to meet 
 you often in life that I am fain to lead you to it on each and 
 all of the many lines on which it is commonly approached, 
 hoping so to shorten somewhat for you the processes of 
 dialectic which stand between us and clear conclusions on 
 the matter. But I shall be doing you little service if I do 
 not set you re-thinking every one of my steps, checking 
 my chain of therefores, and finally looking at the problem 
 with your own eyes, seeking light for yourselves wherever 
 you find me obscure or unsatisfying. In other letters we 
 shall approach the problem on other lines of argument. If 
 I am helping you at all, you will find the pilgrimage not 
 uninteresting. At the time I write these letters for your 
 future reading, you being far too young to understand them, 
 you daily tackle me with countless comical " whys." I 
 would fain hope that when you read these pages you 
 will be as comprehensive in your curiosity, practising, 
 however, on your own faculties as you used to hammer at 
 mine.
 
 LETTER VI. 
 
 We have seen how even a trained logician like the late Pro- 
 fessor Jevons, working with a formal definition in a treatise 
 that aimed at scientifically exact statement, could lose sight 
 of his own chosen terms and thus seriously confuse his 
 argument, even when he was not swayed by any prejudice. 
 Unless, then, formal logic positively miseducates men for 
 argument, great must be the likelihood of such confusion 
 on the part of fluent writers who venture on long processes 
 of argument, over complex issues, without any trained sense 
 of the difficulties of exact reasoning. Especially great are 
 their logical risks when their argument deals, so to speak, 
 with the very stuff of argument, the processes of thought 
 and feeling which constitute opinion. 
 
 A notable instance lies to hand in a very interesting work, 
 which you will do well to read — the History of the Rise and 
 Ijifliience of the Spirit of Rationalism in Europe^ by Mr. 
 W. E. H. Lecky. I must warn you that Mr. Lecky, though 
 an industrious, is a somewhat careless historian. He has 
 done good service in bringing together much interesting 
 information on the history of opinion ; and he is usually to 
 be trusted when he gives details for which he cites his 
 authorities. When, however, he offers general statements on 
 processes of development prior to those which he treats at 
 length, he is at times extravagantly wrong. The assertion, 
 for instance, concerning Averroism, that " the teachings of a 
 stern and uncompromising infidelity flashed forth from Seville 
 and Cordova," ' is utterly misleading ; and no less so is the 
 proposition that " In the towns, paganism had arrived at the 
 
 1 Work cited, ed. 1S87, i., 48.
 
 78 LETTERS ON REASONING Letter VL 
 
 last stage of decrepitude when Christianity arose ; and there- 
 fore in the towns the victory of Christianity was prompt and 
 decisive."^ A "victory" which took three hundred years 
 to accompUsh is not plausibly to be called " prompt " ; and 
 the writer's own context implies that it was not " decisive," 
 inasmuch as " the different elements of paganism continued 
 to exist in a transfigured form, and under new names." 
 
 But what I want now to discuss with you is the intel- 
 lectual confusion revealed in the preface to Mr. Lecky's 
 work. It is, I think, purely a matter of fallacy— that is 
 to say, there is no moral bias at work. He sets out, as I 
 understand him, to maintain a proposition nearly identical 
 with one I have suggested to you in a previous letter — 
 namely, that beliefs and arguments which to men at a 
 certain stage of knowledge are plausible and convincing, 
 are by men of fuller knowledge seen at a glance to be worth- 
 less. The point is that when we get to knoAv a mass of the 
 most relevant facts bearing upon an erroneous traditional 
 belief, we readily dispose of it by arguments which could 
 not be thought of by those who had not those facts before 
 them, and had to deal with the belief, as it were, in vacuo. 
 Mr. Lecky's logical mishaps occur in his effort to state this 
 proposition in terms of the psychological processes of 
 normal conviction. 
 
 "When," he writes, "towards the close of the eighteenth century 
 the decUne of theological passions enabled men to discuss these [theo- 
 logical] matters in a calmer spirit, and when increased knowledge 
 
 produced more comprehensive views it was observed that every 
 
 great change of belief had been preceded by a great change in the 
 intellectual condition of Europe ; that the success of any opinion 
 depended much less upon the force of its arguments, or upon the 
 aijilities of its advocates, than upon the predisposition of society to 
 receive it, and that that predisposition resulted from the intellectual 
 type of the age. As men advance from an imperfect to a higher 
 civilisation they gradually sublimate and refine their creed 
 
 " The pressure of the general intellectual influences of the time 
 determines tlie predispositions which ultimately regulate the details of 
 
 ^ Id., p. 36.
 
 Letter VI. LETTERS ON REASONING 79 
 
 the belief. A change of speculative opinion does not imply an 
 
 increase of the data upon which those opinions rest, but a change of 
 the habits of thought and mind which they reflect. Definite arguments 
 are the symptoms and pretexts, but seldom the causes of the change. 
 Their chief merit is to accelerate the inevitable crisis. They derive 
 their force and efficacy from their conformity with the mental habits of 
 those to whom they are addressed. Reasoning which in one age would 
 make no impression whatever, in the next age is received with 
 enthusiastic applause." 
 
 Thus far the confusion consists in a failure to assign 
 fixed values to terms. What Mr. Lecky ought to have 
 said in the second of the sentences above quoted is that 
 every great change of belief had been preceded by many 
 smaller changes of belief. He writes of " intellectual con- 
 dition" and "intellectual influences" as if these were not 
 in terms of beliefs. Obviously they are. Instead there- 
 fore of saying that pressure of general intellectual influences 
 determines a predisposition which determines beliefs (that 
 is what Mr. Lecky's loose phrasing comes to), one should 
 say that beliefs on great or central issues are prepared or 
 determined by beliefs on smaller issues. 
 
 How, then, are those minor beliefs so altered as to affect 
 major beliefs ? We must answer, Either by simple definite 
 argument or by presentments of fact which evoke and clinch 
 definite argument. To say that definite arguments merely 
 " accelerate the inevitable crisis " is a fresh confusion. 
 There can be no " crisis " until definite arguments are 
 forthcoming. What Mr. Lecky should have said is that 
 definite arguments of an innovating kind on a great or 
 central issue have to be preceded by definite arguments on 
 minor issues if they are to be made acceptable. " Mental 
 habits" are substantially habits of belief. The result of 
 this initial obscurity in Mr. Lecky's thinking is that, with 
 the truth under his eyes, he falls into a definite self-con- 
 tradiction. 
 
 "This tone and habit of thought," he proceeds, " is created, not by 
 the influences arising out of any one department of intellect, but by the
 
 8o LETTERS ON REASONING Letter VL 
 
 combination of all the intellectual and even social tendencies of the 
 age. Those who contribute most largely to its formation are, I believe, 
 the philosophers. Men like Bacon, Descartes, and Locke have pro- 
 bably done more than any others to set the current of their age. They 
 have formed a certain cast and tone of mind. They have inti-oduced 
 peculiar habits of thought, new modes of reasoning, new tendencies of 
 inquiry." 
 
 Here, you will see, the original idea is completely lost 
 sight of. What the philosophers deal in are precisely the 
 "definite arguments " which the writer has told us are " the 
 symptoms and pretexts, but seldom the causes, of the 
 change." If the philosophers can successfully "introduce 
 peculiar habits of thought, new modes of reasoning," all 
 the foregoing argument comes to nothing. The " con- 
 formity with the mental habits of those to whom they are 
 addressed," above declared to be necessary, has disappeared. 
 If a single writer can for his own time " form a certain cast 
 and tone of mind," what is the meaning of the assertion 
 that "a change of speculative opinions" implies "a change 
 of the habits of thought and mind which they reflect " ? 
 How came the innovating philosopher to innovate ? 
 
 The tangle is to be unravelled by going back to the truth 
 obscured at the start. Important changes of opinion, or 
 changes in important opinion, whether on the part of indi- 
 viduals or of numbers, are the result of minor changes of 
 opinion, or changes in minor opinions. Not that any one 
 minor change is necessarily primary in a given process : 
 many minor opinions may be revolutionised as the result of 
 a great change ; but the point is that no great change of 
 belief occurs save as a result of a number of smaller mental 
 adaptations — that is, changes of belief. Mr. Lecky, casting 
 back towards his escaped proposition, proceeds to say : 
 " But philosophical methods, great and unquestionable as is 
 their power, form but one of the many influences that con- 
 tribute to the mental habits of society." To say that a 
 "great" influence is "but one of many" is to misuse 
 language. The "many" cannot all be "great": the more
 
 Letter VI. LETTERS ON REASONING 8i 
 
 items there are in a given total, the less great are they rela- 
 tively. And Mr. Lecky proceeds to give illustrations which, 
 while they will help you to see w^hat he is driving at, do not 
 help out either of the last-cited propositions. 
 
 " Thus the discoveries of physical science, encroaching upon the 
 domain of the anomalous and the incomprehensible, enlarging our con- 
 ceptions of the reign of law, and revealing the connection of pheno- 
 mena that had formerly appeared altogether isolated, form a habit of 
 mind which is carried far beyond the limits of physics. Thus the 
 astronomical discovery that our world is not the centre and axis of the 
 
 material universe, but is an inconsiderable planet revolving with 
 
 many others around a sun which is itself but an infinitesimal point in 
 
 creation [Mr. Lecky means "in the universe"] has avast and 
 
 palpable influence upon our theological conceptions." 
 
 If this be so, the "vast" influence is primarily exerted 
 not by a philosopher's " new mode of reasoning " but by the 
 impact of a realised truth — a belief — on beliefs seen to be 
 correlative with that. [In point of fact, the belief in 
 question has only very gradually affected the theological 
 doctrines on which it properly bears : its influence, there- 
 fore, was not speedily " palpable," and it is not even now 
 " vast " as regards the majority of religious people ; but we 
 must agree that it ought logically to be very great.] 
 
 Mr. Lecky now proceeds to a fresh inconsequence. If his 
 
 views be correct, he says, 
 
 " they establish at once a broad distinction between the province of 
 
 the theologian and that of the historian of opinions The first is 
 
 restricted to a single department of mental phenomena, and to those 
 logical connections which determine the opinions of the severe reasoner ; 
 the second is obliged to take a wide survey of the intellectual influences 
 of the period he is describing, and to trace that connection of congruity 
 which has a much greater influence upon the sequence of opinions than 
 logical arguments." 
 
 Here we come upon the old fallacy as to what constitutes 
 a logical argument. It is quite true that there is a broad 
 distinction between the theologian as such and the historian 
 of opinion ; but it is not of the sort that Mr. Lecky suggests. 
 The theologian's business is to theologise, to state and argue
 
 82 
 
 LETTERS ON REASONING 
 
 Letter VL 
 
 his creed on lines of " revelation " or of a priori assumption 
 and deduction ; the business of the historian of opinion is to 
 show when, how, and why theological and other opinions 
 were commonly formed and altered. But "connection of 
 congruity " is an aspect of the exposition of both ; and it is 
 a gratuitous error to drive the historian of opinion off the 
 W^. \^ ground of " logical argument," though Mr. Lecky in a sense 
 
 is not exactly at home there. The theologian, indeed, is prac 
 tically driven to take note of the very facts in culture-history 
 
 - . r , to which the historian points as explaining changes in theo- 
 \ K i ' logical opinion, and to try to accommodate those facts to 
 '^ 'l his theology, or it to them, in terms of " connection of 
 congruity." 
 
 At this point Mr. Lecky diverges to the great question of 
 free-will, the discussion of which I reserve for another 
 letter : in the meantime let us follow him on his return to 
 the matters above considered. Here he will again serve 
 you as an object lesson in reasoning. 
 
 " Nothing," he writes, " can be more certain to an attentive observer 
 than that the great majority even of those who reason much about 
 their opinions have arrived at their conclusions by a process quite 
 distinct from reasoning. They may be perfectly unconscious of the 
 fact, but the ascendancy of old associations is upon them ; and, in the 
 j overwhelming majority of cases, men of the most various creeds conclude 
 their investigations by simply acquiescing in the opinions they have been 
 ' taught." 
 
 I The last statement is, I think, perfecdy true. Professor 
 Bain, for his part, remarking on the "strong probability 
 that any given individual has never exercised any independent 
 judgment in politics or in religion," declares that a " hundred 
 to one is a safe estimate of such a probability." ^ But this 
 is not the same thing as saying that most " even of those 
 who reason much about their opinions have arrived at their 
 conclusions by a process qjiite distinct from reasoning" 
 Observe Professor Bain's circumspect phrase : " Any inde- 
 
 ^ Induciion, 2nd ed., p. 136.
 
 Letter VI. LETTERS ON REASONING 83 
 
 pendent judgmenl " : he docs not say "any judgment." 
 Any "investigation" is a process of reasoning. What Mr. 
 Lecky should have said is that the processes of reasoning of 
 most people arc incomplete, short-sighted, relatively " un- 
 critical," uncandid. As I have tried to show you, there is a 
 heavy presumption that our errors of reasoning may be 
 traced to faulty moods, to prejudice, to temper, to hasty 
 belief in propositions of fact ; but none the less the process 
 of error is a process of reasoning. Error is a mode of judg- 
 ment : broadly speaking, it is an incomplete process of 
 judgment. 
 
 This holds good, obviously, of errors which are not mere 
 echoings of one's teachers. As Mr. Lecky himself puts it, 
 " the love of singularity, the ambition to be thought intel- 
 lectually superior to others, the bias of taste, the attraction 
 of vice, the influence of friendship, the magnetism of genius 
 — these, and countless other influences all determine con- 
 clusions." He is describing the case of "those who have 
 diverged from the opinions they had been taught"; and he 
 here implies that they are many ; but again he is confused, 
 for the influences of friendship and genius are forms of 
 " teaching." He then goes on : — 
 
 "The number of persons who have a rational basis for their belief 
 is probably infinitesimal ; for illegitimate influences not only determine 
 the convictions of those who do not examine, but usually give a 
 dominating bias to the reasonings of those who do. But it would be 
 manifestly absurd to conclude from this that reason has no part or 
 function in the formation of opinions." 
 
 Quite so, you will say with me. I think that Mr. Lecky 
 would have done better tiot to assert the absurdity as he 
 actually did, when he is thus forced to unsay his dictum. 
 " All that we can rightly infer is," he continues, " that the 
 process of reasoning is much more difficult than is commonly 
 supposed." Precisely ! The pity is that Mr. Lecky did not 
 at this stage profit by the discovery, to the extent of recasting 
 his preface. Just after the avowal he repeats without 
 
 H
 
 84 LETTERS ON REASONING Letter VL 
 
 hesitation his formula that " the opinions of a given period 
 are mainly determined by the infellecUial condition of 
 society"; as if opinions in general were not elements in the 
 intellectual condition. 
 
 We begin to suspect an inaptitude for right statement on 
 Mr. Lecky's part when we find him going on thus : — 
 
 " Those who have appreciated the extremely small influence ot 
 definite arguments in determining the opinions either of an individual or 
 of a nation — who have perceived how invariably an increase of civili- 
 sation implies a modification of belief, and how completely the con- 
 troversialists of successive ages are the puppets and the unconscious 
 exponents of the deep under-current of their time, will feel an intense 
 distrust of their unassisted reason, and will naturally look for some 
 guide to direct their judgment. I think it must be admitted that the 
 general and increasing tendency in the present day is to seek such a 
 guide in the collective wisdom of ?nankind as it is displayed in the 
 developments of history. In other words, the way in which our leading 
 thinkers, consciously or unconsciously, form their opinions is by endea- 
 vouring to ascertain what are the laws that govern the successive modi- 
 fications of belief." 
 
 It is quite true that a study of the strenuous error of past 
 thinkers may fitly set up in us a distrust of our unassisted 
 reason : that is to say, seeing how many false beliefs have 
 been fiercely held because of ignorance of the relevant facts, 
 we are moved to seek on all hands for knowledge on which 
 to build our opinions. But to say that we find a guide in 
 the " collective wisdom of mankind," and that that wisdom is 
 " displayed in the developments of history," is to use glaring 
 misnomers. What constitutes " the collective wisdom of 
 mankind " it is very hard to say ; but certainly there is 
 nothing plausibly to be so called in the mere process of 
 change. Past error is the collective ««wisdom of mankind : 
 present beliefs, in terms of the very doctrine of develop- 
 ment, are to be suspected at every point ; and many of 
 those held by the largest numbers are, in the opinion of 
 minorities, as false as any of the past. To "assist" your 
 reason you turn, not to the beliefs of mankind in mass, but 
 to the reasonings and researches of the studious few. It is
 
 ■^ 
 
 Letter VI. LETTERS ON REASONING 85 
 
 only they who can — at least it is only they who do — think 
 out " the laws that govern the successive modifications of 
 belief." 
 
 Broadly speaking, these laws are : that men tend to believe 
 (i) as they were taught ; (2) as their economic interests lie ; 
 (3) as their knowledge guides them. The last-named factor 
 of necessity operates slowly, knowledge beginning with the 
 few, and only with difficulty reaching the many. To realise 
 this is not to be impressed by the collective wisdom of 
 mankind. Where interests can be directly and forcibly 
 affected, opinion will be affected : thus in the English 
 Reformation the processes of plunder which revolted some 
 made staunch Protestants of others — those who profited by 
 the plunder. They became highly receptive to arguments 
 that formerly they would have rejected. But where new 
 doctrines do not set up or are not helped by an economic 
 interest — where they have against them both vested interests 
 and established teaching, their fortune is of the hardest. 
 The collective wisdom of mankind, in this aspect, is a very 
 poor affair. 
 
 Mr. Lecky would probably admit all this ; but his careless 
 reasoning never lets it clearly appear. He speaks of the 
 developments of science and philosophy and industrial life 
 as going on " till the period when conclusions [which] the 
 reason had once naturally and almost instinctively adopted 
 seem incongruous and grotesque " — here admitting after all 
 that wrong beliefs are reached and held by way of reasoning 
 — and then puts his case thus : — 
 
 *' When an opinion that is opposed to the age is incapable of modifi- 
 cation, and is an obstacle to progress, it will at last be openly 
 repudiated ; and if it is identified with any existing interests, or asso- 
 ciated with some eternal truth, its rejection will be accompanied by 
 paroxysms of painful agitation. But much more frequently civilisation 
 makes opinions that are opposed to it simply obsolete. They perish by 
 indifference, not by controversy. They are relegated to the dim 
 twilight land that surrounds every living faith ; the land of the unrealised 
 and the inoperative." 
 
 H 2
 
 86 LETTERS ON REASONING Letter VL 
 
 I commend this to your notice as a sample of how you 
 ought not to write or talk on sociological questions : a 
 sample of the muddle of notions that may result from the 
 free movement of a rhetorical style. First " an opinion 
 that is opposed to the age " is the loosest of descriptions. 
 It may fitly be applied to a new opinion, a disturbing 
 doctrine, such as were the theories of Copernicus and 
 Newton and Laplace and Darwin and the geologists when 
 they were first put forth. Yet Mr. Lecky means by his 
 phrase " an old or estabUshed opinion which is inconsistent 
 with a number of recently-acquired opinions." By "some 
 eternal truth " he means " religious belief," and the propo- 
 sition would be as true if he had written " popular 
 delusion." He speaks, finally, of opinions that perish, yet 
 do not perish, being relegated to the " land of the unrealised 
 and inoperative." He might have placed in that landscape 
 a good many opinions that are professedly held as sacred — 
 for instance, the saying that we ought to love our neighbour 
 as ourselves. 
 
 Of what was he thinking ? Let us turn to his book for 
 light. He says in the preface : — 
 
 " My object in the present work has been to trace the history of the 
 spirit of Rationalism ; by which I understand, not any class of definite 
 doctrines or criticisms, but rather a certain cast of thought, or bias of 
 reasoning, which has during the last three centuries gained a marked 
 ascendancy in Europe. The nature of this bias will be exhibited in 
 the ensuing pages, when we examine its influence upon the various 
 forms of moral and intellectual development." 
 
 It turns out, however, that he does deal with quite a number 
 of definite doctrines and criticisms, and that he has not 
 followed the effect of the rationalistic cast of thought in its 
 length and breadth. The word Rationalist has had a 
 notable history. In Germany, about 1600, it was applied (i) 
 to certain Aristotelian humanists at Helmstadt by their 
 opponents. In England, in Bacon's day, it was applied to (2) 
 the physicists who proceeded on a priori theory in disregard
 
 Letter VI. LETTERS ON REASONING 87 
 
 of experience. During the Civil War, it was applied to (3) 
 a small sect or group who professed to " follow reason " in 
 their religion and their politics. A little later, we find it 
 applied to Unitarians and Deists. About the end of the 
 seventeenth century and in the first quarter of the eighteenth, 
 it signified one of those Christians who claimed to give 
 rational grounds for their religious tenets. Those applica- 
 tions having disappeared from common usage, a century 
 later the name Rationalism was commonly applied to the 
 critical method of certain liberal theologians in Germany, 
 who, following a method set up by the French Peyrere and 
 the English Toland, sought to find natural explanations of 
 miracle-stories in the Bible. In the terms of the case, it was 
 not of the former modes of Rationalism that Mr. Lecky 
 seeks to trace the history, nor even of the last, though he 
 glances at that. And save when he is dealing with such 
 indirect influences as that of recovered pagan art on the art of 
 the Renaissance, his own pages constantly show that " defi- 
 nite aguments " played a great part in the progress he notes. 
 In his opening chapter, for instance, he deals with the 
 belief in magic and witchcraft, and its decline ; and he 
 himself makes it partly evident that definite arguments 
 served in turn to promote, to maintain, and to discredit it. 
 There was a whole library of works written to propagate the 
 belief in the later Middle Ages ; and it is admitted that, at 
 a time when freethinking arguments could not safely be 
 printed, the freethinkers set their faces against it.^ There 
 must therefore have been a great deal of definite argument 
 before Montaigne avowed his unbelief; and he in turn, by 
 Mr. Lecky's admission, ^ met the superstition by " a mode 
 of argument which was destined long afterwards to assume 
 a most prominent place in theological controversy " — the 
 argument, namely, used by Hume against miracles, that it is 
 more likely that men's senses or their testimony should err 
 than that the normally recognised laws of Nature should be 
 ^ Work cilcd, pp. 80, 97, notes. - lb., p. 92.
 
 88 LETTERS ON REASONING Letter VL 
 
 violated. To say of Montaigne that " the bent and 
 character of his mind led him to believe that witchcraft 
 was grossly improbable" — thus giving the whole credit to 
 his special genius — is to misstate the historical facts. In his 
 youth he was both credulous and fanatical ; and it was only 
 after many years of painfully educative experience that, 
 recoiling alike from Catholic and Protestant fanaticism, he 
 became broadly critical of all credulities.' During these 
 years he must have heard much " definite argument." 
 
 Mr. Lecky, sticking to his thesis, after admitting that the 
 " modes of thought" of Montaigne and his disciple Charron 
 persisted, goes on to insist that 
 
 "Though the industry of modern antiquaries has exhumed two or 
 three obscure works that were published on the subject, those works 
 never seem to have attracted any serious attention, or to have had any 
 appreciable influence in accelerating the movement. It presents a 
 spectacle, not of argument or of conflict, but of a silent evanescence 
 and decay." 
 
 At this stage, the fallacy into which we have seen Mr. Lecky 
 slipping in his preface through sheer laxity of thought and 
 expression has become a positive delusion : so important 
 is it to keep a watch on all our arguments. The process he 
 here speaks of was not and could not be silent : it is absurd 
 so to conceive it. Even if the other books to which he 
 refers were obscure, this would not alter the fact that those 
 of Montaigne and Charron to which he had referred had 
 an immense vogue and influence. To publish such books 
 in those days was a hazardous undertaking : Montaigne 
 profited by royal protection, and Charron ran risks. But 
 though the others were less popular, it is quite gratuitous to 
 say that they " never seem to have attracted any serious 
 attention." The work of Gabriel Naude, Apologie pour les 
 grands hommes soupco7inez de Magie, won a good deal of 
 
 ' See the Iiitroditiliou aiix Essais de Moiita/i^iie, l>y E. Champion, 
 1900, for a study of the great essayist's dcvcl(>])mcnt. Vou will find 
 this book well worth reading if, as I hope will be llie case, you are 
 interested in Montaigne.
 
 Letter VI. LETTERS ON REASONING 89 
 
 serious attention, though, as Mr. Lecky admits, it is 
 " extremely tiresome " — a sufficient reason why it should not 
 be popular, whatever its arguments were. And it " stands to 
 reason," as we say, that the unwritten discussion must have 
 been boundless. What were people to talk about ? what do 
 they ever talk about, if not the matters on which opinions 
 are changing ? Mr. Lecky tells in so many words that " the 
 
 priests continued to anathematise as infidels all who 
 
 questioned " the reality of witchcraft, and that " many of the 
 
 lawyers maintained the belief with equal pertinacity." 
 
 Against whom ? True, many people who heard the 
 unbelievers anathematised might hesitate to speak out ; but 
 outspoken doubters there must have been. It is incredible 
 that a belief once so general and so intense can have 
 declined without abundant discussion ; and further research 
 would have shown Mr. Lecky further traces of that in France 
 in the first half of the seventeenth century. 
 
 So Mr. Lecky decides that in England the able treatise of 
 Reginald Scott (1584) "as a matter of fact exercised no 
 appreciable influence " ; but it is his method, or his fixed 
 pre-supposition, that prevents his appreciating influences like 
 Scott's. They cannot reach the mass of a nation at such 
 a stage of culture as that of Elizabethan England until they 
 have passed by way of the more thoughtful few. Now, 
 there are many traces of educated scepticism in the 
 Elizabethan period, and in the reign of James. Mr. Lecky 
 goes astray because he looks for a kind of trace that the 
 conditions did not admit of — abundant literary allusions. 
 But the very fact that Reginald Scott's treatise was burned 
 by the order of King James proves that it was then dangerous 
 to proclaim such opinions, though on the other hand such 
 a treatise could not have been issued if the author were 
 alone in his views. 
 
 How important exact knowledge is to right reasoning on 
 such matters becomes newly clear when we find Mr. Lecky 
 pronouncing that Shakspere probably believed unquestion-
 
 90 LETTERS ON REASONING Letter VL 
 
 ingly in witchcraft — this on the strength of the witches in 
 Macbeth and the treatment of Joan of Arc in Henry VI. 
 But the experts are now agreed that Shakspere had no hand 
 whatever in the Joan of Arc scenes ; and the witches in 
 Macbeth merely represent the natural resort of play-makers 
 in such an age to any machinery that would attract audiences. 
 There is no more reason to think that Shakspere believed in 
 witchcraft than that he believed in the reality of such a 
 phenomenon as the ghost in Hamlet. Those expedients were 
 common on the stage before his time : he used them like 
 another, save that, in Alacbeth, he very clearly indicates the 
 unreality of the spectre of Banquo seen by Macbeth, and 
 makes Macbeth's purpose arise independently of the witches. 
 But a playhouse was the last place where an open attack on 
 a popular superstition could be thought of. Even to publish 
 a book of such a drift would have been to incur intense 
 odium, and probably personal danger, all the more because 
 under the different stimuli of King James and the Puritan 
 clergy the popular belief was becoming more fanatical. 
 Books against witchcraft were lacking, just as books against 
 the orthodox creed were lacking ; but we have clear testi- 
 mony that in private talk heresy was often avowed. How, 
 once more, should such opinions spread if they were never 
 discussed ? Mr. Lecky notes that the anti-sceptical work 
 of the otherwise sceptical Glanvil, Saddiicistiius Triuwphatus, 
 opens with a picture of the rapid progress of disbelief in witch- 
 craft. How could that disbelief be known if it were not much 
 discussed ? The book, again, " had an extraordinary success." 
 How could this be if its arguments were not talked of? 
 
 It is true that, on the other hand, there were not many 
 books on the sceptical side, and that those published were 
 not by distinguished authors ; but here again Mr. Lecky 
 misreads the case. At that period, heresy of all kinds 
 spread more by manuscript than by books : as he actually 
 notes, Cudvvorth, himself accused of leanings to atheism, 
 declared, as to wizards and magicians, that " our so confident
 
 Letter VI. LETTERS ON REASONING 91 
 
 exploders of them can hardly escape the suspicion of having 
 some hankering towards atheism." Under such circumstances 
 most unbelievers would be unwilling to print, and booksellers 
 to sell, books against the belief in witchcraft ; but how could 
 " confident exploders " be recognised if they did not argue ? 
 
 If you have any remaining doubt on the point, I need but 
 remind you that in the period from 1660 to 1690 there were 
 many declarations by clerical writers to the effect that 
 "unbelief" in general — that is to say, unbelief in Christian 
 dogmas — had become extremely common ; yet there were 
 only two recognised unbelieving authors — Herbert and 
 Hobbes — down till 1679 ; and their arguments do not go 
 very far to assail orthodoxy. It is surely clear, however, 
 that the bitter clerical complaints prove the occurrence of a 
 great deal of private discussion. Yet Mr. Lecky persistently 
 repeats that " the scepticism that was already pervading all 
 classes was steadily and silently increasing." ^ A few pages 
 before ^ he had avowed that " there was manifested in some 
 classes a strong disposition to regard witch stories as absurd." 
 Such manifestation could not take place without definite 
 argument. 
 
 \Vhen we come to Mr. Lecky's own comparatively reason- 
 able account'' of the various influences which promoted 
 scepticism, it becomes still clearer that he has fallen into a 
 confusion of ideas. 
 
 "The reaction against the austere rigidity of the last Government had 
 prochiced among the gayer classes a sudden outburst of the most derisive 
 incredulity. From mocking the solemn gait, the nasal twang, and the 
 aflectcd phraseology of the Puritans, they naturally proceeded to ridicule 
 their doctrines ; and having soon discovered in witchcraft abundant 
 materials for their satire, they made disbelief in it one of the tests of 
 fashion. At the same time the higher intellectual influences were tending 
 strongly to jiroduce a similar movement among the learned. Hobbes, 
 who was the most distinguished of living philosophers, had directed all 
 the energies of his scepticism against incorporeal substances, had treated 
 with unsparing ridicule the conceptions of demons and of apparitions, 
 
 ^/</., p. 121. 2 p_ 108. s pp, 109-110.
 
 92 LETTERS ON REASONING Letter VL 
 
 and had created in his disciples a predisposition to regard them as below 
 contempt. A similar predisposition was formed by the philosophy of 
 Bacon, which had then acquired an immense popularity. The Royal 
 Society had been just established ; a passion for natural philosophy 
 much resembling that which preceded the French Revolution had 
 become general ; and the whole force of the English intellect was 
 directed to the study of natural phenomena, and to the discovery of 
 natural laws. In this manner there was formed a general disposition to 
 attribute to every effect a natural cause, which was soon followed by a 
 conviction of the absurdity of explaining phenomena by a supernatural 
 hypothesis, and which rapidly discredited the anecdotes of witches. 
 There does not appear to have been any very careful scrutiny of their 
 details, yet there was a growing indisposition to believe them, as they 
 were discordant with the modes of thought which the experimental 
 philosophy had produced." 
 
 Observe here how the fallacious formula of " silent " change, 
 without " definite argument," has led the historian to mani- 
 pulate his facts. He limits the anti-Puritan reaction to " the 
 gayer classes," and the influence of Hobbes to " the learned." 
 Now, there is plenty of evidence that Hobbes was much 
 read, and still more quoted, among the "gayer classes," who 
 thus had definite argument as well as personal prejudice to 
 influence and guide them. If the philosophy of Bacon had 
 an " immense popularity," the study of that cannot have 
 been limited to the learned. It is an error, again, and one 
 which damages Mr. Lecky's own thesis, to say that the 
 whole force of the English intellect was directed to natural 
 science : intellects like those of More and Cudworth were 
 admittedly not so directed; and Mr. Lecky had just been 
 telling us that there was a large new literature in support of 
 the belief in witchcraft. Finally, the opposing movement 
 cannot have avoided " any very careful scrutiny " of the 
 orthodox case. If Glanvil's book was widely read, it must 
 have been widely discussed in detail. 
 
 It is for a special reason that I have troubled you with 
 this long analysis of Mr. Lecky's confusion of words and 
 ideas. His thesis that opinions on important issues change 
 " silently " and " without definite argument " is either
 
 Letter VI. LETTERS ON REASONING 93 
 
 adopted or independently set up by many people as an 
 excuse for letting intellectual progress take care of itself. 
 Wherever any odium attaches to the maintenance of new 
 truth, many people who recognise the truth are ready thus 
 to shirk the trouble and possible injury that come from 
 avowing it. In my owti case, I have been advised a 
 hundred times by prudent friends not to take part in the 
 criticism of beliefs which they regard as false and foolish. 
 Such beliefs, they argue, will pass away " silently " ; the 
 corrective criticism is " in the air " ; the true doctrine will in 
 time inevitably drive out the false. Such argument, I am 
 satisfied, is not only fallacious but " insincere," in the sense 
 I have given to that term. 
 
 Those reasoners, I find, never bethink them to apply 
 their rule to disputes over politics where their views accord 
 with those of the majority, or even to disputes over scientific 
 doctrine where no odium attaches to their own position. 
 Precisely where, on their theory, the truth might be left to 
 take care of itself, because the interests of a majority are on 
 its side, they busy themselves on its behalf ; and where no 
 great interests or prejudices are arrayed against a doctrine of 
 which they approve, they see fit to push it. The "air," in 
 these cases, they regard as a non-conductor : it is only when 
 there are storms in it, so to speak, that they counsel a 
 masterly silence. And Mr. Lecky, as I say, tends to keep 
 them in countenance. 
 
 Well, I never found that any truth got into the air and so 
 did its work unless men put it there and kept it there. It 
 is because some men are active and courageous that others 
 can inactively breathe a sound mental air. And I trust that, 
 though that course is not the way to wealth or the widest 
 popularity, you will agree with me in ranking the pioneers, 
 the volunteers on forlorn ho[)es, higher than the prudent 
 ones, and that you will rather fight for the unpopular truth 
 than for the successful one. In any case, I hope you will 
 recognise the errors of reasoning and of historical statement
 
 94 
 
 LETTERS ON REASONING Letter VL 
 
 by which Mr. Lecky has given colour to the counsel of 
 putting your hands in your pockets and your faith in a 
 speechless evolution. 
 
 I do not dispute, indeed, that certain changes of opinion 
 take place with relatively little discussion, in virtue of the 
 alterative force of mere social conditions. In a long 
 period of peace, for example, men have been seen to grow 
 less prone to brutal sports and to duelling. When brutal 
 sports decline, there is likely to be a growth of kindness 
 towards animals. But even in these cases, where the 
 initial force is the simple change in the conditions, no 
 considerable change in opinion takes place without dis- 
 cussion. The literature of such discussion, where it gets 
 into print, is apt to pass out of sight precisely because it has 
 carried its point ; but if you will make a careful research, 
 you will find that there was a great deal of literature on 
 duelling and on the kind treatment of animals before in 
 England the first was discredited and the second made a 
 matter of legislative action. 
 
 To make the point clear, let us take one more of Mr. 
 Lecky's applications of his " no-argument " doctrine. Telling 
 of the movement for the abolition of judicial torture, he 
 writes ^ : — 
 
 " In France, probably the first illustrious opponent of torture was 
 Montaigne, the first of the French sceptics ; the cause was soon after- 
 wards taken up by Charron and by Bayle ; it was then adopted by Vol- 
 taire, Montesquieu, and the Eiicycloprcdists ; and it finally triumphed 
 when the Church had been shattered by the Revolution. In Spain, 
 torture befjan to fall into disuse under Charles III., on one of the 
 few occasions when the Government was in direct opposition to the 
 Church. In Italy the great opponent of torture was Beccaria, the 
 friend of Melvetius and of Holbach, and the avowed exponent of the 
 principles of Rousseau. Translated by Morellet, commented on by 
 Voltaire and Diderot, and supported by the whole weit^dit of the French 
 philosophers, the work of Beccaria flew Iriuniplianlly over Europe and 
 vastly accelerated the movement that produced it. Under the influence 
 
 ' ^"'•, PP- 330-2-
 
 Letter VI. LETTERS ON REASONING 95 
 
 of that movement the Empress of Russia abolished torture in her 
 dominions, and accompanied the abolition by an edict of toleration. 
 Under the same influence, Frederick of Prussia, whose adherence to 
 the philosophical principles was notorious, took the same step, and his 
 example was speedily followed by Duke Leopold of Tuscany." 
 
 Here, one would suppose, the effect of " definite argument" 
 was tolerably clear : read Beccaria, and you will see how 
 ratiocinative he is. But Mr. Lecky bethinks himself that 
 he must riot seem to admit the efficacy of definite argument ; 
 and so he gives us this corrective passage : — 
 
 "Nor is there, upon reflection, anything surprising in this. The 
 movement that destroyed torture 7vas much less an intellectual than an 
 emotional niovemeiit. It represented much less a discovery of the 
 reason than an increased intensity of sympathy. If we asked what 
 positive arguments can be adduced on the subject, it would be difficult 
 to cite any that was not perfectly familiar to all classes at every period 
 of the middle ages. That brave criminals sometimes escaped, and that 
 timid persons sometimes falsely declared themselves guilty ; that the 
 guiltless frequently underwent a horrible punishment, and that the 
 moral influence of legal decisions was seriously weakened — these argu- 
 ments, and such as these, were as much truisms in the eleventh and 
 twelfth centuries as they are at present. Nor was it by such means 
 that the change was effected. Torture was abolished because in the 
 progress of civilisation the sympathies of men became more expansive, 
 their perceptions of the sufierings of others more acute, their judgments 
 more indulgent, their actions more gentle." 
 
 I invite you to dissect this passage in its turn with some 
 closeness. Once again, concern for an erroneous formula 
 has led Mr. Lecky to mistake seriously the facts of history. 
 He would have us believe that the treatise of Beccaria, 
 eagerly commented by Voltaire and Diderot, read with deep 
 interest throughout Europe, and acted on by Catherine and 
 Frederick and Leopold, consisted mainly of truisms, " per- 
 fectly familiar to all classes at every period of the middle 
 ages." This is sad nonsense. The arguments in question 
 were not truisms, and they were not familiar to any class at 
 any period of the middle ages. Mr. Lecky does not give a 
 scrap of evidence to show that they were. He cites the
 
 96 LETTERS ON REASONING Letter VL 
 
 chapter in Augustine's Ciiy of God in which the irrationaHty 
 of torture is very clearly put ; but St. Augustine was the 
 most intellectual of the Christian Fathers ; and his City of 
 God, being an argument against the pagans of his own day, 
 is not at all likely to have been the most familiar of his 
 treatises in the middle ages even for churchmen. Certainly 
 the matter was never so discussed in the middle ages as to 
 make the arguments against torture familiar even to lawyers. 
 When Montaigne urged some of them, late in the sixteenth 
 century, he seems to have been as suggestive to his readers 
 on that theme as on witchcraft ; and I have noticed only 
 three wTiters cited as opposing torture in the century and a 
 half between Montaigne and Beccaria. 
 
 Even in England, where torture had been abandoned 
 after the Revolution of 1688, there is nothing to show that 
 the arguments against it had become familiar. It seems to 
 have been dropped mainly because English common law, 
 not being consciously based, like the law of most of the 
 continental nations, on that of Rome, did not recognise the 
 practice ; and its consequent association with the tyrannies 
 of royal prerogative brought it into political disrepute. 
 Finally, we have Beccaria's own avowal that he had been led 
 to his philosophical views only five years before writing his 
 book, the first influence upon him being that of Montesquieu, 
 and the second that of Helvetius, who, he says, " aroused 
 my attention for the first time to the blindness and miseries 
 of humanity." 
 
 It thus appears that sympathy may be aroused and 
 extended by a ratiocinative appeal. Mr. Lecky's fallacy 
 takes fresh form in his phrase, " much less an intellectual 
 than an emotional movement." It is quite true that some 
 movements are very largely emotional and very slightly 
 intellectual, in the common senses of those terms : but on 
 the other hand there can be no great " movement " of an 
 intellectual kind without its emotional side. Men can be 
 " moved " for a truth, for a reasonable teaching, as against
 
 Letter VI. LETTERS ON REASONING 97 
 
 an unreasonable practice ; and Mr. Lecky's own record 
 shows how essential to the humanitarian movement was the 
 reasoned exposition of Beccaria and his school. I want 
 you to pay special heed to this — that every judgment, every 
 process of reasoning, has its quantum of emotion : the bare 
 recognition of the correctness of a mathematical demon- 
 stration is emotion in minimis : one's first sense of the 
 justice and the irrefragability of a great philosophic or 
 scientific doctrine is a marked emotion ; and you will 
 remember Franklin's account of the intense stress of his 
 feeling when he had experimentally proved his hypothesis 
 that the lightning was the same thing as electricity. In 
 such a movement, then, as that for the abolition of judicial 
 torture, where so much human suffering was involved, there 
 must needs be much play of emotion ; but, once more, do 
 not be led by such reasoning as Mr. Lecky's to suppose that 
 humane reforms get somehow made without being argued 
 for. In this very case of Beccaria, as you may learn from 
 the preface to Mr. Farrer's translation of the Dei delitti e 
 delle pene, a thoughtful Scotch artist of that day was con- 
 vinced that the book belonged to " the category of Utopias, 
 of Platonic Republics, and other ideal governments, which 
 display indeed the wit, the humanity, and the goodness of 
 their authors, but which never have had, nor ever will have, 
 any influence on human affairs." No estimate of the kind 
 was ever further wrong. Beccaria's book had an immense, 
 a speedy, and a durable influence, revolutionising to a great 
 extent the criminal procedure of half of civihsed Europe, 
 and so gaining ground which has never been lost. 
 
 It ought to have been an irksome thought to that artist, if 
 he lived to have it, that such an attitude as his had not only 
 been mistaken, but had perhaps helped to retard a good 
 cause, whose possibilities of success were in the circum- 
 stances really great. I trust you will not have such mistakes 
 to look back upon. When you first read these letters you 
 will perhaps be still at the stage of cherishing the ideals
 
 98 LETTERS ON REASONING Letter VL 
 
 natural to schoolboys. I remember how, about the age of 
 fifteen or seventeen, I was deeply concerned about the 
 fighting strength of our country, not at all realising how 
 immensely more important are its social than its military 
 arrangements. I therefore could not appreciate at that age 
 the men and the movements which sought to better the life 
 of peace, with its " wrongs and shames " ; and it needed 
 "intellectual" experience to enable me to develop my 
 sympathies. You, I hope, will not be slower than I was. 
 The conservatism of boyhood and girlhood is not a very 
 serious matter ; but I know few less estimable political types 
 than the adolescent Conservative who has never attained to 
 hoping highly for human betterment because he has never 
 been able to extend his sympathy to the vast world of 
 toiling and heavy-laden humanity. 
 
 I need not again tell you, then, that I am not making 
 light of the forces of emotion when I urge upon you the 
 vigilant use of your reason. The transmutation of emotion 
 is in fact the end of all curative human action ; and the 
 main moral use of reason is to effect such transmutation. 
 But it is in the train of clear thinking only that emotion can 
 be trusted to run. The wrong argument, the wrong belief, 
 has its emotion like the right ; all the cruelties and iniquities 
 of history have proceeded on emotion ; and, as I have been 
 arguing, it is emotion that inspires the majority of bad 
 arguments and blinds men to truth. 
 
 In those very matters of the belief in -witchcraft and the 
 practice of torture, we have fresh illustrations of the process. 
 Nothing can be clearer than the inconsequence, the irra- 
 tionality, of judicial torture ; but the demonstration was 
 stolidly resisted by many men trained to dealing with 
 evidence. Why? Simply because their habits were fixed; 
 it irked them to hear younger men (it might be) arguing 
 that wliat they had been doing all their lives was unreason- 
 able ; and they tartly resisted the innovating doctrine ; 
 some proceeding, in the common way of conservatism, to
 
 Letter VI. LETTERS ON REASONING 
 
 99 
 
 charge all manner of evil bias on the reformers. Bcccaria 
 was told that he sympathised with crime. 
 
 In the case of the superstition of sorcery, again, we have 
 the signal cases of Glanvil and Bodin, the first a man of 
 markedly critical turn of mind on the side of natural science ; 
 the second one of the most powerful intelligences in the 
 France of his day, and notably rationalistic on the side of 
 religious dogmas. Both set themselves strenuously to 
 maintain the reality of witchcraft when many men of less 
 intellectual grasp and energy were beginning to dispute it. 
 Why did they thus err ? Apparently because they had long 
 been wont to believe unthinkingly in that particular doctrine, 
 because they came late to the facts and arguments which 
 discredit it ; and because their very energy of mind entered 
 into their resentment of the cavils which they may have 
 heard from men of less calibre. Let us say, if we will, that 
 their powers of judgment were unequally developed ; that 
 on some sides their common-sense was feeble ; or that they 
 had not the patience to work out the reasonable induction 
 from the pathological phenomena in which they saw evidence 
 of witchcraft. 
 
 In the same fashion, you will remember, some very able 
 and sceptical men of the Renaissance held by astrology, 
 while some men of far less ability rejected it either on 
 " scriptural " grounds or by reason of mere narrowness of 
 mind. But though the astrologers may often have been 
 able thinkers, we may safely infer, I think, that they went 
 wrong because of the zvish to believe the doctrine that the 
 relations of the stars in space at the moment of our birth 
 determine our careers. 
 
 And there is almost no limit, so far as we can gather from 
 history, to the possibility of perverse error of this kind — 
 error of emotional bias, of prejudice — on the part even of 
 critically-minded men. Gabriel Naud^, above mentioned, 
 was a good deal of a rationalist for his day, being sanely 
 sceptical about witchcraft where the great Bodin was
 
 lOO LETTERS ON REASONING Letter VL 
 
 credulous. Yet his personal and political prejudices led 
 him to condone such an act of ferocious wickedness as the 
 Massacre of St. Bartholomew — a crime denounced and 
 loathed even by many Catholics of the party which com- 
 mitted it. 
 
 It may seem to you, when you reflect on all this, that the 
 prospect before poor human nature is rather hopeless ; that 
 error is ubiquitous ; that there is no way of guarding against 
 it. Certainly none of us can wholly escape it : I shall not 
 push you to what Voltaire called " the insane project of 
 being perfectly wise." But be sure that every one of the 
 errors and perversities of which I have spoken was demon- 
 strable at the time of its committal by the processes of 
 consistent reasoning. Every fallacy is an inconsistency ; 
 every moral error is an insincerity. If you can always 
 remember to revert to the test, "do as you would be done 
 by " — or, more strictly, do not as you would ?iot be done 
 by — you can never err as did Naude. And if you never 
 make an affirmation or a negation that is inconsistent with 
 those on which you normally proceed in your beliefs, you 
 will run small risk of being the champion of a doomed 
 delusion. In short, the safeguard against the risks of 
 reasoning is just — more reasoning. 
 
 If we can correct the errors of Mr. Lecky, we may surely 
 correct some of our own before they have become inveterate. 
 Let us take one more glance at his handling of the theme 
 above considered— the relative influence of argument and 
 of other factors on men's opinions. When he comes to the 
 question of belief in future punishment (hell), he notes 
 rightly enough that theory and practice reacted on one 
 another : that religious theory had promoted cruel practice, 
 and that humanised practice in turn tended to humanise 
 religious theory. Then he goes on : — 
 
 " This gradual and silent transformation o. the popular conceptions 
 is doubtless chiefly due to the habit of educing moral and intellectual 
 truths from our own sense of right, rather than from traditional teach-
 
 Letter VI. LETTERS ON REASONING loi 
 
 ing, which has accompanied the decline of dogmatic teaching 
 
 Descartes, who was the chief reviver of moral philosophy, may be 
 regarded as its leading originator." ^ 
 
 Here once more there is both assertion and denial of a 
 silent process, without definite argument. Mr. Lecky goes 
 on to ascribe " a real though minor influence " to Descartes's 
 " purely spiritual conception of the soul," which made hell- 
 fire a grotesque irrelevance. But, a little further on,^ he 
 arrives at the concession that — 
 
 "When at last Descartes maintained that thought is the essence oi 
 
 the soul he contributed vitich to that frame of mind which made 
 
 men naturally turn with contempt from ghosts, visible demons, and 
 purgatorial fires." 
 
 I shall not ask you to go with me further on the track of 
 this historian's inconsistencies, though you will do well to 
 look out for them when you read him. We have suffi- 
 ciently made out, I think,, the fact that he phrased and 
 reasoned loosely, frequently forgot what he had written, and 
 so committed many errors against which moderate care 
 might have guarded him. These are all normal human 
 failings ; and if I am not very careful, you will doubtless 
 detect some of them in the course of these letters. Let 
 such discoveries, above all things, put you upon avoiding 
 such errors for yourselves. There is no other profit to be 
 drawn from the study of error. 
 
 1 r. 336. ■' P. 343- 
 
 1 2
 
 LETTER VII. 
 
 In a previous letter I said I should discuss with you in 
 another the question o'' " chance." I feel bound to do so 
 because that theme comes up in the course of many lines of 
 reasoning on the serious issues of life, and because it seems 
 to me to set up more confusions of argument than almost 
 any other. Let me warn you then, at the outset, that it is 
 a more difficult problem than it may at first sight seem ; on 
 which score I shall break up my letter into sections. 
 
 § I- 
 
 We saw how Professor Mintoused the expression : " some 
 other cause than chance." That is, he treated chance as a 
 " cause." Now, I think he would on challenge have 
 admitted that the expression was an error ; because to define 
 " chance " as one of a variety of possible causes of a given 
 phenomenon is to deprive of significance the term " cause " 
 itself, and by consequence to nullify the term " chance." 
 
 Let us consider what the word normally means. In such 
 expressions as : "I chanced to meet a friend at the cross- 
 roads"; "a stray dog chanced to enter the shop"; "by 
 chance I lit upon this passage in reading"; or "I chanced 
 to strike my foot against a stone," we mean simply that in 
 such cases the incident is unplanned and unexpected ; or, in 
 other words, that we had not recognised beforehand, or set 
 up by our will, the coincident lines of causation which 
 brought it about. We do not say — unless we speak very 
 thoughtlessly — that it was uncaused : every event, we 
 normally admit, has a cause or causes in previous events. 
 The movements of the stray dog are inferribly the results of 
 causes which we may or may not guess.
 
 Letter VII. LETTERS ON REASONING 103 
 
 The meaning of the term is brought out more precisely 
 when we say over a given phenomenon : this cannot have 
 happened by chance, meaning that we are sure it was pre- 
 arranged by someone's will, or that between two or more 
 of the details of the incident there is a " causal connection." 
 By " causal connection " we mean something that is excluded 
 by the ordinary use of the expression " by chance." If in 
 walking on the street I am nearly struck by a falling slate, 
 I do not infer that it was thrown at me : I surmise either 
 that it had been ill-fastened, and that at length, through a 
 series of minute causes which I cannot trace, it " chanced " 
 to slip loose as I was passing, or that someone working on 
 the roof unintentionally caused it to fall at that moment. 
 If there be a high wind blowing, I readily assign that as 
 the cause. Every step in the whole coincidence is clearly 
 " caused " \ but inasmuch as the forces which moved the 
 slate had no known connection with the motives or forces 
 which brought about my presence at that instant, I call the 
 coincidence a matter of " chance." If, again, it should 
 happen that a man whom I did not know lurched against 
 me in the street, I should reason in the same way, probably 
 surmising that he was tipsy. 
 
 But suppose that, in the course of a morning's walk, 
 several men should lurch against me. If they all seemed 
 to be tipsy, and there were many tipsy people about, I 
 should still infer " chance." But if they were the only 
 tipsy or apparently tipsy people I saw, I should probably 
 begin to suspect at the third encounter that they were not 
 tipsy, but were either pickpockets or persons with some 
 design against me. If the annoyance continued, my 
 suspicion would become a certainty. In excluding " chance," 
 then, I assume '" purpose." Needless to say, non-purposive 
 coincidences can be very remarkable. A few days before I 
 began this letter I invited a friend, whom I shall call A., to 
 dine with me at my club ; and he accepted. At the hour 
 fixed he did not arrive ; and, " chancing " to meet in the
 
 I04 LETTERS ON REASONING Letter VIL 
 
 lobby another friend, B., who knew A., I invited him to 
 wait a moment and dine with us. We waited over a quarter- 
 of-an-hour, when B. explained that he could not wait longer. 
 I then went to the dining-room with him, leaving at the 
 inquiry office a note of my whereabouts. Soon there was 
 brought to me, at dinner, a telegram which ran somewhat 
 thus : " Disappointed, unable to come : missed five train ; 
 no other till ten." The telegram came from a town some 
 distance away. Clearly, then, A., having gone thither on 
 some business, had failed to get back in time ; so B. and I 
 went on with our dinner. Ten minutes later, however, there 
 was brought me a message that someone was waiting for 
 me in the ante-room ; and on my descending, I found, 
 after all, A. The telegram, addressed to me by name but 
 not by initials, was not for me : it was unsigned, and not 
 from A. Though there were only a few other club members 
 with my surname, it chanced that one of them was expecting 
 a friend to dinner just as I was, and while my friend was 
 late, the other man's friend could not come at all. Finally, 
 A. had actually arrived immediately after the telegram ; but 
 by another " chance " the attendant in the inquiry office had 
 just then been relieved, and the new man had overlooked 
 my note, thus allowing A. to wait needlessly. 
 
 § 2. 
 
 The bearing of such an episode on what is called " the 
 logic of chance " will appear when we proceed to deal with 
 that. The ideas called up by discussion of the term 
 " chance " tend to centre chiefly round what are called games 
 of chance, such as dice-throwing : the word " chance," in 
 fact, comes etymologically from the "fall" of the dice; 
 and in connection with the phenomena of these games 
 there has arisen a mode of reasoning loosely called 
 " probability-logic." In the throwing of dice we have a 
 play of what is called "pure chance": that is to say, 
 assuming the dice to be "true," it is quite impossible to
 
 Letter VII. LETTERS ON REASONING 105 
 
 trace or forecast the series of impacts which determine how 
 they shall lie when thrown. From the very fact that we can 
 give no " reason why " certain numbers should turn up, men 
 would begin to grow suspicious if a particular player were 
 to go on many times throwing the same numbers ; and such 
 suspicions, one hears, used at times to lead in dicing-days to 
 the act of breaking the cubes, in order to ascertain whether 
 or not they were " loaded." It is notorious, however, that 
 in all games of chance there occur " runs of luck," inde- 
 pendently of any known fraud ; and this very phenomenon, 
 which might be supposed sufficient to convince us of the 
 incalculableness of " chance," has set both gamblers and 
 philosophers on the attempt to find a " law " thereof. The 
 gambler's theories one knows of only by hearsay : they are 
 reputed to be mostly fantastic ; but the thinkers have framed 
 a simple and, at first sight, satisfying theory, to the effect 
 that given results in games of chance occur in the long run 
 in the ratio of the " objective possibility." To keep the 
 matter as far as possible on non-mathematical lines, let us 
 take the simple case of pitch-and-toss. The tossed-up coin 
 may come down either heads or tails : these are the only 
 " chances." In the language of the theory of chance " the 
 chances are even " : that is to say, we know no reason why 
 either heads or tails should come up. 
 
 But when we say this we simply ascertain and avow our 
 ignorance : we have found and stated nothing whatever as 
 to what will happen. Nevertheless the mathematicians, 
 or many of them, cling to the notion that " in the long run " 
 of tossing, heads and tails will come out even. And in this 
 they are supported by what we may call the instinctive 
 expectation we should all have that when, say " heads " had 
 come ten times in a series of tosses, the next toss would give 
 tails. 
 
 At this point there tends to arise an amusing debate. 
 The late Mr. R. A. Proctor, the astronomer, once had such a 
 discussion in a newspaper with an anonymous antagonist
 
 io6 LETTERS ON REASONING Letter VII. 
 
 who called himself " An Inveterate Gambler." I have before 
 
 me some extracts which will show you how they argued. 
 
 Take, says Mr. Proctor, 
 
 "'An Inveterate Gambler's' idea that if in fifty tossings of a coin 
 there have been forty heads and but ten tails, the odds are four to one 
 that the next toss will be a tail. Of course this is wrong ; the chances 
 for head and tail are even for that as for every other toss." 
 
 He then goes on to say that 
 
 " the science of probabilities comes in and explains, what ought to be 
 obvious, that the next tossing is quite independent of all the past ones, 
 and that the betting should be even on ' head ' and ' tail.' But Buffon 
 and the other fellows who tried the experiment of tossing a coin many 
 thousand times proved this experimentally. For in all those multi- 
 tudinous trials it was found that there was not the slightest trace of a 
 tendency towards ' head ' after {sic) runs of ' tails,' or vice versa. There 
 were many sequences of even (sic) ten or twelve ' heads,' yet following 
 these ' head ' came as often as ' tail.' " 
 
 Here, on reflection, you will at once admit that though, if you 
 had not previously considered the subject, you would have 
 been inclined to say that after many " heads " there is /i^efy 
 to be " tail," Mr. Proctor is right in saying that " the next 
 tossing is quite independent of all the past ones." In other 
 words, being wholly unable to trace the minute mechanical 
 causation which determines the result, we do not know at 
 all how the toss is determined — we are as ignorant after forty 
 tosses as before. But Mr. Proctor in turn fell into a con- 
 fusion which I once found surprising, and which I should 
 now describe as the almost inevitable result of an argument 
 proceeding on the belief that there is a "science of proba- 
 bility " in such matters. 
 
 In the first place, it is a confusion to say that the alleged 
 " science of probability explains " that the fifty-first toss is 
 independent of the preceding fifty. It is our awakened 
 common-sense that explains that. Mr. Proctor's "science 
 of probability" turns out to be a process of arithmetical 
 calculation ; and there is no arithmetical calculation thus 
 far : we pass the judgment without resort to arithmetic.
 
 Letter VII. LETTERS ON REASONING 107 
 
 In the next place, Mr. Proctor falls into pure 
 absurdity when he says that in the experiments of Buffon 
 and the "other fellows" it was found that there was "not 
 the slightest trace of a tendency towards ' head ' after runs 
 of ' tails,' or vice versa." For that matter, a " run of tails " 
 is marked as such by heads before and after : after a run of 
 tails there must be a head : if there be no head we are not 
 " after " the run. Mr. Proctor meant that, as a matter of 
 fact, after you have had tails a number of times you may go 
 on getting tails. But when he added that "even" (why 
 "even"?) after sequences of ten or twelve heads, heads 
 actually did come as often as tails, he was in effect asserting 
 that in the given experiments runs of thirteen heads were 
 exactly as common as runs of twelve ; and if his argument 
 were coherent he was committed to arguing that runs of 
 fourteen were as common as runs of thirteen, and so on 
 indefinitely. Now, as we are dealing with definite numbers, 
 we know, on principles of mere arithmetic, that this cannot 
 have been so. It cannot even have been true that, up to 
 the figure thirteen, runs of each number occurred with 
 equal frequency. Mr. Proctor has succumbed, as I should 
 put it, to the normal fatality of the theory that there is a 
 " law of chance " : he has been inadvertently saying that 
 there is regularity where in the terms of the problem there 
 is none. 
 
 The nature of the fallacy becomes clearer as we follow up 
 the discussion. Mr. Proctor proceeded to explain that in 
 terms of the science of probabilities the chances as it were 
 approximate when we mount into very large numbers. The 
 "gambler," proceeding on his common-sense, avowed that 
 he could not for the life of him see why the mathematical 
 chances should come right in a million tosses any more than 
 in two ; whereupon Mr. Proctor answered : — 
 
 *' He is right enough if he imagines the mathematical cliances point 
 to absolute equality. In a million tossings, the event actually most 
 probable among many millions of millions of millions of possible events
 
 io8 LETTERS ON REASONING Letter VII. 
 
 (the actual number is two raised to the power one million, a number 
 containing 301,031 digits, which I would rather not calculate) is that 
 there should be 500,000 heads and 500,000 tails. Yet this chance, though 
 the largest, is largest among a number of chances which are exceedingly 
 minute. The odds are many thousands to one against absolute equality. 
 But they are also many thousands to one against the numbers of heads 
 or tails so disproportioned as 500,500 to 499,500." 
 
 In short, there are " many thousands to one against " 
 absolute equaUty ; but there are more chances in favour of 
 it. That is to say, there is no " science " whatever of the 
 problem in the ordinary sense of that term. The so-called 
 mathematical " science of probability " can predict nothing 
 in such matters as these ; it ends in measuring chance in 
 terms of chance : that is to say, it cannot measure at all. 
 And yet, when all is said, if we were in any way coerced or 
 persuaded to bet on the result of a toss after heads had 
 already occurred ten times running, we might " reasonably " 
 bet on tails, simply because we cannot suppose runs of 
 eleven to be as common as runs of ten unless we also 
 suppose runs of a thousand to be as common as runs of 
 three. And if we so betted, the next toss might nevertheless 
 yield heads ! Such is " chance " : who shall find its " law " ? 
 
 § 3- 
 In view of this standing dilemma, some have argued that 
 " probability " is to be conceived solely in terms of our 
 tendencies to expect certain results. But Dr. John Venn, 
 in his very able treatise. The Logic of Chance, argues 
 cogently^ that this view negates the possibility of a science 
 of real occurrences, and that unless we are to fall back on 
 mere mathematical calculations of abstractly possible rela- 
 tions of numbers — Permutations and Combinations — we 
 must have regard to actual experience, making that our 
 starting-point and point of return. You will agree with him, 
 I think, that mere calculation of the possible combinations 
 of a pack of cards counts for nothing as a guide to action 
 ^ Work cited, ch. iv. §§ 12, 13.
 
 Letter VII. LETTERS ON REASONING 109 
 
 unless it can be shown that actual experiment more or less 
 corroborates the estimate. If the " science " is to consist in 
 pure calculation, we must " go over to the mathematics, and 
 so lose all right of discussion about the things " : rejecting 
 that course, " we take part with the things, and so defy the 
 mathematics." 
 
 But what is the result of " taking part with the things " ? 
 \Vhat do we or can we know by experiment as to actual 
 averages of such phenomena of " pure chance " — that is, 
 untraceable causation — as the results of tossing coins ? On 
 this head Dr. Venn pronounces so oddly that I feel bound 
 to transcribe the page in full, lest by a paraphrase I should 
 be thought to misrepresent him : — 
 
 " The formula, then, not being demonstrable a priori (as might have 
 been concluded), can it be obtained by experience ? To a certain extent 
 it can ; the present experience of mankind in pence and dice seems to 
 show that the smaller successions of throws do really occur in about the 
 jjroportions assigned by the theory. But how nearly they do so no one 
 can say, for the amount of time and trouble to be expended before wc 
 could feel that we have verified the fact, even for small numbers, is very 
 great, whilst for large numbers it won Id be simply intolerable. The 
 experiment of throwing often enough to obtain ' heads ten times ' has 
 been actually performed by two or three persons, and the results are 
 given by Dc Morgan and Mr. Jevons. This, however, being only 
 sufficient on the average to give ' heads ten times ' a single chance, the 
 evidence is very slight ; it would take a considerable number of such 
 experiments to set the matter at rest. 
 
 "Any such rule, then, as that which we have just been discussing, which 
 professes to describe what will take place in a long succession of throws, 
 is only conclusively j^roved by experience within very narrow limits, 
 that is, for small repetitions of the same face ; within limits less narroio, 
 indeed, we feel assured that the rule cannot be flagrantly in error, 
 otherwise the variation would be almost sure to be detected. From this 
 we feel strongly inclined to infer that the same law will hold throughout. 
 In other words, wc are inclined lo extend the rule by Induction and 
 Analogy. Still there are so many instances in nature of proposed laws 
 which hold within narrow limits, but get egregiously astray when we 
 attempt to push them to great lengths, that we must give at best but a 
 qualified assent to the truth of the formula."^ 
 
 1 /(/. ch. iv. § 9, 2nd ed., p. 97.
 
 no LETTERS ON REASONING Letter VIL 
 
 Here, while coming to a nearly negative conclusion, Dr. 
 Venn lays down the contradictory views that (a) we cannot 
 know the average of chances in " large " numbers of instances, 
 because the labour of the experiment would be intolerable • 
 and (i>) that nevertheless for large numbers (" within limits 
 less narrow " — a very loose phrase — must mean this, or the 
 argument comes to nothing) a variation in actual practice 
 from the theoretical rule " would be almost sure to be 
 detected " ; and in various passages he seems to avow the 
 belief he has above disclaimed. This confusion would seem 
 to be a survival from an early habit of identifying the 
 numerical theory of chances with the actual course of things. 
 
 To that species of miseducation, too, I am inclined to 
 attribute in some degree the attitude of many educated men 
 who, insisting that past chances in no way control future — 
 that after tossing " heads " a hundred times we may just as 
 well have heads as tails — nevertheless insist that in the 
 mathematical " theory of probability " we possess a real 
 science or knowledge. In the terms of their own argument, 
 not only does the theory yield us none but unverifiable pro- 
 positions : its scientific truth consists in their unverifiability. 
 Professing to give a numerical theory of all chances, they 
 yet exclude all counted or known chances. It is difficult to 
 imagine a more nugatory doctrine ; and probably even 
 academic habit could not keep it in countenance if it were 
 not that useful calculations from actual experience — as in 
 life insurance — are made, as it were, by the same mathe- 
 matical machinery. 
 
 What we arrive at, after discriminating between the 
 different classes of events in regard to which we ccMiimonly 
 use the expression "chance," is the conclusion that some 
 kinds of events which singly are " uncertain " or unpredict- 
 able as to time and place, but are either certain to occur 
 once in each person's case (as death) or known to occur 
 frequently (as fires and accidents), do in actual fact occur 
 with such a degree of numerical uniformity in time that
 
 Letter VII. LETTERS ON REASONING iii 
 
 action can profitably be taken on the basis of such uni- 
 formity. Thus the deaths per thousand of the population 
 vary little from year to year, and still less from decade to 
 decade, though a good many of those deaths yearly occur 
 from accidents to strong and healthy people. So with fires : 
 actual experience tells that they occur oftener in certain 
 kinds of business than in others ; hence an adjustment of 
 rates of premium according to " average risk." So with life 
 insurance, and insurance against burglary or accidents. So 
 even with such matters as the number of letters posted 
 annually without addresses : in each order of cases, while it 
 is impossible to predict a given event, we may count on the 
 number of cases being nearly the same in a given period 
 and in similar conditions. Here there is really a " science 
 of probability," albeit one of a very simple order ; and I 
 note that an actuarial expert, after discussing the problem 
 all. round, concludes that "the business of insurance has 
 little or nothing to do with the mathematical calculation of 
 chances." ^ 
 
 §4. 
 These last classes of phenomena have set up fresh con- 
 fusions of thought in regard to "chance" and "causation." 
 Because, for instance, suicides occur in given populations in 
 nearly the same number every year, varying even from season 
 to season with a good deal of regularity, people are apt to 
 say, " This is 7iot chance." What seems to be in their minds 
 is the idea, "These things are caused" ; and by implication 
 they would seem to hold that " to happen by chance " is to 
 happen without being caused. Now it follows from our 
 foregoing reasonings that both of these notions are fallacies. 
 From the point of view of the general onlooker, accidents 
 and suicides and omissions to address letters are cases of 
 " chance," though all events are caused ; and in each case 
 of the sort under notice the chain of causation may on 
 
 1 On Probability and Chatice, and their Connection with the Business 
 of Insurance, by J. B. Sprague, M.A., LL.D., F.R.S.E., &c. (1892).
 
 112 LETTERS ON REASONING Letter VIL 
 
 inquiry be traceable at will. A suicide is the culmination of 
 a series of causes, all of which may be known to some 
 onlooker ; but to the statistician, and to those who do not 
 know the causes, it is a chance. To the word " chance " 
 we must attach a reasoned meaning : that is all. 
 
 Some writers, observing that for many people the word 
 suggests the happening of things causelessly, have con- 
 fusedly impeached the word instead of the fallacy connected 
 with it. Even so great and so acute a thinker as Hume, by 
 reason of a certain " royal carelessness " about his terms, 
 falls into several confusions on this subject. It may be 
 profitable to you to trace his steps. In his early Treatise Oj 
 Human Nature (1739) he first reasons that "chance is 
 merely the negation of a cause," but proceeds to note more 
 clearly that " 'tis commonly allowed by philosophers that 
 what the vulgar call chance is nothing but a secret and con- 
 cealed cause." ^ Yet in his essay Of the Rise atid the 
 Progress of the Arts and Sciences (1742) he writes that 
 "nothing requires greater nicety, in our inquiries concerning 
 human affairs, than to distinguish exactly what is owing to 
 chance and what proceeds from causes." What he ought to 
 have said here is : " to distinguish exactly between causal 
 and non-causal coincidences or sequences " ; that is, between 
 cases of causation and cases in which a given event or 
 series coincides with or follows upon another event or series, 
 but is produced by causes apart from that. Finally, in the 
 first of the two Inquiries (1748) into which he recast his 
 Treatise, he writes, first, that there is ^'' no such thing as 
 Chance in the world " ; ^ an extremely careless way of speak- 
 ing, which he corrects later by the observation that " Chance, 
 when strictly examined, is a mere negative word, and means 
 not any real power which has anywhere a being in Nature." '^ 
 
 ^ Book L, part iii., §§ 11, 12. 
 
 '^ /iiijiiiiy Cuuceniiiig the Human Understanding, sect. \i., Oj 
 Probability. 
 
 ^ Id., sect, viii., Of Liberty and Necessity, part i., last par.
 
 Letter VII. LETTERS ON REASONING 113 
 
 This brings us to the true statement of the case, which may 
 be worded thus: "'Chance' is an expression we use to 
 connote either our ignorance of the causation of a given 
 event or our behef that a given coincidence is not as such 
 causal — that is, has not been planned." 
 
 The remark that " there is no such thing as chance " is a 
 sample of verbal confusion which you may do well to examine, 
 because such muddles occur rather often in discussion on 
 difficult questions. As it stands, the phrase is strictly 
 meaningless : it is a mere counter-sense. " Such a thing as 
 chance " means that there is chance. Hume meant : " There " Cs\..<JvAr^ 
 is no such thing as some people understand by chance." i L^, 
 
 Even the phrase, " There is no such thing as a centaur," is ^***-^ 
 but a passable short cut to the proposition, "There is no '<^ c^"^"^' 
 such thing in Nature g5 men represent by the figure called . " ,'.. ^ 
 a centaur." But Hume's phrase is worse than that ; and it ' ' ^^ . 
 gives a precedent for the now current formula of so-called ' L{/^ ' 
 
 . ' .- ^ . ^r 5 La ^-^ 
 
 pseudo-reasoning without limit. ^^.^mJ 
 
 I need hardly say, after the foregoing argument, that ^--'^^'*^_ r 
 Hegel falls into a far more serious confusion than Hume's U/\^^ * 
 
 when he affirms that " the world is not abandoned to chance 
 and external contingent causes, but a Providence controls 
 it." ^ Hegel, with all his Pantheism, actually implied that 
 Chance is a force which somehow excludes causation ; for 
 he speaks later of " the empire of chance " and " alien 
 necessity and chance." ^ To such a phrase as " the empire 
 of chance " we can attach only one rational meaning : " the 
 region of untraced or untraceable causes." Hegel then 
 might conceivably mean that a number of the phenomena 
 of the universe belong to that region, but that human affairs 
 do not. This, however, would be both false in fact and 
 inconsistent with his general theorem, for there are intelligible 
 
 ^ Philosophy of History, Eng. tr. , p. 13. 
 2 Id., pp. 34, 35. 
 
 " Christian Science," that " there is no such thing as Death • i ^ [\ 
 
 — a sense-destroying locution which opens the way for *\ L^ C-*'^^
 
 114 LETTERS ON REASONING Letter VIL 
 
 and enigmatic aspects in both non-human and human evolu- 
 tion ; and Hegel is committed to affirming that all alike are 
 the expression of the immanent or directing Reason. 
 
 His expression then would seem to be a mere stroke of 
 declamation, a figure of rhetoric, a way of being verbally 
 impressive. And his disciples in this connection illustrate 
 afresh for us the difficulty of being sincere in the sense 
 of being philosophically consistent. " Not every trifling 
 occurrence," says one of them, Gans, in the preface to 
 Hegel's posthumous work, " not every phenomenon pertain- 
 ing rather to the sphere of individual life than to the course 
 of the World-Spirit, is to be ' construed,' as it is called, and 
 robbed of its life and substance by a withering formula." If 
 there be any meaning in words, a "trifling occurrence," to 
 begin with, would be raised to new dignity and significance 
 by being treated as part of the course of the ^Vorld-Spirit ; 
 and if the World-Spirit means anything it means the organic 
 totality of all human occurrences. If, on the other hand, 
 any phenomena can be said to lose their body and sub- 
 stance by being involved in a withering formula, it may fitly 
 be said that Hegel's formula of the World-Spirit does this for 
 the large masses of history, for the immeasurable processes 
 of human experience which he reduces at times to intellectual 
 abstractions. 
 
 Hegel might well complain of " the labour of the notion ": 
 he strained under it ; and some of his confident disciples, as 
 Gans, miscarry badly under the burden. And as I fear that, 
 whether in that way or in another, I am bringing home the 
 sense of the burden in question to you in this letter, I shall 
 here end it, leaving for another the task of grappling with 
 the most important problems in regard to which men use 
 arguments turning on "chance." That task will be the 
 lighter for the relative tediousness of the excursion we have 
 just finished.
 
 LETTER VIII. 
 
 The standing confusion set up by the random use of the 
 word " chance " is seen in full play in the common argu- 
 ment about the obviousness of design in Nature. It does 
 not seem likely that the average reasoner will have got past 
 the plane of that argument in your day, seeing that he has 
 contentedly energised there now for at least some centuries. 
 That the order of the universe cannot have been set up by 
 chance is still, at least, one of the standing formulas of 
 theism ; and it is ordinarily employed with the assumption 
 that it involves the conclusion : " therefore, the order of 
 Nature is planned by an intelligent infinite Person." Let us 
 trace the usual steps of the argument. 
 
 It is common to put first to the non-theist the parable of 
 the watch found by the traveller in a desert : the traveller, it 
 is urged, at once realises that he has found a product of 
 design, somehow brought there, though he sees no other 
 sign of human passage. In the same way, says the theist, 
 the universe itself, being a plexus of law and recognisable 
 order, cannot have come into such form " by chance " ; it 
 must have had a designing maker. 
 
 Observe that at the very outset the argument destroys 
 itself. The watch is recognised as a product of design /// 
 contrast with the desert : it cannot have been produced as 
 the scattered rocks of the desert have been. That is to say, 
 for the theist, the phenomena of the desert are matters of 
 chance. Yet the very purpose of the argument is to prove 
 that the whole cosmos, of which the desert is part, is a pro- 
 duct of design : that there is law and order throughout it. 
 From the play of intelligent volition in the watch we are 
 
 K
 
 ii6 LETTERS ON REASONING Letter VIIL 
 
 analogically to infer the play of intelligent volition in the 
 cosmos : yet the argument assumes for its first step that 
 parts of the cosmos at least show no play of intelligent 
 volition. The modern theist repudiates the ancient theor- 
 ist's notion of a " fortuitous concourse of atoms " ; yet he 
 assumes for his present argument that the desert is actually 
 a fortuitous concourse of atoms. 
 
 Now, taking " chance " in its rational meaning — which is, 
 as we have seen, " uncontrolled or uncontrollable causation 
 of events, inexplicable or unexpected coincidence " — the 
 ancient speculator was really talking reasonably enough. 
 After all the declamation on the subject, it remains a reason- 
 able though not exactly a useful thing to say that "the 
 order of Nature is a matter of Chance"; for that is only an 
 admission of the plain fact that we have no ultimate 
 knowledge or control of the causation of the order of 
 Nature. We can trace the details or sequences of causation 
 only to a certain extent : beyond that we can but predicate 
 causation as co-extensive with existence, admitting ignorance 
 of the further steps of the sequence. 
 
 For the rest, the common argument about the impossi- 
 bility of a bagful of letter-blocks being shaken out so as to 
 form a rational sentence is a peculiarly gross logical con- 
 fusion. It starts with products of express human design — 
 the letters. Now if we were actually to find a number of 
 sticky balls or lumps in a " state of nature," and shake them 
 violently in a box, so that they finally fell into an aggluti- 
 nated heap, we should in the ordinary meaning of the term 
 have before us a result of chance — that is, the result of a 
 process of unanalysed causation ; yet that result would be a 
 form of "order," and as such indistinguishable from much of 
 "virgin nature." It may be answered that the adhesiveness 
 of the lumps is itself a mark of design ; but I can carry 
 back the argument to a process of "chance" which should 
 result in the property of adhesiveness, and so on ad infinitum. 
 Thus the common argument, so complacently put as
 
 Letter VIII. LETTERS ON REASONING 117 
 
 conclusive, is so only for people who never think out what 
 they mean by " chance," who beg the question in the act of 
 framing their formula, and who yet virtually reason as if 
 chance meant "absence of causation." 
 
 That is the position fallen into by the late Professor 
 Momerie in his book On Belief in God, before referred to. 
 It is there affirmed ^ that the laws or processes of Nature are 
 as they are " either through purpose or by chance." Unless 
 " chance " here means " without causation " the proposition 
 is devoid of significance ; and in that sense no one ever 
 advanced it. Seeking to buttress this nugatory theorem, 
 the Professor writes that 
 
 " When only eleven planets were known, De Morgan showed that the 
 odds against their moving in one direction round the sun with a slight 
 inclination of the plane of their orbits — had chance determined the 
 movement — would have been 20,000,000,000 to i." 
 
 Here we have a dramatic illustration of the confusion 
 wrought in men's thinking by the machinery of the calculus 
 of probabilities. As we have already seen, De Morgan's 
 proposition amounts to nothing ; anything with any number 
 of mere so-called " chances " against it ca7i happen, in terms 
 of the theory ; and the phrase " had chance determined the 
 movement" merely stands for failure to attach to the 
 word " chance " any rational meaning. It is somewhat 
 shocking to realise that professional teachers of logic and 
 philosophy can publicly come to such a pass ; but we have 
 already seen, and shall see again, that professed logicians 
 can argue at times as ill as the man in the street. Reduced 
 to significant form, Dr. Momerie's proposition would run : — 
 " Either the process of Nature is planned by an intelligence 
 like ours " [he repeatedly insists on such a resemblance], " or 
 its total causation is unknown to us." And this, which he 
 does not in the least mean to say, is, as I have already 
 remarked, perfectly true. 
 
 Still, I do not advise you to go about saying " The order 
 
 1 Ch. iv. 
 
 K 2
 
 ii8 LETTERS ON REASONING Letter VIIL 
 
 of the universe is a result of the play of Chance," for the 
 average person would in all likelihood say yo?/ were asserting 
 that the order of the universe is uncaused. Besides, the 
 theist would seem to meet you by saying : " It is true we 
 cannot trace the manner of causation beyond a certain 
 point : but we remain certain that the causation is planned. 
 Just as we are sure the watch had a maker, and that it was 
 somehow brought to the desert, though we did not see it 
 made or brought ; just as we should know there was a maker 
 for a new machine such as we had never seen before, and 
 whose working we did not understand, so we know that 
 there must be a maker for such a stupendous machine as the 
 universe." 
 
 Here, to begin with, the theist is unconsciously restating 
 the contradiction above alluded to. \Ve know the machine 
 as distinct from the non-machine ; we know the footprint on 
 the sand as distinct from the non-human markings on the 
 sand : we know the house or hut or tent as contrasted with 
 the cave or the plain or the tree : in other words, the order 
 of Nature is ostensibly undesigned, design being admittedly 
 the mark of the human procedure in contrast with the non- 
 human. But the theist will never submit to the checkmate 
 of the strict process of his own argument : he falls back on 
 the conception that from human design we infer in Nature a 
 design of a different and a higher order. He has committed, 
 and will persist in committing, the fallacy of transforming his 
 terms. 
 
 Let us then see if we can enlighten him on a higher 
 plane of argument. Or, say, let me assume that you are 
 spontaneously impressed by the problem somewhat as he is 
 — that, in the common phrase, you " cannot suppose all 
 this frame of things to be without a mind." This is simply 
 one of the most general forms of what Ruskin called 
 "the pathetic [he had better have said the sympathetic] 
 fallacy " — the instinctive tendency to impute to the aspects 
 of Nature our own emotions - and it is the very generality
 
 r : 
 
 Letter VIII. LETTERS ON REASONING 119 
 
 of the process in this case that makes it so hard to detect 
 as such. A very little thought raises us above the notion 
 that rain tells of sadness in Nature, that thunder and 
 lightning tell of wrath, that sunshine stands for good cheer, 
 and so forth. Once those significances seemed quite ob- 
 vious to human beings : to-day all instructed men put them 
 aside as primitive fancies. But many instructed men still 
 hold by the ^ohil sympathetic fallacy — the ascription to 
 Nature of total purpose, or, in other words, the inference 
 of a quasi-human "design " for the whole immeasurable and 
 eternal process. 
 
 Now, the philosophic rebuttal of all such inferences can 
 be put in a proposition belonging to a series which are of 
 the essence of logic, but which nevertheless much logical 
 practice seems to leave many men incapable of realising. ; 
 That proposition is that f/icre can be no rational ascription of 
 single mode to the totality of things. .\11 propositions of ' 
 mode, all assertions of " manner of happening," in order to 
 be intelligible, must be in terms of contrasted modes. This 
 proposition, again, is of the essence of psychology : all 
 psychology proceeds upon it. Try to frame a proposition 
 of mode, of manner of happening — in other words, of 
 " phenomena " — without assuming contrast with some other 
 mode or phenomenon, and you will see that it is impossible. 
 A term, to be significant, must mark off something else, 
 some other term : a process of happening, to be realised in 
 thought, must be distinguishable from another process 
 of happening : a term of mode, to signify a cogitable process 
 of happening, must imply other processes. Professor 
 Jevons, impressed by this general truth, went so far as to 
 say that even the term " thing " implies the correlative 
 " that which is not a thing " — a proposition which I take to 
 be a philosophic error. The true doctrine, to my thinking, 
 is that " thing " either stands for " everything " or is a non- 
 significant term, like " existence " until it is qualified or 
 quantified. Each of those terms is applicable to every
 
 I20 LETTERS ON REASONING Letter VIIL 
 
 aspect of infinitude. Unfixed, it is like the unapplied 
 vocable " the." 
 
 Now, the one proposition of quality, quantity, or quiddity, 
 that we can rationally make concerning the cogitable 
 universe as a whole is simply — that it is infinite existence ; 
 and that proposition is intelligible in only one way, as the 
 negation of finity, which is the mark of all propositions of 
 mode. To attempt to give to the bare proposition of infinity 
 • — a proposition to which we are shut up by the very nature 
 of thought — any characteristic of mode, is to commit logical 
 suicide, to produce a meaningless proposition. You will 
 see this readily enough if you try to put any proposition of 
 infinite mode in terms of sense perception : you will never 
 consent to describe the infinite as blue, loud, thin, soft, hot, 
 or sweet. But the reasoning which excludes such pro- 
 positions leads logically — that is, by the law of consistency 
 — to an exclusion of such propositions as " the universe is 
 purposive," " the infinite is good." A good infinite, in 
 terms of our psychological " first principle," is thinkable 
 only as against a bad infinite : a designed universe is think- 
 able only as beside an undesigned one. The abstraction of 
 infinity is intelligible as the negation of every aspect of 
 finity : further than that it is not thinkable at all. Just as 
 it negates limit of extension, it negates mode. And the 
 «A inveterate habit of talking of a " loving " and " wise " Infinite 
 
 is simply a persistent refusal to think and speak rationally 
 on such matters. 
 
 By the same reasoning, all propositions of " mode of 
 happening " become meaningless when applied to the 
 infinite. The current " idealist " philosophy, for instance, 
 tells us that all things are " maintained solely by thought." 
 Now, no matter what perplexities be involved for us in the 
 effort to analyse our ideas of reality, nothing can alter the 
 fact that that is a meaningless proposition, inasmuch as it 
 affirms one mode of happening for the infinity of things. In 
 order to condenm it as verbiage, we do not need to con- 
 
 v 
 
 r ^ 
 
 %J
 
 Letter VIII. LETTERS ON REASONING 121 
 
 front it with all the other propositions with which, if 
 accepted, it would have to co-exist in our minds. We say, 
 in terms of our first and last principles, that no mode of 
 universal existence, or universal happening, can be thought. 
 You can think of given phenomena as " maintained solely 
 by thought," when you think of other phenomena as not so 
 maintained. Not otherwise can the proposition have any 
 significance. To say that " all things, all phenomena, are 
 maintained solely by thought," is somewhat like saying " all 
 things are soft." 
 
 How can it be, perhaps you ask, that such propositions 
 are maintained by instructed men if they be really non- 
 significant ? Can I be right, you may ask, in thus quashing 
 as meaningless the formulas not only of theologians but 
 of some philosophers, not committed to theological pre- 
 suppositions ? Press the question on yourselves, I beg of 
 you. If you can by consistent reasoning refute my argu- 
 ments, do so. I am doing my best for you : if you can 
 better my best, it will be to you a measurable gain. 
 
 Meantime, here is my answer. Such self-contradictory 
 or meaningless propositions as I have instanced seem to 
 me to have actually taken rise in the effort to sustain 
 theological presuppositions. The final verbalism, the 
 ultimate meaninglessness, is reached after a number of 
 steps, each one of them plausible, because it deals with a 
 recognised difficulty, but each one none the less fallacious. 
 By a series of minor fallacies or verbalisms, men reach to a 
 sonorous, a clanging verbalism. And it is very hard indeed 
 to discuss metaphysics without verbalism. I am sometimes 
 disposed to think that every metaphysical treatise I know 
 proceeds upon a careless use of words in nearly every page. 
 But it would be vain to attempt to justify such a judgment 
 in these Letters ; and it is equally out of the question for 
 me to undertake to present you with my own metaphysic, 
 my own philosophy, under this form. The production of 
 that is a task for long leisure; and I know not whether
 
 122 LETTERS ON REASONING Letter VIII. 
 
 such leisure will ever be mine. Here it must suffice to 
 work out this one issue of the vogue of meaningless pro- 
 positions about the Infinite. 
 
 The formula, then, that all existence is maintained solely 
 by thought seems to follow upon the formula that " things 
 exist only in our consciousness." Men debate as to whether 
 things exist in our consciousness or outside of owx conscious- 
 ness ; being moved to this debate, historically, by the 
 exigencies of the belief in a controlling Spirit or creative 
 God. Denials of the reality of certain theological concep- 
 tions seem to have led to theological denials of the reality 
 of phenomena accepted without question by the anti- 
 theologians ; and this verbalist formula, " things exist only 
 in our consciousness," is one of the results of that line of 
 movement. 
 
 Now, the question whether things exist in or out of our 
 consciousness is, in terms of the foregoing argument, 
 absolutely meaningless if taken as a demand for an account 
 of the mode of happening of all existence. It is significant 
 solely as a question oi how ive shall define '■'' consciousness.^'' 
 All are agreed that things exist, and that we are conscious of 
 things ; and it really does not matter a straw whether we 
 say that all things exist in our consciousness or outside of it. 
 If we affirm " inside," we imply that there is an outside — the 
 very thing denied. The debate is one that ought never to 
 have taken place ; and it would not have taken place if either 
 side had clearly realised that the question is an attempt to 
 predicate mode of existence of the infinity of things, and 
 that no such predication can be construed in thought. Con- 
 trasted modes of occurrence or existence of phenomena are 
 predicable : absolute mode for infinity is a contradiction 
 in terms. 
 
 But by ignoring or overlooking this fundamental law of 
 rational statement, and by attaching an illusory significance 
 to the formula " things exist only in our consciousness," men 
 reached the further non-significant formula "all things are
 
 Letter VIII. LETTERS ON REASONING 123 
 
 maintained by thought." From that the next step, equally 
 illicit and inconsistent, is to Totality of Thought maintaining 
 an infinite universe = God maintaining all things. Every one 
 of those propositions is a nullity, a verbal ineptitude, the 
 expression of incapacity for coherent thought on the part of 
 men claiming specially to exhibit such capacity. They 
 are simply saying things that cannot be thought. The 
 entire logical residuum of the argument is the implication 
 that God is a name for Everything. 
 
 This is how the case works itself out in that unhappy 
 book of Professor Momerie's, On Belief in God. After 
 advancing a series of admittedly indecisive arguments for his 
 thesis, including that about "purpose" and "chance," 
 before mentioned, the Professor proceeds to complete his 
 unconnected series of contentions by one irrelevant to all of 
 them — the theorem now under notice. After setting out 
 with a plea from the alleged universal instinct or surmise of 
 men, he affects to prove his case by a formula which, on his 
 own confession, runs counter to universal instinct — thus 
 stultifying himself for the second time. The universal 
 theistic instinct, he had urged, " is not very likely to turn out 
 altogether delusive " : a sufficiently futile form of words. 
 Now he asserts that " the popular notion of course is, that 
 matter exists apart from consciousness," adding that reflec- 
 tion will show this to be an error. All the while he is merely 
 asserting the truism that we cannot think of matter without 
 matter's being thought of by us. There is " no such 
 thing," he pleads, in the old way, "as an untasted taste." 
 That is to say, there is no such thing as no thing. Next we 
 have a formula a little less obviously meaningless, our old 
 acquaintance : — " Things exist only in consciousness." Of 
 this proposition the only significant construction is : " We 
 are conscious of things only when we are conscious of them." 
 Any further implication, such as the Professor is leading his 
 inexpert readers to set up, is a lawless absurdity. Ringing 
 the changes on non-significance, he further affirms that
 
 124 LETTERS ON REASONING Letter VIIL 
 
 _ - " Things as we know them are the work of thought," which 
 y^°*^" means simply that tjiings are thought by us as we think 
 Vr-^\ them; and again that "There would be/fr us no material 
 ") world at all except for the activity of our own minds." He 
 
 '^ ' might have added that if there were no material world there 
 
 would be no " us." He has in effect merely asserted that if 
 we did not perceive a material world we should not perceive 
 it. And it is from this string of absolutely empty pronounce- 
 ments that he proceeds to the proposition : — " Another 
 Understanding — infinite, though at the same time in essence 
 identical with our own — is the source of that orderly system 
 ^. of relations which we know." He might as well claim to 
 W / have proved that we can digest only in an infinite digestion, 
 
 ■y\ur^ or have toothache only in a toothache of the Infinite. The 
 1 , '^'^ entire argument is a mere swoon of the reasoning faculty, a 
 b, ^ \}( failure to follow the thread of thought in a small artificial 
 /^ , maze of phrases. But that sort of thing passes for philosophy 
 * ^ -p with many men, and many professors. 
 
 < - «• fT So strong, indeed, is the spell of a philosophic fashion that 
 
 ^ jj#^we find it yielded to by so independently trained a thinker 
 
 , as John Mill. In his Examination of Sir William 
 
 Hamilton's Philosophy, he sets out on his first mquiry 
 
 
 J{\, thus :- 
 
 
 ur^ 11 "All language recognises a distinction between myself — the Ego — 
 
 and a world, either material or spiritual, or both, external to me, but 
 of which I can, in some mode and measure, take cognizance. The 
 . most fundamental (juestions in philosophy are those which seek to 
 
 ■'j n determine what we are able to know of these external objects, and by 
 j/^ e what evidence we know it." ^ 
 
 [j-'\ ^'^ "''^' t^hc proposition in the second sentence seems a 
 
 mistake : it describes the problems of science rather than 
 the problems of philosophy ; but we may let that pass. 
 What we need to note is that Mill here uses the ordinary or 
 popular language about the " external," and this properly 
 enough, inasmuch as it distinguishes between vitally 
 
 ' Fifth cd., p. 6.
 
 Letter VIII. LETTERS ON REASONING 125 
 
 different orders of experience, as dreams and perceptions. 
 Most of the so-called " idealists " would at once object to 
 Mill, as some one has done in a pencilled note on my 
 copy of his book, that " To [say we] perceive something out- 
 side ourselves is a contradiction : we perceive only within 
 us : that which we perceive is only a modification of our- 
 selves, therefore within us." This kind of formula has 
 passed current as philosophy for generations ; and yet it is 
 the merest countersense. He who affirms that we do not 
 perceive anything outside ourselves affirms that there is an 
 outside, else his proposition is meaningless. Then he 
 alleges that there is an outside but that we do not perceive 
 it. We now come to a mere dispute as to what is to be 
 meant by the term "perceive." The "idealists' " argument 
 implies an absolute confidence as to the presence of the 
 "outside." For that confidence we want a forcible term. 
 Has he any more forcible than " perceive " ? If so, let him 
 put it ; but let him not argue as if the dispute were on a 
 question of fact or existence when it is merely over a 
 question of definition or terminology. 
 
 Mill, after the pronouncement above quoted, might be 
 supposed to argue in some such fashion as I have just done. 
 But, proceeding to combat the inconsistent reasoning of 
 Sir William Hamilton, Mill goes on to gainsay himself 
 Hamilton said that the " primary " qualities of matter [as 
 extension, hardness, etc.,] "are apprehended as they are in 
 bodies ; the secondary [as, colour] as they are in us ; the 
 secundo-primary [i.e. mechanical as distinct from geometrical 
 properties of bodies] as they are in bodies and as they are 
 in us." ^ This I hold, with Mill, to be a tissue of intellectual 
 confusion : no such discrimination of our sense-relations to 
 matter will stand logical analysis. But Mill in the course of 
 his discussion, noting the criticisms of other writers on his, 
 says this : — ^ 
 
 > Dissertation!: and Discussions, pp. S57-8. 
 - Ed. cited, p. 41.
 
 126 LETTERS ON REASONING Letter VIII. 
 
 "Mr. Eraser, like myself, believes the Primary Qualities to have no 
 more existence out of our own or other minds than the Secondary 
 Qualities have, or than our pains and pleasures have." 
 
 Here the term " external," as previously used, is com- 
 pletely negated ; and the new proposition is simply not what 
 Mill previously and normally believed and said. He has 
 been led into fantasy by the customary entanglement of 
 terms. The right answer to Hamilton has been missed. 
 The so-called primary and secondary and secundo-primary 
 qualities of matter are alike modes of our total cognition ; 
 and to say that they "have no more existence out 
 of our own or other minds than our pains and pleasures 
 have," I repeat, is to proceed from non-significance to 
 non-sense. The one psychological fact involved is the 
 humble and really unnecessary truism that we cannot think 
 of matter without thinking of it. Matter thought of is 
 matter perceived, remembered, or imagined. Those who say 
 " Matter cannot exist unthought-of or unperceived " are 
 merely putting a vicious or unmeaning form on the truism 
 that when we go through the form, say, of imagining a world 
 without people, we are in imagination importing ourselves 
 into that world, and picturing it as seen by eyes. That is 
 simply one of the enveloping conditions of our nature, 
 one of the constituent facts of our existence. We cannot 
 think of stone save as hard, or of fresh grass save as green : 
 in other words, we think the greenness and the hardness. 
 Precisely so ; but the inconsequent corollary, the mere 
 verbalism that we maintain the thing as we think it, that it 
 ''cannot exist save as perceived," says nothing that is think- 
 able. It is one of those solemn sophisms inventible only by 
 men buttressing false creeds by fallacious systems. This 
 sort of philosophy is of the order of the Nicene Creed : it 
 inherits from the doctrine of the Trinity ; from the Siim/na 
 of Thomas Aquinas and the dialectic of the school-men ; 
 from the ages of pseudo-science which explained Nature in 
 the light of Semitic mythology. But the trick of it, as we
 
 Letter VIII. LETTERS ON REASONING 127 
 
 see, has become second nature to many men who have 
 transcended those stages of the evolution of theological 
 error. They remain under the spell because they are not 
 good enough reasoners ; because they lose their way in an 
 argument ; because words are too slippery for them ; or 
 because they have been so early reared on sophistries that 
 they have never come to the right and full use of their powers. 
 This we are driven to say even of Mill, and in respect of 
 yet another error. Oppugning Hamilton's philosophy, which 
 is often acute and penetrating, but chronically subverted by 
 theological presuppositions, he argues on the one hand that 
 "The Infinite," and "The Absolute," as formulated by 
 Hamilton, are " chimerical abstractions," standing for nothing 
 thinkable ; and on the other hand that Hamilton's denial of 
 the thinkableness of infinite "goodness" and so forth is a 
 fallacy. Now, the first criticism is implicit in Hamilton's 
 own statement, for he avows that his Infinite and Absolute 
 are in every way inconceivable, though he theologically 
 claims that they are to be worshipped under "authority." 
 With such teaching we need not now occupy ourselves. But 
 the thesis that the phrase " infinite goodness " is a placing 
 of infinity in a finite category is perfectly valid so far as 
 it goes, and Mill's denial is a sad lapse into fallacy. He 
 goes about to confute Hamilton,^ first, by the argument that 
 general predicates, such as "good," are all " at least poten- 
 tially infinite " ; and he actually proceeds to prove this by 
 saying that " good " is a term for all who ever did or will or 
 conceivably can be good, and " this is not a limited number." 
 It is difficult to realise that such an irrelevance can come 
 from a thinker of ability. The question of the number of 
 persons who can be termed good, you will at once see, is 
 wholly beside the case. We might as well say the same 
 thing of the potential applicability of the terms " short," 
 " bilious," " red-haired," " dull." That would not affect the 
 fact that these terms are marks of limitation, which is the 
 ' Examination, ed. cited, p. 104.
 
 128 LETTERS ON REASONING Letter VIIL 
 
 conception in hand. Even on his own irrelevant line of 
 thought, he is wrong in saying that the number of applica- 
 tions of the term "good" is "potentially infinite." Here 
 he really does commit the error sometimes charged on him, 
 of confusing the indefinite with the infinite. 
 
 He then goes on to admit that the " comprehension " 
 (= incidence) of such a term as "good " is limited. This 
 is the real difficulty, and he proceeds to meet it by 
 arguing that on Hamilton's view 
 
 " Infinite goodness cannot be thought as goodness, because that 
 would be to think it finite. Surely there must be some great confusion 
 of ideas in the premises, when this comes out as the conclusion." 
 
 All that is conveyed here is a resolute dislike of the chal- 
 lenged conclusion, and I think you will be unable to find 
 anything else in Mill's argument. He takes advantage of 
 Hamilton's error in putting the form " Infinite Space " 
 (= extension) on all fours with "Infinite Goodness"; but 
 beyond showing that we cannot help concluding in the 
 infinity of extension, he never makes out any case on the 
 true issue. So far as his reasoning goes, it is as valid for 
 saying " infinite shortness " or " infinite redness " as " infinite 
 goodness." Once more, " good," like " blonde," is a term 
 of mode intelligised as in contrast with "not-good"; and 
 "infinitely good " is either a rhetorical figure or an abuse of 
 language, like " infinitely sweet " or " infinitely round." 
 
 Beyond this point Mill's argument becomes an attack on 
 Dean Mansel for saying, in effect, " God is infinitely good, 
 but not in our sense of the term goodness." Mill warmly 
 insists that this is a mere vitiation of language ; and here, it 
 seems to me, he is quite right. We might as well suggest 
 a non-human meaning for " heat " or " blue " or " sixpence " 
 as for "goodness." The fitting summary is that the attempt 
 to force theism on the reason must always end in a de- 
 moralising departure from veracity of utterance. But Mill, 
 having committed himself to a vindication of the formula 
 "infinite goodness," is left pushing a fallacy against a
 
 Letter VIII. LETTERS ON REASONING 129 
 
 sophism — a natural result of his setting out with the osten- 
 sible assumption that "God" is a clear and valid term, 
 standing for a clear conception. In a note to the second 
 or a later edition of his work,^ he expressly says that " Con- 
 formity to the standard of right has a positive limit, which 
 can only be reached, not surpassed." That is a cancel- 
 ment of his reasoning about infinite goodness. 
 
 It is on this miscarriage of Mill's that Professor Momerie, 
 in a note to one of the later editions of his book on theism,^ 
 bases his assertion that Mill " exploded" the idea that there 
 is a contradiction in the expression " Infinite Personality." 
 As you see, there has been no such confutation. Our 
 warranted conclusion is that Professor Momerie, who uses 
 the semi-theist Mill to discredit his fellow-theists Hamilton 
 and Mansel, was simply determined to go on speaking of 
 an Infinite Person as he had been taught to do. " We do 
 not degrade the Almighty," he insists, " by saying that He 
 thinks and knows and wills. If the power behind Nature 
 were destitute of these faculties, it would be infinitely 
 inferior to the poorer type of man." Here we have, in effect, 
 the pretence that those who talk (loosely, to my mind) of a 
 " power behind Nature," think of it by a mere process of 
 eliminating human characteristics and leaving in their con- 
 cept those of inorganic Nature. Nobody, of course, ever did 
 anything of the kind, and men at the position of Spencer 
 (who taught that his Unknowable must transcend mere Mind) 
 might tell Dr. Momerie that he in turn degraded his deity 
 below birds and fishes, in that he did not conceive of 
 " Him " as able to fly, or to live in water. 
 
 You will perhaps now be disposed to agree with me that it 
 is not possible to discourse profitably of "The Infinite" and 
 "The Absolute" except by way of showing the said im- 
 possibility. In so far as Professor Bradley, in his very 
 able book on Appearance and Reality^ aims at something 
 more, he does but illustrate the negative, for his Absolute is 
 
 ^ Fifth ed., p. 49. '-^ Rep 1904, p. S4.
 
 I30 LETTERS ON REASONING Letter VIIL 
 
 just the old " All." Mansel saw that " the Absolute and 
 Infinite can be nothing less than the sum of all reality,"^ 
 and Mr. Bradley shows that Reality includes and is all 
 Appearance. To the same conclusion, unwittingly, comes 
 Professor Momerie when he says that " every existence, 
 divine as well as human, is necessarily conditioned by, or 
 related to, every other." What now becomes of Infinite 
 Personality ? We are left facing a collective Everything, of 
 which we cognise a fragment in extension, and a surface in 
 analytic intension. 
 
 There is one last device of fantasy of which you will have 
 to take note — that of shifting, as it were, on to the other foot 
 and affirming that Self is the "sole reality." That is the 
 theme of various German philosophies which I shall not here 
 attempt to characterise. It will suffice to show you how the 
 thesis emerges in that of Fichte, as handled by his devoted 
 interpreter Dr. William Smith of Edinburgh — a most worthy 
 gentleman, whose benignant face I well remember. There 
 can have been few kindlier men ; and I know few more lawless 
 processes of reasoning than that in which he summarises the 
 Fichtean position in his Memoir of FicJite. In Fichte's 
 Wissenschaftskhre, he affirms, 
 
 "for the first time, philosophy becomes, not a theory of knowledge, 
 but knowledge itself ; for in it the apparent division of the subject 
 thinking from the object thought of is abolished, by penetrating to the 
 primitive unity out of which this opposition arises." - 
 
 On the next page he remarks that 
 
 '■'■ \\\it fundamental character of finite being is thus the opposition of 
 itsf if (///^in) and of something opposed to itself" ; 
 
 but that does not trouble him. Immediately he goes on to 
 assert: — 
 
 "The Ego affirms the Non-Ego, and is affirmed in it; the two 
 conceptions are indissoluble, nay^ they are but otte conception modified 
 by different attitudes of the mind. But as these attitudes are in every 
 
 ^ Limits of Religious Thought, p. 30. 
 - Memoir of Fichte, 2nd ed., p. 75.
 
 Letter VIII. LETTERS ON REASONING 131 
 
 case voluntarily assumed by the Ego, it is itself the only real existence, 
 and the Non-Ego, as well as the varied aspects attributed to it, are but 
 different forms of the activity of the Ego." ^ 
 
 Then, after the statement that the Ego's " effort after self- 
 development is the root of our practical existence" — the 
 effort being thus prior to the personality — we have this : — 
 
 "This effort is limited by the Non-Ego — the creation of the Ego 
 \X.%f\i for the purposes of its oum 7noyal life. Hence the practical Ego 
 nnist regard itself as acted upon by influences from without, as restrained 
 by something other than itself — in one word, as finite. But this limita- 
 tion, or in other words the Non-Ego, is a 7nere creation of the Ego, 
 without true life or existence itself, and only assumed as a field for the 
 self-development of the Ego." 
 
 And so on, and so on, in an unspeakable beatific see-saw 
 of unruffled self-contradiction. You see how the process is 
 kept up : the good man has settled his formula in advance, 
 and the whole machinery of thought and speech must be 
 violated to yield it as an ostensible logical result. Technic- 
 ally, it all turns on the affirmation in the word " voluntarily," 
 which is explicitly false. This house of cards goes down at 
 a touch from any side : like all idealisms commonly so-called 
 it can give no philosophical solution whatever of the mental 
 phenomena of insanity. If the madman's thoughts are 
 /V/voluntary, we have the result that a million million men, 
 without compact, coincide voluntarily in their concepts of an 
 " external " order, while the involuntarily thinking minority 
 are all at sixes and sevens. And the theorem leaves us to 
 the impressive conclusion that as the Ego is simple per- 
 ceptionless and sensationless zero when we mentally abstract 
 from it all " influences from without," and as the Non-Ego is 
 a "mere" creation of that zero, without "true" existence, 
 there is no true existence. 
 
 As we have seen, the logical answer to all such pseudo- 
 philosophy is that a reduction of all existence to one mode 
 is the mere negation of thought. You will perhaps not feel 
 
 ^ The italics are mine. 
 
 L
 
 132 LETTERS ON REASONING Letter VIII. 
 
 that you need that canon to justify you in putting aside 
 the Fichtean formula. Our ablest living metaphysician, Mr. 
 Bradley, is fain to reject it, with curious hesitations, on the 
 score that " to make the self absolute is, so far as I see, to 
 postulate in the teeth of facts, facts which go to show that 
 the selPs character is gone when it ceases to be relative." ^ 
 That seems ground sufficient ; but perhaps it would have 
 been needless to set it forth in an Appendix if the true canon 
 of relativity had been implicit throughout in Mr. Bradley's 
 system. 2 
 
 Let us now retrace our steps to the problem of Chance, 
 in the light of the principles we have just been considering. 
 We saw at the outset that " design " was realised in thought 
 by the theist only as contrasted with non-design : we now 
 see that this is a law of thought, and that the theist's 
 attribution of design (a term of limitation) to the infinite is 
 a mere vitiation of words, like " soft infinite " and " good 
 infinite." At every point of the so-called design argument 
 the same fatality appears. When we are told to think of the 
 marks of design in the eye, we are implicitly invited to 
 recognise absence of design in the storm, the desert, the 
 cloud. To say " everything is designed " is to distinguish 
 nothing : if every phenomenon is designed, what marks the 
 watch from the wilderness ? Theological people speak of 
 " providential " occurrences, forgetting that on their professed 
 principles all occurrences are alike " providential." They 
 are in the grip of a logical dilemma. To give the word 
 significance they must make it contradict their whole 
 theology. Where Hegel stumbled, how shall they stand ? 
 
 In other words — in terms of one of our previous arguments 
 and definitions — they are either fixedly confused or they are 
 "insincere," in our special sense of the term, in that they 
 hold by inconsistent propositions from logically irreconcilable 
 
 ^ Appearatice and Reality, 3rd ed., App., p. 559. 
 * As to this compare Mr. B. Bosanquet's Knowledge and Reality, 
 1885, p. 332.
 
 Letter VIII. LETTERS ON REASONING 133 
 
 motives. And we are entitled to say of many of them that 
 their aberration is one of desire, of self-assertion. They will 
 not be content to admit their nescience, their inability to 
 frame rational propositions about the infinite. The theo- 
 logian will not be content to say, " It chanced that I missed 
 taking a train which was destroyed " : he must needs say, 
 " God mercifully provided that I should miss that train " — 
 after he has told us that all of God's provisions are merciful, 
 and has thus committed himself to the implication that God 
 mercifully killed the other people. Theology, in sad truth, 
 subsists on men's primary egoism and on their faculty for 
 false reasoning. 
 
 The one element of truth behind all these false pretences 
 of knowledge is simply the truism of the universality of 
 causation. And the false pretences of knowledge do but 
 obscure that truth. Theology arose in virtue of the non- 
 recognition of universal causation, men going about to say 
 " God did this" because they did not conceive of him as 
 doing that. 'I'hus their very conception of cosmic causation 
 was a delusion when they first attained to it ; and when some 
 denied their patently false assertions of supernatural causa- 
 tion they retorted by charging their gainsayers with denying 
 all cosmic causation. It may be that the charge was at 
 times true ; but it is certain that every effort towards a true 
 conception of causation in terms of universal law or of simple 
 statement of sequence was met by verbiage about things not 
 happening " by chance." Thus the theologian in one 
 breath affirmed and denied universal causation; for " chance " 
 really meant in his vocabulary non-causation, and he had to 
 affirm non-causation as the condition of some events in order 
 to insist on causation in others. This we have seen virtually 
 done by Hegel and his school. 
 
 Now that we realise the universality of causation, we are 
 under no solicitude about affirming the law concerning any 
 phenomenon whatever : our sole concern is to find what the 
 causation in a given case is. And it is part of common 
 
 L 2
 
 134 LETTERS ON REASONING Letter VIIL 
 
 honesty to admit that all causation is on one level ; that it is 
 all of a piece ; and that as all events and coincidences in 
 non-human Nature which we cannot control or anticipate 
 are for us as much matters of " chance " as such events and 
 coincidences in human affairs, " design " and " volition " are 
 on this view again out of the question. Let me explain. 
 JFt' design and will certain things : that is to say, our 
 volitions play a part in the total field of causation. But 
 many things happen to us independently of our volition ; and 
 equally, where we count on and exploit the sequences of 
 Nature, there occur unexpected variations, as in seasons and 
 temperatures, in droughts, floods, earthquakes, blights. If 
 we are to speak of the universe as designed, these are all 
 designed ; but the theist himself does not so regard them in 
 daily life. For the reasoner, who has realised that proposi- 
 tions of mode concerning the infinite totality of things are 
 meaningless, and that design is thinkable only as beside non- 
 design, the assertion is a mere counter-sense. Design is but 
 a mode in the infinite series — a mode appertaining to mind. 
 The infinite transcends design as it transcends mind. 
 
 It may occur to you that in assuming the universality of 
 causation I appear to be sinning against the canon I affirm. 
 But it is only a seeming inconsistency; for causation cannof 
 be conceived as contrasted with non-causation : we simply 
 cannot rationally think that any event is not caused. We 
 can but think of it as not caused or controlled by us. That 
 is to say, causation is not finally a modal concept: it is part 
 of the reasoned concept of sheer existence; whereas design 
 is a modal concept, because we can and do conceive of 
 events as undesigned, and the very asscrtor of universal 
 design habitually does so, and implicitly confesses it. 
 
 And now we come to the twofold crowning anomaly of 
 men's debates on the order of Nature. The very theologian 
 who insists that the whole universe is designed, openly turns 
 round at a certain point and insists that it absolutely lies 
 with us to determine our acts; that our volition is "free";
 
 Letter VIII. LETTERS ON REASONING 135 
 
 and that we are therefore responsible to the universal 
 designer for what we do. In the terms of the proposition, 
 " he " designed the universe of things and events, which 
 includes us and our deeds ; yet after all " we " are the 
 masters of our volition, in the sense that we do wrong when 
 we could have done right, and vice versa. Thus does incon- 
 sistency go from bad to worse. By a quite intelligible 
 sequence, the proposition of " Infinite Design," which is 
 only formally rational, is followed up by a proposition which 
 formally annihilates it : " We are uncontrolled in the exercise 
 of our will," that is, " our acts are not part of Infinite 
 Design." I shall deal with that problem in another letter. 
 Meantime, take note of the intellectual insincerity of the 
 school which affirms this proposition after that other — the 
 non-determinateness of men's wills in the same breath with 
 the fore-ordination of all things. 
 
 And that contradiction is paired with yet another. The 
 same assertor of the designedness of the entire order of 
 Nature practises prayer to the designer, and afiirms its 
 reasonableness. It is important to keep those facts present 
 to your mind when you are discussing the vital problems of 
 belief about the totality of things. You will read what are 
 ostensibly most earnest asseverations, confident arguments, 
 calm claims to have set forth irrefutable doctrines ; and you 
 may find it hard to withstand such a show of sincerity. But 
 it is all the while the literal fact that some of those ostensibly 
 convinced reasoners are affirming contradictions as gross, 
 shibboleths as meaningless, as words can be made to frame. 
 And while I insist that such persistent inconsistency con- 
 notes insincerity, I must proceed to explain the persistence, 
 the insincerity, in terms not merely of egoism, of partisan- 
 ship for one's formed or inherited opinions and dogmas, but 
 of the difficulties of coherent analytic reasoning on the 
 deeper problems of life, the imperfect structure of the 
 thinking faculty, and the still more faulty cast of the tools of 
 reasoning — to wit, words. From among many testimonies
 
 136 LETTERS ON REASONING Letter VIIL 
 
 on that head, let me cite to you that of an able reasoner 
 whom I shall further discuss with you later, the Reverend 
 Jonathan Edwards, of famous theological memory. It is 
 over the problem which wrecks his otherwise well-sustained 
 argument that he is fain to make this confession : — 
 
 " Language is indeed very deficient, in respect of terms to express 
 precise truth concerning our own minds, and their faculties and opera- 
 tions. Words were first formed to express external things ; and those 
 that are applied to express Ihings internal and spiritual are almost all 
 borrowed, and used in a sort of figurative sense. Whence they are, 
 most of them, attended with a great deal of ambiguity and unfixedness 
 in their signification." ^ 
 
 Language is in short an instrument still stamped with 
 
 the ineptitudes of primeval thought, still crooked from the 
 
 fallacious bents of dawning intelligence. This, which we 
 
 should regard as the greatest of all the arts if it were not 
 
 the only one'' acquired by all save the malformed, is the 
 
 least improvable of them all. The other arts can rise on 
 
 stepping-stones of discarded and disregarded models ; but 
 
 this had always to carry in itself the perversities set up by 
 
 outgrown ideals. In painting and sculpture, even in the 
 
 art-forms in which language is the material, we can try new 
 
 methods, discarding a poetic as we do a pictorial convention ; 
 
 but the primary art of speech is only by pitifully minute 
 
 degrees capable of transformation. In some aspects, indeed, 
 
 the evolution is considerable : in the specially artistic aspect 
 
 of expression it is marked and relatively rapid ; but as 
 
 regards the improvement of the labour of sheer thinking, it 
 
 is slow and difficult. For mental analysis, so to speak, we 
 
 must still use the stone axe, the knife of flint : witness my 
 
 very metaphors. We cannot shape a new speech by taking 
 
 thought : developed reason is bound to the use of the 
 
 implements forged by the blundering barbarian. 
 
 You will fuid it, by the way, a useful discipline in think- 
 
 ' hhjuiiy into the Freedom of the Will, cd. 1 790, p. 324, part iv., 
 sec. vii.
 
 Letter VIII. LETTERS ON REASONING 137 
 
 ing to scan carefully anything you may write, for laxities of 
 expression. A habit of correcting these is at least a lead to 
 accuracy of argument. All writers fall into them now and 
 then. George Bentham, in his Outline of a Neiv System 
 of Logic (1827), pointed out a startling instance in Arch- 
 bishop Whately's phrase charging certain cultivators of 
 Natural Philosophy with " over/ookiiig\he boundless field that 
 was before them tvithin the legitimate limits of the science." ^ 
 Yet that absurd " bull " was never corrected by its author, 
 for it stands in my copy of his book, published in 1850. 
 In Mill you may find many less flagrant lapses, for example, 
 such a tautology as "final termination."'^ Some exquisite 
 writers, it is true, are very incorrect thinkers on abstract 
 themes ; but I fancy that watchfulness over phrasing 
 promotes as it is promoted by accurate thinking ; and indeed 
 the correct or finished writer is so by reason of thinking 
 closely so far as the expression of his own meaning goes. 
 
 And as with speech, so it is with conduct. Only in a few 
 may the life of thought develop in even partial independence 
 of the huge pressures of habit and inheritance : error lies 
 about us from our infancy, and we mostly grow up with 
 opinions as we grow up with appetites and tastes. Thus the 
 struggle of opinions takes on the ways of the struggle for 
 life, so headstrong, so little moralised. The struggle, in 
 short, is all one ; and it is all Nature. Men's inconsistency 
 in simple matters of conduct, of personal relation, is in the 
 mass so monstrous, so shocking, that merely to reprobate it 
 is in a manner to evade the confession that moral perversity 
 is a part of the order of things. And it is wholesome to 
 face and confess that fact chronically ; for only thereby, 
 unless we be singularly endowed with love, can we hope to 
 keep in charity with our opponents — a thing the more 
 desirable because, if we can infer the future from the past, 
 
 ' Whately's Logic, intiod. 
 
 '^ Examination of Sir William Hamilton's Philosophy, prof, lo 
 3rd ed.
 
 I3S LETTERS ON REASONING Letter VIIL 
 
 the theologians among them will certainly not keep in 
 charity with us. 
 
 My meaning will become more clear when I discuss with 
 you the problem of Freewill, which has been by some 
 uncharitably called the pons asinorum of moral philosophy. 
 That definition, alas, is much too exclusive in the present 
 state of civilisation. Our bridge is narrow, and really not 
 easy to cross ; and I should prefer to call it in all good 
 faith the pons theologoruin, the bridge at which the 
 theologian (now) balks. To cross it, you need a cool and 
 a clear head, and a quite frank curiosity. Take a rest, 
 then, before you approach it.
 
 LETTER IX. 
 
 The moment the issue of " Free-Will " is posited, we are 
 on the verge of a snare constituted by the crudity of the 
 terms with which men first had to work. Let us watch 
 how they have been wont to stumble into it in the past. 
 
 Ages ago, no one can say when first, men were led to 
 brood on the spectacle of human conduct, the play of 
 passion, the as-it-were flagrant fetality of the course of 
 some ; the dooms of strange misfortune, signal prosperity ; 
 the wide diversities of character which seemed to constitute 
 or determine fate ; and the no less wide diversities of fate 
 which seemed to determine character. Out of the common 
 brooding and guessing arose that dim notion of Fate ; of 
 dooms laid by capricious or conflicting Gods, whose will 
 men might not resist. And among those who doubted the 
 existence of such Gods — a type of intellectual variation 
 that tends to arise in every stage of human evolution, but is 
 hard put to it to prosper in any, thus far — there rose the 
 guess that the inferred compulsion lay in the far-off" but 
 strangely enduring stars, whose courses, with cyclic varia- 
 tions, were seen to last from age to age of transitory 
 humanity. Sometimes the Godites accepted the astrological 
 hypothesis, being undisturbed by a fallacy the more ; some- 
 times they resented it as doing dishonour to the Gods. 
 And when, in special conditions, there rose a special passion 
 for what was held to be righteousness, the men most swayed 
 by it were apt to insist that the Gods or God sought only 
 such righteousness, and that men's misdeeds came of their 
 own depravity, or else of their subservience to evil spirits — 
 Gods discredited for the time being.
 
 I40 LETTERS ON REASONING Letter IX. 
 
 Such ail attitude was proper to narrow and strenuous 
 minds, especially to the worshippers of a single God, much 
 feared in their world ; but in communities of an intellectually 
 freer or more cosmopolitan cast the compulsions of character 
 and circumstance were more justly considered ; and with 
 the mixing of races went an interchange of theories, most 
 civilised peoples harbouring both. Thus the later Jews, 
 drawing on the cosmopolitan thought of Babylon, accepted 
 the dogma of an original, depravity, transmitted fatally from 
 the first man to his descendants ; yet all the while held men 
 as punishable by their Maker for their sinfulness. Similarly 
 the historic Greeks held by the thought of a Fate either 
 imposed by some divine volition, swaying men to their own 
 undoing, or enveloping and determining the very Gods 
 themselves, yet never thought of adjusting to that thought 
 their customary law. Thus was the riddle passed on through 
 the generations, with chronic debate, the stress of it some- 
 times turning on the justice of the Gods, sometimes on the 
 guilt of men. 
 
 In the specialised cults which turned to fresh account the 
 primordial superstitions of sacrifice and atonement, both sides 
 of the contradiction naturally persisted. The worshipper 
 was declared to be under the curse of a general if not of 
 a particular guilt ; and the special sacrifice or ceremony 
 took that guilt away : hence faith in the sacrifice was for 
 the priest the first and last stipulation, the indispensable 
 thing, and as such a virtue in itself. But alike in the con- 
 sciousness of priest and of worshipper, " natural " or quasi- 
 critical ethic tended also to hold its ground ; and there 
 struggled forth in the Christian world in particular a confused 
 compromise, in terms of which all men were prone to sin, 
 and needed grace from on high even to repent, yet were 
 reprehensible for not repenting, for not having grace. But 
 at every conscious adjustment the crux took fresh form : you 
 would get grace if you prayed for it ; yet since many were 
 obdurate, you clearly needed grace to pray. The anomaly
 
 Letter IX. LETTERS ON REASONING 141 
 
 had a firm hold further in respect of most men's conscious- 
 ness of their own moral variability, which the antithesis seemed 
 vaguely to explain, yet did not solve. And as surely as the 
 demoralising bias of ritual and the sacrificial principle 
 reduced morals to mummery, the rebelling conscience of 
 the more thoughtful, with its recurring variations towards 
 common-sense and manliness, insisted that virtue and vice 
 lay in the will-, and were to be measured by right and wrong 
 action. On the other hand, the challenged priesthood, 
 menaced on their central standing ground, reverted in more 
 or less good faith to the philosophic defence that if God 
 ruled he must rule men's wills ; and the pagan demonstra- 
 tion of the cosmic reign of Necessity was employed to 
 confute Christian heretics who sought to throw men back on 
 their own moral judgment and initiative. I do not say that 
 this sketch covers all the ground of the evolution in question ; 
 but it will serve broadly to indicate how the form and fortunes 
 of a particular philosophic problem are determined by a long 
 chain of historic circumstances. 
 
 For the modern European world the problem was set up 
 on high, as it were, through the sect-strifes following on 
 the Protestant Reformation. What at first sight seems odd, 
 the Protestant Churches, professedly revolting at the immo- 
 rality of the Papal, did not set up as against the Catholic 
 doctrine of sin and penance and confession and absolution 
 a doctrine of individual responsibility and retribution such 
 as would make morality a matter of reciprocal human duty. 
 On the contrary they stressed heavily the old Augustinian 
 doctrine of predestination and arbitrary divine grace, which 
 in the fifth century had triumphed over the heresy of 
 Pelagius, who stood for the moral freedom of men. Despite 
 that triumph a confused compromise followed, and four 
 centuries later the predestinarian doctrine of Augustine was 
 condemned as a heresy in the person of the monk Gottschalk 
 — so signally do heresy and orthodoxy change places under 
 changed circumstances. It would seem that the authority
 
 142 LETTERS ON REASONING Letter IX. 
 
 of Augustine was able to carry the dogma of predestination 
 in his day because in that time of poHtical ruin the fatality 
 of things had come very deeply home to the southern clergy, 
 part as they were of the falling fabric of Rome ; while the 
 northern Pelagius, nourished on Greek lore, expressed the 
 surviving spirit of human energy. When in the ninth 
 century Gottschalk, a brooding monk, insisted vehemently 
 on the high Augustinian doctrine, he was angrily resisted 
 both by the scholarly Bishop of Mayence, Rabanus Maurus, 
 and by the Gallic Archbishop Hincmar ; and at the same 
 time the temper of self-assertion, natural to a tempestuous 
 barbarian world, found again in John the Scot (i.e., the 
 Irishman) a mind, trained like that of Pelagius in Greek 
 thought, to give it nobler expression. In so far, however, as 
 the question WcT.s ecclesiastically decided against Gottschalk, 
 who was flogged and imprisoned, it was judged partly on 
 economic grounds (Churchmen feeling that Gottschalk's 
 doctrine struck at their power as dispensers of saving 
 sacraments), and partly on grounds of regional jealousy. 
 Thus the problem was never philosophically or even officially 
 solved, inasmuch as John's very remarkable rationalistic 
 pantheism is not at this point clarified ; and his startling 
 rationalism heightened the reaction in Gottschalk's favour, 
 with the result that Hincmar in turn was condemned as a 
 heretic in two councils, and the Church in general receded 
 towards the anti-Pelagian teaching of Augustine. 
 
 In the sixteenth century the lines of opinion were again 
 determined largely by political conditions, for though 
 Luther's early study of Augustine had much to do with his 
 creed, his opposition to indulgences and his relation to the 
 Church of Rome counted for more. Early reformers like 
 Pelagius had sought to upset ceremonialism by denying 
 predestination : Luther and Calvin, with the same aim, 
 affirmed it. The reason was that Protestantism, if it was 
 to fight Romanism with any success, had to stress at once 
 the scriptural side of Christianity and the belief in the
 
 Letter IX. LETTERS ON REASONING 143 
 
 principle of atonement, insisting only that the atonement 
 operated not through the functioning of a special priesthood, 
 but through absolute divine fiat laid on the whole nature of 
 things. Thus divine foreordination became a Protestant 
 tenet ; and there ensued the singular spectacle of another 
 general reversion in the Church of Rome to an ancient 
 heresy of common-sense, in contravention of the teaching 
 of its first great theologian, Augustine. Either way, the 
 object was the same, to establish the special claims of the 
 historic Church as against dissidents from its practice. 
 
 On both sides the moral confusion was boundless, 
 essentially because both sides started from premisses which 
 could not be reconciled with rational moral practice — the 
 premisses of an omnipotent and all-good Creator who fore- 
 ordained human sin while reprobating such sin, and who 
 satisfied his own sense of justice by sacrificing his son as an 
 atonement for the sin thus at once foreordained and con- 
 demned. On this ground, the most powerful intelligences 
 fought without the slightest approach to solution, merely 
 revolving in opposite circles. Start from their premisses 
 and you will see the fatality. If the posited God be 
 administrative and omnipotent, he must have foreordained 
 all things : if he willed that men should sin, he can have no 
 rationally intelligible right to condemn or punish them. 
 Over this dilemma many disputants took up opposite 
 positions which equally annihilated rational morality : one 
 side insisting that the potter was free to do what he would 
 with the clay, which could not without blasphemy challenge 
 him ; the other arguing that what God willed could not be 
 wrong, and that the provision of the atonement was made 
 expressly to balance all human acts, which were thus all 
 alike permissible. But the natural moral-sense or common- 
 sense in turn, though still bound by theology, revolted from 
 both extremes : the more orderly predestinationists on their 
 side insisted in the same breath that God foreordained sin 
 and repentance, salvation and damnation, at his free will,
 
 144 LETTERS ON REASONING Letter IX. 
 
 and that all the same he was perfectly righteous and a hater 
 of sin ; while on the other side there arose new Pelagians, 
 notably the followers of Arminius, who more or less clearly 
 argued that sin and salvation were not foreordained, and 
 that all men were free to seek and win salvation, but 
 that nonetheless God was omnipotent and foreknew all 
 things. 
 
 From these opposing circles you will see there is no 
 rational escape while the premisses stand. You have your 
 line A B, the dogma of creation by a Good Infinite One (a 
 twofold contradiction in terms, since Infinity is that which 
 cannot be added to and cannot be thought as having finite 
 mode), and you simply describe your circle either from the 
 point A or from the point B, neither argument ever absorb- 
 ing or excluding the other. Start from God's creative 
 omnipotence and you have the thesis of Calvin : start from 
 God's Infinite Goodness and you have that of Arminius. 
 Both points of the premiss are vain figments ; yet fixed and 
 sacred premisses they remained. And they so remained for 
 this among other reasons, that, fatally false as they were, 
 they distortedly represented for men the two fundamental 
 facts of moral science : the absolute continuity of causation 
 in human as in non-human phenomena, and the necessary 
 reference of all moral judgment to individual character. 
 
 What, then, is the rational solution ? Shall we afiirm 
 either with the Arminians that "man's will is free," or with 
 Luther that "man's will is slave"? Or, coming to the 
 modern aspect of the struggle — for it goes on to-day as 
 between theists and naturalists — shall we say with the 
 Determinists, " man's will is determined," or with the 
 theologians, who as a body have performed one more 
 complete change of front, "man's will is left free by God"? 
 At first sight there is something bewildering in this per- 
 petual changing of sides ; heterodoxy and orthodoxy alter- 
 nately ranging on this side and that, as in a dance. It is 
 worth while, before we come logically to close quarters with
 
 Letter IX. LETTERS ON REASONING 145 
 
 the problem, to realise why those repeated transformations 
 have latterly come about. 
 
 Broadly speaking, the old Pelagians, as aforesaid, denied 
 predestination by way of maintaining rational moral motives 
 against a dogma which was destroying them; and the Church 
 held by predestination because that doctrine on the whole 
 best consisted at the time with its hold over men's minds. 
 Later, in a much divided world of feudalism, that very 
 doctrine was seen to lend itself to anti-priestly heresy, and 
 was condemned ; till the rationalism of some of its 
 opponents drove orthodoxy back towards the predestinarian 
 position. The first Protestants, in turn, held by predesti- 
 nation because, when once men had on other grounds 
 broken away from Rome, that doctrine was as good for 
 them as for it, and was as it were a bulwark of conviction 
 against Papal pretensions. Rome then returned to quasi- 
 common-sense because that could be used to discredit 
 Protestantism, with its licentious sects ; and Protestantism, 
 in turn, prudently alternated its theoretic predestinationism 
 with a popular asseveration of all men's freedom to come to 
 grace. 
 
 Newer rationalism, in turn, rebelling against the doctrine 
 of eternal torment, adhered to by Catholic and Protestant 
 alike, insisted on the one hand on the enormity of eternal 
 punishment for foreordained sin, and on the other hand 
 pressed against arbitrary theology the truth that belief is not 
 a matter of volition, the will being absolutely determined by 
 motives conditioned by structure of the mind. In the very 
 different hands of Spinoza and Collins this line of thought 
 led to powerful philosophic demonstrations, which struck at 
 the very bases of the Christian creed of salvation by faith. 
 And as the.se demonstrations brought out constantly the 
 principle that all events and acts, including all volitions, are 
 determined by antecedents, it was clearly the cue for 
 Christian theologians to retort that such Necessitarianism 
 (so it was called) made an end of all human responsibility ;
 
 146 LETTERS ON REASONING Letter IX. 
 
 that it would thus reduce society to anarchy ; that morahty 
 depended on the fact of the freedom of the will ; and that 
 therefore God had left men's wills free. All the while their 
 own theology had twice over, in principle, destroyed the 
 basis of rational morals — by its doctrine of predestination 
 by a Good Infinite and by its doctrine of atonement and 
 salvation by faith. But these circumstances, of course, 
 were not put side by side with the new formula of free will. 
 It was the general cue cf theology to appeal to the sanctions 
 of quasi-common-sense morals, and this was duly done. 
 
 Now let us come to the logic of the problem. It is 
 philosophically to be solved at once on the fundamental 
 logico-psychological principle indicated in my last letter — 
 that of the thesis that propositions of single mode cannot 
 rationally be made of the infinity of things. On the same 
 grounds with that proposition stands this, that terms of 
 relative mode cannot rationally be used absolutely of an endless 
 series absolutely considered — or, in other words, that terms of 
 relativity become non-significant when employed to exclude 
 relativity of mode in an infinite series which is itself not 
 thought of as one of two or more correlative modes. Let 
 me explain in detail. When we think of, say, crows, we 
 mentally define them against other birds ; and the statement 
 " all crows are black " is perfectly significant. But when 
 we think of volitions as such, we can define them against no 
 correlative ; they constitute for us an absolute or infinite 
 series ; being coextensive with thought, they belong to no 
 wider species. Any ascription of mode to them, then, is 
 intelligible only as connoting other modes ; as " weak," 
 implying " strong." That is to say, we can never describe 
 them all by a relative term — always excepting the terms of 
 causation, as already explained. And since I can say 
 " weak volition," only because I can recognise also " strong 
 volition," so I can affirm " free volition " only if I recognise 
 also " unfree volition." To say "all are free" is like saying 
 " all are weak." The moral world is the world of moral
 
 Letter IX. LETTERS ON REASONING 147 
 
 judgments, choices, the world of volitions : if, then, we 
 apply a term of moral mode to all volitions (an infinite 
 series, having no defining contrast,^ since non-moral 
 phenomena are wholly out of the question) we reduce it to 
 absolute non-significance. But, equally, " unfree " cannot be 
 predicated of the totality of volitions : the formula " the will 
 is not free " = " all volitions are non-free " is just as meaning- 
 less as the other. Common-sense has here come to grief 
 by reason of the primitively unfit tools with which it worked. 
 " Free " is a term significant solely as contrasted with 
 " unfree " ; it arises as a description of a state or mode of 
 being and doing, and is applicable only to certain states of 
 being and doing as contrasted with different states. Now, 
 it is not suggested on either side of the free-will dispute 
 that some volitions are free and others not free : that notion 
 is excluded by all as unthinkable ; therefore it is at once 
 logically clear that the terms have no rational bearing on 
 volitions at all. 
 
 One philosopher of good standing, Dr. Shadworth 
 Hodgson, has indeed endeavoured to reconcile inde- 
 terminists to the determinist view by showing that, given 
 determinism, there is a " real free-will " in respect of certain 
 acts as compared with others ; that is to say, there is not 
 free-will in such a case as an intensely thirsty man's resolve 
 to drink the first liquor he sees, that being " action deter- 
 mined by a single unresisted motive." Processes of deliberate 
 choice, on the contrary, he describes as " really free " 
 volitions. But unless we are to suppose on the thirsty 
 man's part a mere reflex action (" consensual reflex action " 
 is a phrase apparently meant to convey such a view), in which 
 case there is no "choice" at all, this discrimination is only 
 one more fallacy, for the difference between the thirsty 
 man's choice and that of the less thirsty man is only one of 
 degree of intensity of motive, his choice being " any drink- 
 
 ^ As it is put by Spinoza at the beginning of his Ethica : '■'•At corpus 
 non terniinatur cogitatione, nee cogitatio corpore." 
 
 M
 
 148 LETTERS ON REASONING Letter IX. 
 
 able liquor rather than more thirst." Unless he be insane, 
 however, he will not drink boiling water, or a known deadly 
 poison. The problem, then, really remains as it was. If there 
 be anything in the argument before us, it would be equally 
 valid as regards the case of a man with an overwhelming 
 bias to drink alcohol to excess when he might drink water : 
 of his case too it would have to be said that it showed " no 
 trace of free-will." But this is not suggested, or apparently 
 seen, by Dr. Hodgson. He has erred in the usual way in 
 stating the problem, for he seeks to confute those who, he 
 says, " deny the freedom of volition," and so would " rob 
 the words duty, conscience, right, and wrong of all distinctive 
 meaning." The logical course would be, as above con- 
 tended, to point out that denial of " freedom of volition" is 
 equally with assertion of it a spurious predication, a 
 meaningless proposition. If any determinists, so-called, 
 talk of "necessity" and "compulsion" in the process of 
 volition, they are morally reverting to the theological fallacy 
 (which we shall consider later), and logically reverting to the 
 fallacy of predicating one term of a relation concerning an 
 absolute series to the exclusion of the other ; for 
 " necessity " and " compulsion " are only synonyms for the 
 negation of freedom.^ But Dr. Hodgson implicitly commits 
 the same fallacy. For lack of a true logical analysis, in short, 
 a whole series of recent thinkers have, as we shall see 
 further, added fresh confusion to a problem which they 
 should have formally cancelled as illicit. 
 
 What men are confusedly pointing at when they speak of 
 
 ^ J. S. Mill {Logic, B. vi., c. ii., §§ 2, 3) rightly condemns the use of 
 the word necessity in this connection, but does not show, as he ought, 
 how it is theoretically or logically wrong, and in his anxiety to oppose 
 the Owenites he himself falls into confusions. The former objection 
 applies also to the otherwise decisive argument of T. H. Green 
 {rrolegomeiia to Ethics, B. ii., c. i., §§ 106-110) ; but he avoids Mill's 
 confusion, though the editorial synopsis of § 108 obscures its argument 
 in Mill's sense.
 
 Letter IX. LETTERS ON REASONING 149 
 
 " free " or " non-free " volition is the pseudo-problem whether 
 volitions are caused or uncaused. But " non-free " is not 
 equivalent with "caused," and "caused" is not the anti- 
 thesis of " free " : they conceptually quadrate or compare 
 in no way whatever, any more than " black " and " long." 
 Causation, as I said before, is not finally a modal or differ- 
 entiating conception at all : it abstracts itself in thought as 
 an element in every rational conception of existence or 
 happening ; and there is no psychic or ideal contrast to it, 
 since non-causation is merely a word to which on reflection 
 no concept whatever can logically be attached. " Un- 
 causedly " can be made to do rational duty only as a bad 
 synonym for " inexplicably " or " unintelligibly " caused, as 
 when we talk of a " causeless caprice " or " causeless 
 anger." 
 
 It may occur to you here that I am " begging the 
 question " as regards volitions — taking for granted the 
 thing in dispute. Rather I am applying a logical law which 
 quashes the form of the dispute, even as it would dispose 
 of a dispute as to whether infinite length is hard or soft.^ 
 But as it happens that even men who ostensibly recognise 
 the co-extent of causation with existence, the aspect of 
 causation in all happening (our tools here, you see, are still 
 sadly clumsy !), yet stumble over the figment of uncaused 
 volitions, I will now go with you over the ground a posteriori 
 in order to bring home to you the reality of causation in the 
 world of volition as it does come home to us when we begin 
 handling the problem from within. 
 
 We cannot do better than take it up in company with 
 Jonathan Edwards, who approached the problem from 
 within, recognising indeed the logical meaninglessness of the 
 
 1 J. S. Mill argues, as against Professor Bain (System of Logic, i-vol. 
 ed., p. 26, ttote) that "we can certainly predicate of a sound, or a 
 smell, that it is not white." We certainly caii utter verbiage ; but we 
 can also refuse to debate about it ; and meaningless propositions should 
 not rank as predications. 
 
 M 2
 
 ISO LETTERS ON REASONING Letter IX. 
 
 ordinary formulation,^ but reasoning nevertheless from point 
 to point of the concrete case with a calm coherence of 
 power which marks him a born thinker. He might have 
 found in Spinoza and in Hobbes the finished form of the 
 argument he was compassing ; but as a devout Christian he 
 had abstained from reading Hobbes and Spinoza ; and he 
 forced his way through the argument in the character of a 
 Calvinist determined to clear up the issue as between him- 
 self and the Arminians. His demonstration is broadly in 
 this wise. 
 
 Criticising Locke, he notes that " A man never, in any 
 instance, wills anything contrary to his desires, or desires 
 anything contrary to his will " ; all alleged cases of that kind 
 being so merely through miscarriage of the terms. The 
 will, then, is determined or set to work by something, and 
 unless it is " a cause that acts and produces effects upon 
 itself " it is determined by motives : that is to say, " the will 
 always is as the greatest apparent good is," or " the will is 
 always determined by the strongest motive." But it is 
 obvious that different men are very differently affected by 
 the same appeal, that a given motive acts in them in very 
 different degrees, and that it may act in the same man with 
 different force at different times. Some men, in fact, are 
 either always or occasionally " morally unable " to respond 
 to a given moral appeal. This may be a matter either of 
 instruction or of natural bias. A good man, broadly speak- 
 ing, is one whose bias is good. In other words, his will is 
 predisposed to good actions, so that he does without struggle 
 of inclinations what another does only after struggle, or, for 
 lack of sufficiently strong good inclination, does not do at 
 all. 
 
 ^ E.g., When he remarks that " in propriety of speech neither liberty 
 nor its contrary can properly be ascribed to any being or thing but that 
 
 which has will For the will itself is not an agent that has 
 
 a will ; the power of choosing itself has not a power of choosing" 
 (Pt. I., § v.). This was previously pointed out by Slrutt and by 
 Locke.
 
 LeUcr IX. LETTERS ON REASONING 151 
 
 But here arises a crux for both the naturalist and the 
 theist. If I happen to have strong inclinations to evil and 
 yet conquer them, am I more or less " virtuous " than one 
 whose inclinations are wholly or almost wholly good ? The 
 question is often answered by both theologians and non- 
 theologians with an award of special merit to the character 
 which triumphs over its evil inclinations. Theology used 
 almost normally to make virtue consist in self-denial ; so 
 that the man with the most complete inclination to good 
 must presumptively be good with less " virtue " than one 
 who had strongly conflicting inclinations. But common- 
 sense, though apt to endorse the view that the latter type 
 deserves special praise, is apt also to be staggered by the 
 consequence that a man gets credit for special virtue on the 
 strength of the bad elements in his nature ; and both the 
 latter-day theologian and the early rationalist have been 
 moved to " hedge " over the problem. 
 
 Then there is the further theological crux : If the bad 
 man is so by reason of an over-plus of bad inclinations 
 which are innate, how can he be deserving of punishment 
 as compared with the good man, who, so to speak, could 
 not go wrong ? The first instinct of both the theologian 
 and the natural moralist, on seeing the point, is to fall back, 
 however inconsistently, on the notion that all men start alike, 
 and that it is by " their own doing," so to speak, that they 
 come to have an overwhelming bias towards evil. But this 
 position, in which many theologians stolidly persist, is a 
 mere evasion of the problem ; for we are left asking : If all 
 start alike, why do some begin to diverge to the wrong side ? 
 Why does any one will towards evil? On this view, Oest le 
 premier pas qui coute. If A from infancy leans to the right 
 side and B from infancy leans to the wrong, what is the 
 primary cause ? Observe that if you say there is no pre- 
 disposing cause (and many theologians in effect say this) you 
 have at once denied the principle of causation in the 
 universe and quashed the theological theorem of design, as
 
 152 LETTERS ON REASONING Letter IX. 
 
 well as that of " original sin." Theologically speaking, God 
 on this view did not design anything whatever in regard to 
 men's wills : non-theologically speaking, causation is given 
 up, and reasoning ceases. 
 
 All this Edwards saw with perfect clearness ; and accord- 
 ingly he is firm in his Calvinistic doctrine that " God " did 
 fore-ordain how men should will and act. It was for him, 
 as a devout Christian, the only way of conceiving causation, 
 the only way of being rational and consistent in his thought. 
 But in virtue of his consistent theism he inevitably proceeds 
 to propositions which utterly repel the civilised moral sense ; 
 and, seeking to salve his conscience at a critical point, he no 
 less inevitably destroys his previous argument. Up to a 
 certain point he proceeds securely, save that he needlessly 
 abandons his logical foot-hold as to the meaning of " liberty." 
 Inasmuch as liberty means the power to do as one pleases, 
 he said, one must " please " or desire something in order to 
 experience liberty. This bias, then, is clearly not a negation 
 of liberty : it is the condition of the realisation of liberty ; 
 and it was logically unnecessary, as regards the human 
 problem, to argue as Edwards does after Locke, that no one 
 can conceivably desire to have been without predisposition. 
 That would be "to desire to have no desires" — a contra- 
 diction in terms. 
 
 As regards virtue and vice, again, Edwards sees and shows 
 that both consist in bias or predisposition,^ and that it is 
 vain to say a man is not wicked or reprehensible because 
 his evil bias is innate, since it is just markedly evil bias that 
 constitutes wickedness or reprehensibility. But as soon as 
 he comes to close quarters with his theology he begins to do 
 alternate violence to good feeling and to rectitude, to moral 
 instinct and to consistent reasoning. Much of his argument 
 is given to purely " scriptural " considerations, to the recon- 
 
 ^ Mis formula of virtue as "benevolence towards all being," ])y the 
 way, was put before him, in almost his words, by Bishop Cumber- 
 land.
 
 Letter IX. LETTERS ON REASONING 153 
 
 ciling of Christian doctrine and narrative with the nature 
 of things ; and here his logic is necessarily mere quibbling. 
 But we reach something worse than quibbling when we 
 come to his reconciliation of the fact of the existence of sin 
 with his dogma of a punishing deity. 
 
 " It was meet," he writes, " «/" Sin did come into existence, and 
 appear in the world, it should arise from the imperfection which 
 properly belongs to a creature, as such, and should appear to do so, that 
 it might appear not to be from God as the efficient or fountain. But 
 this could not have been, if man had been made at first with Sin in 
 his heart ; nor unless the abiding principle and habit of Sin were first 
 hUrodiiced by an evil act of the creature. If sin had not arose from 
 the imperfection of the creature, it would not have been so visible that 
 it did not arise from God as the positive cause and real source of it." 
 
 Here we have one of the acutest of reasoners committing 
 one of the grossest conceivable confusions of argument. 
 Not only does he here flatly negate his own main doctrine, 
 he relapses into sheer nullity of phrase. The whole point 
 of the passage is that as Sin evidently did arise "by an evil 
 act of the creature," therefore it 77iust have done so. On 
 that simple plan, the argument could have been stopped by 
 an opponent at any stage : the Arminian had only to say 
 that men's wills are evidently not predetermined, and there- 
 fore cannot have been. And Edwards, hypnotised as he is 
 at this point by his dogmatics, feels that something is wrong, 
 for he goes on to protest that, as regards the objections of 
 the Arminians, 
 
 " No additional difficulty is incurred by adhering to a scheme in this 
 manner differing from theirs, and none would be removed or avoided by 
 agreeing with and maintaining theirs. Nothing that the Arminians say 
 about the contingence or self-determining power of man's will can serve 
 to explain, with less difficulty, how the first sinful volition of mankind 
 could take place, and man be justly charged with the blame of it." 
 
 This is quite true : all theists are in the same dilemma ; and 
 on theistic lines the strife of Calvinism and Arminianism is 
 absolutely insoluble. But that is the condemnation of both.
 
 154 LETTERS ON REASONING Letter IX. 
 
 The rational solution is one that annihilates the whole 
 theistic premisses. All of them alike— Creation, Fore- 
 ordination, Infinite Goodness, Infinite Justice, Grace, 
 Original Sin, and Eternal Punishment — are pure counter- 
 senses, the result of applying the categories of the finite and 
 relative to the concept of the All ; and they held their place 
 in such a mind as that of Edwards, as they do in weaker 
 minds to this day, in virtue of the capacity of man to be 
 hypnotised by traditionary error— call it Myth, Religion, 
 Authority, or what you will. Tantum relligio potuit suadere 
 malorum. 
 
 To-day, a mind of such power as belonged to that of 
 Edwards, so placed as to be fairly open from childhood 
 onwards to rational influences, could not hold his creed ; 
 could not stand with him on the doctrine of the sin and fall 
 of Adam, and the consequent predeterminate "guilt" of 
 the whole human race, to the point of justifying the dogma 
 of infant damnation. It was the peculiar emotional habit 
 set up by early and continuous superinduced hysteria on 
 Christian lines that so warped such an intelligence. Had 
 he even met in his youth with the Eihica of Spinoza, where 
 are set forth so many of his arguments on will, he might 
 have been emancipated by the one penetrating thought, ad 
 Dei naturam neque intellectum neque voluntatem pertitiere, 
 "to the nature of God neither intellect nor will pertains." ^ 
 That indeed is not a final deliverance from the fallacy of 
 theism ; for Spinoza himself, while thus logically excluding 
 from his doctrine of the Infinite those finite modes, proceeds, 
 in affirming the necessary determination of all things, to 
 ascribe to the infinite " omnipotence " — all-powerful?z^5j-, a 
 mere abstraction of the idea of force with "All," and 
 " capacity for " superadded. He might as well have affirmed 
 " All-will." In the infinite indeed inheres all force, but the 
 abstraction " all-forcefulness " is a mere verbalism ; and the 
 " he " of Spinoza is a perpetual paralogism. 
 
 ' Et/iica, Pt. I, Prop, xvii, Schol. Propp. xxxi, xxxii.
 
 Letter IX. LETTERS ON REASONING 155 
 
 Great, however, was the moral gain from the great step 
 Spinoza did take, for in his ethic does the better spirit of 
 humanity first begin to find its coherently reasoned justifica- 
 tion. Hitherto the best instincts of men had been over- 
 ridden by creeds which divinised the ethic of savagery, 
 placing beside the ordinance of compassion the dogma of 
 retribution. The paralogism of a Good God had lent itself 
 to every cruelty, the animal bias of men glorifying and 
 sanctioning itself by attributing to Infinity its own appetite 
 for protracted revenge, and its own expedients of imprison- 
 ment and torture. The whole of that evil ethic followed 
 on the irrational conception of Infinity as a Will, a Creator, 
 a Governing Person, who hated and punished sin. Thus 
 have religious people, at all times, with the best intentions, 
 as the phrase goes, been worsening life with the idea of 
 bettering it, shutting a door on moral progress by way of 
 terrifying sinners into righteousness. 
 
 When once we substitute for the irrational conception of 
 a Personal Infinite the rational conception of universal 
 causation or causal sequence, the problem of morals presents 
 no difficulties save those (certainly numerous) involved in 
 being consistent despite of tradition or bias and habit. We 
 recognise at once that men vary congenitally in moral bias 
 and capacity as in every other respect. Where the theologian, 
 in defiance of the whole spectacle of life, insists that the 
 worst evil-doers knoiv they do evil precisely as others know 
 it, making at most a grudging and inconsistent exception in 
 favour of the unquestionably insane, the naturalist recognises 
 many degrees of moral incapacity from " insanity " to quasi- 
 perfection. It is the theological habit of charging men with 
 "sinning against the light" that involves men in such 
 absurdities : they will not surrender the carnal joy which 
 such denunciation gives them. I have heard a theologian 
 thus publicly declare that a criminal who manifestly took 
 pleasure in atrocious murders knew perfectly well how 
 criminal he was ; and a number of theists, thus fooled to
 
 156 LETTERS ON REASONING Letter IX. 
 
 the top of their bent, applauded him. Such words, to mean 
 anything in terms of consciousness, must mean that the 
 criminal in question found his crimes as loathsome to enact 
 as we should find them, yet went on committing them 
 because he chose to do so. Here we have the inveterate 
 absurdity of the moral absolutist. 
 
 It will not, I hope, put you to much trouble to be met 
 by the old parry of the theologian, to the effect that, though 
 there are many cogent arguments in favour of the causedness 
 of volition, yet they are set aside by the primordial fact that, 
 nevertheless, " we know [or feel] that we are free to will." 
 It is the fact, however, that this argument has given trouble 
 to some ostensibly non-theological reasoners, as Professor 
 Fowler, and even to Professor Sidgwick. The former has 
 said ^ that the conflict between the feeling of Hberty of 
 volition and the unanswerable arguments for determinism is 
 insoluble, because determinism in his view excludes praise 
 and blame, which are nevertheless irreducible ; and the 
 latter has avowed similar difficulty, only professing ^ to offer 
 a "practical solution" by way of the avowal that we 
 " cannot use determinist conceptions " when we are making 
 a choice. It may be well, then, to put briefly the logical 
 solution of their difficulties. 
 
 With the question of praise and blame I shall deal later : 
 meantime let us note first how both Professors miscarry 
 because they have never looked to their logical funda- 
 mentals. Had they done so they would have noted that 
 the verbal predications of "freedom" and "unfreedom" can 
 have no rational bearing singly on the absolute category of 
 volitions, inasmuch as they are admittedly not both applic- 
 able ; and that accordingly the impossibility of " using a 
 determinist conception " when making a choice is no more 
 of a philosophical difficulty than the impossibility of 
 
 ^ Principles of Morals, pt. ii, 336. 
 
 2 Article on " Some Fundamental Ethical Controversies," in Mind, 
 October, 1889.
 
 Letter IX. LETTERS ON REASONING 157 
 
 applying to choice the concept of extension, or to hght 
 the concept of gravitation. But as Professor Sidgwick had 
 missed this consideration, he would probably not have been 
 convinced by it ; and to one in his position we should have 
 to put the matter a posteriori. As thus. It is only at 
 first sight that our volitions seem even to common-sense to 
 be " free " ; for if Professor Sidgwick had been invited to 
 commit a series of murders for gain he would have found in 
 his moral bias to the contrary an invincible force. Even if, 
 however, this bias be itself defined as a volition, a choice, 
 and the " difficulty " thus reasserted, the shock of the logic 
 of determinism is to be got over just as is the shock of dis- 
 covering that the rising and setting of the sun are only 
 apparent motions of the sun. When I decide to take a 
 walk I certainly do not " use " the scientific conception of 
 the gravitation of my body towards the earth's centre, or 
 that of the rotation of the earth on its axis, because those 
 conceptions have nothing to do with my purpose. My 
 "common-sense" knows of no downward pull upon me, 
 because my whole existence is conditioned by that down- 
 ward pull ; and it knows of no whirling of the earth, pre- 
 cisely because of my physical relation to the earth. But 
 who has now any difficulty in acknowledging the gravitation 
 and the rotation ? 
 
 At first, no doubt, it was otherwise : to most men of the 
 sixteenth century the doctrine of Copernicus seemed utter 
 folly ; and even to this day we practically do not " use " the 
 conception of gravitation for a single act, though men of 
 science use it for research into the processes of the cosmos. 
 But simple familiarity with the conception excludes all sense 
 of " difficulty " ; and it will be just the same when men have 
 become rationally familiar with the conception of deter- 
 minism. As we move in gravitation or rest in the earth's 
 movement, we choose in the process of determination. 
 Satisfaction in this view is substantially a matter of habit of 
 mind : as Mill put it, " what persons can and what they
 
 158 LETTERS ON REASONING Letter IX. 
 
 cannot conceive is very much an affair of accident." ^ We 
 shall not say that we are " free " to will any more than that 
 we are " free " to fly or to live under water. We shall 
 continue to say that we are or desire to be free to act as we 
 will or would, having willed in terms of motives. Thinkers 
 who are still at Professor Sidgwick's point of view would do 
 well to go back to Locke's query whether anybody wants 
 to be " free " to will without any motive or bias. 
 
 And the moral gain from the expUcit acceptance of the 
 law of causation in volition is this, that just as the concep- 
 tion of gravitation immensely advances our comprehension 
 of the relations of our earth to the cosmos, so the conception 
 of determinism immensely advances our comprehension of 
 our moral relations to each other ; or rather, let us say, it 
 visibly will so advance us. Step by step the theologian has 
 been driven from the insane ethic of infant damnation, the 
 insane practices of heretic-burning and witch-burning, the 
 blind cruelty of torturing witnesses and flogging madmen, to 
 a relative sanity and decency of method. But we are still, 
 in Emerson's phrase, " at cockcrow and the morning star." 
 The late Dr. Martineau, a too eloquent apostle of bias, was 
 wont to argue that the determinist view of Hfe altered its 
 whole moral dynamics. In a sense (the expression is too 
 loose to be fixable) it will do that ; but only gradually, even 
 for the determinist ; and the indeterminist is still to reckon 
 with. When the reasoner urges on the man of instinct that 
 all volition is determined by capacity, training, and circum- 
 stances ; that all punishment is mere animalism unless it be 
 rationally calculated to work reformation, and that most 
 punishment actually does not and cannot work reformation, 
 the man of instinct throws up new defences — certain 
 sophists and certain imperfect reasoncrs helping him — in the 
 shape of (i) protests that such rationalism takes away the 
 sense of "responsibility," and (2) questions as to how we 
 continue to bestow praise and blame, to feel admiration or 
 ^ System of Logic, B. iii, c. v, § ii : i-vol. ed., p. 237.
 
 Letter IX. LETTERS ON REASONING 159 
 
 repulsion, or to be capable of remorse. Here, as we have 
 seen, a thinker quite friendly to determinism, Professor 
 Fowler, capitulates to the difficulty. " For the last enemy 
 to be destroyed is fallacy." 
 
 By this time, however, you (I hope) will have little 
 difficulty in defending yourselves against the new assault of 
 verbalisms. " Responsibihty " is a general synonym for 
 duty, either explicitly accepted or legally and generally 
 imposed. Responsibility on the part of a " creature " to its 
 alleged " creator," of the clay to the potter, is unthinkable : 
 as well talk of the responsibility of a bullet to a gun, or to 
 the marksman. Such an idea is one more application of a 
 category of discrimination to an absolute series without its 
 correlatives. But on the part of man towards man there are 
 responsibilities general and special ; and inasmuch as all 
 men are on one plane (with endless diversities of capacity 
 and opportunity) as regards the determinateness of their 
 wills, each has the right, in general, to make contracts with 
 others, and all collectively have the right to frame laws for 
 the equal protection of all. By " have the right " I mean that 
 so to act is in consistency with the most generally accepted 
 of all moral propositions — that we ought to do as we would 
 be done by, or, more strictly, ought not to do what we know 
 we should resent having done to us. Difficulties begin to 
 arise round that principle when we consider the case of the 
 wrongdoer who would like to be left unblamed; and we 
 proceed to solve them by framing rules in what seem to be 
 the general interest, deciding that he who commits an initial 
 aggression on another is fitly to be restrained or deterred. 
 
 And now comes more clearly into view the immense 
 moral superiority of the ethic of reason over the ethic of 
 religion. While the traditions and devices of barbarism are 
 accepted as " revelation," and fortified by the fallacies of 
 theism, it is held that he who does wrong must be made to 
 suffer because " sin ought to be punished " — that is to say, 
 the proposition is merely restated in terms of revelation and
 
 i6o LETTERS ON REASONING Letter IX. 
 
 primitive instinct. The theist of to-day is content merely to 
 confine the offending madman as not being " responsible," 
 but is still resolute to make the bad man suffer. Now, so 
 far as we know, madman and bad man differ only in this, 
 that the one's brain is diseased or injured, and the other's is 
 congenitally ill-balanced. It may be argued, indeed, that 
 the dependence of mind on brain is "only an inference," 
 not a certainly ; but unless we are to revert to the Arminian 
 chimsera of a personality without bias or predisposition, we 
 must at least say that the bad man is somehow predisposed 
 to evil. Rationally stated, then, the problem is this : Is the 
 badness such as is proved to constitute an intolerable risk 
 to society ; and if yes, how shall society best be protected ? 
 Given alternative modes of protection, that is rationally to 
 be preferred which causes least suffering to the wrongdoer, 
 unless the other would (a) demonstrably tend to reform him, 
 or ib) burden society unduly — two problems calling for 
 much careful calculation. 
 
 To act otherwise is to be intellectually inconsistent, and 
 temperamentally brutal. If all evil-doing ought to entail 
 suffering, the madman should not escape : indeed, it is 
 probable that some admitted madmen are to some extent 
 more amenable to threats and suffering than some certificated 
 " sane " criminals. But why, seeing evil done and suffering 
 caused, should I seek therefore to cause more suffering? 
 The instinct in that direction is no rational pretext : like the 
 instinct of the criminal himself it is sheer animalism : only 
 the hope of doing good can rationalise an act of retribution. 
 Now, the question as to how far punishments have ever done 
 good or may be made to do so, either by reforming offenders 
 or "deterring others," is one of social art, so to speak; one 
 of comparison and research, of sociology and psychology. 
 Having discussed it elsewhere, I shall not occupy you with 
 it here : the essential point in this connection is that we are 
 bound in reason thus to substitute the conception of human 
 utility for the conception of that duty of social revenge in
 
 Letter IX. LETTERS ON REASONING i6i 
 
 the name of Deity which is argued for by the very people 
 who affirm the same Deity's prohibition of individual 
 revenge, and quote the text, " Vengeance is mine, saith the 
 Lord." By putting all such insincerities aside, we are 
 capacitated to combine compassion with self-preservation, 
 to act as rational believers in the causation of all conduct 
 without foregoing a single means of influencing conduct for 
 the better. 
 
 This last claim you may hear denied by those who say that 
 the doctrine of determinism must tend to corrupt character 
 by teaching men to regard their weaknesses and vices as 
 incurable. Now, if that were true, mankind must long ago 
 have been corrupted beyond remedy by the doctrine of the 
 Infinite and All-Good Creator. Some measure of such 
 corruption has doubtless taken place. ^ But such is the 
 natural instinct of men to self-esteem and to hope for 
 betterment, that, though bad men may have caught at 
 predestination as a formal excuse for their sins, and some 
 are bewildered by theology into antinomianism, men of 
 average character seem almost always to count on becoming 
 better either by " grace " or by good resolutions. And 
 while a weak theist may be argued into recklessness by a 
 propounder of predestination, no sane rationalist can be 
 moved to " give way " to his worse propensities by the mere 
 thought that they are innate, when all the while he knows 
 his better tendencies are equally innate. Nothing can prove 
 that he is determinate to a given end until that end is 
 reached : his volition is always the point of determination ; 
 and in the terms of the case his next volition is unknown ; 
 the die is never cast, so to speak, till it no longer matters 
 how he argues. When a man has lost hope for betterment 
 the machine is already out of gear. 
 
 ^ Rabanus Maurus, for instance, declared that the teaching of 
 Gottschalk had driven many people to moral despair, and others to 
 reckless licence (cited by Waddington, Hist, of the Church, 1833, 
 p. 259, n. ).
 
 i62 LETTERS ON REASONING Letter IX. 
 
 So obvious is this that the opponents of determinism are 
 fain to fall back on another plea. Granted, they say — and, 
 as we said, they can quote the admission of a would-be 
 determinist on their side— granted that a determinist is not 
 shut up to inanition on his own part by his belief, at least 
 he is shut up in consistency to abstaining from praise and 
 blame of others, thus losing a powerful means of affecting 
 others for the better. This again is pure fallacy, the corre- 
 lative of the fallacy about responsibility. What we have 
 affirmed about conduct is its causation : now, the recognition 
 of causation in an aspect oi physical nature has never with- 
 held men from the language of praise, though (and this is 
 hopefully significant) it has withheld them from the language 
 of moral blame. We " praise " beauty in a landscape, we 
 praise beauty in a face — though not to the beautiful person, 
 save in circumstances which I need not detail. If, then, we 
 take avowed joy in the sight of physical beauty, knowing 
 that it is predeterminate in the person as in the landscape, 
 what should conceivably withhold us from taking avowed 
 joy in the sight of goodness, moral beauty, seeing that there 
 is practically no restriction of prudence or taste — save by 
 way of expression — on the avowal ? Praise is fundamentally 
 the expression of pleasure : it is only on theological pre- 
 suppositions that moral praise is made to seem something 
 
 else. 
 
 As regards blame, again, where is the trouble ? The plea 
 of causation is indeed a rational rebuttal to cruelty, to mere 
 vengeance ; and so much the better ; but it is no bar to 
 self-preservation. The most devoted champion of the 
 indeterminateness of the human will will grant that a tiger 
 is rather plainly pre-disposed to prey ; but will probably not 
 go on to say that because the tiger is thus "not responsible" 
 we ought to let him eat us. It is the same with a 
 thug or a thief : our right of self-preservation is primordial, 
 and no argument can undermine it. And so, in degree, 
 with blame. Certainly the consistent determinist (no easy
 
 Letter IX. LETTERS ON REASONING 163 
 
 thing to be, I confess) will withhold blame where he believes 
 blame can do no good, just as a physician will not scold a 
 lunatic, or even (at the moment of diagnosis) a fever-patient. 
 But where he has reason to think blame cafi do good, where, 
 feeling wronged, he may reasonably reckon on making the 
 wronger realise his injustice and inconsistency, his blame is 
 as rational as it is spontaneous. That he should control his 
 expression of it by his sense of purpose — whether to influence 
 the wrong-doer or to counteract his tendencies by warning 
 others — is pure moral gain. And where, finally, we recognise 
 that he who wrongs us or others is too badly biassed to be 
 influenced by our blame, we have one of two courses rationally 
 open to us. Either the wrong-doer is liable to restraint under 
 law or he is not. If he is, verbal blame is superfluous : the 
 " punishment " is blame put in action ; and the scoldings 
 administered by judges to offenders in the act of decreeing 
 punishment on them are neither rational nor seemly. 
 
 If, again, the wrong-doing is of a kind that is not provided 
 for by law (and sometimes even if it is !) our rational course 
 is either to keep aloof from the offender, thereby expressing 
 our blame with tolerably practical emphasis, or, if from com- 
 passion or a higher sense of duty we go on enduring his 
 society, to show by our natural sadness or restraint what we 
 feel. Beyond this, on no reasonable pretence can we be 
 called upon to go. No more than we can go into raptures 
 over physical ugliness can we feel happiness or trustful 
 comradeship in the presence of moral deformity ; nor does 
 the principle of determinism commit us to pretending that 
 we do. With an ugly person, certainly, we are committed 
 by good feeling and right reason to suppressing all sign of 
 discomfort : to do otherwise would be to inflict avoidable 
 suffering with no compensating gain. With an insane 
 person we act as does the physician, for the reason last 
 given as well as out of compassion. But with a bad person 
 we feel distrust ; and as the indication of this is the last 
 chance of influencing him for good, we are under no duty 
 
 N
 
 i64 LETTERS ON REASONING Letter IX. 
 
 whatever of restraining it, save as regards "economy" or 
 the judicious expression of it. The rationale of the matter 
 is — I will not say, as clear as on the theistic view : it is 
 incomparably clearer. 
 
 Yet, I have heard a distinguished clerical writer childishly 
 argue that on the determinist view all praise and blame must 
 amount to a mere " moral stratagem," as who should say 
 that it is a stratagem to smile with pleasure, or to wince 
 when you burn your hand ; and there may still be in circu- 
 lation in your day a work in which a statesman of to-day ^ 
 argues, just as childishly, to my thinking, that a determinist 
 cannot rationally feel remorse for wrong-doing. It is as if 
 one should say that a determinist cannot rationally be sorry 
 to find he has cancer, or because he was born with a club 
 foot. Mr. Balfour, it is true, tries to give a philosophic 
 aspect to his argument by urging that we cannot enjoy 
 beauty of colour or sound unless we conceive such beauty 
 as something akin to a transcendental beauty somewhere 
 perceived by a transcendental mind ; and that for that kind 
 of reason we shall fail to discriminate with praise and blame 
 between either moral or physical charm and ugliness when 
 we conceive them all as equally determinate moments in the 
 infinite whole. But until Mr. Balfour can assure or convince 
 us that he enjoys his dinner only because he thinks a 
 transcendental power transcendently enjoys transcendental 
 meals, I shall take leave to regard his argument as the most 
 remarkable display of philosophic insincerity made in my 
 time. 
 
 These devices of theological polemic, you will note, are 
 all instances of the fallacy of applying relative terms singly 
 to unrelated categories, affirming one term of a contrast 
 singly of an uncontrasted series, to the exclusion of the 
 other term. Did men reason rigorously, such fallacies 
 could never arise ; but even a loose reasoner, were he not 
 fixed in his presupposition, would see the futility of arguing 
 ^ Mr. Balfour in his Foundations of Belief .
 
 Letter IX. LETTERS ON REASONING 165 
 
 that determinism ought logically to make us indifl'erenl to 
 our moral conditions, as being determinate, while tacitly 
 recognising that it cannot conceivably make us indifferent 
 to our physical conditions, which are equally determinate. 
 The sophism is indeed doubly absurd, for, in the very terms 
 of the determinist case, we start from a moral bias, which is 
 the negation of moral indifference. The anti-determinist in 
 effect says. Because you believe you have a bias, you ought 
 not to have a bias. Such ultimate absurdities come of the 
 spirit of partisanship, against which I have before warned 
 you. All the while, the partisan here is on other grounds 
 committed, like theologians in general, to the doctrine of 
 foreordination in its theistic form : he oppugns it simply 
 when it is stated in its non-theistic form ; in which case he 
 will not scruple to negate causation. Now, it is precisely 
 the theist who is stultified when he condemns any aspect of 
 things, moral or physical, since his most solemn and most 
 general doctrine is that all things happen by God's holy 
 will, under which they work together for good. It has been 
 repeatedly declared that the essence of religion is submission 
 to the omnipotent divine will. But the primordial figment 
 commits its acceptor to perpetual inconsequence : in the 
 name of religion he is to be remorseful for certain of his own 
 acts which are alleged to reverse the omnipotent divine will, 
 and he is to condemn other men's acts on the same score 
 even when accepting them as part of that will. For the 
 naturalist there is no countersense of divine will in the case, 
 and there is no inconsequence in his rational relation to 
 the order of things. He resists moral evil as he resists 
 physical evil, on the score that his own moral judgment, 
 playing on all the knowledge he can attain, is for him 
 at each moment the crowning stage of moral volition in the 
 cosmos. 
 
 We can now solve without difficulty the old problem as to 
 whether the man most to be admired is he who has no evil 
 propensities to resist, or he who has them, but in general 
 
 N 2
 
 l66 LETTERS ON REASONING Letter IX. 
 
 or finally overcomes them. Our admiration is properly the 
 expression of our pleasure after due reflection, and we can 
 take no pleasure in the thought that any one has evil pro- 
 pensities. What we do is to rejoice with him when he over- 
 comes them — as, for instance, when a man with a bias to 
 alcoholism is so moved by his perception of the harm it 
 does, or by the blame or contempt or appeals addressed to 
 him, as to acquire a stronger bias, controlling the first. To 
 rejoice with him is not to say he is a better man than one 
 quite soundly constituted ; but, on the other hand, for the 
 determinist to say that he is a faultier man is not to raise 
 any question of his liability to either punishment or reward : 
 it is to bring his controlled propensity into line with a lame- 
 ness, as a thing calling for compassion or sympathy and no 
 longer for blame, though blame, intelligently adjusted, may 
 again become fitting if the propensity reasserts itself. 
 
 Long as this letter has grown, I must not end it without 
 reverting to that old theorem which we have seen employed 
 by Hegel to vindicate theism — the theorem, namely, that 
 " evil is non-ens " or " not positive." The thought is one 
 that still appeals to earnest and reflective people ; but you 
 will see on a moment's consideration that it is wholly be- 
 side the problems of morals, and that it is but a half- 
 thought at best. It was employed by such a gifted thinker 
 as John the Scot, who got it from earlier thinkers, to solve 
 the old dilemma of God's authorship of evil ; and it is so 
 employed still. But the whole procedure is stamped with 
 fallacy, for it is precisely the existence of evil that sets men 
 on thus verbally proving its non-existence. In some forms 
 of the argument, as in that put by Shaftesbury in his 
 Characteristics^ we are told that evil is so only to us : from 
 the point of view of the Infinite it is not evil. So be it. 
 Then l.)y the very same reasoning good is non-ens : it is good 
 only to us ; from the point of view of the Infinite it is not 
 good. We are at the old futility of turning the explicitly 
 relative into an attribute of the All.
 
 Letter IX. LETTERS ON REASONING 167 
 
 You will agree with me, I think, that " the point of view 
 of the Infinite " is not a point at which the finite can place 
 itself; and that the verbalising semblance of doing so, 
 whatever gymnastic value the process may have, will certainly 
 not help morals, save insofar as it may help men to realise 
 the very truth we have been discussing, that conduct is a 
 form of cosmic causation. As we have seen, however, the 
 normal effect of supposing you are at the point of view of 
 the Infinite has not been to make men either sane or 
 merciful. Remember that all those disputes as to the 
 punishability of men by their Maker did not arise by way of 
 sequent reasoning from the concept of Infinity : they arose 
 as attempts to justify a barbaric conception of a "future 
 life " in which good people are to be ostentatiously rewarded 
 and bad people brutally punished, just as on earth. It is 
 because for ages men have set out on the path of philosophy 
 thus burdened with the credences of ancestral ignorance that 
 they have so constantly stumbled. The burden was one of 
 bad morals as well as of bad logic ; and it belongs to the 
 situation that the bad morals should help to keep the logic 
 bad. The more reason to feel assured that right reasoning, 
 when they attain to it, will infallibly make them better men. 
 To believe otherwise is to beUeve that insincerity promotes 
 virtue ; for to insincerity, as we have defined it, we always 
 come in our analysis of logical error. 
 
 Let me here avow once more, however, that the risk of 
 insincerity is not all on one side. What I mean by the 
 word, in the special force I have given it, will often be 
 chargeable on the rationalist — on me, on you. For even as 
 the supernaturalist is caught in the spell of a strong " will 
 to believe," which commits him to insincerities of argument, 
 so are we all apt to be held in the net of an instinctive 
 tendency that is rebel to our reason, and plays it false. No 
 instinct is more spontaneous than that of blame, of retribu- 
 tion : even those of us who instinctively recoil from such a 
 "revenge" as aims deliberately at injuring an enemy for
 
 i68 LETTERS ON REASONING Letter IX. 
 
 vengeance' sake are open to the satisfactions of severe 
 censure ; and few there be who scientifically adjust their 
 blame to the remedial or self-guarding purposes which ought 
 in reason to control it. No one, I think, can do so at all 
 times : we all vary in our capacity to be judicial ; and I 
 confess to finding the difficulty constant at this point. That 
 is to say, I find myself frequently lapsing into "insincerity" 
 as I have defined it, in that my reasoned conviction and 
 my act do not consist. 
 
 The inconsistency, as you will see, is not at all that 
 charged by those who deny to the determinist the right to 
 praise and blame at all : that charge we have seen to be 
 pure paralogism. It lies in the non-restriction of praise and 
 blame, but much more particularly blame, to our reasoned 
 perception of its fitness, of its utility. We are inconsistent, 
 insincere, not when we call the bad man bad, or the thief a 
 thief, or the liar a liar, still less (if possible) when we repent 
 of having ourselves done a cruel or a faithless act, but when 
 we merely vilify and humiliate the liar without seeking to 
 do good, or add insult and ill-usage and hard antipathy to 
 our imprisonment of the thief. It is when we merely gratify 
 our resentment, in an animal temper, that we are false to our 
 avowed recognition of the causedness of the resented action, 
 the determination of will by antecedents. If, when attacked 
 by a mad dog or any wild beast, I should do any more than 
 defend myself to the extent of killing the animal, if I should 
 seek to enjoy revenge by keeping it in a state of protracted 
 torture, I should prove myself more of an animal, so to 
 speak, than of a thinking man. It is the same wuth human 
 offences. If we resent them further than (i) to safeguard 
 ourselves, (2) to affect the offender for the better by making 
 him either afraid or sorry, or (3) to put others on their guard 
 against him, we are, as determinists, insincere : we are approxi- 
 mating to the sophisticated cruelties of the theologian. 
 
 Doubtless the keen resentment of iniquity and baseness 
 which has underlain much theological denunciation of sin
 
 Lctlcr IX. LETTERS ON REASONING 169 
 
 may be justified on the third of the grounds I have just 
 given : it stirs up the better feelings of the spectators, turns 
 what might have remained apathy into an antipathy that 
 involves sympathy with the right, and reveals to some dull 
 intelligences their past inconsistency. At times such an 
 utterance of blame, as it seems to me, is the most pressing 
 of public duties ; and when it involves much more unpopu- 
 larity than applause there is a certain presumption that, if 
 done with sanity and competence, it is not a mere gratifica- 
 tion of malice. But remember that where the theologian's 
 moral philosophy ends the rationalist's should as such begin ; 
 and that, if you do not rest on judgment where he rests on 
 instinct and fiction, you are in a fair way to be as insincere 
 as he. 
 
 When all is said, perhaps, your best advantage over him 
 will be in the power to detect and check your insincerities. 
 I do not mean that he does not do this : his literature is full 
 of avowals of backsliding and unworthiness. But these very 
 avowals he will instantly disavow when it is a question of 
 accrediting his creed and discrediting "unbelief" : it is his 
 doom to be for ever at the mercy of his false premisses. 
 Delivered from those premisses, ours should be the happier 
 as it is the clearer case. But part of its happiness will con- 
 sist, not in any sense of superiority (that is a poor stay for 
 normal comfort ; nay, a sure source of intellectual demorali- 
 sation, though it may be the supreme stay in a moral 
 struggle), but in the perpetual reminder of the critical 
 reason that in respect of the primary snare of insincerity we 
 are all kin ; that where we ourselves have reached, all may 
 one day be ; and that there is thus no limit in man's nature 
 to a perpetual betterment.
 
 LETTER X. 
 
 In reading the foregoing letters it may have occurred to 
 you at times that I seem to give a very low place to 
 "instinct," and, again, that I finally accept it as the ultimate 
 discriminator between good and evil, the determinant of 
 virtue and wickedness. I recognise that there is something 
 there to be cleared up. Unless we logically solve the 
 apparent contradiction, there will remain a possible source 
 of much confusion in argument on moral questions. 
 
 From the discussion on the so-called freedom of the will 
 we reached the conclusion that a general moral bias under- 
 lies all our moral judgments ; that people are in large 
 measure born predisposed to become good or bad ; and that 
 a varying degree of susceptibility to given motives constitutes 
 in large part the " determination " of our wills. When 
 we see people with a marked cast of benevolence or of 
 cruelty, of truthfulness or of the reverse, albeit similarly 
 educated and situated, we recognise this clearly enough. 
 But even in the discussion on these moral problems which 
 hinge on that of the causation of will, we could not but see 
 that moral bias is set up by doctrinal training ; and we are 
 reminded that it is also set up by mere associations, and 
 that a special set of temptations may further make all the 
 difference between the outbreak and the control of the worse 
 forms of "instinct" in many persons. Nay, we have proof 
 that certain kinds of physical shock or damage to the brain 
 may greatly alter a man's bias or balance, even turning a 
 good character into a bad one. All the while the " deter- 
 mination " of the will is clear ; but what exactly are we to 
 understand by " instinct " in these different aspects?
 
 Letter X. LETTERS ON REASONING 171 
 
 Let us clear the position point by point. The determinist 
 argument commits us to Hume's position (impHcit in those 
 of some theists before him, as Cumberland and Edwards), 
 that in every inquiry into the bases of morals we come 
 finally to a " sentiment," a feeling, a bias, which cannot 
 further be analysed.^ That sentiment or bias is a deter- 
 mination of our wills. If we like very much to help people, 
 why, we like it ; if we like very much to hurt them, there is 
 no getting away from that fact, whether or not we are 
 at the same time sufficiently cautious to keep clear of the 
 criminal law. But it is also a fact that strong moral volitions 
 or dispositions of feeling may be developed by a course of 
 training, and that the special bias so shaped is to the mind 
 that feels it indistinguishable at bottom from any other kind 
 of moral judgment. 
 
 We may say, indeed, that some " instincts " are primary, 
 and others superinduced. Thus all men hke to have their 
 own way ; and it would be an enormously difficult thing to 
 train anybody to enjoy being kicked, though some slaves 
 may perhaps be made tolerably callous to the sensation. 
 But whereas in a free country, as the phrase goes, a feeling 
 of indignation at the act of kicking an " inferior " may be 
 practically universal, the inhabitant of a free country may 
 after a short residence in Egypt or parts of South Africa 
 become quite complacent over such an act, whether done 
 by himself or by others. The well-bred Englishmen, coming 
 unexpectedly on a scene of servant-kicking for the first 
 time, will recoil from it, and the recoil will have to his own 
 mind all the aspects of " instinct " ; but the Englishman 
 who kicks is feeling and [acting just as instinctively. We 
 arrive at this generalisation, that we all resent or dislike that 
 act or judgment which sharply conflicts with our judgment, 
 but that our judgment is largely alterable by training and 
 circumstances. 
 
 In Mark Twain's admirable story, "Huckleberry Finn," 
 ^ Appendix L to Enquiry Concerning the Principles of Morals.
 
 172 LETTERS ON REASONING Letter X. 
 
 there is a capital picture of the state of mind of the boy 
 who feels that he is doing a " low-down " thing in helping a 
 negro slave, the property of a friend of his, to escape. A 
 boy brought up to think slavery an infamous wrong would 
 have a quite contrary feeling in the same case : he would 
 feel he was doing a praiseworthy act, and be jubilant. Each 
 boy would be exhibiting moral feeling and a fairly conscien- 
 tious cast of mind, the first no less than the second : nay, 
 the first boy might be at bottom the better of the two, might 
 be in general more trustworthy, more kind-hearted. Apart 
 from such a superiority, the difference between them is to 
 be recognised as one of moral education ; and the same 
 verdict holds good of the widely divergent sentiments of 
 multitudes of people respectively born in slave-holding and 
 " free " countries. But on the other hand there have grown 
 up in slave-holding communities a certain number of people 
 who revolted at slavery, and in free countries a number of 
 people who either positively sympathised with the slave- 
 holders or cared nothing about the matter ; such persons in 
 both cases coming to their conclusions in virtue of a primary 
 bias, not being led to their views by their teachers. Here 
 we note a specially marked determination of moral feeling 
 or of insensibility, a special cast of sympathy or of unimagi- 
 nativeness. To sum up, so far, some men strike us by their 
 bias, others, so to speak, by their biasableness— both 
 phenomena being of course forms of determination of the 
 will. Whether I grow up markedly disposed to gentleness or 
 to cruelty, or markedly open to persuasion by those about 
 me, either way I exhibit predisposition. 
 
 It begins to be clear, then, that we can make concerning 
 " instinct " what we have otherwise seen to be a rational or 
 logical proposition, that it may be either good or bad, best or 
 worst. We can broadly say with Kant that there is nothing 
 better than a good will, that is, a bias to good— by which, 
 of course, we do not mean that such a will may not be 
 bettered by increase of knowledge. Rather, we realise that
 
 Letter X. LETTERS ON REASONING 173 
 
 instinct is variously subject to modification by training and 
 by reflection on freshly seen truths ; and, in particular, that 
 our (instinctive) judgment as to what is good or bad in 
 instinct is modifiable by training and knowledge, though every 
 new moral argument in turn works down to a basis of feeling. 
 In other words, instinct is modifiable in terms of instinct, in 
 the sense that all reasoning is a complex of feelings, that 
 all feeling relates to knowledge, and that new knowledge 
 means new feeling. Always a man seeks that which he 
 likes or wants : in that primary sense we are all egoists. 
 But some of us like or want to help others to their satis- 
 factions ; and this sort of egoism differentiates sharply from 
 that which cares little for other people's happiness. To 
 that, accordingly, we restrict the name in its censorious 
 sense. And yet that narrow and self-regarding egoism inay 
 expand towards the large and altruistic, in virtue of the sheer 
 fact of desire for satisfactions. 
 
 Now for our ethical crux. If all moral judgments thus 
 root in feeling or " sentiment " or bias, it follows that there 
 may be irreducible conflicts of moral feeling. If, for 
 instance, a man is so constituted as positively to enjoy 
 torturing animals, his pleasure therein will make him 
 insusceptible to my protest that he is exhibiting a low 
 animal passion, unless indeed he happens to be still more 
 susceptible to the pain of blame ; and he may meet me 
 with my own proposition that a fundamental feeling is the 
 last standard of right and wrong. Similarly a theologian 
 may tell me that he feels that sin ought to be punished, and 
 that from this feeling there is no appeal. Now, it should 
 at once be admitted that a certain degree of strength of 
 bias is not alterable by argument, and that if an opponent 
 is not affected in his fundamental feeling by our argument, 
 the argument as against him has failed. What follows? 
 This : that the moral ideals and teachings struggle for 
 survival, and gain or lose adherents in virtue of their greater 
 or lesser attractiveness, or conformity to interest and bias.
 
 174 LETTERS ON REASONING Letter X. 
 
 A community will readily accept a moral code which 
 sanctions what most of the community want to do, unless 
 it thereby chances to bring into clear relief their inconsistency 
 by affirming something that they usually deny. In that case 
 many will disavow the code while acting on it. The whole 
 chance for good morals lies in this, that at bottom and in 
 the long run it is useful for the mass of men to do as they 
 would be done by. But inasmuch as it so often seems 
 useful to the individual or the community as such to do 
 otherwise, the fortunes of morality depend proximately on 
 the variations of moral bias. That is to say, a great gift for 
 consistency or for comprehension of the true interests of 
 society is a " variation " of the nature of genius ; and 
 though the men who constitute such moral variations in the 
 human species do not survive in virtue of their goodness, 
 they may so persuade a community as to secure by changed 
 action a marked increase in happiness, in which case their 
 superior ethic gains ground, and tends to become the 
 " instinctive " form of judgment of the average persons who 
 would otherwise have judged differently. 
 
 Thus the higher morality prospers simply as a variation 
 towards higher rationality, for it is only through appeals to 
 the instincts for truth, consistency, and betterment that the 
 average will is to be brought from a lower to a higher 
 choice. It is quite true that, while the theologian feels he 
 ought to make the lives of criminals miserable in prison, he 
 is impervious to the arguments which prove the contrary ; 
 and while any man feels only or mainly in terms of primary 
 egoism he cannot be made a good man (in terms of our 
 feeling) by any reasoned demonstration. But experience 
 shows that by degrees the better types of theologian can be 
 enlightened by knowledge and argument to the point of 
 feeling differently about punishment, present and future ; 
 and it is historically very certain that as regards the criminal 
 and primary egotistic types we are not a whit worse off as 
 determinists than our fathers were as theists. The failure of
 
 Letter X. LETTERS ON REASONING 175 
 
 supernaturalist ethic and practice to lessen crime and to 
 promote sympathy is written over the whole surface of 
 history : it is stupendous ; and so far as experience has yet 
 gone everything tends to show that a scientific handling of 
 moral problems will give immeasurably better results than 
 the theological ever did. 
 
 Broadly speaking, progress in morals consists in extension 
 of sympathy ; and though sympathy is not co-extensive with 
 knowledge, it is clear enough that in general extended 
 knowledge is the way to extended sympathy. Genius for 
 sympathy is rare : most of us depend for advance there on 
 culture and widened experience. And if it be true that to 
 do as we would be done by is to maximise well-being, then 
 the more ready people are to act on that principle the better 
 the world will be. 
 
 Here, of course, we take for granted that such betterment 
 '\s,felt to be a desirable thing. With anyone who does not 
 or cannot be led to feel it, there is no arguing : we can but 
 seek to discredit him, and, if need be, restrain his anti- 
 social acts. His is the egoism we are driven to regard as 
 evil : if it cannot be transmuted by the appeal to consistency 
 or by the experience of retaliation in kind, it is to be 
 reckoned with as an anti-social appetite. But a fresh 
 confusion is being set up in ethics by some who, reverting 
 to the test of total utility as between conflicting forms of 
 bias, wholly or partially repudiate on that score the rule of 
 doing as we would be done by. We are to consider, they 
 say, whether a given act will advantage society in the future : 
 if we can show that it will, it is thereby justified, even if in 
 achieving it we do to non-aggressors what we should greatly 
 resent having done to us. As commonly put, this argument 
 is limited to acts by the community as such : I have not 
 seen any professed student of morals openly argue that the 
 individual will do well to act on the same principle ; and 
 this restriction of the thesis is significant. If the principle 
 be sound, it surely ought to be laid down for all action.
 
 176 LETTERS ON REASONING Letter X. 
 
 That it is suggested only as a code for communities is 
 apparently due either to a lack of sincere belief in it on the 
 part of those who urge it, or to the perception that, while an 
 unrestricted form of the proposition would get a man into 
 trouble as encouraging crime, the restricted form will 
 keep him on good terms with a community that is 
 disposed collectively to do otherwise than as it would be 
 done by. 
 
 Now, this teaching is not sufficiently disposed of by 
 merely insisting that the law of reciprocity is primary and 
 unchallengeable, though we may hope that one day it will 
 be felt to be so. If our ethic is to be rational we must be 
 prepared to meet all reasoned opposition ; and if it be 
 argued that civilisation is to be bettered by our ceasing in 
 any set of cases to do as we would be done by, we must see 
 what the argument is worth. A common illustration of the 
 theory is the case of ancient slavery. It seems on retrospect 
 that the possession of slaves was a means of leisure to 
 studious men, and it is inferred that slavery thus made 
 possible a development of letters, arts, and sciences which 
 otherwise could not have taken place. Those developments, 
 in turn, being passed on to posterity, have greatly promoted 
 human happiness in ages which have been able to abolish 
 slavery. Ergo, it is urged, slavery is not always or necessarily 
 wrong. Here we have a very interesting problem in sociology, 
 and an important exercise in reasoning. 
 
 The problem, you will see, is twofold, and the proposition 
 before us recognises only one of its aspects. First we have 
 to ask, what constitutes a wrong ? next, what ivould have 
 happened in antiquity had there been no slavery ? On the 
 first head we say that a wrong act is in the first place a 
 departure from an agreed-on code, wrong-doing being a 
 relation between man and man. Now, all communities 
 have agreed among themselves to certain necessary restric- 
 tions or definitions of the law of reciprocity, notably the 
 provisions for dealing with theft, fraud, and violence. We
 
 Letter X. LETTERS ON REASONING i77 
 
 do not say : " If I stole, I should like to be let off; there- 
 fore I shall let off thieves " ; we agree rather to affirm that 
 thieving is an act we shall never commit, and consequently 
 one in respect of which we shall never desire reciprocity : 
 and as it is on the other hand seen to be an intolerable 
 aggression on society, we agree to " punish " or otherwise 
 resist it. But in many communities the systematic definition 
 of the law of reciprocity has gone much further than this ; 
 and in those of antiquity it went to the length of permitting 
 men (a) to sell themselves into slavery in payment of 
 gambling debts or for other considerations, {l>) to sell their 
 children, and (c) to enslave and sell captives and the children 
 of captives. 
 
 Now, it is not to be denied that, given these legal practices, 
 slavery could not anciently have figured as a " wrong " in 
 the sense in which it does for us to-day. Men's relation 
 towards it was rather such as our relation to (a) cases in 
 which men are ruined by paying gambling debts, (If) cases 
 in which the children of a ruined or dead man are left to 
 sink from comfort to poverty, and (c) the innumerable cases 
 in which men see others suffer from poverty and do nothing 
 to help them beyond giving small alms or supplying public 
 "relief." To most of us to-day this degree of apathy 
 seems natural and fitting, while the ancient apathy on 
 the subject of slavery seems strange or even revolting. I 
 hope, and I hold it likely, that one day our apathies will 
 seem to our posterity strange and revolting, and that it will 
 be held a matter of course that no one shall under any 
 circumstances be left to suffer physically from poverty ; that 
 mere poverty, in fact, will be socially made impossible ; and 
 that in particular the acceptance of a "gambling debt" 
 which can impoverish anyone will be regarded as an odious 
 action. On this view, you see, moral judgments are very 
 clearly " relative," varying from age to age ; and we may at 
 once agree that it is misleading to speak of ancient slavery 
 as if its maintenance stood for relatively the same degree of
 
 178 LETTERS ON REASONING Letter X. 
 
 moral apathy as the maintenance of slavery would imply 
 to-day. 
 
 Let us not forget, however, that the ancients were just as 
 much liable to " insincerity " as we ; and let us not hesitate 
 to suppose that Aristotle, for instance, was insincere when 
 he met as he did the protests of the more sympathetic men 
 of his time against slavery.^ The fact was that those men 
 represented a variation towards a higher moral instinct than 
 his, a larger sympathy, a greater capacity for the " good life" 
 at which he professed to aim. I at least cannot doubt that 
 he was, as we say, false to his lights, that he fell back on his 
 habits, his prejudice, and would not reduce his doctrine to 
 consistency. With his large and lucid intelligence, he could 
 see that the Platonists were inconsistent in justifying the 
 enslavement of captives and opposing other forms of slavery : 
 he on his part challenged the enslavement of captives, but 
 proceeded to theorise that some were slaves by "nature" in 
 respect of their need to be governed, their unfitness for self- 
 government.2 But as soon as he goes into detail he is forced 
 to acknowledge that such need and unfitness are universal, 
 merely varying in degree,^ so that his appeal to "nature " is 
 quashed ; and when he proceeds to treat the problem of 
 practical management he simply suppresses the ethical 
 challenge which his own admissions invite, taking it for 
 granted that ■■" the members of every well-regulated state 
 should be free from servile labour," but admitting that 
 nobody had yet hit on the proper way of managing slaves, 
 since these everywhere tend to be insubordinate.* That a 
 great political thinker could avow this, and yet not see the 
 explanation in the anti-social character of slavery, is to be 
 explained only by regarding him as standing wilfully to his 
 acquired bias, his social habit. 
 
 When this is realised, we have gone some way towards 
 solving the utilitarian problem : Had there not been slavery 
 
 ' Sec the Politics, i. 3, 5, 6. '^ Id., i. 2, 6. 
 
 "^ Id,, i. 13. ^ Id., ii. 9.
 
 Letter X. LETTERS ON REASONING 179 
 
 in the ancient world, how would civilisation have gone ? 
 In one aspect the question is quite idle, for slavery was the 
 outcome of the total ethical, political, and economic life ; 
 and to suppose a social evolution without slavery is to suppose 
 all these conditions different. But for this very reason, you 
 will see, it is idle to argue that slavery was a good thing 
 inasmuch as it promoted civilisation. To have done without 
 it, say in Greece, men in general must have been better, 
 juster, wiser, nobler ; and such nobler men could well have 
 evolved a higher civilisation than was actually produced. 
 They could have provided variously for the leisure of their 
 students, and so have escaped the pernicious idleness which 
 actually injured Greek city life ; and they might have made 
 their polity permanent. And inasmuch as some men strove 
 to persuade the others to renounce the evil thing, they were 
 as truly right, and their opponents as truly wrong, as men 
 ever are in any strife of ideals. 
 
 Slavery was as certainly a means of social corruption and 
 dissolution asit was a meanstothe existence of a leisured class : 
 Aristotle's admissions say as much ; and those who opposed 
 it were, as we can now see, on the line of social preservation. 
 This is the specifically sociological side of the case. They 
 were further more sympathetic, more self-critical, more sincere 
 in this regard than the champions of use and wont. This is 
 the specifically moral side of the case. To say, then, that 
 slavery was " not wrong " is to make the words right and 
 wrong meaningless. It may indeed be plausibly argued that 
 both terms are in such a case inapplicable ; that a wrong 
 act is that which is seen to be such in terms of the recognised 
 ethic of an age ; and that Greek slavery was not so regarded, 
 since even the slaves themselves, if freed, would readily 
 have enslaved others. But on this view the thesis that 
 ancient slavery was right is already disposed of; and for the 
 rest, to argue that it was " neither right nor wrong " because 
 most people did not see it to be wrong is to confuse the issue 
 as to what constitutes wrongness. \Ve all agree — all non- 
 
 o
 
 i8o LETTERS ON REASONING Letter X. 
 
 burglars, I mean — that objectively a burglary is wrong even 
 if the burglar holds that he is as justifiably employed as a 
 trader. His cast of mind is a reason for not seeking to 
 make him suffer more than is involved in protecting society 
 from him ; but we must all the same class his act as anti- 
 social. It is not by counting the number of people for and 
 against a given course that we can decide whether it is right 
 or wrong : a majority vote proves at most that the course 
 in question is permitted by the majority. But if we are 
 convinced that the majority in so deciding is inconsistent 
 and unjust ; that it is trampling down a minority's " rights " ; 
 that it is not doing as it would be done by, and is passion- 
 ately ignoring the fact, then we who are not of the majority 
 will rationally pronounce its action to be wrong. 
 
 Certainly it is not of much importance to insist that 
 ancient slavery was wrong when nobody is saying the 
 contrary ; but when some do, as aforesaid, say the contrary, 
 we are near to a vital moral issue. For the only practical 
 purpose of such an assertion is to induce us to do otherwise 
 than we would be done by in our own day, on the plea that 
 in so doing we may advance civilisation. This plea we have 
 now partly considered ; but we must consider it further. 
 The preliminary answer is that all such calculations are in 
 the first place grossly inconsistent, since they never admit 
 the notion that 2ve on the same score ought not to resent a 
 felt " wrong " done to us, and in the second place grossly 
 fallacious because they never take into account the reactions 
 of the course proposed. I might put the matter more briefly 
 by saying that such arguments are the negation of all 
 morality, since morality actually consists in a code held to be 
 reciprocally binding. This is clear enough to all, I suppose, 
 when an individual goes about to square his conduct on the 
 assumption either that his gain should be his sole moral 
 criterion, or that his gain must in the end be a gain to 
 civilisation In the social relation we positively must have 
 reciprocity if we are to have continuance. If a man lies
 
 Letter X. LETTERS ON REASONING i8i 
 
 to us or cheats or robs us, we shall not be induced to tolerate 
 his acts by any hypothesis on his part that his lying and 
 cheating furthers civilisation by helping him to live. 
 
 Eugene Aram, of famous memory, is said to have reasoned 
 that he might fitly commit a murder in order to turn to the 
 promotion of his useful studies money which its owner would 
 not have put to any good account. He was reasoning 
 exactly as do those theorists who say we may fitly overthrow 
 by violence a backward State in order to hasten the develop- 
 ment of its civilisation. Well, the rational answer to Eugene 
 Aram (apart from the fact of the danger he incurred) would 
 have been, firstly, that he could not know that his possession 
 of the money would benefit mankind : he could only guess 
 that it would ; secondly, that the same argument would, on 
 his own showing, entitle anybody to murder him ; and, 
 thirdly, that he had entirely overlooked the question of the 
 possible moral reaction of his act upon himself. A great 
 psychological novelist of our day, Dostoyevsky, has wonder- 
 fully imagined a case of the kind, in which a needy student, 
 mentally shaken by hardship, kills a greedy old woman, 
 thinking to rid the world of a worthless being and turn her 
 money to his own good purposes. He is, however, morally 
 paralysed by the unforeseen horror of his own act, and lives 
 in a mere trance of new suffering until he gives himself up 
 to the police. Now, it is not to be assumed that this is 
 what would happen in every case of a crime so motived ; 
 but the mere recognition of the possibility is enough to 
 withhold any sane man from such a deed. On the other 
 hand, it is obvious that no such proof can ever conceivably 
 be given of good results from private murder and robbery 
 as could induce a community to dream of applauding them. 
 All this, I say, is obvious enough : no one argues other- 
 wise. But, as I have remarked, it is not uncommon at 
 present to find men arguing that as between communities 
 the moral principle of reciprocity does not hold ; that a State 
 may be justified in international burglary and what we may
 
 i82 LETTERS ON REASONING Letter X. 
 
 term political garotting on a mere calculation of probable 
 gain to future humanity from its act. A given community, 
 they say,^ is backward, ignorant, and conservative : let us 
 conquer it, no matter at what cost of slaughter and devasta- 
 tion, and of cruelty to non-combatants, seeing that when 
 the bloodshed is over a new generation will reap an abundant 
 profit in improved conditions. We have here a reversion, 
 within certain limits, to an old species of doctrine, commonly 
 but dubiously fathered on the Jesuits as a body — the doctrine 
 that the end justifies the means. And some men give the 
 doctrine a superficially fresh aspect by identifying it with 
 Utilitarianism, 
 
 Now, there have been committed many oversights of 
 
 argument in the name of Utilitarianism, as in the 
 
 name of every other principle ; but no reasoning can 
 
 get away from the central facts that all moral systems 
 
 assume the ultimate coincidence of utility with right 
 
 action ; and that men in the long run are persuaded 
 
 to act on any code only in the belief that it is useful either 
 
 to themselves or to others. We must then be ready to meet 
 
 rationally any such attempt as that under notice, to prove 
 
 that a course repulsive to our developed notions of goodness 
 
 and righteousness is in reality useful to humanity. I have 
 
 personally a good deal of sympathy with " Intuitionists " 
 
 who resent and denounce such attempts to carry back all 
 
 acquired moral notions to a calculation of consequences. 
 
 The chaim and the profit of association with good people 
 
 lie largely in the very fact that as regards most things they 
 
 have evolved past such calculation ; that they never dream 
 
 of calculating whether it will " pay " them to be just, or 
 
 straightforward, or kind, or courageous ; and I do not think 
 
 that even one who falls short of that spontaneity of goodness 
 
 — unless he fall very far short of it indeed — ever admires 
 
 others of his kind, or fails to admire in his heart those who 
 
 ^ China is sometimes pointed to in my day. I wonder how that case 
 will have gone in yours.
 
 Letter X. LETTERS ON REASONING 183 
 
 attain it. A man who should either profess or seem lo 
 calculate in every case whether honesty is the best policy, 
 or whether he will gain by speaking the truth, would, I 
 fancy, establish in us a tolerably deep distrust. 
 
 The historical and psychological fact is that certain habits 
 of reciprocity have come to be normal and " instinctive " 
 with at least a large number of us, so that, though in the 
 case of certain lines of action we admit and apply the test 
 of utility, the "ethic of consequences," in respect of these 
 established habits we recoil from the suggestion of any resort 
 to such a test. Such questions as those of the fitness of 
 indissoluble or official marriage, the secularisation of Sunday, 
 the payment of certain taxes, and many other matters of 
 politics, must be tried by the test of utility, which, by the 
 way, is no simple matter. Those who affect to settle them 
 by appeals either to authority or to instinct are moralists to 
 whom we need pay little attention. It is otherwise with the 
 case of a person who invites us to join him in a safe fraud, 
 or in an act of physical or moral cruelty : there most of us 
 feel disgust that the utility of such courses should still be 
 thought to be arguable. And I confess to feeling some such 
 disgust when I am asked to believe that the members of 
 a community may usefully further civilisation by wantonly 
 doing to another community what they would passionately 
 resent having done to their own. When, however, the 
 intuitionist declares (if, indeed, he does so express himself: 
 at times he appears to be intuitionally on the side of the 
 pseudo-utilitarians) that he " feels " such an act is wrong, 
 and that nothing will induce him to be a party to it, he sets 
 us reflecting on the fact that men have as avowedly and as 
 obviously " felt " the rightness of many acts that to us now 
 seem hideous wrongs — such as the slaying of heretics, the 
 enslavement of captives and their children, the selling away 
 of slaves' wives and families, the torturing of witnesses, the 
 flogging of madmen, the special degradation of women for 
 acts in which they were partners with unpunished men, and
 
 i84 LETTERS ON REASONING Letter X. 
 
 a hundred others of the long list of instituted social 
 iniquities. This (to the study of which I shall return) is the 
 fatal weakness of professedly " intuitional " morality. While 
 one intuitionist may see that a given act is a breach of the 
 law of reciprocity, and as such may denounce it, another, 
 too prejudiced or self-interested to see as much, may " feel " 
 that the act is perfectly righteous. 
 
 No doubt the intuitionist (or, as he might be more fitly 
 termed, the Apriorist) may insist that the law of reciprocity, 
 the obligation to do as we would be done by, is absolute, 
 and may argue that this is not denied by the opposing 
 intuitionist, who claims that he is doing as he would be done 
 by. But we have noted that the law of reciprocity has to be 
 defined as between individuals ; and the process of definition 
 is inevitably an appeal to utilitarian tests. Let us not then, 
 I repeat, hesitate to meet any challenge to define it as 
 between communities. For my own part I am satisfied, 
 with Mr. Spencer, that the so-called instinct of reciprocity 
 as between individuals emerges most often as a one-sided 
 sense of our own rights, and that only by pressures of external 
 criticism and of the reflection which comes after experience 
 does it take a properly balanced shape. Mr. Spencer puts 
 it that the " sentiment " of justice comes before the " idea " 
 of justice : meaning by the former our sense of our own 
 rights, and by the latter our recognition of the correlation of 
 other people's. It would be better, I think, to say simply 
 that a self-regarding idea of sentiment of justice usually 
 precedes an other-regarding one ; but, terminology apart, 
 the proposition seems to me essentially true. And if it be 
 so, it need not be surprising to us that, in the relations of 
 communities, the idea or sentiment of other-regarding justice 
 should lag far behind the earlier and simpler notion. 
 
 In the case of individuals, I do not doubt, the full sense 
 of the force of the law of reciprocity is partly built up by 
 experiences and calculations of utility. I think I can see 
 this in the case of you two children from day to day ; and I
 
 Letter X. LETTERS ON REASONING 185 
 
 think I can in a measure trace the process even in my own 
 adult experience, to say nothing of my youthful recollections. 
 Most if not all of us tend to be inconsiderate in a hundred 
 ways ; and a sense of the charm and benefit of another's 
 considerateness, or the shock of another's want of it, in a 
 given case, is really an educative perception of the utility of 
 such considerateness. It is conceivable that even a perception 
 of the utility of honesty and candour may set up a choice 
 in favour of honesty and candour on the part of one not 
 strongly predisposed to them. If this were not so, human 
 prospects would be darker than they are. I should indeed 
 have no great hope of converting to my own inclination 
 anyone who argues that we ought to disavow or conceal 
 all unpopular opinions for the sake of our comfort or our 
 incomes. Such a reasoner I should take to be devoid of 
 the bias which involves {a) the finding of a satisfaction 
 ( = utility) in standing for an unpopular truth even unsuccess- 
 fully, and further involves {b) pain at the thought of having 
 turned one's back on it for ease or gain. But I can conceive 
 that even one without this bias may by early education, early 
 instruction as to the ultimate social utilities of straightfor- 
 wardness, be prepared to take the braver and better course, 
 even as one not naturally much endowed with considerate- 
 ness may be educated by others so endowed to act more 
 sympathetically than he otherwise would. It may not be 
 quite unprofitable, then, to prove that moral reciprocity is 
 the best policy as between nation and nation. 
 
 To begin with, much reasoning on the subject is false by 
 reason of the loose or illogical use of terms. The one word 
 " civilisation," in the hands of prejudiced or superficial 
 people, is made to play many tricks. Implicitly assuming 
 that civilisation is always better than incivilisation, they 
 reason that any process which sets up the material phenomena 
 of civilisation in a given place or race where before they 
 were lacking is an unquestionable gain to mankind. Applying 
 an ostensibly utilitarian test, they do not attempt any
 
 l86 LETTERS ON REASONING Letter X. 
 
 utilitarian analysis. They do not ask whether the change 
 adds to average happiness, or whether, supposing it to make 
 some people happier, it does not relatively impoverish and 
 degrade others. Yet it is almost a commonplace of practical 
 sociology that myriads of people in civilised cities exist in 
 worse life conditions than those of an average savage. 
 
 The same reasoners are apt to take it for granted that 
 civilisation by way of conquest is just as good a thing as 
 civilisation by way of peaceful contact and progress, and 
 that because in past history conquest has entered into the 
 development of most nations it either must or profitably may 
 be deliberately practised in future in many if not in all cases. 
 They never ask (any more than do the pro-slavery moralists 
 in regard to ancient slavery) how civilisation would or might 
 have gone in the past if men could have been educated to 
 abstain from conquest on principle. Now, a careful study 
 of history will reveal to you that all the civilisations of 
 antiquity underwent decadence ; that this decadence is 
 always associated with both the infliction and the suffering 
 of conquest ; and that it was precisely where civilisation was 
 most advanced^where men had become in the largest 
 proportion of cases capable of reasoning on the morality of 
 it — that the fatality is plainest. 
 
 At the stage at which men practise conquest most 
 instinctively, with least pretence of having philanthropic 
 reasons, with the most spontaneous belief that " he should 
 take who has the power " — then it is that conquest is most 
 often compatible with gain to civilisation. It is when, having 
 become capable of realising that conquest is the doing as 
 we would not be done by, they brazen out the act with 
 insincere Virgilian formulas, that it is seen, intelligibly 
 enough, to work their own swift demoralisation. That which 
 for primitive man is an unreflecting animal activity is for more 
 developed man a lawless act. Modern Europeans turned 
 pirates are much lower types than ancient Greeks who never 
 were anything else. So with nations which revert to the
 
 Letter X. LETTERS ON REASONING 187 
 
 spirit of tribal savagery. They have repudiated the normal 
 basis of their reasoned ethic ; and the result is domestic as 
 well as international. Give up the law of reciprocity as 
 between States, and you are already inclined to give it up as 
 between diverging groups within the State : cancel it as 
 between them, and you are prepared to annul it as between 
 individuals. And this actually happened in antiquity again 
 and again. Read the history of Rome from the conquest 
 of Carthage to the end of the Republic, and you will get 
 the most fully documented case. 
 
 Consider next this, that of all the so-called barbarian or 
 semi-civilised States effectually over-run by Rome, not one 
 was made capable of maintaining itself when the central 
 part of the empire gave way, and you will partly realise how 
 vain a thing is a forcibly imposed civilisation. But that, 
 after all, is a less impressive thought than the perception 
 that for any State to set up the pretension of being the 
 forcible civiliser of others is to set up in every other, in the 
 degree of its knowledge, the spirit not only of hostility, but 
 of a hostility which feels that towards such a threaten- 
 ing force it need not be scrupulous. The State which 
 collectively harbours the thesis that // may fitly decline to do 
 as it would be done by has justified every other State in 
 planning its overthrow. To justify us in supposing that any 
 State can thrive by avowing its readiness to do as it would 
 not be done by, we should need evidence that a community 
 any more than a man can permanently thrive by the ill-will 
 of neighbours. Was such evidence ever forthcoming ? 
 
 Were it not that I am seeking rather to make you reasoners 
 than to furnish you with historical demonstrations, I could 
 add much to similar effect ; but I have said enough to 
 indicate to you how thoughtless, how ill-reasoning, are many 
 of those who undertake to give rational grounds for deeds 
 which outrage the developed sense and the normal standard 
 of righteousness. They profess to estimate future utilities, 
 a thing that could be rationally done only after the closest
 
 i88 LETTERS ON REASONING Letter X. 
 
 study of past utilities ; and they make no calculation what- 
 ever, falling back on verbal formulas which beg the question. 
 They are, in fact, playing fast and loose with reason while 
 professing to stand or fall by its light. And to realise this is 
 in a measure to realise that you need not be afraid to let 
 rational tests be put to any of your moral convictions. 
 Observe, you may happen to have moral convictions which 
 will not stand rational tests. If you had been brought up, 
 for instance, to believe in Sabbath observance, and had been 
 challenged to explain rationally why one should object to 
 play any game on a .Sunday, you could not do so : you must 
 fall back on dogma or admit that you held to mere habit. 
 If, again, you were trained to think your own country must 
 have been in the right in every one of its wars, and were 
 searchingly questioned by one who knew better, your 
 " feeling " would not save you from argumentative discom- 
 fiture. And the only way to make sure that any of your 
 moral positions is not thus arbitrary is just to test it when it 
 is challenged. But it would be disturbing to feel that the 
 " instincts " on which you speak the truth to your fellows 
 and render good for good might in some particular instance 
 be open to repudiation ; and I advise you to reach moral 
 security once for all by testing every pretence to that effect. 
 I venture to promise you that you will find them as unsatis- 
 factory to the head as to the heart. 
 
 There is indeed one stock case of opportunist ethics 
 which seems to give some people trouble. By the time you 
 read these letters you will probably have heard of the problem 
 whether we can ever be justified in telling an untruth. 
 Some argue that it is always unjustifiable : many, on the 
 contrary, are satisfied that if by telling an untruth to a would- 
 be murderer — say in the case of his asking us which way a 
 fugitive had gone— we can prevent a crime, we ought to tell 
 it ; and that if by deceiving a sick person we can save a 
 life, or if by withholding a horrible truth in some special 
 circumstances we can save some innocent person intense
 
 Letter X. LETTERS ON REASONING 189 
 
 moral suffering, we ought so to suppress the truth. Most 
 reasonuig men, I think, take the latter view ; but the 
 philosopher Kant insisted on the former. He argued that 
 if you see a man with a weapon running after another who 
 has passed near you, and the pursuer should ask you which 
 way the fugitive went, you are morally bound either to refuse 
 to tell him anything, or to tell him the truth. If a murder 
 should result, says Kant, you are then not responsible ; 
 whereas if you should chance, by telling an untruth, 
 unexpectedly to enable the murderer to overtake the fugitive 
 in the case of the latter having changed his course after he 
 has passed you, you would be guilty of facilitating the 
 murder. 
 
 Let us argue the case out. We may simplify the problem 
 at once by supposing that the pursuer asks us such a cjuestion 
 (as " Did a man run that way ? ") that if we refuse to answer 
 it he will readily infer the truth. And we may considerably 
 strengthen Kant's rather flimsy case by raising the hypothesis 
 that for all we know the flying man may be a dangerous 
 criminal or madman, and that the pursuer is carrying a 
 weapon not for slaughter but for self-defence. How is such 
 a problem to be solved ? 
 
 I answer, by utilitarian tests, and by them only. Let us 
 first put a somewhat different case — that of an obvious mad- 
 man who is threatening our life. There is a story that a 
 madman who had got to the top of London Monument said 
 to a sane man whom he found there : " I am able to throw 
 you down to the street, and I am going to do it " ; where- 
 upon the sane man coolly replied : " Oh, that is nothing : I 
 could throw you up here from below, which is a far harder 
 thing." The madman, in the story, consented to go down 
 to have the experiment tried, whereupon he was secured by 
 the police. Now, I fancy that even Kant would in that case 
 condone the untruth. If he refused, I should simply waive 
 further argument, taking as my ultimate position the primary 
 right of self-defence, and affirming that my right to deceive
 
 I90 LETTERS ON REASONING Letter X. 
 
 the madman in self-defence is as clear as my right to resist 
 him if he tries to kill me. 
 
 Kant, I suppose, would have assented in this case on the 
 score that a madman is as such incapable of normal moral 
 relations ; that he must be regarded as for the time being in 
 the relation of a wild animal to us. Well, exactly on that 
 ground I should justify the act of deceiving the man 
 apparently bent on murder. In the case put, we might and 
 indeed should in reason question the questioner, in order to 
 gather if possible whether he was or was not bent on murder ; 
 and if we were satisfied that he was, it would be our clear social 
 duty (if we recognised any such duty) to deceive in order to 
 baffle him, as a person making war on the social system. I 
 suppose Kant would admit our right to resist the would be 
 slayer if in our presence he sought to commit murder, 
 though in so resisting him we might do him severe bodily 
 injury. To say that we are justified in breaking his head, 
 but not in deceiving him, seems to me a rather gross 
 absurdity. If, on the other hand, our attempt to deceive him 
 should by mischance really lead him in the path the fugitive 
 had actually taken, and so facilitate his crime, we should 
 indeed be bitterly grieved ; but as we had done our very 
 best, in our view of the circumstances, to prevent it, we 
 should certainly not be guilty. If, on the contrary, we had 
 facilitated the act either by telling the truth or by maintaining 
 a silence from which the murderer inferred the truth, we 
 might well reproach ourselves bitterly, inasmuch as we had 
 not striven to prevent the crime, and had sacrificed good to 
 the form of good, the spirit to the letter. Thus to separate 
 the formula of right from all consideration of moral conse- 
 quences is to put the very essentials of morals in jeopardy 
 by leaving it without rational sanctions. 
 
 Kant, perhaps, would argue further that by telling an 
 untruth in any one case we weaken our moral nature. I 
 sliould answer that we certainly do so if we transgress the 
 normal law of reciprocity and deceive for mere gain or
 
 Letter X. LETTERS ON REASONING 191 
 
 personal convenience ; but that in the case put we no more 
 vitiate our sense of truth than we brutaUse ourselves when 
 in an emergency we use violence to save a friend from worse 
 violence. If, for instance, I saw either of you attacked by 
 a vicious dog, and in order to save you beat or even killed 
 the dog, 1 should be at least suffering less moral deterioration 
 than I should undergo by standing by and doing nothing ; 
 and it really need not follow that I should develop a proclivity 
 to killing dogs. 
 
 So far, on the other hand, is Kant's moral fanaticism from 
 safeguarding the moral sense that we actually find him, the 
 absolutist, the professed ascetic of veracity, grossly trans- 
 gressing his own law where there is hardly a shadow of 
 excuse for it. It might seem incredible, but it is the fact, 
 that he who pronounced it unjustifiable to deceive a would-be 
 murderer in order to save a life, expressly condoned the 
 utterance by a priest of a religious doctrine in which he does 
 not believe, —this on the score that " populus vult decipi," 
 and that the priest cannot help himself.^ I know no stranger 
 anomaly in the Uterature of ethics ; and I do not see how we 
 can escape the conclusion that Kant was in this connection 
 profoundly " insincere." He seems to have framed his 
 argument against deceiving the armed pursuer in order to 
 buttress a formula he liad already espoused, knowing that 
 there was small risk of having his doctrine put to the test 
 in his daily walk in the avenue of Konigsberg. But where 
 his own convenience, or that of many of his disciples, was 
 seriously imperilled by the command to speak the truth and 
 shame the devil, he virtually gave them and himself carte 
 blatiche where you and I, I hope, would refuse to accept it. 
 
 Let us take that case in turn on its merits. It is the fate 
 of many men to be committed to an ecclesiastical career in 
 youth, before they can rightly judge of the truth of the creed 
 
 ' Religion innerhalb der Grenzen der blossen Veniwift, B. II L 
 Apotome i. sect. 6; B. IV. Apot. ii. , preamble and sect. i. 3, 4. Cp. 
 the cuiiimenl of Baur, Kiichengcschichte des igten Jahrh., 1862, p. 65.
 
 192 LETTERS ON REASONING Letter X. 
 
 they are taught, and to find themselves in later years pain- 
 fully convinced of its untruth, yet unable to abandon their 
 clerical vocation without putting in danger of extreme hard- 
 ship those who are dearest to them, and for whose well-being 
 they are responsible. For men so placed I have nothing 
 but sympathy ; and so long as they have the sincerity to 
 abstain from maligning or belittling those who are free to 
 and do avow heretical opinions, I should never dream of 
 disparaging them. Economic pressure in our ill-framed 
 society is too ubiquitous and too powerful a force to be 
 fairly regarded as morally negligible. My honoured friend, 
 the late Charles Bradlaugh, was in the course of his life 
 privately consulted by quite a number of clergymen so 
 placed, and his habit was to urge on the single men that 
 they should take their fortunes in their hands and give up 
 their posts at any risk, but to tell the married ones that their 
 first duty lay in providing for their wives and children, and 
 to warn them that it might be very difficult to do so if they cut 
 adrift from their office and sought to earn a living as simple 
 teachers of the truth. 
 
 Thus did a wise utilitarian reason, who in his own case 
 had always made truth the first consideration, and had 
 throughout life unflinchingly faced the endless penalties 
 incurred by those who steadfastly impugn popular error. 
 But the transcendentalist, the preacher of uncompromising 
 veracity, is found not merely commuting his commandment 
 on the slightest pressure of his personal interest, but virtually 
 encouraging the disillusioned priest to persist in make-believe 
 when he might without injury to others or even to himself 
 renounce his false position and take one where he could 
 preach truth with no other hindrance than that which conies 
 of popular prejudice. 
 
 It is not now necessary, I hope, to remind you that by the 
 utilitarian test Kant's counsel is as indefensible in this case 
 as his veto was in the other. The deliberate resort to make- 
 believe in religion does work moral evil in a hundred ways.
 
 Letter X. LETTERS ON REASONING 193 
 
 Men often guess, indeed, that evil-doers are to be restrained 
 by threatening them with hell-fire and bribing them with the 
 proffer of vicarious salvation ; when they might ascertain by 
 inquiry that the vast majority of criminals have always been 
 believers in such religious doctrines, and that on the other 
 hand the deliberate and calculated practice of deception, 
 under no coercion of a grave need, does corrupt the men 
 who resort to it. But such are the incongruities of human 
 nature that you are likely to meet in your lives with many 
 moralists who on the one hand profess to repudiate the 
 morality of consequences, and on the other justify by 
 utilitarian pleas courses which you would be ashamed to 
 take. 
 
 I have heard such a moralist, a man of culture and 
 apparent moral earnestness, denounce as demoralising the 
 " naturalistic " view of human actions, and in the same 
 discourse bitterly reproach rationalists for saying that we 
 should always seek and teach the truth for its own sake. 
 " Would you," he asked, " insist on telling an innocent child 
 of the wickedness of his mother? " That is exactly what a 
 moral-minded utilitarian would not do ; and the pretence 
 that the demand for true leaching about religion is on a 
 par with such an act is itself an act of sheer moral baseness. 
 By "teach the truth" we mean "let that which you teach 
 be truth," and it does not even imply that a true doctrine is 
 to be forced on people who do not want to listen to it. To 
 such basenesses of misrepresentation, however, men 
 frequently resort in their determination to defend their 
 beliefs anyhow ; and I warn you on the one hand to be 
 prepared for such basenesses in your intercourse with men, 
 and on the other to be anxiously on your guard lest you 
 should ever descend to similar courses. Let your loyalty to 
 your ostensible " intuitions " be such as to preserve you 
 from the unworthy devices of dialectic to which professed 
 intuitionists so often descend, and which are so poorly 
 justified by what Mr. Spencer calls " the profoundest of all
 
 194 • LETTERS ON REASONING Letter X. 
 
 infidelity — the fear lest the truth should be bad " — bad, 
 that is, for mankind in mass and in perpetuity. 
 
 I have now, I hope, made clear to you how it is that on 
 one hand I hold good instincts to be the root of right action, 
 and the complete lack of them to be incurable by any sort 
 of moral doctrine, though a moderate endowment of them 
 may be much improved upon by rational appeals ; while on 
 the other hand I repeat that a reference to moral instinct as 
 a standard independent of all utilitarian tests is futile from 
 the point of view of moral persuasion, and is as readily 
 resorted to in a bad cause as in a good. In point of fact, 
 almost none of those who profess to hold by such a standard 
 do really adhere to it. I referred above to the late Dr. 
 Martineau's declaration that a substitution of a utilitarian for 
 an intuitionist ethic makes a complete change in our moral 
 dynamics. This, you will see on reflection, is really a 
 utilitarian argument, an appeal to the test of consequences 
 in the very act of repudiating such a test. " If you become 
 a utilitarian," says Dr. Martineau in effect, " you will become 
 a worse man." I have tried to show you how what is really 
 moral in Dr. Martineau's recoil from rationalistic methods is 
 perfectly preserved when you take the step he is afraid to 
 take. He was somewhat in the position of the Paduan 
 professor described by Galileo, who resolutely refused to 
 look through Galileo's telescope at those satellites of Jupiter 
 in which he declined to believe. Your loyalty to the evolved 
 higher " instincts" is not, I think, impaired by the demon- 
 stration that they really make for the welfare of mankind ; 
 and the recognition of this fact will one day, I think, make 
 an end of the strife between so-called intuitionalism and 
 utilitarianism. 
 
 We have still to deal, however, with some forms of bad 
 reasoning resorted to by well-meaning men who are at Dr. 
 Martineau's point of view ; and with the positions of one 
 such reasoner I shall deal in my next letter.
 
 LETTER XI. 
 
 When you read these letters, I hope, the dispute about 
 what is called " hedonism " will have got past its present 
 stage ; but I fear the anti-hedonists will still be making 
 play with their unhappy distinction between wanting things 
 and liking them. When men began to reduce Ethics to 
 reason after the Reformation, they made the mistake of 
 laying the stress of the distinction between naturalist 
 and supernaturalist ethics on the fact that all moral codes 
 express preferences, and that the end of all action is pleasure 
 or happiness. The proposition, as meant, is quite true ; it 
 is, in fact, when understood, a truism ; but this circumstance 
 enabled some critics to disparage it as such, and others to 
 misrepresent it on the score that it could not have been so 
 intended, and must be otherwise construed. 
 
 Thus Locke, for instance, contemptuously remarked that 
 to say we always seek our pleasure is merely to say that we 
 want what we want ; and thinkers of another type proceeded 
 to construe " pleasure " as meaning physical gratification. 
 Pope, certainly not a very good man, could successfully 
 asperse Epicurus, who was, by such a line as 
 
 " The fattest hog in Epicurus' sty " ; 
 
 and, as you know, the term Epicurean has come to mean for 
 the average man a mere lover of good eating, which the 
 average man himself often is, and Epicurus and the first 
 Epicureans certainly were not. And the trouble persists. 
 Carlyle met the doctrine that all action aims at happiness 
 by saying that the good man seeks " not happiness, but 
 blessedness," which one would be inclined to regard as an 
 idle quibble if one did not surmise it to be an expression 
 
 p
 
 196 LETTERS ON REASONING Letter XL 
 
 of the average intuitionist's real inability to realise that 
 intuitions differ. Carlyle simply meant that what was 
 happiness for some people was not happiness for him ; and 
 because they called their blessedness happiness he insisted 
 on calling his happiness blessedness. That their sincere 
 happiness on what he felt to be a low plane must have been 
 on his own theistic principles as much " God-implanted " as 
 his own ideals, he was not intellectually sincere enough to 
 acknowledge ; and hi^ habit of " taking his pleasures sadly " 
 further obscured the case for him. 
 
 If, however, the intuitionists have not been either thought- 
 ful or candid, the Naturalists have not been exactly profound, 
 or even sagacious on the side of tactics. They ought, I 
 think, to have taken up Locke's position that " hedonism " 
 is a truism ; that the professed intuitionist strives as truly for 
 what will satisfy him as does the professed Epicurean or 
 utilitarian ; and that the moral problem is to know what we 
 do well to find our pleasure in ; whereas the adoption of the 
 name " hedonism " gives the verbalising intuitionist his 
 chance by seeming to stand for the doctrine that we ought io 
 seek our pleasure, and that the hedonist is opposing a self- 
 sacrificing or austere school who say the contrary. For a 
 moralist to call himself a " pleasurist " ^ is to give a some- 
 what gratuitous opening to an enemy who has never been 
 zealous to speak well of those who gainsay him. 
 
 You will have noticed that there are three battlegrounds 
 between rational and dogmatic or theological morals : the first 
 being the question of " free-will " ; the second, that of the utili- 
 tarian test of conduct ; the third, that of the aim or end of 
 right conduct ; and in every stage of the struggle the super- 
 naturalist who is on the defensive imputes vicious propensi- 
 ties to his opponent, though as a matter of fact there are 
 always some supernaturalists opposed to him on the 
 philosophic issue. There have been, as we saw, theological 
 
 * In the New Testament ihe word rjhovi) is always used as signifying 
 sensuous and ignoble pleasures.
 
 Letter XI. LETTERS ON REASONING 197 
 
 determinists ; there have been professed theological utili- 
 tarians, to say nothing of the fact that the ruck of theologians 
 have always stressed the doctrines of heaven and hell, 
 which somewhat directly address our self-interest ; and finally 
 there are theologians who are content to avow that blessed- 
 ness is happiness, and that all men seek it. But all this 
 will not deter the other theologians from speaking evilly of 
 determinism, utilitarianism, and hedonism. 
 
 What is more, the formula " We always seek our pleasure " 
 is not strictly accurate when tried by the tests which the 
 theologian applies to all formulas but his own. As many men 
 and many utilitarians have seen and said, pleasure or happi- 
 ness is not always, or even normally, attained by con- 
 sciously seeking for pleasure or happiness regarded as such : 
 on the contrary, it is an ancient commonplace that mere 
 " pleasure-seekers " do not get most pleasure in life, dis- 
 counting pleasure as they do by dwelling on the thought of 
 it. Happiness is often attained as a result {a) of gratifica- 
 tion of a desire which did not take pleasure conceptually 
 into account ; {b) of choosing what may be loosely termed the 
 nobler of two pains ; {c) of doing what is felt to be an irksome 
 or hard duty ; and so on. All this, observe, is known and 
 admitted by the " hedonist " ; only his name for his doctrine 
 does not suggest as much;' and the su[)ernaturalist, who is apt 
 to be partisan first and truth-seeker afterwards, takes his cue 
 accordingly, and professes to draw a vital distinction where 
 there is at bottom no difference. The dispute is essentially 
 a verbal one, partly over the signification of " pleasure," 
 and partly over the " aim " and " end " of action. (Irant that 
 "pleasure" includes the act of seeking to fulfil a desire, 
 such act being a refusal of the pain of feeling it without 
 
 ' J. S. Mill, after making tlie admission in his Utilitarianism, goes 
 
 on to say that "To think of an o!)ject as desirable and to think 
 
 of it as pleasant are one and the same t/iini;." This is inaccurate, and 
 lays him open to Green's criticism. Tlie true solution is noted below. 
 Unfortunately, Mill's handling of Utilitarianism is flawed by more 
 serious fallacies. 
 
 P 2
 
 198 LETTERS ON REASONING Letter XL 
 
 effort to gratify it — and the dispute is at an end. And it is 
 quite reasonable thus to recognise a negative as well as a 
 positive pleasure, in respect that both are antitheses to pain. 
 Clearly the "nobler of two pains " is relatively agreeable, and 
 is for that cause chosen by the one who so regards it ; and 
 the irksome duty is accepted by us because we should feel 
 more distressed by the consciousness of having evaded it. 
 To say this is not to fall into J. S. Mill's mistake of saying 
 that "pleasures" differ in nature or quality as well as in 
 quantity. There is difference in the nature or quality of 
 experie?ices which alike give pleasure ; but when we say " the 
 nobler pleasure " or " the nobler pain " we are using a meta- 
 phor or trope, and should always remember that the psycho- 
 logical fact in view is " greater pleasure " and "less pain," 
 the adjective " nobler " being properly applicable to the 
 course of action or endurance which yields pain or pleasure. 
 It is always degree or stress of satisfaction that is ultimately 
 looked to. And when common-sense lays down the maxim, 
 " Have an extraneous interest or interests in life : live for 
 others as well as for yourself; thereby you will better attain 
 happiness than by always thinking of getting happiness 
 directly," it obviously enough puts happiness as the end of 
 rational action while warning us against making it the 
 immediate ah>i. 
 
 Ethical teachers who have been theologically educated, 
 however, find it very hard to break the habit of thinking of 
 man's end in terms of the " will of God," or of thought about 
 " God " ; and even when they have realised that " the will of 
 God " can mean for us only {a) what other people (ancient 
 Semites or others) have thought best, or {b) what we think 
 best, they still recoil from admitting that even a change of 
 terms would identify them with " hedonism." One such 
 teacher, the late Professor T. H. Green, protested that 
 hedonistic doctrines "offend the unsophisticated con- 
 science." * I shall revert to that test by-and-by in examin- 
 
 * ProUgommia tj Ethics, 3rd. cd., p. 164, § 157.
 
 Letter XI. LETTERS ON REASONING 199 
 
 ing Professor Green's own moral practice ; but for the 
 present let us note, first, that the unsophisticated conscience, 
 if it will only be candid, can easily be satisfied as to the 
 rectitude of reasoned hedonism so-called ; and, secondly, 
 that when hedonism is renamed and verbally corrected as 
 far as necessary, there can remain no excuse for any theo- 
 logical pretence either that the standard of right lies outside 
 our intelligence, or that it is independent of experience and 
 the calculation of consequences. 
 
 As I have said, the practical problem of ethics is to dis- 
 cover what conduct is really best fitted to yield happiness ; 
 and the first truth the inquirer reaches is that men's reasoned 
 convictions, as well as their primary tastes, differ indefi- 
 nitely. Now, what is best for free men is to be conceived 
 in terms at once of their desires and of their capacities ; and 
 if we feel freedom to be essential to our own happiness, we 
 clearly ought in consistency to desire the freedom of all men 
 in so far as they do not commit aggression on their fellows. 
 And as our happiness is so largely determined by our 
 desires, our aspirations, our conceptions of what is just and 
 good and beautiful, it follows that their different ideals and 
 aspirations will yield them a different view of what is the 
 best conduct. But, again, we all depend indefinitely on 
 each other for the realisation of our ideals, even if one of 
 our ideals should be that of being left alone ; and so far as 
 we wish to have the social or political conditions of life 
 differently arranged, it is our business as moralists to per- 
 suade our fellows, if we can, to desire what we desire, as it 
 is theirs to persuade us. All such persuasion necessarily 
 accepts as the end of all action the good of all (including, 
 with the specific restriction proper to their case, even that 
 of the men to whom, as violators of our rights, we feel 
 bound to refuse freedom). If it does not do this, it is a 
 mere conspiracy on the part of a few to seek their good in 
 disregard of that of others. And a moralist who addresses 
 his fellows in general, without limitation of class, sex, sect.
 
 200 LETTERS ON REASONING Letter XL 
 
 party, or nation, is of course raised at once above any such 
 conspiracy. Under such conditions, that is to say, our 
 "egoism." coincides with our "altruism": our argument 
 necessarily is, " such a course is good for all of us." Here 
 we may settle once for all the question as to the nature of 
 egoism. " The egoist " seeks his own pleasure ; but a good 
 man, whom we call "unselfish," finds his pleasure in pro- 
 moting the happiness of others. Thus selfishness is properly 
 to be defined not as seeking our own pleasure, but as failing 
 to find it in the fashion aforesaid. 
 
 Now, this principle or test not only may be, but usually is, 
 avowedly accepted by teachers proposing opposite lines of 
 action, and by supernaturalists as well as naturalists ; and as 
 this principle is exactly what is meant by " universalistic 
 hedonism," there is so far no true theoretic dissidence ; and 
 the disputes which actually go on about the bases of ethics 
 are, as we have seen, largely verbal. As for the men who 
 do not consent or profess to seek the good of all, they do 
 not as a rule trouble us with theories of morals ; but in so 
 far as they do we can by argument convict them of having 
 on their own showing no appeal to make to the rest of us, 
 and so either persuade them of the futility of their pro- 
 paganda, or discredit them with readers not decisively 
 biassed on their own account to a disregard of the good of 
 others. In point of fact, even theories which disparage the 
 principle of " altruism " usually do so on the pretext that it 
 is injuring the race, and that a disregard of it will in the end 
 produce a race of happier people. 
 
 This last, then, is simply one of the innumerable modes 
 of dissidence as to " the good." Facing these, as I have 
 before suggested, a rationalist will avow that moral bias 
 varies as the physical type varies, and will confess that the 
 moral world of man is a scene of struggle of types of ideal 
 for survival. There is no getting away from the fact that 
 each man thinks his own aspiration the best, even if he con- 
 fessedly aspires for the few as against the many, or for
 
 Letter XI. LETTERS ON REASONING 20l 
 
 physical goods rather than mental ; and in stressing this 
 truth hedonism brings out a cosmic fact which present-day 
 theology tends to obscure, though the theology of yesterday 
 expressly insisted on the bias to sin as a primary fact in 
 human nature. The theologian or theist seems to shrink 
 from the avowal that his " higher " ideal is simply his taste, 
 his proclivity, his bent, as another man's " low ideal " is his. 
 Green so hesitated, though for the purpose of his anti- 
 hedonistic propaganda he actually argued laboriously that 
 an end is pleasurable to us because we desire it, and that 
 we do not desire it because it is pleasurable. The admission 
 in the first clause is decisive as to the variety of moral bias 
 in men, and its ultimate character in all ; but what Green 
 was thinking of in particular was the proposition of the 
 second clause. You will readily see, I think, that it is 
 finally beside the case, inasmuch as we never desire 
 anything that we think would not be either directly pleasur- 
 able or ultimately pleasure-producing. By Green's own 
 admission, " We cannot think of an object as good — />., 
 such as will satisfy desire — without thinking of it as in 
 consequence such as will yield pleasure." This actually 
 yields the whole case, even with exaggeration ; ^ and it does 
 not alter the case to say further that while with hedonists 
 the good generically is the pleasant, the true view of the 
 good is that it satisfies some desire. What is evil to my 
 judgment is good to another's, and satisfies his desire. 
 Green seems to have felt that, as it was his duty to teach 
 young men to choose high ideals rather than low, he 
 did well thus to impress upon them that ideals are to be 
 determined by argument. I have already urged on you 
 that in a measure they may be, but I have urged you also to 
 
 ^ Prolegomena to Ethics, 3rd ed., p. 178, § 1 7 1. Green actually 
 makes the very slip for which he had criticised Mill. We cati think of 
 an object as being what we want without thinking of it as in con- 
 sequence pleasure-yielding. What he should have said of it is that we 
 cannot think of it as not pleasure-yielding.
 
 202 LETTERS ON REASONING Letter XL 
 
 recognise that where another man's ideals are intractable 
 to yours, he is as much as you a part of the cosmos, and 
 that it is pure delusion to suppose your ideals are part of a 
 "divine plan," and his outside it. 
 
 Green exemplified the weakness of all theism, and of most 
 pantheism, in that he insisted on giving his aspirations a sort 
 of divine cachet as against others. That they were for the 
 most part very good aspirations I gladly admit : he seems to 
 me to have been a notably good man ; but he cannot have 
 been more sure of their representing " God's will " than 
 many other men have been that theirs did so when they 
 were working what I regard as wickedness. It is nothing 
 short of startling, at times, to realise how theists and 
 pantheists can stultify their own philosophic doctrine by 
 their ethic. Thus Fichte, a thinker of great capacity, used 
 to urge on his colleagues in the university board, as a 
 reason for accepting his policy on a given point of adminis- 
 tration, that he felt the All was speaking through his mouth, 
 when on his own showing the All was equally speaking 
 through the mouths of his opponents. Green, less naively, 
 exhibited the same infirmity. He argued diffusely about the 
 " Spiritual Principle " of the universe, by way of finding 
 justifications in that formula for his moral bias, which was 
 what he meant by the spiritual principle. It was only a 
 grandiose way of saying, with a gratuitous fallacy appended, 
 what Hume said simply — that a sentiment, a bias, is the 
 starting point of all our moral judgments. The appended 
 fallacy is the division of the forces of the moral world into 
 spiritual and non-spiritual, divine and non-divine. Philoso- 
 phically speaking, wickedness is just as cosmic as goodness. 
 To such verbalism do men come when they will not consent 
 to avow that the test of right action is its effect on human 
 well-being, and that the judge of human well-being is just 
 humanity, or each for himself. 
 
 Swerving thus. Green actually laid down the proposition 
 that " it is not an illogical procedure, because it is the only
 
 Letter XI. LETTERS ON REASONING 203 
 
 procedure suited to the matter in hand, to say that the 
 goodness of man lies in devotion to the ideal of humanity, 
 and then that the ideal of humanity consists in the goodness 
 of man." ^ In other words, right action consists in devotion 
 to right action. Had any philosopher opposed to him ever 
 ventured thus to trifle with the intelligence of his readers, 
 Green, I fancy, would have expressed himself on the subject 
 rather more severely than he ever had occasion to do in his 
 actual criticism of other writers. He is here acting as if he 
 held that logical consistency is a thing to be sternly exacted 
 of the Leweses and Spencers, who are not theists, but is not 
 always incumbent on the man who is doing the will of God. 
 His very words, " not an illogical procedure because it is the 
 only procedure suited to the matter in hand," show that he 
 knew he was playing fast and loose - with logic, and held 
 he was entitled to do so by his special inner light. The 
 " matter in hand " was not rational argument, but the 
 summing up of moral philosophy in a formula which should 
 not admit that the final criterion of right action is human 
 happiness, but should affirm a " spiritual principle " indepen- 
 dent of all action or sensation as such. The didactic result 
 is either pure verbiage — the implicit proposition that 
 " devotion to goodness lies in devotion to goodness " — or the 
 implication : " we must not conceive of goodness as the 
 caring about our own or other men's physical or even 
 mental goods as utilitarians do " — this after a line of 
 argument which all along implicitly admitted that what we 
 are seeking is a guide to conduct. I leave it to your "^ unso- 
 phisticated conscience " to say how this procedure strikes it. 
 Apparently he had felt it necessary to justify Kant's 
 phrase, " the categorical imperative " — a formula of which 
 you can judge the practical value when you remember how 
 Kant counselled the unbelieving priest. Green himself, 
 
 ^ Prolegomena to Ethics, p. 206, § ig6. 
 
 ^ This very expression he twice used a Httle before (pp. 190, 193) in 
 such a way as to show he felt the danger of such a charge.
 
 204 LETTERS ON REASONING Letter XL 
 
 after a few verbalising sentences, admits that " when we ask 
 ourselves what it is that this imperative commands to be 
 done, we are met with just the same difficulty as when asked 
 to define the moral ideal or the unconditional good," and 
 that no particular duty can be so specified, whence arises " a 
 suspicion that, after all, there is no categorical imperative, 
 no absolute duty, at all." But he goes on : 
 
 " After the explanations just given, however, we need not shrink from 
 asserting as the basis of moiality an unconditional duty, which is yet not 
 a duty to do anything unconditionally except to fulfil that unconditional 
 duty. It is the duty of realising an ideal which cannot be adequately 
 defined till it is realised, and which, when realised, would no longer 
 present itself as a source of duties, because the should be would be 
 exchanged for the iV." 
 
 I should like to watch an unsophisticated conscience 
 digesting that pleasing formula, of which the significance is 
 that we emphatically ought to do whatever we ought because 
 we ought, though it is often a toss-up what we ought to do ; 
 and that when we have realised the ideal of oughtness by 
 doing all we ought to do there is no ought left. The edifying 
 upshot in practice is, for instance, that if a savage feels he 
 ought to get dead drunk at a certain religious festival 
 precisely because he knows he ought, he is obeying the 
 categorical imperative ; and when he realises his ideal by 
 getting dead drunk it ceases to be a source of duties. 
 Which would be, in one sense, only too true. 
 
 A similar determination to make out a foregone con- 
 clusion appears to be the explanation of the passages in 
 which Green constructs a verbal semblance of reason for 
 believing in something that may be called God. The idea 
 of human development, he argues, implies " the eternal 
 realisation for, or in, the eternal mind of the capacities 
 gradually realised in time "; and also " that the end of the 
 process of development should be a real fulfilment of the 
 capacities presupposed by the process^ ^ Now, " develop- 
 
 1 As cited, § 187.
 
 Letter XI. LETTERS ON REASONING 205 
 
 ment " is a term for a certain process of change cognised as 
 having happened and (it may be) as likely to continue ; 
 and when used in regard to the human species the term 
 signifies our hope or belief that in the mass it becomes 
 gradually wiser and better. The " capacities pre-supposed 
 by the process " are simply part of the total phenomenon. 
 To say, then, that this conception in particular implies an 
 eternal realisation in an eternal mind is to make an absolutely 
 gratuitous assumption. If one development implies it, so 
 does every other. The idea of development from acorn to 
 oak must equally call for accommodation in " the eternal 
 mind "; and no less must the development of disease, 
 madness, wickedness, and the decadence and destruction of 
 civilisation. The whole infinite cosmos, in short, must be 
 similarly thought of; and the sole citation of the case of 
 humanity is either pure fallacy or pure evasion. 
 
 As we shall see later, the complete thesis is sometimes 
 formulated. But Green's sole idea in this connection seems 
 to be that of foisting the God-idea on his argument some- 
 where by main force. The outcome is this memorable 
 proposition (the italics are Green's) : — 
 
 "When we speak of any subject as in process of development accord- 
 ing to some law, we must mean, if we so speak advisedly, that that 
 into which the subject is being developed already exists for some con- 
 sciousness A state of life or consciousness not yet attained by a 
 
 subject capable of it, in relation to that subject we say actually is not ; 
 but if there were no consciousness for which it existed, there would be 
 no sense in saying that in possibility it is, for it would simply be nothing 
 at all. Thus, when we speak of the human spirit being in itself, or in 
 possibility, something which is not yet realised in human experience, 
 we mean that there is a consciousness for and in which this something 
 really exists, though, on the other hand, for the consciousness which 
 constitutes human experience it exists only in possibility." 
 
 1 feel bound to say that if this process of quasi-reasoning 
 be " philosophy," that subject would, as has been said of a 
 bad novel, "repay careful avoidance." I cannot recall a 
 more monumental example of verbal inconsequence. The
 
 2o6 LETTERS ON REASONING Letter XL 
 
 " therefore " of the argument, as you will see, might have 
 been framed as a practical joke by the author of " Alice in 
 Wonderland." Observe, it is perfectly true that " there is 
 no sense in saying that in possibility it is"; and the rational 
 sequitiir is, " therefore don't say it." The phrase is simply 
 an abuse of language, a bad pun on the word " is." But on 
 the possibility of committing this abuse of language Mr. 
 Green founds a demonstration of the existence of an eternal 
 mind in which all imaginable futurity is at every moment 
 cognised as existent ! 
 
 Think, too, of the logical implications of the proposition. 
 Unless it be confessed that the eternal mind depends, as it 
 were, on men's use of the specially absurd form of words, 
 " human betterment in possibility u," every man who asserts 
 that a specified development is possible means ("if he so 
 speaks advisedly ") that the conceived state of things 
 actually exists in the eternal consciousness. And when Mr. 
 Gladstone said that in a given case three courses were 
 possible, he must have meant that all three developments 
 were then already existent, though his express purpose was 
 to prevent two of them from coming into existence. The 
 religious comfort derivable from Mr. Green's view, again, 
 must be peculiar when it is realised that on his principle 
 every sin men can conceive of themselves or others as 
 " possibly " committing has been already committed in the 
 eternal consciousness. It is a small matter, in addition, 
 that when a Fire Insurance Company formally avows its 
 belief that your house may possibly be burned, it is either 
 not acting advisedly or is satisfied that in the eternal con- 
 sciousness the house is burned already. 
 
 It seems necessary to point out, in the interests of rational 
 philosophy, that all this is gratuitous nonsense ; that the 
 vocable " is," in the phrase " it is possible," is not a predica- 
 tion of an existence ; that what is philosophy for Mr. Green 
 would cease to be philosophy in French, where one would 
 say " cela peut etre "; that even in English, when the truth
 
 Letter XI. LETTERS ON REASONING 207 
 
 is told, the proposition, " such-and-such a development is 
 possible," merely means " we should not be surprised if it 
 does happen "; and that even when we say " it is certain 
 that we shall die," we do not mean that in our opinion we 
 are already dead in the eternal consciousness. I know, of 
 course, that it has been verbally demonstrated that in the 
 eternal mind there is neither past nor future, but only an 
 eternal present comprising past and future ; but I am fain to 
 suggest to you that that is simply a way of showing that we 
 had better not talk about an eternal mind. Propositions 
 should be at least partially cogitable. 
 
 And the crowning trouble for those who think (or phrase) 
 with Mr. Green is that after in effect affirming that all of us 
 are to the eternal consciousness already dead and buried at 
 least, Mr. Green's doctrine comes finally to formulating 
 "the true notion of the spiritual relation in which we stand 
 to God" thus : — 
 
 "He is not merely a Being who has made us, in the sense that we 
 exist as an object of the divine consciousness in the same way in which 
 we must suppose the system of nature so to exist, but that He is a Being 
 in whom we exist ; with whom we are in principle one ; with whom the 
 human spirit is identical, in the sense that He is all which the liuman 
 spirit is capable of becoming." 
 
 That is to say, Nature fiierely exists in the divine conscious- 
 ness, but does not exist in the divine Being, whereas we do 
 both, whatever that may mean ; and our spirit is identical 
 with " Him " " in the sense " that He is all that all of us are 
 capable of becoming, be it good, bad, or diabolical, whereas 
 he is ?iot all that is capable of coming to pass in Nature, 
 though all future developments of Nature all the sarne exist 
 in " His" consciousness. 
 
 I am fain to repeat that I cannot recall such an elaboration 
 of verbal nullity in serious literature. And yet this is part 
 of the philosophy that is said to pass current most widely in 
 my day among English students. It would seem, then, that 
 Green's influence has not been wholly fortunate, whatever
 
 2o8 LETTERS ON REASONING Letter XL 
 
 may have been his personal excellence. When, after long 
 progressions of such reasoning as this, we find him still 
 laboriously contending for such verbal distinctions as that 
 between "productivity of pleasure" and "productivity of 
 desirable consciousness," ^ we are entitled to say, I think, 
 that he has turned away from the purpose of ethics, the 
 rectification of conduct, to the establishment of a series of 
 formulas which shall safeguard his theism, which in turn is 
 made to safeguard his conformism. And when we realise 
 that his most tangible doctrine as to good action — that it is 
 that which is done with the view of promoting ultimate 
 human perfection— may be turned in practice to all manner 
 of iniquity by men utterly averse to Green's own ideals, 
 men who can persuade themselves that the injustices and 
 cruelties they commit will secure an ultimate gain to civilisa- 
 tion — when we realise this we are entitled to say that no 
 more in his hands than in Kant's has the theistic metaphysic 
 of ethics been made a sure means to the betterment of 
 life. 
 
 That theological bias was the cause of such futile 
 reasoning on Green's part seems to me clear ; but it may 
 become clearer to you how disloyal he could be to the 
 test of the " unsophisticated conscience " if I quote one 
 of his remarks in his address on " Faith " : — 
 
 " Inal)ility to adopt the creeds of Christendom in their natural sense 
 — and in any other sense they are best left alone — need not disqualify us 
 from using its prayers They are not meant to be heard of men."'** 
 
 What the last phrase means it is somewhat hard to say. 
 The object of the suggestion about using the prayers of 
 Christendom is ostensibly to persuade sympathetic unbe- 
 lievers to join in public worship ; and in another address 
 Green urges on the same kind of audience the use of 
 "Christian ordinances."^ To say, then, that the prayers 
 " are not meant to be heard of men " cannot mean that the 
 
 ^ Prolegomena, § 364. ^ Works, iii. 274. " lb., 251.
 
 Letter XI. LETTERS ON REASONING 209 
 
 unbeliever is to pray privately, so as to avoid deceiving 
 believers about his opinions. The only meaning I can 
 attach to it is that the non-Christian need not trouble him- 
 self about the inconsistency of praying publicly in the name 
 of Christ, since his fellow-men are not entitled to judge him 
 in respect of prayers not addressed to them — a proposition 
 which I can but characterise as astonishing. But even if 
 the last-cited sentence be either put altogether aside or 
 regarded simply as meant to enforce the other, one thing 
 remains clear, that Green was quite ready to outrage the 
 " unsophisticated conscience " where it was at once socially 
 convenient and arguably conducive to good discipline to do 
 so. In other words, he counselled religious make-believe as 
 good both for those who practise it and for those in whose 
 company they dissemble. It is the inconsistency of Kant 
 over again. To some such moral perversity we seem 
 always to come in the case of men who insist on making 
 the moral law somehow transcend reason and experience 
 and utility. 
 
 It is a fine question why Green thought fit, in the very 
 act of counselling make-believe, to protest that if the creeds 
 of Christendom be not accepted in their natural sense they 
 are best left alone. What he advised was just the acceptance 
 of them at one point in a non-natural sense ; and after that 
 measure of acceptance it is hard to see why, logically speak- 
 ing, he need draw the line anywhere else. His plea that 
 men are not entitled to condemn inconsistency in a formal 
 prayer because it is not addressed to men, is about as 
 extensive an excursion into non-natural ethic and argument 
 as can well be made. But I suppose that he was salving 
 his conscience by condemning one rather obvious and 
 familiar form of dissimulation while he was justifying another 
 by an abnormal argument. The fact remained that he, who 
 did not believe in the creed of Christendom, held an office 
 which was understood to exact such belief, and that he, thus 
 standing in a morally false position, advised other men to
 
 2IO LETTERS ON REASONING Letter XL 
 
 dissemble to the extent of joining in the official prayers and 
 other " ordinances." 
 
 If we were to treat " religious " men in the fashion in 
 which they too often treat others, we should proceed to say 
 that such a moralist, in his appeals to the unsophisticated 
 conscience, is a hypocrite; and that he has no right to our 
 attention. But a good rationalist in morals will be more 
 magnanimous, more sympathetic towards the transcenden- 
 talist than the latter has ever been towards him ; and will 
 recognise that a deviation from intellectual rectitude under 
 heavy social pressure may be made by a man otherwise 
 scrupulous. I have already urged you to make allowance 
 for the position of a disillusioned priest. It is not to be 
 forgotten, further, that men placed like Green spontaneously 
 tend to hope, and to assume, that good may be done by the 
 maintenance of rites whose religious basis is illusory. Green, 
 knowing that he remained deeply conscientious in other 
 relations, and also that some sincere believers in the current 
 creed were equally conscientious in general, committed 
 (apparently) the common fallacy of assuming that to do as 
 he and those believers did was a security for conscientious- 
 ness. In point of fact, of course, he and those believers 
 were conscientious in general, not because of either the belief 
 or the conformity to it, but rather in spite of such belief and 
 such conformity, since neither is logically conducive to 
 conscientiousness, and many believers and make-believers are 
 notoriously unscrupulous. When you see an exceptionally 
 good man in either camp you may be sure that he has a gift 
 for goodness, a strong innate bias to conscientiousness. And 
 Green had such a bias ; only it was overruled, as so often 
 happens, in the direction of a religious conformity to which 
 he was pressed by his position, and obscurely led by his 
 theism, and in which he was confirmed by the faculty for 
 fallacy which theism implies and conserves. 
 
 I'allacy, as we have repeatedly seen, always works out as 
 inconsistency. Had Green been true to the logical implica-
 
 Letter XI. LETTERS ON REASONING 211 
 
 tions of his own doctrine of the " Spiritual Principle," which 
 was simply his name for his aspiration towards the good, 
 he would have admitted that the instinct for truth is to be 
 respected to the utmost, and would have recognised that all 
 movement to a higher good takes social shape as a variation 
 from the common standard of good. If a student, for 
 instance, should have first spontaneously recoiled from 
 religious make-believe as an unworthy act, and should then 
 have been reconciled to it by Green's persuasion, Green had 
 so far lowered that student's standard of aspiration, and 
 lessened the amount of energy available for human progress. 
 It is always possible to make out a more or less plausible 
 utilitarian argument for insincere conformity. Such argu- 
 ments, we have seen, can be framed for acts of gross inter- 
 national violence. But their ostensible utilitarianism is 
 always fallacious : the induction is always viciously incom- 
 plete. The true working rule for conduct is that the law of 
 reciprocity and consistency is inflexible, save in cases that 
 can be seen to be essentially exceptional, such as some of 
 those we have discussed ; and these cases are to be fully 
 made out on their merits. If there were no general pre- 
 sumption, no felt rule, every act of our lives would be a pain- 
 ful or at least a precarious calculation of probabilities ; and 
 a premium would be put on anti-social egoism. 
 
 I do not dispute that even an act of make-believe may at 
 times be quite justified on utilitarian grounds, as when a 
 man dissembles to avoid causing intense pain to an already 
 suffering mother, or to save his life among fanatics, or 
 even to avoid creating a certainly useless disturbance among 
 beUevers whom he has strong ground for holding to be 
 incapable of new ideas, and whom he may hope to influence 
 for an ulterior good if he does not horrify them on the side 
 of their creed. As regards the course of keeping silence 
 where it is not socially fitting to discuss serious things, no 
 question of conformity properly arises : it is no man's duty 
 to force discussion in such circumstances. But if we extend 
 
 Q
 
 212 LETTERS ON REASONING Letter XL 
 
 such justification to every act of conformity on the bare 
 ground that it preserves concord and is useful to the con- 
 formist, we have effaced all distinction between unconscien- 
 tious and conscientious conduct in such matters. And that 
 is at once the most fundamentally anti-utilitarian and the 
 most anti-intuitionist of all courses. 
 
 Let us apply the historic test. In the ancient Semitic 
 world there were practices of sanctified vice, one of them 
 being the law that every woman in certain cities must sell 
 herself once in the temple of the Goddess of Love. Once 
 such a law had become established, any man or woman who 
 revolted against it would create much disturbance and even 
 distress : parents would be shocked and terrified ; " religious" 
 people would sincerely regard the revolt as blasphemous ; 
 priests would so denounce it , and there would be a general 
 expectation of divine vengeance if the blasphemy were 
 tolerated. It would take a strong will and a great gift for 
 moral judgment to persist in the blasphemy ; and it would 
 take an enlightened eye for utility to see that the protest 
 was useful in the highest degree. In all likelihood such 
 protests were at times made, and punished by death ; and 
 many who were disposed to make them would be either 
 menaced or persuaded into conformity. Yet we know now, 
 if we know anything of sociology, that such protests were 
 stirrings of the so-called Spiritual Principle, affirmations of 
 the Will to Good, and valuable movements towards the 
 highest social utility — the production of higher character 
 and more refined feeling, the enlargement of sympathy and 
 reciprocity, the substitution of reason for superstition. 
 
 Now, what reason have we for doubting that in the 
 (certainly less serious) case of the revolt either against such 
 an irrational act as prayer in general, or against the special 
 inconsistency of ostensibly professing in prayer a creed one 
 does not hold, the " spiritual " and social utilities are not 
 equally on the side of honesty ? None whatever. Granted 
 the marginal cases — and such cases would arise in connection
 
 Letter XI. LETTERS ON REASONINH 213 
 
 with the most vicious sanctities of the ancient world — the 
 rational presumption is overwhelmingly on the side of 
 honest and courageous conduct. There is no other agency 
 by which vicious conventions can ever be discredited. 
 Every reform in human history has had to be made against I 
 outcry ; and the greater the reform the greater the outcry. 
 Broadly speaking, the outcry takes the shape of stoning the 
 prophet ; and every generation raises sepulchres— or statues 
 — to the memory of the stoned prophets of the last age, 
 yet proceeds confidently to stone its own prophets in the 
 historic way. 
 
 Doubtless there are more false prophets than true. We 
 must always revert to the utilitarian test : I am doing so now. 
 But whether or not the conservatives are right when they 
 say of the prophet " he lies," you have ground throughout 
 all history for suspecting that they are wrong when they say 
 "he blasphemes"; and you may be quite sure that those 
 are working evil whom you hear saying under their breath 
 " it doesn't pay," or " let us keep in with the majority." For 
 even if such dissemblers chance in one case to be opposed to 
 an error, it is by pure hazard : their principle is one which 
 paralyses the very faculty of distinguishing between error 
 and truth ; and their lives, rationally considered, are moral 
 failures. 
 
 To be quite scientific, let us say that in such cases there 
 is an apparent conflict of utilities. There is an obvious 
 utility of present comfort, and a less obvious one of future 
 higher harmony resulting from an extension of consistency 
 in human relations. But that is not all. There is the 
 conflict between the present utility of social comfort and 
 the present happiness of conscious loyalty to our own 
 perception of truth ; and your life will be determined in 
 large part by the measure of reality which those differing 
 utilities possess for you. 
 
 If there be any rational meaning attachable to such 
 phrases as " spiritual principle " and " spiritual life," they 
 
 Q 2
 
 214 LETTERS ON REASONING Letter XL 
 
 should apply, on Green's own principle, to the habit of 
 finding our highest good in faithfulness to this intellectual 
 view or estimate of good. But an ideal, remember, may be 
 either right or wrong. It may be held by an insane fanatic 
 who regards himself as the mouthpiece of God, or by the 
 thinker who has taken the utmost pains rationally to under- 
 stand the good, and is loyal to his larger knowledge. Where 
 I differ from Green, as I understand him, is in recognising 
 that every ideal of good is to be tried by the test of utility 
 in the light of the largest obtainable experience. My counsel 
 to you then is, once more, to try your aspiration stringently 
 by that test, and, if it can bear the trial, to stand by it in 
 confidence. The more real your thought is to you, the 
 more decisive will be your satisfaction in living by it, and 
 the more " spiritual " your level of life. 
 
 Unless, however, your experience be very different from 
 mine, you will often find the professedly " spiritual " people 
 the furthest from such spirituality. Even when they meet 
 the definition in respect that their thought, their ideal, is 
 more real to them than the normal notion of pleasure, they 
 often hold it as a mere unchastened bias, or a devotion to a 
 pleasurable state of quasi-hysteria. Any worshipper of an 
 idol or icon is apt to be reckoned spiritual in comparison 
 with a viveur and a money-getter, in that he contemns or 
 disregards their pleasures. But inasmuch as he is simply 
 devoted to or hypnotised by a fixed idea, he has not mentally 
 transcended their plane : he merely heads a different way. 
 And so with many pietists who habitually dwell on the idea 
 of their deity which they have gathered from early teaching, 
 or from pictures, or from a little reading and reflection. 
 They have never challenged themselves, never tried their 
 thought in the fire of reason, never held it up to the large 
 light of knowledge. It is thus simply a variety of carnal 
 inclination, no more necessarily conducive to human well- 
 being than any obstinate man's persistence in a given course 
 merely because he has a will to it. When such a man
 
 Letler XL LETTERS ON REASONING 215 
 
 shows that, in disregard of the happiness of others, he finds 
 his happiness rather in his loyalty to a revengeful or unsocial 
 course once taken, even at the cost of loss and odium, 
 than in tasting the pleasures and profit of society, 
 we do not on that account certificate him as a lofty 
 and spiritual personality. In his case we have applied the 
 broad tests of human utility. But they ought to be applied 
 in all cases. Faithfully applied, they bring us and keep us 
 to the conclusion that the proper purport of spirituality is 
 just intellectuality, largeness and consistency of thought, the 
 power to find in the processes and results of reason a more 
 durable reality than in the satisfactions of the lower or less 
 developed mind. 
 
 As for the final risk that a man may find his highest 
 " reality " in a delusion, it is a risk we must all consent to 
 take. Looking around me, I see many men combating year 
 after year for causes that in my eyes are not merely lost, but 
 false; battling for errors; spending life and treasure to 
 restore the credit of the incredible, and to set up an order of 
 things hostile alike to good morals and to well-being. Such 
 men would of course see me in the very light in which I see 
 them ; and no thoughtful man can look long on those 
 antitheses of ideals and convictions without searchings of 
 heart. In this dilemma we have only the resource on which 
 I have been insisting — that of applying the severest tests of 
 utility and consistency ; but at least those who avow and 
 bear fealty to such tests have a safeguard that is lacking to 
 those who insist on " looking within " for guidance first and 
 last. Within, certainly, is the spirit that judges ; and who 
 shall say how far spirit may differ from spirit ? But between 
 the spirit open to knowledge, to reason, and the spirit that 
 wilfully closes ear and eye to the lessons of things, there is 
 all the difference that subsists between culture and fanati- 
 cism, between fortitude and savagery, between sanity and 
 hallucination. 
 
 What wisdom can do, you have done when you actively
 
 2i6 LETTERS ON REASONING Letter XL 
 
 recognise that a course is not rationally to be called right 
 merely because your bias or your interest lies that way; that 
 no stress of mere passion can serve to make it higher or 
 better ; that your sympathies may be astray, and your ideals 
 ill-developed. But when you have tried the case by the 
 tests as you understand them, when you have weighed bias 
 in the scales of utility and of reciprocity, and have rationally 
 explained to yourself in terms of error or interest the diver- 
 gence of other men from what you are convinced is the right 
 course, then the measure of your character, of your spiritu- 
 ality, of your moral stature, is just your persistence in your 
 doctrine. You will never, I hope, go about saying like 
 Fichte that the All is speaking through you in particular ; 
 but you may fitly say to yourselves that whatever force for 
 good there may be in the cosmos is as truly incarnate in you 
 as it can ever be subjectively known to be. We who speak 
 and act are for the moment the consummation of the cosmic 
 drift ; and there is no higher law or revelation for men.
 
 LETTER XII. 
 
 Round that question of "reality," to wliich we have come 
 by way of the problem of conflicting ideals and the conflict 
 between ideals and expediencies, there lies a whole cycle of 
 philosophic dispute. Coming to it by way of morals, we 
 see, broadly speaking, that that is relatively most " real " to 
 a man for which he cares most ; and if we are heedful of the 
 implications of that statement we recognise that things 
 differ for us not, strictly speaking, as real and unreal, but as 
 more or less real. The word "real," however, in itself, 
 raises difficulties as soon as we set about defining it. It has 
 as its correlatives " unreal " and " seeming " or " merely 
 apparent"; and when we seek to frame for ourselves a com- 
 plete philosophy, we arc bound to reckon with the problem 
 those words involve. 
 
 Well, as I have already remarked, I am not undertaking 
 in these letters to provide you with a philosophy ; but this 
 ultimate problem gives so many tough exercises in reason- 
 ing that I am fain to offer you hints as to how they are to 
 be met. No debate, I believe, more fully illustrates the 
 truth of what has been said about the imperfection of the 
 instrument of language, though some of the most elaborate 
 discussions of the subject make no confession to that effect ; 
 and I have long held that the controversy might be very 
 much shortened if the verbal intricacies of the case were 
 narrowly scanned to begin with. Professor Bradley, who 
 has written a brilliant treatise on "Appearance and Reality," 
 says in one place ^ : — 
 
 " I confess thai I shrink from using metaphors, since they never can 
 suit wholly. The writer tenders them unsuspiciously as a possible help 
 
 ^ Third edition, page 194.
 
 2i8 LETTERS ON REASONING Letter XIL 
 
 in a common difficulty. And so he subjects himself, perhaps, to the 
 captious ill-will or sheer negligence of his reader." 
 
 There is another way of looking at the matter. Professor 
 Bradley, who says he shrinks from using metaphors, seems 
 to me to use them in greater profusion than almost any 
 other metaphysician ; and though he is certainly not a sloth- 
 ful thinker, his metaphors often strike me as evasions of a 
 difficult analysis, or, at best, attempts to get an easy way out. 
 The result is great brilliancy of style and argument, with 
 endless obscurations of the issue. Apparently he does not 
 realise how metaphorical his method is ; for when he pro- 
 ceeds in this case avowedly to venture on a "metaphor" he 
 supplies, as he himself says, a "fiction"; and it is one to 
 which I, for my part, can attach no logical purport whatever. 
 That done, he recommences to coruscate in metaphors, 
 showering them (to speak in his own manner) on every step 
 of his argument, apparently without knowing it. 
 
 Now, this is a problem which suffers particularly from a 
 metaphorical method, simply because metaphor is in its 
 nature a creation of new verbal complications, and this is 
 a dispute which begins in a need for verbal analysis. All 
 language, indeed, as has often been observed, is funda- 
 mentally metaphorical : that is precisely why philosophy is 
 so hard a task ; but for that very reason we are not helped 
 by multiplying the concrete figures which stand between 
 us and abstract ideas. To begin with, we find the philo- 
 sophers of India laying a stumbling-block in the first step 
 on the path of this very discussion by saying that "All is 
 illusion." Such a proposition, as I have before pointed 
 out, is strictly meaningless. " Illusion " is significant for us 
 solely as correlative to "reality" or some other term standing 
 for "non-illusion"; and to apply it to the infinity of pheno- 
 mena is to reduce it to nullity by cancelling the correlative 
 which defines it. To think of illusion at all, we must think 
 it against known reality; as we conceive "mirage" to be 
 a visibly transient phenomenon in contrast with more
 
 Letter XII. LETTERS ON REASONING 219 
 
 permanent things. And to argue that all of those more 
 permanent things are but mirage in relation to a spiritual 
 reality "beyond," does not help us when the alleged 
 reality is admittedly a simple inference, and impossible of 
 conception. 
 
 Of course we know what the Indian philosophers are 
 driving at. They use the term for an admitted deception 
 of the senses to force upon us a doubt of the whole 
 testimony of the senses, a recognition of the transiency of 
 all known things. But the service thus done is in the next 
 instant undone if we rest in the counter-sense they have 
 framed. If "all is illusion" their doctrine is illusion: in 
 other words, it is an illusion to think that all is an illusion. 
 On that line of reasoning we get "no forwarder"; and to 
 make progress we must first look sharply to our terms. 
 
 The form and pressure of the problem come upon us 
 very clearly in the study of elementary physics. We con- 
 sidered it in a former Letter as it arises in the theological 
 handling of the so-called primary and secondary qualities 
 of matter. Let us consider it anew from the scientific side. 
 When you got your first lessons in the science of "light," 
 you probably learned that colour is not an attribute or 
 quality of an object, but is a state of perception set up in us 
 by the fact that a given body reflects certain of the light rays 
 upon our retina, absorbing the others. In my time, however, 
 the physicists, while taking care to deny that a green leaf 
 is "in itself" green, were wont to speak of the "red and 
 green rays," having apparently no difficulty in crediting 
 colour to a ray "in itself" I have never been able to rest 
 in that sort of solution ; and as little can I find satisfaction 
 in saying that colour exists "only in the eye," or "only 
 as perceived." That method lands us in the old logical 
 impasse of ascribing a single mode of happening to an 
 infinite category, for if we are consistent we cannot rest 
 with specifying only colour as a "visionary" or subjectively 
 given attribute of things. As colour is seen, so are hardness
 
 220 LETTERS ON REASONING Letter XII. 
 
 and warmth felt ; and if the colour exists "only as perceived," 
 equally the coldness and hardness of stone exist "only as felt." 
 
 Now, in a sense those propositions are reasonable, for 
 we have no way of thinking colour, and hardness, and so 
 forth, save through our sense-perceptions ; and to imagine 
 one is to imply the other. But when we think of the past 
 history of the earth, as inferred from the data of the sciences, 
 we are left with a residual conviction that there was an 
 earth of a certain sorl when there were no eyes or other 
 human sense-organs to perceive it ; and this conviction is 
 not reducible by any formula about the dependence of 
 existence on perception. The conviction is more deeply 
 realised than the argument, which subsists only as a verbal 
 sequence unaccompanied by clear ideas. Some thinkers at 
 this point resort to the very simple device of saying that 
 " God " was there to perceive everything, just as Mr. 
 Green says " He " is there to make future development 
 possible by perceiving it before it happens ; but you will see 
 that, on the one hand, their God thus exists, in terms of 
 their own case, on that mere tenure of their retrospective 
 imagination which they declare to be insufficient to vouch 
 for the world which he was required to perceive ; while, on 
 the other hand, if you reason the matter out, you will find 
 once more that the God which thus perceives everythmg 
 must logically consist in the entire cosmos, including all the 
 orders of life with their different orders of perception, and 
 all the orders of things perceived. 
 
 This last process of thought leads us, I think, to a sound 
 philosophical position ; but nothing save confusion, so far as 
 I can see, can be reached by the plan of calling the theoretic 
 cosmos " God." That term has for ages signified, by the 
 confession of every single thinker who uses it, an infinity of 
 delusions, for not only have no two theists had the same God- 
 idea, but it is quite destructive of their case to say that the 
 different God-ideas stand merely for the different aspects of 
 a perceived reality, in the way in which we all admittedly see
 
 Letter XII. LETTERS ON REASONING 221 
 
 different rainbows, and slightly different forms of any given 
 object under simultaneous observation. If it be argued, for 
 instance, that Jupiter or the whole Greek pantheon is but 
 the Hellenic perception of a reality seen by Semites as so 
 many Baals and Goddesses, or as Yahweh, we must either 
 entirely give up the discrimination of illusion and reality or 
 put aside that fashion of terminology. And you will find 
 that it must be put aside if you are to carry on argument ; 
 for the discrimination between truth and error, or sound 
 reasoning and fallacy, stands or falls with the discrimination 
 between reality and illusion. If we were to say that all 
 illusions are but particular aspects of " the " reality, we must 
 next say that all fallacies are but aspects of the true con- 
 clusion ; and in that case we are all alike right in every 
 dispute. 
 
 Nevertheless, there remains a problem to solve in the 
 spectacle of the world's million God-ideas : a problem which 
 on analysis turns out to be just the general problem of 
 reality and illusion. Even as the "attributes of matter" 
 turn out on analysis to depend as it were on our senses, 
 leaving us wondering what is permanent or real, so does the 
 aspect of our moral world (as has already in part appeared 
 in our discussions) turn out to depend on the cast of our 
 minds ; and we are led to ask, Is there any less reality in 
 an intensely-held conception of one or many personal Gods 
 or Goddesses than in a perception of colour ? Colour seeins 
 to consist or arise in a relation of light, object, and eye ; 
 and when we call that man " colour-blind " who sees as 
 green what we see as red, are we doing aught but saying 
 that our vision must be the right vision ? Many different 
 groups of God-worshippers have in that fashion called the 
 Gods of the other groups "false," meaning sometimes that 
 they were bad Gods, sometimes that they did not exist. 
 When we in turn say that all their ideas alike were illusions, 
 have we placed them on a different philosophical footing 
 from our sense-perceptions ?
 
 222 LETTERS ON REASONING Letter XIL 
 
 Here we begin to find standing-ground. While men's 
 God-ideas shift and flow hke clouds as seen from age to age 
 and from place to place, their colour-ideas seem to remain 
 permanent ; and the variation called " colour-blindness " is 
 also known to follow in the main one or two lines. Thus, 
 without settling in advance the question as to the absolute 
 meaning of reality, we recognise, to begin with, that some 
 perceptions or conceptions are in one aspect more real than 
 others. We see that a man brought up to believe in the 
 God Krishna or the God Yahweh can be taught to realise 
 that these God-ideas are delusions, and to cease to ascribe 
 reality to them ; while, though he may be taught to say that 
 a blade of fresh grass is "not really green," he certainly 
 believes that it is, because he has no other way of thinking 
 about fresh grass. However he may philosophise, he is at 
 one with the rest of mankind in thinking of stones as hard : 
 nay, mankind are thus at one even while they continue to 
 ascribe reality to their own God-ideas and deny it to those 
 of all others. 
 
 Shall we then say, on the one hand, that what is per- 
 manent is " the " real, or on the other hand that that is real 
 as to which the mass of mankind are agreed ? To say this 
 would be to land ourselves in new dilemmas ; for we have 
 learned to regard as illusion the apparent rising and setting 
 of the sun, as to which the mass of mankind for ages had 
 an unquestioning conviction of reality ; and on the other 
 hand mere permanence is but a narrow mark of reality. 
 Is the cloud less real than the mountain ? Tennyson 
 answers : — 
 
 " The hills are shadows, and they flow 
 
 From form to form, and nothing stands ; 
 They melt like mist, the solid lands, 
 Like clouds they shape themselves and go." 
 
 Impermanent things, we reflect, are not what we mean by 
 illusions. The mirage itself, one of the types of illusion, is 
 for the time a real phenomenon, perceptible to a number of 
 
 3
 
 Letter XII. LETTERS ON REASONING 223 
 
 men at the same moment : it is illusory only as an image in 
 a mirror would be if we did not know we were looking in a 
 mirror. \Vhat, then, shall we say of the spark or flash we 
 see when something strikes our eye ? It is like the mirage 
 in respect that, after one experience, we are not deceived by 
 it : we know there has been no flash for the eyes of others 
 beside us : yet we recognise the visual phenomenon. Whether 
 or not we call the flash thus produced a real flash, the per- 
 ception of it is for the moment a reality. 
 
 Once more we regress to " real illusions " — the illusions 
 we have found out as such for ourselves, and those we recog- 
 nise as continuing in the beliefs of others. Lord Tennyson, 
 who was very confident over his not very profound reason- 
 ings, has suggested that those who did not see eye to eye 
 with him in religious and philosophic matters were there 
 victims of the illusion of " the straight staff bent in a pool." 
 But inasmuch as Lord Tennyson held in the main with the 
 beliefs of the average man of all ages, it seems at least as 
 likely that he was deceived by appearances as that his 
 antagonists were. The rod bent in water is the type of the 
 simplest form of illusion — that which is but momentary, and 
 creates no deception after one experience. It is possible, 
 indeed, that a savage may sincerely believe that the ocular 
 illusion is wrought by a spirit in the water ; and it is 
 quite likely that a savage who had been once allowed by 
 a civilised man to look through a field-glass would believe, if 
 so told, that the white man was a great magician. All men, 
 however, are capable of being taught the truth on such 
 points ; and, once taught, they are at one. Either, then. 
 Lord Tennyson had access to occult forms of teaching or he 
 was resorting to bluster for lack of the ability to convince 
 his fellows of the truth of his conclusions. And if it be 
 found that in an increasing number of cases those of his 
 way of thinking can be led by reasoning to the other, but 
 not vice versa, there is at least room for a rational presump- 
 tion that the illusion lay on his side.
 
 224 LETTERS ON REASONING Letter XIL 
 
 Still the fact would remain that many men live and die 
 convinced of the reality of their conceptions of things 
 declared to be supernatural and undemonstrable ; and we 
 who may hold them utterly deluded may on our own part 
 live and die convinced of the importance of ideals and the 
 truth of theories which the majority of our fellows regard as 
 " visionary." For us, I have argued, those beliefs are " real " 
 in the measure in which they suffice us for motives of con- 
 duct or subjects of emotion and contemplation. Yet it 
 follows from or is part of some of those very beliefs, that 
 others which similarly suffice for other men are illusions; 
 and it lies on the face of the case that some of our " realities " 
 are as such much less easily demonstrable than some which 
 we treat as of less importance, such as the satisfactions of 
 popularity, wealth, and physical comfort or luxury. 
 
 Let us, then, keep the problem as far as possible clear at 
 the outset by stating it under two aspects : that is to say, let 
 us note that "reality " has in usage two orders of meaning, 
 one that is avowedly metaphorical and self-regarding, and 
 another that has regard to objective facts. On the subjective 
 side, we recognise that " reality " is a name for degree of 
 conviction, or bias, or preoccupation ; that ideals are " real " 
 for a man in the measure in which he is content to live for 
 them ; and that his inner life may in this sense be more real 
 for him than physical comfort. All the while, we recognise 
 that his cherished reality may be a delusion ; and that 
 indeed it is very likely to be so if he does not at all times 
 keep it open to the tests of reason and evidence — evidence 
 of consistency and utility in matters of morals ; evidence of 
 error or actuality in regard to propositions of existence. 
 
 If, having acquired certain traditional beliefs in the exist- 
 ence of "spirits " and deities, he clings to them in virtue of 
 emotional habit and imperfect judgment, we who have 
 studied the grounds and origins of such beliefs say he 
 believes in unrealities ; and no stress or vividness of feeling 
 on his part can make them aught else : the sole "reality" in
 
 Letter XII. LETTERS ON REASONING 225 
 
 the matter is just his stress of feeling, of delusion. Some 
 philosophers, desirous of emphasising the " reality " of the 
 feelings use loose expressions to the effect that " the God " 
 believed in '■'■exists as an idea in the minds of his worship- 
 pers " ; but this, I suggest to you, is a mere yielding to the 
 snares of language, in the manner of the Platonic philosophy, 
 which laid on humanity burdens of verbiage hard to bear, 
 and removable only after many generations of weary dispu- 
 tation. If we predicate "existence " in ienns of admittedly 
 false conceptions, then there exist many versions of every 
 one of us — to wit, the different ideas we have of each other. 
 On that line we reach simply the old countersense : " All is 
 Reality "is as meaningless a proposition as "All is Illusion." 
 If we are to argue coherently about reality, it must be by 
 way of aiming at a. discrimination among ideas. We are 
 part of a world of relativities. 
 
 This perception is at once the solution and the statement 
 of the dilemma of " subjectivity." I have argued that the 
 way (if there be any way) to escape absorption in subjective 
 delusion is to keep every belief in touch with tests — to refer 
 the idea of the Good back to the ideas of the True, the 
 Useful, the Good-for-All. It can be answered that the true 
 and the useful and the good-for-all are simply perceptions in 
 terms of our idea of the good ; and that there is thus no 
 final security against subjective delusion. But this I 
 admit, and advise you to admit. What I claim is 
 simply that, when we do of choice make the circuit of 
 the tests of the true, the useful, and the good-for all, we 
 have widened the scope of our intellectual and moral life ; 
 and that in making it we are actually conscious of correcting 
 or modifying our idea of the good. Here again it may be 
 answered that the Hindu Yogi is just as happy or as pain- 
 less in his introspective method as I can be in the circum- 
 spective. But this I again admit to be possible. My claim 
 is that he is less likely to help other people to be happy. 
 And if yet again it be urged that the preference of that
 
 226 LETTERS ON REASONING Letter XIL 
 
 result is merely the statement of my peculiar way of being 
 happy, I answer : Agreed. We come back finally to that 
 fact of moral bias. I am simply seeking to show you the 
 arguments for the belief that the social bias is absolutely 
 consistent with the doctrine that utility is the proper test of 
 happiness, the end of all rational action. If I thought you 
 were devoid of that bias, I should not expect to persuade 
 you to acquire it. 
 
 On the side of simple belief, again, it is not hard to 
 show that he who omits to apply the tests of evidence, of 
 consistency, runs the risk of suffering for his omission. If, 
 of course, he can be sure of never having his present confi- 
 dence or self-satisfaction shaken, if he can maintain to the 
 end his conviction that he has the truth within, he is in the 
 terms of the case safe. But how many men do we see 
 standing for a while to their prejudice and out-facing all 
 criticism, yet ultimately undergoing the chill of disillusion- 
 ment, the distress of knowing that they had resisted the 
 truth and propagated error. Professor Green, for instance, 
 held by conventional views as to the Christian Sacred Books 
 even after his reason had realised the error of much of their 
 content ; and he discoursed elaborately for his students on 
 some of the Pauline Epistles, on the assumption that they 
 were all genuine documents. Had he lived till to-day, 
 paying any heed to the special researches of critical scholars, 
 he would probably have realised that his labour was sadly 
 mis-spent. What is more, some who do realise it may be 
 thereby led to look askance at his moral ideals, to doubt 
 whether it is worth while heeding the teachings of one so 
 hypnotised by tradition. His friends, I should think, would 
 in that case suffer on his behalf. 
 
 I do not, of course, suggest to you that any man is thus 
 to be disregarded because on certain sides his sense of 
 reality was ill-trained. The point is that our happiness is 
 likely to depend to some extent on the accuracy of our 
 view of reality in the matter even of inferences as to past
 
 Letter XII. LETTERS ON REASONING 227 
 
 events, just as it certainly depends very much on the 
 accuracy of our perception of concrete reahty. I need not 
 urge upon you that, if you mistake an open window for an 
 open door, the consequences may be serious. But many 
 men are slow to realise that we do well to keep an open 
 mind for all reality. It has been argued, for instance, that 
 the very firmament is for us but a fretwork of points of 
 light, a far-away pattern which it concerns us little to know 
 the nature of; and, doubtless, it were better for us, if we 
 had to make such a choice, so to regard it than to let its 
 divined vastitude obliterate for us the realities of human 
 love and duty and experience. Granting that we might 
 happily dream a long dream of the stupendous realities of 
 the universe, we should sooner or later stumble disastrously 
 against the neglected facts of our environment, as Thales is 
 said to have fallen into a pool on his walk while gazing at the 
 midnight heavens. Here the argument from " the good " 
 admits of no dubiety.^ But we can brood on the mystery 
 of the firmament in due hours without loss of faculty for 
 love and labour; nay, with a new sense of the need for 
 love in our little lives, and the need for labour to let love 
 prevail ; whereas the life that knows no such brooding is 
 either a forced mill-round of toil and care or a narrow or 
 frivolous absorption in small or mindless joys. 
 
 Depend upon it, though some minute or "artificial" 
 knowledges ^ may hardly, if at all, meet the test of usefulness, 
 
 ^ Observe that tlie argument is often absurdly put, by first making 
 and then unconsciously revoking the admission that what a man can go 
 on finding good is for him as good as our different good is for us. When 
 you read this you will have appreciated your mother's story of the old 
 lady who said she was glad she didn't like beans, because if she liked 
 them she would eat them, and as she didn't like them they would be 
 very disagreeal)le to her. A good many serious arguments in sociology 
 follow exactly that formula. 
 
 "^ Such as that of the number of times the word " the " occurs in the 
 Bible. No known result thus far can justify the labour expended in 
 that investigation. 
 
 R
 
 228 LETTERS ON REASONING Letter XH. 
 
 every large perception of reality will bear it triumphantly. 
 It is not merely that the first star-gazers began the knowledge 
 which underHes and has evolved all philosophy, and has 
 alone made possible the compassing of our planet, but that 
 the further our imaginations reach into the immensity of 
 space the more sustaining is the solemnity of the emotion 
 gained. Our griefs are less wounding, our desires less 
 dazzling, when we place them in that tremendous vista; 
 and all the while oui better joys lose nothing of reality, 
 though the unworthier may, when we return to them with 
 minds purified by that deep contemplation. 
 
 Here, perhaps, you will be conscious rather that I am 
 preaching than that I am reasoning. Return with me, then, 
 to the argumentative problem, at the point of "objective" 
 reality. I have taken for granted, in the last few pages, 
 that objective realities are knowable as such, though we 
 have still to settle exactly what we mean by reality, and 
 have recognised that the attributes of what we call matter 
 are expressions of or for our relation to matter. It is usual 
 on this head to say they are " only " expressions of our 
 relation to matter, or that they exist "only" in our con- 
 sciousness. But the word " only " ought to be scrutinised 
 just as closely as names for attributes ; and those who thus 
 use it will find it hard to say what they mean. We have 
 already looked at the problem on one line : let us look at 
 it on another. 
 
 If I say, " There is only one plate on the table," there is 
 no difficulty about my meaning : there had been supposed 
 to be and might conceivably be two or more plates on the 
 table, and I am correcting a possible misconception : I am 
 substituting the idea of a less for a more. But if I say, 
 " Colour is only my sensation," do I substitute a less for a 
 more ? If so, what is the more ? Had anyone previously 
 thought colour was two or more things, or was in two places 
 at once ? No, it may be answered ; but we had been wont 
 to suppose colour existed outside of our consciousness.
 
 Letter XII. LETTERS ON REASONING ilq 
 
 Then what is meant by " outside of our consciousness " ? 
 We saw, you will remember, that such a definition is 
 logically void as regards the conception of consciousness ; 
 but let us waive that conclusion, and consider the thesis in 
 respect of the conceptions attaching to the descriptive terms 
 employed. Is there, let us ask, a place or world outside of 
 our consciousness ? In terms of the proposition before us, 
 either there is or there is not. If there is not, how came 
 we to have the idea of it ? And in that case, what is meant 
 by " outside " ? I can understand how men came to have 
 the idea of a God or Devil or angel or heaven or hell that 
 does not exist : it is simply a modification of their ideas of 
 men and of pleasant or unpleasant surroundings ; but how- 
 came they to think of an " outside " to their consciousness 
 if it be unthinkable ? 
 
 It is no use to answer that outside means " external to " : 
 what, on this view, is " external " ? If " outside " stands for 
 nothing at all, there is no meaning in the assertion that 
 things do not exist outside our consciousness. To make 
 any beginning of argument we must make outside and 
 inside mean something : and we have just to let them mean 
 what we always mean by them, as when we say, " The cart 
 is outside the house," "The horse is inside (or in) the 
 stable." Now, to say that the horse can be in or inside my 
 consciousness as it is in or inside the stable is impossible for 
 any theorist, whatever be his philosophy. The fact is that 
 the term " inside " is totally irrelevant to the idea of per- 
 ception ; and, that being so, the term " outside " is equally 
 irrelevant. They ought never to have been so used : the 
 verbal dispute before us ought never to have arisen. We 
 said before that it could mean no more than a dispute as to 
 the definition of consciousness : we now see that it can 
 lead to no rational definition whatever. Language, as we 
 agreed, is a clumsy set of instruments at best ; but it is a 
 gratuitous folly to use the demonstrably wrong instruments 
 in an exceptionally difficult analysis. Such disputes, called 
 
 R 2
 
 230 LETTERS ON REASONING Letter XIL 
 
 by the name of metaphysics, perfectly justify the account of 
 it as "hke looking in a dark room for a black hat that isn't 
 there." They are conducted by enthusiasts who really have 
 not mastered the art of reasoning, though they often seem 
 to regard it as their specialty. 
 
 Making the best we can of our admittedly primitive and 
 troublesome tools, let us approach the problem watchfully. 
 And first as to colour. The rational statement is neither 
 that colour exists " only " in or in relation to consciousness, 
 nor that it exists apart from consciousness. But neither, on 
 the other hand, is it rational to say that it aumot exist apart 
 from consciousness. Here, if you have heard something of 
 the forms of scholastic logic, you may be ready to answer 
 that as " A is either B or not B," we must hold either that 
 colour does or that it does not exist apart from conscious- 
 ness. But the formula "A is either B or not B," as we have 
 repeatedly seen, is relevant only where A can conceivably 
 be one or other, and where B is a definite term. The pro- 
 position, " Grass is either happy or not happy," is meaning- 
 less ; and so is " Socrates is either a chord or a discord." It 
 is sufficiently idle to say that A cannot be both B and not B 
 in a case where it cannot be thought as either. Now, in 
 the case before us, we are disputing to begin with as to how 
 A is to be conceived, and any proposition as to A's mode of 
 existence, in order to be intelligible, would have to be 
 preceded by an agreement as to what we are to understand 
 by A. 
 
 If, then, we cannot agree as to what colour is, how can we 
 agree as to how it must be} On the other hand, we should 
 be no further forward if, say, we agree to call it " a relation 
 between subject and object," unless we can conceive of a 
 relation as visually perceptible. " Relation " is a name for (i) 
 the state of consciousness as recognised by ourselves, and 
 (2) for the perceived fact that all things in some degree or 
 sense influence or affect each other ; but though we realise 
 things as "related," the idea of relation is one that accrues
 
 Letter XII. LETTERS ON REASONING 231 
 
 upon and is not identical with the idea of a thing. It is in 
 fact, like causation, one of the constituents of our reasoned 
 idea of existence. To say, again, that colour exists only in 
 or in relation to our consciousness, is not only irrational in 
 view of the meaning of " only " and " in," it is futile as a 
 practical description of our knowledge, for no man believes 
 that the colours we can " see " when our eyes are shut are 
 of the same order of reality as the colours of objects, any 
 more than that our dreams are of the same order of reality as 
 our waking experience. Further, as we noted before, if colour 
 is to be spoken of as "existing only" in relation to con- 
 sciousness, so are the hardness of rocks, the fluidity of water, 
 and all the other attributes of things perceived. Either, 
 then, we extend the irrational formula " exist only " to the 
 whole universe, or we must return to the starting-point 
 of the fact of perception and try another path. 
 
 Shall we say, then, that colour exists in objects not in 
 relation to consciousness ? We agreed, I hope, that we can 
 l)elieve in the past existence of an earth somewhat like ours 
 without any people on it, or vaguely picture to ourselves the 
 aspect of an unpeopled moon ; but on the other hand the 
 moment we think of colour our consciousness is posited, so 
 that, while our words would be meaningless if we said that 
 our consciousness created and annihilated things, it is 
 equally meaningless to say we can think of colour not in 
 relation to consciousness. It amounts to saying how we 
 think of colour when we do not think of it. And for this 
 reason, finally, it is equally meaningless to say that colour 
 cannot exist save as a fact of or in relation to conscious- 
 ness. That proposition stands for no possible conception 
 or perception : it is a process of words calling up no 
 justifying set of ideas. When I say, " A fish cannot swim 
 or live on dry land," I have recollection of the perceived 
 fact that a fish will painfully flounder and speedily die when 
 taken out of the water : I am describing the effect of a real 
 transition from state to state. But in the terms of the case
 
 232 LETTERS ON REASONING Letter XIL 
 
 there is no imaginable transfer of objects from the state of 
 perception to "a state of non-perception." You cannot 
 think of things as ceasing to exist when they cease to be 
 perceived; "ceasing to exist" has no meaning for us save 
 as "ceasing to live''' or "passing from one form to another," 
 as when snow becomes water, or an object is reduced to 
 ashes or gases by burning. 
 
 In fine, as we saw before on other hnes of discussion, 
 every proposition of absolute mode for an infinite category 
 of existences is irrational, as it asserts no cogitable idea. 
 Therefore no such propositions should ever be made : they 
 constitute mere pseudo-philosophy. The true philosophic 
 lesson of the discussion over colour is simply this : that we 
 cognise or infer objective existence under the forms of 
 perception ; and that existence finally means that which we 
 cognise or infer. Having cognised coloured objects, we 
 necessarily re-cognise or recollect them thus so long as our 
 colour-sense or colour-memory subsists ; even as a blind 
 man, having cognised objects by touch and taste alone, must 
 so re-cognise them while he remains as he is. We are told 
 that a man born blind has been led to conceive of a bright 
 colour as " like the sound of a trumpet." There you have 
 dramatically expressed the futility of the notion of conceiv- 
 ing existence otherwise than as we have perceived it. The 
 blind man's effort is relatively rational : he thinks of an 
 unknown sense-perception by analogy with sense-perception ; 
 but the process of argument which ends in afiirming that our 
 act of perception creates, or that our cessation of perception 
 annihilates, the perceived thing, is a relatively irrational 
 because demonstrably self-contradictory attempt to put the 
 aspect of infinite and non-relative perception on a finite and 
 relative perception. 
 
 And this is historically, I believe, the explanation of much 
 of the vain debate we have been considering. Recognising 
 wistfully the transitoriness of all thin^^s, and conceiving 
 change to be the negation of " reality," because it is of the
 
 Letter XII. LETTERS ON REASONING 233 
 
 nature of admitted illusions to be transient, men sought for 
 reality in terms of something intransitory, alternately sur- 
 mising (a) a " substratum " or " materia substantia "^a 
 "something under" the aspects of things which was 
 permanent but imperceptible " matter " — and (d) a " Spirit " 
 which remained the same through all the mutations of the 
 universe. That imagined fundamental "substance" it was 
 that Berkeley rejected as a figment ; and he was perfectly 
 right. A sub-stantia of which the characteristic is to be 
 absolutely imperceptible, and to be no^ that which is per- 
 ceived " over " it — this is just a roundabout way of trying to 
 realise " nothing." You have a physical analogy to the 
 process in the game of grasping your right thumb with your 
 left hand, and then trying to catch the point of said thumb 
 with the right hand — a performance declared by a scientific 
 friend of mine to be the dumb show of all metaphysic. 
 But Berkeley, in turn, must needs continue the search for an 
 unchanging reality in terms of Spirit, and so set up for us 
 the equally nugatory formula of a God in whose " mind " all 
 existence subsists as an idea. All the while, the God-idea 
 is thus the most absolute negation of that unchangingness 
 which the thinkers set out to find. They sought first an 
 antithesis to " appearance " ; and all they attained on that 
 line was an unknown "something that never appears." But 
 in the God-idea they avowedly conceive an Unchanging One 
 as an Infinite Consciousness of All Change — the eternal 
 cognition of change. 
 
 That this mere abstraction of the concept of Change 
 should figure as more satisfying or consolatory to men than 
 the consciousness of Change itself, is due to economic and 
 social conditions. That is to say, economic interests are 
 bound up with theological doctrines ; and doctrines once 
 established find prejudiced defenders ; so that what began 
 as a simple search for truth is maintained as a factitious 
 machinery for justifying or buttressing a religious system.
 
 H^ tu-A^U^ ^ /\JL*^^ VI v/V^uuv ^ 
 
 234 LETTERS ON REASONING Letter XIL 
 
 When the search is sincerely and consistently carried on, it 
 is readily seen that the thought of an Infinite Consciousness 
 of Eternal Change is no refuge from the thought of mutation, 
 being simply an indefinite extension in thought of our own 
 consciousness of change ; and that accordingly no solution 
 of the problem of reality is so to be found or formulated. 
 We must revert rather to our matured perception that on the 
 one hand all things are at varying rates in process of change, 
 and that on the other our consciousness itself is variable and 
 fallacious. 
 
 From the two principles we reach the conclusion that 
 reality, considered as what is contrasted with illusion, is 
 proved by a state of consistency between our perceptions 
 and conceptions at different moments, as tested by repeated 
 comparison ; and that reality considered as contrasted with 
 " appearance " in general is merely an abstraction of our 
 perceptions of existence. " Appearance " is either a synonym 
 for illusion, as when we say that in a certain physiognomy 
 there is the appearance of wisdom without the reality, or a 
 name used as implying that every attribute or aspect of 
 things is mutable — that worlds are reducible to gases, and 
 that nebulae can condense into stars, and so on. In this 
 latter sense, then, the term rationally signifies not " unreality " 
 ( = " lesser degree of reality") or "illusion," but just our 
 knowledge of the mutation of reality. 
 
 One day, walking with you children by the lake in 
 Regent's Park, I sought for a definition of my notion of 
 reality ; and I found one in the proposition that " the 
 measure of reality is degree of relatedness."^ That seemed 
 and still seems to me to express the different aspects of the 
 problem as well as a formula can be made to do it, leaving 
 as it does a verbal loophole for the " transcendental " method. 
 
 - This, I find, squares with Spinoza : — "Quo plus realitatis aut esse 
 unaquaeque res habet, eo plura altributa ei competunt " {Ethica, Pt. i, 
 prop. ix).
 
 Letter XII. LETTERS ON REASONING 235 
 
 We may say if we will that the ultimate reality is the sum 
 of all relations ; and while the Naturalist will recognise that 
 that phrase stands for the cosmos, the " idealist " will 
 doubtless proceed to afifirm a consciousness in which all 
 relations are "contained." Nugatory as that affirmation i.s, 
 it is perhaps less fatal to coherent thinking than the pro- 
 position that reality is " something ^^/^/«^ all appearance " — 
 a form of words which makes reality the unknown symbol of 
 a symbol. 
 
 This, as it happens, is the logical upshot of a favourite 
 formula of some men of science, to the effect that our 
 scientific ideas of the processes of nature are but a " short- 
 hand notation of what actually happens." Either they 
 know the " longhand," the "what actually happens," or they 
 do not. If they do, it is their business to state it. If they 
 do not, they cannot conceivably know that any account of 
 it is a "shorthand notation." They have but made one 
 more attempt to show that " we may know more than we 
 can know " ; even as does Professor Bradley when, by way 
 of preface to a treatise of which the effect is to show 
 that the terms " deity" and " divine " signify illusions, he 
 suggests that some men may have in them more of " the 
 Divine " than others. What his treatise entitles him to 
 say is simply that some men have wider perceptions of 
 relations than others. As against such reversions to illusory 
 terms, our formula may perhaps be not useless. 
 
 We end, then, with the truth of the reality of mutation and 
 the mutation of reality. From that truth we can seem to 
 get away only by means of mock propositions, framed in 
 the interests of beliefs otherwise reached, and in possession 
 of reverence before dialectical philosophy came to champion 
 them. And the mock propositions themselves, on analysis, 
 are found to be but restatements of the problem under the 
 guise of verbal solutions. Depend upon it, then, that you 
 lose none of the more durable joys of life by facing reality
 
 236 LETTERS ON REASONING Letter XII. 
 
 under every aspect, synthetic or analytic, under which we 
 can coherently think it. It is by seeing as largely as may 
 be the objective truth of things that you can best make your 
 inner life stable. The more widely you are related to things 
 the more real is your own existence, tried by the tests of 
 your reason.
 
 LETTER XIII. 
 
 'When I look back over these dozen letters, I am set 
 wondering how far my plan may have helped you to use 
 your reason aright. I am fain to trust that, even if I 
 should have at times argued amiss, the ramble you have 
 here made with me (so I like to fancy it) will have prepared 
 you to follow a formal treatise on logic more alertly than 
 you otherwise might, and may further have made you more 
 watchful over the making of your opinions in general. I 
 can conceive that the processes of argument through which 
 we have gone could be reduced to symbolic form, if they 
 have not all been already. Symbolic logic, they tell me, 
 has been in great measure recast in the past ten or twenty 
 years ; and perhaps the fallacies I have been trying to 
 expose to you in the last two or three letters have already 
 been symbolically disposed of. But when I talked recently 
 with a master of the newer symbolical methods, I found 
 that he and I differed unexpectedly and surprisingly on such 
 a point of moral judgment as the view we ought to take of 
 suicide. 
 
 Now, as I tried to show you in one of my letters, and 
 have taken for granted in the others, all reasoning, all logic, 
 is finally a testing of the consistency of our beliefs. The 
 truth of a given proposition is finally to be proved only by 
 showing that it is not inconsistent with any other proposition 
 which we profess to hold as certain. That is what truth 
 means : it has ultimately no other meaning. Of course we 
 cannot in practice compare any one beUef with all our other 
 beliefs ; and it would not be worth while, many having 
 no obvious correlations ; but until a new opinion has been 
 tested by comparison with older ones which have a clear
 
 238 LETTERS ON REASONING Letter XIIL 
 
 relation to it, we have it by a precarious tenure. And 
 questions of right and wrong, as we saw, come ultimately 
 under the same logical and psychological law with questions 
 of true and false. The question as to how we ought to 
 speak of suicide, then, is to be solved in terms of our whole 
 conceptions of moral causation. It is not a very important 
 one, I fancy : but it may not be quite a waste of time to 
 consider it here, by way of one more exercise in the 
 coordination of opinions. 
 
 If you agree with me in recognising the natural determi- 
 nation of all conduct, and at the same time the inevitable- 
 ness of praise and blame, you will come to the problem with 
 a certain philosophic bias, or pre-conception ; for those 
 opinions involve others. Let this bias be duly avowed, so 
 as to keep the discussion lucid. Regarding a man's character 
 and actions as the outcome of his physical structure, train- 
 ing, and circumstances, we regard, say, his anti-social acts as 
 something we have a right to protect ourselves against, and 
 consequently to denounce. But we also see the futility of 
 going on denouncing the wrong-doer when we have protected 
 ourselves by shutting him up : in that position, we having 
 given up the hope of controlling him by our blame, he is to 
 us very much as a mad person in confinement. We cannot 
 love him, but we can think of him with. compassion. 
 
 Now, that seems to me to be the consistent attitude in the 
 case of the suicide. Let us put it either that he killed 
 himself while "sane" or that he did it while "mad." In 
 the latter case there can be no reasonable dispute : if mad- 
 ness is a ground for our compassion, the act of self-destruc- 
 tion by the madman is so. But let us say the suicide was 
 sane, as I think a good many suicides must be held to be, 
 unless we " beg the question " by defining suicide as an 
 insane act. Now, the man who has slain himself is so 
 perfectly disposed of that, though he may have caused some 
 of us a very painful shock by his act, he stands to us as does 
 the confined criminal, " only more so," as the slang goes.
 
 Letter XIII. LETTERS ON REASONING 239 
 
 It is doubly idle to denounce him when, so to speak, there 
 is no " him " left to denounce. 
 
 But, it may be answered, the act of suicide is anti-social, 
 cowardly, evasive, and ought to be so spoken of, just as we 
 denounce any cowardly or anti-social act which does not end 
 in self-destruction, both because we detest it and because 
 the expression of our detestation is so far the creation of 
 a social code. Well, I grant that by holding up cowardice 
 to contempt we prevent each other in some measure from 
 playing the coward ; but when cowardice stands helplessly 
 confessed — when the coward avows his incurable weakness — 
 is it not in decency and consistency a subject for compassion 
 rather than for further vituperation ? Ls it not morally on a 
 par with dipsomania and kleptomania ? 
 
 There remains the residual anti-social aspect of the act — 
 the possible effect of it in laying heavier burdens upon others. 
 On this score, it may be urged, it is rightly to be held up to 
 odium of all kinds, and all the more because the suicide is 
 dead, and cannot suffer from our censure. I should answer, 
 however, that this is a sophistication of our moral feeling ; 
 that it evades the question as to the actuality of the imputed 
 cowardice, and begs the question as to results ; and that, as 
 a matter of fact, some who loved the suicide may suffer very 
 severely from our blame. We must therefore press the 
 analysis further. 
 
 It seems to me, on a full survey, that an act which is 
 intended to end the doer, leaving him neither pleasures nor 
 pains, is to be conceived differently from an act of cowardice 
 or evasion or selfishness by which the doer lives and hopes 
 to profit. It does not appear to me to be either logically or 
 psychologically describable as cowardly. A coward will hide 
 or run away in battle ; but did any coward ever shoot him- 
 self in battle to avoid being shot ? Of two detected criminals, 
 one shoots himself to avoid the disgrace of exposure or the 
 pain of imprisonment ; the other thinks of shooting himself, 
 but cannot nerve himself to the act. It is rather odd to call
 
 240 LETTERS ON REASONING Letter XIII. 
 
 both cowards ; and we do not solve the anomaly by saying 
 the first is a moral and the second a physical coward. The 
 second may very well be both, in daily life. You might 
 allow this, and then pronounce the suicide a moral but not 
 a physical coward. But still the spirit of consistency, so to 
 speak, would demur. If a man may continue to be a moral 
 coward after detection and exposure, it seems idle to say 
 that the suicide is a ivorse moral coward. On the other 
 hand, his act may stand for a passion of self-contempt and 
 self-detestation of which the average moral coward is 
 incapable. When, for instance, we read in some romance 
 of one member of a proud " old family " calling on another 
 who has dishonoured himself to spare the family a worse 
 disgrace by committing suicide, we cannot very well say the 
 disgraced one is a coward for consenting. 
 
 I ought perhaps to explain here, for the sake of other 
 readers if not for yours, that I have not the faintest idea 
 of committing suicide ; and that I should be horrified at the 
 thought of either of you doing so. Your mother and I do 
 our best to give you healthy bodies to begin with ; and if we 
 can teach you also to act and reason well there will be small 
 risk of your ever seeking to end your lives. The question 
 is one that may be and ought to be considered disinterestedly, 
 on the assumption that none of those of us who discuss 
 it is at all disposed to commit "the rash act," as the 
 journalists used to call it in the days before they cultivated 
 humour and a hand-made style. 
 
 The rational verdict, then, seems to me to be this, that 
 suicide is an "xoXsiii getieris^ to which the terms of opprobrium 
 passed upon the acts of non-suicides are inapplicable. We 
 are dealing here with another of those confusions of thought 
 which consist in applying to certain categories terms of mode 
 which have no relevance to them. And if the practical 
 question be held to be as to how our criticism of the act may 
 affect the action of other men, it will be found that the 
 attitude of vituperation is in many cases inconsistent.
 
 Letter XIII. LETTERS ON REASONING 241 
 
 Suppose, for instance, that a man has committed a fraud, 
 for which, on detection, he will be subjected to a very long 
 imprisonment. Knowing that in prison he can do no good 
 to anybody, and will have no joy in life, he forestalls 
 us by shooting himself. If we thereupon call him a ""'^^ 
 coward, are we not in consistency bound to give praise 
 for courage to the criminal who in a similar position does 
 not shoot himself? Yet who ever does award such 
 praise ? 
 
 My friend the logician pointed to the case of the man who 
 simply finds life intolerably hard or joyless, and kills himself 
 for relief. He ought, says my friend, rather to bear up and 
 help other people to bear their troubles. Far be it from us 
 to make light in the least degree of the prescription to help 
 the heavy-laden and comfort the life-weary. But I cannot 
 think that a life-weary man is likely to be able, do what he 
 will, to brighten life for others ; and I really cannot find it 
 in my heart to call him ill names. When his life-weariness 
 is due to a desperate and incurable disease, and no one 
 suffers any intense shock from his suicide, no untrammelled 
 reasoner, I think, will denounce him : only on a dogmatic 
 and authoritarian view of " duty " can he be condemned. 
 But even when he has merely ceased to take any satisfaction 
 in life, and thus without any physical torment suffers the 
 sorrows of life without getting any of its joys, the consistently 
 rational feeling towards the act of suicide on his part seems 
 to me to be one of pure compassion. 
 
 Strictly speaking, indeed, such a suicide is in all likelihood 
 the expression of physical decay of some kind, unless it be 
 the result of some terrible bereavement. In the latter case, 
 indeed, the springs of life are certainly affected in a physical 
 way. And while every one of us would feel it a duty to try 
 to help such a bereaved person to live, knowing as we do 
 that he or she may recover enough interest in life to be 
 useful to others and to be ultimately capable of a measure 
 of happiness, we should in consistency have to admit that
 
 242 LETTERS ON REASONING Letter XIIL 
 
 suicide committed in the stage of utter despair is morally 
 analogous to a suicide committed in a state of fever. 
 
 Suicide in general, the statisticians tell us, is very much 
 determined by temperature; and when very high temperatures 
 coincide with periods of commercial depression and loss, 
 the figures mount exceptionally. This should set us upon 
 thinking that suicide might be made very much less frequent 
 by a bettering of our social system on the one hand, and the 
 more rational management of our bodies on the other. And 
 I am quite sure we should be much better employed in 
 striving for those consummations than in writing " coward '' 
 against the name of any suicide. If deterrence be our 
 object, are we doing anything in that way to attain it? 
 Nothing at all, I should say. Mr. Spencer tells of a certain 
 Sir Peter Laurie who expressed a determination to " put 
 down suicide " ; and the accepted view is that he was not 
 very wise. A policy of rhetoric, I fancy, will fare no better. 
 When a man has come close to the point of committing 
 suicide, he is hardly likely to be affected by the thought that 
 anybody will say " coward " over his grave. While he can 
 wince at such a prospect, liis impulse to suicide cannot be 
 very strong. 
 
 I was once put in a very painful position by an appeal to 
 my feelings on this point. An esteemed friend of mine, 
 who was slowly dying of a painful and incurable disease, 
 and was consciously a burden to others, asked me for help 
 towards committing suicide. The only conceivable way 
 was to try to procure poison ; and if I did that I should 
 have been liable to trial and punishment for murder. I 
 confess I had not the courage for that ; and I felt also that 
 the law is quite right in refusing to permit a private act of 
 the kind,^ seeing that the pretence of such a motive, if 
 
 ' In one memorable case, where there was no room for doubt as to 
 the moral innocence of one who aided another to die, the coroner's 
 jury, I am glad to say, recognised that the act ought not to be punished.
 
 Letter XIII. LETTERS ON REASONING 243 
 
 recognised, would certainly be used to screen acts of 
 absolute murder. But I conceive that a cjuite rationally 
 ruled society would provide a public means of euthanasia 
 for the relief of sufferers like my poor friend, taking ample 
 precautions against acts of mere passion on the part of 
 sufferers, or of fraud on the part of any who may wish their 
 death. 
 
 If the reasonableness of this be disputed, I should ask 
 how it is that no law nowadays punishes a man for self- 
 mutilation. It is surprising to me, I confess, to read that 
 in foreign countries where conscription rules it still is, or at 
 least recently was, possible for a youth to escape military 
 service by cutting off his thumb : one would suppose that 
 such an act would be regarded on any legal theory as an act 
 of fraudulent evasion against the State. But at least no law 
 punishes a man for self-wounding in general as it does for 
 wounding another : it is only when the self-injuring person 
 is supposed to be aiming at suicide that he is liable to 
 punishment. The conclusion seems to be that it is solely 
 on the score of religious veto that the act of suicide is 
 in Christian countries regarded as " criminal " and those 
 failing in the attempt are punished — as if to remind them 
 that it is a mistake to do such things by halves. If the veto 
 and the penalty were rational, they would attach to acts of 
 self-flagellation and of such fasting as tends to shorten life. 
 But here religion prescribes the course of action it elsewhere 
 denounces : the suicide is buried at cross-roads " with a stake 
 in his heart " ; while the monk, prematurely worn out by 
 self-castigation and self-mortification, is revered as a saint. 
 Doubtless there is a partially unifying point of view in the 
 sentiment that the suicide would not bear trouble, and that 
 the saint would ; but the germ of that sentiment is the 
 fantasy that by inventing sufferings for oneself one enhances 
 one's virtue ; and unless jurisprudence is to take that point 
 of view it had better eliminate from its precepts the tradi- 
 tional judgment upon suicide, and that penalty for failure 
 
 s
 
 244 LETTERS ON REASONING Letter XTII 
 
 which is in effect a broad hint to the unsuccessful to be 
 more thoroughgoing next time. 
 
 Now, to return to my friend the master of symbolical 
 logic, this problem is one which even his improved symbolic 
 notation does not appear to solve for him to my satisfaction ; 
 hence I am led to persist in my notion that the reasoning 
 faculty is usefully to be exercised and disciplined by non- 
 symbolical and non-scholastic methods. I shall not here 
 attempt, therefore, to carry out an idea of my own for 
 formulating abstractly all the different forms of fallacy ; 
 because, even if I could do it well (which I doubt), there 
 would probably be more use in my dealing with the subject 
 as I am doing. My view is that the logicality {i.e. the con- 
 sistency) of moral judgments is to be tested by a regression 
 to the bases of morals in life ; and the logicality of proposi- 
 tions concerning existence by a regression to the bases of 
 perception and conception. Symbols, I suspect, however 
 perfected, will serve only as the " shorthand notation " of 
 such processes of reasoning ; and longhand must come 
 before shorthand. 
 
 I conclude, then, as before, that a good reasoner is to be 
 made as a healthy physique is made, by doing things in 
 general, not by learning to split hairs. The beginnings are 
 made in childhood, in daily life ; and fresh steps are taken 
 every time you think out a problem in arithmetic, a theory 
 in science, an argument in politics, or an interpretation in 
 history. Right reasoning is a deepening of such experience. 
 A clear view of the bases of morals in life is reached by 
 thinking loyally about life ; and so with the bases of percep- 
 tion and conception. Not that I want you to give your 
 youth to analytical psychology : that does not come naturally 
 to youth. I mean simply that when you are faced by pro- 
 positions from the philosophy of perception and conception 
 you will do well never to be put off with A — B formulas as 
 being authoritative, but to look earnestly into your own 
 mental processes.
 
 Letter XIII. LETTERS ON REASONING 245 
 
 Beyond that, it is hard to say what species of intellectual 
 interest is best fitted to develop the reflective judgment. 
 Certainly there must be scientific study if the development 
 is to be at all complete. A scientific mind, broadly speaking, 
 is a mind watchful of the implications of its knowledge ; 
 and even if we somewhat narrowly defined a logical mind 
 as simply one watchful of the implications of its language, it 
 would be clear that the one faculty reacts on the other. A 
 thoroughly scientific mind, in fact, is a thoroughly logical 
 mind, as I understand science and logic. But it is hardly 
 possible even for a man of science, so called, to have a 
 thoroughly logical mind, were it only because the kinds of 
 perception involved in the physical sciences are so different 
 from those involved in the moral or social sciences. 
 
 I have sometimes wished, for my own part, that my life's 
 work lay in natural science, because of the comparative 
 stability, so to speak, of the subject matter, and the quasi- 
 absoluteness of the result. In a scientific research you may 
 look to an ultimate agreement if your view be right and your 
 demonstration clear ; and the question between you and 
 your opponent or instructor is now rarely one of unscientific 
 pre-conception. Fresh generalisation, indeed, is usually a 
 matter of philosophical bias for the time being ; but so long 
 as theology does not enter, there is a good chance of agree- 
 ment. The phenomena of nature " stay put," so to speak : 
 they are there to experiment upon, to be seen, to be measured, • 
 to be dispassionately known. But in the moral sciences you 
 are, as it were, in the matter you are studying : you are 
 analysing passions and proclivities, challenging prejudices 
 and vested interests ; and the people in whom the interests 
 and prejudices and proclivities vest are those to be enlight- 
 ened. The truth of the thesis is no security for its accept- 
 ance : even of those who will take the trouble to investigate, 
 many do but restate their proclivities in the form of con- 
 clusions. And when it comes to urging new lines of action, 
 you have to persuade passions rather than judgments. Thus 
 
 s 2
 
 246 LETTERS ON REASONING Leller XIII. 
 
 there is an inherent elusiveness and incertitude in the 
 problem ; and in comparison those of the chemist and the 
 naturahst seem tangible and soluble. 
 
 Yes, but there is a two-fold answer. For one thing, we 
 simply cannot help aiming at a sociology, a science of 
 organised humanity — those of us, that is, who care about 
 making the world better, or even keeping it from growing 
 worse; and if some of us did not cherish that interest, it 
 would be a bad place for the men of science. If, then, the 
 inquiry is to be undertaken at all, it must needs be handled, 
 by some at least, scientifically. On the other hand, the 
 problems of physical science are insoluble at the other end 
 in a sense in which those of the moral sciences are not, even 
 at their most baffling point. Even in detail, the natural 
 sciences have plenty of enigmas. An expert in botany has 
 just been telling me that a certain scarlet fungus or quasi- 
 mushroom which grows in pine woods is in this country 
 quite poisonous, while in Russia the same plant, growing in 
 a similar environment, is harmless and edible. The facts, 
 he says, are quite certain. If so, the botanist has his cruces, 
 and they seem to implicate the chemist. But it is in respect 
 of the ultimate problems of physics that science is finally 
 the statement of an enigma, and to those ultimate problems 
 it is always luring us ; while the problem of sociology, which 
 is the right adjustment of life in view of experience, is 
 relatively definite and manageable. 
 
 All the same, to return to a previous topic, the right study 
 of the moral sciences, of which sociology should be the 
 integration, is a matter for expertise like another ; and in 
 view of its profound importance it is essential to realise this. 
 It will not avail to come from the physical sciences and say 
 that their method is the key to this. In the abstract all 
 method is the same : it consists in induction and deduction, 
 with the help of hypothesis ; but merely to see facts is a 
 difficult accomplishment ; and the faculty of relevant hypo- 
 thesis is one that grows only out of a vital familiarity with
 
 Letter XIII. LETTERS ON REASONING 247 
 
 the matter in hand. Hence the grotesque inadequacy of 
 the generalisations made by many experts in the physical 
 sciences when they come to the moral. Each expert knows 
 that his own business needs a long special study ; but each 
 is apt to suppose that the ultimate business of all, the under- 
 standing of organised man, is within the reach of his 
 unspecialised common sense. 
 
 In a laudable little book by Mr. Lowes Dickinson, entitled 
 The Meaning of Good} a man of science is represented as 
 undertaking to state the whole problem of ethics in terms 
 of generalisation as to the drift of Nature. You have only 
 to ascertain, he says, which way Nature is tending on the 
 whole in the matter of human and social evolution, and to 
 seek to cooperate with that tendency. Then you will be in 
 harmony with Nature : otherwise you will be vainly fighting 
 against her. I do not know whether any man of science 
 has actually so expressed himself; but from what I have 
 observed of the sociology even of great naturalists, as 
 Darwin, I fear that something of the kind has been said. 
 Darwin, for instance, talked to the same effect when he 
 argued that because civilisation has progressed by way of 
 the struggle for existence, it will fall away if men bethink 
 themselves so to limit their families that the struggle shall 
 be made easier.^ The nature of the fallacy in these propo- 
 sitions is worth noting. 
 
 Broadly speaking, it consists in forgetting that everything 
 is Nature ; that men do not cease to be part of Nature when 
 they acquire science ; and that what they want to do is just 
 as much a cosmic tendency as anything else. Even, indeed, 
 if we agree to make " Nature " mean all the cosmic forces 
 and phenomena except ourselves and those we determine, 
 the fallacy is no less gross. By that definition Nature is the 
 sum of the irrational forces we are to exploit ; and to look 
 
 ^ Glasgow : Maclehose & Sons, 1901. 
 
 - See his letter cited in Mrs. Besant's Autobiographical Sketches, 
 first edition, 1885, p. 136.
 
 248 LETTERS ON REASONING Letter XIIL 
 
 for the revelation of rational purpose on precisely the non- 
 rational side of the equation is to stultify the assumption at 
 the start. But on Darwin's view we are part of Nature ; 
 only we must take care not to use our reason beyond a 
 certain point. The same idea, I regret to say, is implicit in 
 some of the reasoning of Mr. Spencer on politics. ^ Here, 
 then, we see the two men who have done most in the past 
 generation to coordinate our ideas of Nature as a whole, 
 falling into a bottomless fallacy, so to speak, when they 
 deal with the application of their knowledge to human 
 conduct. 
 
 Logically, the fallacy may be exposed by the tests of 
 consistency. To begin with, it is reason, once more, that 
 offers the warning against relying on reason : both Spencer 
 and Darwin have to posit their judgment of Nature in telling 
 us to follow Nature and not our judgment. Nor do they 
 in general at all recognise the principle that any one natural 
 tendency is to be endorsed simply because it is natural. 
 Darwin knew very well that social and moral evolution had 
 in many ways modified the struggle for existence : penal 
 law was from the beginning such a modification : but he 
 never dreamt of proposing to abolish penal law any more 
 than of suggesting a return to cannibalism. Mr. Spencer 
 knows perfectly well that militarism is an expression of a 
 very powerful tendency in Nature ; but his whole moral 
 influence has been rightly used to discredit and resist 
 militarism. When, then, either evolutionist fell back on the 
 mere past course of Nature as creating a model for our future 
 imitation, he was resorting to a fallacy very much in favour 
 in the opposite camp, for the common reason, that he could 
 not put a better face on his case. 
 
 The sense in which we must " follow Nature " is sufficiently 
 obvious. We are part of Nature, and as such determined 
 by a thousand conditions which we cannot evade. The 
 
 ' See his Study of Sociology, ed. 1873, pp. 351, 395-7, 401-3; and 
 his Man versus the State, p. 64.
 
 Letter XIII. LETTERS ON REASONING 249 
 
 nature and pressure of those conditions it is our first business 
 . to understand and adapt ourselves to ; and the first principle 
 of sane living is the acceptance of the unalterable. But 
 when we as a result of our past evolution acquire aspirations 
 towards a better life, and partly see our way to such 
 betterment, it is an inversion and not an application of 
 the wisdom of experience to say. We have reached our 
 present state through the operation of certain processes in 
 themselves evil, therefore we must conserve those processes 
 in order to conserve what good we have. Darwin doubtless 
 thought that his prescription was on a par with the pre- 
 scription against tight corsets — a counsel founded on 
 experience. But this it clearly was not. Tight corsets have 
 been tried, and it is found that people are healthier without 
 than with them. Now, in the terms of the case the 
 experiment of modifying the struggle for existence by 
 limiting families had not been tried ; and Darwin's position 
 logically consisted in warning us against trying it because it 
 had never been tried. 
 
 The case lies in a nutshell thus : All lines of action are 
 natural : all new departures, human or sub-human, may for 
 the purposes of this inquiry be conceived as Nature's 
 experiments. Those we observe to fail we take care not to 
 try again — that is, any fresh attempts in the same direction 
 seek by new devices to guard against the danger experienced. 
 But to say of a proposed social experiment that it ought not 
 to be tried because Nature did not work in that way in the 
 past, is logically to veto all social change whatever, and, 
 indeed, to deny (what is at the same time affirmed) that 
 there is change in Nature. If a proposed experiment can 
 be shown to be fiot a novelty, but a repetition of an attempt 
 that has actually failed, then it suffices to point to that fact 
 in order to discredit it, and there is no need to talk of the 
 method of Nature. If Darwin knew of such a past case in 
 point he would have cited it. 
 
 So with Mr. Spencer's citation of the method of Nature.
 
 2SO LETTERS ON REASONING Letter XIIL 
 
 Where he can actually show that a given legislative 
 experiment has entirely failed in its aim he does so, and 
 finds in that a sufficient argument against persisting in the 
 course in question. It does not occur to him here to note 
 that past Nature is what has been done. Only when men 
 propose a course that he thinks bad, but can ?iot show to be 
 a mere repetition of past error, does he bethink himself of 
 telling us to follow what he calls the course of Nature. 
 The answer is that Nature has been manifested as much in 
 failure as in success. Pestilence, earthquake, decadence, 
 are aspects of Nature or nothing. In short, the formula 
 " follow Nature " is strictly meaningless. Either it means : 
 " Profit by experience of past miscarriages," in which case it 
 had much better be put in that way ; or it stands for the 
 inability to use that argument, and merely signifies, " Avoid 
 doing this, because it has not been done." 
 
 I need hardly point out to you that the argument of the 
 man of science in Mr. Dickinson's dialogue falls by the same 
 reasoning. We cannot find a code of conduct in the mere 
 process of Nature, because all kinds of conduct go on in 
 Nature ; and to say that the kind of conduct which is for the 
 moment succeeding is Nature's plan, amounts to saying 
 either that there is no change in Nature or that whatever is 
 done is right. The pretence of finding the total plan of the 
 Cosmos is a fatuity : we can at best discern how evolution 
 is going with a given State or society or cause or ideal 
 relatively to us. Suppose then that we had been Romans of 
 the period of decadence and had discerned the decadence, 
 would it have been the rational course to seek to promote 
 it ? On that view we should try to spread the small-pox 
 when we see it about, and, when ill, try to die. A famous 
 personage of my day, the late Mr. Rhodes, laid down the 
 doctrine under notice, in an argument to the effect that by 
 looking to see how political movement and expansion are 
 actually going we may divine how " (iod " wants things to 
 go, if there be a ^^God." And Mr. Rhodes professed to
 
 Letter XIII. LETTERS ON REASONING 251 
 
 shape his course by that light. But as it happened he died 
 in the prime of life from heart disease ; and on his own 
 principles the proper inference is that God, alias the Great 
 Perhaps, wants us to superinduce heart disease in politicians 
 of Mr. Rhodes's way of thinking. 
 
 I may seem to you to be trifling when I spend time in 
 exposing such absurdities of doctrine : indeed, I hope that 
 your generation will be rational enough to make the 
 discussion a superfluity. But I can assure you that many 
 well-meaning people in my day have thought, or at least 
 talked, in this fashion ; and, as we have seen, some of our 
 profoundest naturalists have at times fallen into ways of 
 reasoning which logically come to the same thing. 
 
 It is all rather discouraging ; but in life you must lay your 
 account with discouragement ; and I need hardly counsel 
 you to remember that even failure to realise a given ideal, 
 or the discovery that the cause in which you believe is not 
 succeeding, is not a proof that the ideal or the cause is 
 wrong. The great line of Lucan, 
 
 Vidrix causa diis placuit, sed victa Catoiii, 
 
 translated in terms of the Cosmos of reason, is as good a 
 watchword as any, in the end, not merely because it can 
 serve to turn defeat into victory in concrete fact, but because 
 it lifts us to a height from which we can see that our actual | 
 defeat may prepare the victory of those to come. But even 
 that expectation is not necessary to a consistent rationalism. 
 On this last line of reasoning we come once more to the i 
 ennobling conclusion that, as we are " Nature," so our 
 weighed and tested ideal has for us the sanction of the i 
 whole order of things. For us, it is Nature's last word, 
 her newest message. If it should bear the tests of later 
 knowledge and wisdom, it is well in exaisis ; if not, as Mr. 
 Spencer would say, it is still well, though not so well.^ What 
 would be wholly pitiable would be our consenting to let the 
 ' See First Principles, end of Part I.
 
 252 LETTERS ON REASONING Letter XIIL 
 
 knave and the fool, the bigot and the barbarian, have the 
 advantage of us in daring to think that Nature was for them 
 because we dared not think she was for us. And so, with 
 Goethe, Wir heissen Euch, hoffen ! 
 
 One of the ways in which the less rational seek to quell 
 the confidence of the more rational is by insisting that 
 reason will never enable men to transcend their lower 
 instincts — that this can be accomplished only in some 
 rapture of obedience tc a command coming from an adored 
 figure ; and the figure commonly hailed among us still by 
 such reasoners is the gospel Jesus. They speak both 
 ignorantly and heedlessly, knowing apparently nothing of 
 the lessons of pre-Christian history, and disregarding the 
 immense mass of facts in their own every-day life which 
 confute their claim. When you read this, I trust, you will 
 already have read the story told in Plutarch's Lives of how 
 Lykurgus, the legendary Spartan legislator, forgave and 
 reformed the brutal young aristocrat who, detesting one of 
 his proposed laws, joined a party which attacked him in the 
 street, and struck him a blow that destroyed one of his 
 eyes. The indignant citizens took the young man and gave 
 him to Lykurgus to slay or enslave as he would. Lykurgus 
 took him home and did neither, but, while tending his own 
 wound, rationally instructed him, and so won his affectionate 
 admiration. Then, according to one version of the tale, 
 Lykurgus took the young man to the people and said : 
 "You gave me a bad citizen ; I give you back a good one." 
 There is no finer tale in the gospels than that ; and it is told 
 of a " heathen," in a half-civilised age. It shames most 
 Christians and most civilised men, simply because it is such 
 a perfect vindication of reason. 
 
 You will doubtless hear, further, from men of physical 
 science and from men of no science, that it is vain to seek 
 to better the world by reason, and therefore vain to seek to 
 compass a science of organised man. You are likely to hear, 
 I fear, many asseverations of the revived doctrine of force.
 
 Letter XIII. LETTERS ON REASONING 253 
 
 whether to the end of affirming the supremacy of the indivi- 
 dual vvill-to-Hve or to that of promoting " ultimate human per- 
 fection " by tyranny and slaughter. But these fulminations, 
 you will probably find, do not as a rule come from men who 
 know much about history in detail. If they could steep them- 
 selves for a day in the consciousness of impotence which pos- 
 sessed many an ancient autocrat, they would see the problem 
 of conduct in a new light ; and if they knew a little more of 
 what has been actually done in the world by wise persuasion, 
 and of what has been undone by violence, they would per- 
 haps reconsider their simple creed of brute force. 
 
 Two generations ago, when there existed among the 
 Khonds of India the practices of human sacrifice and 
 female infanticide, there came to that people a Scotch 
 officer named Macpherson, a man of intellectual tastes and 
 training, who had entered the service of the East India 
 Company because his weak eyesight shut him out from more 
 congenial careers. By the exercise of tact and rational 
 persuasion he succeeded in turning the bulk of Khonds, of 
 their own will, to an abandonment of both infanticide and 
 sacrifice, never using the arm of the civil power where opinion 
 was concerned. "A conquest of force," he wrote, "is no 
 conquest at all. ^Vhenever I heard that any man of 
 consequence was spreading opinions opposed to mine, I 
 called a council, and prayed him to state his views there. 
 He did so ; and I could always either convince him or make 
 him confess himself silenced — a great triumph for my 
 partisans." ^ 
 
 In the matter of female infanticide, he found that the 
 tribes chiefly implicated justified the practice first by an 
 alleged injunction of their Deity, secondly by asserting that 
 it increased the male births, and thirdly by pleading the 
 trouble caused by the frequent separations and divorces 
 
 1 Memorials of Service in India. From the Correspondence of the 
 late Major Samuel C. Macpherson, C. B. Edited by his brother, 
 William Macpherson. London : Murray, 1865. P. 214, note.
 
 254 LETTERS ON REASONING Letter XIIL 
 
 that occurred even when women were few. To the last 
 argument he quietly replied that 
 
 " Infanticide produces scarcity of women, which raises marriage 
 payments so high that tribes are easily induced to contest their adjust- 
 ment when dissolutions of the tie occur ; while these dissolutions are 
 plainly promoted by that scarcity, which prevents every man from having 
 a wife. On the cessation of infanticide, women would become] abun- 
 dant, and the marriage payment would become small. Every man would 
 have a wife in these districts, as elsewhere ; women would have less 
 power to change, and when they did there would be no difficulty in 
 making the requisite adjustment of property." ^ 
 
 To the argument from divine injunction he answered that 
 the alleged injunction, when examined, was found to be a 
 permission, given in the conditional case of any tribe finding 
 themselves unable to " manage " their women ; and such 
 inability, he subtly suggested, would mean incapacity for the 
 normal duties of manhood. At the same time he assured 
 them that, if they reckoned up, they would find that no more 
 male children were born when female children were killed. 
 Thus, by appealing at once to their reason, their pride, and 
 their faculty for rational doubt, he gained his end. And so 
 with the practice of annual human sacrifice, which among 
 certain tribes was regarded as the highest religious duty. In 
 his own words : — 
 
 " I have invariably appealed directly to the clear reason and the strong 
 affections of this natural and truthful people, avoiding the least offence 
 to their pride, or wound to their self-love — the awakening to hostility 
 of any sentiment or prejudice or passion which I could not control. I 
 have thence, in the first instance, denounced neither of the practices 
 which I have laboured to abolish as a crime : I have but arraigned them 
 as deplorable errors, in which many portions of mankind, including our 
 forefathers, have participated, but from which they have been successively 
 delivered, elevated by their own reason and experience, or by those of 
 others, as we desire to elevate the Khonds." 
 
 In the words of his brother : — 
 
 "The result was obtained by making the abolition a Khond move- 
 ment ; by sapping the religious conviction of the necessity of human 
 
 ' /</., p. 222.
 
 Letter XIII. LETTERS ON REASONING 255 
 
 sacrifice, and showing that all the material interests of the tribes, and 
 those which had most weight with the best men, would be promoted by 
 abolition. No doubt it would have been easy to proclaim — easier than 
 to execute — a crusade against these devil-worshippers and murderers ; 
 but (besides being inaccessible) they were men, and had human reasons 
 and motives for their conduct. The question was, by what inducement 
 could they be led to desist ? and to solve this, of course, it was necessary 
 in the first place to learn what were the reasons and motives which 
 were to be overcome. ... By continued and patient observation of the 
 people — in the forests, in the prisons, under circumstances the most 
 various — some knowledge of their inner life was at length painfully won. 
 . . . The effect was produced, not by introducing any new theory, but 
 by building on the old foundation ; by calling into healthy action the 
 principles already recognised among them, and suited to their character ; 
 by making them feel that the change was their own experiment for 
 their own interest — not the experiment of the Government for ends of 
 its own." ^ 
 
 The malpractices, you see, had been " natural " : they 
 were simply the results of bad reasoning ; but doubtless 
 many Khonds felt, like Darwin, that they must adhere to 
 the course on which " Nature " had hitherto proceeded. 
 Macpherson, on the other hand, showed them where they 
 reasoned amiss, and so quietly persuaded them to begin a 
 great step in civilisation, which, once accomplished, has 
 never been retraced. 
 
 I almost hesitate to say to you how much more I esteem 
 this man and his act than the men of force and the kind of 
 conquests they effect. I can but hope that you will come 
 independently to share my opinion. In my day, Macpherson's 
 doctrine appears to be in little favour : it is the doctrine of 
 domination and violence that wins most ears ; and the tran- 
 scendentalists, I notice, are pretty much on that side. 
 Britons, feeling themselves to be "educated," are less easily 
 won to reason than were the Khonds : the conceit of race 
 and the conceit of personal sagacity in combination make a 
 rather bad soil for rational persuasions. Macpherson, as it 
 happened, was not one of those who suppose that without 
 
 ' Id., pp. 214-216.
 
 256 LETTERS ON REASONING Letter XIIL 
 
 taking any pains they can understand the problems of 
 society : he took, as we have seen, endless pains to under- 
 stand the mental life of a barbarous people ; and " he used 
 to say that it was from the study of Guizot that he had 
 
 J learnt how to reclaim the Khonds." ^ I fear that the average 
 representative of the empire today is not given to seeking in 
 the study of sociology for light on the problems of adminis- 
 tration ; and the prospects of civilisation are none the 
 
 I more hopeful. You v/ill perhaps be able to judge, when you 
 read these pages, what have been the relative gains from 
 Macpherson's method and from the other. But what it speci- 
 ally concerns me to urge upon you here is that, if I am 
 right as to the superiority of his ideal, the difference 
 between that and the other is at bottom the difference 
 between good and bad reasoning. And that good reasoning 
 will conduce to good action is quod erat demonstrandum. 
 
 ^ Work cited, p. 351.
 
 INDEX 
 
 Abacus, Jevons's, xxiv 
 
 Action and doubt. 28 
 
 Aesthetic thinkini^, 15 
 
 Animals, doubt in, 18 ; humane 
 
 treatment of, 94 
 Appearance and reaHty, 217 sq. 
 Aram, Eugene, 181 
 Argument against oneself, gain 
 
 from, 39 
 Aristotle, xxii, 178, 179 
 Arminius and Arminianism, 144, 
 
 150, 153.. f6o 
 Arnauld, xvii-xix, xxiii, xxvi 
 Arthur, King, 42 
 Astrology, 99, 139 
 Augustine, 96, 141-3 
 Authority, value of, 47-59 
 
 Bacon, Francis, xviii-xix, xxvi, 91 
 
 Roger, xix 
 
 Bailey, S., 59 
 
 Bain, xix-xxii, xxv, xxviii, 19, 82, 
 
 149, note 
 Balfour, A. J., 164 
 Bayle, 94 
 Beccaria, 94-97 
 Belief, when culpable, 23, 36 ; 
 
 nature of, 52, 54, 79 sq. 
 Bentham, G., xx, xxii, 137 
 Berkeley, 233 
 
 Bias, 30, 152, 165, 166, 171, 185 
 Blasphemy, 212, 213 
 Bodin, 99 
 
 Bosanquet, B., xvi, 132 
 Bradlaugh, 192 
 Bradley, Prof, xvi, 129-130, 132, 
 
 216 sq., 235 
 
 Calvin and Calvinism, 144, 150, 153 
 Candour, difficulty of, 31, 33-34 
 Carlyle, 195-6 
 Carnal inclination, 30, 31 
 Categorical imperative, 203 
 Catherine the Great, 95 
 
 Catholicism and Protestantism, 20- 
 24, 50-52, 141-4 
 
 Causation, ill sq., 1 34, 149 
 
 Celts and Teutons, 20-22 
 
 Chance, logic of, xxix-xxx ; mean- 
 ing of, 20-21, 102 sq., 115 sq. 
 
 Change, 233-6 
 
 Charity, 137 
 
 Charlemagne, 42 
 
 Charron, 88, 94 
 
 China, 182 
 
 Christian ordinances, 208 sq. 
 
 Civilisation, 1S5 sq. 
 
 Clericalism, 146 sq., 209 
 
 Clittbrd, W. K., 23-24, 27, 37 
 
 Coincidence, 104 
 
 Collins, 145 
 
 Colour, problem of, 125, 219, 228, 
 
 Condillac, xviii-xx 
 
 Condorcel, xx 
 
 Conduct, relation of to thought, 6, 
 
 152 sq., 247 sq. 
 Conquest, 186, 253, 256 
 Conscientiousness, rarity of, 29 ; a 
 
 help to right reasoning, 39 
 Consciousness, 122, 229 
 Conservatism, 98 
 Consistency in thought the gist of 
 
 right thinking, 6-7 ; in politics, 
 
 10 ; meaning of term, 35-36 ; 
 
 law of, 2 n 
 Contradiction, pain of, 26 ; risks of, 
 
 32-33 
 Conversion, logical, xxiii 
 
 Copernicus, 157 
 
 Credulity, error through, 17 ; our 
 
 bent to, 19 
 Croll, Dr. James, 43-45, 47, 49 
 Cudworth, 90, 92 
 Cumberland, Bishop, 152, note ; 171 
 
 Darkness, fear of, 1 7 
 Darwin, 247-9
 
 258 
 
 INDEX 
 
 De Morgan, Prof., 2, 109, 117 
 Descartes, xvii-xix, xxvi, loi 
 Design argument, 1 15-135, 144 
 Determinism, 13959., 161, 170 sq. 
 Dickinson, G. L., 247 
 Diderot, 94-95 
 Disbelief, duty of, 30 
 Divine, tlie term, 235 
 Dostoyevsky, 181 
 Doubt, 17-20, 26 ; gain from, 27 
 Drobisch, xxvii 
 Duelling, 94 
 
 Economics, 61-62 
 
 Edwards, Jonathan, 44, 45, 136, 
 
 148 sq. 
 Ego and Non-Ego, the Fichtean, 
 
 130 sq. 
 Egoism and altruism, 173-5, -°° 
 Emerson, 26, 158 
 Emotion, 12, 94-96, 98 
 Englishmen and lower races, 171 
 Epicurus, 195 
 Error, sources of, 17, 37 ; when 
 
 guilty, 36 
 Ethics, problems of, 139 sq., 170 sq. 
 Evil, 70-73, 166 
 Experience, effect of on beliefs, 41 
 
 Faith, 48, 208 
 
 Fallacies, logical discussion of, xxii ; 
 as to reason, 3-4; as to logic, 
 9-1 1 ; as to emotion and imagina- 
 tion, 12 ; of verbal confusion, xx, 
 13, 79 sq., 122 sq., 154, 161, 185, 
 229, 230 ; from false data, 20-24, 
 174 ; of argument, 63-64, 121 ; 
 from early bias, 43-45 ; of de- 
 parture from definition, 61-62, 
 65-66, 112, I16-7; of counter- 
 sense or negation of category, 
 228 sq. 
 
 119, 146 
 
 118 sq. 
 Faraday, 43 
 Fate, 139 
 
 Fichte, 130 sq., 202 
 Food, opinions about 
 Fowler, Prof., 156, 159 
 Franklin, 97 
 Eraser, Prof., 126 
 Frederick the Great, 95 
 Freedom, 156 
 P'ree-will, 134, 139 S(]. 
 
 "pathetic," 
 
 48 
 
 Galileo, 194 
 Gambling debts, 177 
 Gans, 114 
 Gladstone, 206 
 Glanvil, 90, 92, 99 
 God, 123, 127-9, 133. 139. 141, 
 143-4, 152-5, 204 sq., 214, 220-2, 
 
 233 
 
 Goethe, 252 
 
 Good and Evil, 70-73, 152, 166, 
 
 225 
 Gospels, study of, 41 
 Gottschalk, 141-2, 161, note 
 Gravitation, 157 
 
 Greece, civilisation of, 178-9, 186 
 Green, T. H., xvi, 148, 198-^9, 201 
 
 sq. , 226 
 Guizot, 25 s 
 
 Halifax, quoted, 28 
 
 Hamilton, xx, xxiii, 125 sq. 
 
 Happiness, 214. See Hedonism 
 
 Hedonism, 195 sq. 
 
 Hegel, 68-76, 1 13-4, 132, 133 
 
 Hell, doctrine of, 145 
 
 Helvetius, 96 
 
 Herbert, Lord, 91 
 
 Hesitation, 27 
 
 Hincmar, 142 
 
 History, problems of, 41-42, 1 14, 
 
 186 sq. 
 Hobbes, 91, 92, 150 
 Hodgson, Shadworth, 147-8 
 Honesty, intellectual, 38, 1S5 
 Hiickleberiy Finn, 1 7 1-2 
 Hume, 87, 1 12-3, 171, 202 
 
 Idealism, I19 sq., 203 sq., 228 .sq. 
 Ideals, 2, 15, 172, 199, 211, 214, 
 
 251-2 
 Illusion, 218-219 
 Imagination, 12 
 Immorality, intellectual, 7, 29 
 Inconsistency, meaning of, 6, 35 ; 
 
 commonness of, 37, 56, 135, 137 
 Induction and deduction, xxv 
 Infinite, fallacies about the, 1 20, 
 
 128, 144, 154, 161, 233 
 Infinity, 227-8 
 Injustice, self- forgiveness for, 26, 
 
 38 ; proclivity to, 39 
 Insincerity, 34 sq.,66, 167, 168, 186
 
 INDEX 
 
 259 
 
 Instinct, 170 sq. 
 Interest, bias of, 30 
 Inluitionism, 170 sq., 182 sq., 193 
 Irrationalism, 15 
 
 lesus, historicity of, 41-42 
 Jevons, Prof., xxiii-xxiv, 13-14, 
 
 61-65, 77, 109. 119 
 John the Scot, 142, 166 
 judgment, duty of, in morals, 49, 
 
 Justice, 184 
 
 Kant, xxviii, 40, 172, 189 sq., 203, 
 
 208 
 Kelvin, 43 
 Khonds, the, 253 sq. 
 Kidd, B., 4, 5 
 
 Language, imperfection of, 136, 229 
 Lecky, W. E. H., 77, sq. 
 Leopold, Duke, 95 
 Lionardo da Vinci, xviii 
 Locke, 24, 150, 152, 158, 196 
 Logic, teaching of, xvii sq. 
 
 meaning of, 3, 8-11 
 
 function of, xx, xxiv, 8, 14, 
 
 40, 244 
 
 • symbolic, 64, 237, 244 
 
 formal, xxi, xxvi, 3, 63 
 
 Logiqiu de Port Royal, xvii sq. 
 Lucan, 251 
 Luther, 23, 142, 144 
 Lykurgus, 252 
 
 Macaulay, 9-10 
 
 Macpherson, Major S. C, 253-6 
 
 Malice, errors through, 39 
 
 Mansel, 128, 129 
 
 Martineau, 158, 194 
 
 Matter, existence of, 123 sq., 219 sq. 
 
 Metaphors, 136, 218 
 
 Militarism, 248 
 
 Mill, J. S., xx-xxii, xxvi, 8, 59, 
 
 124-9, 137, 148, note ; 149, note; 
 
 197, note ; 198 
 Minto, Prof., xviii, xxii, S, 14, 
 
 20-23, 102 
 Miracles, 87 
 Mirage, 218-9, 222-3 
 Mode, fallacies of, 119 sq., 148, 
 
 232 
 
 Momerie, 67, 117, 123, 129 
 
 Monarchism, 1 1 
 
 IMontaigne, 87-88, 94, 96 
 
 Montesquieu, 94, 96 
 
 Morals, 49, 148, 156 sq., 172 sq., 
 
 195 sq. 
 More, H., 92 
 Morellet, 94 
 Moslems, 19 
 
 Nature, meaning of, 246-252 
 Naude, 88, 99 
 Necessitarianism, 141, 144 
 Newman, 50-51 
 Newton, 13 
 
 Pain of contradiction, 26 
 Pauline epistles, 226 
 Pelagius and Pelagianism, 14 1, 145 
 Peyrere, 87 
 Platonism, 225 
 Pleasure. See Hedonism 
 Politics, reason in, 3, 9-1 1, 24S 
 Pope, 195 
 
 Praise and blame, 162 sq. 
 Prayer, 135, 208 
 Predestination, 141 sq. 
 Presuppositions, 25 
 Probability-logic, xxix, 104-110, 
 Proctor, R. A., 105-8 
 Propositions, 52-53, 54 
 Protestant ethics, 141 
 Providence, 113, 132-3, I45 
 Punishment, 159 sq. 
 Puns, philosophic, 206 
 Puritanism, 91 
 
 Quibbling, 8, 14 
 
 Rabanus Maurus, 142, 161, note 
 Race, theories of, 20-24 
 Rationalism, meaning of, 86 sq. 
 Reading and judgment, 46 
 Reality, problem of, 213-5, 217 
 
 sq., 234 sq. 
 Reason, meaning of, 3, 82 ; Hegel's 
 
 doctrine of, 70-76 ; and morals, 
 
 167, 187 sq. 
 Reasoning, value of exercise in, 5, 
 
 14 ; common slovenliness in, 5-6, 
 
 14, 82 : relation of, to culture, 79 
 Reciprocity, 184, 211 
 
 T
 
 26o 
 
 INDEX 
 
 Reformation, the, 23-24, 85 
 Religion and race, 20-24 
 
 and conduct, 31, 68 
 
 definitions of, 65-66 
 
 Resentment of criticism, 26-27 
 Responsibility, 162 sq. 
 Revelation, 50, 68 
 Revenge, 160, 162, 167 
 Rhodes, Mr., 250-1 
 Right and wrong, 173 sq. 
 Rome, history of, 1S7 
 Royal Society, 92 
 Raskin, 218 
 
 Sacerdotalism, 56, 212 
 Sacred vice, 212 
 Scholasticism, xx 
 Science, discipline in, 245 sq. 
 Scott, Reginald, 89 
 Shaftesbury, 166 
 Shakspere, 89-90 
 Sidgwick, A., xv-xvi, xxiv-v 
 
 Prof. H., 156-8 
 
 Sigwart, xvi 
 
 Sin, theological view of, IS3-I55 
 
 Sincerity, difficulty of, 31, 34-38, 
 
 60, 167-8 
 Slavery, ethics of, 171-2, 176 sq. 
 Slavonic race and religion, 20-21 
 Smith. Prof. Robertson, 65-66 
 
 Dr. W., 130 sq. 
 
 Sociology, necessity of, 225 sq., 248 
 
 Socrates. 42 
 
 Somers, Lord, 10 
 
 Specialism, 47 
 
 Spencer, 184, 242, 248-250 
 
 Spinoza, 145, 147, 150, 154-5 
 
 Spirits, 224 
 
 "Spiritual principle," 202, 205, 
 
 212-4 
 Sprague, Dr., cited, 11 1 
 
 Statistics, iii 
 
 Strutt, 150, nole 
 
 Subjectivity, 225 
 
 Substantia, 232 
 
 Suicide, problem of, 237 sq. 
 
 Switzerland, race and religion in, 21 
 
 Syllogistic figures, the, xxviii 
 
 Symbolic logic, 64, 237, 244 
 
 Sympathy, 175 
 
 Taste, authority and, 59 
 
 Tell, 42 
 
 Tennyson, 222, 223 
 
 Teutons, theories about, 20-24 
 
 Thales, 227 
 
 Theology, 132, 159, 16S-9 
 
 Thought, pleasures of, 29 
 
 Toland, 87 
 
 Torture, abolition of, 94-96 
 
 Trinity, doctrine of, 53-56 
 
 Truth, duty of telling, 188 sq. 
 
 Ueberweg, xxi, xxvii-xxviii, 32-33 
 "Unbelief," 1 8- 19, 30, 90 
 Utilitarianism, 174, 182, 197 
 Utopias, 97 
 
 Variation in morals, 174 
 Venn, Dr., xxix-xxx, I08-IIO 
 Virgil, 186 
 
 Virtue, theological view of, 151 
 Volition. See Free-will 
 Voltaire, 94, 95, lOO 
 
 Watch argument, 1 15 sq. 
 Wealth, definition of, 61-62 
 Whateley, xx, 137 
 Will-to-believe, 30 
 Wisdom, collective, 84 
 Witchcraft, 87-91 
 Women, mental life of, 3, 16
 
 ALL LIBERAL THINKERS 
 
 WHO HAVE NOT ALREADY JOINED 
 
 THE RATIONALIST PRESS ASSOCIATION, LTD., 
 
 ARE EARNESTLY INVITED TO DO SO. 
 
 The Chairman of the R.P.A. is the veteran reformer and 
 Rationalist, George Jacob Holyoake, and among its Honorary Associates 
 are the eminent scientists, Haeckel and Berthelot, together with Edward 
 Clodd, John M. Robertson, F. J. Gould, Dr. F. J. Furnivall, Dr. 
 Stanton Coit, Dr. Washington Sullivan, and other noted pioneers of 
 mental and moral progress. 
 
 The Objects of the R.P.A. are to stimulate the habits of reflection 
 and inquiry, and the free exercise of individual intellect ; to promote a 
 rational system of secular education ; to assist in publishing the works 
 of capable thinkers, and in popularising the great discoveries of 
 modern science and scholarship ; to re-issue, in cheap form, notable 
 books of a critical, philosophical, or ethical character ; and generally to 
 assert the supremacy of reason, as the natural and necessary means to 
 all such knowledge and wisdom as man can achieve. 
 
 Membership of the R.P.A. can be secured by payment of an 
 annual subscription of not less than ^s., renewable in January of each 
 year. In addition to receiving current literature of the Association 
 up to but not exceeding the value of their subscription, and acquiring 
 a vote at all General Meetings, Members who aie resident in the 
 United Kingdom can borrow books from the Association's Lending 
 Library through the post (Rules free, Catalogue 3^.), or can resort to 
 the Reading Room at Nos. 5 & 6, Johnson's Court ; while Social 
 Gatherings of Members and their friends are arranged from time to time. 
 
 The Publications of the R.P.A. are forwarded, as issued, to 
 all Members, on account of their current year's subscriptions, unless 
 directions to the contrary are given. Books are sent to all parts of the 
 world post free. Members who prefer to receive books by request only 
 can make their own selections from the notices of new publications 
 which will be sent to them in due course. 
 
 Full Particulars of the R.P.A., including the latest Annual 
 Report and any special information desired, can be had gratis by 
 applying to the Secretary, Charles E. Hooper, 5 & 6, Johnson's Court, 
 Fleet Street, London, E.C. 
 
 i
 
 WORKS BY JOHN M. ROBERTSON. 
 
 Courses of Study. viii.-5i6 pp. ; cloth, 8s. 6d. net, by 
 post gs. 
 
 "An encyclopedic book by a well-known scholar, and likely to be 
 of much value." — Times. 
 
 " Ought to be in the library of every student, and, indeed, of every 
 person aspiring to knowledge." — Leicester Mercury. 
 
 " Forms a valuable book of reference for reading men, and should 
 have a place of its own in every well-stocked library." — Scotsman. 
 
 " The best book of its kind."— Zo«(/o« Opinion. 
 
 Christianity and Mythology. xviii.-484 pp. ; 8^. ed. 
 
 net, by post gs. 
 " This magnificent work will be welcomed. . . It is a reference library 
 in itself upon the subjects with which it deals. The reading, the 
 research, the critical comparisons shown, are a matter for envy and 
 unbounded admiration." — The Reforiiier. 
 
 Pagran Christs : Studies in Comparative Hierology. 
 
 xviii.-442 pp. ; Sj. (id. net, by post gs. 
 " The book is of intense interest to the student of comparative 
 religion, and it is marked by the industrious ability which has enabled 
 Mr. Robertson to gain the ear of many who dissent widely from his 
 views." — Literary World. 
 
 A Short History of Christianity. 400 pp. ; cloth, 6j. 
 
 net, by post bs. i,d. 
 In this work the author endeavours to present dispassionately a 
 coherent theory of the true origins of the Christian cult, and to explain 
 its growth in terms of all the sociological elements of the case. 
 
 Letters on Reasoning". XXX.-260 pp. ; cloth, ^s. 6d. 
 
 net, by post 3^. lod. 
 
 What to Read. Suggestions for the Better Utilisation of 
 Public Libraries. Strong paper cover, printed on best paper, 
 24 pp., 4(f., by post 5(/. 
 
 To be issued September, igoj. 
 
 A Short History of FreethOUg'ht. (Thoroughly revised 
 and partly re- written.) In two large volumes. 
 
 Agents of the Rationalist Press Association, Limited : 
 Watts & Co., 17, Johnson's Court, Fleet Street, London, E.C. 
 
 ii
 
 ' . — -H 
 
 
 , Ik. 
 
 
 
 ^ i.A V Co .wj-iflH/L-j 
 
 
 
 ■'/- 
 
 \
 
 UC SOUTHERN 
 
 REGIONAL LIBRARY rACU-lTY 
 
 AA 000 512 118 1 
 
 L 006 339 329 2