This book is DUE on the last date stamped below 
 
 Oct kn 
 ore 
 
 
 
 1011 
 
 m^ 
 
 ;.Ar i 4 1931 
 
 1^30 
 
 WAR 1 9 1935 
 
 \ 
 
 m 
 
 ^ •' 1^^?^ 
 
 SEP 13 19§f< 
 
 E-ir 
 
 JAN 1 2 196C
 
 THE SCIENCE OF ETHICS.
 
 BALLANTYNE, HANSON AND CO. 
 EDINBURGH AND LONDO.V
 
 ^ ? Y 
 
 THE 
 
 SCIENCE OF ETHICS, 
 
 BY 
 
 LESLIE STEPHEN. 
 
 jjj j^ JJ5 J. 
 
 J > 
 
 » ,• 
 
 M * ^ > 'i t •» 
 
 J ' > 
 
 3 3 
 
 
 NEW YORK: 
 
 G. P. PUTNAM'S SONS, 
 
 27 & 29 WEST TWENTY-THIRD STREET. 
 
 1882. 
 ^5^ .
 
 t t ( c c
 
 3J" 
 looi 
 
 PREFACE. 
 
 A PREFACE is generally the most interesting, and not seldom 
 the only interesting, part of a book. It is useful to the hasty 
 critic who wishes to avoid the trouble of reading at all, and 
 to the more serious student who wishes to have the clue to 
 the author's speculations put into his hands at the earliest 
 possible period. I should be glad to be useful to both classes, 
 to save some readers the trouble of getting through more 
 than a couple of paragraphs, and to point out to others what 
 is the kind of result which they may expect from a perusal of 
 the whole work. 
 
 My ethical theory, then, when I first became the conscious 
 proprietor of any theory at all, was that of the orthodox 
 utilitarians. J. S. Mill was the Gamaliel at whose feet I 
 sat, and whose authority was decisive with me on this as 
 on other matters. In this, of course, I was simply following 
 the example of the majority of the more thoughtful lads of 
 my own generation. At a later period my mind was stirred 
 by the great impulse conveyed through Mr. Darwin's Origin 
 of Species. I shall always, I hope, be proud to acknowledge 
 the great intellectual debt which I, in common with so many
 
 vi PREFACE. 
 
 worthier disciples^ owe to his writings,^ So far as ethical 
 problems were concerned^ I at first regarded Mr. Darwin's 
 principles rather as providing a new armoury wherewith to 
 encounter certain plausible objections of the so-called In- 
 trusionists^ than as implying any reconstruction of the utili- 
 tarian doctrine itself. Gradually^ however, I came to think 
 that a deeper change would be necessary, and I believe that 
 this conviction came to me from a study of some of Mr. 
 Herbert Spencer's works. It became stronger during a 
 subsequent attempt at a brief historical examination of the 
 English moralists of the eighteenth century. Whilst I was 
 finishing that task, I read Mr. Henry Sidgwick's Methods of' 
 Ethics, then just published. As I differ upon many points 
 from Mr. Sidgwick, and especially upon the critical point of 
 the relation of evolution to ethics, I am the more bound to 
 express my sincere admiration for his book. It set me think- 
 ing when it failed to make me think with him. The result 
 of my thinking was a resolution to set down as systematically 
 as I could a statement of the ethical theory which had com- 
 mended itself to me. I resolved to begin at the beginning 
 as well as I could, and trudge steadily through the alternate 
 platitudes and subtleties into which every moralist must 
 plunge. My views were, of course, more or less modified 
 in the process, and though they have not substantially 
 changed, I hope that they have gained in consistency and 
 clearness. At any rate, my labours are embodied in the 
 
 ^ It is with a pang of deep regret that I must add to-day (April 24th, 1882) 
 that I can no longer cherish the hope of fully acknowledging it to Mr. Darwin 
 himself. I was withheld from speaking formerly by the feeling that anything 
 like a compliment (sincere though it might be)'seemed incongruous in presence 
 of that exquisitely simple and modest nature. Yet I could wish that I had 
 been less diffident.
 
 PREFACE. vii 
 
 following pages, which may be briefly described as an 
 attempt to lay down an ethical doctrine in harmony with 
 the doctrine of evolution so widely accepted by modern men 
 of science. 
 
 To this statement it would be desirable to add an acknow- 
 ledgment of my debt to other writers. I find it impossible 
 to do so_, for the simple reason that I am altogether ignorant 
 of the extent of my obligations. It has happened to me, as, 
 I presume, to almost all writers upon such topics, to discover 
 that arguments which had apparently sprung up spontaneously 
 in my own mind had really been expressly stated by my 
 predecessors, and, moreover, stated in books with which I 
 had been familiar; and were, therefore, in all probability, 
 intellectual waifs and strays upon which I had unconsciously 
 laid hands. In less direct ways, I have, of course, been influ- 
 enced to a degree which I am quite unable to estimate. I 
 have no fear that my obligations to writers belonging to 
 what I may call my own school, to Hume, Bentham, the 
 Mills, G. H. Lewes, and Mr. Herbert Spencer, will be over- 
 looked or underrated ; and I would gladly name others to 
 whom I am more opposed, were it not that in many cases 
 an acknowledgment would look like a claim to affiliate 
 my speculations upon men who would regard the claim as 
 offensive. I can, however, obviate any objection which may 
 be made to want of fuller acknowledgment by the explicit 
 and perfectly sincere admission that I do not believe (though 
 ajrain I cannot be certain even of this negative statement) 
 that there is a sin2;le ori2:inal thou2;ht in the book from 
 beginning to end. By original, I mean of course a thought 
 which has not occurred to others; though I, of course, also 
 claim to have made every thought which I utter my own 
 by reflection and assimilation.
 
 viii PREFA CE. 
 
 I am the more bound to make this statement because I 
 have made it a rule never to mention proper names. I have 
 done so partly because I think that any book which aims 
 at scientific method should contain within itself all that 
 is necessary to the immediate issues^ and should avoid the 
 appearance of anything like an appeal to authority ; and 
 still more because I have observed that, as a matter of fact, 
 any such references are apt to introduce digressions, and to 
 lead one aside into disputes as to the rightful interpretation 
 or correct affiliation of the principles of other writers, which, 
 however interesting, really involve irrelevant issues. 
 
 Another question may be suggested by this avowal. Why 
 publish a fresh discussion of so ancient a topic if you have 
 nothing new to say? The general answer is simple enough, 
 namely, that problems of this kind require to be dis- 
 cussed in every generation with a change of dialect, if not 
 with a corresponding change of the first principles ; and, 
 further, that it is desirable that all points of view should be 
 represented. This last remark may suggest some answer to 
 the more special question, whether my book has not been 
 made superfluous by the discussion of the same topic upon 
 the same assumptions by the leading exponent of the 
 philosophy of evolution in Mr. Herbert Spencer's Data of 
 Ethics. To this I reply that I differ from Mr, Herbert Spencer 
 in various ways ; and moreover, that we really stand at 
 different points of view. Mr. Spencer has worked out an 
 encyclopaedic system, of which his ethical system is the 
 crown and completion. T, on the contrary, have started 
 from the old ethical theories, and am trying to bring them 
 into harmony with the scientific principles which I take for 
 granted. My aim is more limited, though we ought to 
 coincide in results so far as we cover the same ground. I
 
 PREFACE. ix 
 
 have^ as it were, surveyed the province from within, without 
 attempting to pass the frontiers, whilst he reaches the pro- 
 vince after surveying the whole empire of scientific thought ; 
 and therefore I have laid stress upon some matters which 
 he treats with comparative lightness, whilst in other cases 
 the relation is reversed. 
 
 In fact, however, I hold that there is ample room for any 
 number of labourers in this field ; nay, that there will be 
 room for the labourers of many generations to come. I 
 have no doubt that ethical problems will be debated long 
 after I (it would be impertinent to consider the case of Mr. 
 Spencer) am dead and forgotten. It is enough and more 
 than enough if one can communicate the very slightest 
 impetus to the slowly grinding wheels of speculation. At 
 times, I have been startled at my own impudence when 
 virtually sitting in judgment upon all the deepest and acutest 
 thinkers since the days of Plato. But I easily comfort myself 
 by remembering that the evolution of thought is furthered 
 by the efforts of the weak as well as of the strongest ; and 
 that if giants have laid the foundations, even dwarfs may 
 add something to the superstructure of the great edifice of 
 science. So far as my reading has gone, I have found only 
 two kinds of speculation which are absolutely useless — that 
 of the hopelessly stupid, and that of the hopelessly insincere. 
 The fool who does not know his own folly may be doing 
 nothing, and the philosopher who is trying to darken know- 
 ledge may be doing worse than nothing ; but every sincere 
 attempt to grapple with real difficulties made by a man not 
 utterly incompetent has its value. I claim to come within 
 that description, though I claim nothing more. And I have 
 the satisfaction — not a very edifying one, it may be said, for 
 a professed moralist — to reflect that if my book does no
 
 X PREFACE. 
 
 good to anybody else^ it has provided me with an innocent 
 occupation for a longer time than I quite Hke to remember; 
 whilst I hope that there is nothing in it — if I may apply 
 to myself what a discerning critic has said of Dr. Watts' 
 sermons — " calculated to call a blush to the check of 
 modestv." 
 
 LESLIE STEPHEN. 
 
 London, April 18S2.
 
 CONTENTS. 
 
 CHAPTER I. 
 
 Purpose and Limits of the Inquiry. 
 
 I. The Starting-Poijit. 
 
 PAGE 
 
 1. Nature of ethical controversy ..... i 
 
 2. Conciliation requires a right method .... 2 
 
 3. And a postponement of metaphysical problems . . 3 
 
 4. Metaphysical doubts are irrelevant in the sphere of science 4 
 
 5. Including the moral sciences ..... 5 
 
 6. Truths in moral science independent of metaphysics . 6 
 
 7. The sphere of science ...... 7 
 
 II. Difficulty of Moral Science. 
 
 8. The free-will difficulty irrelevant ..... 9 
 
 9, Moral science, though not self-contradictory, may be 
 
 unattainable . . . . . . . .10 
 
 10. Imperfecti6irx)f science generally . . . . .11 
 
 11. Hopeless complexity of the problem of individual conduct 12 
 
 12. Even average conduct too complex for calculation . . 13 
 
 13. Absence of a scientific psychology . . . .14 
 
 14. Special difficulties from varying character and motive . 15 
 
 15. Intermittent action of consciousness .... 16 
 
 16. Sociology as vague as psychology . . , . .17 
 
 17. Narrow limits of moral sciences ... . . 20 
 
 III. A ttainable Results. 
 
 18. Empirical knowledge of conduct . . . . .21 
 
 19. Its relation to scientific knowledge .... 22 
 
 20. Difficulty of obtaining definitive rules .... 23
 
 xii CONTENTS. 
 
 PAGE 
 
 21. Statistical method insufficient ..... 24 
 
 22. Method of political economy insufficient ... 26 
 
 23. Appearance of fatalism due to false assumptions . . 28 
 
 24. The permanent social structure . . . . .30 
 
 IV. Theory of Social Evohction. 
 
 25. Importance of study of social structure . . . . 31 
 
 26. New significance of phenomena when regarded as indi- 
 
 cative of organic growth . . . ... 32 
 
 27. The true statement of the problem of sociology . . t^t^ 
 
 V. The Ethical Problem. 
 
 28. Our problem is to discover the scientific form of morality 35 
 
 29. Distinction between actual and ideal morality . . 36 
 
 30. Actual morality here concerned ..... 39 
 
 CHAPTER II. 
 
 The Theory of Motives. 
 
 I. The Problem. 
 
 1. An examination of first principles necessary ... 40 
 
 2. Excluding, however, the metaphysical principles . . 40 
 
 II. 77^1? Emotions as Deiermining Conduct. 
 
 3. Happiness determines conduct in a scientific sense . 42 
 
 4. Some common objections are irrelevant ... 42 
 
 5. The formula, though vague, is not meaningless . . 43 
 
 6. It excludes an erroneous theory, but requires more 
 
 precision ........ 44 
 
 7. Theory of happiness as the sole end of conduct . . 46 
 
 8. Supposes a strictly inconceivable process ... 47 
 
 9. And in many important cases gives an erroneous formula 
 
 — Objective and subjective ends .... 47 
 
 10. Misrepresents the inseparable connection of emotion and 
 
 reason ......... 49 
 
 11. Pain and pleasure as ultimate determinants of conduct . 50 
 
 12. In all cases pleasure corresponds to stable and pain to 
 
 unstable equilibrium . . . . . .51
 
 CONTENTS. xiii 
 
 Page 
 
 13. Analogy of least resistance , . . . . .52 
 
 14. Nature of an act of choice ...... 53 
 
 15. Difficulty of discovering the sphere of volition . . 54 
 
 16. And of knowing our own desires ..... 56 
 
 17. True statement of the greatest-happiness formula — The 
 
 ■ province of reason . . . . , . .56 
 
 III. T/ie Reason as Determining Conduct. 
 
 18. All conduct " reasonable " so far as giving an assignable 
 
 law 58 
 
 19. Senses in which conduct may be more or Iq^s reasonable 59 
 
 20. Reason not opposed to feeling as such .... 60 
 
 21. Sense in which it may dispense with feeling ... 61 
 
 22. Latent feeling . . . . . . . .61 
 
 23. Feeling by signs ........ 62 
 
 24. Reason and feeling are mutually involved, and naturally 
 
 develop together . . . . , . .63 
 
 25. But may be considered separately — Reason as implying 
 
 accuracy of judgment ...... 64 
 
 26. And consistency of motive insufficient to determine 
 
 conduct 65 
 
 27. Frohlem o{ the su7nmie?n ionum . . . . .67 
 
 28. Given a single end of conduct, the problem is definite . 67 
 
 29. But men have in fact many partially inconsistent ends , 68 
 
 30. The unity of the agent implies di: facto commensurability 
 
 of ends ......... 69 
 
 31. Reasonableness thus implies harmony and consistency of 
 
 conduct ......... 70 
 
 32. But as character varies, the end is still indeterminate . 71 
 2)1. So far, therefore, no sole end follows from reasonableness 73 
 
 \Y. Types. 
 
 34. IMeaning of " type" illustrated .... 
 
 35. A typical instrument implies the solution of a problem 
 
 36. And a maximum efficiency of given means to a given end 
 
 37. Esthetic value of types .... 
 
 38. The same principles apply to a living organism 
 
 39. What corresponds in this case to the "end" , 
 
 40. The evolutionist theory suggests an answer 
 
 41. The evolution of vitally efficient types . 
 
 42. Extreme intricacy of any accurate statement . 
 
 74 
 11 
 IS 
 76 
 76 
 78 
 7S 
 80 
 80
 
 xiv CONTE^WTS. 
 
 V. T/ie Principle of Utility. 
 
 PAGE 
 
 43. Conduct may be useful as " pleasure-giving " as to " life- 
 
 preserving :" the two must tend to coincide . . 82 
 
 44. Habits and instincts may be considered in either aspect . 83 
 
 45. Necessity of distinguishing between organisms and aggre- 
 
 gates ......... 84 
 
 46. Utility must be understood with reference to whole 
 
 organism , . . . . . . .85 
 
 47. Ultimate theory of pain and pleasure irrelevant . . 86 
 
 48. They must be functions of working condition of organism 86 
 
 49. Correlation of painful and pernicious, pleasurable and 
 
 beneficial ........ 88 
 
 50. Presumption in favour of more essential instincts . . 89 
 
 VI. Social and Individual Utility. 
 
 51. Instincts essential to the race may be prejudicial to the 
 
 individual ... ..... 90 
 
 52. Hence necessity of considering relation of race to indi- 
 
 vidual ......... 92 
 
 CHAPTER III. 
 
 Theory of Social Motives. 
 
 I, The Individual and the Race, 
 
 1. Importance of the distinction . . . . • 93 '^ 
 
 2. Properties of the organic whole not inferable from pro- 
 
 perties of constituent units . . . . • 93 . 
 
 3. Dependence on race an essential property of man . . 95 \/ 
 
 4. There must be a tacit or express reference to this in every 
 
 theory of human nature . . . . . . 96 v/ 
 
 5. Some properties may admit of independent variation . 97 
 
 6. In simplest case, the organism may vary whilst social a 
 
 organisation is constant . . . . . .98 
 
 7. In most complex, organic variation implied by change 
 
 in social organisation . . . . . -99 
 
 8. The most complex social organisation varies whilst indivi- 
 
 dual organisation is constant . . . . .100
 
 CONTENTS. 
 
 XV 
 
 1 1. Society atid Men. 
 
 9, The last is true of human societies 
 
 10. Social development may take place 
 
 variation .... 
 
 11. The accumulation of experience . 
 12. 
 
 13- 
 
 without orjranic 
 
 Inheritance of capital and skill 
 Changes implied in language as a product of the 
 factor ....... 
 
 14. Necessity of considering the social factor 
 
 1 5. The mutual dependence of the members of society 
 
 16. The omnipresent social discipline . 
 
 17. Society to be regarded as an organic growth . 
 
 social 
 
 PACE 
 lOI 
 
 103 
 104 
 
 105 
 107 
 108 
 109 
 I 10 
 
 III. Social Organisation. 
 
 18. Social organs . . . . . . . .112 
 
 19. Corporate sentiment implied in an organ . . .113 
 
 20. Various developments of corporate sentiment . . . 115 
 2 I. Every such sentiment implies the existence of more general 
 
 sentiments . . . . . . . .116 
 
 22. The hierarchy of social instincts . . . . .117 
 
 23. The constitution of each organ dependent upon contingent 
 
 circumstances . . . . . . .119 
 
 IV, Social Tissue. 
 
 24, 
 
 2$. 
 
 " Social tissue " is the substratum of social organs . 
 
 Theory of evolution requires definition of units 
 
 Relation of this to problems involving reference to social 
 
 factor . . . 
 
 Social development spread laterally as well as lineally 
 How this affects the struggle for existence 
 The unit of operation cannot be the state 
 30. Otherwise our problem^ would be the evolution of typical 
 
 states ........ 
 
 Nature of the social tissue ..... 
 
 Malthusian competition ..... 
 
 Complex nature of the actual process ... 
 The vitality of the tissue has conditions capable of inde 
 
 pendent statement ...... 
 
 35. No precise test of identity of tissue 
 
 26. 
 
 27. 
 28. 
 29 
 
 3t. 
 32. 
 33. 
 
 34. 
 
 120 
 120 
 
 121 
 122 
 123 
 123 
 
 125 
 
 126 
 127 
 127 
 
 129
 
 xvi CONTENTS. 
 
 V, T/ie Family. 
 
 PAGE 
 
 36. The family not co-ordinate with other forms of association 131 
 
 37. It corresponds to more intimate and underlying order of 
 
 forces ......... 
 
 38. It represents the forces of cohesion of the social tissue , 134 
 
 39. Recapitulation to show necessity of these distinctions . 135 
 
 132 
 
 CHAPTER IV. 
 Form of the Moral Law. 
 
 I. Law and C^istom. 
 
 1. Customs and habits . . . . . . .137 
 
 2. They are "laws of nature " in the making . . .137 
 
 3. Positive laws imply custom, including a potential element 138 
 
 4. They must rest ultimately upon custom, not upon coercion 140 
 
 5. Every society must have organic laws, intelligible only 
 
 through itself. . . . . . , ,142 
 
 6. It is dependent upon the vitality of corresponding social 
 
 instincts . . . . . . . .143 
 
 7. We must therefore study the conditions of this vitality 
 
 separately . . . . . . . .145 
 
 II. The Moral Law. 
 
 8. The moral law must be an approximate statement of these 
 
 conditions . . . . . . . .147 
 
 9. This does not necessarily give the reason, but only the 
 
 cause of morality . . . . . . .148 
 
 III. The Moral Law is Natural. 
 
 10. Hence we may deduce certain attributes of morality — It 
 
 is "natural" ........ 149 
 
 11. Possibility of an artificial development of actual morality 150 
 ~I2. Morality must be a function of the organic instincts . 150 
 
 13. In what sense it can be affected by a moral legislator . 151 
 
 14. Morality only susceptible of secular variations . .152 
 
 15. The "immutability" and "eternity" of the moral law 
 
 deducible from this . . . . . .153
 
 CONTENTS. xvii 
 
 IV. Morality is Natural. 
 
 PAGE 
 
 1 6. Change of form from " Do this " to " Be this " . -155 
 
 17. Why is the change important ? . . . . -155 
 
 18. The external and internal rules may approximate but can- 
 
 not be coincident . . . . . . .156 
 
 19. A general rule of conduct must be a rule of character . 157 
 
 20. Extrinsic and intrinsic motives . . . . .158 
 
 21. Some classes of conduct without intrinsic motive . . 159 
 
 22. Most rules imply a variety of possible motives . .160 
 
 23. Possibility of external rules due to habit and actual incon- 
 
 sistency . . . . . . . .160 
 
 24. They may correspond approximately to constant motives 161 
 
 25. The extrinsic motive of respect for authority implies 
 
 occasional deviations . . . . . .162 
 
 26. Hence reference to authority or extrinsic motive must be 
 
 eliminated from a true moral law . . . .163 
 
 27. The development of such laws is a process of induction . 164 
 
 28. Hence the attribute of ^f^y^riT/f? " supremacy " of moral law 166 
 
 V. Basis of Morality. 
 
 29. Recapitulation , . . . . , . .167 
 
 30. Possible deduction of moral law . . . . .168 
 
 31. Many laws which satisfy these conditions not moral . 169 
 
 32. The distinction of the organism and aggregate to be borne 
 
 in mind , . , . . . , .170 
 
 33. Hence we have to seek for the difference of the moral law 17 i 
 
 CHAPTER V. 
 
 Contents of the Moral Law. 
 
 I. The Law of Nature and Morality. 
 
 1. The law of nature includes prudence and morality . . 172 
 
 2. Deduction of special virtues . . . . . .173 
 
 3. The cardinal virtues . . . . '". . .174 
 
 1 1. Virtue of Courage, 
 
 4. Social welfare dependent upon strength, and especially 
 
 upon courage . . . . . . . .175 
 
 5. Development of chivalrous sentiment . . . . 'i^77
 
 XVIU 
 
 CONTENTS. 
 
 PAGE 
 
 6. Development of " moral courage " . . . .178 
 
 7. Limitations to utility of courage . . . . .179 
 
 8. Question whether courage is properly a virtue . . 1 80 
 
 9. Courage regarded by common sense with a quasi moral 
 
 approval , . . , . . . .181 
 
 It is regarded as a necessary but not sufficient condition 
 
 of virtue . . . . . . . .182 
 
 The development as a process of implicit induction . 183 
 
 Question whether approval implies perception of utility of 
 
 courage . . . . . . . . .184 
 
 The distinct perception emerges in later stages . .185 
 
 Approval of conduct as useful must itself be useful , .186 
 
 Possible non-fulfilment of this condition . . , .187 
 
 16. Tendency to its complete fulfilment . . . .188 
 
 17. Application to other virtues of strength . . . .189 
 
 10. 
 
 1 1. 
 12. 
 
 13. 
 14. 
 15- 
 
 III. Viftue of Tempera7ice. 
 
 18. Temperance either prudential or strictly virtuous . .190 
 
 19. Change from external to internal rule . . . .191 
 
 20. The antisocial quality of sensuality produces the moral 
 
 disapproval . . . . . . . .192 
 
 21. The sentiment goes beyond the utilitarian perception of 
 
 "consequences" . . . . . . .195 
 
 22. The development of the ascetic theory . . . .195 
 
 23. Ascetic theory implies a one-sided but sometimes accurate 
 
 assumption . . . . . , . .196 
 
 24. Ordinary and heroic virtues , . . . . .197 
 
 25. A tacit recognition of utiUty implied in these theories . 198 
 
 26. In what sense all "consequences" are to decide morality 199 
 
 27. The whole organic constitution must be taken into account 200 
 
 28. Physiological element in the moral sentiment. . .201 
 
 IV. Virtue of Trjcth. 
 
 Social welfare dependent upon truthfulness . . . 202 
 Slow development of toleration . , . , .204 
 Possibility of an absolute statement of moral law of truth- 
 fulness ......... 205 
 
 Still the virtue is based upon social welfare . . . 206 
 
 Casuistical difficulties from an absolute statement . . 207 
 34. The development of truthfulness resembles that of other 
 
 virtues ......... 208 
 
 29. 
 30. 
 31. 
 
 32. 
 33
 
 CONTENTS. xix 
 
 V. T/ie Social Viriues. 
 
 PAGE 
 
 35. These virtues developed in the same way and based upon 
 
 social welfare , . . . . . . .210 
 
 36. Justice and benevolence distinguished . . . .211 
 
 37. Justice means acting on sufficient reason . . .212 
 
 38. The sufficient reason means social welfare . . .213 
 
 39. Hence the duties really consistent are a development of 
 
 the same principle . . . . , . .214 
 
 VI. The Social Definition of Virtue. 
 
 40. Recapitulation of the principles asserted . . .215 
 
 41. Origin of problem of obligation . . . . . 217 ^_^ 
 
 CHAPTER VI. 
 
 Altruism. 
 I. Esioisfic Instincts. 
 
 '<b^ 
 
 1. The possibility of self-sacrifice disputed . . .219 
 
 2. The egoist denies the possibility of conscious self-sacrifice 220 
 
 3. Beneficial consequence does not necessarily imply benevo- 
 
 lent intention . . . . . . . .221 
 
 4. Gradual development of truly benevolent intention . . 222 
 
 5. Doctrine of " association " gives an apparently inadequate 
 
 explanation , . . . . . . .223 
 
 6. Egoistic theory that the happiness of others can never be 
 
 an ultimate end . . . . . . .224 
 
 7. Implied truth that all motives may be stated objectively 
 
 or subjectively . . . . . . .225 
 
 8. The objective statement must express some condition of 
 
 the subject , . . . . . . .226 
 
 9. If it includes the happiness of others, some further analysis 
 
 is needed 227 
 
 II. Sympathy. 
 
 10. " Objective" is that which exists for you as well as for me 228 
 
 I I. To think of another man is to have representative feelings 229 
 
 12. Sympathy has a logical as well as an emotional value . 230 
 
 13. This is obscured by the ambiguities of ordinary language 230
 
 XX 
 
 CONTENTS. 
 
 14. Nor does it giv'e a full account of the process 
 
 15. Cruelty as identified with stupidity 
 
 16. Or as inattention to implicit consequences 
 
 17. Complications introduced by antipathy . 
 
 18. Sympathy the fundamental fact 
 
 19. Pleasure in pain as pain 
 
 20. Normally sympathy means self-identification with others 
 2T. The difficulty is to understand the reverse feeling . 
 
 PAGE 
 
 236 
 
 237 
 238 
 
 III. Altr7iis7n. 
 
 22. Sympathy the condition of altruism 
 
 23. All conduct conditioned by the agent's feelings 
 
 24. It is not therefore selfish .... 
 
 25. Association does not explain it . 
 
 26. It must be conditioned by the agent's knowledge of 
 
 men's feelings ...... 
 
 27. But is not therefore removable at pleasure 
 
 28. The ideal conditions as fixed as the non-ideal 
 
 29. An egoistic interpretation still possible . 
 
 other 
 
 239 
 
 239 
 240 
 241 
 
 242 
 243 
 245 
 245 
 
 IV. The Rule of Conduct. 
 
 30. The true question, what is the law of feeling ? 
 
 31. The " greatest-happiness " principle of conduct 
 
 32. Virtual altruism of unreasoning instincts 
 
 33. How affected by intellectual development 
 
 34. The purely prudential maxim never absolutely observ- 
 
 35. But approximately observable when sympathy is neglected 
 
 36. Sympathetic motives are not necessarily divergent . 
 
 37. But can only conform fully so long as sympathy is dormant 
 
 or factitious ........ 
 
 38. Systematic disregard of sympathetic virtues implies idiocy 
 
 39. Their law implies regard to happiness of others as an 
 
 ultimate motive ....... 
 
 40. The prudential maxim is taken up into the higher as society 
 
 develops ........ 
 
 41. Fallacies involved in egoistic statement . 
 
 42. The true meaning of " disengagement " . 
 
 43. Every end must correspond to a real motive . 
 
 44. Benevolence not explicable by a tacit reference to self 
 
 45. Hence comparison of sympathetic and non-sympathetic 
 
 agents ......... 
 
 246 
 247 
 248 
 249 
 
 250 
 251 
 252 
 
 253 
 255 
 
 256 
 
 257 
 
 259 
 260 
 260 
 261 
 
 262
 
 CONTENTS. 
 
 XXI 
 
 CHAPTER VII. 
 Merit. 
 
 I. T//e Conception of Merit. 
 
 1. Virtue implies more than simple benevolence . 
 
 2. Approval of virtue implies more than gratitude 
 
 3. Hence the various problems of obligation 
 
 4. Merit is proportioned to the intrinsic goodness of the agen 
 
 5. It represents a claim upon the universe . 
 
 6. The value set upon virtues varies with their utility . 
 
 7. Analogy in this respect to economical theory of value 
 
 8. Merit applies only to voluntary conduct 
 
 9. And to qualities capable of training 
 
 10. The merit of an action means the virtue proved by it 
 
 1 1. Morality cannot be independent of motives when true 
 
 morality is developed ..... 
 
 12. Difficulties of distinguishing material and formal morality 
 I 3. The genuine moral law must classify actions by motives 
 14. Recapitulation ....... 
 
 I'AGE 
 264 
 265 
 265 
 266 
 267 
 268 
 270 
 27 I 
 272 
 
 274 
 275 
 276 
 277 
 
 II. Free-will. 
 
 15. The nature of the free-will difficulty .... 278 
 
 16. The merit must depend upon the motive . . . 279 
 
 17. Hence "voluntary" means free from coercion, not from 
 
 determination. . . . . . . .280 
 
 1 8. Free-will theory, so far as separating motive from conduct, 
 
 is destructive of merit , . . . . .281 
 
 19. This includes responsibility . . . . . .282 
 
 20. The statement that we can form our own characters is 
 
 irrelevant . . . . . . . .283 
 
 21. T\\& causa causcB argument . . . . . .284 
 
 22. Causation does not lessen responsibility . . . 285 
 
 23. The difficulty arises from a false interpretation of causation 286 
 
 24. Cause and effect imply mutual dependence, and therefore 
 
 implication, but not coercion as something additional. 287 
 
 25. Necessity is the antithesis of chance, and both subjective. 290 
 
 26. Fore-knowledge is consistent with responsibility . .291 
 
 27. Thus the ^rt^ifji^ ^ra/^i'iS? argument is really irrelevant . .292 
 
 III. Effort. 
 
 28. Merit implies discipline . . . . • 
 
 29. A standard of action and membership of a class 
 
 30. The change of the standard gives rise to ambiguities 
 
 294 
 294 
 295
 
 xxii CONTENTS. 
 
 PAGE 
 
 31. Difficulty of deciding the Standard applicable to infants . 296 
 
 32. And to madmen ; but motive in all cases the criterion . 297 
 
 33. The difficulty arises from the complexity of the facts, not 
 
 from ambiguity of principle . , . . .299 
 
 34. Hence the fallacy of measuring merit by effort without 
 
 distinguishing the cases . . . . , .299 
 
 35. The man is most meritorious to whom virtue is easiest, 
 
 provided that he belongs to the class . . .301 
 
 36. Hence the problem whether essential instincts can be bad 301 
 
 37. Development implies more than moral development . 302 
 
 38. Effort so far essential to virtue that liability to temptation 
 
 is essential ........ 303 
 
 IV. Knowledge. 
 
 39. The love of virtue for its own sake .... 304 
 
 40. How far does conformity to a law imply a recognition of 
 
 law? ......... 
 
 4 1 . Reason, and therefore knowledge of law, implied in develop- 
 
 ment ......... 306 
 
 42. But the implicit obedience to law may precede conscious 
 
 acceptance of law ....... 306 
 
 43. Thus the problem becomes one of fact, not of principle , 308 
 
 44. Transition and recapitulation ..... 309 
 
 305 
 
 CHAPTER VIII. 
 
 The Conscience. 
 
 I. Theories of Conscience. 
 
 1. Merit implies intrinsically virtuous motive or conscience . 311 
 
 2. Difficulty of explaining the nature of conscience . .311 
 
 3. The deduction from pure reason ambiguous . . -313 
 
 4. The theory of a primitive faculty implies a wrong psycho- 
 
 logy 314 
 
 5. The moral judgment is a function of the whole character 315 
 
 6. Love of virtue for its own sake must be understood con- 
 
 sistently with this . . . . . . ,316 
 
 7. A separate faculty not a conscience, but a conscientious man 317 
 
 8. Conscience is therefore derivative, and not the less real 
 
 though generally weak in point of fact . . .319
 
 CONTENTS. xxiii 
 
 II. Sense of Shame. 
 
 PAGE 
 
 9. Conscience implies an intellectual and emotional factor . 319 
 
 10. The sense of shame seems to be one element . . 3-i 
 
 11. Shame is excited by offences other than moral , . 322 
 
 12. By simply ridiculous, and even by right conduct . . 323 
 
 1 3. The general law depends upon heightened self-conscious- 
 
 ness .......•• 3-3 
 
 14. Hence implicated in conscience, but the relation to con- 
 
 science variable . . . . . . -3-4 
 
 15. The law of the feeling is not deducible from logical 
 
 definition . . . . . . . -3-5 
 
 16. It gives rise to quasi intuitions, which, however, are 
 
 variable . . . . . . . -3-6 
 
 17. Its possible variation limited by individual organisation . 327 
 
 18. And by the law of the social organism .... 3-8 
 
 19. Hence it is involved in moral development . . . 329 
 
 20. Its law and its utility definable only through social factor 330 
 
 III. Esthetic Judgments. 
 
 21. This distinctive quality of the ethical judgment consistent 
 
 with its composite character . . . . -33- 
 
 22. Esthetic judgments are equally composite . . . 332 
 
 23. An aesthetic judgment may involve every primitive instinct 333 
 
 24. It is differenced by form of gratification, not by instincts 
 
 gratified ........ 334 
 
 25. Hence the moral judgment must include an £esthetic 
 
 element . . . . . . . . -335 
 
 26. The assimilation of the moral law impHes the growth of 
 
 such a sensibility ....... 33^ 
 
 27. Hence the so-called moral sense, which outruns the direct 
 
 moral judgment ....... 337 
 
 28. Questions of "taste" include a reference to objective 
 
 facts ......... 33S 
 
 29. The moral judgment an implicit judgment of types . . 339 
 
 IV. The Conscience. 
 
 30. The conscious ground of morality may vary widely from 
 
 the cause . . . . . . . .339 
 
 31. Every association implies the development of a highly 
 
 complex instinct of loyalty . . . . -341 
 
 32. The sense of duty as the corporate instinct of the '-social 
 
 virtue" is necessarily vague ..... 34- 
 
 33. But it is definite because all organisations involve the 
 
 fundamental instinct ...... 343
 
 xxlv CONTENTS. 
 
 PAGE 
 
 34. And because the true school of morahty is the family . 344 
 
 35. The reference to the virtue itself unconscious though 
 
 essential ........ 345 
 
 36. Morality is rooted in family sentiment, though divergence 
 
 is possible ........ 346 
 
 37. Two points of view are thus necessary .... 348 
 
 38. The conflict of duties possible, though the question is 
 
 more of fact than morality ..... 349 
 
 39. Conscience is thus a modification of all primary instincts 349 
 
 40. Application to the case of maternal love . . • 35i 
 
 41. Transition to problem of happiness .... 352 
 
 J 
 
 CHAPTER IX. 
 
 Happiness as a Criterion. 
 
 I. Utilitarianism. 
 
 From the cause we infer the reason of morality . . 353 
 
 Problems of the criterion and sanction of morality . . 354 
 
 The utilitarian criterion . . . . . -355 
 
 4. Irrelevant objections ....... 356 
 
 5. Relevant objections to uncertainty, lowness, externality, 
 
 expediency of code . . . . . . -358 
 
 II. Limits of the Method. 
 
 6. Utilitarianism based on the empirical philosophy . . 359 
 
 7. Theory of " lots" of happiness ..... 360 
 
 8. Tendency to regard society as an aggregate, not as an 
 
 organism . . . . . . . .361 
 
 9. Inadequacy of method to give a scientific formula . . 363 
 10. It makes a wrong assumption as to constants . . 364 
 I I. The analogy of sanitary rules which refer to health instead 
 
 of pleasure ........ 365 
 
 I 2. The evolutionist morality makes the same change in the 
 
 moral criterion ....... 366 
 
 13. And thereby makes a code flexible and rigid in the right 
 
 places ......... 367 
 
 14. Utilitarianism represents instantaneous morality, but re- 
 
 quires correction when the organism varies . . 368 
 
 15. Analogy of individual and social growth . . . 370 
 
 16. Importance of similar considerations elsewhere . . 370
 
 CONTENTS. 
 
 XXV 
 
 III, Variability of Morality. 
 
 1 7. The standard must be affected by the average estimate of 
 
 happiness ........ 
 
 1 8. But does not assume the desirability of a maximum of 
 
 such happiness ....... 
 
 19. This imphed in the nature of moral development . 
 
 20. The moral law implicitly commands an elevation of ideal 
 
 happiness ........ 
 
 21. Another cause for lowness of utilitarian standard 
 
 PAGE 
 
 373 
 374 
 
 375 
 
 375 
 
 IV. Extrinsic Morality. 
 
 22. The utilitarian rule tends to be external 
 
 23. The egoistic utilitarian explains benevolence by extrinsic 
 
 motives .......•• 
 
 24. Illustration of the miser ...... 
 
 25. External morality thus explicable, but not internal . 
 
 j/ 
 
 76 
 
 377 
 378 
 
 379 
 
 27. 
 28. 
 
 34. 
 35- 
 36. 
 37. 
 38. 
 
 39- 
 40. 
 
 V. Expediency. 
 
 The altruistic utilitarianism is still said to rest on expe 
 
 diency ..... 
 It cannot make the conscience absolute 
 Importance of observing general rules as understood by 
 
 the utilitarian 
 Consequences are in fact relevant 
 A " categorical imperative " not applicable to external 
 
 morality ....... 
 
 The sphere of casuistry when facts are not disputed 
 
 Utilitarian theory generally regarded as insufficient . 
 
 The analogy of military duty 
 
 Hence true principle in this case 
 
 Which applies to the moral law 
 
 The case of suicide 
 
 Moral pedantry . 
 
 In what sense " expedieney '' must be recognised 
 
 A fixed criterion still remains 
 
 Transition to problem of sanction 
 
 380 
 382 
 
 383 
 
 384 
 
 384 
 386 
 
 387 
 3881^ 
 
 390 
 390 
 391 
 392 
 393 
 394 
 395
 
 xxvi CONTENTS. 
 
 CHAPTER X. 
 
 Morality and Happiness. 
 
 I. T/ie Sanction, 
 
 PAGE 
 
 1. Why should a man be virtuous ? ..... 396 
 
 2. Morality unconditionally useful to the society, not to the 
 
 individual ........ 397 
 
 3. The problem of Job ....... 397 
 
 4. Presumption that virtue gives happiness to the virtuous . 398 
 
 5. The standard of the virtuous not a standard for others . 399 
 
 6. Nor is it possible to lay down an absolute and uniform 
 
 standard ........ 400 
 
 7. Various senses of question ...... 402 
 
 8. Assumption that honesty is the best policy . . . 403 
 
 II. Happiness and Virtue. 
 
 9. The temptation to exaggerate the coincidence . . 404 
 
 10. Health the primary condition of happiness . . . 404 
 
 1 1. Extension of the principle beyond bodily health . . 405 
 
 12. Deduction from correlation of painful and pernicious . 406 
 
 13. Equally true of the socially pernicious .... 407 
 
 14. It suggests the line of argument ..... 408 
 
 1 5. Happiness rewards obedience to law of nature, not to moral 
 
 law simply . . . . . ■ . . . 408 
 
 16. A neglect of this leads to many of the perplexities . . 409 
 
 17. This consistent with coincidence of happiness and virtue 410 
 
 18. A man cannot be too virtuous for happiness normally . 411 
 
 19. Confirmation from particular moral rule of temperance . 412 
 
 20. The extension of the principle to other moral rules . .413 
 
 21. This applies equally to sympathetic and non-sympathetic 
 
 motives . . . . . . . . .414 
 
 22. Question of virtue in advance of actual standard . .416 
 
 23. The coincidence less perfect in this case . . .417 
 
 III. Moral Discipline. 
 
 24. Natural aptitude for virtue . . . . . .418 
 
 2 5. The fact that virtue has to be acquired is generally irrelevant 4 1 9 
 
 26. Moralisation part of a process which has other advantages 419 
 
 27. Extension of the ordinary prudential argument . .421 
 
 28. Unity of virtue ........ 422
 
 CONTENTS. 
 
 XXV 11 
 
 29. Moral idiocy possible .... 
 
 30. No external morality possible in such cases 
 
 31. The prudential argument still valid 
 
 32. A moral world generally desirable for all 
 
 PAGE 
 423 
 423 
 
 425 
 
 33. 
 
 34. 
 35. 
 36. 
 37. 
 
 IV. Self-Sacrijice. 
 
 Self-sacrifice sometimes a necessary consequence of virtue : 
 the case of Regulus ...... 
 
 No argument but the gallows available in some cases 
 
 This too follows from admission of real altruism 
 
 The ultimate discord ....... 
 
 The general statement limits, but does not include it 
 
 426 
 429 
 430 
 431 
 432 
 
 I. 
 2 
 
 3. 
 4. 
 
 5. 
 6. 
 
 7. 
 8. 
 
 9- 
 10. 
 
 II. 
 12. 
 
 13. 
 
 14. 
 
 15- 
 16. 
 
 17. 
 18. 
 
 19- 
 20. 
 21. 
 
 CHAPTER XL 
 
 Conclusion. 
 
 Further controversies .... 
 
 The insoluble difficulty 
 
 The art and practice in other sciences . 
 
 In ethics both a science and an art 
 
 " Ought " and " is " . 
 
 Beliefs about conduct imply laws of conduct 
 
 The scientific theory of positive laws 
 
 And of the moral law .... 
 
 The belief to become effective must stimulate the emotion: 
 Morality not universally binding de facto ^ but only dc ju}\ 
 Scientific ideal of a perfect correlation . 
 Progress implies discord ..... 
 
 Can the limits of the scientific inquiry be transcended? 
 Futility of ontology ...... 
 
 Importance of disentangling the scientific question . 
 Metaphysical inquiry means the elimination of ultimate 
 
 fallacies, not the discovery of facts 
 Metaphysics related to ethical as to other sciences . 
 The moral code is formed by scientific observation . 
 INIetaphysics cannot affect this conclusion 
 The ontological and metaphysical theories are irrelevant 
 The theological theory adds nothing to the scientific theory 
 
 of permanence of morality . . , . . 
 
 434 
 435 
 435 
 436 
 437 
 438 
 
 439 
 440 
 
 441 
 
 442 
 
 443 
 444 
 446 
 
 447 
 447 
 
 448 
 450 
 450 
 
 451 
 452 
 
 455
 
 XXVUl 
 
 CONTENTS. 
 
 22. The denial of the reality of evil . . . . , 
 
 23. The science of morality does not give the art, nor, there 
 
 fore, the sanction of morality .... 
 
 24. Religion and morality ...... 
 
 25. Belief affects conduct, but not the ultimate condition 
 
 26. The cause of belief must be considered . 
 
 27. Change of dialect not a change of ultimate belief . 
 
 28. The basis of morality unalterable .... 
 
 PAGE 
 456 
 
 456 
 
 457 
 458 
 
 459 
 460 
 
 460
 
 THE SCIENCE OF ETHICS. 
 
 CHAPTER T. 
 
 PURPOSE AND LIMITS OF THE INQUIRY. 
 I. The Starting-Point. 
 
 I. At the outset of any inquiry it is proper to take stock 
 of the results obtained by previous explorers of the same 
 field, and in particular to ask how far they have reached that 
 unity of doctrine which affords a presumption, though not a 
 conclusive proof, of the attainment of definitive truth. If 
 we examine the history of ethical speculation for this purpose, 
 we are confronted by a fact which, were it not so familiar, 
 might appear to be anomalous. In one sense moralists are 
 almost unanimous ; in another they are hopelessly discordant. 
 They are unanimous in pronouncing certain classes of con- 
 duct to be right and the opposite wrong. No moralist denies 
 that cruelty, falsehood, and intemperance are vicious, or that 
 mercy, truth, and temperance are virtuous. Making every 
 deduction from this apparent unanimity — allowing that 
 similar names may be interpreted in very different senses; 
 that the general outlines of a moral code may be the same 
 whilst its spirit varies widely ; and that the moral codes 
 accepted at different times and places have been as different 
 as has ever been seriously maintained — yet it remains true 
 that there is an a})proximation to unity. The difference 
 between different systems is chiefiy in the details and special 
 application of generally admitted principles. It is not such
 
 2 THE SCIENCE OF ETHICS. 
 
 as we might anticipate from a radical opposition both of 
 method and principle. But if we turn from the matter to 
 the form of morality; if, instead of asking what actions 
 are right or wrong:, we ask, what is the essence of rio;ht 
 and wrong? how do we know right from wrong? why 
 should we seek the right and eschew the wrong; ? — we are 
 presented with the most contradictory answers ; we find 
 ourselves at once in that region of perpetual antinomies, 
 where controversy is everlasting, and opposite theories seem 
 to be equally self-evident to different minds. 
 
 2. This remark has long been familiar, but perhaps we 
 have not, even yet, familiarised ourselves with some of its 
 consequences. We must consider this protracted antagonism 
 in the light of a principle which grows more prominent with 
 the advance of historical methods of inquiry. Every widely- 
 spread opinion deserves respect by the bare fact of its exist- 
 ence. It is itself a phenomenon which requires to be taken 
 into account. We can no long^er be content with refuting; 
 our opponents; we are also bound to explain them. The 
 vitality of any doctrine supposed to be erroneous proves that 
 it cannot be entirely erroneous. It must have an element of 
 truth for which it is necessary to provide accommodation in 
 any satisfactory system. It is a recognised criterion of 
 successful speculation that it should explain not only the 
 phenomena considered, but the illusions due to a partial view 
 of the phenomena. The fact that the Ptolemaic system of 
 astronomy can be fitted into the Copernican system as re- 
 presenting an imperfect view of the facts, whilst the relation 
 cannot be reversed, is itself part of the evidence on behalf of 
 the more comprehensive system. The necessity of applying 
 this reflection to ethical doctrine has been widely recognised, 
 though the dif^culty of applying it is still enormous. Some 
 writers aim at the desired conciliation by eclectic systems, 
 which turn out to be mere tessellations of inconsistent frag- 
 ments, instead of being; harmonious wholes which absorb into 
 themselves the partial and so far erroneous doctrine. More 
 frcquentlv, writers who try to explain the theories of their
 
 THE STARTING-POINT. 3 
 
 antagonists, explain only their own version of those theories; 
 \vhich is apt to be a very different thing. And this suggests 
 that, in order to attempt a satisfactory conciliation, we have 
 first to raise a preliminary question as to the most promising 
 method for reaching such a conciliation. 
 
 3. Now, if we ask what is necessarily implied in the 
 attainment of a theory at once satisfactory in itself and wide 
 enough to exhibit all preceding theories as imperfect and 
 one-sided views of the whole truth, the prospect is rather 
 alarming. For ethical controversies spring from the ultimate 
 problems of all thought. Full conciliation can only be reached 
 when we have attained a definitive system of philosophy. 
 To reconcile moralists, we have to solve the problems which 
 have ac^itated without beino- set at rest ever since men beran 
 to reflect upon the nature of truth in general. We have to 
 decide upon controversies- where the advance to unanimitv, 
 if it exists at all, is foreshadowed not so much in the un- 
 equivocal triumph of either partv, as in the gradual modifica- 
 tion of the old issues in such a sense that mutual understand- 
 ing is becoming rather more possible. The problems are 
 hardly being answered, but perhaps are being put into a form 
 in which they may some day admit of an answer. If an 
 agreement between moralists is to be adjourned until the day 
 in which all metaphysical puzzles will have been solved, we 
 must be content to hand them over to a future generation ; and 
 to say the truth, I believe that our hopes of a perfect conciliation 
 must be adjourned till that indefinitely distant millennium. 
 But another less disheartenino; conclusion is also suo-o-ested 
 by the reflection that, in spite of these difficulties, there has 
 been in other respects so much agreement. Is it not, in fact, 
 conceivable that we may so disentangle the perplexing threads 
 of controversy a\ to separate the questions which admit of 
 some approximate; answer from those which we must be 
 content not only to leave unsolved, but to consider as of 
 such a nature that even the method of solution, and the kind 
 of solution possible and desirable, must be regarded as hither 
 to unsettled problems ^
 
 4 THE SCIENCE OF ETHICS. 
 
 4. This will be clearer if we remark that the difficulty 
 in question is by no means peculiar to ethical speculations. 
 On the contrary, we may say that it is common to all 
 branches of knowledge. We possess a large body of estab- 
 lished scientific truths — established, that is, in such a sense 
 that their validity would be questioned by no school 
 of metaphysicians. Geometry, for example, includes a 
 vast body of propositions, which are admitted by all 
 reasoners to possess the highest degree of certainty. Yet 
 geometry involves the conception of space, and the true 
 nature of that conception is precisely one of the problems 
 upon which metaphysicians have disputed most eagerly and 
 interminably. The geometrician proceeds without troub- 
 ling himself in the least about such argumentation, and 
 in the judgment of the metaphysicians themselves he is 
 perfectly right. Space, according to some thinkers, has only 
 a phenomenal reality ; if so, the truths of the geometer are 
 only applicable to phenomena. Space, as we know it, may 
 perhaps be only one out of various possible kinds of space. 
 If so, the doctrines of geometry can only be regarded as 
 certain for that kind of space with which we are conversant. 
 The geometer postulates space, and (it may be) one particu- 
 lar kind of space ; but the validity of his reasoning, so long 
 as that postulate is granted, remains unaffected. He is not 
 concerned to inquire whether, in a different universe or in 
 some transcendental world, his theorems mav require modifica- 
 tion or mav cease to be in any way applicable. It is fortu- 
 nate that this is so, for otherwise we should have no know- 
 ledge until we had reached the ultimate jroal of kn()wledg:e. 
 Similar remarks are applicable to all the physical sciences. 
 We mav say, in mathematical language, that the formulae 
 obtained by the scientific reasoner include constants reo;arded 
 by him as ultimate elements, though the metaphysician 
 may discover that these constants themselves require further 
 analysis, and are possibly dependent upon some further 
 condition for their apparent constancy. But the formulae 
 may be equally valid within the sphere of science what-
 
 THE STARTING-POINT. 5 
 
 ever may be the metaphysical interpretation of these con- 
 ceptions. The relation holds though the nature of the 
 things related mav be uncertain. If space be only phenome- 
 nal, the limitless properties will not hold in the noumenal 
 world ; but in this world, where phenomena are of so much 
 importance, they will serve as well as ever for our guidance. 
 Thus we may sav, in legal metaphor, that the man of science 
 does not go behind the record. He takes the authority of 
 his immediate perceptions as final, and leaves it to the meta- 
 physician to discover, if he can, some further justification of 
 the proceeding. 
 
 5. I am content, for the present, to assume this much, with- 
 out attempting to define more accurately the relation between 
 metaphysical and what are generally called scientific reason- 
 ings. However that line may be drawn, we may admit that, 
 in the case of the physical sciences at least, we can obtain 
 knowledge which, within its own sphere, is entirely indepen- 
 dent of the metaphysician's theories. Is not this true in all 
 cases, and therefore in those cases in which the science is 
 concerned with the conduct and character of human beings? 
 May we not discover propositions about the relations of men to 
 each other and the internal relations of the individual human 
 being which will be equally independent of metaphysical 
 disputes? As we assign the relations between parts of space 
 without asking what is space in itself, mav we not determine 
 rules about men without asking what is meant, for example, 
 by personal unity, or what is the true mode of distinguish- 
 ing object from subject? It is true that in these more com- 
 plex investigations we are constantly on the verge of meta- 
 physical abysses. The inquiry into the laws of thought is 
 very apt to complicate itself with inquiry into the nature of 
 thinking beings. The difficulty has given rise to many 
 controversies. From one point of view, metaphysics has 
 seemed to be merely a department of psychology, as the 
 nature of men's thoughts seems to be dependent upon the 
 constitution of the men who think. From another point of 
 view, the opposite extreme seems to be more tenable, as the
 
 6 THE SCIENCE OF ETHICS. 
 
 laws of particular modes of thought must form a special de- 
 partment of the inquiry into the laws of thought in general; 
 and thus psychology must be taken up into and absorbed in 
 metaphysics. And, in fact, as the case of ethical speculation 
 itself proves, it is very difficult to reason about any question 
 involving psychology without gliding into the ultimate meta- 
 physical problems. And yet a discrimination of the two 
 elements, though harder to maintain, may be as conceivable 
 here as in the physical sciences. For it is a fact — and a 
 fact for which we may again appeal to these ethical specula- 
 tions — that there are a large number of propositions, which 
 we are daily and hourly using, which refer to human 
 nature, and which are presumably independent of meta- 
 physical speculation. Like the geometrical truths, they 
 involve elements which may require further analysis^ but, 
 like the geometrical truths, they will hold good, or at any 
 rate hold good within a very wide sphere, whatever the 
 results of such analysis. 
 
 6. Take, for example, any one of those theorems of which 
 we constantly assume the truth. Most mothers, we say, love 
 their children. That is a statement which conveys a distinct 
 meaning, and a statement of which we can discover and test 
 the truth without appealing to any metaphysical system. In 
 order to express it at all, we have to use some conceptions which 
 the metaphysician professes to explain. We tacitly assume a 
 distinction between subject and object, a distinction between 
 persons, a causal relation between certain phenomena, and so 
 forth ; and metaphysicians will proceed to deal with these dis- 
 tinctions and relations after their own fashion. They may tell 
 us, for example, that we can transcend the distinction of subject 
 and oljjcct; that persons, though phenomenally different, are 
 different manifestations of an identical substance; that the rela- 
 tion of cause and effect is a mere illusion, or that it implies a 
 transcendental power behind the phenomena. But all such 
 statements, true or false, have not the very slightest influence 
 upon our belief in the asserted fact. They state some- 
 thing supposed to be implied in the fact; but something also
 
 THE STARTING-POINT. 7 
 
 remains unaltered, whatever be implied. Nor is there the 
 least reason to suj)pose that the phrase in question conveys 
 a different meaning to metaphysicians of opposite schools. 
 Solomon's famous experiment upon motherly love was not 
 suggested by nor did it require to be reinterpreted by metaphy- 
 sical theorems. The follower of Hegel means in all probability 
 precisely the same thing as the follower of Hume when he 
 says that a mother loves her child; though when they come 
 to reflect upon certain ulterior imports of the phrases used, 
 they may come to opposite conclusions. The formula 
 remains the same ; for all purposes of conduct it evokes the 
 same impressions, sentiments, and sensible images, and it there- 
 fore represents a stage at which all theories must coincide, 
 though they start, or profess to start, from the most opposite 
 bases. 
 
 7. I say, therefore, that without making anv metaphysical 
 assumptions, there is a region in which all metaphysical tenets 
 are indifferent. This is the recrion of science: and thoucrh 
 I cannot here attempt to define its limits, I consider that 
 there is at least a presumption that some of the moral ques- 
 tions of which we are to speak, fall fairly within this region, 
 and thus within a region to which the ordinary scientific 
 methods are applicable. In order that we may reach a 
 scientific conclusion in the sense here suggested, two con- 
 ditions are necessary. The first is that with which we have 
 been dealing, namely, that we should be able to discover rela- 
 tions that are unaffected by the metaphysical elements which 
 they contain, which shall be true so far as they go, whatever 
 our metaphysical theories. I do not dwell longer upon 
 this question, because the most satisfactory way of showing 
 that it can be complied with, is to comply with it; that is, 
 to lay down propositions to which, in fact, metaphysical 
 inquiries are plainly irrelevant. And this I shall attempt in 
 the following pages. But there is another condition upon 
 which I must dwell more fully. Scientific knowledge means 
 simply that part of knowledge which is definitive and 
 capable of accurate expression. It is merely the crystallised
 
 8 THE SCIENCE OF ETHICS. 
 
 core of the vagi;ue mass of indefinite and inaccurate knovvlcdire. 
 It reaches the highest or most strictly scientific stage when it 
 admits of being stated in precise propositions of unconditional 
 vaUdity. By unconditional, I mean, of course, that the con- 
 ditions under which it holds are given in the proposition 
 itself. It may be as much a scientific truth that all mam- 
 malia are air-breathing animals as that all particles of matter 
 gravitate to each other. It is not necessary for the scientific 
 character of the proposition that the existence of mammalia 
 should be unconditional, but only that the existence of one 
 property should be the sole and sufHcient condition for the 
 existence of the other. But this unconditionality and pre- 
 cision represents an ideal which is seldom if ever realised; 
 and propositions are generally called scientific which make 
 some approximation to this quality. They are approximately 
 scientific when they are precise enough to afford a gener- 
 ally reliable rule, and require conditions which are gener- 
 ally fulfilled ; or, in other words, when they do not fully 
 reveal and formulate a "law of nature," but make some 
 approximation to that completeness of statement. We 
 should not call the proposition " Mothers love their children " 
 a scientific proposition in the highest sense, because it is 
 incapable of precise statement as an unconditional truth. 
 Some mothers do not love their children. We have no 
 means of giving any quantitative measure of the passion of 
 love, and when we compare this statement with a purely 
 scientific statement — as, for example, that gravitation varies 
 inversely as the square of the distance — we are at once 
 sensible that our formula falls altogether short of the re- 
 quirements of science as we understand it in other branches 
 of inquiry. Still the statement presumablv includes an ap- 
 proximation to son)e scientific truth, and it is worth while 
 to consider how far it may be rendered scientific, for in 
 the endeavour we shall throw some light upon the problem 
 before us.
 
 DIFFICULTY OF MORAL SCIENCE. 
 
 II. Difficulty of Moral Science. 
 
 8. I shall only touch in the briefest way upon one objection 
 to the possibility of a scientific treatment of moral questions 
 which would be fatal, could it be sustained. I refer to the 
 objection founded upon the doctrine of the so-called freedom 
 of the will. I must decline to enter upon a controversy 
 already thrashed out to the very last fragments of chaff. It 
 is enough in such a case to indicate one's own position, and 
 to refer for the arguments by which it may be supported to 
 the innumerable writers who have considered it at length. 
 Hereafter, indeed, I shall be obliged to say something of 
 the questions under another aspect. For the present, I 
 may observe that the theory of free will is relevant so far 
 as it affects or is supposed to affect what has been called 
 the universal postulate. In all reasoning about facts, this 
 postulate, whatever its true nature, must be invariably 
 assumed. Whether we speak of the uniformity of nature, 
 or of the principle of sufficient reason, or of the universality 
 of causation, we are adopting different phrases to signify the 
 same thing. To me, indeed, it appears that the theorem, in 
 whatever form it may be most fitly expressed, is not so much 
 a distinct proposition, the truth or falsehood of which can be 
 discussed, as an attempt to formulate the intrinsic process of all 
 such reasoninjr. Unless we assume that identical inferences can 
 be made from identical facts we are simply unable to reason 
 at all. The alternative to making the assumption is not to 
 admit some other possibility, but to cease to think. If there 
 is nothing arbitrary in nature; if a thing can at once be and 
 not be ; or if the same cause may produce different effects, the 
 very nerve of every reasoning process is paralysed. We can 
 no more argue as to phenomena than we can make a formal 
 svllogism if we suppose that contradictory propositions are 
 not mutually exclusive. Further, I can see no ground what- 
 ever for excluding the case of human conduct. I infer a 
 man's actions from his character and circumstances, or his
 
 lo THE SCIENCE OF ETHICS. 
 
 character from his actions with the same confidence as I infer 
 the path of a planet from the known determining forces, or the 
 forces from the path. If two men act differently there must 
 be some corresponding difference in the character or circum- 
 stances, as if two bodies produce different reactions there 
 must be some corresponding difference in their chemical com- 
 position. Now, if the doctrine of free will be inconsistent 
 with this theory, I must simply say that I reject it. If it be 
 consistent with the theory, I have at present nothing more 
 to do with it: for it is only in so far as it is inconsistent that 
 it affects the possibility of a scientific treatment of ethics. 1 
 shall only add that, in any case, to reason about conduct is to 
 assume that it is determinate. If actions be intrinsically 
 arbitrary, or in so far as they are arbitrary, a theory of action 
 must be a contradiction in terms. And thus, as it has been 
 said, that whether we are or are not free, we must act as 
 though we were free, I may say that whether conduct be or 
 be not determinate, we must reason about it as though it 
 were determinate. 
 
 9. Passing over this difficulty, the question still remains, 
 \vhether any science of human nature be possible in any other 
 sense than this : that its existence does not imply a contradic- 
 tion in terms. We have already the names of such sciences 
 as sociology and psychology. Are they anything more than 
 names, or is there any reason to suppose that they will ever 
 be anything more ? These problems are not impossible in 
 the sense in which the problem of perpetual motion is im- 
 possible; for there is an answer to them, if only we could 
 find or express it; but they may be impossible, in the sense in 
 which it is impossible for an infant just learning arithmetic 
 to calculate the movements of a planet. It can be made 
 intelligible to him that the proposed operation has a meaning 
 and an answer for more developed minds ; but even an 
 approximate calculation is altogether beyond his powers. 
 At present the so-called sciences consist of nothing more 
 than a collection of unverified guesses and vague generalisa- 
 tions, disguised under a more or less pretentious apparatus of
 
 DIFFICULTY OF MORAL SCIENCE. n 
 
 quasi-scientific terminology. Can they ever be made precise 
 and certain ? 
 
 lo. It is easy to show how far we are at present from any 
 such consummation. The accepted test of true scientific 
 knowledge is a power of prediction. In the typical science of 
 astronomy that test has been satisfied. We can trace back- 
 wards and forwards through many ages the series of pheno- 
 mena of the solar system. That we can do so is owinci- to 
 the simplicity of the data and the conditions. Given the 
 law of gravitation and the masses and initial position of two 
 moving bodies, we can apply our formulae and work out their 
 movements. Yet even for this typical science it required 
 a vast genius, and such a genius inheriting the labours of 
 many generations of careful observation and acute reasoning, 
 to work out the elementary problem. Complicate the con- 
 ditions in the slightest deo-ree, add the attraction of another 
 body, and the problem becomes so unmanageable that the 
 severest labours of the acutest minds, during generations of 
 unparalleled progress, are unable to give us more than an 
 approximate solution. The law which is embodied in the 
 movements of a planet lies, so to speak, on the surface of 
 things; it is given in the greatest simplicity with the 
 minimum of over-ridino; and conflictino- influences: and 
 therefore it is that we are able so far to disentangle the 
 phenomena and reach an applicable result. In other cases, 
 we have to perform a process of abstraction which is here 
 done to our hands. Elsewhere we can obtain simple rules only 
 by disregarding the heterogeneous mass of modifying circum- 
 stances with which the simple process is inextricably involved 
 hi concrete cases. And therefore we can only obtain cer- 
 tainty of prediction when we can ourselves so modify circum- 
 stances as to make them correspond to our ideal constructions. 
 We can make a conditional prediction in regard to many 
 concrete events; we can say that such and such processes 
 will develop themselves if no interference takes place from 
 without ; or we can predict if we are able to obviate the risk 
 of such interferences. We can be sure of an experiment in
 
 12 THE SCIENCE OF ETHICS. 
 
 a laboratory, but our knowledge in regard to any series of 
 events not voluntarily prepared is limited by the constant 
 possibility of the interference of conditions which we have 
 not foreseen or are unable to appreciate. And, thus, even 
 where science is most perfect, our powers of proving it are 
 as a rule bound within narrow limits. 
 
 II. Is it not, then, a mockery to join in one phrase such 
 ■words as science and human nature? Will Caesar cross the 
 Rubicon ? The problem is as determinate as the problem. 
 Will a projectile fall on the hither or the further bank ? But 
 consider for a moment the conditions which must be taken 
 into account in a solution. We must know the whole range 
 of operative forces which will allow play for his will to act, and 
 therefore have some neoative information at least as to the uni- 
 verse at large ; we must know what are the political, social, 
 geographical circumstances which may possibly affect his de- 
 cisions; we must know, again, how these complex facts 
 mirror themselves in his mind; what are his calculations, his 
 ambitions, prejudices, hopes and fears; we must know even 
 what freaks of memory or sudden associations of names 
 and incidents may stir the current of his meditations; we 
 must understand the laws of his character, and how it 
 depends upon the state of his liver, his digestion and the 
 electric tension of the atmosphere ; for " character " is the 
 name of an undecipherable mass of sensibilities, inherited and 
 acquired habits of reasoning and feeling, changing from day 
 to day, baffling all calculations and eluding the shrewdest 
 observer. The wiliest diplomatist trusts to crude generalisa- 
 tions, which fail as often as they are verified; our most 
 confident anticipations of the friend whom we knew to his 
 heart's core are put out by a toothache or a grain of sand in 
 the wrong place ; history is a record of " fears of the brave 
 and follies of the wise," of confident anticipations falsified, 
 and revelations of character which astonish no one more 
 than the agent himself. Is it not as futile to apply the name 
 of science to our guesses about the shifting and uncompre- 
 hensible network of thouoht and feeling which makes a
 
 DIFFICULTY OF MORAL SCIENCE. 13 
 
 character as to give it to the vague guesses of a shepherd who 
 hopes a fine morrow from a red sunset? Even the "scien- 
 tific" meteorologist can scarcely foretell a day's weather; and 
 vet his problem is simple and his knowledge accurate when 
 compared with the vague surmises which we dignify by the 
 name of sciences of human nature. 
 
 13. But let us look a little closer. We are pitching our 
 standard too high. The test of predicting the result of any 
 concrete set of processes is, as I have said, rarely if ever satis- 
 fied. There is always the tacit condition, that no interference 
 will take place with our assumed data. I know that a stone 
 thrown into the air will descend ; I can calculate its path 
 with minute accuracv, and define its range and velocity. 
 But this is, of course, upon the hypothesis that it meets 
 with no obstruction. As every event in the universe 
 may be said to be more or less connected with every other 
 event, it is plain that this condition can rarely be secured 
 with absolute certainty. But I do not refuse the name 
 of science to the general propositions which enable me to 
 make this conditional prediction. On the contrary, they 
 form the great mass of scientific knowledge. We must, there- 
 fore, consider more closely how far we can discover such rules 
 about human conduct. It is impossible to say whether 
 Ceesar will or will not pass the Rubicon. It may be pos- 
 sible to sav that most men would or would not cross ; 
 or, asrain, to sav that Caesar will cross under certain conditions, 
 which again may correspond to the conditions most com- 
 monly fulfilled, and which may therefore give a certain de- 
 gree of very useful knowledge. We may make some such 
 statements with absolute confidence. Man, so far as he is a 
 ump of matter, obeys certain mechanical laws. Throw him 
 out of a window, and he will fall as though he were a lump 
 of lead or a corpse. Chemical changes may be worked in 
 his tissues as assignable as though he were simply a gas or a 
 salt. So far as he is a living organism, we have to consider 
 another set of conditions. He is the embodiment of a set of 
 processes which are always taking place, and which are not
 
 14 THE SCIENCE OF ETHICS. 
 
 tlcduciblc from the external circumstances. To know how he 
 will act, we must sometimes know at what stage they happen 
 to be. When he is thrown out of a window, this considera- 
 tion is irrelevant. If we want to know how he will act when 
 food is presented to him, they become relevant. We must 
 know whether he is hungry or satisfied, awake or asleep, and 
 so forth. But further, as he reasons and has motives, he res- 
 ponds to any external stimulus after a more complex fashion. 
 His actions are regulated not only by the immediate action of 
 his senses, but by innumerable anticipations of the future, 
 recollections of the past, and inferences as to distant objects. 
 So far as he is a thing or an animal, it is comparativ^ely easv 
 to determine his conduct. Given a starvino; doo- and a 
 lump of meat in contact, and vou can predict the result. 
 But to determine the behaviour of a human being with a glass 
 of water presented* to his parching lips, vou must be in posses- 
 sion of an oro;anon for calculating; the action of human motive, 
 and be able to unravel the tanoled skein of thought and feel- 
 ing which varies enormously from one man to another, and 
 is determined by innumerable subtle influences, each of which 
 sets all calculation at defiance. 
 
 13. 7\) attack such problems as these, we require some 
 tenable psychology. We may say that conduct is determined 
 by pain and pleasure. W^e may repeat all the little maxims 
 which have been made popular by the observers of human 
 character. But, to gain anvthing like scientific knowledge, 
 we require some mode of estimating these pains and pleasures, 
 and the faculties to which they appeal. If we had but a 
 sincrle passion, if we were only locomotive stomachs like a 
 polyp, the problem would be simple. But man is a complex 
 of numerous passions, a little hierarchy of conflicting, co- 
 operating, mutually interacting emotions, which it is simply 
 impossible to tabulate and measure. Numerous classifica- 
 tions have been made by acute observers. Yet I do not 
 suppose that anv of these can be regarded as more than a 
 rouah account of the most obvious facts. To ask which are 
 the primitive and elementary passions, how they are related.
 
 DIFFICULTY OF MORAL SCIENCE. 15 
 
 and how the derivative passions are compounded, is to ask 
 questions which admit of no definite answer. We are simply 
 feeling about in the dark ; putting rough guesses into preten- 
 tious shapes, or dressing poetical symbols in the language of 
 science; but we are far, indeed, from anything which can 
 for a moment deserve our confidence as a scientific analysis. 
 
 14. If, again, we could rely upon any such analysis, we 
 should still be at the very outset of our task. Who can say 
 what is the relative importance of the various parties in the 
 little internal parliament which determines our policy from 
 one moment to another, or by what subtle and inscrutable 
 ties they are connected? Take, for example, the simple 
 physical appetite of hunger. No doubt hunger determines 
 many human activities, and we have a strong probability that 
 a hungry man will eat when he has a chance. But how are 
 we to say what felt hunger will play in any given organisa- 
 tion? What are the conditions, external or internal, which 
 determine its strength and its influence upon our actions? 
 It depends upon innumerable and indefinable conditions ; 
 upon the state of the constitution ; the condition of the palate, 
 and upon the various habits which we have acquired. It 
 may be suspended entirely by grief, or by the affection of 
 some bodily organ not obviously or normally concerned in 
 deteijmining its strength. Under certain conditions it be- 
 comes an overpowering impulse, capable of overcoming every 
 moral and prudential restraint. It causes mothers to desert 
 their children, and breaks up the bonds of the discipline which 
 has become a second nature. Under other circumstances, it 
 loses all power. The glutton turns with loathing from the 
 food which enthralled him a moment before, not merely if he 
 is satiated, but in obedience to some accidental association 
 which has called up a spasm of disgust. What is true of 
 hunger is even more obviously true of the higher and intel- 
 lectual emotions, of which we can only say that their com- 
 position is apparently complex in the highest degree, and the 
 relation to other parts of the system vitally important and 
 yet hopelessly obscure. So, again, if we know something of
 
 -16 THE SCIENCE OF ETHICS. 
 
 the constituent passions, we should still have to investigate 
 the laws of what may be called mental perspective. It is a 
 familiar fact, that the intensity of a passion varies in some 
 way with the proximity of the appropriate object. The 
 prospect of a pain at a certain distance scarcely affects us at 
 all, though we may be perfectly convinced of its approach; 
 and the immediate prospect of the same pain may be over- 
 powering. A man may be utterly indifferent when he knows 
 that he is to be hanged to-morrow, and yet be a hopeless 
 coward on the gallows. His neighbour may tremble at the re- 
 mote prospect, and yet meet death with all imaginable courage. 
 The law, according;: to which distance diminishes the corre- 
 sponding feeling is inscrutable In itself, and varies according 
 to inscrutable differences of temperament. And again, 
 pains and pleasures of different classes vary in this intrinsic 
 power of calling up the premonitory foretaste. A purely 
 physical enjoyment, intense at the moment of fruition, Is 
 often unintelligible in the absence of the immediate condi- 
 tions. A drunkard can easily deny himself his pleasures in 
 prospect, and yet may be impotent in the presence of tempta- 
 tion. At a distance the dlssfrace of drunkenness is strono-er 
 than the appetite for drink. When the drink Is at hand 
 the appetite overpowers the most vivid sense of degrada- 
 tion. The same is equally true of the higher motives. 
 A man habitually careless of duty and little inclined to 
 benevolence may be capable, at moments of excitement, of 
 heroic self-sacrifice. He may, like Falstaff, make a mock of 
 honour when he is cool, and yet when under the stress of 
 immediate shame may throw away life rather than break a 
 code of honour which he regards as ridiculous. 
 
 15. To mention one more familiar difficulty — for I am 
 saying nothing which has not been a commonplace with all 
 observers — I have hitherto spoken as though we always acted 
 upon some conscious motive. But in a large part of our lives 
 we are mere automata. We go through many, perhaps 
 most, of our dally occupations with little more consciousness 
 than a machine; and therefore it is impossible to foretell
 
 DIFFICULTY OF MORAL SCIENCE. 17 
 
 whether a given passion will or will not operate under certain 
 circumstances. Were a man's consciousness always awake to 
 all the consequences of his actions, we could say that he 
 would act in a certain way. But consciousness acts fitfully ; 
 it goes to sleep or rouses itself in obedience to conditions 
 which we are unable to assign. A man would act in one 
 way if the results of the action were present to his mind, but 
 he acts in another way simply because it is easiest for him to 
 act to-day as he happened to act yesterday, l^he momentum 
 of his past life causes him to continue in the groove in which 
 he has placed himself. We are therefore unable to infer 
 conduct from character unless we know what are those 
 accidental chains by which it is bound. A man gives money 
 or savs his prayers, not because he is charitable or devout, 
 but because he has been brought up to do so. Thus the 
 data from which his conduct must be calculated include not 
 only his true character but the special circumstances of his life. 
 His conduct is probably a kind of diagonal between that 
 which would be dictated by his instincts when fully conscious 
 and that which is determined by another independent set of 
 causes. If he were purely mechanical or purely rational in 
 his actions, we could predict his behaviour; but he is alter- 
 nately one and the other, and passes fitfully from the automatic 
 to the conscious stage in obedience to conditions which alto- 
 gether elude our powers of observation. 
 
 16. The conduct of man is as much dependent upon his 
 moral and mental orcjanisation as the various reactions of a 
 body upon its chemical composition. To make this general 
 admission the ground of a science, we must know something 
 more. We must know what are the elementary faculties; 
 what their relative stren2;th ; what relations hold amonirst 
 them in virtue of the unity of the subject ; and what are the 
 laws by which they respond to any given stimulus, come 
 into the foreground of conscious motive, and again pass for a 
 time out of sight, to leave blind habits to occupy their vacant 
 places. And upon such points we are in ignorance so profound, 
 that, far from knowing the answers to our questions, we 
 
 B
 
 1 8 THE SCIENCE OF ETHICS. 
 
 scarcely know how to put them intelligibly. If our psycho- 
 logical armoury is so scantily provided, can we find effective 
 weapons in the nascent science of sociology? Can we reach 
 certainty by studying the social instead of the individual 
 organism ? We do not, as might at first seem probable, find 
 that the difficulties increase in proportion to the extent of 
 the field of observation. We can often lay down trustworthy 
 propositions in regard to an aggregate or an organic whole 
 when we know little of the sejmrate parts. We can deter- 
 mine the properties of a body though we should find it hope- 
 less to trace the movements of a particular molecule; and the 
 idiosyncrasies which make the study of individual character 
 so perplexing neutralise each other, or become unimportant 
 when we are dealing with a whole society. But this only 
 proves that the difficulties do not increase in a compound 
 ratio; it does not show that they are actually less. In 
 truth, when we have to deal with a societv, and try to 
 tabulate and to measure the forces by which it is held to- 
 gether, we find ourselves in presence of perplexities resem- 
 bling those which beset our psychology, whilst the vastness 
 of the data required and the indefinite possibilities of 
 :hange introduce other and apparently insoluble problems. 
 What is the relation, for example, between the religious 
 creed of a nation and its political sentiments? What is the 
 force of the selfish or antisocial passion, and the force which 
 restrains it within such bounds as are consistent with order ? 
 What are the conditions of moral and intellectual progress, 
 and what will be the probable action of any new belief, or of 
 a chanjre introduced in the material conditions of life? To 
 ask such questions is to suggest the extreme vagueness of all 
 our guesses and the immense complexity of the problem. I 
 need not say how shortsighted are the ablest statesmen, and 
 how constantly that which happens is precisely the one thing 
 which nobody foresaw, but which, after the event, appears to 
 have been just what every one should have foreseen. What 
 will be the effect of tcachinc; readino; or writino; ? If the
 
 DIFFICULTY OF MORAL SCIENCE. 19 
 
 question be asked in regard to an individual, we are perplexed 
 because the variability of character and understanding makes 
 it difficult to judge of the effect upon a given intellect. We 
 can speak more confidently of the average effect, because the 
 individual variations will correct each other when we are 
 dealing with a mass. But when we ask the further question, 
 in what way this ascertainable effect (we suppose it to be 
 ascertainable) will react upon the whole social structure? we 
 have another set of problems of the most intricate kind. 
 Will the increase of knowledge make men content or dis- 
 contented ? Will it confirm or shake the beliefs upon which 
 the social order depends? Will it simply strengthen the 
 impulse towards a higher culture, or will it also increase the 
 tendency to self-indulgence and weaken the bonds of discip- 
 line ? If we can give some vague answer to such questions, 
 it is clearly not such an answer as can be called scientific, or 
 as enables us to give any definite prediction of results. It is 
 difficult to predict what will be the effect of special circum- 
 stances upon an individual; but if the society of which he is 
 a member remains approximately unaltered, we can give a 
 fair guess as to the probable consequences of an ascertained 
 modification of his character. But that which is a very slight 
 change in each individual may involve a vast and incalcul- 
 able alteration in the social equilibrium when we suppose 
 that a whole class is affected and the conditions of national 
 life modified through its whole extent. When we reflect 
 upon the extreme difficulty of obtaining the necessary know- 
 ledge, of appreciating the state of mind of millions of men, 
 of discovering the latent passions which may be smouldering 
 amonirst them, their state of accessibilitv to new ideas and 
 new conditions of life, we may well feel the untrustworthiness 
 of our so-called scientific methods. The discovery of a new 
 principle in mechanics or the promulgation of a new reli- 
 gious creed may alter the whole social state or bring about 
 political and social convulsions. But how can we predict 
 new discoveries or new creeds ? To foretell a discovery is to 
 make the discovery yourself, and to make it before its time.
 
 20 THE SCIENCE OF ETHICS. 
 
 To foretell the new creed is to be yourself the potential religious 
 reformer ; and to foretell the effect of both upon the society 
 in which they are promulgated is to trace out a complex series 
 of actions and reactions, to appreciate the state of mind of 
 masses of men, and the mode in which they will be affected 
 by a given thought, and^ moreover, to be provided with a 
 quantity of information as to facts for which there can hardly 
 be room in any individual brain. Any one who should have 
 prophesied the history of the present century at its beginning 
 with any precision would have had himself to foresee the 
 course of science, the attitude taken by the greatest thinkers, 
 the influence upon men's imaginations of new conceptions of 
 the world, and to have traced out an incalculable series of 
 chang-es in the relations of classes, and to determine the effect 
 of all these changes upon the material conditions of existence. 
 Any shortcoming might omit some essential point. In short, 
 the prediction of the course of history, even in general terms 
 and for a brief period, would require an intellect at least as 
 much superior to that of a Socrates as the intellect of a 
 Socrates is superior to that of an ape. Indeed, we may say the 
 greater the intellectual development, the greater is the difficulty 
 of foreseeincr the results which the intellect will obtain. And 
 therefore there is a kind of natural limit to the development 
 of the powers of human nature. For as the intellect becomes 
 more capable of grappling with the old problems, the com- 
 plexity of the problems presented tends itself to increase. 
 
 17. I have dwelt upon such considerations partly to 
 illustrate the nature of the difficulty, and partly to obviate 
 the impression that in using the word " science " I supposed 
 that a science of human nature could either now or at any 
 future time make any approach to the accuracy or certainty 
 of the physical sciences. A shorter road to the same con- 
 clusion might have been taken by simply challenging be- 
 lievers, if such there be, in such sciences, to produce any pro- 
 position which possesses, or even claims, the same sort of 
 authority as belongs to the doctrines of a fully constituted 
 science, or to justify such claims by adducing any instance
 
 ATTAINABLE RESULTS. 21 
 
 of scientific prevision. But I do not anticipate any serious 
 objections to this part of the argument, and it is perhaps 
 more important to consider the more positive results. We 
 have briefly surveyed the ground upon which our super- 
 structure is to be reared, with the result, so far^ that it is too 
 treacherous and unstable to bear any solid edifice. Let us 
 consider next whether anv trustworthy space is left after all 
 these deductions, and what kind of confidence may fairly be 
 challenged for any philosophic construction. 
 
 III. Attainable Results. 
 
 1 8. After all that has been said, it must be admitted, as I have 
 incidentally admitted, that we do in fact possess a consider- 
 able deo-ree of knowled<re as to the conduct of our fellow- 
 creatures. A confidence that our neighbours will act in 
 accordance with certain anticipations is essential to almost 
 every part of our conduct. I do not carry weapons in 
 London now, as I should carry them in barbarous towns, 
 because I am as certain that the passengers will behave 
 peaceably as I am that the houses will not fall on my head. 
 I trust my whole fortune with complete confidence to my 
 bankers or lawyers, because, though I know that there are 
 such thincTs as knavish men of business, the risk is not ereat 
 enough to afiect my conduct. I act as confidently upon the 
 assumption that mothers love their children as upon the 
 assumption that London Bridge will bear the weight of the 
 passing crowds. I have as little doubt that a toothache or 
 a liver complaint will diminish a man's happiness as I have 
 that a stone is indigestible. Moreover, such knowledge 
 sometimes reaches a hioh decree of discrimination. The 
 great dramatist, we say, knows the human heart ; he does not 
 know it as the man of science knows the properties of a 
 chemical compound; that is, he cannot draw out any specific 
 set of propositions to express his knowledge — he cannot give 
 chapter and verse for his conclusions; but he can feel, though 
 he cannot explain, how a selfish or a heroic character will
 
 22 THE SCIENCE OF ETHICS. 
 
 think and act under given circumstances. His power over 
 our sympathies is proportioned to the truth of his divinations, 
 and therefore to the degree in which we too recognise the 
 truth of his portraiture. So the diplomatist or the attorney 
 who watches every phrase and gesture of his antagonist 
 draws conchisions as to their feelincrs and character which 
 are often of startHns; accuracy. He relies, it may be, too 
 much upon certain principles which are frequently inaccurate 
 and never infallible. He can assign them and consider them 
 collectively to embody a knowledge of the world. And 
 though the shrewdest of such men, statesmen, courtiers, 
 solicitors, or confessors, is liable to innumerable blunders, 
 the very capacity of reasoning upon such matters implies 
 that there is some such knowledo-e. There is, in fact, a vast 
 
 O 7 7 
 
 difference in the acuteness with which men judge of character; 
 and as some are acuter than others, it must be that they 
 have at least implicit canons of judgment which are not 
 entirely valueless. 
 
 19. The knowledge thus assumed does not differ in kind 
 from scientific knowledge. There is, in truth, only one 
 kind of knowledge; and knowledge gradually passes into the 
 scientific state as it becomes more definite and articulate. 
 We can hardly distinguish in respect of certainty between 
 much of the knowledge called scientific and much of this 
 floating and indefinite knowledge. A very vague proposition 
 may often give rise to certai-nty inasmuch as it excludes many 
 conceivable hypotheses. I do not know exactly how long 
 a man may live, but I am quite certain that he will not live 
 for a thousand years. If I know a man to be a coward, 
 I cannot say exactly what will frighten him, but I can be 
 very certain that he would run away from a fiery flying 
 dragon. But knowledge, however certain, remains in the 
 unscientific stage so long as a proposition is of such a nature 
 that I cannot define the conditions under which it will hold 
 true. My certainty that a boat into which I am about to 
 step will support my weight is not greater than my certainty 
 that a mother will try to save her child from drowning. In
 
 ATTAINABLE RESULTS. 23 
 
 both cases my knowledge as to the particular fact is imme- 
 diately founded upon a rough estimate of the facts before me. 
 I rehearse the events in imagination, and my imagination 
 may be able to rehearse them accurately. But there is this 
 difl'erence : that in the case of the boat I can state the general 
 conditions of floating and sinking with an accuracy which 
 may pass for absolute. I know that the result will depend 
 upon the relations between the weight of the boat and the 
 weight of the water displaced, and upon other precisely mea- 
 surable and ascertainable conditions. I cannot make a similar 
 statement in the case of the mother. My psychology aifords 
 me no tenable analysis of the passions called into play, and 
 no measure of their intensity. There are such things as " un- 
 natural " mothers, and I cannot say what innate sensibilities 
 or what subsequent culture are required in order to develop 
 the more normal intensity of feeling. Thus, though the 
 strength of my conviction in a given case may be fully justifi- 
 able, it does not afford a safe generalisation. I know that 
 the rule will fail occasionally, and I cannot tell when. If, 
 therefore, I try to discover a principle of universal validity 
 under which the particular case may be ranged^ I am at a loss. 
 I may, in my ignorance, be omitting the essential conditions 
 and retaining those which are contingent. I explain the sink- 
 ing or floating of the boat by referring to a theory which, upon 
 whatever grounds, the man of science holds to be immutably 
 or unconditionally true. But I cannot regard the principle. 
 Mothers love their children, as having this degree of validity, 
 as it is liable to exceptions and to limitations dependent upon 
 some hitherto unknown principle. Thus the highest degree 
 of certainty obtainable is represented by the assertion that 
 most mothers, and not that all mothers, love their children. 
 
 20. Now for many practical purposes this conclusion is 
 amply sufficient, and I must refer for its logical justification 
 of the belief which it represents to works upon the theory of 
 induction. In any case, it represents one of those beliefs 
 upon which we are in daily and hourly reliance. But the 
 difficulty which I have noticed is one which must be felt
 
 24 THE SCIENCE OF ETHICS. 
 
 when we try to give any scientific theory of human conduct, 
 and which, as I need hardly observe, is constantly brought 
 forward against all empirical systems of morality. For the 
 scientific reasoner must endeavour to show not only that 
 things are so and so, but that they could, not have been 
 otherwise. I do not mean by this that he is called upon to 
 give an a priori deduction of the laws which he investigates. 
 It is clear enough, after my previous remarks, that this, in 
 the case of psychological inquiries at least, would be to 
 demand impossibilities. But he must at least give some 
 answer to the question why are things so and so ? So long 
 as his generalisation merely amounts to a statement of 
 average behaviour of the phenomena, it does not give us 
 a law. It does not tell us under what circumstances the 
 rule will hold good, or what would be the consequences of a 
 deflection from the rule. In the assumed case, it is only 
 when a man can tell us, however va2:uelv, what are the 
 conditions of maternal love, or what organic change is im- 
 plied by its absence, that we approach a scientific theory of 
 the passions. It is just this question which is made so diffi- 
 cult by the absence of any tenable psychology. The bare 
 fact that, according to our experience, maternal love is the 
 rule, does not enable us to say whether it is a mere accident, 
 or fashion, something which might be present or absent 
 without any material difference, or, on the other hand, a 
 passion rooted in deepest grounds of human nature, and only 
 to be removed by a radical alteration in the whole organic 
 system. In short, human nature, vipon this first superficial 
 examination, seems to be a mere aggregate of faculties which 
 take one form at one moment and another at the next, but 
 which does not afford a sufficiently plain ground for scientific 
 inquiry. 
 
 31. Without asking at present how far this affects any 
 ethical system, I may next observe that the diflSculty is 
 sometimes attacked by simply extending the area of observa- 
 tion or increasing its accuracv. The facts of conduct are, 
 
 it is suggested, determinable, and determinable with a great
 
 ATTAINABLE RESULTS. 25 
 
 dcoree of accuracy. Little as we know of the psvcho- 
 lo'TV of maternal love, we may discover, by simply counting, 
 what proportion of mothers commit infanticide in a given 
 space of time, and we may prophesy as to the number who 
 will commit infanticide in a similar period of the future. 
 Relying upon such facts, we may pass over the psychological 
 difficulties already noticed, or the metaphysical puzzles about 
 free will; for we have a definite basis of objective fact, which 
 at any rate supplies us with many necessary data. I do not 
 for a moment desire to under-estimate the importance of such 
 a procedure, and yet it must be noticed that it does not help 
 us very far towards a solution of our problem. The general 
 fact which is revealed by such observations is one which we 
 have already taken for granted. It is proved, for example, that 
 the number of infanticides bears a fixed proportion to the 
 population. If it were proved, as people sometimes seem to 
 fancy, that this proportion is always observed whatever the 
 social condition in other respects, we should indeed be in 
 presence of a very startling fact. But it need hardly be said 
 that this is not proved, and that the very purpose of statistical 
 inquiries implies a belief that it is not a fact at all. What 
 is proved is what we have taken for granted, namely, that 
 under the same circumstances — including both external cir- 
 cumstances and the character of the race — human conduct will 
 not alter. This is the doctrine of the uniformity of nature, 
 already assumed in all our arguments; and the utmost that 
 can be said is that statistical inquiry proves that circumstances 
 are more constant than we might have supposed. There is 
 nothing strange in the fact of our guesses on such matters 
 turnino- out to be more or less erroneous. Our problem, 
 however, lies beyond this. We count the number of infan- 
 ticides because we hope to find the law according to which 
 the number varies; to show how it depends upon the density 
 of the population, its comfort or misery, its state of civilisa- 
 tion or barbarism, and so forth. Unluckily the facts which 
 we have thus ascertained with numerical precision only 
 briniT us in face of new difficulties. We are trying to find
 
 26 THE SCIENCE OF ETHICS. 
 
 an answer to a very complex problem, and are applying what 
 is called the method of " concomitant variations." It appears, 
 let us suppose, that the number of infanticides is greater in 
 one place and period than in others. Then we have to ask, 
 why is it greater? And we immediately discover that the 
 relevant conditions are so numerous and intricate that we are 
 scarcely further advanced than before. The religious convic- 
 tions of a race, its social arrangements, its state of material 
 prosperity, and so forth, may all exercise an influence, and we 
 are generally left to conjectures, every one of which mio:ht 
 excite controversy. It may certainly be doubted whether 
 any plausible theory has really been suggested or confirmed 
 by this method ; nor will this cause surprise if we consider 
 how numerous and varied are the experiments necessary to 
 reveal a new property in some substance capable of scientific 
 treatment, and how exceedingly limited is our power of trying 
 scientific experiments upon human society. The difficulty, 
 in fact, is so great, that the ablest representatives of empirical 
 methods have agreed to give up the effort as impracticable. 
 
 32. They have sometimes professed to meet the difficulty 
 by a method illustrated by the case of political economy. 
 We may, it is said, discover certain truths by an indirect 
 method, though the application of direct methods is im- 
 practicable. Political economy is said to be founded on 
 the hypothesis that human beings are actuated by the desire 
 for wealth, and thus involves conclusions which are true so 
 far as that assumption is true. Inversely, therefore, we may 
 discover from the conformity of our deductions to observed 
 facts how far our primary assumptions were justifiable. To 
 me it appears that the statement is inaccurate. Economists 
 undoubtedly assume that most men prefer a guinea to a 
 pound in commercial transactions; but they also assume 
 beyond this that men are influenced by all those passions, 
 whatever they may be, which enable them to live together 
 peaceably, to co-operate in innumerable ways, and to put con- 
 fidence in each other's dealings. This implies far more than a 
 desire for wealth — as, for example, respect for property — and
 
 ATTAINABLE RESULTS. 27 
 
 therefore the existence of all that complex system of regula- 
 tions which prevents the desire for wealth from manifesting 
 itself in cutting throats; and thus the science — if it deserve 
 the name — may give results which are valid so long as the 
 existing organisation holds together, that organisation being 
 manifestly dependent upon countless instincts, beliefs, and 
 so forth, which lie altogether beyond the scope of the 
 economist. But the organisation may change, and has in 
 fact changed within historical times, in consequence of the 
 development of processes upon which he can throw no light. 
 Thus his power of dealing with any series of social pheno- 
 mena is confined within narrow limits. It is concerned, I 
 should say, with the external and mechanical relations 
 between different parts of the social organisation, not with 
 the principles of vital growth. The economist can investi- 
 gate with great advantage such problems as the incidence of 
 a tax or the effects of free trade. He can show what 
 are the channels alonec which wealth circulates whilst the 
 existing system is unaltered ; but as soon as he goes further, 
 he is in presence of the old difficulties. He shows, for 
 example, that an increased population means a diminished 
 share of wealth to each person, and that an increase 
 of wages means a diminution of employers' profits. For 
 short times, and assuming no correlative social change, 
 his doctrine may be unassailable; but if we ask what will 
 really be the effect of limiting the population or of a 
 sudden rise of wages, we have to ask other questions of most 
 vital importance. Will the increased prudence mean a 
 general lowering of energy? Will it imply the adoption 
 of practices fatal to national morality, and therefore to 
 its industry ? Will an increase of wages increase drunken- 
 ness or stimulate saving? and if so, within what limits and 
 with what results ? These problems, again, involve innumer- 
 able moral, social, and religious questions, to which the eco- 
 nomist, as such, has no answer to give ; and yet, if unable to 
 answer them, he cannot fully solve even the economical 
 question. These considerations are enough to show the
 
 28 THE SCIENCE OF ETHICS. 
 
 feebleness of the supposed attempt to isolate a particular im- 
 pulse and consider its consequences apart. The intricate 
 actions and reactions between different elements of the indi- 
 vidual and the social organisation set all such attempts at 
 defiance. The economist is really forced to consider as con- 
 stant precisely those forces which the sociologist has to regard 
 as incessantly changing, and whose mutual relations are the 
 very subject-matter of his inquiry. And thus the results of 
 economical inquiry, however valuable in themselves, are 
 restricted to a limited sphere, and take for granted the very 
 points which we are concerned to investigate. 
 
 23. Meanwhile, however, the establishment of the fact that 
 such laws are ascertainable, even if their significance be 
 exaggerated, has undoubtedly been of great service towards 
 advancing scientific methods. The remark generally elici- 
 ted by their enunciation, though founded on a miscon- 
 ception, explains their real value. They are regarded, in 
 fact, as tending to confirm a fatalistic theory. If murders, 
 it is said, are constant in number, and yet murderers are 
 not moved by any desire to make up a given tale of crimes, 
 the observed uniformity must be due to some mysterious 
 agency. Some dark fate must pick out a certain number of 
 men every year and order them to cut (or, as we must add in 
 equity, to abstain from cutting) their neighbours' throats. 
 And the criticism would be just if it were asserted that the 
 number was constant whatever the social state. The paradox 
 results from the fact that, on the one hand, a regularity in a 
 number of events implies that the events have a direct influ- 
 ence upon each other, whilst, on the other hand, it seems 
 clear that in the supposed cases each man acts in complete 
 independence of his neighbour. There would be nothing 
 strange in the fact of a fixed number of murders if there 
 was such a relation between the events that each murder 
 successfully achieved diminished the temptations in some 
 given proportion; if, for example, there was a limited number 
 of tyrants who united together, and the murder of one enabled 
 the others to guard themselves more effectually. But it is
 
 ATTAINABLE RESULTS. 29 
 
 easy to remark that uniformity of this kind may equally well 
 result where each event is a collateral and independent eflect 
 of certain fixed causes. One mother does not commit in- 
 fanticide, nor does another give birth to a blind child, because 
 another has murdered her child or had a blind babv. But 
 the numbers will be uniform so long as the predisposing 
 causes are uniform and act upon similar material. If cer- 
 tain unhealthy conditions of life remain constant, so will 
 the diseases which they produce. Where there is a certain 
 quantity of undrained marsh, there will be a certain quan- 
 tity of agU'';, and (as some philosophers have urged) a cer- 
 tain quantity of belief in hell-fire, until the discovery of the 
 virtues of bark may eradicate both evils. Now the condi- 
 tions which produce murder are not of this palpable and 
 material kind, but it does not follow that thev are not 
 equally fixed. Society, in fact, is a structure which, by its 
 nature, implies a certain fixity in the distribution and relations 
 of classes. Each man is found with a certain part of the joint 
 framework which is made of flesh and blood instead of bricks 
 or timber, but which is not the less truly a persistent structure. 
 There is room for so many rich and so many poor, for such 
 and such fixed numbers of lawyers, and clergymen, and 
 scavengers. The structure can be modified, and is always 
 being gradually modified; but it can only change gradually, 
 for each change involves the reorganisation to some extent 
 of the rest of the body, and a complex system of action and 
 reaction. The social body is no more liable to arbitrary 
 chancres than the individual body, thoucrh its oraanlsatlon is 
 not so extern-ally conspicuous ; and it is really no more 
 surprising that there should be throughout long periods a 
 fixed proportion of paupers than that through much longer 
 periods animals should be produced with the same bodily 
 proportions. Whilst this is so there will be an approximately 
 constant number of persons in the same bodily and mental 
 state, exposed to the same conditions and temptations; and 
 therefore, again, there will result a number of minor uni- 
 formities which appear surprising so long as we take each
 
 30 THE SCIENCE OF ETHICS. 
 
 man separately and regard him as a being independent of 
 his neighbour, but which are perfectly intelligible in so far as 
 they are the natural products of many underlying uniformities 
 due to the social structure. Consider each atom of a tree bv 
 itself, as the plaything of an intricate chaos of forces of 
 bewildering perplexit)^, and you may wonder that the same 
 number of leaves is produced every year; but when you 
 remember that each atom is part of a structure bound to- 
 gether by a number of mutual relations, the wonder vanishes. 
 And it is equally true of that more obscure structure which 
 we call human society. 
 
 24. Thus our previsions as to human conduct are not simply 
 statements of an averacre result. If I know that there are 
 ninety-nine blanks in a lottery and one prize, I am certain that 
 only one of a hundred drawers will be successful, though quite 
 uncertain as to which it will be. I could know only if I knew 
 a number of details as to the distribution of tickets and the 
 movements of men's hands of which I know nothing. In 
 regard to manv social phenomena, I have the same kind of 
 knowledge. I know that there must be a fringe of destitute 
 people hanging upon the skirts of society, a small number 
 of prizes to be won by the most energetic or fortunate, and 
 a number of intermediate places into which the remainder 
 will be distributed by mutual pressure. There are so many 
 officers, privates, and hangers-on to an army, though I 
 carmot say which place will be occupied by a given man. The 
 members of a given society are forced to accommodate them- 
 selves to certain fixed conditions as much as the iron which 
 is poured into a given mould. There is a causal connection 
 underlying the apparently arbitrary movements of the indi- 
 vidual. The social struggle is thrusting down the weaker 
 into the position where want of food is most felt and stealing 
 most tempting; and character is being determined in count- 
 less indirect ways by the mutual pressure of the various 
 classes. Men are not only more or less alike, and so far approxi- 
 mate to a certain average, but they are also being constantly 
 educated in a thousand ways by the persistent conditions of
 
 THEORY OF SOCIAL EVOLUTION. 31 
 
 the social organism; and thus there are secondary or derivative 
 laws of conduct dependent upon these conditions and pro- 
 ducing uniformities not affected by the variation of indi- 
 vidual idiosyncrasy. 
 
 IV. Theory of Social Evolution. 
 
 25. A full realisation of this truth, which is of course 
 a very old truth in substance, a perception that society is 
 not a mere aggregate but an organic growth, that it forms 
 a whole, the laws of whose growth can be studied apart 
 from those of the individual atom, supplies the most 
 characteristic postulate of modern speculation, and we 
 may note its general bearing upon the problem before 
 us. The explicit and constant recognition (for there has 
 always been a kind of recognition) that society forms such 
 an organic growth, that its properties can be studied 
 separately, and cannot be inferred directly from the 
 characters of the component individuals, does not in- 
 deed relieve us from the difficulties already noted — for 
 by recognising that the laws are there we certainly do not 
 discover what the particular laws are — nor does it, in any 
 case, suggest the possibility of any a priori theory upon the 
 subject. We cannot say, previously to any study of the 
 oro-anism, that it is so, and that it could not have been 
 otherwise. On the contrary, we may suppose that many 
 other forms of society besides those with which we are 
 acquainted may be possible, and may be actually realised 
 under difierent conditions. If anybody should maintain that 
 in some other planet the propagation of the inhabitants may 
 be carried on by a totally different method, and that there 
 may be no distinction of the sexes, I do not see how it would 
 be possible to confute him, and under such circumstances the 
 whole social structure would be organised upon very different 
 principles. The utmost to which we can aspire is to show 
 how different parts of the structure mutually imply each 
 other; so that, given the whole, we may see that any particular
 
 32 THE SCIENCE OF ETHICS. 
 
 part could not be otherwise ; and this, though a less ambitious 
 conclusion, would be amply satisfactory for scientific purposes. 
 I know of no a priori proof, and for my part I cannot con- 
 ceive one, which should establish the necessity for a human 
 being consisting of just so many organs, of a stomach, lungs, 
 heart, legs, and so forth ; still I am quite able to show that 
 under existing circumstances each part of the framework is a 
 highly convenient part of the whole, and I can more or less 
 determine the conditions which — still upon that understand- 
 ing — best fit it for playing its part. Nor, again, has the 
 assumption which we are considering enabled us to acquire 
 that degree of knowledge which would enable us to predict 
 the future course of history, nor — if I am ri<rht in what I 
 have already said — ^made it even conceivable that such know- 
 ledge will ever be attainable. 
 
 26. What advantage, then, is gained by accepting this 
 theory? The first gain is the simple recognition that there 
 must be laws, and that there may be discoverable laws of 
 social growth which are essentially relevant to our inv^esti- 
 gation, but which previous methods of inquiry have tended 
 to ignore. So long as reasoning has been conducted upon the 
 tacit assumption that social phenomena can be satisfactorily 
 explained by studying their constituent atoms separately, 
 attention was diverted from some most important principles. 
 If we could have studied the body on the assumption that 
 each organ had an independent vitality which required no 
 reference to the other organs to make its laws of growth 
 intelligible, we should gain a good deal by simply recognising 
 the existence of the whole orc-anism. There are cases in 
 which we may study a number of units separately, and 
 thence infer the properties of a whole figured from such units. 
 There are other cases in which the properties of each part 
 are so dependent upon the whole that it is impossible to 
 understand them separately without reference to the proper- 
 ties of the whole. If the problems of human conduct really 
 fall under the second category, and if at the same time we 
 assume them to belong to the first, we shall manifestly
 
 THEORY OF SOCIAL EVOLUTION. 33 
 
 neglect some essential conditions. The symptom of the error 
 will be, that, as we have omitted a reference to certain 
 regulative principles, some of our data will appear to be 
 arbitrary, simply because we have not attended to the condi- 
 tions by which they are actually determined. My statement 
 must necessarily be vague at this stage of the argument, but 
 I shall endeavour to work it out hereafter. I may, however, 
 add already, that the assumption in question gives a new 
 sio-nificance to the facts before us. It shows how facts 
 which we w^ere previously content to leave unexplained, and 
 perhaps to set down as ultimate and inexplicable data of the 
 problem, may be important embodiments of the principles to 
 be discovered. Vast importance has been given to many 
 apparently trifling facts by the theory of evolution, as applied 
 to all the sciences which have to do with organised beings. 
 What was formerly set down as a freak of nature, or 
 dismissed from the sphere of the explicable by some verbal 
 reference to a special creation, turns out to be an important 
 link in a chain of evidence as to past conditions of organic 
 life. And so in sociological inquiries, we find that some 
 apparently trifling and arbitrary custom, wliich to certain 
 observers passed for a mere barren curiosity, gives a sudden 
 and effective illustration of remote social states. And what 
 is true of exceptional peculiarities is equally true of the 
 more permanent properties of social organisation. I am 
 now enabled to see that a statement which seemed only to 
 describe the averao-e mode of behaviour of independent beings 
 has really a vast significance when considered as describing 
 a quality of a persistent organism. For the theory of evolu- 
 tion brings out the fact that every organism, whether social 
 or individual, represents the product of an indefinite series of 
 adjustments between the organism and its environment. In 
 other words, that every being or collection of beings which 
 forms a race or a society is part of a larger system -, that it is a 
 product of the continuous play of a number of forces con- 
 stantly shifting and rearranging themselves in the effort to 
 maintain the general equilibrium, and that consequently every 
 
 C
 
 34 THE SCIENCE OF ETHICS. 
 
 permanent property represents, not an accidental similarity, 
 but a correspondence between the organism and some per- 
 manent conditions of life. 
 
 27. It may thus be said that the whole history of the 
 world and its inhabitants represents a problem of stupendous 
 magnitude. A being of vastly superior powers might con- 
 ceivably work out that problem, if he were acquainted with 
 the data at any particular moment, and capable of unravel- 
 ling all the inconceivably complicated forces which are con- 
 flicting and co-operating, and calculating their resultants. To 
 us, who are only infinitesimal parts of the system, the im- 
 possibility of any such attempt is almost ludicrously obvious. 
 We work out the problem bv living, or rather we work out 
 our own little bit of the problem. We are utterly incompe- 
 tent to grasp the whole, or to rise above it and say why such 
 and such data must have been given, and what will be the 
 further stages of the process. But when we once recognise 
 the fact that the problem is being worked out, we see also 
 that an answer is actually given in some degree by the very 
 facts before us. Our own lives are the answer. We can 
 thus obtain certain results, a posteriori, by recognising the 
 sense in which the evolution of history is really the solution of 
 the problem. If we can succeed in putting the question fairly, 
 we shall find the answer to it ready written upon the facts. 
 This is really the nature of the change in the point of view 
 implied in the acceptance of the evolution theory. It tells us 
 what is the proper mode of interrogating nature. When we 
 know the laws of gravitation we can prove that planets must 
 move in ellipses. If we knew nothing of the law of gravitation 
 or were unable to calculate its effects, but yet knew, in general 
 terms, that there was some such law, and that it was the sole 
 law in operation, we might be able to infer that elliptical move- 
 ment was a necessary consequence of its action. The planets, 
 it has been said^ are constantly engaged in working out complex 
 differential equations. They give us the answer to a set of 
 problems which no human brain may be sufficiently com- 
 prehensive to solve without such experience, lliis kind of
 
 ■STATE NOfiMAi SCHOOL. 
 
 THE ETHICAL PROBLEM, 'os Angeles, €3135 
 
 solution is equally given by the incomparably more complex 
 phenomena of human life ; and when we fully recognise the 
 fact that a problem is being solved, we have only to gain 
 some appreciation of its general nature and conditions in 
 order to obtain some important, though it may be very 
 limited, conclusions as to the meanino- of the answer which 
 may fairly be called scientific. They are not scientific in 
 the sense of giving us quantitative and precise formulae, but 
 they may be so far scientific as to be certain and verifiable. 
 
 V. The Ethical Problem. 
 
 28. We may apply this briefly to the special problem before 
 us. That problem is, in fact, to discover the scientific form 
 of moral itv, or, in other words, to discover what is the 
 general characteristic, so far as science can grasp it, of the 
 moral sentiments. The difficultv by which we were met 
 was the apparently arbitrary and fluctuating nature of all 
 human instincts. We could see our way to saying that, as a 
 general truth, and in the average, people did in fact have 
 such and such feelings. But the uncertainty of the data 
 seemed to paralyse any further inferences. We had no 
 apparent reason for saving why they might not have an 
 entirely different set of feelings. And, in one sense, the 
 difficulty is irreconcilable, or at least unanswerable from the 
 purely scientific point of view ; for I do not now ask how 
 far it mav be met by an assumption of transcendental or 
 metaphysical principles. But the conception which I have 
 just endeavoured to explain may show how, when we consider 
 human beings to be the product of a long series of processes 
 of adaptation or adjustment, acting either upon the individual 
 or the social organism, we may hope to discern that anv giv'en 
 set of instincts corresponds to certain permanent conditions, 
 and how one part of the organism implies another, or how, 
 the whole being given, the relation between its facts follows, 
 and thus how the general system hangs together. We can 
 attack the problem, what part do the moral instincts play in
 
 ^6 THE SCIENCE OF ETHICS. 
 
 the general system of human society, which is itself part of 
 the wider system of the world in which we live ? And this 
 brings us back to a proposition from which I started. For, 
 in fact, it is obvious that the variation in moral beliefs which 
 presented itself as a great difficulty must be a part of the 
 problem to be solved. We have not to consider a number 
 of different and irreconcilable opinions held by equally compe- 
 tent persons, and to strike a rough balance between them, and 
 arbitrarily exclude some from a hearing. We must ask, in fact, 
 what is the cause of these opinions, which, again, is some- 
 thing very different from the reason as perceived by those 
 who hold them. The evolution of opinion is part of the 
 whole evolution; and it mav appear upon further investiga- 
 tion that opinions which present themselves as radically 
 opposed, are, when properly considered, nothing more than 
 the partial views of the truth which have commended them- 
 selves to persons under different conditions. The true law 
 of belief will account for the erroneous as well as for the 
 accurate opinions. If the evolution of moral sentiment is 
 a work in which manv minds in many generations co-operate, 
 it is plain that the opinions held at any given period will 
 correspond in some way to the corresponding stage of evolu- 
 tion, and that the variation will appear not as arbitrarv, 
 but as confirmino- the o;eneral law. 
 
 29. It must be the work of the following chapters to give 
 more precision to these general remarks. One observation, 
 however, must be premised, of which I have had to feel the 
 importance. Starting, as I profess to do, from the scien- 
 tific point of view, it follows that I have to deal, in the first 
 place, with facts of observation. I have, that is, to con- 
 sider the moral sentiments which have, as a historical fact, 
 exercised an influence in the world, and to ask what part 
 they play in the general process of evolution. I hav^e, there- 
 fore, nothing to do, in the first instance, with those moral 
 principles which are or profess to be deduced from transcen- 
 dental considerations or from pure logic independent of 
 any particular fact. The distinction, which cannot be fully
 
 THE ETHICAL PROBLEM. 37 
 
 considered at this stage, is one which must yet be noticed, 
 because I am convinced that a neglect of it leads to a vast 
 amount of ambiguity and misunderstanding. That there 
 is, upon any theory, a great difference between actual and 
 ideal moralitv, I take to be an admitted fact. In anv given 
 society there are, as a rule, several moral standards : there is 
 that which is tauirht in churches ; that — not alwavs identical — 
 which actually determines our approval or disapproval ; that 
 which is current in the most cultivated, and that which is 
 held by the most barbarous classes; that which is approved 
 by the advanced thinkers, and that which commends itself to 
 the thorouo-ho-oino- conservatives, and so forth. The difference 
 
 COO '' 
 
 is universally allowed to exist. The utilitarian moralist 
 
 considers that to be moral which makes for happiness, but 
 
 he admits that the average calculation of happiness is often 
 
 very wrong. His opponent holds that moral principles are 
 
 deducible from pure reason, but he admits that most men 
 
 are very poor hands at pure reasoning. Perhaps he appeals 
 
 to the voice of conscience, but he does not assert that the 
 
 voice is not capable of misinterpretation. The distinction is 
 
 recognised on all hands, but the ordinary mode of recognition 
 
 leads to much confusion. Each man thinks that his own 
 
 moralitv is the rio-ht moralitv, and that the ordinary standard 
 
 is mistaken and immoral so far as it deflects from it. He 
 
 does not say that your moralitv is erroneous, but denies it 
 
 to be morality at all. I do not object at present to this 
 
 mode of speech, but it may lead to misunderstanding. Thus, 
 
 for example, one moralist asserts that the moral code varies, 
 
 whilst another says that it is fixed. Yet they mav, and 
 
 sometimes do, mean the very same thing; for both may 
 
 allow that the actual code varies, and both may agree that, if 
 
 men were better reasoners or better calculators of happiness, 
 
 the code would be fixed : the variability is predicated of the 
 
 actual, the fixity of the ideal code. Thus the question of 
 
 what ought to be moral (if I may be allowed the phrase) is 
 
 often confused with the question of what actually is moral. 
 
 The various controversies as to international law illustrate
 
 38 THE SCIENCE OF ETHICS. 
 
 the confusion. When a writer holds that a certain custom 
 ought to be observed by different nations, he at once declares 
 that it is the law. He attributes to it a kind of potential 
 existence, though he allows that it is not actually operative. 
 His opponent denies it to be the law, because he gives that name 
 only to laws which are actually observed, though he may be 
 willing to admit that it is desirable that it should be the law. 
 A similar misunderstanding perplexes many problems in 
 casuistry. So, for example, we ask what judgment we are to 
 form of conduct which is "materially" wrong but "formally" 
 right; or, as the words are sometimes interpreted, of conduct 
 which is in conformity with the moral law as accepted by the 
 agent, but not in conformitv with right moralitv. A man 
 deserts his wife, sincerely believing that marriage should not 
 be permanently binding. Such cases are often perplexing; 
 and at present I have only to remark that the perplexity is 
 increased by the ambiguity in question. If we understand 
 that by 7'ight we mean right according to the ideal code, and if 
 the ideal code — however constructed — implies indissolubility 
 of marriage, then the man clearly acts wrongly. We may 
 still ask whether his non-recognition of the true principle 
 makes his conduct better or worse; whether a man who 
 knowingly breaks the law from conviction is better or worse 
 than a man who breaks it unknowingly? This, in my 
 opinion, amounts to asking which man would, on the whole, 
 observe the genuine moral code with fewest exceptions. 
 And this, again, is a question of fact to be settled by psycho- 
 logists and direct observation. If, on the other hand, by 
 right we mean simply conformity to the moral code which is 
 actually operative at the given time, it is simply a question 
 of fact whether that code does or does not prescribe indis- 
 solubility of marriage. The further question then remains, 
 whether the new rule upon which the man acts is a better 
 or a worse rule than the accepted rule ? And this is to ask 
 whether it would or would not form a part of the ideal code, 
 however that is to be constrvicted ? The problem is then of 
 necessity complicated, but it becomes hopelessly perplexed
 
 THE ETHICAL PROBLEM. 39 
 
 ^vhen we have not settled as to the sense in which we use tlic 
 word moral, for the man may be right in one of the senses 
 and wrono; in the other. 
 
 30. It is desirable to avoid this ambiguity as much as pos- 
 sible ; and therefore I give notice of the fact that, until I 
 state the contrary, I wish to be understood as referring in all 
 cases to the actual law. I mean by the moral code that set of 
 rules which, as a matter of fact, is respected in a given society, 
 and so far determines the ordinary approvals and disapprovals 
 as to be an effective force in governing conduct. It will be 
 an important inquiry at a future period, what is the relation 
 of this to the ideal code, and what it precisely is that I 
 understand by the ideal code. I mention the fact explicitly, 
 because I think that few ambiguities in the whole inquiry 
 have been more fruitful of misunderstanding.
 
 ( 40 ) 
 
 CHAPTER II. 
 
 THEORY OF MOTIVES. 
 
 I. The Problem. 
 
 1. Ethical speculation must, as thus understood, be Impli- 
 cated in psychological and sociological inquiries; that is to 
 say, its foundations must be laid In the treacherous region 
 where the vague doctrines of common sense have not yet 
 crystallised into scientific coherence. Perplexing ambiguities 
 beset the simplest primary propositions. We accept from 
 the language of ordinary life some statement which serves 
 well enough for a working maxim. As soon as we apply It 
 to purposes of accurate inquiry, we find that it savs too mvich 
 or too little ; that it causes a hopeless ambiguity or pushes an 
 undeniable truth into obvious error. We are trying to split 
 hairs with a carving-knife. We must, therefore, look closely 
 to our initial steps, and scrutinise some apparently obvious 
 and harmless propositions. 
 
 2. We have to deal with human conduct. So far as a 
 man is a material thing, his conduct. If It can properly be 
 called conduct, is determined by purely mechanical conditions; 
 so far as he Is a sentient being, his feelings, and so far as he 
 is a rational being, his reasoning powers, must be taken into 
 account. We are already in presence of a difficulty. How 
 are we to conceive of the relation between reason and feel- 
 ing, or of the relations between the vital and the mechanical 
 forces? As we understand these words, in one sense or 
 another, we find ourselves impelled towards one of the two 
 opposite poles to which all ethical speculations converge. 
 According to the principles already stated, wc must distinguish
 
 THE PROBLEM. 41 
 
 between the scientific and the underlying metaphysical pro- 
 blems. The metaphysical are, for my purpose, irrelevant. I 
 do not, for example, enter into the discussion between the 
 materialist and the idealist. The materialist regards man as 
 an automaton worked by mechanical forces, which, accord- 
 ino" to him, are the sole realities, and considers consciousness 
 to be in some sort a superfluous spectator. Without dis- 
 cussing the tenability of such a theory, I take it for granted 
 that pains and pleasures have an influence upon human con- 
 duct; that men eat because they are hungry, strike because 
 they are angrv, and act upon reasoned plans because they 
 have certain convictions as to the nature of the world and the 
 consequences of their actions. The materialist may show, if 
 he can, that all the processes thus described have a mechanical 
 aspect; that hunger corresponds to certain organic states 
 which involve certain cerebral changes, transforming impulses 
 received from the nervous system into messages determining 
 mechanical changes in the various organs of the body. In 
 any case, however, we are in complete darkness as to the 
 nature of these mechanical changes, and are long likely so to 
 remain ; and, moreover, in any case — according to my con- 
 tention — a statement of them will not supersede the other 
 mode of statement. The proposition, that is, that hunger 
 makes men eat, will express a truth whatever material impli- 
 cations may be involved in that simple statement. The 
 ultimate reality may be something different, but at least the 
 formula gives, and, so far as I can see, is always likely to give, 
 an intelligible and simple account of the facts. And the 
 same will be true if we adopt the opposite theory of idealism, 
 and in some sense deny reality to the mechanical entities 
 called brain and stomach and food. They may be simply 
 clusters of sensations, but the formula will still express a 
 relation which will hold good as to their reciprocal action. 
 We have to inquire what is the most general formula of 
 this kind discoverable.
 
 42 THE SCIENCE OF ETHICS. 
 
 II. The Emotions as Determining Conduct. 
 
 3. After emerging from the purely metaphysical regions, we 
 have still to look carefully to our footing. Conduct, I have 
 said, is determined by feeling 5 we fly from pain; we seek 
 pleasure; life is a continuous struggle to minimise suffering 
 and to lay a firm grasp upon happiness ; " good " means every- 
 thing which favours happiness, and " bad " everything that 
 is conducive to misery; nor can any other intelligible mean- 
 ing be assigned to the words. To some minds these pro- 
 positions appear to be self-evident; they cannot be denied 
 without self-contradiction; the difficulty of proving them is 
 the difficulty of proving any of those primary doctrines for 
 which we must appeal to what is called the direct testimony 
 of consciousness. Yet, in laying them down, I have already 
 made assertions which seem to beg the question, and the 
 appeal to consciousness may be rejected as virtually an appeal 
 to common sense, or, in other words, to unreasoning prejudice. 
 I shall, however, venture to assume that the assertions are 
 in some sense valid ; and I make that assumption upon the 
 ground already taken — the ground, namely, that no meta- 
 physician would really deny that they express truths, though 
 he may deny that they express ultimate truths. The only 
 real question is in what sense they are valid. Nobody who 
 has ever had a toothache — nobody, one might rather say, 
 who has ever had a sensation — will deny that he avoids pain 
 as such and seeks pleasure as such. And, in fact, if we 
 examine the ordinary criticisms of these simple propositions, 
 we shall find that the critics deny not so much the propositions 
 themselves as certain interpretations which tacitly introduce 
 some further and more questionable assertion. 
 
 4. Thus, for example, it is frequently admitted that pain 
 deters and pleasure attracts, but it is at the same time denied 
 that pain and pleasure are the sole deterrent or attractive 
 qualities. We shrink from fire or the knife, but some other 
 motive may overpower our spontaneous dread. What, then.
 
 THE EMOTIONS AS DETERMINING CONDUCT. 43 
 
 is the nature of this different motive? It must, by the 
 argument, be something which admits of a comparison with 
 pain and pleasure, which has its equivalent in terms of 
 feeling, and which represents, therefore, a force of the same 
 order. Is this not virtually to admit that it is still a pleasure 
 or a pain ? The dread of shame or remorse overcomes the 
 martyr's dread of the fire. Is not that because shame and 
 remorse are themselves painful, and in some men more 
 exquisitely painful than physical torment? The pain and 
 pleasure may be higher in kind, but it is still a pleasure or a 
 pain. The true statement is that one emotion may be over- 
 come, not by a something which is altogether disparate from 
 emotion, but by an emotion of a different kind ; and this is 
 of course indisputable. It does not traverse the proposition 
 that emotion can be limited by nothing but emotion. Or, 
 again, the conclusion is sometimes evaded by showing, not 
 that pain or pleasure do not determine conduct, but that 
 conduct is not determined by certain modes of estimating 
 pain and ])leasure. A man, it is often said, may deliberately 
 prefer a life of pain to annihilation, and in so doing he 
 would choose a clear balance of painful sensation. But if we 
 look closer, this is simply to say that the prospect of annihila- 
 tion is more painful at the moment than the prospect of 
 future misery. This, indeed, is highly probable. The 
 instinct which revolts ao;ainst the thought of annihilation is 
 so powerful that the imagination of future evils is unable to 
 overcome it. Perhaps this proves that the instinct is a 
 foolish one; but who ever denied that we had foolish instincts — 
 instincts, that is, which lead us to sacrifice a great future to 
 a small present pleasure, or to choose an evil rather than 
 make the effort to wield the bare bodkin ? All that is 
 proved is that we are not always determined by a calculation 
 of pleasure to come ; or, again, that the estimate of future 
 pleasure does not always produce a corresponding pleasure; 
 and this will be admitted on all hands. 
 
 5. We have to meet these objections by the sacrifice of 
 much of the apparent import of our statement. We are com-
 
 44 THE SCIENCE OF ETHICS. 
 
 pelled to widen the significance of our words until we admit 
 every conceivable form of agreeable or disagreeable feeling ; 
 and we are forced to allow that pain and pleasure may deter- 
 mine us to act so as in the long-run to sacrifice happiness 
 and court misery. And hence arises another, and in some 
 ways a more formidable, criticism. Granting, it is said, 
 that happiness is the sole aim of all human conduct, what 
 are we the wiser ? The proposition is valueless unless 
 universally true; and if universally true, it is nugatory. 
 Happiness is known to us solely as that which men desire : 
 to say, then, that they desire happiness is to say that they 
 desire what they desire. The force of this familiar argu- 
 ment is undeniable. The proposition, I agree, is nothing if 
 not universal ; it must cover all the actions of all human 
 beings, at every moment of their lives and throughout their 
 whole range of conscious motive; it must be equally true of 
 our sensual appetites, our purest emotions, and our intellec- 
 tual activities. Happiness guides us when we are eating 
 our dinners, or studying metaphysics, or feeding the hungry; 
 when we sacrifice all prospects of future happiness to the 
 loftiest or to the most grovelling motives; when we destroy 
 our health and ruin our families for a glass of gin, or walk 
 up to a battery to buy one more chance of victory for a good 
 cause. The love of happiness must express the sole possible 
 motive of Judas Iscariot and his Master ; it must explain the 
 conduct of Stylites on his column, of Tiberius at Capreae, 
 of A Kempis in his cell, and of Nelson in the cockpit of 
 the Victory. It must be equally good for saints^ martyrs, 
 heroes, cowards, debauchees, ascetics, mystics, cynics, misers, 
 prodigals, men, women, and babes in arms. Truly it must 
 be an elastic principle. Can it be more than this that men, 
 so far as they act from motives, have some motives for their 
 action ? 
 
 6. I admit that the proposition is one which would have 
 been scarcely worth an explicit assertion had it not been 
 explicitly denied. But it is not an empty proposition. It 
 asserts, in the first place, the general proposition already laid
 
 THE EMOTIONS AS DETERMINING CONDUCT. 45 
 
 down, that human actions are no more the product of 
 chance than any other observable phenomena. But beyond 
 this, it makes another assertion which it is essential to bear 
 in mind. To assert, in fact, that the quality of any conscious 
 state in respect of its pain or pleasure is the essential con- 
 dition of its being desired or dreaded, is to deny that the con- 
 dition of conduct can be found in anything; which has no 
 relation to pain or pleasure. And as attempts have been 
 made to find the condition elsewhere, it becomes necessary 
 to deny the possibility of their success. To say that action 
 is determined by feeling is not to enable us to determine 
 which particular feeling will be most efficient. As the state- 
 ment is true of all motives, it throws no light upon the 
 relation between different motives. Nor, again, does it lead 
 us to any fertile proposition as to the nature of conduct in 
 general. Pain and pleasure are words which it is impossible 
 to define ; as they are names of the highest class and not a 
 species of some more general class, we can no more define 
 them by genus and difference than we can define light and 
 darkness, or subject and object, or past and future : but we 
 can still know what they mean ; and it is conceivable that 
 we might be able to assign the physiological conditions of 
 their existence, in which case we might be able to advance to 
 some fertile conclusions from a statement which is at present 
 nugatory by its excessive generality. At present, we can 
 only say that psychologists distinguish — and their distinction 
 has at least a scientific if not a metaphysical validity — between 
 the emotions and the intellect, or between feeling and 
 reasoning. We assert that conduct is determined by the 
 feelincrs, not as denyino; that it is also in some sense deter- 
 mined by the reason, but as maintaining that a state of con- 
 sciousness which is neither painful nor pleasurable cannot be 
 an object either of desire or aversion, and that, so far as it is 
 painful or pleasurable, it will produce one or the other. We 
 shall, I think, see that this doctrine, however vague, is by no 
 means without an important bearing upon ethical specula- 
 tion 5 but it also suggests the necessity of entering more
 
 46 THE SCIENCE OF ETHICS. 
 
 fully into the meaning of the distinction in question. Since, 
 as I have said, man is both a sentient and a reasoning being, 
 we have to ask how we are to conceive of the mode in which 
 these separable, or at least verbally distinguishable, faculties 
 co-operate in determining conduct. The inquiry will lead to 
 a fuller answer to the difficulties upon which I have briefly 
 touched in the last paragraph. 
 
 7. Moralists who regard happiness as the sole aim of all 
 conceivable action have accepted a simple analysis of 
 reasoned conduct. Conduct, as the old reasoners said, fol- 
 lowed the "last act of the judgment." Though men fre- 
 quently, perhaps generally, act without much conscious choice, 
 the typical and fully developed act of decision consisted in a 
 judgment that one action would produce more happiness 
 than another. The intellect, to use the ordinary metaphor, 
 weighed different "lots" of happiness in its balance; the 
 will inevitably inclined to that conduct which weighed the 
 heaviest in these scales. Reasoned conduct differed from 
 merely instinctive conduct in that it implied an adaptation 
 of means to ends, and therefore a possibility of following 
 courses of conduct not agreeable in themselves, but promising 
 a greater total of happiness. The characteristic of the reason- 
 ing being was the power of acting with a view to distant ends, 
 instead of being the slave of immediate impulse. Now I do 
 not deny that it may be possible to use this phraseology so as 
 to express the facts with tolerable accuracy. But I think 
 that it is apt to sanction certain erroneous hypotheses which 
 have done much to perplex the whole subject. 
 
 8. We may have to consider some of these hereafter. For 
 the present I may observe that at the very outset there is 
 apparently one defect in the statement. It seems to assume 
 that a deferred pleasure is as potent as an immediate pleasure. 
 So long as "I "am to have the pleasure, it matters not, so 
 far we have gone, whether it be the " I " of to-day or the " I" 
 of next year. The influence of mental perspective appears 
 therefore to be entirely ignored ; and yet there can be no 
 doubt, so long as we are dealing with facts, that this influence
 
 THE EMOTIONS AS DETERMINING CONDUCT. 47 
 
 is of the highest importance, and that every removal of a 
 pleasure in time and place has an immense effect upon its 
 power of determining conduct. But if we try to rectify our 
 formula so as to allow for this difficulty, we become aware 
 that it is of the very essence of the statement. The 
 happiness which determines the will is always regarded as 
 future, though it may be in the immediate future. The 
 normal state of man is that of the proverbial ass, who con- 
 stantly pursues and never obtains the wisp of hay fixed at 
 the end of the shaft. The popular maxim, " Man never is, 
 but always to be blest," is converted into a scientific formula. 
 Conduct, you say, is always determined by feeling; but the 
 doctrine consistently applied is fatal to the analysis assumed. 
 For on this analysis, reason, regarded as a faculty altogether 
 distinct from the emotions, determines conduct by reference 
 to a feeling which does not yet exist, or, in other words, does 
 not exist at all. The actual feeling is taken to be a mere 
 consequent which does not affect conduct. The feeling which 
 is hereafter to be is the sole determining influence. 
 
 9. To express ourselves accurately, and to carry out con- 
 sistently our initial assumption, we must modify the state- 
 ment. No feeling can affect us except so far as it is felt. 
 An unfelt feeling is a nonentity. Many of our feelings, it 
 is true, are regarded by us as prophetic foretastes of a feeling 
 to come. We shudder at the edge of a cliff because we fore- 
 see the consequences of a fall, and thus the shudder implies 
 the belief in those consequences. But though this intellectual 
 perception is an essential part of the process, it would not 
 affect our conduct if it were a perception entirely divorced 
 from feeling. It affects us because the perception is itself 
 painful, because it involves an anticipatory realisation of an 
 approaching pain. It is more accurate to say that my con- 
 duct is determined by the pleasantest judgment than to say 
 that it is determined by my judgment of what is pleasantest. 
 And though in such cases the remark may seem to be 
 hypercritical, and one formula to be equivalent to the other, 
 there are many cases of the highest importance in which it
 
 48 THE SCIENCE OF ETHICS. 
 
 is essential to any intelligible account of the process. A 
 man, for example, acts in a given way under the influence 
 of vanity. Although he knows that he would be happier 
 in the long-run if he abandoned some burdensome honour, 
 his self-esteem nails him to his post. He is governed by 
 the instantaneous feeling, by the immediate sense that a 
 resignation would expose him to ridicule or contempt. And 
 though this generally includes a foretaste of future suffering, 
 he is determined by the actual painfulness of the choice, 
 not by the belief that he is choosing something which will 
 produce pain. This is equally true of all the feelings which 
 we may roughly call aesthetic feelino-s — feelings which contain 
 no special reference to the future, but which are prompted 
 by immediate perception. The same, again, may be said 
 of a large class of actions which seem puzzling to many 
 observers, and which is conveniently illustrated by the case 
 of games. When we speak of conduct as determined by the 
 pursuit of ends, it seems to be contradictory that a man 
 should take a keen pleasure in the pursuit of an end admit- 
 tedly trifling, the slaughter, for example, of a fox, or the 
 solution of a conundrum. The true explanation, however, 
 seems to be different. The pleasure of a game, speaking 
 generally, is the pleasure of a discharge of accumulated 
 energy. We are "bored," that is, we suffer a painful sensa- 
 tion, if we do nothing when we are in vigorous health, and 
 therefore we receive pleasure in doing anything, even if that 
 anything brings us no additional pleasure in itself. The 
 so-called end is pleasant simply because the pursuit affords 
 a convenient pretext for a regulated discharge of energy. 
 The life and death of a fox is absolutely indifferent; the 
 exercise of the powers necessary for catching and killing him 
 may be momentarily delio-htful. The stimulus comes from 
 within, not from without; from a change m our state, not 
 from a change in our circumstances. We might try to 
 describe the difference by distino-uishiniT between " sub- 
 jective" and "objective" ends. In some cases, we should 
 then say, we wish to relieve our feelings; in others, to bring
 
 THE EMOTIONS AS DETERMINING CONDUCT. 49 
 
 about some state of things which does not at present affect 
 our feelings. The distinction would be erroneous, because 
 every end must be at once objective and subjective. It must 
 be objective, because in all cases our conduct has certain 
 objective consequences, that is, consequences other than a 
 state of our own consciousness; and it must be subjective, 
 because it cannot be an end unless we desire it. But the 
 distinction corresponds to a real difference. In every case of 
 rational action we foresee a certain set of consequences as 
 dependent upon our conduct. The "intention" of the agent is 
 defined by these foreseen consequences. His "end" is defined 
 by that part of the foreseen consequences which he actually 
 desires, and the end defines the "motive," that is, the feeling, 
 which actually determines conduct. Now it is true that in 
 some cases the immediate feeling is all that is consciously opera- 
 tive, and we care comparatively little for the consequences 
 in any other relation, though we may distinctly foresee them, 
 and even call them our end. In other cases, on the contrary, 
 the conduct is only adopted in view of certain pleasures to 
 be obtained hereafter. But even in that case we adopt it 
 because the foretaste of the pleasure is itself pleasurable. 
 We like the labour which is to bear fruit hereafter. Hence 
 we easily speak of ends in the latter case, because we over- 
 look the existing pleasure which exists only as a foretaste of 
 the future pleasure, whilst in the former case the reference 
 to an end seems to involve a desire for an end not in itself 
 desirable. In fact, however, though one element or the 
 other may be more or less prominent, though the motive 
 may be more easily definable in objective or subjective 
 languao-e, and the pleasure be dependent or independent 
 upon the representation of future states of feeling, the two 
 elements must always be present, and the motive be an 
 actually existing feeling. 
 
 10. I think, then, that the metaphor of the balance is 
 misleading, and that the conception to which it corresponds 
 leads to many fallacies. It is misleading because it implies 
 an erroneous analysis, and separates different aspects of a
 
 so THE SCIENCE OF ETHICS. 
 
 single process, as though they were different processes succeed- 
 ino; each other in time instead of beino^ mutually involved. 
 The intellect and the emotion are in reality related as form and 
 substance, and cannot be really divided. To judge of plea- 
 sures is to feel the pleasures themselves, or to feel represen- 
 tative pleasures. The process is at once feeling and think- 
 ing, and may be regarded from either point of view. The 
 emotion is something determinate, and therefore its logical 
 formula may be given, but the logic without the feeling 
 would be a mere blank nonentity, a form without substance. 
 This is equally overlooked when we regard the pleasure as a 
 kind of independent thing which the intellect weighs and 
 measures, or whether we try to resolve it into a perception in- 
 stead of regarding it as bound up in the perception. It seems 
 to me, for example, inaccurate to say that vanity is the belief 
 that we are superior to our neighbours. It is equally inaccu- 
 rate to say that it is a pleasure which results from that belief 
 as a consequent in time following an antecedent. It is the 
 emotional side of a process which, upon its intellectual side, 
 is a belief; it is a feeling evolved by bringing together in 
 certain ways a perception of our own qualities and those of 
 our neighbours, and implies equally a conviction and an 
 emotion. 
 
 II. Falling back, then, upon the original principle, I repeat 
 that pain and pleasure are, according to me, the determining 
 causes of action. It may even be said that they are the sole 
 and the ultimate causes. They are the sole causes in this 
 sense, that where two courses of conduct are otherwise pos- 
 sible, and the choice of one depends upon the agent's own 
 decision, his will is always determined by the actual painful- 
 ness or pleasantness of the choice at the moment of choos- 
 ing, and that there is no different kind of motive. They are 
 ultimate in this sense, that it is impossible to analyse pain and 
 pleasure into any simpler elements. We might conceivablv 
 find the physiological causes of pain and pleasure; we might 
 show, that is, that each state corresponded to certain organic 
 conditions; but such a discovery, however important, would
 
 THE EMOTIONS AS DETERMINING CONDUCT. 51 
 
 not alter the fact that the pain and pleasure are the imme- 
 diate causes of conduct, and that they admit of no further 
 analysis from the subjective point of view, whatever may be 
 the physical correlates. We mav, again, investigate the his- 
 torical causes of the pain and pleasure, that is, we may trace 
 the previous stao-es throu2;h which the sentient bsino- has 
 come to be what it is. But this, thoucrh a^rain a most inter- 
 esting inquiry, would not alter the fact that at the given 
 moment the feelino-s, beino- such as thev are, determine the 
 conduct. If we could analyse, in short, the forces which 
 maintain the existing order, the feelings considered as painful 
 or pleasurable would represent the ultimate elements in the 
 analysis. Any attempt to analyse further only lands us in 
 contradiction and confusion. The fact that any state of con- 
 sciousness is painful, is so far a sufficient reason for avoiding 
 it — a reason which may be overpowered by others, and which 
 may in its turn admit of explanation; but so far as it goes, 
 it represents a true cause, and one which must always operate. 
 I admit that so far as the act of choice implies a weighing of 
 pains and pleasures, the pains and pleasures are alwavs repel- 
 lent and attractive, and the sole measure of the repulsion and 
 attraction. I only deny that this gives an adequate account 
 of the intimate nature of the process. 
 
 13. How, then, must we conceive of the mode of action 
 of these forces? The most obvious fact seems to me to be 
 that in all cases pain as pain represents tension, a state of 
 feeling, that is, from which there is a tendency to chano-e; 
 pleasure represents so far equilibrium, or a state in which 
 there is a tendency to persist. Some such statement repre- 
 sents most nearly the characteristic of all emotional states, 
 from the lowest upwards. The worm writhes on the hook, 
 and we inevitably interpret the writhing as indicative of ao-onv. 
 The mind (if I may say so) writhes under a painful emotion. 
 It makes an effort to writhe into some more tolerable posture. 
 Fear implies an effort to get away from the painful state of 
 anticipated pain, and as it becomes intense, the spasmodic 
 struggle becomes so desperate as to render any definite action
 
 52 THE SCIENCE OF ETHICS. 
 
 impossible, even though we know that our only hope lies In 
 cool and regulated effort. The pain of continuous illness 
 produces a restlessness, a vague unguided effort to secure 
 some change, even though no specific change presents itself 
 as desirable. Similar remarks miirht of course be made about 
 
 O 
 
 the inverse case of content and satisfaction. When I choose 
 I go through a process of a complex but still similar kind. 
 Two modes of feeling are more or less distinctly present to 
 my mind. Each is recognised perhaps as the foretaste of 
 a feeling about to become actual. So far as my power of 
 volition goes, I am decided by the relative painfulness or 
 pleasurableness. I try each mode ideally, and settle down 
 into that which is on the whole the easiest. What in the 
 actual feeling is the "writhing" and spasmodic struggle is, 
 in the representative feeling, a difficulty in admitting the 
 suggested course of conduct. When I am satisfied, my mind 
 acquiesces and slides into that mode of action which causes 
 the least disturbance. 
 
 13. The analogy which naturally offers itself and seems to 
 give the best account of the facts is the mechanical principle of 
 least resistance. A body under the influence of several forces 
 moves in that direction in which the resultant of the opposing 
 forces is a minimum. Similarly the various desires operate 
 in such a way that the volition discharges itself along that 
 line in which the balance of pleasure over pain is a maximum. 
 The pleasurable state is one of stable, the painful one of un- 
 stable equilibrium. Alter the conditions slightly in the latter 
 case, and the agent spontaneously adopts a new attitude, 
 like a spring relieved from pressure. Alter them in the first 
 case, and there is a tendency to revert to the previous state. 
 In one case we have a body balanced on a ridge by external 
 forces, and ready to fall at once if the supports be removed; 
 in the other we have a body so balanced that after a slight 
 change it will always tend to fall back to the original position. 
 It is, perhaps, not superfluous to remark, in order to avoid 
 possible misconstruction, that the volition may exercise a very 
 small influence, even when the limiting conditions are in
 
 THE EMOTIONS AS DETERMINING CONDUCT. 53 
 
 great part ideal. The more painful is not necessarily the 
 less permanent condition. It is one in which there is an 
 additional chance against permanence. A painful thought 
 may fascinate the attention. Terror sets up, as I have said, 
 so disturbed a condition that the mind cannot settle into any 
 definite course : the hand shakes when it would aim the 
 necessary blow. To determine even the subjective conditions 
 of conduct we have to consider the fascination as well as the 
 mere painfulness and pleasurableness of the emotions. We 
 can no more alter arbitrarily the circumstances of our micro- 
 cosm than of the external world. The painful thought cannot 
 be included simplv as painful if it excites an intense emotion 
 and is associated by many links with all our habitual objects 
 of thought, even though we may know that the pain can- 
 not suggest a remedy. It is as difficult to avoid brooding 
 over vain regret as to evade a physical constraint. The will 
 becomes paralvsed, and the brave man feels that the best 
 course is to face the pain till his sensibility becomes deadened 
 by persistent familiarity. 
 
 14. To explain the theory fully it would thus be necessary 
 to consider the sphere of volition, or to analyse what, accord- 
 ing to one theory, is called an act of Free Will. I have 
 to make some simple choice, say, between a glass of wine 
 and a glass of water. The process, according to me, consists 
 essentiallv in rehearsing the two modes of conduct which pre- 
 sent themselves to my mind. I am moved by the foretaste 
 of the pleasure of drinking, the difficulty of reaching either 
 crlass, the dislike to expense, the moral and medical scruples 
 in which T indulge, and so forth ; and the decision takes 
 place according to the principle of least resistance. To 
 analyse all the operative motives may be impossible even 
 in so simple a case; to weigh and compare their importance 
 in general terms may be impossible ; but in any case, I 
 decide by the simple process of feeling one course to be the 
 easiest. It often happens that each course will be the easiest 
 if it happens to be the one most directly contemplated ; and 
 it will in that case be a matter of chance — that is, it will
 
 54 THE SCIENCE OF ETHICS. 
 
 depend upon other circumstances than my will — whether I 
 take the wine or the water. If now I am desh'ous to know 
 what will be the result of some process entirely independent 
 of my volition — sa}'', the choice of another person in the same 
 case — I shall equally rehearse all approaching signs of events. 
 But in the latter case my mind works automatically upon 
 the data provided. It takes certain facts, arranges them 
 according to logical rules, and brings out a conclusion which 
 does not follow with more or less certainty because the result 
 is painful or pleasurable. My volition determines, of course, 
 whether I shall or shall not reason. If I choose to reason, I 
 apply a fixed rule which (if rightly applied) brings out a fixed 
 conclusion. One and only one result can follow from the 
 one set of data. In the other case, my intellect shows two 
 courses to be equally possible. The sequence of events which 
 terminates in wine is as consistent with all the facts as the 
 sequence which terminates in water. One of the series then 
 follows by introducing the additional fact that it is the 
 pleasantest. I may say that it follows because I know or 
 perceive it to be the pleasantest. But this, as I have explained, 
 seems to me to be a roundabout way of saying that I feel 
 it to be pleasantest, or, more simply, that it is the pleasantest 
 to me ; for the knowledge is here the very same thing as the 
 feeling in a reflective mind ; or, if knowledge be used in a 
 wider sense to include the represented feeling, the feelino- is 
 the fundamental fact, and the knowled<re the reflected feelinjr. 
 We may then say that in this case the prediction fulfils itself. 
 In the other case, I foresee a fact because it is about to 
 happen. In this case, it is about to happen because I foresee 
 it. My anticipation is the essential condition of its happen- 
 ing; and in this case the foresight becomes a volition. 
 
 15. The believer in free will (according to me) misinterprets 
 these facts. The foresio-ht of events is of the same nature 
 in the two cases up to a certain point, and at that point a 
 further condition is required in one case which does not apply 
 in the other. As this condition depends upon the person 
 who foresees, it appears to be arbitrary, although in fact it is
 
 THE EMOTIONS AS DETERMINING CONDUCT. 55 
 
 simply an expression of his constitution. When we see another 
 man about to act, we assume his taste to be one of the data, 
 and calculate his action as we calculate any other future 
 event. When it is our own, we fancy in some obscure sense 
 that it is in our power that it should exist or not exist, and 
 therefore regard the whole decision as arbitrary. I do not 
 wish, however, to dwell upon this argument, but to point 
 out another fact which is sometimes neglected — the difficulty 
 we have in drawing; the distinction between the two kinds of 
 foresight accordingly. Slow experience teaches us what is 
 the true sphere of volition. Our prevision of our own conduct 
 is exceedinfrlv fallible. Nothing:; is more common than to 
 mistake wishes for anticipations and dislikes for disbelief. A 
 perfectly logical mind would draw conclusions unbiassed by 
 pain and pleasure. The hatred of error would overbalance 
 the painfulness of anticipation. Its emotions would decide 
 it to add up its accounts; but they would be unable to 
 persuade it that two and two make five. But nothing is 
 easier than to find a mind which never permits its anticipations 
 to intrude beyond their proper sphere. The logical mill once 
 set going must grind out results irrespective of their pain or 
 pleasure; but we dread to set it going or tamper with its 
 action. We are apt, in vulgar phrase, to "cook" our accounts. 
 The reluctance of the mind to gaze upon painful facts prevents 
 us from setting the sum precisely; and as we tamper with the 
 materials at every stage of the process, we end with that mass 
 of contradictions and baseless prejudices which we know as 
 human beliefs. It is only by long experience, in short, that 
 we learn what are the predictions which can fulfil themselves, 
 and those which have no effect upon the future. If it is still 
 hard to resist the illusion that a thing will happen because 
 wc desire it, it is intellio-ible how all the reli(2;ions which are 
 rooted in early stages of mental development sanction the 
 propensity to hold that fate can be conquered by will, and 
 that prayers — the embodiment of desires — can govern the 
 stars in their courses. Anticipation and volition spring from 
 the same root, and it is by a very gradual and difficult process
 
 56 THE SCIENCE OF ETHICS. 
 
 that we learn to assign to each its proper sphere in our inental 
 operations. 
 
 1(5. The difficulty of knowing what we desire is as great as 
 the difficulty of estimating the efficacy of our desires. The 
 play of motive is beyond measure complicated. We may 
 meet even with apparently contradictory cases, such as the 
 apparent delight of many people in contemplating the darkest 
 probabilities. The full explanation must be left to psycho- 
 logists. It is partly perhaps that there is an actual pain in 
 trying to reconcile the belief in a particular form of good 
 fortune with a general reputation of evil. It requires a 
 painful effort to shake off doubts and fears which have become 
 familiar. More frequently perhaps it is the shrinking by 
 anticipation from the shock, so trying to a sensitive nature, 
 which follows the disturbance of a pleasant dream ; we would 
 rather have bad dreams than encounter the possible jar of a 
 sudden awakening ; or vanitv, for some reason, makes the 
 effort to throw off the painful emotion still more painful 
 than persistence ; or perhaps, most frequently, we are simply 
 making a convenient pretext for a display of bad temper or 
 melancholy which has been generated by some other cause. 
 To follow out such arijuments would lead me too far. I must 
 conclude this discussion by repeating once more the main 
 conclusion which I have sought to establish; for it seems to 
 me that a clear theory of ethics can only be attained by a 
 clear understanding of a proposition which, as misunderstood, 
 contains the ererms of countless fallacies. 
 
 17. The true proposition, then, that conduct is determined 
 by the feelings, has been constantly confounded with the 
 erroneous proposition that it is determined by the agent's 
 judgment of his happiness. This is expressed in the form 
 that the will is determined bv a kind of syllogism. The 
 major premiss is invariably — I will adopt the course of con- 
 duct which will j^roduce the greatest balance of happiness. 
 I am unable to admit the accuracy of this statement, 
 although I do not deny that in many cases it is an 
 approximate statement of the case. So far from admit-
 
 THE EMOTIONS AS DETERMINING CONDUCT. 57 
 
 ting the second proposition to be an expansion of the 
 first, I hold them to be really inconsistent. The feeling 
 which determines conduct is not a judgment at all, though 
 it is inseparably bound up with serious judgments. It is a 
 simple unanalysable fact. If we would, not define, but de- 
 scribe the feelino- in other words, we should rather call it a 
 psychical force. Love and hatred, desire and aversion, deter- 
 mine our conduct as physical forces determine the move- 
 ments of a bodv. Thev have a definite more or less ascer- 
 tainable value, of which we can form a judgment, but which 
 is not itself a judgment. The sense of fatigue, for example, 
 puts an absolute limit upon my energies, and implies no 
 direct reference to any future pains or pleasure. It is 
 there in such and such a degree, and we can say no more 
 about it. It is, or may be, the subjective correlate of 
 certain objective phenomena which might be described in 
 terms of nervous force, muscular tension, and so forth, and 
 is equally definite in its relations. The same is equally true 
 with all the direct emotions and sensations. They are 
 what they are, and underlie all judgments, instead of being a 
 product of those judgments. The possibility of forming any 
 judgment as to future pains or pleasures presupposes their 
 existence as the existence of forces is presupposed in calculat- 
 ing their actions and reactions. So far, again, there is no 
 ground for supposing that our pains and pleasures are in any 
 way regulated with a view to future enjoyments or suffer- 
 ino-s. This consideration beg-ins to su^rgcst itself so flir as our 
 immediate pains and pleasures are recognised as foretastes of 
 pains and pleasures to come. In that case, which is, of course, 
 more or less the case of nearly all the actions of reasoning 
 beings, the actual enjoyment, and therefore the conduct, is 
 more or less affected by our previsions of the future. A 
 secondary action is set up which tends to regulate the play 
 of the passions. It may happen, but, so far as we have gone, 
 it may also not happen, that the passions may be so regulated 
 that the conduct dictated by our immediate feelings may coin- 
 cide with that which would be dictated by a judgment of our
 
 58 THE SCIENCE OF ETHICS. 
 
 total happiness. And this leads us to the next problem to be 
 attacked. We can only be affected by the prospect of the 
 future in so far as we are reasoning beings. We must, there- 
 fore, consider in what sense the mere blind action of imme- 
 diate feelino; is o;overned and reo-ulated bv the reason: for 
 the principles hitherto considered do not imply a conscious 
 regard to general rules. A man's action at each point is 
 determined by his feelings, and therefore his whole course de- 
 pends upon the varying feeling, or, in other words, upon its 
 law. But this is equally true of man and beast, and does not 
 show what is implied in the acceptance of a general rule. 
 
 III. The Reason as Determining: Conduct. 
 
 18. We have therefore to consider in what sense the 
 conduct of every man is determined by his reason. I must 
 begin by taking note of one obvious but important distinction. 
 There is a sense in which we all admit that the universe is 
 reasonable throughout — the sense, namely, in which we simply 
 assert the validity of the universal postulate ; a sense in which 
 there is no distinction in point of reasonableness between the 
 fill of a stone and the workino: of a loo-ician's brain. As we 
 rise from the inoro;anic to the oro-anic, and a2:ain from the 
 lowest to the highest organism, the working formula, if I 
 may give that name to what is generally called the law of 
 the phenomena, becomes continually more complex, and 
 soon surpasses all our powers of statement. Every conceiv- 
 able event is, however, in this sense reasonable; which is 
 only to endorse once more the familiar assertion that chance 
 is but a name for ignorance. In this sense, again, no conduct 
 could be more or less "reasonable" than any other. The 
 most foolish vagaries of the most illogical mind have their 
 cause, and would be explained if we could look into the 
 mind of the agent. The errors of the puzzleheaded have 
 their ^' reason " as much as the soundest judgments of the 
 clearest thinker. It is plain, therefore, that we must use the 
 word in a different sense when we make it the basis of a
 
 THE REASON AS DETERMINING CONDUCT. 59 
 
 distinction between one thing and another. When we sav 
 that the conduct of Plato is reasonable, and that the dog or 
 the stone acts instinctivelv or mechanicallv, we incan that 
 Plato has a mind capable of apprehending a general rule, and 
 carrying on certain logical processes, and that these circum- 
 stances determine his conduct. He is a conscious and 
 reasoning subject as well as an object; and the distinction 
 might perhaps be expressed by saying, that whereas everv 
 conceivable phenomenon is objectively reasonable, or, in the 
 common phrase, has an assignable " law," intelligent beings 
 alone are subjectively reasonable, or determined in their 
 conduct bv perceptions and inferences. In this latter sense 
 men may be reasonable or unreasonable in the most varying 
 degrees. In the former, the unreasonable is the contradictory, 
 or, in other words, the non-existent. 
 
 19. Taking reason, then, in the only sense in which it 
 affords a basis of distinction, we shall find, I think, that 
 conduct is generally called reasonable in several connected 
 but not identical senses. Sometimes we are told that a man 
 is reasonable in so far as his reason controls his passions. The 
 reason is a separate faculty which in some way rules or sup- 
 plants the other faculties. A man, again, mav be called 
 reasonable so far as he possesses the faculty in a high degree, 
 as he apprehends certain general rules, and as they affect his 
 conduct; and this, again, may imply, on the one hand, that 
 he reasons accurately, and so has conceptions of the world 
 corresponding to the realitv, or, on the other hand, that his 
 judgment of what is most desirable is fixed and consistent. 
 In this, too, it is implied that he acts in such a way as really 
 to brino; about what he desires, or that he knows how to 
 proportion means to ends. And, finally, we may mean 
 something beyond all this, namely, that his ultimate ends 
 are "reasonable" or worthy in some sense not yet defined. 
 Reason, in short, may be opposed to passion or to want of 
 thought, or to want of proportion between mean and ends, or 
 may indicate consistency and loftiness of purpose. The animal, 
 the man of sudden and incontrollable impulse, the simple
 
 6o THE SCIENCE OF ETHICS. 
 
 dunce, or the consistent and calculating man who yet aims 
 at some purely sensual or selfish end, may each be called 
 unreasonable. In trying to answer the question before us, 
 what, namelv, are the characteristics of reasoned as dis- 
 tinguished from unreasoned action ? we shall incidentally 
 consider these various senses. 
 
 30. The supposed conflict between reason and passion is, as 
 I hold, meaningless if it is taken to imply that the reason 
 is a faculty separate from the emotions, and contemplating 
 them as an external spectator. Reasoning and feeling are, 
 according to me, bound together in an inseparable unity. 
 Every act of choice is a struggle between passions involving 
 more or less reasonins:, but not resolvable into an emotionless 
 process. The man who is distracted between the charms of gin 
 and duty is not divided between passion and reason, but 
 between a sensual pleasure and the love of home, or the fear 
 of hell, or the diso;ust of conscious degradation. If resistance 
 to these emotions gave him no pain he would not resist. 
 There must be emotion on both sides, as well as reason on 
 both sides, or a struggle would be impossible. It may be 
 that a reasonable man is more likely than an unreasonable 
 to resist the temptation to drunkenness — that is, that the 
 greater the development of the reasoning faculties the less 
 the love of gin — as the inverse proposition is pretty generally 
 true. But this is a very different statement, and points to a 
 different process of indisputable reality. Reason, in short, 
 whatever its nature, is the faculty which enables us to act 
 . with a view to the distant and the future. Consequently, 
 in so far as a man is reasonable, he is under the influence of 
 motives which would not be otherwise operative. The 
 immediate bodily appetite is held in check by a number of 
 motives to which only the reasoning being is accessible. His 
 mind's eye sees not only the public-house but the absent wife. 
 The reason so far leads a man to multiply the data from 
 which his conduct is calculable, or the more reasonable a 
 man may be the greater the variety of ways in which his 
 feelinsrs will be interested.
 
 THE REASON AS DETERMINING CONDUCT. 6i 
 
 2T. The impression, however, which leads us to oppose 
 feeling to reason is partly due to another fact. A large part 
 of our conduct is automatic; it is either not detcraTiined by- 
 conscious motives, or it is determined by motives which, 
 though they rise for a moment to the surface of consciousness, 
 are forgotten as soon as felt. Of our conscious conduct, 
 again, part may be caljed instinctive and part reasonable, 
 according as the motive does or does not include some refer- 
 ence to ulterior ends. These modes of action pass into each 
 other by imperceptible degrees. We say that the bird 
 building a nest acts instinctively, meaning to denv that it 
 has any view of the consequences, and to assert that it builds 
 because the act of building is in itself pleasant or the not 
 building painful. A man may, in the same wav, eat his 
 dinner instinctively, that is, without contemplating the con- 
 sequences to his health, and the same act may be reasoned 
 when those consequences are taken into account; w^hen, 
 for example, he eats purposely to gain strength for a special 
 contingency. The instinct may be converted into reason as 
 the consequences become manifest, and the reasoned action 
 become instinctive as the consequences are left out of account. 
 So, again, the instinctive action becomes automatic when it 
 is done without leaving any trace upon consciousness. It 
 raav still be voluntarv in the sense that the ag^ent may be 
 able to refrain if his attention happens to be aroused. 
 Habitual actions pass through all these gradations. We do 
 a thing first with some conscious motive; we come to do 
 it without reference to other consequences, and even without 
 consciousness; and the difference is, that w^hereas the action 
 previously required some effort, it now requires an effort to 
 refrain from the action. We are like a man walkino- in 
 chains, who for2:ets their existence until somethins: leads 
 him to diverge from the line imposed, and consequentlv to 
 be galled by the restraint. The motive is thus a latent force 
 which only manifests itself in consciousness under some 
 opposing action. 
 
 22. Let us now return to our gin-drinker. He feels the
 
 62 THE SCIENCE OF ETHICS. 
 
 desire, we may suppose, but the desire is instantly quenched 
 by the bare thought, say, of duty, or of the injury to his family. 
 We say, in popular language, that his passion has been 
 quelled by reason. The mere intellectual perception, however, 
 could have no effect if the sense of duty and love of family did 
 not represent a strong fund of emotion capable of being called 
 into vigorous operation. His conscience or his family affec- 
 tions have generated a habit. He has become an intellectual 
 automaton. He acts by a certain rule without calling into active 
 play the feelings by which the habit was orginally generated. 
 It now reqnires an effort to break through that rule as it 
 formerly required an effort to act in accordance with it. The 
 thought of family or duty is enough to convert the act of 
 drinking, which just before was simply the satisfaction of a 
 natural instinct, into a breach of a settled rule. The habit 
 probably acts without converting more of the latent force into 
 an active form than is necessary for the purpose of counter- 
 acting the temptation. The intellectual operation brings his 
 action under a new category, and so far it is quite true that 
 his passion is conquered by reason ; but it is not in virtue of 
 the purely logical operation, but of the fact that the reason 
 reveals a new set of forces ready to spring into action to the 
 necessary degree. 
 
 23. We may thus be said to feel by signs as well as to reason 
 by signs. When we are conducting a logical operation we 
 easily call up the full meaning of the symbols by which the 
 operation is conducted. Symbols are useful precisely because 
 they enable us to dispense with that laborious process. In 
 the same way our feelings are determined without callino- 
 into active operation the full meaning which they may con- 
 vey. The sight of a red flag may deter me from crossing a 
 rifle range without calling up to my imagination all the 
 effects of a bullet traversing my body. If the motive which 
 prompts me to run the risk be strong, it may be necessary to 
 convert a greater volume of latent into active emotion; and 
 as we frequently fail to do this, we often run risks which we 
 should avoid were the consecjuences distinctly contemplated-
 
 THE REASON AS DETERMINING CONDUCT. 63 
 
 We steer our course bv au apparcutly insiguificant rudder, and 
 only call out forces sufficient to overcome the actual resistance. 
 But it would be a great error to assume that the mere call 
 would be sufficient if the force were not at hand and ready to 
 supply the necessary assistance. And thus, if I may say so, 
 the game is generally decided as experienced players decide 
 games, bv a simple show of cards on both sides without 
 actuallv playing out the moves. 
 
 24. No theory can be tenable which virtually asserts reason 
 and feeling to be two separate and independent faculties, one 
 of which can properly be said to govern the other. The 
 reason is not something superinduced upon the emotions as 
 something entirely new. There is no absolute gap between 
 the lower and the higher organisms. The animal instinct 
 may be regarded as implicit reason, or the reason as a highly 
 developed instinct. Instinct is reason limited to the imme- 
 diate, and incapable of reflecting upon its own operations ; 
 and reason an extended instinct, apprehending the distant 
 and becoming conscious of its own modes of action. The 
 development of the whole nature implies a development both 
 of the intellectual and the emotional nature. The growth of 
 new sensibilities implies a power of detecting new qualities 
 and new relations between phenomena; and the growth of 
 mechanical power implies the capacity of bringing things into 
 fresh combinations, and so developing new sentiments. The 
 increased range of thought due to the power of forming 
 abstract conceptions and reasoning by symbols is associated 
 with an equal growth in the complexity and variety of the 
 corresponding emotions. That we may have a nature capable 
 of being stirred by such words as patriotism, philosophy, and 
 religion, we must not only have an intellect capable of opera- 
 tions utterly beyond the reach of the lower being, but we 
 must also have sensibility capable of a vast variety of equally 
 new sensibilities. It is natural, however, that we should 
 look upon the intellectual development as constituting the 
 essence of the change, and interpret the finer methods by 
 which the conflict of feeling is decided in the more developed
 
 64 THE SCIENCE OF ETHICS. 
 
 nature as implying that feeling is superseded by reason. 
 This interpretation, however, is really as erroneous as would 
 be the opinion that the finer instruments which enable us to 
 attain great mechanical results by a trifling expenditure of 
 force enable us to dispense with force altogether. A child 
 can raise a vast weight, with the help of certain machinerv, 
 by touching a spring or starting an electric current. We do 
 not infer that the effort is produced without force. Similarly, 
 the delicate and complex mechaiiical operations of a highly 
 oro-anised intellect may govern the conduct of the a(»;ent 
 without evolving any great expenditure of emotion. But it 
 is not that he acts without emotion, only that his emotions 
 act by more complex and refined methods. A word governs 
 them where the duller mind would require the actual stimu- 
 lus of a powerful passion. 
 
 25. Our problem, then, is not how does a reasoning differ 
 from an unreasoning being, but how does the being at a 
 higher stage of development, both intellectual and emotional, 
 differ from the being who is in both respects at a lower 
 stage? But it must also be observed that, in spite of the 
 close connection between the two elements, it is possible to 
 consider them separately. The intellectual faculties, for ex- 
 ample, may vary, whilst the emotional remain constant. It 
 is common to speak as though the reason might grow at the 
 expense of the emotions, and vice versa : and though we 
 may criticise the form of the statement, it represents the 
 undeniable fact that emotional activity is, in particular cases, 
 vuifavourable to certain forms of intellectual activity. In- 
 capacity for certain modes of fueling must limit our sphere of 
 experience in the matter of knowledge, but great excitabilitv 
 of the same feelings may hinder us from performing the 
 logical operations which always require a certain calm. The 
 most obvious measure of intellectual power is the accuracy 
 with which our ideal constructions represent the actual world 
 outside, and this may vary greatly in different moods with- 
 out a corresponding variation in its emotional quality. We 
 may conceive, for example, that two beings might be equally
 
 THE REASON AS DETERMINING CONDUCT. 65 
 
 absorbed by the passion of hunger; both might regard the 
 world as a gigantic kitchen, and arrange all possible objects 
 under the simple categories of eatable and non-eatable. But 
 one mi<j;ht difl'er widely from the other in the accuracy of his 
 judgments as to what was and what was not eatable, and his 
 skill in devising; means for conveying; the eatable into his 
 stomach. The reasoning- being; lives in a larg;er world than 
 the unreasoning, and he is so far more reasonable as his world 
 is more real. The height of unreason, in this sense, is repre- 
 sented by the case in which a man is under an illusion, and 
 supposes his limbs to be glass, or fancies that he can hold a 
 fire in his hand by thinking; on the Caucasus. Still the most 
 perfect sanity and the most delicate adjustment of means to 
 end, though it includes certain modes of conduct, leaves a 
 very wide margin of character; and men may be, in this 
 sense, equally reasonable whilst varying indefinitely in con- 
 duct. 
 
 2,6. The same considerations apply if we take a different test. 
 The accurate representation of the world implies an accurate 
 representation of our own feelings. A reasonable man, we 
 say, knows his own mind, a knowledge which, as satirists 
 and philosophers agree, is of the rarest. The sentiment 
 excited by any general name should represent that which is 
 actually excited in each particular case. If I accept the 
 principle of loving all men, I am inconsistent in hating any 
 individual of the class man. It would be extravagantly " un- 
 reasonable " to be charitable to every sufferer unless he had, 
 say, red hair or a black skin. In many such cases the eccen- 
 tricity amounts to a temporary change of character. The 
 sight of some hated symbol has a kind of physiological influ- 
 ence which rouses the wild beast within the man, and for the 
 time turns the philanthropist into a ruffian. The man, as 
 the phrase goes, is no longer himself. So, in other cases, the 
 man who is a stoic in his ideal world gives way to senti- 
 ment in the world of realities. This is, of course, possible 
 and common enough, and the development of reason will 
 tend to restrain such oscillations, as the widened area of
 
 66 THE SCIENCE OF ETHICS. 
 
 observation, the increased tendency to view the world as a 
 connected whole, and to bring all observations under general 
 principles must give continuity and consistency to the conduct 
 of life. We perceive that we act at different moments upon 
 different principles; that at one time our conduct is deter- 
 mined by such purely superficial considerations as the dislike 
 to a particular colour, whilst at another we go beyond the 
 superficial consideration to the interests of the human being 
 which are independent of such trifles. Reason will tend to 
 make such real inconsistencies vanish. But it does not ap- 
 pear that the bare condition of logical consistency is suffi- 
 cient to eliminate eccentricities. There is always some cause 
 for the wildest vagary or the most unreasonable prejudice. 
 To give a merely formal consistency to my conduct, it is suffi- 
 cient that this cause should become a reason; that the motives 
 by which I am actually determined should be represented in 
 the general rules which I frame. If hatred to the red-haired 
 actually influences me, I have only to dislike the red-haired 
 man in theory as much as I dislike him in fact to make my 
 conduct consistent in a formal sense. It is, of course, true that 
 this process of conscious representation of my actual character 
 will tend to modify the character itself; for it may be that 
 the motives by which I am in fact prompted will no longer 
 commend themselves when I try to generalise the principles 
 which they embody and to fit them with a code of conduct. 
 Still it would seem that the most eccentric set of pre- 
 judices might be stated coherently inasmuch as they do 
 in fact represent real qualities of a single individual. In 
 other words, the bare condition of logical consistence, of the 
 absence of direct contradiction from the feelings by which 
 I am prompted in my ideal constructions, and in the actual 
 occurrence of the case represented, does not suffice to define 
 my conduct any more than the accuracy with which my 
 ideal constructions correspond to the outside world. And 
 for this reason, as it seems to me, it is impossible to deduce 
 laws of conduct frovn the bare condition of consistency. 
 Any kind of conduct will satisfy that condition if the objective
 
 THE REASON AS DETERMINING CONDUCT. 67 
 
 reason or course of actioii be converted into the subjective 
 cause or reason for action. 
 
 27. But the development of our reason doubtless implies 
 more than a mere process of formulating our existing senti- 
 ments — something beyond a mere superioritv in the adapta- 
 tion of means to ends or in the knowledge of our own 
 characters and the external world, l^hat this must be so 
 follows, indeed, from the connection between the intellectual 
 and the emotional development. The man differs from the 
 beast, not simply by the addition of a faculty which enables 
 him more effectually to gratify the same passions, but by the 
 growth of a new set of emotions, which only exist in germ, 
 if they exist at ail, in the inferior beings. We have to ask, 
 then, how far this consideration will help to determine the 
 nature of reasoned conduct. We come upon the track of 
 the old discussion as to the siimmum bonum, that chief o-ood 
 which, it was conceived, must be desired by every one in 
 virtue of his being reasonable. If, in fact, it were possible to 
 define any such end as is implied in all reasoned conduct, we 
 should be at least on the way to a highly important conclu- 
 sion. Let us see how the difficulty presents itself at this 
 point of the inquiry. 
 
 28. We will assume for a moment that we may consider all 
 human conduct as determined by some single end ; that there 
 is some constant and definable object towards which all 
 desires converge, and which is common to all men. This is 
 to assume, if we look at the problem from the other side, 
 that all the feelings by which we are prompted may be reo^arded 
 as modifications of some single instinct. Man will resemble 
 the conventional miser whose sole object in life is to make 
 money, or, in the other mode of statement, whose sole passion 
 is avarice. So far as this can be re<rarded as at all representing- 
 the facts of the case, the influence of increased rationality 
 will be obvious. For it is plain that so far as the man 
 becomes more reasonable, all his activities will be brouo-ht to 
 converge upon a single line of action. Since the various 
 subsidiary ends have a common measure, his reason will enable
 
 68 THE SCIENCE OF ETHICS. 
 
 him to compare them, as the miser compares different courses 
 of action, simply in respect of the pecuniary profit. This 
 increased rationality will thus manifest itself in an increased 
 skill in proportioning means to ends, and in a more com- 
 plete knowledge of his own character. He will have to learn 
 by experience how far new combinations of circumstances 
 will gratify his ruling (or rather his sole) passion, which is 
 the same thing as discovering what is the law of the passion. 
 He is comparing modes of feeling identical in kind, and to 
 discover how he feels under given circumstances is to discover 
 what circumstances he prefers. This may often be difficult, 
 inasmuch as it mav be difficult to brins; tocrether in thouo-ht 
 two different states; but the criterion is always simple, 
 namely, the balance of gratification in one way or the other. 
 Finally, the fullest development of the reason would not, so 
 far, serve to determine the problem what is this sole or 
 master passion. That still remains as an arbitrary or entirely 
 undetermined element in the problem ; though if it could 
 once be fixed, the remainder of the problem would be deter- 
 minate. Given the love of money, or of any other definite 
 object as determining all the activities, and the rest is a 
 question of calculation. All the rules of action would appear 
 as co-ordinated into a systematic whole as different applica- 
 tions of the one fundamental principle, " Get money ; " as in 
 scientific inquiries all the particular cases appear as em- 
 bodying (and in that sense obeying) certain simple universal 
 rules. 
 
 29. Here, then, the criticism recurs that the happiness 
 which all men desire is not a simple end, but a name for 
 many and radically different forms of gratification. The de- 
 scription just given would hold good in strictness of nothing 
 but a polyp, an organism swayed by a single desire, say, a 
 love of warmth and a dread of cold, and representing not a 
 whole composed of parts but an indivisible atom. If such a 
 beinti^ could be supposed to be endowed with reason, its 
 actions would lu) doubt have a single end, and there could 
 be no dispute as to its siunmum honum. But as the growth
 
 THE REASON AS DETERMINING CONDUCT. (g 
 
 of reason implies the development of a vast complexity of 
 feeling, this apparent unity is illusory. Man, in fact, is a 
 microcosm as complex as the world which is mirrored in 
 his mind ; he is a federation incompletely centralised, a 
 hierarchy of numerous and conflicting passions, each of 
 which has ends of its own, and each of which, separately 
 considered, would give a different law of conduct. He is in 
 some sense a unit, but his unity is such as to include an 
 indefinite number of partly independent sensibilities ; and 
 consequently it is impossible to lay down any single end, 
 even for the individual, and still more obviously impossible to 
 lay down such an end as that which is in fact desired by all 
 individuals whatever their constitution. Whether we can 
 lay down such an end as that which all " ought " to desire is 
 at present not the question; for we are simply dealing with 
 the facts, and have not come within view of the meaning to 
 be attributed to " ought." 
 
 30. Now the difficulty thus suggested undoubtedly makes 
 it difficult to lay down any simple formula of conduct : it 
 rather tends to prove that any formula which professes to be 
 simple must be illusory. And, moreover, we have this diffi- 
 culty, that our psychology is at present utterly inadequate to 
 decide what are the elementary passions of which the organic 
 federation is composed, or in what sense they can be regarded 
 as distinct. We must be content with the vaguest and most 
 general propositions upon the subject. We feel that the 
 different physical appetites are in some sense distinct, and 
 perceive them to correspond to different bodily organs. Yet 
 thev are not independent ; for each organ and each appetite 
 depends at every moment upon the whole organism, or upon 
 the state of the other organs of which it is composed. The 
 difficulty becomes still greater when we come to the higher 
 or intellectual instincts, whose limits and differences are so ill 
 defined, and which have no distinct physical correlates. Leav- 
 ing it to the psychologists to describe as well as they can the 
 complex constitution of this mysterious federation, we may here 
 be content with one or two remarks. We may assume that we
 
 yo THE SCIENCE OF ETHICS. 
 
 are roughly describing facts when we say that the man, consi- 
 dered upon the emotional side, is built up of a certain small 
 number of primary sensibilities, each of which may be stimu- 
 lated in a vast variety of ways, and is subject to a number of 
 intricate actions and reactions from its neighbours. More- 
 over, it is the very condition of organic unity that they form 
 in some sense a whole, so that the action of each is dependent 
 at every moment upon the state of the others. A superior 
 being who could examine our characters would be able to lay 
 down the formula of our conduct, in which the determining 
 instinct would appear as the resultant of various subordinate 
 instincts, each actino; according to the stimulus of external 
 circumstances, and limited by its dependence upon the other 
 constituent parts of the organism. However different the 
 feelings may be in kind, they must be commensurable : they 
 have a certain value in terms of each other, and as parts of a 
 single whole they have a single and (by the superior being) 
 definable resultant. The formula would be beyond measure 
 complex : it would not depend upon the conditions of grati- 
 fication of a single instinct, but upon the gratifications of 
 several instincts, themselves connected by complex laws ex- 
 pressing the organic constitution of the agent. Instead of 
 supposing, according to my former illustration, that a man's 
 conduct is determined simply by the love of gold, we should 
 have to suppose that he wished for gold and silver and any 
 other metals in certain varying proportions, dependent upon 
 the laws of his own character, whilst the pursuit of each in- 
 volved more or less incon'sistent modes of action. But how- 
 ever complex the resulting formula, and of course it would 
 be indefinitely more complex than the illustration suggests? 
 it would still give determinate rules; and though the end 
 would be harder to define, it would be still definite. 
 
 31. Hence, when we return to the question of rationality, 
 we may say that the general result is not affected. For the 
 operation of reason will still tend to bring about a certain 
 unity in the result. So far as any instinct, whether simple 
 or complex, is dominant, the reason will tend to proportion
 
 THE REASON AS DETERMINING CONDUCT. 71 
 
 means to ends, and so far to bring about unity of action and 
 purpose. The various actions directed to the gratification of 
 that instinct will form parts of a coherent and intelligible 
 system. Further, as the emotions are closely connected, as 
 they blend with each other, and the whole process of develop- 
 ment is a process of forming a certain hierarchy in which the 
 separate and special instincts are subordinated to the more 
 central and massive, the reason will develop, if not a unity, at 
 least a harmony of action. For, so far as we reason, the action 
 of each separate instinct is controlled by a constant reference 
 to the requirements of others. We may act like the lower 
 animals under the immediate impulse of hunger, but our 
 huno-er is restrained, not onlv by the foresight of to-morrow's 
 appetite, but by the knowledge that this indulgence may be 
 at the expense of other pleasures. The passion is regulated 
 and restrained bv our desire of a more intellectual or emotional 
 enjoyment. Reason supplies, if we may say so, the flywheel 
 which makes every part of the machinery act under the in- 
 fluence of the other parts with which it is organically con- 
 nected. A caprice — such as the dislike to the red-haired — is 
 checked, because such a caprice implies obedience to the 
 sincrle emotion without reference to other considerations. 
 We act absurdly because we forget that the redness of 
 hair is not inconsistent with the persistence of the same 
 feelings and svmpathies which determine our conduct in 
 the case of men of a different complexion. So far as we 
 are reasonable we are able to consider our conduct, and 
 to consider in anv given case all that is implied in it. In- 
 stead of being guided exclusively by the superficial considera- 
 tion, we see in the mind's eye all the deeper and more im- 
 portant circumstances which it involves; and therefore the 
 strongest feelings, whatever they may be, the most permanent 
 emotions, are being constantly called into play to correct and 
 dominate the trifling ones. Each instinct has its voice in 
 determining the action of the federal government, but no one 
 is allowed to take the command exclusively without reference 
 to the wishes of the others.
 
 72 THE SCIENCE OF ETHICS. 
 
 32. But such considerations, which might be expanded 
 to much greater length, and which would require expansion 
 in a psychological treatise, still leave the important question 
 open. The reason brings the whole conduct into harmony 
 and unity; it forbids us to pursue trifling objects at the 
 expense of important; for instead of allowing each instinct 
 to operate exclusively in turn, it subjects each to the implicit 
 and explicit control of the others. And the very mode by 
 which our rational and emotional nature is evolved implies 
 this constant centralisation proceeding pari passu with our 
 reason in the complexity of thought and feeling. The great 
 central emotions become stronger as the discovered laws, 
 whether of our own or of external nature, become ranged 
 into a consistent hierarchy of principles. But we may still 
 ask, how is the relation between the different instincts de- 
 termined ? What settles the influence exerted by each member 
 of the federation ? What do we mean when we speak of one 
 interest as trifling and another as important? One passion, 
 as a matter of fact, is less powerful than another under given 
 circumstances, but the weaker is not therefore suppressed. 
 Hunger will still overcome love under certain conditions, 
 though the relative importance of the part which it plays in 
 the whole organism will be diminished (let us assume) as 
 the reasoning power increases. The character is determined 
 for each individual by its original constitution, though the 
 character is modified as the reason acts, not only because 
 reason accumulates a constant reference to certain motives, 
 and so gives them greater influence in determining conduct, 
 but because it enables us after a time to judge even of our own 
 character as a whole, to rehearse not only particular acts but 
 moods, and so become spectators of ourselves, and regard our 
 own feelings with disgust or complacency. Every such reflec- 
 tion tends to modify future action by revealing to us more dis- 
 tinctly its social consequences, and by investing it with cer- 
 tain associations of approval or disapproval. But, after all, 
 we start with a certain balance of feeling, with certain fixed 
 relations between our various instincts; and however these
 
 THE REASON AS DETERMINING CONDUCT. 73 
 
 may change afterwards, our character is so far determined 
 from the start. Again, it is plain that this varies greatly 
 with different people and gives rise to different types. In 
 one man the sensual passions have a greater relative import- 
 ance than in his neighbour, and so forth. And the question 
 arises, whether we can determine which of these types, or 
 anv of them, is most reasonable ? 
 
 ^^. So far as we have gone, I do not see how any conclu- 
 sion of this kind can be drawn. Assuming a certain end, as 
 I have said, we may say what is the most reasonable mode 
 of conduct, and what therefore will be the conduct of the 
 most reasonable man with a view to that end. But nothing 
 hitherto stated will enable us to define the end which is itself 
 most reasonable, or to give any meaning to the phrase. And 
 this argument becomes, if anything, stronger when we admit 
 that the end is exceedingly complex, instead of being simple 
 and uniform. Each tvpe of character has its own end, which 
 may determine the persistent and harmonious action. The 
 sensualist has purposes as definite and intelligible from his 
 point of view as the ascetic. The rule, " Let every one care for 
 me," is quite as simple, and, in a logical point of view, defines 
 conduct as consistently and reasonably, as the rule, "Love your 
 neighbour as yourself." That motive is most important for 
 any man which corresponds to his strongest and most fre- 
 quently stimulated instinct; but we have so far no means of 
 saying why reason should determine any particular relation 
 between the instincts, or why any one character should not 
 be just as reasonable as any other. In short, there is so far 
 an arbitrary element in our data which we have found no 
 means of determining. Man is both a reasoning and a 
 feeling animal. We can say what it is to be more reason- 
 able, in so far as reason implies a conformity between the 
 actual world and the ideal world by which it is represented ; 
 and this holds good whatever the precise meaning attributed 
 to those words by the metaphysician. We are still at a loss 
 to determine the other element of the problem : what is 
 that criterion, if any, by which we can judge of feelings?
 
 74 THE SCIENCE OF ETHICS. 
 
 They exist, and so far cannot be called true or false. They 
 are actual or they are nothing. If actual, in what sense is 
 one feeling, or set of feelings, or one type of character, better 
 than another ? That, of course, is the vital problem of all 
 ethical speculation, which will recur in various forms here- 
 after. To answer it at all, we must, as I hold, have recourse 
 to a different set of considerations, and they must be those 
 already indicated in the previous chapter, where I have 
 pointed out that the arbitrariness of our data is corrected by 
 the theory of evolution. 
 
 IV. Types. 
 
 34. Every reasoning agent, I have said, represents a certain 
 tvpe. Since his conduct is regulated in each particular by a 
 certain regard to the purposes of his life, and his sentiments 
 are constantly correcting and blewding with each other, there 
 is a certain unity of character in spite of the complex consti- 
 tution of his nature. Each impulse is subordinated to the 
 whole, and there is a subjective unity of sentiment corre- 
 sponding to the objective unity of organisation. The phrases 
 thus used are necessarily vague, and a kind of mystical sense 
 has been sometimes imported into them. We must try to 
 give them a little more precision; and, in particular, we must 
 ask what is meant by a " type," an expression which fre- 
 quently recurs in these discussions. Consider, for example, 
 a simple mechanical contrivance, such as a bow. The bow 
 was, we mav suppose, originally discovered by some simple 
 experience which revealed to some primitive savage the 
 properties of a special combination of wood and sinew. 
 Successive experiments gradually led to improved forms, by 
 some such process, it may be, as that which is illustrated in 
 Lamb's philosophical apologue of roast pig. Every improved 
 form would be used because it saved trouble. To have the 
 best form of bow was a matter of importance to each tribe 
 for use against beasts or enemies. The tribe which had the 
 best bow would have an advantage in the struggle for exist-
 
 TYPES. 75 
 
 ence. Whenever accident revealed an improvement, it 
 would be adopted whenever the powers of observation w^ere 
 keen enouo-h to note the savings of labour required for a 
 given effect. As men became intelligent, experiments 
 would be tried with more conscious and deliberate purpose, 
 and in time they might be carried out upon system, and 
 general formulae would be laid down. Discovery w^ould 
 become invention as the foresight and powers of calculation 
 developed, and in time mechanical theorists would become 
 able to lay down general rules, stating the relation between the 
 power and the materials of bows and arrows, and the range, 
 accuracy, penetration, and so forth, of the weapons. 
 
 ^^. Without pursuing this conjectural history, it is clear 
 that in any case there is a definite problem to be solved. 
 Given the materials and the purpose to be attained, one form 
 of bow would represent the maximum of efficiency. If, for 
 example, the archer has at his disposal a particular kind of 
 wood and has to pursue a particular kind of game, he will 
 have to satisfy a particular set of conditions which might be 
 expressed in mathematical formulae. Whether he has to work 
 out the best form of bow by rule of thumb, or whether he 
 is able to apply general rules, the process is essentially the 
 same, though in the latter case he can reason about whole 
 classes of cases as well as about particular cases, and can 
 understand more precisely what he is about, and be guided 
 against attempting impossibilities or satisfying contradictory 
 conditions. Even in so simple a case the complexity of the 
 data is so great that it might exceed the powers of the most 
 skilful mechanist to deduce the result from general principles. 
 But in any case the problem is determinate. The bow, we 
 may say, is there ; and in some way or other it may be formed, 
 however tentative or systematic the method. 
 
 ;^6. The bow, then, which represents the best solution of 
 the problem may be called the typical bow. It is that form 
 to which bows approximate so far as perfect, and every devia- 
 tion from which implies a defect. The actual may differ, 
 either as simply worse in all respects, or as gaining some
 
 76 THE SCIENCE OF ETHICS. 
 
 advantage at the price of others, such as a longer range with 
 less accuracy. The problem as to the degree in which it is 
 worth while to sacrifice range to accuracy can only be deter- 
 mined by its use, and presumably by a series ot" experiments. 
 And assuming the use and the materials to be constant, we 
 may regard the evolution of bows as a long series of attempts, 
 extending, it may be, over many generations, to solve the 
 jKoblem. The discovery of a new material or the applica- 
 tion of the bow to fresh purposes would set up a new process 
 of development. No bow is an actual realisation of the 
 type, for every bow, like every man, has its faults; nor is 
 the average the same as the typical bow, for deviations may 
 take place more frequently in one direction than another. 
 But the tvpical and ideal bow represents that maximum of 
 efficiency to which the bow tends to approximate, either by 
 the direct perception of its utility by the archers, or by the 
 fact that the archers who have the best bows are likely to 
 exterminate the others. 
 
 37. It is not irrelevant to observe that an instrument 
 which thus approximates to a type as the product of a slow 
 elaboration has a special beauty. The idea of the bow has 
 been rolled in countless minds till it is rounded to perfection, 
 like the pebble on the sea-beach. Everything superfluous 
 has been removed, as the trained athlete gets rid of super- 
 fluous tissues. It is exquisitely adapted to its purpose; and 
 the pleasure of contemplating it implies that we can perceive 
 this adaptation and derive from it that subtle sense of harmony 
 which governs our intellectual emotions. The pleasure, in 
 , fact, is another aspect of the skill implied in the so-called 
 "rule of thumb." Though we cannot calculate, we can feel 
 the utilitv. The recognition of the perfect adaptation reveals 
 itself to our feelino^s as aesthetic satisfaction. The instinct 
 outruns our power of scientific statement of the conditions; 
 for, in fact, the scientific knowledge corresponds only to that 
 small part of our knowledge which is capable of definite 
 articulate statement. 
 
 38. We may trace a similar principle in forms infinitely
 
 TYPES. 77 
 
 more complex. A Greek statue, we say, represents a type 
 of physical excellence. We cannot explain precisely how 
 the sculptor discovered the type, nor how we recognise it 
 when presented; but we do, in fact, recognise the solution 
 of an amazingly complex problem. The anatomist can tell 
 us, within certain limits, why one combination of bone and 
 muscle is more efficient than another. The artist is able to 
 present us directly with the form which represents the maxi- 
 mum of strength and agility possible to human flesh and 
 blood. The figure, we say, is perfectly graceful, because it 
 can perform any given task with the least expenditure of 
 strength, or make a given amount of strength do the most 
 work. Nothing is superfluous and nothing deficient. 
 The being represented could leap a given height with the 
 least exertion, or leap the greatest height with a given 
 exertion. And here too we see that the form is the best rela- 
 tively to certain conditions. The Hercules or the Apollo 
 may each be the best form as we require the greatest possible 
 strength to wield a club or the greatest possible activity of 
 locomotion. So, in fact, every athlete is generally best 
 at some particular exercise, or in some special department 
 of that exercise. The typical athlete represents rather a 
 typical group than a single form. Most defects are disquali- 
 fications for every kind of excellence, but, as between diff'erent 
 representatives of the typical group, we cannot say positively 
 that one is best unless we can define the function excellence in 
 which is to supply the criterion. And when this function 
 is determined, we have equally to assume certain data. If 
 the best athlete is to be the best runner, or say the best runner 
 for a given distance, we still have to remember that the athlete 
 has not only to be a runner, but a breathing, thinking, digest- 
 ing animal, to have lungs, brains, and stomach; and we tacitly 
 assume, in speaking of the typical form, that the resulting 
 conditions have to be satisfied. We do not speak, or rather 
 we cannot speak with any intelligible meaning, of the 
 greatest speed absolutely, but of the greatest speed obtainable
 
 78 THE SCIENCE OF ETHICS. 
 
 consistently with the general conditions implied in the other 
 wants of the human frame. 
 
 39. We can give, then, a precise meaning to the word 
 " type " so long as the end and the conditions can be regarded 
 as fixed. The bow has to bring about a certain result under 
 certain conditions. It is part of a whole, and may be con- 
 sidered as a supplementary limb discharging a given function 
 in the organism. But when we take the whole organism, 
 the meaning requires modification. The eye is useful to the 
 hand and to the whole organic system; the hand, again, to 
 the organism, including the eye. Each part is good or bad 
 relatively to the whole; but what is implied by the whole? 
 When there is some external criterion the answer is easy. 
 We can say what is the typical sheep from the point of view 
 of the butcher, or again from the point of view of the wool 
 merchant; but, what is the typical sheep considered ab- 
 solutely ? If the word " absolutely " be used in such a sense 
 as to exclude a reference to any conditions whatever, the 
 question ceases to have any meaning. It is intelligible to ask 
 what is the greatest speed of an animal, or of a steam-engine, 
 or of an electric message, because in each case I assign a 
 limit of possibility; but to ask what is the greatest speed 
 possible for anything is to ask a question which by its nature 
 has no answer. What then is the criterion bv which the 
 sheep is to be judged? We may say that it is the sum total 
 of all the sheep's relations to the external world, for we have 
 no means of regarding one relation either than another. 
 Again, each form of sheep plays its own part in the general 
 system, and the judgment of butcher, wool merchant, or hunter 
 will differ according to his particular needs. How are we to 
 decide which judgment is right or in what sense we can apply 
 risrht or wrons; to such estimates? 
 
 40. The answer to this question, which I must accept, for 
 I cannot inquire into the ultimate grounds, is that supplied 
 by the evolutionist. That theory, I take it, implies the 
 following statements. The organism — sheep or man — is not 
 a simple aggregate of independent parts which might be put
 
 TYPES. 79 
 
 toc-cthcr accordins: to one confio-uration or another. In that 
 case all that we could say of it would be, that an organism 
 was composed of such and such parts or that a conscious being 
 had such and such instincts; but that, for anything we could 
 say, the same elements might be put together in any other 
 way. The ultimate constitution would remain as a perfectly 
 arbitrary set of data. We should be unable to get beyond 
 the difficulty in which we have been already landed. But 
 we learn from the theory of evolution that as the individual 
 organism is composed of mutually dependent parts, and that 
 its existence involves the maintenance of a certain equilibrium, 
 so each organism supports itself as a part in a more general 
 equilibrium, and that its constitution depends at every 
 moment upon a process of adaptation to the whole system 
 of the world. And this may be expressed by saying that 
 every animal represents the solution of a problem as well as 
 a set of data for a new problem. As the bow is felt out, the 
 animal is always feeling itself out. The problem which it 
 solves is how to hold its own against the surrounding pressure 
 and the active competition of innumerable rivals. A species, 
 indeed, does not simply adapt itself to absolutely fixed con- 
 ditions, like wax poured into a rigid mould. In altering 
 itself it alters to some extent its environment. By extir- 
 pating a rival race it sets up a whole series of actions and 
 reactions, implying a readjustment of the whole equilibrium 
 amonffst all the races with which it is in contact, and to 
 some extent an alteration of the inorganic world. A new 
 state of things slowly substitutes itself for the old, but in 
 such a way that each species is continuous with the preceding, 
 and has been slowly remoulded by an incessant series of 
 unconscious experiments conducted under the constant con- 
 dition that failure means extirpation. Hence, though we 
 cannot say that either the end or the conditions are absolutely 
 constant, and though any full statement would have to be 
 unendingly complex in consequence, the whole process is 
 describable as a slow elaboration of types. The material at 
 any moment is a species, a group of organised beings, capable
 
 So THE SCIENCE OF ETHICS. 
 
 of varying within certain limits, placed under conditions 
 which, for the moment, are apparently fixed, and succeeding 
 in so far as it realises the condition of maximum total 
 efficiency. The capacity for holding its own replaces the 
 condition of fitness for a fixed external end. 
 
 41. We assume, then, that, although we cannot apply an 
 d priori method, although we cannot define the materials 
 of which men are made, or the end which they have to 
 fulfil, we can determine to some extent their typical excel- 
 lence. Recognising the general nature of the great prob- 
 lem which is being worked out, we can discov'er what is 
 implied in some of the results. The process of evolution 
 must at every moment be a process of discovering a maxi- 
 mum of efficiency, though the conditions are always varying 
 slowly, and an absolute maximum is inconceivable. At 
 every point of the process there is a certain determinate 
 direction along which development must take place. The 
 form which represents this direction is the typical form, any 
 deviation from which is a defect. This typical form, again, 
 so long as evolution continues, cannot represent an ultimate 
 result ; it is not (if I may say so) at the summit of a peak, 
 but on the summit line of a perpetually ascending ridge. 
 The actual forms are almost always more or less defective ; 
 the individual is below the ridge, not at a higher point of a 
 ridge. Gradually, indeed, if development continues, the type 
 itself will improve ; one particular form will differ from the 
 previous variations by being at a higher point of the ridge, 
 and in that last it will represent the advance towards a new 
 species. 
 
 43. To take into account all the corrections which would 
 be necessary for an accurate statement would be very diffi- 
 cult, and for my purpose the task would be superfluous. 
 The exposition and establishment of the theory of evolution 
 lies beyond the ethical problem, and is one of the data 
 which we must be content either to repudiate or (as I do) 
 take for granted. I will merely add one remark which 
 may indicate some of the difficulties involved in a complete
 
 TYPES. LosAtiguu^ 8 1 
 
 statement. I have said that every type must be relative to 
 the assumed end. As one form of athletic excellence is best 
 for one exercise and another for a different exercise, so in 
 the higher qualities there is room for an indefinite variety 
 of qualities, each of which may be best in certain relations. 
 There may, indeed probably there must be, properties com- 
 mon to all the typical forms, but they must be such as to 
 be reconcilable with great individual variation. So, for ex- 
 ample, a man may be a poet, a philosopher, a statesman, and 
 so forth, and we may say that to each function there corre- 
 sponds an appropriate type. Now it is conceivable that the 
 highest excellence in different departments of conduct may 
 imply consistent conditions. The greatest philosopher may 
 also be the greatest athlete and the greatest poet. It is 
 equally clear that there is no necessary connection. Brains 
 of abnormal power may be associated with puny muscles. 
 The sensibility of a poet, the preoccupation with abstract 
 principles of a philosopher, may unfit either for business. 
 The statement, indeed, is always ambiguous ; for it does not 
 follow that the poetic sensibility might not be combined 
 with the qualities of a man of business, as (apparently) was 
 the case of Shakespeare. It seems, however, to be a highly 
 general rule, that great excellence has, as it were, to be bought 
 at a price, and that efficiency in one direction implies de- 
 ficiency in others. Here, then, occurs the question — only 
 to be solved by experiment — What is the relative value of 
 different kinds of efficiency ? And a complete answer might 
 bring out the fact, which seems on other grounds probable, 
 that it is an advantage to a race to include a great variety 
 of different types. To take this into account would require 
 a new complexity of statement. It is enough, however, to 
 say here, that by speaking of a type I do not mean to assert 
 that there is one special constitution, conformity to which 
 by any individual of a race is a condition of efficiency, but 
 simply that the process of evolution is always the working 
 out of a problem which implies the attainment of general 
 efficiency by the acquisition of certain general qualities. 
 
 F
 
 82 THE SCIENCE OF ETHICS. 
 
 IV. The Principle of Utility. 
 
 43. These considerations^ however, lie rather outside of 
 the relevant argument. Before proceeding, it may be well 
 to call attention to the nature of the position here adopted. 
 I have, in fact, shifted my point of view. Speaking of 
 motives, I have argued that conduct is determined by pain 
 and pleasure. The painfulness of a state is a sufficient, ulti- 
 mate, and sole reason for avoiding it. But in speaking of 
 the elements of character, I have substituted for the considera- 
 tion of pain and pleasure the consideration of conditions of 
 existence. The fact is simply that the constants in one 
 problem are variables in the other. Given a certain char- 
 acter, the agent does what gives him pleasure. But if 
 we ask how he comes to have that character, the only mode 
 of answering is by referring to the conditions of existence. 
 His character must be such as to fit him for the struggle of 
 life. The reason of conduct is always its quality in terms 
 of pain or pleasure. The cause of its being painful or 
 pleasurable is the constitution of the agent ; and for this 
 constitution we can only account, so far as we can account 
 for it at all, by considering it as a variable, dependent upon 
 the conditions of life. Only in this way does the problem 
 from which we started become determinate. If we take 
 character as fixed, the development of reason can only imply 
 a harmony, an adaptation of means to ends, and so forth, 
 leaving the end or the dominating instinct in itself a positive 
 or arbitrary datum. As character varies, so will the ends 
 vary; and from the simple consideration of consistency, or 
 of pain and pleasure, we cannot by any ingenuity determine 
 what will be the general law of conduct. Hence the neces- 
 sity and the importance of the other mode of investigation, 
 as enabling us to assign more or less precisely the conditions 
 of this constitution, which, from the other point of view, 
 must be taken for granted. This double mode of reasoning 
 involves, therefore, no inconsistency, and is forced upon us 
 
 ^-^
 
 THE PRINCIPLE OF UTILITY. 83 
 
 by the conditions of the case. We may regard conduct 
 either as painful and pleasurable, or as conducive or not 
 conducive to the permanent existence of the agent. And 
 hence we have the consideration that there must be a corre- 
 lation between painful and pernicious actions on the one 
 hand, and pleasurable and temporal on the other. A man 
 will do what pleases him, and, if he is to live, must do what 
 is good for him, or at least what is not destructive. The 
 " useful," in the sense of pleasure-giving, must approximately 
 coincide with the " useful " in the sense of life-preserving. 
 This is a fundamental doctrine from the evolutionist point 
 of view, and requires a little further consideration. 
 
 44. We may remark, in the first place, that all conduct 
 may be considered as a set of habits, to each of which^ so far 
 as it is voluntary, there is a corresponding instinct. I use 
 both words in the widest possible sense. By a habit I mean 
 any mode of conduct which can be brought under a general 
 rule, and this, of course, would include the automatic as well 
 as voluntary actions. I use instinct, again, to include all 
 conscious impulses to action, whether including more or less 
 reasoned choice, and whether innate or acquired. Some 
 mode of feeling conditions all conscious action. Whether 
 the immediate action is pleasant in itself, or pleasant because 
 regarded as part of a whole system of actions, involving 
 remote consequences discoverable by refined calculations, it 
 still involves in the last result some pleasurable emotion. 
 The operative instinct may be an animal instinct, such as 
 hunger, or an instinct involving high intellectual develop- 
 ment, such as patriotism or religion. Where an action is 
 simply an end to some remote purpose, and absolutely in- 
 different in other respects, it is the foretaste of the pleasure 
 derivable from the remote consequences which determines 
 the action; and, of course, it is often difficult to decide 
 what may be the true motive. But in any case, there is 
 some motive, and the instinct means the sensibility to that 
 class of motive. We may then compare habits and their 
 corresponding instincts, in respect of their painfulness and
 
 84 THE SCIENCE OF ETHICS. 
 
 pleasurableness, or in respect of their utility, so far as they 
 imply total efficiency or the reverse; which means, again 
 (when we consider the organism primarily and not the con- 
 sequences of action), in respect of their essentiality or super- 
 ficiality. 
 
 45. There are, in fact, some habits which are essential to 
 the organisms — such habits (if the word may be used in 
 so strained a sense) as breathing or digesting. They are 
 the processes which constitute life rather than maintain it, 
 and are, for the most part, automatic, or beyond the direct 
 reach of our volition. We can kill ourselves as we can kill 
 another man, and so suspend digestion; but our volition 
 does not immediately affect the digestive process in our 
 own more than in another man's stomach. From this point 
 of maximum necessity the habits graduate to the most super- 
 ficial and transitory. Some may be changed or abolished 
 by a transitory change of feeling, whilst others can only be 
 altered by a reconstruction of our whole character. And 
 this suggests a remark, familiar enough, but worth the 
 explicit statement, because many fallacies spring from its 
 neglect. It is simply this : that every theory about an organ 
 or any property of an organism must be understood with a 
 reference, tacit or express, to the mutual interdependence 
 of every part of the organic whole. If, for example, I say 
 that a certain habit is pleasant or painful, useful or perni- 
 cious, I must remember that the habit cannot be simply 
 deducted as from a sum total of independent wholes, and 
 that, on the contrary, its removal must involve to some 
 extent a readjustment of the whole organic equilibrium. It 
 is therefore essential to take into account all the indirect or 
 underlying consequences involved in this consideration, as 
 well as to strike an average of the good or bad effects of the 
 habit itself. In order that the proposition, ''This habit 
 is a bad one," may have any real meaning, we must assume 
 that the organism can exist without the habit. If this be 
 possible, then wc must take into account all the differences 
 necessarily involved. Wc must compare the whole man as
 
 THE PRINCIPLE OF UTILITY. 85 
 
 he exists with a certain quahty to the whole man as he exists 
 without it. To suppose that any given characteristic can 
 be simply subtracted, is to argue as though cutting off a leg 
 had no other consequence than removing a crutch. In 
 one case the whole character of the man is affected, in the 
 other it remains constant. The fallacies which result from 
 an oversight of this obvious truth are so common and pesti- 
 lent, that it is necessary to take note of their nature. We 
 are always tempted to assume that we can take off a habit 
 as we can take off a coat, and to suppose that some political 
 or social evil can be remedied by simply removing the most 
 obvious source of the mischief without troubling ourselves 
 to ask what other organic changes will be set up. This 
 error is indeed so common as to be almost the master-error 
 of a crude sociology, and it may be described as a virtual 
 confusion between oro;anic and merelv mechanical wholes. 
 
 46. It is important to remember this in considering the 
 principle just laid down of the correlation betweeen painful 
 and pernicious habits. Some such process, I have said, is 
 necessarily implied in the evolutionist theory, and is implied 
 at every stage. The highest organism has been built up from 
 the lowest under the constant stress of this condition. From 
 the vague wriggle of the worm who acts upon the implicit 
 formula, "Any change may be for the better," though one 
 wriggle may save and another slay, as it takes him into or 
 out of the beak of the bird, up to the most refined motives 
 and delicate calculations of the philosopher or statesman 
 who meets a danger by a specific series of carefully co- 
 ordinated actions, the general principle must still be the 
 same. So far as the instincts of any agent lead him into 
 danger, they are a point against him; so far as they lead him 
 into safety, a point in his favour in the ceaseless competition. 
 In every case, too, we must consider the total effect. The 
 growth of a new instinct brings fresh dangers as well as fresh 
 advantages. The problem which is always being worked 
 out is whether the new form is on the whole more or less 
 capable of holding its own. It does not simply drop out one
 
 86 THE SCIENCE OF ETHICS. 
 
 instinct and insert another^ but is more or less modified 
 throughout, so as possibly to develop unsuspected qualities 
 in directions very remote from that which is most conspi- 
 cuously concerned. 
 
 47. This suggests the problem as to the true theory of 
 pain and pleasure. Granting the general truth of this prin- 
 ciple of correlation between the two kinds of utility, we 
 may ask whether it cannot be pushed further? May it not 
 be possible to show from the nature of the case that pain 
 and pleasure are necessarily connected with pernicious and 
 beneficial processes respectively? It has in fact been main- 
 tained, and with a considerable show of evidence, that there 
 is some intimate connection between pleasure and a state of 
 heightened vitality on the one hand, and between pain and a 
 state of lowered vitality on the other. I will not go into an 
 argument which is not strictly relevant to my case, and 
 which would lead me far beyond the limits of my knowledge, 
 and, as I imagine, beyond the limits of all established science. 
 The connection, whatever it may be, can only be discovered 
 by careful and prolonged investigations; it takes us into the 
 perplexing region of obscure physiology; it gets mixed up 
 with controversies as to the ultimate metaphysical problem 
 of the relation between subject and object. I will only venture 
 one observation, bearing upon the limits within which the 
 principle is applicable. We may certainly say, in general 
 terms, that there is as close a connection between health 
 and happiness as between disease and misery, and that the 
 anomalies which present themselves in attempting to gene- 
 ralise this theory might be cleared up by a more accurate 
 investigation. But this does not tend, properly speaking, 
 to explain the connection, nor docs it show with any pre- 
 cision how close the connection must be. The principle 
 may reveal a tendency and as exhibiting certain conditions 
 of organic life, but it does not in any sense explain the 
 existence and the intimate nature of pain and pleasure. 
 
 48. Thus, in the first place, if we use the word "utility" 
 in the subjective sense, it means, as I have said, nothing
 
 THE PRINCIPLE OF UTILITY. 87 
 
 but painful and pleasurable. Every pleasure is so far good, 
 every pain so far bad; and it is a contradiction in terms to 
 speak of a pain as useful. The only admissible meaning of 
 such a phrase emerges when we regard the system as a whole, 
 and admit that a certain pain is useful because a loss of the 
 susceptibility to the pain in question would involve a greater 
 pain on the whole. The being enjoys the maximum happi- 
 ness possible under its conditions of life, although its con- 
 stitution involves a certain admixture of pain. If, again, we 
 speak of utility in the other and objective sense, a pain may 
 be useful so far as it determines to preservative actions, 
 although this, again, only implies utility in the first sense 
 so far as preservation implies a balance of happiness. Now 
 we must assume that, in some way or other, painful or 
 pleasurable states are dependent upon the material conditions 
 of the organism, and, moreover, dependent upon its actual 
 state at each moment, and upon its mode of working, not 
 upon its absolute strength or ability. The absolute strength, 
 in fact, means its potential capacity of resisting any strain 
 to which it might be subjected. If the strain is not actually 
 present, the capacity to bear it may not be required. Many 
 diseases are painless; weak lungs, for example, often give 
 little pain unless the patient is called upon for some exertion. 
 The healthy lung, we must suppose, has great reserves of 
 power; but so long as they are not called into action, it 
 matters not whether they exist or not. It is the right work- 
 ing of the machinery which is relevant, not the power of 
 the machinery to bear work under different circumstances. 
 Till the crash comes, the weak fibre which does all that 
 is needed is as useful and produces no more pain than the 
 strong fibre. And so, again, the fact that any mode of conduct 
 may lead to bad consequences hereafter does not of itself 
 produce either pain or pleasure, though to the reasoning 
 animal the anticipated evil is already an evil. In the absence 
 of that anticipation we have only to ask whether the organic 
 equilibrium is maintained, and have no concern with results. 
 I walk with equal comfort whether I am advancing to a
 
 88 THE SCIENCE OF ETHICS. 
 
 pitfall or along a firm path, or whether I am or am not 
 weakening some vital part, so long as I am not weakening it 
 in such a way as to interfere with the maintenance of a cer- 
 tain equilibrimn. 
 
 49. We must suppose, then, that pain and pleasure are 
 the correlatives of certain states which may be roughly 
 regarded as the smooth and the distracted working of the 
 physical machinery, and that, given those states, the sensa- 
 tions must always be present. So far we do not come in 
 sight of any question of utility or of any sense in which 
 pain can be regarded as useful. We reach that point of 
 view where we take into account the principle already stated, 
 but incapable (as it seems to me) of any proof except that 
 of observation, that pain corresponds to unstable and 
 pleasure to stable states of equilibrium. So far, that is^ as 
 our feelings determine our conduct, they determine us to 
 avoid pain and retain pleasure. And this is quite enough 
 to show that there is a tendency to correlate painful and 
 pernicious, pleasurable and beneficial. The agent which 
 delights in states which generally love pernicious conse- 
 quences is so far self-destructive. An animal which liked 
 poisonous food would be unable to exist when the food was 
 easily obtainable. And it is clear that this condition, which 
 has been always operative from the earliest stages of evolu- 
 tion, must maintain the correlation within certain limits. 
 It is equally clear that the correlation is very far indeed 
 from perfect. The growth of a new taste involves new 
 dangers as well as new advantages, and may establish itself if 
 the new animal as a whole is better than the old as a whole. 
 If external circumstances change, new relations are developed 
 which may bring out new evils. The taste of a savage for 
 stron<i" drink may be harmless because dormant until the 
 civilised man introduces firewater. There is, again, no 
 obvious reason why a perfectly useless pain may not endure 
 indefinitely. The agony caused by certain diseases must 
 remain as Ions; as the constitutional state which it reflects is 
 producible under actual circumstances; and to bring about
 
 THE PRINCIPLE OF UTILITY. 89 
 
 such a change in the organism as would make this' state 
 impossible may involve such a total change in its whole 
 construction, involving all manner of collateral changes, as 
 may never be realised, or be realisable only at the price of 
 developing other evils. And, again, many actions once 
 pleasurable tend to become automatic, and so there is 
 apparently at least a loss of happiness without any obvious 
 compensation. We may perhaps guess that there is even 
 in this case an advantage, inasmuch as the automatic per- 
 formance of any action tends, so to speak, to set the con- 
 sciousness free for other purposes; and thus, although the 
 most highly developed agent is the most perfect automaton, 
 he has also for that very reason a greater range of conscious 
 thought and feeling. Such considerations are enough to 
 remind us how limited is our knowledge and how rapidly we 
 are reduced to vague conjectures. This only may be said, that 
 the correlation in question does not imply that all our 
 pleasures and pains are useful in the sense of tending to 
 self-preservation, as indeed such a correlation would be 
 manifestly contradictory to all our experience; and, again, 
 it does not tend to explain in any way whatever the 
 existence of pain and pleasure, for this existence is presup- 
 posed, and is presumably dependent upon a direct relation 
 between the feelings and the organic processes, which is 
 to us absolutely inscrutable ; but at the same time it re- 
 presents a highly important condition, which must regulate 
 at every stage the process of evolution. It only justifies us 
 in saying, that so far as any agent takes pleasure in things 
 conducive to his preservation, he has a better chance of 
 survival, and, therefore, that we may regard compliance 
 with this condition as one cardinal point in the theory of 
 organic development. 
 
 50. How much further can we proceed ? One remark is 
 of course obvious. So far as any instinct is deeply seated, 
 so far, that is, as a being would require a complete recon- 
 struction to exist w^ithout it, we may consider that it has a 
 presumption in favour of its utility. It may be absolutely
 
 90 THE SCIENCE OF ETHICS. 
 
 essential to the existence of the organism ; and even if 
 plainly non-essential, we may fairly argue against its being 
 pernicious. An institution which has flourished in many 
 different ages and races under the most various conditions 
 must presumably fulfil some want and correspond to some 
 deeply seated instinct. So, to take a familiar example, it is 
 possible that the taste for stimulants may be injurious, but 
 it has in its favour the number of healthy and vigorous races 
 in which it is found, showing that at least it cannot be 
 destructive of vigour under all circumstances ; and, again, 
 we must admit that it satisfies a very widely spread desire, 
 which may be dependent upon a profound constitutional 
 necessity, and which will have to find its satisfaction in 
 one way or another. The problem, therefore, is not solved 
 by a simple summation of good and bad results, but involves 
 an inquiry as to the place filled by a desire for stimulants 
 in the whole economy of life. Whether any conclusions 
 assuming to be called scientific can be reached by such 
 methods need not be considered at present. The illustra- 
 tion may at any rate give some general indication of the 
 true import of the question to be put and the data required 
 for a satisfactory answer. 
 
 V. Social ajid Individual Uiiliti/. 
 
 51. I have spoken of instinct, habit, and so forth, as being 
 essential or non-essential to the organism, and have added 
 that between these two relations there may be an indefinite 
 number of gradations. What, then, is precisely meant by 
 *^' essential " in this connection? At a 2;ivcn moment anv 
 instinct may be essential in this sense, that the agent 
 could not exist without it. A man's life may depend upon 
 his possessing abnormal speed, strength, and eyesight. The 
 life of the organism depends at every instant upon his rela- 
 tion to the surrounding world, and by varying them we may 
 vary the requirements indefinitely. In speaking, therefore, 
 of any organism generally, we tacitly assume the existence
 
 SOCIAL AND INDIVIDUAL UTILITY. 91 
 
 of the appropriate medium. A capacity is only essential if 
 it is essential under these normal conditions. A power 
 of breathing air is essential to certain classes of animals. 
 They cannot live in any other medium, and they cannot 
 live in air if they are incapacitated for this function. The 
 bare existence of the animal implies the existence of certain 
 conditions and of certain corresponding powers, and those 
 faculties are properly essential without which it could not 
 live anywhere, not those without which it could not liv^e in 
 some particular set of cases. Admitting this, there occurs 
 another difficulty. The process by which the correlation 
 of pernicious and painful states is worked out is one which, 
 by its very nature, must take a considerable number of genera- 
 tions. Races survive in virtue of the completeness of this cor- 
 relation. But the quality which makes a race survive may 
 not always be a source of advantage to every individual, nor 
 even, if we look closer, to the average individual. Since 
 the race has no existence apart from the individual, qualities 
 essential to the existence of each unit are of course essential 
 to the existence of the whole. If one animal cannot live 
 without lungs, a million cannot. But the converse proposi- 
 tion does not hold. In fact, there is a large and important 
 group of instincts in regard to which it is manifestly untrue. 
 The sexual and parental instincts are essential to the race, 
 for without them the race would cease in a single generation. 
 It is equally certain that they are not essential to the indi- 
 vidual. An animal deprived of them may not only live and 
 thrive, but will avoid many dangers to which it is exposed by 
 possessing them. The " unnatural " mother has the great 
 advantage that she will not give her life for her young. And 
 it is at least conceivable, though it may not happen actually, 
 that some creatures thus devoid of passions upon which the 
 continuance of the race absolutely depends may be not only 
 happier under certain special conditions, but may, on the 
 averas:e, enjoy more happiness than their neighbours. In such 
 cases, the parents virtually sacrifice themselves for the good 
 of the race. They may be unconscious of the sacrifice, and
 
 92 THE SCIENCE OF ETHICS. 
 
 we cannot call them unselfish. They are rather in a state 
 of mind in which^ as devoid of all prevision of consequences, 
 the question of selfishness and unselfishness has not yet pre- 
 sented itself. They act in a way not calculated to bring in 
 the greatest amount of happiness, but they act in obedience 
 to an instinct not guided by any calculation as to the fall of 
 the balance. 
 
 53. Here then is a case, and one of the very highest 
 importance, for it concerns the germ of all social life, in 
 which we see that the correlation between the benefi- 
 cial and the pernicious must be interpreted in a sense 
 different from that which we miijht at first siijht take for 
 granted. An instinct, that is, grows and decays not on 
 account of its effects on the individual, but on account 
 of its effects upon the race. The animal which on the 
 whole is better adapted for continuing its species will 
 have an advantage in the struggle, even though it may 
 not be so well adapted for pursuing its own happiness. 
 Here, then, it becomes desirable to attempt to bring into 
 greater distinctness the true meaning of the contrast between 
 the individual and the race, in order that we may endeavour 
 to determine in what sense there can or cannot be a conflict 
 between the individual who is the product of the race, and 
 the race which is itself formed of individuals, and in what 
 way the principle already laid down must be explained or 
 modified when we take this distinction into account.
 
 ( 93 ) 
 
 CHAPTER in. 
 
 SOCIAL MOTIVES. 
 I. The Individual and the Race. 
 
 I. The last chapter has brought us to a distinction of 
 vital importance. We have had to distinguish between the 
 effects of the interests of societies and the interests of 
 individuals. An action or an instinct may, it seems, be of 
 essential importance to the whole, and of little or none to 
 the individual, and this distinction will obviously affect 
 our reasoning at many points. If we assert that the survival 
 of an instinct is determined by its utility^ we must 
 further decide how the utility is to be measured, whether 
 we are to consider the utility to the individual or the utility 
 to society, and in what sense any distinction is possible. 
 Other difficulties may reveal themselves as we proceed. We 
 will, therefore, begin by attempting to define as clearly as 
 we can the true meaning of the distinction in question. 
 We may thus be able to understand the nature of the social 
 bond so far as is necessary for our purpose, and to consider 
 what language will be best adapted to express the relations 
 involved. 
 
 ■2. It is of course obvious that the individual and the race 
 are not two separate things capable of coming into collision. 
 The individual, as I have said, is the product of the race ; 
 and the race the sum of the individuals. And the fact that 
 the immediate interests of an individual may be incom- 
 patible with those of the race does not necessarily affect 
 our statement. The existence of such incompatibility is 
 of course only too familiar a fact. The prosperity of a
 
 94 THE SCIENCE OF ETHICS. 
 
 Napoleon may involve the degradation of his country. But 
 when I speak of " the individual " as being better or worse 
 adapted to his circumstances, I am not speaking of any 
 particular person, but of the average person. A Napoleon 
 may conceivably thrive by possessing qualities which are 
 injurious to his fellows. But it would be something very 
 like a contradiction to suppose that the average man might 
 be improved by conferring upon each qualities prejudicial 
 to the rest. If the average person is more intelligent and 
 richer, the whole of which he forms a part ,^ssesses a 
 greater sum of intelligence and wealth. If the qualities of 
 any society may be regarded — and in some sense it seems 
 that they must be regarded — as the sum of the qualities of 
 its constituent units, it follows that whatever strengthens or 
 weakens the average unit must also weaken or strengthen 
 the whole. This, indeed, would be accurately true if 
 we were justified in considering each unit as so far inde- 
 pendent that what is true of each might be applied to 
 all by direct multiplication. The difficulty begins to show 
 itself when we regard society, not as a mechanical aggregate, 
 but as an organic whole. In that case, we cannot regard 
 the efficiency of the whole as a simple sum of separate 
 efficiencies. The qualities of each unit may then be depen- 
 dent for their nature and for their efficiency upon their 
 relations to the organism. Thus, for example, if we take 
 an army, its efficiency will depend partly upon the strength 
 of the soldier, and partly upon his discipline. Double the 
 marching power of each soldier, and you will double the 
 length of marches possible for the army. But an increase 
 in the spirit of discipline produces an effect upon the army 
 which cannot be determined by simple multiplication ; for 
 not only may the efficiency of the army be enormously 
 increased by a slight increase in the spirit, but such an 
 increase may operate very differently under different states 
 of the army. It might diminish its efficiency in warfare 
 which required separate action, as it might enormously 
 increase it where the first necessity was unity, and there-
 
 THE INDIVIDUAL AND THE RACE. 95 
 
 fore blind obedience. A quality useful to the whole when 
 acting together might be prejudicial to the individual mem- 
 ber when acting independently. Hence, it would seem, 
 we have in this case a datum which could not be determined 
 without some knowledge of the total organism, as well as of 
 its separate parts. The distinction, therefore, between the 
 race and the individual, though not a distinction as of two 
 separate things, may be of great importance as correspond- 
 ing to a distinction in the mode in which the efficiency is 
 affected hy^ different qualities. For some purposes a body 
 may be regarded as an aggregate, whilst for others it can 
 only be understood as an organic whole. And this has 
 evidently an important bearing upon our reasoning. 
 
 3. Let us see how this difference must be expressed upon 
 the theory already laid down. I have spoken of qualities, 
 instincts, organs, and so forth, as being either essential or 
 non-essential to an organism. It is plain that it w-ould be 
 idle to ask what any organism would be without any of its 
 essential qualities. We should in that case be referring to 
 a mere nonenity. When we say inan, we mean, amongst 
 other things, a living being with a stomach, and environed 
 by eatable matter. To say that a man would be better or 
 worse if he had no stomach, is to put together words which 
 have no real meaning whatever, or, in any case, to speak of 
 some creature so radically different from a man for most 
 purposes, that it would lead to mere confusion to apply the 
 same name to it. You might describe a statue as a man 
 without organs, but this is simply to play with words, 
 unless we confine our reasoning to properties dependent 
 exclusively upon external forms. By "man," we mean a 
 being belonging to a given class, and varying within the 
 limits determined by the essential properties of the class ; 
 and amongst these essential properties we must, of course, 
 reckon dependence upon a race. Man means a being born 
 of woman, and perhaps a being ultimately descended from a 
 monkey. It would, therefore, be sheer nonsense to speak 
 of a man as if he either might or might not be in some
 
 96 THE SCIENCE OF ETHICS. 
 
 respects independent of society. He may be in the position 
 of a Robinson Crusoe, and living in a desert island ; but 
 even so he must have been begotten, born, kept alive 
 through infancy, and have inherited whatever qualities are 
 implied in those processes. A man not dependent upon a 
 race is as meaningless a phrase as an apple that does not 
 grow upon a tree. The words have no sense in a purely 
 arbitrary sense. And further, it is equally clear that the 
 best type of man must mean the best type of man developed 
 under those conditions. The best kind of bread means the 
 best food that can be made out of grain; and though a lump 
 of granite might have some qualities which in a different 
 relation are better than those of bread, it would be a mere 
 juggle if we said that the bread least likely to spoil was 
 " bread " made of granite. 
 
 4. It follows that the distinction drawn between the 
 social and the self-regarding qualities, or^ again, between 
 qualities as useful to the race and useful to the individual, 
 cannot possibly be ultimate distinctions. Every man is 
 both an individual and a social product, and every instinct 
 both social and self-regarding. To say that a man is an 
 organism is to say that each of his organs is so dependent 
 upon all the others that it cannot be removed without alter- 
 ing the whole organic balance ; or, as I have said, that a leg 
 is not, or is not solely, a crutch. If we speak, then, of one 
 instinct as referring to the society and another as referring 
 to the individual, we must always remember that each of 
 necessity implies the other. In speaking of them apart, we 
 are using the artifice of the mathematician who considers 
 one set of symbols to be variable and another as constant, 
 not as meaning that the quantities which they represent are 
 really fixed or in reality independent, but simply as enabling 
 him to calculate more easily by disentangling separate sets 
 of consequences. The social qualities are developed on the 
 invariable condition that the self-regarding qualities exist, and 
 vice versd, and the " best " qualities mean the best consistent 
 with this condition. We may, as I have already said, con-
 
 THE INDIVIDUAL AND THE RACE. 97 
 
 sidcr any organ by itself, and, for example, say that one man 
 has perfect lungs ; but the perfection is relative to their form- 
 ing part of an organism which has also a stomach, whilst the 
 best form of stomach means also the best for a man with 
 lungs. And precisely the same reasoning applies also to the 
 mental or emotional faculties. Whatever distinctions may 
 afterwards be drawn between them, we must never ignore 
 their necessary connection or mutual implication. As the 
 man is an individual, the process by which he is developed 
 is a single process; and in speaking of any one instinct, 
 property, organ, or faculty, there is always a tacit or express 
 reference to the whole organisation. 
 
 5. How, then, does the distinction arise ? The answer 
 may be suggested by the illustration just vised. The depend- 
 ence of an apple upon a tree is absolute. It admits in this 
 sense of no degrees. I cannot say, therefore, that an apple 
 owes certain qualities to the fact of its growing upon a tree, 
 for it owes all its qualities to that fact. The non-tree-grown 
 apple is a nonentity. But it is equally plain that, in another 
 sense, the dependence admits of many degrees, for the 
 possibility of distinguishing between the two classes implies 
 that for some purposes they are separable. And this must 
 mean that the apple has certain qualities which are inde- 
 pendent of its relation to the tree, not in the sense that they 
 would exist if that relation were abolished, but in the sense 
 that they may vary whilst that relation remains approxi- 
 mately constant. In some respects I may treat of the apple 
 as though it were an independent unity, because it may 
 change without a corresponding change in the tree ; in 
 other respects, I can only understand the changes in the 
 apple by taking into account its dependence upon the tree. 
 And hence, as these properties and the proportion between 
 them may vary in different kinds of apples, I may say that 
 some apples are more dependent upon the tree than others ; 
 not as denying that in every case the dependence is absolute, 
 but simply as asserting that in some cases the qualities 
 which are only intelligible through that dependence, or 
 
 G
 
 98 THE SCIENCE OF ETHICS. 
 
 which vary directly with its variation, are more prominent 
 than others. The comparison is not drawn between an 
 apple growing on a tree and others not growing upon trees, 
 but between the apple in which these properties immediately 
 dependent upon that relation are more prominent, and 
 others in which they are less prominent. And this general 
 statement will hold equally true in regard to the essential 
 properties of any organism whatever. 
 
 6. Let us see, then, how this applies to the general pro- 
 blem of the relation between the individual and the race. 
 There are, or there may be, organisms in which the distinc- 
 tion so far disappears that we need not take it into account. 
 Every living thing must be capable of propagating its kind, 
 and must so far have a property useful to the race. But we 
 may suppose the existence of organisms in which the rela- 
 tion between individuals is limited to this reproduction. The 
 insect may lay eggs which come to life in the next season ; 
 the successive generations may inosculate without overlap- 
 ping ; whilst during life each insect may exist in complete 
 independence of its fellows. If we further suppose that the 
 production of eggs be essential to the insect life — which, 
 whether an actual case or not, is a conceivable case — we 
 should have a case in which the interests of the race and the 
 individual would be identical. The successive individuals 
 would be so many links in a chain, and each would potentially 
 contain the whole series of descendants. Whatever hurt the 
 individual would necessarily so far hurt the race. Now, it 
 is to be observed that even in such a case the dependence 
 of the individual upon the race is absolute. Without repro- 
 ductive powers the race would not exist. And, further, 
 since an insect with superior powers of reproduction would 
 so far be more efficient, the type must be elaborated with 
 reference to this condition ; an increased power of locomo- 
 tion might be a disadvantage on the whole if it involved a 
 diminished power of reproduction. But we may be justified 
 in supposing that a variation In the locomotive faculties may 
 take place, whilst the reproductive power remains approxi-
 
 THE INDIVIDUAL AND THE RACE. 99 
 
 mately constant. This is expressed by saying tliat the insect 
 would be better so far as it could fly better. We do not for 
 the moment attend to the collateral results ; and so far the 
 efficiency due to better flying is measurable by the results 
 to the individual insect. Supposing the occurrence of a 
 new form with greater powers of flight, it would so far be 
 a superior insect as those powers adapted each insect better 
 for obtaining food, avoiding its enemies, and so forth. 
 
 7. To advance a step, we may take the famous case of bees, 
 which has afforded so many parallels to poets and philo- 
 sophers. In such a case the sexual relations, though they 
 cannot be more essential — for essential does not strictly 
 admit of degrees — are more prominent. The queen-bee, the 
 drones, and the workers are each dependent upon the others 
 for their continued existence. The individual insect is not 
 intelligible by himself. The race can only be continued by 
 the co-operation of different individuals with corresponding 
 differences of organisation. The best insect must now mean 
 the best relatively to the society of which it forms a part. 
 That form of bee will flourish which forms the most efficient 
 hives. The hive, in other words, will be the unit which 
 must be taken into account in considering the general pro- 
 blem of surviv^al. It would be therefore as idle to ask which 
 would be the best form of bee considered apart from the 
 hive, as it would be in the previous case to ask which would 
 be the best form of insect considered apart from its power 
 of reproduction; for in either case we are abstracting from 
 an essential property. Here, again, as in the former case, 
 we have certain faculties which may be supposed to vary 
 whilst the social qualities remain fixed. The bees which fly 
 better are so far better, assuming the power of flight to be 
 gained without a compensating loss of the qualities which 
 fit the bee for society. The difference is that we now have to 
 consider an organism which has more functions dependent 
 immediately upon the whole of which it forms a part, and 
 intelligible only through that whole. The hive, we may say, 
 is at once an aggregate and an organic whole, and we may
 
 loo THE SCIENCE OF ETHICS. 
 
 consider it in either character for purposes of analysis, 
 though we must not overlook the tacit implication that 
 each set of qualities is valuable only by reference to its com- 
 patibility with the others. 
 
 8. At this point, however, occurs a consideration which 
 is of vital importance to the argument. The individual 
 bee, I have said, is intelligible only through its relation 
 to the hive. Its properties, the individual as well as the 
 social, are developed, either indirectly or directly, by refer- 
 ence to the constitution of the hive. Now, in every case, 
 every quality of the organism is intelligible only through the 
 environment. The environment of the queen-bee consists 
 partly of the drones and working bees, and partly of the air, 
 flowers, and so forth. It is dependent upon the one just as 
 it is dependent upon the other. What, then, is the reason 
 for distinguishing? The answer is, that the distinction may 
 or may not be of vital importance, according as we are 
 considering one or other problem. Given the organisation 
 of bees, the behaviour of the queen-bee, for example, will 
 depend partly upon the flowers and partly upon the drones 
 and working bees amongst which it is placed. So long as 
 the organisation remains fixed, we may count both the 
 remainder of the society and the surrounding objects as parts 
 of the '^environment" between which it is unnecessary to 
 make any distinction. And for many purposes this assump- 
 tion is accurate. But if we consider the organisation as 
 variable, as we must do if we are considering the problem as 
 to the merits of a particular kind of bee, the distinction 
 at once becomes important. For in that case an organic 
 variation in the queen-bee necessarily supposes a correlative 
 variation in the organisation of all other members of the 
 hive. Wc cannot, for example, suppose the queen-bee to 
 acquire better wings, without supposing a correlative change 
 to take place in the wings of all its descendants. We may, 
 on the other hand, suppose, and in some cases we must 
 suppose, the flowers to remain unaltered, and to be part 
 of the fixed conditions to which the swarm adapts itself.
 
 SOCIETY AND MAN. loi 
 
 And hence the distinction is needless, or is vitally impor- 
 tant, according as we are dealing with changes which do or 
 do not imply a fixed organisation. At any given moment 
 the organisation is fixed ; if we speak of periods dnring which 
 evolution introduces sensible changes, it is not fixed. But 
 there is yet a further case. If it were admissible to suppose 
 that the hive was capable of acquiring new properties whilst 
 the organisation of its members remained fixed, we should 
 be forced to introduce a reference to this varying condition 
 also. To determine the conduct of the bee, we must know 
 not merely its organisation and its environment, but the 
 state of the hive. We might, again, deal with problems into 
 which this fresh datum entered either as a constant or a 
 variable, and we should in each case have to decide before 
 we could lay down a satisfactory formula of bee life, which 
 new theory did or did not involve a reference to this vari- 
 ability of this element. Now, as we shall see directly, this is 
 a consideration of essential importance in theories of human 
 society. 
 
 II. Society and Man. 
 
 9. This follows from a consideration of some very familiar 
 truths. We have sufficiently shown that we cannot make 
 a comparison between man in a social state and the nonentity 
 man independent of society, the real comparison being be- 
 tween man at an early and man at a comparatively late 
 stage of social development. So it would be idle to discuss 
 the effect of light upon the eye by comparing an eye which 
 is sensitive with an eye which is not sensitive to light, for 
 such an insensitive eye would not be an eye at all; but 
 we may determine very profitably how the eye which is 
 highly sensitive difi'ers from the eye which is but slightly 
 sensitive; and, in the parallel case, we have to compare 
 men at remote stages of social development in order to 
 determine the eflfect of this element in their constitution. 
 Now two assumptions may be made ; we may in the first place 
 take for granted that between the savage and the civilised
 
 102 THE SCIENCE OF ETHICS. 
 
 society there is a vast difference^ including, amongst other 
 things, the presence of a recognised and formulated moral law. 
 It is, indeed, a question for the philosophical observer how 
 far rudimentary systems of morality may be recognised even 
 amongst the rudest savages; and we may assume that, as 
 will hereafter be stated, germs of moral sentiment, the feel- 
 ings and instincts which in a more highly developed state give 
 rise to the moral law, are to be found not only amongst 
 savages, but in some sense even at a far lower stage of develop- 
 ment than the human. I assume simply that the explicit 
 recognition of certain general rules of conduct^ the observ- 
 ance or breach of which is attended with moral approval 
 or disapproval, is comparatively a recent phenomenon. The 
 relations of man to woman, of parents to children, of the 
 individual to the primitive social unit, whatever it may be, 
 exist at the lowest point of the scale, and no doubt corre- 
 sponding modes of conduct were regarded with some kind of 
 sentiment as far back as we need go in the history of the 
 race. But a distinct recognition of general regulative prin- 
 ciples is only possible when the reflective and reasoning powers 
 have become developed and some sort of theory of human 
 life has gained acceptance. And, in the next place, we may 
 take for granted that this difference does not imply a cor- 
 responding difference in organisation. There is no reason 
 to suppose that the innate faculties of a modern European 
 differ essentially, or that they differ very greatly, from 
 those of the savages who roamed the woods in prehistoric 
 days. There is clearly no reason to suppose that the brain 
 of a modern English baby is intrinsically more developed than 
 that of an ancient Athenian baby. Yet there is a vast 
 difference in many ways between the morality of the adult 
 Englishman and that of the adult Athenian, and still more 
 between the morality of the Englishman and that of the 
 Scandinavian pirate or the wielder of flint implements. It 
 is presumable, therefore, that the moral development is not 
 to be explained solely as corresponding to any organic change 
 in the individual.
 
 SOCIETY AND MAN. 103 
 
 10, Hence, for a historical solution of the problem of the 
 moral instincts, it would be necessary to compare man and 
 society as it now exists with that which existed previously 
 to the evolution of a distinct moral system, and to show how 
 the social change had been brought about without a corre- 
 sponding change in the individual organisation. In other 
 words, we may for this purpose consider man — that is, the in- 
 dividual as born with certain capacities and characteristics — 
 as approximately a constant, and then show how the society 
 which is constituted of similar raw material comes to differ 
 so materially in the properties of the manufactured article. 
 To trace the process fully would be to give a complete history 
 of morality, which is both beyond my powers and irrelevant 
 to my immediate purpose. It will be sufficient if I consider 
 briefly how such a process is conceivable, and what is implied 
 in its realisation. 
 
 11. Social development takes place without a correspond- 
 ing change of individual organisation. A modern gunboat 
 could crush the fleets which fought at Salamis, and a modern 
 child could solve problems which bewildered Archimedes; 
 and in whatever way we may explain this change, we 
 certainly cannot interpret it as implying that the average 
 child of to-day is born with faculties radically superior to 
 those of Archimedes or of Themistocles. The change ob- 
 viously depends upon the ancient and familiar truth that 
 man can accumulate mental and material wealth; that he 
 can learn by experience, and hand over his experience to 
 others. It may be that germs of this capacity are to be 
 found in the lower animals, but we shall make no sensible 
 error if we regard it, as it has always been regarded, as the 
 exclusive prerogative of humanity. An unreasoning animal 
 can only adapt itself to new circumstances, except within 
 a very narrow range, by acquiring a new organisation, or, 
 in other words, by becoming a difiercnt animal. Its habits 
 and instincts may therefore remain fixed through countless 
 generations. But man, by accumulating experiences, can 
 virtually alter both his faculties and his surroundings with-
 
 I04 THE SCIENCE OF ETHICS. 
 
 out altering his organisation. When this accumulation 
 extends beyond the individual^ it implies a social devclop- 
 mentj and explains the enormous changes wrought within 
 historical times^ and which define the difference between the 
 savage and the civilised man. Let us consider for a moment 
 some of the conclusions which may be inferred from this 
 cardinal fact. 
 
 12. Imairine an exhaustive statement of the differences 
 
 O 
 
 between modern England and the England of thirty cen- 
 turies back — of the England inhabited by twenty mil- 
 lions of civilised beings and the England which was the 
 hunting-ground of a few tribes of wandering savages. We 
 should have to begin by noticing vast material changes — 
 rivers embanked^ marshes drained^ forests felled, roads con- 
 structed, houses built, fields tilled, wolves supplanted by 
 sheep — and making a calculation of the industrial capital 
 and the artistic treasures accumulated. All this, we should 
 have to observe, implies the possession not only of different 
 materials, but of a new set of tools. The modern, though 
 not inheriting greater faculties at his birth, can compel 
 material force to co-operate with him. He has gunpowder, 
 steam-engines, and printing-presses where a wheeled carriage 
 or an iron nail would have astonished his predecessor as a 
 miracle of art, and would have been unattainable bv an 
 equal expenditure of intellectual energy. This, again, implies 
 the inheritance, not only of materials and of tools, but of 
 skill. To the savage a telegraph or a book would be so much 
 iron wire and rags. The modern has at his disposal vast 
 accumulations of knowledge. He knows the properties of 
 substances, the form and character of his dwelling-place, 
 the history of his race ; innumerable products of previous 
 intellectual energy in the shape of discovered laws of 
 nature, mathematical formulas, philosophical, religious, and 
 political speculation, are at his service. The knowledge 
 existing in different times has become incomparably too vast 
 for any single brain. Much of it, we may even say, exists 
 in no brain, and yet is potentially there. It is externalised
 
 SOCIETY AND MAN. 105 
 
 in countless books and papers laid up in accessible places. 
 We inherit not merely the tangible products of labour but 
 the-methods of labour. Our ancestors transmit to us both 
 results and the means of obtaining fresh results ; they trans- 
 mit their mechanical skill and their logic, although they 
 do not transmit any modification of structure. The infant 
 always starts at the same point of intelligence, but the path 
 has been cleared for him, so that he can reach an enormously 
 more distant goal. A child is not born a clockmaker now 
 any more then he was three thousand years ago, but not 
 the less does he inherit the power of making clocks. 
 
 13. The most striking illustration of this process is to be 
 found in language. Language is clearly a product of the 
 social organism in this sense, that its development does not 
 imply any essential change of the organs, but a social develop- 
 ment. It was not bestowed upon men from without, nor 
 is it a necessary product of our organisation, or worked out 
 by each man for himself. It has been gradually elaborated 
 from some simple germ by the race under the pressure of 
 social needs; and each child accepts it as a ready-made instru- 
 ment from his nurse and transmits it to his successors, impress- 
 ing upon it at rare intervals some relatively trifling improve- 
 ment. If we notice how much a child learns in simply learn- 
 ing to speak, we shall see how much it owes to the society in 
 which it is placed. To learn to speak is to learn a number 
 of signs with which to fix in the memory a number of things 
 or aspects of things which would else be forgotten, and to 
 enable ourselves to recall them easily to the memory of others, 
 and to have them recalled by others. It is, again, to have 
 both outward objects and the emotions which they excite 
 arranged in groups, so as to facilitate the reproduction of old 
 impressions and to render accurate and speedy the complex 
 processes involved in every act of reflection. Language is 
 so essential an instrument of learning, that it is very difficult 
 for the mature mind to conceive of any but the simplest 
 process of thought taking place without it. Thus in learning 
 a language we learn a logic; for the structure of language is 
 
 v-»
 
 io6 THE SCIENCE OF ETHICS. 
 
 determined by the elementary methods of reasoning, which 
 in its turn determines the methods of those who speak it. 
 As every instrument supposes certain methods of using it, 
 the mechanism of language implies the acquisition of corre- 
 sponding mental habits. Thus, again, language implies the 
 unconscious absorption of a philosophy, as is abundantly 
 clear to any one who will trace the use of such words as 
 matter, form, suhstance, spirit, and so forth, and observe how 
 all speculation has started from the attempt to analyse what 
 was already implicitly given by words embodying previous 
 results of thought. We are metaphysicians in the cradle, 
 and distinguish object and subject by methods instilled into 
 us by our nurses. We start with an implicit psychology, 
 as the names of the various emotions imply a rough classi- 
 fication of the primary elements of character. The same is 
 true to some extent of every branch of inquiry. The child 
 learns the Ptolemaic system of astronomy as soon as he can 
 talk about the moon and the stars. A philosopher who 
 wishes to introduce a new conception has to invent a new 
 terminology, which is yet always a modification of the old 
 symbols, and in the very act imposes fetters on his own mind, 
 and provides moulds in which thought is to run hereafter. 
 Often, it need hardly be said, he introduces some insidious 
 sophistry from which it is very difficult to escape. Philosophy 
 is in great measure a series of attempts to escape from the 
 erroneous conceptions thus tacitly introduced in the very 
 earliest forms of speech. And, finally, it may be observed 
 that language naturally affects our feelings as well as our 
 conceptions. Words not merely denote an object, but 
 associate it with certain emotions. We catch the subtle con- 
 tagion of prejudice from the language which it has impreg- 
 nated. We hate a race because its name has been used 
 as a term of abuse. Papist amongst Protestants, heretic 
 amongst Catholics, Jew amongst Christians, are words which 
 have been used to propagate bitter hatred combined with an 
 almost complete ignorance of the hated object. Briefly, to 
 teach a child to speak is to educate it, to prepare it for
 
 SOCIETY AND MAN. 107 
 
 association with others^ to lay it open to all manner of 
 influences, to start it with a mass of knowledge already 
 elaborately organised, to teach it methods of thinking and 
 imagining, to insinuate into its mind philosophical and 
 religious principles, and to inoculate it with innumerable 
 associations which must be important elements in the 
 development of its character. 
 
 14. The child, then, starts with an organisation for 
 thinking and speaking as the bird leaves the egg with the 
 organs of flying or swimming. Whether and in what sense 
 this organisation is the product of a previous elaboration is 
 not at this point a relevant question. In each case, at any 
 rate, the infant starts with certain faculties. But the 
 essential diiference is clear. llie eagle learns to fly, the 
 duck to swim, the child to walk as eagles, ducks, and children 
 have flown, swaim, and walked since the species appeared ; 
 from the organisation we can safely infer the function; 
 but in the case of language this is no longer true. The 
 child as it grows up inherits not only the faculties, but the 
 modifications and elaborations of its faculties which have 
 been developed by the society in which it lives. Its organs 
 of speech, its intellectual powers, might be equal to those 
 of Homer or Plato ; but if Homer or Plato had been born 
 amongst the Hottentots, they could no more have composed 
 the "Iliad" or the ^'Dialogues" than Beethoven could have 
 composed his music, however fine his ear or delicate his or- 
 ganisation, in the days when the only musical instrument 
 was the tom-tom. The instrument, whether it has or has not 
 a material embodiment, is equally essential in both cases. The 
 analogy of vocal music would serve as well as that of instru- 
 mental. The art of singing has to be created by the labours 
 of successive generations of musically endowed generations, 
 whether the instrument be the human throat or the fiddle. 
 Hence, the activity of the individual is essentially conditioned, 
 not merely by his individual organisation, but by the social 
 medium. His predecessors have created a new world. The 
 physical basis may be the same, but the man develops under
 
 io8 THE SCIENCE OF ETHICS. 
 
 a set of influences which profoundly affect his intellect, his 
 emotions, and his activities. The material world is not more 
 altered by cultivating fields and manufacturing tools than 
 the world of thought by the development of language and 
 all that system of logically organised methods of reasoning 
 which determine the lines of discharge of intellectual energy. 
 15. Human conduct, then, depends essentially upon the 
 social factor; we must study the properties of the social as 
 well as of the individual organism in order to understand it ; 
 and in this is already implied a further condition of vital 
 importance. The individual, that is, is dependent at every 
 moment upon his contemporaries as well as upon his ances- 
 tors. When I have learnt a language, I have an instrument 
 which will serve me in solitude. I can sit down in my 
 study and speculate or imagine as I please. My thoughts, 
 it is true, are modified at every instant by the instrument 
 elaborated by others, but the instrument remains, when once 
 acquired, a constant factor. But this, of course, is an in- 
 finitesimal part of the ordinary use of language. As it 
 has been developed by the need of communication, it also 
 serves at every moment as a means of communication, and 
 it is as governing my relation to my fellows that it exercises 
 themost palpable and continuous influence; and this isequally 
 true of all the other social faculties. Almost every action of 
 my life is dependent more or less directly upon the co-opera- 
 tion of others, and the more so as I become more civilised. 
 I cannot think without assuming the knowledge attained by 
 others. I see that my fire is low; I feel that I am too cold ; 
 I infer that I should put on coals. Even in so simple a case I 
 use inherited results of the experience of others, and especially 
 of the great discovery of fire and its properties. But I am 
 also dependent upon the continued co-operation of others. 
 I could not arrange the details of a day's work without 
 taking into account the conduct and the continuous action, 
 for example, of those processes which determine my supply 
 of coal. I cannot think to any purpose without taking for 
 granted the veracity and intelligence of innumerable fellow-
 
 SOCIETY AND MAN. 109 
 
 men, and fitting my own results into the vast mass of results 
 attained by others. Each individual, in whatever depart- 
 ment he labours, assumes that others are labouring in tacit 
 or express co-operation. If millions can live in a region 
 which formerly supported a few thousands, it is because each 
 of the millions has millions of co-operators. If I can devote 
 myself to write an ethical treatise, it is because thousands of 
 people all over the world are working to provide me with 
 food and clothes, and a variety of intellectual and material 
 products. If another man lives by putting one brick on 
 another, it is because he can trust the discharge of other 
 essential functions to the numerous classes who are contri- 
 buting more or less directly to his support, protection, and 
 instruction. Briefly — for it is useless to dwell upon this 
 familiar topic — the growth of society implies that division of 
 functions which has been more or less recognised by every 
 one who has considered the question since the dawn of 
 speculation. This vast social organisation is the work of a 
 vast series of generations unconsciously fashioning the order 
 which they transmit to their descendants. It is only neces- 
 sary to take note of the familiar fact, because some of the 
 consequences which it implies have been neglected or insuffi- 
 ciently emphasised. Briefly, society exists as it exists in 
 virtue of this organisation, which is as real as the organisa- 
 tion of any material instrument, though it depends upon 
 habits and instincts instead of arrangements of tangible and 
 visible objects. 
 
 16. It is, again, obvious that as every man is born and 
 brought up as a member of this vast organisation, his char- 
 acter is throughout moulded and determined by its pecu- 
 liarities. It is the medium in which he lives, as much as the 
 air which he breathes or the water which he drinks. And 
 this implies not merely, from the facts already noted, that 
 his intellectual furniture, his whole system of beliefs, pre- 
 judices, and so forth, are in a great degree acquired by 
 direct transference, and that consciously or unconsciously he 
 imbibes the current beliefs and logical methods of his fellows.
 
 no THE SCIENCE OF ETHICS. 
 
 but also that he is educated from infancy by the necessity 
 of conforming his activities to those of the surrounding 
 mass. If his feeHngs or behefs bring him into conflict with 
 his neighbours, he is constantly battered and hammered 
 into comparative uniformity. To deviate from the beaten 
 track is to expose oneself to incessant collisions. If it is the 
 custom to keep to the right in a street, I can only go to the 
 left at the risk of being trodden under foot. And thouirh 
 some results of this process are lamented by many reasoners, 
 it is clear that in some degree it is a necessary condition of 
 all progress. It is as necessary to conform to certain rules 
 as to accept certain beliefs. If I insisted upon trying for 
 myself the effect of every kind of food, I should not survive 
 my first crucial experiment upon arsenic. If I deviate from 
 ordinary rules, I so far deprive myself of the advantages of 
 co-operation, because my neighbours are unable to foresee 
 my conduct or to act in harmony with me. If I insist upon 
 dining only in the middle of the day, I am so far debarred 
 from society ; if I object to the ordinary forms of worship, I 
 cannot have the stimulus of common prayer. If to gain 
 the advantages I accept the rules, my character is modified 
 accordingly. Be the result bad or good, all organisation 
 implies uniformities of conduct, and therefore continuous 
 discipline. We are born, not into a chaotic crowd, but into 
 an organised army, and we must learn to keep step and rank 
 and to obey orders. But to appreciate more clearly the 
 nature of the discipline, we must consider some broad facts 
 as to the constitution of the army. 
 
 17. Sociology treats of the social organism, and those con- 
 siderations already set down may serve to show what is 
 meant by this statement, and how far the word is used in 
 the same sense when we speak of the individual and the 
 social organism. It is enough to say here, that it implies at 
 least that some important rules are equally applicable in both 
 cases. It is as true that man is dependent upon his fellows 
 as that a limb is dependent upon tlie body. It would be 
 as absurd to ask what would be the properties of a man
 
 SOCIETY AND MAN. ill 
 
 who was not a product of the race, as to ask what would be 
 the properties of a leg not belonghig to an animal ; or to 
 ask what would be the best type of man without considering 
 his place in society, as to ask what would be the best kind 
 of leg without asking whether it belonged to a hare or a 
 tortoise. And in the next place, it is true that the proper- 
 ties of a society cannot be deduced from the independent 
 properties of its members in the same sense as it is true that 
 the properties of any living body cannot be deduced from 
 the mechanical and chemical properties of the elements of 
 which it is composed. Destroy the life in either case, and 
 the remaining properties of the dead materials do not enable 
 us to assign their properties when forming an associated 
 whole. We cannot infer the properties of a society by 
 supposing it to be an aggregate of beings independent of 
 society, because such beings are mere nonentities. The 
 residue, when you had abstracted from it all social qualities, 
 would in any case be merely the material elements, the 
 hydrogen, carbon, and so forth, to which the body may be 
 reduced; and from these you could infer neither the indi- 
 vidual nor the society. But further, any given society has 
 properties of its own which cannot be deduced from those 
 properties of the individual which are common to men in 
 all social states, for those properties may, as we have seen, 
 remain constant when the social organisation varies. The 
 inference from the part to the whole would be as impossible 
 as the inference to the properties of any given animal from 
 the tissues which it possesses in common with other animals. 
 The properties depend upon the way in which they are 
 combined, and we therefore require other data to determine 
 the results of the combination. It is, therefore, necessary 
 to speak of society as an organism or an organic growth 
 which has in some sense a life of its own. And this, it is 
 to be repeated, implies no mystical or non-natural sense. 
 Society is not an organism with a single centre of conscious- 
 ness. It is not something which has any existence apart 
 from the existence of the individual members. But the
 
 112 THE SCIENCE OF ETHICS. 
 
 name marks the essential fact, that although at any time 
 the properties of the constituted whole are the product of the 
 constituting units, those vniits have gained their proper- 
 ties in virtue of belonging to this whole. The society as a 
 whole acquires new characteristics at different stages of 
 growth which are only explicable through its history ; and 
 therefore, though we may properly speak of any particular 
 social phenomena as resulting solely from the character of 
 the individual agents, we must also tacitly assume that their 
 character is to be dependent upon their relations to a society 
 at the given stage of growth. We may no doubt try to 
 explain this fact by assuming that latent properties have 
 existed in the individual at earlier stages, as we may suppose 
 that the material elements of a body have latent properties 
 which only reveal themselves through the vital union. But 
 since we can only know them as they are manifested, we 
 can only give an intelligible account of the society by 
 regarding them as properties of the social organism. The 
 phrase "latent property" is only intelligible as marking the 
 fact that a man (or a baby), if transplanted from one society 
 to another, would acquire the corresponding characteris- 
 tics, which is, indeed, assumed in our statement. So that, 
 whatever the ultimate facts, we must be prepared to find 
 qualities developed through the social union which are not 
 immediately deducible from the properties of the individual 
 without reference to that union. The necessity of this as- 
 sumption and of the corresponding terminology will appear 
 as we proceed. 
 
 III. Social Organisation. 
 
 1 8. An organism implies organs. The society, like the 
 individual, has its organs of self-defence and nutrition, its 
 apparatus corresponding to the brain, the stomach, and so 
 forth, though it would be absurd to press the analogy too 
 far. There is this much resemblance, at any rate, that the 
 society develops associations each of which has its separate
 
 SOCIAL ORGANISATION. 113 
 
 function^ and implies the development of corresponding 
 associations with other functions. Political, ecclesiastical, 
 and industrial oro;ans become more distinct and more inter- 
 dependent as society advances. These organs, indeed, are 
 neither mutually exclusive nor generally conterminous in 
 respect of the individuals who compose them. Every man 
 may be, and generally is, a member of several organs. He 
 may belong to a church, a state, a commercial company, 
 and so forth. Sometimes membership of one may be in- 
 compatible with membership of another, especially of the 
 same kind ; but the various associations overlap in complex 
 ways, and 'do not imply, so to speak, a fissure down to 
 the foundations. Industrial associations may be formed 
 of the subjects of different states. The map of Europe 
 would diifcr according as we marked the ecclesiastical or the 
 political divisions. Englishmen belong to many churches, 
 and the Catholic Church includes subjects of many states. 
 And, of course, each organ, though it may be regarded 
 as a unit for some purposes, is for others a highly com- 
 plex structure ; as, for example, the state includes mili- 
 tary, judicial, and administrative organs, each of which is 
 again an organised structure. 
 
 19. Now, although this division into organs or associa- 
 tions does not correspond to a division of society into sepa- 
 rate groups, but to a division of the various activities of 
 a single group, it is equally true that each organ may be 
 regarded as havins: in some sense a life of its own. It 
 persists as a body persists in spite of, or rather in virtue 
 of, a continuous change in the constituent particles. The 
 associations graduate by indefinite degrees from the most 
 ephemeral and accidental up to the most permanent and 
 essential. Man, according to the old formula, is naturally 
 a social animal. Whenever two men meet there is the 
 germ of a possible alliance or antagonism. The most trivial 
 bond may serve. Two travellers who happen to meet in 
 crossing a ferry have already a certain community of 
 interest. They are liable to common dangers and have a 
 
 H
 
 114 THE SCIENCE OF ETHICS. 
 
 certain sympathy. If they are passed by another boat, they 
 will regard the other travellers as rivals, and be prepared to 
 take a pride in the success of the embryo society, even in 
 defiance of the reflection that their own merits have no 
 influence upon its success. Johnson's hatred of the "dockers," 
 when, for the moment, he adopted the local prejudices of 
 Plymouth, is a humorous illustration of a general principle. 
 The germs of association are everywhere present, and every 
 temporary cohesion supplies the necessary medium for their 
 rapid development. There are, indeed, two cases which 
 correspond to very different kinds of association. We may 
 regard the human organisation with no more interest than 
 we should feel for a material instrument. We may regard 
 a servant, as we regard the bell which summons him, as a 
 food-brin^ino; machine. A master may re2;ard his slaves 
 with the same feeling as he regards his horses or his ploughs. 
 They are simply means to an end, and the association would 
 vanish if the end were no longer desirable. In the opposite 
 case, we sometimes reverse the process, and come to entertain 
 for material instruments some of the affection which is 
 normally the product of sympathy; but the feeling springs 
 up naturally, and acquires importance when we have to do 
 with our fellows. In such cases the corporate feeling ceases 
 to have any accurate proportion to our desire for the osten- 
 sible end of the association. The man loves the school in 
 which he has been brought up with very little reference to 
 its merit as an educational institution. We come to love a 
 corporate body as though it were a real person. We speak 
 of a church or a state as " she," and when we are told that 
 a corporation has no soul, the remark strikes us as if it were 
 the revelation of a new and unsuspected truth. The osten- 
 sible end of an association is often the least part of its value 
 for us. We really love it because it supplies us with a 
 means of cultivating certain emotions and of enjoying the 
 society of our fellows ; and it would be an entirely inadequate 
 account of the whole statement if we regarded it as simply
 
 SOCIAL ORGANISATION. 115 
 
 the means of attaining that pleasure which has given the 
 pretext for its formation. 
 
 20. The corporate sentiment thus developed is more 
 complex and less capable of analysis in those associations 
 of which we have become members without any conscious 
 volition. When we have been brought up from our infancy 
 as members of a state or churchy or have been made members 
 of some society before the age of reflection^ our emotions 
 and activities become so thoroughly identified with those of 
 the body, that we are incapable of assigning any specific end 
 as supplying the dominant motive. The sentiment^ for 
 example^ of patriotism is one which defies analysis. The 
 state, we may say, discharges a certain social function. 
 According to some theorists^ it is useful simply as a means 
 of protecting its members against violence. Even upon 
 this supposition the corresponding sentiment would be 
 complex^ inasmuch as a desire for protection implies a 
 desire for gratifying any of the instincts which might expose 
 us to danger. But the bare desire for protection would be 
 a very inadequate explanation of the emotion roused by the 
 thought of a man's native country, which possesses a com- 
 plexity of a far higher order^ and seems to be an instinct 
 in which every part of his nature is more or less directly 
 concerned. It may possibly be true that if the need of 
 protection could be removed the instinct would decav, but 
 in any case it has an independent vitality, the conditions 
 and nature of which could only be unravelled by elaborate 
 psychological and historical investigation. At the opposite 
 end of the scale are generallv the industrial associations, 
 which generate very little corporate sentiment. The actual 
 co-operation which exists between different parts of the 
 industrial mechanism calls out very little distinctive senti- 
 ment towards the whole association. The end alvvavs 
 remains distinct. A bank, for example, is a highly im- 
 portant part of the machinery by which one set of people 
 are enabled to co-operate in the industrial occupations of 
 another set. But although the utility of the bank is the
 
 Ii6 THE SCIENCE OF ETHICS. 
 
 cause of its persistence, the banker and his customers on 
 both sides are not induced to carry on their operations by 
 any conscious desire for their neighbours' advantage, but 
 simply by a desire for their own comfort. Each man con- 
 fines his view of consequences to himself. His intention is 
 simply to keep or acquire wealth. By doing so he facilitates 
 the operations of his neighbours for the same purpose ; but 
 he does not, or need not, think of that result at all, any 
 more than the farmer in clearing the ground considers 
 the interests of the cattle which he is about to raise. Their 
 interests coincide with his own up to a certain point, at any 
 rate ; but he is not prompted by any sympathy with their 
 feelings. 
 
 21. But this difference does not necessarily or generally 
 affect the power of such persistence of the organ. An elabo- 
 rate industrial organisation is necessary to the life of a 
 civilised nation, and each member of the nation is interested 
 in maintaining it ; and thus it is maintained, although it may 
 be that no man feels any more enthusiasm about a bank or 
 a railway company, considered as a factitious person, than 
 he does about a steam-engine or a printing machine. Though 
 merely machines, they are necessary machines. In any 
 case, too, another conclusion is equally manifest. Every 
 association, that is, to whatever type it belongs, necessarily 
 implies the existence of certain more general instincts 
 dependent upon the whole social development. If any 
 social organ were or could be conditioned by a separate 
 instinct; if it supplied the means by which alone one specific 
 faculty of our nature could find its gratification, then a study 
 of that instinct or faculty would be the study of the corre- 
 sponding organ. Or rather, the study of one or the other 
 would be stated as inner and outer, as being objective and 
 subjective .theories of the same phenomena. But it is plain 
 that this does not correspond to the facts. The church, 
 for example, depends upon the religious instinct. It will 
 flourish or decay with the rise and fall of all that is implied 
 in that name. But the religious instinct is the name of a
 
 SOCIAL ORGANISATION, 117 
 
 highly complex set of scnthnents which have other mani- 
 festations than that of the ecclesiastical organisation, and 
 depend upon much wider conditions than those to which 
 it is subject in that capacity. The religion of a country de- 
 pends, amongst other things, upon the growth of specula- 
 tion, upon the philosophical and scientific conceptions which 
 have established themselves, and which react in countless 
 ways upon the beliefs which we call religious, and the prac- 
 tices to which those beliefs give rise. So, again, the state 
 depends upon the loyalty of its members; but the senti- 
 ment which we call loyalty is again involved in innumerable 
 ways with all the other modes of social development. It is 
 closely bound up with religious beliefs, as the decay of a 
 religion may involve a decay of the political order which 
 shares its sanctity. Every political change has an ecclesiastical 
 reaction; whilst in the other direction, again, the development 
 or decay of political institutions is closely connected with 
 the changes in the industrial organisation. The existence of 
 complex industrial mechanism implies necessarily the exist- 
 ence of a mutual confidence. It could not exist unless men 
 were ready to trust their fortunes to their neighbours, and 
 to rely in various ways upon their co-operation. This 
 implies the existence of a political order in which peace will 
 be preserved and contracts enforced. And further, there is 
 a close connection between the industrial state and the 
 political and religious condition of a country, as the state of 
 prosperity or misery of the mass of the population has a 
 direct and vitally important bearing upon their relations to 
 their rulers and teachers. Briefly, therefore, we may say 
 that the existence of any specific organ implies the existence 
 of an organism provided with other organs discharging cor- 
 relative functions; and therefore it implies the existence, 
 not onlv of a certain instinct to which it owes its own 
 vitality, but the existence of more general instincts to which 
 it is related as a special manifestation to a general sentiment. 
 23. Let us recapitulate these conclusions for a moment, 
 with a view to their bearing on our problem. We have
 
 ii8 THE SCIENCE OF ETHICS, 
 
 seen that the various properties characteristic of a given 
 social state may be regarded as corresponding to three suc- 
 cessive degrees of generaHty. For, in the first place, we 
 have those properties which belong to a society in so far as 
 it consists of men, that is, of men organised as men are now 
 organised, or, since we have assumed that this may be 
 regarded as a constant element, of the properties which 
 belong to all societies since a period antecedent to the growth 
 of an explicit moral system. In the second place, we may 
 consider the properties of a society at any given stage — and 
 of course I shall speak generally of the civilised man of the 
 present day — in so far as they depend upon these primary 
 instincts^ when modified and converted into a virtually new set 
 of instincts by social development and inheritance. These 
 instincts must be common to all societies at that stage, and 
 to the members of such societies whatever their special 
 relations to the society. And, thirdly, we have yet another 
 set of properties which belong to the society as organised, 
 and which are still modifications of the more general instincts, 
 but which correspond to the particular organs into which 
 the society is distributed. Thus, to take a particular example, 
 we may consider a man in so far as he is born with certain 
 sensibilities which render him capable of sympathy with 
 his neighbours. These sensibilities, by the hypothesis^ may 
 be found in the rude as well as in the more civilised state; 
 but in the higher state they will give rise to what may be 
 called a new set of faculties and instincts in virtue of the 
 process already described: the savage chief is transformed into 
 the civilised statesman. But, in the next place, the instincts 
 so developed will qualify him for being a member of various 
 forms of association, political, ecclesiastical, and so forth; 
 and, as acted upon by the special circumstances in which he 
 is placed, will again determine his fitness for one of these 
 functions rather than another, and further for some parti- 
 cular type of ecclesiastical, political, or industrial organisation. 
 And in each case, it nuist be observed, the more special 
 form of instinct must be regarded as conditioned by the
 
 SOCIAL ORGANISATION. 119 
 
 more general; not as if they were separate forces, one of 
 which must be conceived as controlHng the other, but simplv 
 as implying that the particular is an embodiment of the 
 general under certain specific circumstances, and that the 
 general rule must therefore be stated independently of the 
 particular case, as including, not as controlling it; and 
 further, that the conditions upon which, in fact, the exist- 
 ence of the general instinct depends must always be more 
 general than those which are given in the particular mani- 
 festation. 
 
 23. The grounds for this distinction have perhaps been 
 sufficiently indicated; the reason for insisting upon them 
 may appear more clearly by introducing another consider- 
 ation. It is impossible, as has been sufficiently said, to 
 determine the properties of any given society directly from 
 the innate properties of the individual members. We 
 must regard its properties, and thus the special characteristics 
 of its various organs, as determined by those properties which 
 are developed through the social union. But, again, we 
 cannot determine these organs directly from the social 
 properties of the individual; for, in fact, it is plain that 
 the organs are differently constituted according to the 
 special environment. In other words, the existence of a 
 particular political order implies necessarily the existence of 
 a certain stage of social development; but the inference 
 cannot be inverted. Each social stage is compatible with 
 many forms of political organisation according to the 
 circumstances in which it is placed. Under one set of 
 circumstances there is a greater, and under others a less 
 degree of centralisation; in one country the democratic, 
 and in another the aristocratic, element may be more 
 prominent, and this without any important difference in the 
 intimate or underlying social constitution. The various forms 
 of political organisation possible at a given stage are related 
 as varieties to a species, which may differ slightly, though 
 by altering the circumstances each might be transformed 
 into the other. The deeper and more permanent charac-
 
 I20 THE SCIENCE OF ETHICS. 
 
 teristics are manifested under certain relations in the political, 
 under others in the industrial organs, and so forth; and, 
 again, they may be compatible with varying conformations 
 in each particular manifestation. 
 
 IV. Social Tissue. 
 
 24. In order to mark this distinction, I will venture to 
 speak — applying an obvious analogy — of social "tissue." 
 The tissue is built up of men, as the tissue of physiology is 
 said to be built up of cells. Every society is composed of 
 such tissue; and the social tissue can no more exist apart 
 from such associations than the physiological tissue exist 
 apart from the organs of living animals. The distinction 
 does not correspond to things separable as concrete pheno- 
 mena, nor can it be compared to the distinction between 
 a coat and the cloth of which it is made ; for unorganised 
 social tissue does not exist, and the tissue develops new 
 properties according to the mode in which it is arranged. 
 Thus, if you will, the distinction may be regarded as 
 merely a logical device ; and yet, without taking into account 
 in some form or other the facts which it is intended to 
 describe, it seems impossible to give an adequate account of 
 the process which we have to consider. 
 
 25. The process is the social evolution. The typical 
 organism is by our assumption that organism which is best 
 fitted for all the conditions of life, or in other words, which 
 has the strongest vitality. Now the difficulty which meets 
 us in attempting to extend to human society the principle 
 which may be accepted as regulating the evolution of infra- 
 human species is the difficulty of determining the units. 
 The theory of evolution cannot be clearly apprehended or 
 applied to this case until this point is settled ; for every 
 such theory supposes a double set of processes. Every change 
 involves on the one hand a readjustment of the organic 
 equilibrium within the organism and a readjustment of its 
 relations to the internal world. The two processes are in
 
 SOCIAL TISSUE. 121 
 
 constant correspondence, and each Is regulated by the other. 
 But we should manifestly fall into hopeless confusion if we 
 did not know what was the unit of operation. Since the 
 process is one in which certain changes are mutually 
 implied, so that to suppose one to take place without 
 another is to suppose impossibilities, it is obviously essential 
 to determine as far as we can the limits of the organic 
 solidarity. We might otherwise fall into the absurdity of con- 
 sidering the evolution of legs without taking into account 
 the correlative changes of the rest of the organism. 
 
 26. Now the difficulty scarcely reveals itself in the evolu- 
 tionist theories as applied to the lower species. We assume 
 an organic change to occur — no matter how — in certain 
 individuals of a species, and that change to be inherited by 
 their descendants ; and thus two competing varieties to 
 arise, one of which may be supplanted by the other, or each 
 of which may supplant the other in a certain part of the 
 common domain. Some such process is clearly occurring 
 in the case of human variations. Everywhere we see a 
 competition between different races, and the more savage 
 tribes vanishing under the approach of the more civilised. 
 Certain races seem to possess enormous expansive powers, 
 whilst others remain limited within fixed regions or are 
 slowly passing out of existence. So far as human develop- 
 ment supposes an organic change in the individual, we may 
 suppose that this process is actually going on, and that, for 
 example, the white man may be slowly pushing savage races 
 out of existence. I do not ask whether this is the fact, 
 because for my purpose it is irrelevant. We are considering 
 the changes which take place without such organic develop- 
 ments, not as denying the existence of organic developments, 
 but simply because they are so slow and their influence so 
 gradual that they do not come within our sphere. They 
 belong, as astronomers say, to the secular, not to the periodic 
 changes. Confining ourselves, therefore, to the changes 
 which are, in my phrase, products of the '' social factor," and 
 which assume the constancy of the individual organism, we
 
 122 THE SCIENCE OF ETHICS. 
 
 have to ask what is the unit ? And here the theory of 
 variation just stated seems to require some modification. 
 
 27. Suppose, in fact, that an individual acquires a new 
 instinct or faculty in the sense already explained. He 
 makes, let us suppose, a scientific discovery which gives a 
 fresh command over the natural forces. Or we may suppose 
 that a change is brought about in the social machinery 
 which substitutes friendly co-operation for a state of anta- 
 gonism, and therefore of wasted energy. What will be 
 the effect of such a change upon the society in which it 
 occurs ? Obviously it will have an effect strictly analogous 
 in one respect to that of the organic change. Improved 
 artillery, like improved teeth, will enable the group to which 
 it belongs to extirpate or subdue its competitors. But in 
 another respect there is an obvious difference. For the 
 improved teeth belong only to the individuals in whom 
 they appear and to the descendants to whom they are 
 transmitted by inheritance ; bvit the improved artillery 
 may be adopted by a group of individuals who form a 
 continuous society with the original inventor. The inven- 
 tion by one is thus in certain respects an invention by all, 
 though the laws according to which it spreads will of course 
 be highly complex. It may be confined to a particular class 
 or nation, or it may spread through the whole race so far as 
 it has reached the necessary stage of development. To find 
 an analogy in the case of the individual, we must imagine 
 some molecular change to occur in one part of the organism 
 and to spread by some kind of fermentation or contagion 
 throughout the whole frame. And further, it may briefly 
 be noted that any such change, like the organic change, 
 involves a whole series of correlative changes, as the inven- 
 tion of artillery had a profound efi'ect upon the social and 
 political organisation of Europe, or as the development of 
 more effective teeth more or less alters the whole constitution 
 of the animal race in which it occurs. And this suggests that, 
 for certain purposes at least, the whole race, or the whole 
 race which has arrived at a certain stage, must be regarded
 
 SOCIAL TISSUE. 123 
 
 as a single organism, or rather a continuous organic growth, 
 and that any modification arising in one part is propagated 
 throughout the whole system. 
 
 . 28. It is indeed clear that this process does not exclude 
 the action of the " struggle for existence." An invention, 
 that is to say, is propagated, in part at least, because the 
 possession gives an advantage to the possessors. When one 
 people has big guns or effective steam-engines, another 
 people makes them in order to hold its own in the com- 
 mercial and political competition. But there is the impor- 
 tant difference that the other race ca?i make them. If some 
 animals acquire better teeth, their rivals cannot at once 
 improve their teeth in order to meet the new difficulty; 
 but men in the same social state can adopt the same inven- 
 tions. And, moreover, the process takes place in great part 
 by a direct method. A new discovery spreads through the 
 social tissue as a fermentation spreads through a continuous 
 fluid. It is always regulated by the struggle for existence. 
 Beliefs which give greater power to their holders have so 
 far a greater chance of spreading as pernicious beliefs would 
 disappear by facilitating the disappearance of their holders. 
 This, however, expresses what we may call a governing or 
 regulative condition, and does not give the immediate law 
 of diffusion. A theory spreads from one brain to another 
 in so far as one man is able to convince another, which is a 
 direct process, whatever its ultimate nature, and has its 
 own laws underlying the general condition which determines 
 the ultimate survival of different systems of opinion. 
 
 29. This, at any rate, is enough to show that it is of vital 
 importance to understand what are the conditions of this 
 mutual accessibility. In so far as one part of the race is con- 
 tinuous with another, and can directly receive any modifica- 
 tions which may arise elsewhere, the race must be considered 
 as forming for some purposes a unit, though for other 
 
 purposes it may still be a multitude of mutually competing 
 individuals. The problem would be simple if we were 
 
 entitled to regard the race as broken up into independent
 
 124 THE SCIENCE OF ETHICS. 
 
 groups — a case partly noticed in the case of some savage 
 tribes. In this case nations would be only related as hostile 
 units, the existence of each being maintained by a constant 
 struo:o;le aerainst its nei2:hbours, and new inventions or beliefs 
 being incapable of spreading beyond its borders. A modifi- 
 cation due to the social factor would be for our purposes 
 in the same position as an organic modification. The dis- 
 covery of a new weapon or the growth of stronger teeth 
 w'ould equally give an advantage to the group in which it 
 occurred (assuming it to be diffused through that group)^ and 
 we might regard the group as a unit in our reasoning. Those 
 groups would survive in either case which were the most 
 powerful in the struggle for existence, or in relation to all 
 external circumstances, including its hostile neighbours. Now 
 it is impossible to regard political states as forming in this 
 sense ultimate social units. It is true that the struggle for 
 existence of the race which corresponds to a general regu- 
 lative condition may occasionally lead to an internecine 
 struggle between nations. One survives by extirpating its 
 neighbour. But this is no longer the case in any moder- 
 ately civilised state of the world. War is there a com- 
 paratively subordinate phenomenon. It is part of a larger 
 series of events. War decides how races are to be grouped, 
 not which race is to survive; or sometimes it merely de- 
 cides what are to be the commercial and other relations 
 between different groups. A conquest is the extinction of a 
 political organisation, and the commonest result is that the 
 qualities of the resulting group are determined as much by 
 those of the conquered as by those of the conquerors. The 
 race is not extirpated but incorporated. The struggle for 
 existence still necessarily implies the supplanting of the 
 weaker by the stronger, and is therefore not represented by 
 the international struggle, which must be taken rather as 
 the means by which certain relatively superficial organisa- 
 tions arc determined, though this organisation, again, will 
 have a bearing upon the general process of which it is a 
 part.
 
 SOCIAL TISSUE. 
 
 125 
 
 30. This, again, falls in with a remark already made. If, 
 in fact, we could take states for units, and regard their 
 struggles as the manifestation of the struggle for existence, 
 it would follow that the problem worked out by that struggle 
 would be identical with the problem of the best form of 
 state. I do not mean to say that even upon this hypothesis 
 the qualities of the group, considered under its political 
 aspect, would give an exhaustive statement of its character- 
 istics ; for, in any case, its relation to other groups would be 
 in every case only a part of its relations to the whole external 
 world, and its power of maintaining itself would also depend 
 upon its power of obtaining food and so forth. But the 
 military power would always be the essential criterion of its 
 efficiency, since its life would depend at every moment upon 
 its power of satisfying the condition of military efficiency. 
 Its capacity for holding its own in the struggle would be 
 the central faculty round which all others would cluster. 
 Butthis, it is plain, would give a totally inadequate result. For 
 the best form of state, like the best form of army, is relative 
 to the general stage of social aevelopment, as the best form 
 of leg is relative to the organism in which it is included; 
 and thus the external pressure, though it may always 
 supply one condition, supplies only one amongst many. To 
 state the case fully, we should require to know many other 
 properties of equal importance, and directly dependent upon 
 an entirely different set of conditions, such as the industrial 
 capacity of the state, its geographical position, and so forth. 
 A state, in fact, may be developed when the external pres- 
 sure is little or nothing. The English constitution has no 
 doubt been profoundly modified by the relation to foreign 
 states ; but if for a certain period all communication with 
 the outside world had been absolutely prohibited, the English 
 people would still have developed under different conditions 
 to some different result. The conditions which would have 
 determined their organisation are still of the highest impor- 
 tance, and require to be taken into account when con- 
 trolled or modified by the influence of external relations.
 
 126 THE SCIENCE OF ETHICS. 
 
 Hence — though it may be superfluous to insist upon a 
 tolerably obvious conclusion — we should be led into hopeless 
 confusion if we identified the so-called social organism with 
 states^ regarded them as the units in our calculation, and 
 considered that the special modifications were all directly 
 determined and moulded under the influence of the mutual 
 competition. 
 
 31. We may thus consider the race as forming what is 
 called a social organism, or, as T have preferred to sav, as 
 forming social tissue. The reason for preferring the latter 
 phrase is simply that it implies a different kind of unity. 
 To use the word '^oro;anism" is to sua;g-est that the whole 
 body is capable of combining its efforts in order to bring 
 about some common end; as we may say (with certain 
 reservations) that a whole nation may combine to carrv on 
 a war or a single society to build a house. We cannot in 
 this sense predicate unity of the so-called organism. It is con- 
 tinuous but has not this unity. Its limits are fixed not by its 
 internal constitution but by external circumstances. It there- 
 fore is not analogous to the higher organism which forms a 
 whole separated from all similar wholes, but to an organism 
 of the lower type, which consists of mutually connected 
 parts spreading independently in dependence upon external 
 conditions and capable of indefinite extension, not of united 
 growth. The unity which we attribute to it consists in this, 
 that every individual is dependent upon his neighbours, and 
 thus every modification arising in one part is capable of being 
 propagated directly in every other part. The so-called organs, 
 again, correspond to special combinations of parts of the 
 tissue, the configuration of which is determined by the 
 special circumstances in which it is placed, the physical 
 geography of its habitat, and so forth, but which do not so 
 break it up into distinct fragments as to destroy its continuity. 
 The organs which we call states correspond to the most 
 prominent and most deeply marked lines of distinction; but 
 even this demarcation is still relatively superficial. The rela- 
 tions between members of different states are bv no means
 
 SOCIAL TISSUE. 127 
 
 those of simple antagonism^ but also of direct sympathy 
 and cohesion. 
 
 32. If, now, we ask how the struggle for existence will 
 manifest itself, the answer follows from the considerations 
 already set forth. There is, in the first place, a constant 
 competition, more or less overt, between different parts of 
 the race, which may or may not coincide with the struggle 
 between different nations. If we limit ourselves for a 
 moment to a fixed area inhabited by a strictly homogeneous 
 race, implying, therefore, a particular "social tissue," we 
 have the case of what may be called Malthusian competition. 
 Given the faculties and character of the race, there is room 
 for a certain quantity of population. It resembles an elas- 
 tic body pressing steadily against fixed limits. So far as it 
 accumulates capital or acquires new capacities, there is room 
 for a larger population. If, on the other hand, we suppose 
 its sensibilities to increase or its standard of comfort to be 
 raised, or, in other words, that it refuses to accept existence 
 except upon higher terms, there will be room for a smaller 
 population. And perhaps I may add — though I cannot 
 here discuss the importance of the consideration — that an 
 inverse rule may sometimes hold good. The increase of 
 population in a given area may sometimes be an advantage, 
 in so far as it facilitates a more elaborate organisation. A 
 million people may be able to live better in a given space 
 than a thousand, because they will be able to establish a 
 better division of labour, and, so to speak, to tell off more 
 labourers to some essential purpose. Still, however intricate 
 the problem, the data are determinate. It is usually a question 
 of the maintenance of an equilibrium of a certain kind and 
 quantity of social tissue, that is, of men endowed with certain 
 faculties, amongst which this power of forming organised 
 associations is one of the most important, under fixed external 
 circumstances. 
 
 ^^. When we extend the hypothesis to represent the 
 actual race, when we suppose the possibility of extension 
 into different areas and the competition with different races,
 
 128 THE SCIENCE OF ETHICS. 
 
 the conditions increase in complexity. Considering only the 
 question of the competition of races, which is the problem 
 which concerns us, it may obviously take various forms. 
 Thus, if we suppose the competing races to represent 
 different stages of social development or to be composed of 
 different tissue, the competition may be one of extermina- 
 tion. The feebler race may vanish before the approach of 
 its stronger rivals, as savage tribes vanish from Australia 
 and America. Such extermination, it may be added, need 
 not imply direct slaughter, but simply the gradual asphyxia- 
 tion of a race by confinement within narrower limits, and 
 the loss of energy which seems to result when the power of 
 expansion is destroyed ; but it may be one of subordination, 
 as weaker races have become the slaves of their superiors, 
 or, in some cases, their dependants without being in the full 
 sense slaves ; or, finally, it may be one of assimilation, which 
 is manifested in the relations between races at nearly equal 
 stages of civilisation. For, as in this case, we may suppose 
 the innate faculties to be approximately identical, and pro- 
 bably the acquired faculties to be not very widely different, 
 the race which is put under a stress by the competition of 
 its neighbours will end by acquiring their capacities, and 
 the equilibrium will be restored instead of one being extir- 
 pated. The varieties of circumstance are, of course, in- 
 definite, and may modify the result in a thousand ways. 
 For us, however, it is enough to note one essential fact, 
 namely, that to give any intelligible account of the process, 
 we are forced to bear in mind the distinction already drawn 
 between the fundamental properties in virtue of which the 
 race belongs to a certain kind of social tissue, and those 
 relatively superficial properties which are implied in the 
 particular organisation. Wherever there is mutual pressure 
 lietwecn two parts of a race, we should first have to consider 
 how far they consisted of the same or of different social 
 tissue ; for upon this it would depend whether the struggle 
 should be one of extirpation or assimilation ; and, secondly, 
 if the issue of the strugsrle should in any particular case
 
 SOCIAL TISSUE. 129 
 
 depend upon various conditions, amongst which of course 
 must be reckoned the special organisation of the competing 
 societies. Thus a nation inferior in its intrinsic quahties 
 may be able to extirpate its superiors because it has great 
 advantages of numbers and of geographical position. Or, 
 again, the industrial organisation may determine whether a 
 particular group will be absorbed by its neighbours or 
 remain in an isolated position. Now, in any case, we 
 can oflly speak of the merits of any particular organ with a 
 tacit or explicit reference to the qualities of the constituent 
 tissue. The military power is due not simply to the fact 
 that a nation has numerous armies, but that it has the 
 qualities which enable it to organise numerous armies ; and 
 thus its military power is always relative to the more intrinsic 
 and general qualities. And, secondly, we may assume that, 
 although in many particular cases the more civilised may be 
 supplanted by the less civilised, the race of higher intrinsic 
 qualities by that of lower qualities, these accidental and 
 contino;ent advantao;cs will be eliminated on the averao;e, 
 and the general tendency will be to the predominance of 
 those races which have intrinsically the strongest tissue. 
 
 34. And now, putting aside this question of competi- 
 tion between races at a different stage of development, 
 we may observe that the same principles apply in the case 
 where we assume a perfect continuity or homogeneity of 
 tissue. We must still, that is, distinguish between the 
 tissue and the organ; for in every case we assume that 
 some evolution is and has been taking place. Manv races, 
 perhaps the numerical majority of all races, are indeed in a 
 stationary state ; but in any case where any morality exists 
 we must suppose that there has been an evolution of the kind 
 in question (that is, independent of organic changes in the 
 individual) in the past, as there may be in the future. Such 
 evolution, it is possible, may have its limits. There mav be 
 a period, though for many reasons it would seem to be inde- 
 finitely distant, at which everything has been made out of 
 the given materials which is possible^ and at which further 
 
 I
 
 I30 THE SCIENCE OF ETHICS. 
 
 progress is therefore possible only on the hypothesis of an 
 organic change. In any case, and whatever the stage of 
 evolution or the rapidity with which it is proceeding, the 
 changes which have occurred or are occurring are made 
 possible through the inheritance of instincts and faculties 
 from our ancestors, used for application to new purposes. 
 Intellectual growth is conceivable so long as we are able to 
 acquire the knowledge already formulated, and to extend it 
 over fresh provinces ; and it seems impossible to fix any 
 limits to this process^ especially when we take into account 
 the division of intellectual labour made possible by the social 
 factor. In every case, again, the race considered as a whole 
 has to maintain its equilibrium in the general system of 
 nature, and the particular organisation which is the neces- 
 sary condition of maintaining as well as of extending the 
 dominion of the race at large must always be considered 
 relatively to those properties which have already become 
 organic in the race. The efficiency of any given organ 
 whatever must be estimated relatively to the organic pro- 
 perties which are the product of a previous evolution, though 
 at any given period they may be stationary ; and to deter- 
 mine the laws of variation, which include the case in which 
 the equilibrium is simply preserved without being altered, 
 we must always distinguish between these intrinsic qualities 
 and their more special modes of application. This is a logical 
 necessity, in whatever sense it may correspond to phenomena 
 separable in fact. 
 
 ^^. There remains an important question upon which 
 somethins: must be said. For it may be asked, what are the 
 characteristics in virtue of which we should declare that two 
 races are or are not of identical social tissue ? The question 
 is one which does not admit of a precise answer any more 
 than the parallel question, what amounts to a specific dif- 
 ference as distinguished from mere individual differences or 
 differences between varieties ? It is a question of fact, where 
 the facts graduate by imperceptible degrees. We might 
 perhaps accept as a sufficient criterion the capacity of dif-
 
 THE FAMILY. 131 
 
 ferent races to blend with each other. The organisatiou 
 of individuals would be regarded as identical when one 
 individual could be transplanted into another race with 
 perfect facility. The same test should be applicable to the 
 organic social organism. The experiments which have 
 been tried upon a vast scale in America would afford ample 
 illustrations. When streams of population, all drawn from 
 ev^ery part of the European continent, are poured into a 
 common receptacle, they rapidly blend together, so that all 
 distinction rapidly disappears. The difference is maintained 
 for a time by such differences as those between modern 
 languages, which we may call accidental because they corre- 
 spond to no essential social difference. They can speedily 
 be removed, and each race readily accepts the political and 
 other modes of organisation which it finds in the existing 
 organism. On the other hand, the process is exceedingly 
 slow in the case of some races separated by more funda- 
 mental distinctions. The Chinese and the negro remain 
 side by side with the other populations instead of speedily 
 losing their separate identity. In such cases there is pre- 
 sumably a difference of tissue, although it must be observed 
 that so long as there is no organic difference sufficient to 
 make fusion impossible, the acquired instincts may gradually 
 be adopted by one stream of population, and thus they may 
 be assimilated to their neighbours, with the result either of 
 supplanting one kind of tissue by the other, or of forming 
 a third, differing in some respects from both of the parent 
 stocks. 
 
 V. The Famihj. 
 
 '^6. This, however, is a question into which I need not 
 enter further. But one remark has still to be made which 
 is of vital importance towards giving a clear statement of 
 the case. I have spoken of the state, the church, and in- 
 dustrial bodies as presenting different kinds of social organ?. 
 They are, in fact, the most conspicuous cases, and of the
 
 132 THE SCIENCE OF ETHICS. 
 
 highest importance in any theory of society. But there is 
 another form of association, namely, the family, which is 
 frequently mentioned as though it were another co-ordinate 
 group. This mode of speech, though it may be admissible 
 for some purposes, seems to me to lead to much confusion ; 
 and until we have recognised the essential differences, I do 
 not think that we can have a clear conception of the bearings 
 of our theorv. 
 
 37, The distinctions, indeed, between the family and 
 other forms of associations are too obvious to be neglected ; 
 for, in the first place, we have here to do with a simple and 
 primitive instinct which is presupposed in all associations, 
 instead of being in any sense a product of them. The 
 sexual and parental passions are not only present at every 
 stage of human society, but their germs at least must be 
 present in the lower animals. The sentiment of loyalty 
 to a state is clearly a deriv^ative sentiment ; it is the product 
 of many instincts or modes of feeling, each of which has its 
 own laws, independently of this special application. It is 
 not a separate instinct with a definite object. The family, 
 on the contrary, depends at once upon the most primitive 
 iiistincts of our nature, which are the direct products of our 
 organic constitution. The love of man and woman or of 
 mother and child constitutes a bond which requires and 
 admits of no further explanation by reference to other emo- 
 tions. It is, of course, true that other instincts, and indeed 
 every instinct of which we are capable, come to group them- 
 selves round this central instinct and strengthen the primi- 
 tive tie. But that tie is more or less the ground of every 
 other, an antecedent assumption in all human society, and 
 therefore not explicable as a product of other modes of 
 association. Again, not only the association but its form 
 is determined within narrow limits by the organic structure 
 of the individual. The state or the church is an organism 
 capable of indefinite growth ; it may consist of any number 
 of individuals, as its limits are not fixed by the nature of the 
 units, but l)y various contingent circumstances. A man is not
 
 THE FAMILY. 133 
 
 born a king or a subject^ or in any other social group, but 
 the parts arc indefinitely interchangeable — a statement which 
 is obviously not predicable of husbands and wives, mothers 
 and children. Hence, although the form of the family 
 group has varied in such important respects, it is relatively 
 constant. The same form of family association has con- 
 tinued through many ages, and extends through many 
 different races. And from one point of view this implies 
 the important distinction that whereas a given state of the 
 social tissue is compatible with a vast variety of forms of 
 government, with large or small associations, with despotic, 
 aristocratic, or democratic systems, a change in the family 
 associations implies a corresponding change of vast import- 
 ance in the very intimate structure of society, or in 
 other words, in the social tissue itself. The change, for 
 example, from a society in which polyandry or the forcible 
 capture of women was the rule to a state of monogamy and 
 comparative sexual equality marks one of the greatest con- 
 ceivable changes of social growth ; and thus, from a scien- 
 tific point of view at least, the family is not in any case the 
 product of the political arrangement, but rather one of the 
 primitive conditions which determine the nature of the 
 state. A state may, of course, make a marriage law, and 
 the society, acting through the political organ, may affect 
 the nature of the family association. But though the law 
 determines certain rules about the family tie which may be 
 of great importance, it cannot create it or modify it beyond 
 certain narrow limits. For, in the first place, the state being 
 itself the product of individuals associated in families, its 
 action will always be determined by the existing sentiment ; 
 and, secondly, the influence of any law must always be subor- 
 dinate. Chastity and fidelity are not to be made by any law. 
 No state can force men and women to marry, or really put 
 down licentious habits, even if it makes the attempt; and, 
 on the other hand, the marriage tie might be equally re- 
 spected in fact even if there were no law in regard to it. 
 The law, in fact, recognises one kind of association of the
 
 134 THE SCIENCE OF ETHICS. 
 
 sexes, and bestows certain privileges upon those who are so 
 associated, but it would be a hopeless inversion of consequent 
 and antecedent to suppose that it can really originate it. 
 It may be added, again, that we need not suppose any other 
 form of association to be essential. It is quite possible to 
 suppose men living together without any political associa- 
 tion; and according to some theorists this is not only possible, 
 but represents the ideal state of things. But some associa- 
 tion between the sexes, however temporary and casual it 
 might be, and some protection of infants by parents is 
 absolutely necessary to the continuance of the race beyond 
 a single generation — that is, some sort of family union must 
 always exist. It need hardly be added here, that, as a fact, 
 the family represents that kind of association which is 
 beyond all comparison the most vitally connected with the 
 happiness of the individual, and the condition of which 
 most immediately affects the gratification of all his strongest 
 instincts. For a great part of every one's life the family is 
 the whole world. 
 
 38. If, then, we look at society as a whole, we see that 
 the division into families does not properly represent a mode 
 of organisation co-ordinate with the other social organs. 
 It represents, on the contrary, the immediate and primitive 
 relation which holds men together. The family affections 
 are the bonds which hold individuals together, and the 
 primitive cohesions in virtue of which society becomes 
 possible, the molecular forces which form the separate cells 
 into a continuous tissue, the elementary property in virtue 
 of which society is woven together, to be afterwards formed 
 into different groups. The larger organs are relatively few, 
 and their conformation depends partly upon the contingent 
 circumstances in each case, and partly upon the mutual 
 competition which leads to the survival of some and the 
 blending together of others. They survive the individuals 
 of which they are composed, as every organic group lives 
 by the continuous elimination of some particles, and their 
 replacement by others. Such statements are obviously quite
 
 THE FAMILY. 135 
 
 inapplicable to families. They are multitudinous; and as 
 they represent the primitive cohesion of the constituent 
 particles, they represent a cross division running through all 
 the other groups into which societies may be formed, and 
 constantly disappear with the death of the member, though, 
 as it breaks up, new families are incessantly formed ; and the 
 social competition affects the institution, not through the 
 competition between rival families, but in so far as the for- 
 tunes of the whole group are constantly dependent upon its 
 structure, which is determined in a very great degree by the 
 nature of its family unions. 
 
 39. A brief recapitulation of the principal conclusions 
 will now suggest the conclusion which is relevant to our 
 inquiry. For I have tried to show, in the first place, that to 
 determine the "law^' or ''form" of any instinct developed 
 through the social factor, it is necessary to go beyond the 
 individual to the organism of which he is a member. A 
 man acts from patriotic instinct; we may explain the force 
 of this instinct in any given case partly by his innate quali- 
 ties and partly by his circumstances. The latter considera- 
 tion necessarily introduces the social factor ; for the instinct 
 has been elaborated by the evolution of the social organism, 
 and its force and nature are determined by conditions not 
 given when the innate character is given, and therefore not 
 deducible from the individual organisation. So long, there- 
 fore, as we regard that organisation by itself, we must have 
 an arbitrary datum, the " law " of which is only ascertain- 
 able by introducing the wider considerations. Secondly, we 
 have observed that we have to classify the various social 
 instincts by reference to the complex structure of society, 
 which implies a distribution into mutually dependent organs, 
 each of which has its own conditions of persistence, and dis- 
 charges a more or less definite function ; and further, that 
 any examination of such organs shows that their laws of 
 growth and vitality are always relative to the underlying 
 properties of the " social tissue," each representing a special 
 modification of that tissue in view of some particular func-
 
 136 THE SCIENCE OF ETHICS. 
 
 tion, and varying in its structure according to various con- 
 tingent circumstances. Again, whilst such organs may be 
 adapted to a specific end, and may therefore be regarded 
 simply as instruments for gratifying some independent in- 
 stinct, and their existence is proportioned to their efficiency 
 in securing the gratification, they always tend to become 
 something more ; they acquire a vitality independent of any 
 special end, and become organs discharging a complex func- 
 tion, and imply the existence of a correspondingly complex set 
 of instincts; whilst, finally, the social tissue is its own end, 
 or depends upon the whole system of instincts possessed by 
 man as a social and rational creature. And hence, thirdly, 
 since the social tissue represents the general material or all- 
 pervading substance from which the subordinate associations 
 are constructed, w^e must consider the conditions of its 
 vitality independently ; and therefore we see that it is the 
 primary unit upon which the process of evolution impinges. 
 The social evolution means the evolution of a strong social 
 tissue ; the best type is the type implied by the strongest 
 tissue ; and the correlation between painful and pernicious, 
 pleasurable and beneficial, is to be understood by interpret- 
 ing the pernicious and beneficial with reference to the tissue, 
 whilst painful and pleasurable refer to the instincts gener- 
 ated in the socialised being. It is the vigorous tissue which 
 prevails in the struggle, and fitness for forming such tissue is 
 therefore the criterion of a successful elaboration of the 
 type. This, it may be noted, applies equally to the qualities 
 in respect of which society is an aggregate, and those in 
 respect of which it is an organism ; those which cannot vary 
 without a corresponding variation of social structure, and 
 those which vary so as to imply a change in the average 
 faculties. For the process of evolution is simple, and each 
 set of properties is always developed with reference to the 
 other. This statement, however defective, will, I hope, have 
 cleared the ground sufficiently to enable us to consider more 
 fully what is implied as to the nature of those social in- 
 stincts which gave rise to the moral law.
 
 ^37 
 
 CHAPTER IV. 
 
 FORM OF THE MORAL LAW. 
 
 I. Law and Custom. 
 
 1. Society is an organic structure, dependent for its 
 existence upon the maintenance of certain relations between 
 its members, more complicated in proportion to the com- 
 plexity of the whole. /Its development implies, therefore, 
 the development of customs in the race and habits in the 
 individuals. There must be certain rules of conduct which 
 are observed by all in order that corresponding rules may be 
 observed by each. To trace the development of the society 
 is at the same time to trace the development of the custom, 
 and conversely to trace the custom is to trace the social 
 development. 
 
 2. I use these words "custom" and "habit" without 
 meaning to imply what perhaps is generally implied by 
 those words, that the modes of conduct designated are 
 variable or indefinite. We oppose a custom to a "law of 
 nature" because it is not unconditionally observed, and we 
 oppose it to the positive law of a state because its observance 
 is not enforced by the policeman. But the difference is 
 not fundamental. A habit may often be regarded as a law 
 of nature in the making. It passes into such a law by 
 imperceptible degrees. We do not say that living men have 
 a habit of breathino;, because we hold it to be somethino; 
 more than a habit; it is, as we hold, a law of nature that 
 living men should breathe. We hold, that is, that the group 
 of phenomena denoted by human life inseparably and un- 
 conditionally includes the phenomena of breathing. We 
 speak of habit only in cases where the continuance of tlie
 
 138 THE SCIENCE OF ETHICS. 
 
 conduct depends in some degree upon the control of the 
 will, and can be suspended in cases where it becomes painful 
 beyond a certain point. But the degree in which the co- 
 operation of the will is a condition may vary indefinitely 
 and become a vanishing quantity, as the habit becomes a 
 definitely organised mode of conduct. If we could extend 
 habit to cover any mode of conduct which can be brought 
 under any general formula and is practised under assignable 
 conditions, we might apply it both to conduct which must 
 continue as life continues and to that which implies in 
 addition to life a particular state of desire and aversion in ^ 
 the living being. The latter is essential not to all men, but 
 to all men under given conditions, which may, of course, be 
 rare, or may be all but universal. What is true of habits - 
 is of course also true of customs. I will only repeat that 
 customs may be essential to a society which do not corre- 
 spond to a habit essential to the individual. The existence 
 of a given stage of social development is dependent upon 
 certain customs which are not exemplified in every member 
 of the society. The most civilised country includes nume- 
 rous savages, who are in it but not of it, foreign matters con- 
 tained in the organism, and such that it would fall to pieces 
 were it not for their restraint by the more civilised mem- 
 bers of the community. 
 
 ^3. Customs, again, have a relation to positive laws which 
 has led to some perplexities. In primitive states of society 
 the distinction is imperceptible. A tribe or clan is bound 
 together by certain customs which are regarded by its 
 members as ultimate and indefeasible. They are observed ; 
 but it has not yet been asked how they acquired nor why 
 they should retain authority. In an early stage of reflec- 
 tion they are perhaps regarded as imposed by the authority 
 of the gods, which really supposes the question may be 
 asked, but implies that it cannot be answered. As they do 
 not change, or change only by imperceptible degrees, there is 
 no question as to the mode in which they can be changed, 
 or, in other words, of the authority which imposes them.^
 
 LAW AND CUSTOM. 139 
 
 Historical inquirers, for example, find the traces of such a 
 state in regard to the system of customs which regulate the 
 possession of land. The mode in \\ hich the various mem- 
 bers of a tribal community arc to cultivate its domains or to 
 distribute the fruits of cultivation is settled by a system of 
 rules which; in so far as they persist without reflection upon 
 their utility and the possibility of changing them, have 
 somethino- of the character of animal instincts. Thev are 
 analogous to the tacit agreement by which different herds 
 of gregarious animals may divide a district between them, 
 yriiere is a continuous progress from this social condition to 
 one in which the land laws form an elaborate code or an 
 intricate mass of regulations, the provisions of which are 
 expounded by judges and altered from time to time by the 
 legislature. Now, however great may be the difference 
 between the two stages of development, we have at every 
 stage a social organisation dependent upon custom. The 
 custom, indeed, has gained a new order of complexity. For, 
 in the first place, the law actually obeyed is not a simple set 
 of rules understood and spontaneously accepted by every 
 one concerned. It contains what we may call a potential 
 element. For there are a vast number of regulations which 
 are only brought into play upon rare contingencies and 
 which are known only to a few specialistsX^ It would be 
 straining words improperly to speak of a custom which is 
 absolutely unknown to ninety-nine in a hundred of the 
 persons concerned, and which perhaps does not determine 
 conduct once in a generation. ,>The custom which is essential 
 at all times is that which directly governs conduct from 
 day to day, and which is implied in the mutual confidence 
 of proprietors and respect for the known rights of property, 
 and, moreover, that of obedience to certain constituted 
 authorities when called upon to settle disputes./; I can 
 hardly be said to be in the habit of observing certain rules, 
 the very existence of which never enters my head, but I 
 may be in the habit of accepting the decisions of certain 
 persons who, on occasion, tell me what are those rules.
 
 I40 THE SCIENCE OF ETHICS. 
 
 This, it may be observed, is a general principle already 
 noticed. As knowledge becomes too elaborate for any 
 single head, we acquire a kind of potential knowledge. We 
 do not know the rule, but we know where it is to be found. 
 We are not guided simply by our instinct of locality, but by 
 our confidence in the Nautical Almanac. A judge has the 
 same use in the social organism as a general word in 
 language. 'We accept the general rule that we are to fulfil 
 contracts. In most cases this may be sufficient, but to 
 mark out the rule in a complex case we have to accept the 
 results obtained by persons specially qualified. Thus a 
 custom may regulate conduct in cases where it does not 
 imply a corresponding instinct in the minds of the agent, 
 because there is this elaboration of the potential element, 
 which he can call into play by appealing to the authorities4(^^ 
 
 4. The authority, however, must itself rest upon custom, 
 which, again, is far more elaborate than in the primitive 
 state. For the chief or elders of the primitive tribe we have 
 the whole complex organisation of a modern state. The 
 custom of obedience, again, carries with it much more than 
 is actually present to the mind of the average citizen. He 
 obeys the king, the judge, and the policeman, and has pro- 
 bably a very vague conception of the precise relations 
 between their various privileges and the relations of their 
 offices and functions. There must, however, be a certain cus- 
 tom of obedience to constituted authorities, without which the 
 whole state would be a rope of sand. And in this sense the 
 law is not something more than custom, but simplv a par- 
 ticular case of custom. A law, as jurists tell us, is the com- 
 mand of a sovereign enforced by a sanction ; and the essence 
 of law, therefore, depends upon the ultimate appeal to 
 coercion, or, in other words, upon the circumstance that, if 
 you do not obey the law, you may be made to obey it. A 
 custom, on the other hand, depends always upon voluntary 
 obedience, and exists only so long as people choose to 
 comply with it. Now, for purposes of jurisprudence, the 
 distinction may be important, but it is not ultimate from
 
 LAW AND CUSTOM. 141 
 
 the scientific point of view. It explains only one collateral 
 result from the general system. The lawyer takes for 
 granted the constitution of the state ; he is satisfied that a 
 rule is a law when certain legislative processes have taken 
 place, and he is content with the conclusion that ultimately 
 the judge can appeal to the hangman. He does not go 
 further, nor ask how the state holds together, nor in virtue 
 of what principle the judge can depend upon the hangman's 
 fulfilment of his duty. When we ask that question — as we 
 are bound to do for scientific purposes — we see at once that 
 this possibility of physical coercion cannot give an ultimate 
 answer. How is coercion possible? What will happen if 
 the hangman does not obey the judge ? We may go for our 
 answer to the nursery rhyme about the old lady who could 
 not drive her pig to market. When the butcher would not 
 kill the pig and the rope would not hang the butcher, she 
 had to appeal to the fire to burn the rope, and so forth. She 
 depended upon coercion at some point because she had to 
 deal with a pig, but to get the coercion she had to find some 
 agency set in motion without coercion. Now in the case of 
 the state, it not only may happen, but it is undoubtedly true 
 in all cases, that coercion must be at hand, so to speak, to 
 maintain order in case of necessity ; and by coercion I mean 
 the application of physical force, or the reduction of a man 
 to a mere thing, so that his condition is determined by forces 
 entirely independent of his own volitions. In that case the 
 theory of the state is the theory of the man whose actions 
 determine the action of his neighbours. The hangman and the 
 jailer are doubtless necessaries so long as men are what they 
 are, and savages form a part of a civilised community. But 
 it is equally true that other forces are essential to anything 
 like a civilised society. A temporary association may be 
 formed where one man treats another simply as if he were a 
 tool, or rules him by the threat of so treating him. And, 
 again, a race may be ruled from without by oppressors who 
 appeal only to motives of this class. But this fact does not 
 prove that physical force, or the dread of its application, is
 
 142 THE SCIENCE OF ETHICS. 
 
 in any special sense essential to political society. For every 
 society, beginning from the simplest germ of social union, 
 where the state is not yet differentiated from the family, 
 requires the action of all the social instincts. The more 
 elaborate the structure, the greater the number and force of 
 the instincts which must be called into play. All that is 
 implied in loyalty, patriotism, respect for order, mutual con- 
 fidence between man and man, is essential to the vitality of 
 a complex social organisation. A bond which rested solely 
 upon fear would give, not an organic compound, but a tem- 
 porary association, ready to collapse at every instant. Coer- 
 cion itself is only possible by virtue of the co-operation 
 which implies the existence of every other social motive. 
 We may say of any stone in an arch that it is the keystone, 
 if by abstracting it the arch would fall into ruin, and 
 coercion is in that sense a keystone in the social structure; 
 but so are all the other forces which are essential to the 
 structure so soon as it attains any permanence or magni- 
 tude. A power of flogging may be essential to the discipline 
 of an army, but an army held together solely by dread of 
 the whip would be comprised within a circle defined by the 
 smart of the lash. It is so little essential, indeed, that a state 
 of society is conceivable in which its actual application 
 should disappear altogether. Men might be willing to obey 
 their rulers simply from respect and affection ; judges might 
 be arbitrators whose decisions would always be accepted by 
 mutual consent. The power of applying coercion in case of 
 need must no doubt increase as the strength of the social 
 bond increases ; but that bond is also the stronger in propor- 
 tion as the need of applying it becomes less. 
 
 5. The whole social structure, then, must rest in the last 
 resort upon the existence of certain organic customs, which 
 cannot be explained from without. They depend for their 
 force and vitality upon the instincts of the individual as 
 modified by the social factor; and it would be a fallacy to 
 single out any one of these instincts as the essential one 
 when the co-operation of all is necessarily implied. From
 
 LAIV AND CUSTOM. 143 
 
 this point of view, therefore, it is necessary to distinguish 
 between the organic instincts of a state and the correspond- 
 ing customs on the one hand, and the derivative and secondary- 
 instincts which are the product of that state on the other 
 hand ; otherwise we fall into the absurdity, not rare in 
 political speculations, of implicitly assuming that a state 
 can somehow make itself. A legal sanction may of course 
 be added to any custom whatever, and thus it may seem 
 that a state can make its own constitution and define its own 
 organic laws. In reality, however, the power of making a 
 constitution presupposes a readiness to act together and 
 accept certain rules as binding, and thus again implies a 
 whole set of established customs, such as are necessary to the 
 constitution and authority of a representative body. Law- 
 yers are apt to speak as though the legislature were omnipo- 
 tent, as they do not require to go beyond its decisions. It 
 is, of course, omnipotent in the sense that it can make what- 
 ever laws it pleases, inasmuch as a law means any rale 
 which has been made by the legislature. But from the 
 scientific point of view, the power of the legislature is of 
 course strictly limited. It is limited, so to speak, both from 
 within and from without ; from within, because the legis- 
 lature is the product of a certain social condition, and 
 determined by whatever determines the society ; and from 
 without, because the power of imposing laws is dependent 
 upon the instinct of subordination, which is itself limited. If 
 a legislature decided that all blue-eyed babies should be 
 murdered, the preservation of blue-eyed babies would be 
 illegal ; but legislators must go mad before they could pass 
 such a law, and subjects be idiotic before they could submit 
 to it. 
 
 6. Considering, therefore, any society as a natural growth, 
 or, in other words, regarding it from the scientific point of 
 view, we see what is implied in a law. We assume that 
 certain organic instincts have been formed, corresponding, 
 in my language, to a given state of the social tissue; and 
 involving a certain body of customs essential to the life of the
 
 144 T^I^E SCIENCE OF ETHICS. 
 
 society, and giving rise to a special organisation according 
 to the various internal circumstances. We may, then, trace 
 the manifestations of the social properties in two ways — 
 either as implying a certain social structure, or as implying 
 a certain type of character in the members of the society. 
 There is, on the one hand, a political organisation which 
 acts in certain definable ways, and, for example, has an 
 apparatus for hanging convicted murderers. The "law" 
 may be regarded either as a statement of the relations exist- 
 ing between the various parts of the political organism, or 
 may be viewed as a command and a threat, implying a 
 notice to murderers that they will be hanged if caught. The 
 same facts, regarded from the other side, necessarily imply 
 the existence of an internal law. The individual must 
 acquire certain instincts in virtue of which he respects the 
 authorities and dislikes murderers. He must acquire them, 
 that is, in order to be an efficient part of the social organisa- 
 tion ; and the law may be expressed as threatening him with 
 whatever consequences — other than the legal consequences 
 — result from imperfect harmony with his social medium. 
 The society exists by virtue of the vitality of these instincts. 
 Both kinds of law are expressions of the same general fact ; 
 the essence of the former being that the individual is subject 
 to a pressure tending to enforce a correspondence between 
 his actions or feelings and those of his neighbours. Some 
 such process must take place in every association, from a 
 state to a gang of thieves, whatever the method by which 
 conformity is produced; and wherever it is produced we may 
 speak of a social law. It may not be possible to consider 
 the two modes of action separately. Every law of conduct 
 more or less affects the character of the persons subject to it, 
 so long as it is enforced ; and necessarily every variation in 
 the character more or less affects the sentiments from which 
 the external law derives its force. The correspondence, 
 however, is not so intimate^that one mode of statement can 
 always be rendered into the other. For, as I have said, laws, 
 and indeed elaborate codes of law, arc developed which
 
 LA]V AND CUSTOM. 145 
 
 scarcely aflect the general character of the underlying cus- 
 toms, and which represent latent modifications of the social 
 structure not implying any sensible modification of the 
 instinct of order. And in the same way the instincts may 
 vary widely without producing any normal change in the 
 external order, though they may in some degree affect the 
 mode in which it works. The change may be too fine to be 
 expressed in terms of external relation. 
 
 7. If, then, the essence of any law is in the mutual pres- 
 sure of the different parts of the social structure, by what- 
 ever means it is carried out, and to whatever process it owes 
 its vitality, we have still to consider how the various codes of 
 law must be classified from our point of view. To every kind 
 of association, even the most ephemeral, there corresponds, as 
 I have said, some kind of custom, and therefore of law. And 
 what has been said in the last chapter will give a sufficient clue 
 to the right mode of regarding these various associations and 
 the external processes. If, in fact, we take any association with 
 a given end or function, its structure and the laws of con- 
 duct and character imposed upon its members will be deter- 
 mined by reference to that end and to the society of which 
 it forms a part. An army, for example, may be called the 
 fighting organ of a given nation. It resembles a machine 
 constructed from given materials for a given end. From 
 such data we could determine its structure, the discipline 
 necessary to its existence, and thus the various regulations 
 by which the external relations of the whole are defined, 
 and the corresponding instincts in the units. The statement 
 showing how men could be compounded into a fighting 
 machine would show also how they could be most efficiently 
 combined for a particular kind of fighting or from a par- 
 ticular kind of social organisation ; and the question of how 
 far any particular army fulfilled these conditions would be 
 determined by the specific conditions of time and place. 
 When we pass from the organ to the "tissue," the problem 
 changes. We still have an organic structure with certain 
 rules of conduct and corresponding instincts, but we have no 
 
 K
 
 146 THE SCIENCE OF ETHICS. 
 
 longer a definite end nor a fixed material. The material, 
 that is^ is to he regarded as developing and determining the 
 development of the subsidiary organs. The organ is intel- 
 ligible by its relation to the organism, and the end or the 
 function is assigned by that relation ; but the organism 
 itself is at once means and end; every part depends upon 
 every other part, and the end is intelligible only as the sum 
 of all the correlated instincts. The statement, therefore, 
 beccmes different. We now have to remember that the 
 organism develops without any change (or any correspond- 
 ing change) in the constituent units. It develops pro- 
 perties, therefore, which are not essential to the individual, 
 for he can exist in a ruder state without them ; and which, 
 therefore, imply the growth of a social law — that is, of quali- 
 ties developed in him through the social pressure. And 
 further, we see that some of these properties are essential to 
 the society. Its growth is a process of developing such pro- 
 perties ; and, as we have seen that the most efficient society 
 is that which normally survives, we may inversely infer from 
 the survival of a society that it has developed the properties 
 upon which its efficiency depends. For the end, as before 
 understood, we have now to consider the society as capable 
 of maintaining itself in the general equilibrium, whether by 
 competition with weaker societies, or as supporting itself by 
 its direct action upon the external world, and as capable of 
 doing so by virtue of the social properties which have been 
 developed. We may regard them, therefore, either as the 
 conditions of the social vitality, or as imposing a certain law 
 upon its individual members. In order that the society may 
 exist or develop it must have certain qualities and customs, 
 and must in some way impress the corresponding instincts 
 and habits upon its members. Hence we have to find the 
 qualities which are essential to the society at a given stage of 
 development, though not essential to the individual, and we 
 may then state them either as conditions of the vitality of 
 the social tissue, or as constituting the law imposed upon the 
 individual as a member of society, that is, as a constituent
 
 THE MORAL LAW. 147 
 
 part of that tissue. The actual law, again, may not represent 
 the greatest degree of efficiency possible for a certain stage of 
 social growth ; for, as we have observed, the qualities may 
 vary within certain limits consistently with the persistence 
 of the society, but they must be an approximate statement 
 of the essential conditions. 
 
 II. The Moral Law. 
 
 8. Hence, without further elaboration, we may approach 
 to a definition of the moral law. This much at least is 
 obvious: the morality of a society or of an individual im- 
 plies at least a certain modification of the most important 
 relations and instincts. We may say of any suggested 
 regulation that it is too trifling to have any moral signifi- 
 cance; we cannot possibly say that it is too important. 
 Morality, it may be, is not interested in a mere question of 
 manners and fashions ; but rules which affect the very 
 existence of a society or a human being do not, by that 
 circumstance, lie beyond the sphere of morality. Another 
 principle closely connected with this is equally undeniable. 
 "^ The moral law is understood as applying to all men, in so 
 far as they have reached a certain stage of development, not 
 in so far as they belong to any particular class of society. 
 The same moral law is applicable to all adult men and 
 women, whether they are rich or poor, in one or other 
 profession, or, briefly, belonging to any category compa- 
 tible with a full development of their faculties. Of course, 
 each man has special duties corresponding to his particular 
 position in life, and in some positions there is a greater 
 demand for certain kinds of morality than in others. But 
 this means simply that the same general principle is appli- 
 cable in an indefinite variety of relations. The moral law is 
 alwavs capable of being stated in the form, "All men must 
 do so and so," not all lawyers, or soldiers, or sailors must 
 do. You come within its operation in so far as you have 
 the fundamental qualities common to all members of the
 
 1 48 THE SCIENCE OF ETHICS. 
 
 society, not in so far as you have this or that particular 
 contingent quahty. This, then, is to say that morality 
 defines some of the most important qualities of the social 
 tissue. It does not apply to those qualities which are 
 essential to the life of the individual, for immoral people 
 clearly exist, and the law, in this sense, implies the pos- 
 sibilitv of disobedience. On the other hand, it does not 
 apply to the more special and superficial qualities which fit a 
 man for this or that position without affecting his fitness to 
 be a member of society in some position; and therefore 
 we may assume, from our previous statement, that the moral 
 law is under one aspect a statement of the conditions, or 
 of part of the conditions, essential to the vitality of the 
 social tissue. It may be more than this in various ways ; 
 but it must be this, whatever else it is. The process by 
 which society has been developed implies that the most 
 important characteristics developed in the individual by the 
 social pressure correspond to the conditions of existence of 
 the society. The moral law defines some of the most im- 
 portant characteristics so developed, and is, therefore, a 
 statement of part of the qualities in virtue of which the 
 society is possible. It is not an exhaustive statement, for 
 other qualities may be essential ; nor an absolutely accu- 
 rate statement, for societies exist in which the morality 
 varies within wide limits. But so far as it goes it 
 must be an approximate statement of part of the condi- 
 tions.y 
 
 9. In saying this, I do not mean either to assert or deny 
 that this gives the form in which the moral law presents 
 itself to the members of the society in which it inheres. I am 
 considering the cause, not the reason, of our moral senti- 
 ments. Our moral judgment must condenm instincts and 
 modes of conduct which are pernicious to the social vitality, 
 and must approve the opposite ; but it does not necessarily 
 follow that it must condemn or approve them because they 
 are perceived to be pernicious or beneficial. The question 
 indeed remains, how it comes to pass that we condemn what
 
 THE MORAL LAW AS NATURAL. 149 
 
 is pernicious if \vc do not think it to be pernicious ; and 
 this cannot be fully answered at the present stage of the 
 argument. Here I will only observe, that there is no 
 absurdity in supposing that the cause of our likes and dis- 
 likes may in some sense be the fact that they are useful to 
 us, although we may not be conscious of their utility. 7'his, 
 indeed, must be to some extent the case with all beings below 
 the reasoning stage. 
 
 III. The Moral Law as Natural. 
 
 10. The same principle accounts for the qualities most 
 obviously connoted by the term "moral." The moral law is 
 often distinguished from other groups of law on the ground 
 that it is divine, not human, natural, not artificial, or that 
 it grows instead of being made. The distinction is not an 
 ultimate one. Art to the scientific observer is, as Shakespeare 
 says, a part of nature; everything springs, mediately or im- 
 mediately, from the divine power as here understood, and 
 all human development is a kind of growth. The distinction 
 is only relevant at a lower stage of analysis. We mean bv 
 it this much at least, that whereas the law of the land is 
 determined by the will of the legislature, the moral law is as 
 independent of the legislature as the movements of the 
 planets. King, lords, and commons, bv going through certain 
 forms, may determine whether theft or lying shall be criminal; 
 they cannot in any degree decide whether it shall or shall not 
 be wicked. This seems to be in one sense equally evident 
 in all conceivable moral systems. If, with one set of 
 thinkers, you resolve morality into reason, a law to alter 
 morality would be as absurd as a law to repeal a proposition 
 in Euclid ; if you adopt the utilitarian theory, such a law 
 would be as absurd as a law to alter the pleasure derivable from 
 the consumption of stimulants. Upon the doctrine here advo- 
 cated, it would be as absurd, let us say, as a law to make in- 
 toxication healthv. The action of anv set of people can no 
 more change the nature of facts than of logical necessities.
 
 ISO THE SCIENCE OF ETHICS. 
 
 11. Bat this does not entirely meet the case; for I am 
 here deahng with moraHty as it actually exists, not with 
 morality as it ouo-ht to be. So far as morality either is or 
 necessarily implies a statement of certain facts, it is of course 
 true that morality cannot be made; but actual morality 
 corresponds to men's theories about facts, and varies in 
 proportion as they are fallible. It may therefore deviate 
 from that which would be the code if they were incapable of 
 error, and we may ask how it is possible to define the possible 
 amplitude of their oscillations. . May not the code of rules 
 by which our moral judgments are guided vary as widely as 
 any other code of rules? May it not be swayed by prejudices 
 or altered by respect for some constituted authority ? Though 
 a legislative enactment could not make murder rio-ht, mio-ht 
 it not as a fact determine the sentiment about murder ? This 
 is necessarily a question of fact to be settled by historical 
 inquiry, but the principles laid down may suggest some limit 
 to the possible oscillations. 
 
 12. It is plain, in fact, that though morality varies, it must 
 vary within incomparably narrower limits than other systems 
 of law, because its variation is determined by far more general 
 conditions. It is the variation of the most intimate struc- 
 ture and the deepest instincts, not of the superficial senti- 
 ments or of the special modifications of society. In the 
 earliest stages of growth, when certain rigid customs represent 
 the germs both of moral and other codes, the custom develops 
 in all cases by a slow growth rather than by constant modi- 
 fication; and even in the most civilised periods a similar 
 process may take place in regard to certain rules as modified 
 imperceptibly by judicial interpretation. But as the difference 
 becomes more palpable, the moral law alone retains the 
 characteristics of divine, indefeasible, and so forth. The 
 diflerence then appears between the organic laws of the 
 tissue and the special laws of any particular organisation. 
 One class of laws maintains itself by the direct action of the 
 organic instincts; others by the application of these instincts 
 to special circumstances, or by respect for the authority which
 
 THE MORAL LAW AS NATURAL. 151 
 
 developed by such application. In a given case the two 
 kinds of motive may be inextricably blended. I may obey a 
 given law either because of the authority which enforces it or 
 on its own account. I may keep a promise because I think 
 it right, or because I am afraid of the penalties imposed upon 
 a breach of contract, and in the latter case I keep it for the 
 same reasons which would induce me to wear a prescribed 
 costume. But the instincts which induce me to act morally 
 are co-ordinate with those which induce me to obev authority, 
 and can only be altered by a radical alteration of my whole 
 character. The others are derivative, and may therefore vary 
 as the particular action of the authority varies. And thus 
 we mav assume that the organic variations belong to an 
 entirely different order, and are relatively strong compared 
 with those of the secondary or derivative instincts. 
 
 33. This applies to the case in which we may regard 
 even a moral law as beincr in some dcQ-ree made by a process 
 not unlike that of actual legislation. Such a case is, in 
 fact, more or less illustrated by every great moral teacher. 
 If the Gospels revealed a new system of morality, their pro- 
 mulgation may be regarded as a case of moral legislation. 
 It would be admitted, indeed, by every believer, that for such 
 a case nothing less is needed than a divine interposition ; the 
 direct intervention of a power which can modify the organic 
 as easily as the secondary instincts. But the question, so far 
 as it comes within the sphere of scientific inquiry, is simply 
 one of facts — how far, namely, the promulgation of a new 
 moral principle can alter the accepted moral code? If we 
 could conceive of the moral chano-e wrouo"ht by Christianity 
 as a case of obedience to a power revealed by miracles, the 
 immediate change would not be so much a change of morality 
 as a change in the sanctions of morality. It would be re- 
 vealed to men that certain kinds of conduct had consequences 
 of which they were not previously aware. Such a change 
 might, no doubt, affect indirectly their whole moral character. 
 But it is needless to discuss such a theory, except to take note 
 of the virtual assertion that nothing short of supernatural
 
 152 THE SCIENCE OF ETHICS. 
 
 interference could bring about such a result. So long as we 
 remain within the limits of scientific inquiry, we must admit 
 that the influence of the greatest moral teacher depends, 
 not upon his authority, but upon the congeniality of his 
 teaching to the sentiments by which the social medium is 
 already permeated. He succeeds in so far as his teaching is 
 in harmony with the prevailing instincts. He could not 
 teach if he were not in advance of his fellows, nor find a 
 hearing; unless he were ffivino- articulate shape to thoughts 
 
 O DO I O 
 
 obscurely present to countless multitudes. Like Socrates, he 
 must be something of a "midwife;" he facilitates the birth 
 of the new ideas with which the world is already in travail, 
 and is reallv the interpreter and the mouthpiece of thought 
 seeking for utterance, and representing a slow process of 
 elaboration. The poet and the philosopher, and the religious 
 teacher no less than these, depend for their power upon this 
 unconscious co-operation ; and the more men study the history 
 of the world, the more importance thev come to attach to 
 this occult process of dumb preparation. 
 
 14. When we sav, then, that morality grows and is not 
 made, we really point to this fact, that it is the fruit of a 
 gradual evolution of the orjzanic instinct continued through 
 many generations. Each individual imbibes the moral senti- 
 ments as he grows up and regards them as primitive because he 
 has accepted them without conscious reflection. To alter the 
 code thus elaborated is to alter the most deeply rooted modes 
 of thought and feeling, which are imbedded in the whole 
 scheme of life, and accepted bv the race as its theory of the 
 external world. The reformer must start with these senti- 
 ments ingrained in his character, and must sympathise with his 
 fellows before he can influence them. New discoveries about 
 the external world, new wants due to the growth of society, 
 the gradual accumulation of natural and intellectual wealth, 
 may necessitate some modification of the organic instincts, but 
 such changes must always be slow, and involve many blind 
 gropings after a solution before any tolerable equilibrium can 
 be reached. At each particular stage of the process, the
 
 THE MORAL LAW AS NATURAL. 153 
 
 ordinary mind resists any change in the principles instilled 
 into it from birth, and is only induced to revolt by some very 
 sensible evil. To alter a speculative opinion is hard enough 
 v^'hen its alteration involves any deeply seated change in our 
 system of thousrht, but it is far harder to alter the opinions 
 which have a direct bearing upon our conduct, and still more 
 to modify profound prejudices in the sluggish minds of the 
 e:reat mass of mankind ; and therefore even in the case where 
 the supposed change involves a real improvement in the social 
 adjustment, and has all the advantages, direct and indirect, 
 resulting from that fact, it implies modifications so far-reaching 
 in their character, and requiring the tacit co-operation of so 
 many minds, that it must resemble one of the slow natural 
 processes rather than the sudden change which may be 
 wrought by a single discovery in a particular department of 
 thouQ;ht, or the change produced in comparatively superficial 
 arrano-ement by a leeislative action. 
 
 15. From this observation, too familiar to require further 
 exposition, we may pass to the cognate attribute of the moral 
 law, its eternity and immutability. From our point of view, 
 these phrases must be understood in a sense compatible with 
 the admission of evolution. The actual moral law develops, 
 and therefore changes, whatever may be said of the ideal law. 
 We must recrard the moral instincts as dependent upon 
 human nature or human society, and therefore liable to 
 vary in so far as their subject is liable to vary. When 
 the older school of metaphysicians speak of the immutability 
 of the law, they may either mean that the law will always be 
 the same under the same conditions, which is no doubt true, 
 but gives us no real information ; or that it would be in 
 some sense the same even if the conditions of human life 
 were radically altered, which is either false or refers to a 
 transcendental region of real existences altogether separate 
 from the phenomenal world, and therefore has no intelli- 
 gible bearing upon scientific theories. We cannot mean by 
 eternity or immutability that the moral law will remain 
 unaltered even if the conditions upon which it depends
 
 154 THE SCIENCE OF ETHICS. 
 
 be altered; but only that these are the most fundamental 
 conditions assignable, the permanent conditions of social 
 vitality, which remain constant through an indefinite series 
 of more superficial changes in the social organisation. If 
 we assume that these conditions may be entirely different 
 in some different world, the morality in that world would 
 presumably be also different. If in some distant planet 
 Ivino- were as essential to human welfare as truthfulness 
 is in this world, falsehood might there be a cardinal virtue. 
 The possibility of such a state of things may be denied by 
 those, at least, who profess omniscience ; but if you admit the 
 possibility of such a change, you must admit also the possi- 
 bility of the correlative change in the morality. The morality 
 then is as permanent as the conditions of existence, though 
 there may be a dispute as to the permanence of those condi- 
 tions. A being radically different from man would no doubt 
 have a different code of behaviour. The degree of constancy 
 which we attribute to the moral code will appear more plainly 
 when we come to a more direct deduction of the specific 
 virtues. At present, we may say that any change must be 
 relatively small in proportion to the permanence of the 
 deepest organic instincts as compared with their modification 
 under particular conditions. But we may also go one step 
 further. The variation, we may say, whatever it is, must 
 correspond to a process of evolution, not to what could be 
 called arbitrary modification. If one form of living being is 
 evolved from another, there must be a certain community 
 of plan. Though the two may differ, certain fundamental 
 properties are exemplified in both. Similarly, when a code 
 of law is developed from a simpler code, the latter, though 
 richer in content and more varied in application, must contain 
 certain first principles already given in the first, and extend 
 them by generalising instead of repealing them. There is a 
 continuity though not an identity of sentiment. The type 
 of character which is approved at one stage must be always 
 on the line towards the type approved at a more advanced 
 stao-e. And therefore, although we may raise our standard,
 
 MORALITY AS INTERNAL. 135 
 
 and consider the good man of one period as an average, or 
 less than an average man in a higher period, we may always 
 so far approve of the standard which we have left behind as 
 to consider that its relative judgments were always correct. 
 The qualities approved may have been in everything better 
 than the qualities disapproved, though the highest qualities 
 conceivable were not equal to those now demanded. But a 
 clearer view of the principle follows from another characteristic 
 of morality, which appears from a historical point of view 
 to be of primary importance. 
 
 IV. Morality as Internal. 
 
 16. The clear enunciation of one principle seems to be a 
 characteristic of all the great moral revolutions. The recogni- 
 tion amounts almost to a discovery, and would seem to mark 
 the point at which the moral code first becomes distinctly 
 separated from other codes. It may be briefly expressed in 
 the phrase that morality is internal. The moral law, we may 
 sav, has to be expressed in the form, " be this," not in the 
 form, " do this." The possibility of expressing any rule in 
 this form may be regarded as deciding whether it can or 
 cannot have a distinctively moral character. Christianity 
 gave prominence to the doctrine that the true moral law 
 says " hate not," instead of " kill not." The men of old 
 time had forbidden adultery ; the new moral legislator 
 forbade lust ; and his greatness as a moral teacher was 
 manifested in nothinsi; more than in the clearness with which 
 he gave utterance to this doctrine. It would be easy to show 
 how profoundly the same doctrine, in various forms, has been 
 bound up with other moral and religious reformations in 
 many ages of the world. 
 
 17. What, then, is implied in the change? Conduct may 
 be regarded as a function of character and circumstance. An 
 adjustment of internal to external relations is, it is said, the 
 very definition of life. It follows, hence, that every law as to 
 conduct carries with it a rule as to character, and vice versa.
 
 156 THE SCIENCE OF ETHICS. 
 
 Reofulate a man's feelings or his actions, and you necessarily 
 affect his actions or his feelincrs. Induce a man not to hate 
 his brother, and he will be slow to kill him; and if you 
 persuade him not to kill, vou necessarily limit to some degree 
 the force of his hatred. As it is easier for the primitive mind 
 to accept the objective than the subjective definition of con- 
 duct, the primitive rule takes the corresponding form, and 
 only prescribes qualities of character indirectly by prescribing 
 methods of conduct. 
 
 18. Where, then, is the importance of making the distinc- 
 tion ? If to every mode of feeling there corresponded a 
 definite mode of conduct, the two rules would imply each 
 other. It is possible to suggest certain cases in which this 
 would be approximately true. It seems, at least, to hold true 
 of the appetites as distinguished from the intellectual emo- 
 tions ; for while every appetite has a definite physical organ 
 to correspond to it, the mode of feeling and the mode of 
 acting are mutually implied. To regulate thirst is to regulate 
 drinkino-; but this fails to hold crood so soon as we deal with 
 the emotions which do not discharge themselves by a fixed 
 or narrow channel. If I try to define any mode of action 
 solelv by its objective characteristics, that is, solely by .those 
 qualities through the perception of which I recognise the 
 existence of a world external to myself, I find that the coin- 
 cidence cannot be maintained. Any action so defined may 
 be due to the most varvino; motives, and the same motive 
 prompt the most various actions. Killing generally implies 
 hatred, but in certain cases I may kill from a sense of duty, 
 from a desire of monev, or even from love of the person 
 killed; when I wish, for example, to "put him out of his 
 misery." Therefore, though the prohibition of killing gene- 
 rally forbids the same acts as are forbidden by a prohibition 
 of hatred, the two prohibitions will diverge in an indefinite 
 number of cases. If I wish to forbid all the actions which 
 spring out of hatred, the definition by the internal character- 
 istic is simple and exhaustive, whereas the other kind of de- 
 finition must be indefinitely complex, and must always be
 
 MORALITY AS INTERNAL. JS7 
 
 more or less defective. I may modify the prohibition of kill- 
 ing by permitting it in particular cases, as, for example, in 
 war; and again by adding a number of subsidiary prohibi- 
 tions, forbidding other means of gratifying hatred, such as 
 mere insult or the production of conditions indirectly un- 
 favourable to life. But, after all, it would of course be 
 idle to attempt to sum up all the indefinite variety of ways 
 by which one man can inflict pain upon his neighbours. 
 The difference between the two methods is like the difference 
 between marking a circle by the revolution of a fixed line 
 round a given centre and trying to make an approximate circle 
 by causing a number of other figures drawn from external 
 points to intersect in such a way as more or less to indicate 
 the circumference of the circle. If, in short, I wish to forbid 
 all conduct which is the fruit of a certain disposition, I shall 
 do so at once by forbidding that disposition ; and, in that 
 case, the rule of conduct will tend to become simpler as I 
 briny; new classes of action under this oreneral rule. IL on 
 the other hand, I proceed by the opposite method, and try to 
 give the external characteristics of the same conduct, my rule 
 will become constantly more complex as I endeavour to make 
 it applicable to the indefinite variety of possible cases. 
 
 19. Now every conceivable rule of conduct must be a rule 
 of character. Since action is (upon my assumption) alwavs 
 determined by pain or pleasure, a uniform rule implies uniform 
 feelino- in regard to the conduct prescribed. Rules of action 
 imply a classification of things in general according to their 
 relation to the feelinirs of the assent, and thus the formation 
 of primary rules corresponding to his primitive sensibilities 
 and harmonised by the unity of the organisation constituted 
 by those sensibilities. These primary rules must be capable 
 of statement as rules of character, and for this reason, as we 
 have just seen, cannot be adequately stated as rules of external 
 conduct; for a man, considered as an agent, must be re- 
 garded as simply an organised group of feelings. The calculus 
 of motive cannot include anv other data besides the feelings. 
 The universe only comes into consideration in fixins; certain
 
 158 THE SCIENCE OF ETHICS. 
 
 internal conditions of feeling, whilst the character is the ex- 
 pression of the internal conditions. Now, in anv given case, 
 the conduct will of course depend upon the special stimulus 
 which comes from without as well as upon the character. 
 To know how the character acts, we must know what in- 
 fluences are operative; we must understand the instrument, 
 and we must also know what are the movements of the 
 external player. But to obtain the general rule we abstract 
 from all those particular accidents, and consider only what is 
 essential. We want to know what are the conditions imposed 
 by the structure of the instrument itself, whatever conditions 
 may be imposed from without. But this is to take into 
 account only those external circumstances which are constant, 
 and whose existence is assumed when we assume the exist- 
 ence of the agent. The organism implies the environment as 
 a persistent and universal condition, to which therefore no 
 explicit reference need be made. If breathino- is necessary to 
 life, air must be necessary ; but I can define the laws of breath- 
 ing, its relation to other functions, and its place in the whole 
 organic equilibrium, without making any further reference to 
 this implied condition. So a psychologist might regard a man 
 as a compound of certain primitive emotions, and give their 
 relations as constituting his character without any reference 
 to the external conditions which are necessary to the exist- 
 ence of such a being. The general rules so given would be 
 implied in every particular action and their mode of operation, 
 and each case would depend upon the special stimulus applied ; 
 but the rules would not themselves include any datum of 
 external fact. In the special cases where a given mode of 
 feeling has a fixed external correlative, the two modes of stat- 
 ing the rule would be interchangeable; but as in the general 
 case no such equivalent presents itself, we must necessarilv 
 state the rule in the direct and only possible form, namely, 
 as a rule of character. And since morality is, as we have 
 said, concerned with these general rules, the onlv mode of 
 stating the moral law must be as a rule of character. 
 
 20. The point niav be made clearer by takino; Into account
 
 MORALITY AS INTERNAL. 159 
 
 another consideration. If we consider any given class of ac- 
 tions, we may sav that the intrinsic motive, instinct, or mode of 
 feeling is that which being given, the action follows, or which, 
 if the action takes place, must be present. Hunger we may 
 say (without taking note of possible exceptions) implies eating 
 under certain circumstances, and vice versd the consumption 
 of food under these circumstances implies hunger. And, again, 
 there will be a true intrinsic motive if certain conditions 
 always produce a given desire, even when other conditions 
 may prevent the desire from leading to action. That is the 
 case if hunger is always produced by certain conditions, 
 alth ough other considerations, e.g., the fear of poison, may pre- 
 vent eatin 2:. For a volition we then have only a velleitv. There 
 is, on the other hand, an extrinsic motive when the conduct is 
 desired solely with a view to some further result, which may 
 or may not be present, and is therefore not desired when that 
 result is not anticipated. In such a case the motive is not 
 only overcome, but entirely suppressed by a change of circum- 
 stances, and the apparent end turns out to have been only 
 desirable as a means to some further end. 
 
 21. Assuming this, we may say that there are many classes 
 of conduct to which there is no possible intrinsic motive. 
 This is true of all classes of conduct which can be defined in 
 purely objective terms; for I must understand by such terms, 
 terms which have no direct singificance for the feelings what- 
 ever, or, in other words, which refer solely to the mathematical 
 relations of the object. I have no intrinsic motive for going 
 east rather than west, up rather than down, or round two 
 sides of a triangle rather than along the third. The physio- 
 logist considers all phenomena from this point of view and 
 classifies them entirely by reference to changes in time and 
 place. But when we are speaking of conduct, it is plain that 
 the rules so obtained are only interesting in so far as they can 
 be used by our feelings. They enable us to calculate what 
 will be the consequences of conduct, and reveal their whole 
 significance when we apply them to calculate such conse- 
 quences. The knowledge that one side of a triangle is less
 
 i6o THE SCIENCE OF ETHICS. 
 
 than the two others has no interest in itself, but as soon as I 
 am in a hurry to reach a given point, or wish to avoid reach- 
 ing it too soon_, it of course determines the mode in which my 
 feelings will prompt my conduct. 
 
 22. In the general case, however, the statement will be 
 different. A given class of actions will be pleasant or painful, 
 and may therefore sugirest a rule of conduct of one kind or 
 other. Now, generally speaking, we may say of a certain 
 class of conduct partly defined by external considerations, 
 that it must correspond more or less closely to some intrinsic 
 motive. It may be such that one kind of conduct alone will 
 be desirable on the assumption that the agent is accessible 
 to certain motives. That is, if we have a certain character, 
 we shall act in the way supposed. But it will also generally 
 be true that the same conduct may be prompted by other 
 motives; so that we cannot certainly infer the motive from 
 the conduct. And besides this, it will generally be true that 
 the desirability of the conduct depends to some extent upon 
 circumstances not expressed in the rule. So that some kinds 
 of conduct, although falling within the class definition, do 
 not present themselves as desirable. That is to say, it has no 
 intrinsic motive precisely corresponding to it. So, to repeat 
 the former illustration, certain actions are Q:enerallv the 
 result of kindness. So long as I am kindly I shall not kill ; 
 but the abstinence from killing may be the result of many 
 other motives, such as fear of the gallows ; and in some rare 
 cases kindliness might even prompt to killing. To say 
 this is only to repeat the previous statement of the impossi- 
 bility of making the external and internal codes precisely 
 coincident. 
 
 23. We may ask, then, how any external rule of conduct 
 is possible; for if all conduct is conditioned bv feeling, and 
 uniform conduct implies uniform feeling, whilst external 
 uniformity could only be secured in practice by internal 
 variation, and vice versa, it would seem that no external rule 
 can express a real rule of conduct. In one sense this seems 
 to be rigidly true, 'iliat is, there is no rule which a human
 
 MORALITY AS INTERNAL. i6i 
 
 being will obey under all circumstances in spite of all conceiv- 
 able conflicting motives. This is simply an admission that 
 the strength of the will is finite. It merely asserts that other 
 motives may override those implied in the observance of the 
 rule ; the rule may still be an operative force, though not the 
 sole or dominant force. But the question is how any rule 
 of the external kind can express even a uniform desire, and 
 if not, whence it can derive any permanent influence? To 
 this we may say, that, in the first place, the observance of 
 such a rule may become a habit. The essence of a habit is, 
 as we have seen, that I act in a certain way in obedience to 
 certain signals, without calling up all the feelings implied. 
 If I am in the habit of getting up to breakfast when a bell 
 rings, I may get up when a bell rings which I know to be 
 the indication ; but I must suppose that in this case the 
 knowledge is more or less of the potential kind; that is, that 
 I mio-ht know if I reflected or brouo-ht into vivid conscious- 
 ness all the thoughts connected with the given symbol. And, 
 in the second place, the observance, and still more frequently 
 the breach of a rule, may be due simply to the fact that I am 
 an unreasonable and inconsistent being. This means virtually, 
 as we have already seen, that I may have a different character 
 at different times, and perhaps allow a set of feelings which, 
 on other occasions, are relatively superficial, to overpower 
 feelings which are at other times the most powerful. I have, 
 however, spoken sufficiently of this, which In fact is part of 
 the general problem as to the reasonableness of action. So far 
 as I become reasonable this kind of irregularity will disappear, 
 and I shall be governed in the same cases by the same 
 motives, and cease to apply rules in cases where they are not 
 applicable. 
 
 24. Let us suppose, then, that I act reasonably in the sense 
 that, as I always judge by the same principles or am actuated 
 by the same set of feelincrs brouQ-ht into harmony and sub- 
 ordination, and therefore that a uniform rule does in fact 
 correspond to a uniform mode of feeling, then I may accept 
 the rule as affording a sufficient presumption under ordinary 
 
 L
 
 1 62 THE SCIENCE OF ETHICS. 
 
 circumstances. It serves as an indication that the facts are 
 such as would determine me to act in a given way. So, 
 for example, my objection to killing may be founded upon a 
 dislike to giving pain. That is the intrinsic motive of a class 
 of actions which cannot be defined by any absolutely coin- 
 cident, external, correlative. Still it gives me a strong pre- 
 sumption against killing, because in almost all cases killing 
 gives pain. If I see a man, therefore, and know nothing 
 about him beyond the fact that he is a man, I shall refrain 
 from killing him. I shall again refrain unless the presump- 
 tion is rebutted by evidence that killing will diminish pain ; 
 and in that case, I shall kill if, in fact, the dislike to giving 
 pain is the intrinsic and sole applicable motive of my conduct 
 in relation to my neighbours. On this supposition the general 
 rule is a conditional one, although the conditions may not be 
 distinctly formulated. 
 
 25. Again, the rule may be accepted from an extrinsic 
 motive ; that is to say, from a motive not implied in the 
 definition of the class of conduct commanded or prohibited. 
 This, of course, is the case wherever the rule is accepted, not 
 for itself, but from regard to the authority by which it is 
 imposed. In this case I do not object to killing, but to some 
 consequence not necessarily or invariably connected with it. 
 I may object to kill because killing leads to the gallows or 
 because it leads to damnation. Were I certain to escape the 
 hangman and to obtain spiritual absolution, I might still 
 be ready to kill. This, of course, is a highly important case 
 in practice. The legislator is forced to classify conduct by 
 its objective manifestations. He, therefore, is necessarily 
 limited by the considerations already suggested. He can- 
 not forbid all the possible manifestations of a passion such 
 as hatred, but only those which produce certain tangible 
 and visible consequences. However elaborate his code, there 
 will still be iimumerable devices by which a man whose 
 character prompts him to take the forbidden courses can 
 gratify his passions by indirect methods. And where the 
 moralist and the religious teacher is misled by the analogy,
 
 MORALITY AS INTERNAL. 163 
 
 and instead of forbidding the passion tries to classify all the 
 modes of conduct to which it may lead, he gets into the same 
 difficulty. lie permits what he does not prohibit, and is 
 therefore in danger of producing hypocrisy instead of virtue, 
 and stopping a few holes in a sieve instead of stopping the 
 stream at its source. And here we have the secret of the 
 immense importance attributed by all the higher moralists 
 to the other mode of statement. 
 
 26. Hence we may define the spheres to which rules, which 
 have or have not an external reference, are necessarily limited. 
 So long as any external element is present in the formula, it 
 must be a formula of the organ, not of the organism, appli- 
 cable under particular conditions or circumstances, and not 
 belonging to the man simply in respect of his intrinsic 
 motives. For if the respect for the law is really a case of 
 respect for the imposing authority, that authority is itself a 
 product of the primary instincts acting under special con- 
 ditions, and determined by them and by the properties of the 
 ''tissue" of which it is a modification. Therefore a rule 
 of conduct which tacitly or implicitly depends upon some 
 principle of authority must by its nature define, not a property 
 of the tissue, but of some special product, determined by exter- 
 nal circumstances and variable from time to time. And if 
 we take the case of any rule which does not coincide through- 
 out as to the motive and the conduct, it can only give a 
 conditional rule of conduct, the condition being not the 
 general conditions implied in the existence of the agent, but 
 in some special set of facts which may vary whilst those con- 
 ditions are fixed. In either case, the rule virtually implies 
 acting by some criterion according to which we classify the 
 particular case before us. To have some such criterion is of 
 course an essential in all conduct, since conduct of every 
 kind involves some set of external events. Now, if the 
 criterion shows that the case belongs to a class of events 
 corresponding to a certain kind of feeling, there are always 
 points at which it fails us. If through hai)it or inattention 
 we continue to observe it, we act inconsistently; if we are
 
 1 64 THE SCIENCE OF ETHICS. 
 
 fully consciovis of what we are doing and are guided by 
 uniform principles^ we shall not observe it. Therefore the 
 rule^ properly stated^ does not give a law of the character, 
 but only of the character as affected by certain special and 
 variable circumstances. If, again, the criterion marks the 
 case as belonging to a class of actions forbidden by a certain 
 authority, we do not ask whether the classification itself 
 depends upon any intelligible principle or implies anything 
 beyond a classification by external characteristics. We 
 are consistent in obeying the law, whatever it prescribes, 
 even though the classification assumed by the law be from 
 other points of view inconsistent. In this case, therefore, 
 the true rule is in the form: Do as you are bid by somebody. 
 And as here, again, the instinct of obedience to any authority 
 whatever is necessarily dependent upon the particular cir- 
 cumstances under which the authority has grown up, and is 
 a deduction from the primary instincts in a particular appli- 
 cation, not the principle from which they can be deduced, 
 we still have a rule of character as affected by special con- 
 ditions, not a rule which corresponds to the organic relations 
 of character. Hence any external law whatever fails to give a 
 law of character simply ; and, on the other hand, the organic 
 law must also transcend all these special applications, however 
 general some of them may be. The moral law as a law of social 
 *' tissue/' or as a law concerned only with the development of 
 character by the intrinsic properties of the social organism in 
 presence of fixed external conditions, can only be adequately 
 expressed in terms which have no external reference. 
 
 27. The process, then, by which the moral law (or rather 
 the law of conduct of men considered simply as constituting 
 the social tissue, for this law includes but is not coincident 
 with the moral law) is developed, is a process of generalisa- 
 tion. It corresponds to a vast induction carried on by the 
 race as organised in society. It is a gradual disengagement 
 of certain primary instincts, and a distinct perception of their 
 value and mutual relations from the perplexing complexity 
 of their particular manifestations — a process which is more
 
 MORALITY AS INTERNAL. 165 
 
 complex because it involves a modification of the emotions 
 and of the whole character^ as well as a simple intellectual 
 process. Certain modes of conduct are seen to be bad, that 
 is, they are disliked for some reason or other by the persons 
 concerned. Society tries to put them down as it tries to 
 extirpate dangerous animals. It develops in proportion to 
 its success in this undertaking, which implies, again, a 
 development of the feelings hostile to such practices, and at 
 the same time of a social structure capable of applying 
 effectual restraints. Now, so far as the growth of a certain 
 body of sentiment is implied, the question emerges. What is 
 the common principle in virtue of which this, that, and 
 the other bad practice is hateful ? As men grow more rea- 
 sonable, they are constantly comparing and correlating their 
 feelings; some such process is involved in all conscious action; 
 they are thus classifying the various external phenomena 
 in respect of the feelings excited. Gradually, it appears 
 that this process leads to a necessary divergence of the 
 two methods of classification. The same feeling is excited 
 in countless wavs according:; to the endless combination of 
 external facts. No definite class of external facts can be 
 assigned which precisely corresponds to one intrinsic feeling. 
 Hence, as the ultimate principle of classification must for all 
 purposes of conduct be by the primary feelings, we find that 
 the most general rules of conduct must be expressed in terms 
 of character, and the other rules, which contain the appli- 
 cation of those general rules to more special cases, must take 
 a subordinate position, and be regarded as being only of con- 
 ditional value. This, again, is the same thing as to say that 
 these general rules express the properties of the social tissue, 
 or those properties of the organic growth which underlie 
 all the special arrangements which can only be regarded as 
 comparatively superficial products of the tissue, determined 
 by its necessary external relations. Finally, it may be re- 
 peated that they must necessarily correspond within narrow 
 limits to a statement of the conditions of vitality of the 
 tissue which they characterise.
 
 1 66 THE SCIENCE OF ETHICS. 
 
 28. This, agaiu^ gives an intelligible sense to another sense 
 in which we may understand another predicate frequently 
 attributed to the moral law. Moralists frequently assert its 
 supremacy, and appear in so doing to fall into a vicious circle. 
 A legislator orders me to lie, and the moral law orders me 
 to tell the truth. Then, it is said, I ought to obey the moral 
 law. But if ^^ouoht" means it is rio-ht, and rio;ht means in 
 conformity to the moral law, this appears to be equivalent to 
 saying that I ought to do what I ought to do. It is difficult 
 if not impossible to escape from this dilemma so long as we 
 are speaking of a supremacy de jure; but if we speak of a 
 supremacy de facto, the statement may bear a tenable inter- 
 pretation. It is conceivable, in fact, that any law belonging 
 to a given association may be regarded as more or less con- 
 ditional. I may agree to obey its rules so long as those rules 
 do not conflict with the laws of a higher authority. We 
 might conceivably have a state which in this way did in fact 
 recognise the moral law, so that no law would actually be 
 enforced when it conflicted with the moral sense of the 
 community. How far this is the case in any really exist- 
 ing state I do not presume to say. It is undoubtedly very 
 difficult to enforce laws when they palpably oflend the 
 recognised morality of a country ; and the conceivable reply 
 of a lawyer, that they are still laws, is a mere verbal reply 
 when we are dealing with facts. But however this may be 
 in particular cases, the general principle remains true. The 
 laws of a state, along with all its other arrangements, are, 
 generally speaking, the product of the social medium from 
 which it springs, and generally, therefore, reflect the preva- 
 lent moral feeling. And further, the moral feeling is itself 
 dependent upon conditions of a higher and more general 
 order than those by which the political organisation is deter- 
 mined. We may therefore say briefly that the morality of 
 a race, as it depends upon the most permanent conditions, 
 represents its fundamental characteristics, and that the 
 subordinate rules of conduct, whatever they may be, must 
 be regarded as springing from them, and not vice versa.
 
 BASIS OF MORALITY. 167 
 
 Many laws, indeed, exist which arc regarded as more or less 
 immoral by large classes of the persons bound by them. 
 The law does not cease to be a law because it is immoral, 
 but certainly it has less chance of being anything but a law 
 in name ; and the only general principle must be that the 
 characteristics which are most deeply seated and dependent 
 upon the most permanent conditions must tend, however 
 long the process, to override those which are relatively super- 
 ficial and contingent. Further, it may be added that, in an 
 ideal state of society, every general principle would also be 
 recognized in every particular rule. This is a result, indeed, 
 to which we must expect a gradual approximation rather 
 than anticipate its actual attainment. So, for example, if 
 the moral law commands kindliness and some particular 
 rule prescribes a cruel action, we may say that if the 
 society is progressive (a condition which is of course neces- 
 sary) some uniform rule must be worked out. Then, by 
 hypothesis, kindliness has been discovered to be a quality 
 characteristic of social vitality, and the rule can be laid down 
 absolutely ; whereas the rule which prescribes cruelty is the 
 product of some particular combination of circumstances, 
 and can only be stated conditionally. Hence we may say 
 that the general tendency must be to bring about such a 
 modification of sentiment that the superficial and excep- 
 tional rule may be superseded by one consistent with the 
 general principle. The statement, however, as to the supre- 
 macy of morality and the conscience is generally, as I think, 
 understood in a different sense, and as applicable to particular 
 cases of conduct rather than to the position occupied by the 
 moral sentiments in the general process of evolution. We 
 shall come to this in a later chapter. 
 
 V. Basis of Morality. 
 
 29. So far, then, the argument has justified some of the 
 predicates most generally applied to the moral law, though 
 it imposes a certain interpretation upon them. By saying
 
 1 68 THE SCIENCE OF ETHICS. 
 
 that a law is moral, we mean that it belongs to human beings 
 as such, and not as belonging to any special class. This, in 
 my view, amounts to saying that the moral law defines a 
 property of the social tissue. Hence it must be natural, not 
 artificial ; it must grow, and not be made ; for these pro- 
 perties are the intrinsic and underlying properties implied 
 in all special societies, incapable of being abruptly altered 
 by the action of any particular person, or in obedience to 
 any subordinate series of events, and gradually developed as 
 the society grows instead of being the fruit of special con- 
 tingencies. The law must be eternal so far as anything 
 human can be eternal, for it must be an approximate expres- 
 sion of the conditions of social vitality, as the instincts to 
 which it corresponds are the instincts by which the life of 
 the society is maintained ; and it must therefore be as per- 
 manent as those conditions themselves. It varies only by 
 development, as each step in the social evolution represents 
 a fuller solution of the problem of adapting a society formed 
 of given materials and acting under fixed conditions to the 
 needs which those conditions impose. Again, it must be 
 capable of expression as a law of internal character, not as a 
 law of external facts ; for the only variable element is the 
 character, and the problem to which it supplies an answer is 
 the determination of the most effective qualities of character 
 which can be developed in a given agent to make him an 
 efficient member of society. In the infinite variety of cir- 
 cumstance, these qualities may manifest themselves in a 
 corresponding variety of methods, which can never be 
 adequately summed up or classified by external character- 
 istics. And, finally, since these qualities represent the most 
 general rules of action, such alone as can be stated absolutely 
 — that is, without reference to varying circumstances — the 
 law must be supreme. It deals with the first principles, the 
 primary reasons to which every particular case must present 
 a special application. 
 
 30. Supposing this to be admitted, we have still the critical 
 problem before us. For the natural method would now be
 
 BASIS OF MORALITY. 169 
 
 to deduce from the general principle the particular rules 
 of conduct, and to show that they do in fact lead to the 
 recognised moral law. There is, as I began by saying, a 
 tolerable agreement as to the contents of that law, however 
 wide mav be the divergence as to its form. This being so, 
 it would be absurd, as it would be really misleading, to affect 
 the method of an a priori deduction. We know what are 
 the conclusions to be reached, and need not speak as though 
 we had before us nothing but the premisses. It will be 
 enough to show that the general principles of morality can 
 in fact be deduced from the theory laid down, without sup- 
 posing that we are starting from that theory with perfectly 
 unprepossessed minds, in search of any principles that may 
 turn up. Certain general remarks, however, may be pre- 
 mised, to elucidate the nature of the investigation. We are 
 to see how certain rules have been reached by the evolution 
 of society from a period at which, as we assume, though attribu- 
 ting to our assumption no more than an approximate accuracy, 
 the individual man had his present organisation, but at which 
 societv existed only in germ, and the custom upon which it 
 depends had not been distinctly elaborated nor consciously 
 accepted. We have to deal, then, only with the rules which 
 have been created by society, or rather which have been 
 evolved as society has evolved, the internal and the external 
 processes being necessarily correlated, and not related as though 
 one had appeared first and the other been moulded upon it. 
 
 31. Hence we have nothing to do with certain rules of 
 conduct which are implied in the very constitution of the 
 race, and which would certainlv have to be stated as primary 
 conditions of existence. Men must have certain appetites, 
 hunger, the sexual instinct, and so forth, without which the 
 race could not survive for a day. They are, again, implied in 
 every later development; but we have only to do with the 
 later modification, and not with the initial state. But, in 
 the next place, it is clear that we might lay down many rules 
 of conduct as necessarv to the existence of society which 
 cannot be regarded as properly moral. Thus, for example.
 
 I70 THE SCIENCE OF ETHICS. 
 
 the most obvious condition of the social or individual vitality 
 is what we may roughly call the instinct of self-preservation. 
 If men had no instinct which kept them from walking over 
 precipices or swallowing fire, they would have a very pre- 
 carious tenure of life. Nobody would be called moral for 
 obedience to the rules formed from such instincts, nor even 
 for obedience to the higher rules which are developed from 
 them as society grows. The instinct of self-preservation 
 becomes finer and more sensitive as the emotional and intel- 
 lectual faculties are developed. We become aware of a greater 
 number of conditions, measure them by more delicate tests, 
 and are more sensitive to remote consequences. For the mere 
 avoidance of fire, precipices, poisons, and so forth, we come 
 to observe with more or less regularity a complex set of rules 
 calculated to preserve our bodily health ; but such rules are 
 not generally regarded as moral at all. We do not say that 
 a man is good because he takes care of his digestion or makes 
 it a principle to take a certain quantity of exercise daily. 
 Such conduct is denied to be moral, although we may call 
 it prudent, because it is consistent with selfishness, cruelty, 
 falsehood, and other bad qualities. Briefiy, it is admitted 
 that, in some sense or other, morality implies action for the 
 good of others; but to define that sense accurately is to solve 
 some of the most vital moral problems. We may ask, in 
 fact, whether it is or is not possible for any man to aim at 
 the good of others as an ultimate end ? and, again, whether 
 it is necessary to moral conduct that the good of others should 
 be consciously intended, or whether it is sufficient that it 
 should be a natural consequence of the conduct in question ? 
 To give some answer to these and the allied questions will be 
 one object of the following pages. 
 
 32. We may first recall a distinction, already stated, which 
 will be relevant to this inquiry. Society, as I have said, may 
 be regarded both as an asfo-recrate and as an oro;anism. There 
 are certain qualities which we may suppose to vary in the in- 
 dividual without necessarily involving a change in the social 
 structure. 71ie relations of the parts may remain sensibly
 
 BASIS OF MORALITY. 171 
 
 the same although the general vigour — the sum-total of the 
 enero-ies involved — is increased or diminished. On the other 
 hand, there are qualities in respect of which the reverse is true. 
 No change can take place in them without implying a corre- 
 sponding change in the character of the social union. The 
 distinction is not absolute; for every change in anv part of 
 the organism must have some reaction upon all its other 
 parts ; but the distinction may be, or rather must be, made for 
 purposes of classification. We must distinguish, that is, be- 
 tween such a quality as loyalty, which cannot be supposed to 
 increase or diminish without alterino; the essential character- 
 istics of the social tie, and such a qualitv as mere personal 
 prudence, which would no doubt have a great influence upon 
 the whole social organism, but which may be considered apart 
 from that set of consequences and which immediatelv affects 
 the total power of the community rather than the relation 
 between its members. Some reference, whether erroneous or 
 not, to this distinction seems to be implied in the criterion by 
 which we judtre whether a given rule does or does not belon*)- 
 to morality proper after we have admitted it to be a rule of 
 social vitality. And this distinction may be made without 
 considering the further and most important question how far 
 the quality which is actually and permanently essential to 
 society considered as an organism, rather than to societv con- 
 sidered as an affo-reg-ate, does or does not involve anv conscious 
 reference to the welfare of the societv, or of others besides 
 the agent. This is a question which demands a special 
 discussion, and which, therefore, it is desirable to reserve as 
 much as possible. 
 
 ^^. The question, therefore, which now has to be considered 
 is the deduction of the moral rule from the general principle 
 of social vitality, and with a reference to the question how 
 they are distinguished from other rules deducible from that 
 principle. If we have rightly assigned the germs to which 
 the moral law belongs, we have still to consider how these 
 germs may be divided into species, and what is the specific 
 difference.
 
 ( 1/2 ) 
 
 CHAPTER V. 
 
 CONTENTS OF THE MORAL LAW. 
 
 I. The Law of Nature and Morality. 
 
 1. The law of nature has but one precept, "Be strong." 
 Nature has but one punishment, decav, culminating in death 
 or extirpation^ and takes cognisance of but one evil, the 
 weakness which leads to decay. From this, the most general 
 point of view, we can make no distinction between the various 
 instincts except in so far as they do or do not imply the 
 vitality of the organism to which they belong. But when 
 we regard the individual as an organism within an organism, 
 the law takes different forms and requires to be diff'erentlv 
 stated, according to its mode of impact. In one great class of 
 cases it applies to the instincts in respect of which society is 
 an aggregate, and the conduct of each individual may varv 
 without implying a corresponding variation in the social 
 organisation. In the other class it applies to those instincts 
 which are the vital forces of that association, and cannot varv 
 without a corresponding variation in it. In one case the 
 effect upon the individual is the primary efl^ect, and the society 
 is affected through its constituent units; in the other case, 
 the units are affected through the soqiety, and the law cannot 
 be intelligibly stated without taking the social factor into 
 account. This may be expressed again by saving that the 
 great law, "Be strong," has two main branches, "Be prudent " 
 and "Be virtuous." To assio;n the mutual relations of these 
 resulting codes, which, although distinguishable in abstract 
 analysis, are so closely connected in the concrete, is the task 
 upon which we must now enter. By some thinkers morality
 
 THE LAW OF NATURE AND MORALITY. 173 
 
 has been resolved into a particular case of prudence ; according 
 to others, prudence may be resolved into morality, or both 
 into right reason or into a desire for happiness. Let us 
 consider how the case must be stated from our point of 
 view. 
 
 2. This statement takes for granted the general nature of 
 the distinction. It is, in fact, admitted that by the moral 
 code we mean to refer not merely to the predicates already 
 noted, such as the eternity, supremacy, and so forth, of the 
 code, but also to its havino; in some sense or other a reference 
 to the welfare of the society. What we have now to do is to 
 substitute for that "some sense or other" a more precise 
 definition; and the task would be accomplished if we could 
 deduce the particular laws of conduct from the laws of nature, 
 and then show which of these laws coincide with the moral 
 law and why? There is here the difficulty tliat the moral law 
 has not been, and, if I am right, cannot be accurately codified, 
 even if the agreement as to its contents were still closer than 
 is actuallv the case. If we classify conduct by external marks, 
 we have a variety of general rules, none of which are precise or 
 unconditional. Thus, for example, we have such virtues and 
 vices as generosity and avarice, which refer to a particular class 
 of actions, namely, in this case, to dealings in money matters. 
 If we proceed to ask for a more explicit account of what is 
 meant by avarice, we find that moralists do not mean to 
 condemn the love of money sim p licit er ; for money represents 
 every material object of human desire, and to condemn them 
 all would be to condemn life in the world. Thev really 
 condemn certain excesses, and more especially those which 
 imply selfishness. Avarice is the love of money so far as it 
 implies a disregard of the claims of others, or, again, so far as 
 it implies a defective appreciation of the higher enjoyments. 
 To condemn avarice, then, is to condemn one kind of selfish- 
 ness, and bv implication to condemn it in many relations of 
 life which have no relation to money. A classification pro- 
 ceeding by the various external applications of the internal 
 principle would be endless, and involve repetitions and cross
 
 174 T^HE SCIENCE OF ETHICS. 
 
 divisions. If, on the other hand, we take the other mode of 
 defining morahty, we find that the law reduces itself to one 
 or two simple principles — to the statement, for example, that 
 we should love God and our neighbour; that we should hurt 
 no one, and do to others as we would that they should do 
 to us. But such statements are too general to be available for 
 our present purpose. 
 
 3. To find a classification of the virtues which will not run 
 into infinite detail or be a simple affirmation of the general 
 principle, we may observe that the internal mode of classifica- 
 tion suggests a method which will be sufficient for our purposes, 
 and which corresponds to the ancient doctrine of the cardinal 
 virtues. We may begin by considering the qualities which 
 belong to the individual primarily, and ask how far they have 
 any moral significance. The general formula of such virtues 
 is, " Be strong," or, as we may put.it, " All weakness is an evil." 
 The simplest organism may be considered in respect of its 
 strength or weakness, that is, its power of preserving its life 
 under the various conditions of existence, and that before 
 any complexity of social organisation has been reached. As 
 the social development affects these qualities as well as the 
 others more directly involved, the social pressure constitutes 
 a law which may have some moral character. In the next 
 place, the development is of two kinds, or may be regarded 
 under two main aspects — the emotional and the intellectual. 
 To each of these there belongs a characteristic moral law; 
 for as soon as we can regard the individual as a complex 
 organism, made up of different instincts and capacities for 
 feeling, one main condition of vitality must refer to the 
 strength and mutual relation of those instincts. Hence we 
 have the virtues of which the general formula is, "Be tem- 
 perate," or the correlative statement that all excess is an 
 evil. These qualities, again, however modified in the higher 
 phases of development, must exist in germ even before the 
 animal is capable of anything that can be called reasoning, or 
 of a conscious reference to the distant or the future. When 
 we consider the intellectual development, we have a third class
 
 VIRTUE OF COURAGE. 175 
 
 of virtues refcrrino; to the conditions of intellectual efficiency, 
 the jrcneral formula beinsi; in this case, "Be truthful," and 
 " All falsehood is evil." And, finally, as the social organisation 
 becomes developed, and has special moods of sentiment corre- 
 sponding to it, we have the virtues which correspond directly 
 to a condition of social vitality. The formula may be ex- 
 pressed in the social or commn assertion that all injury to our 
 fellows is an evil. I will not inquire whether this classification 
 can be regarded as accurate or exhaustive. It will give us, at 
 any rate^ a clue to the inquiry quite sufficient for our purpose ; 
 that is, in fact, for showing how the specific difference under- 
 stood by the word '^ moral " is brought out in the code actually 
 formed by our approvals and disapprovals, and what is the 
 nature — so far as it can be defined — of the process by which 
 the development is effected. 
 
 II. Virtue of Courage. 
 
 4. " Be strong " is, I have said, the general precept of the 
 law of nature. The strength of the society, again, is in- 
 creased by the strength of its individual members. So far as 
 each unit is stronger, braver, more energetic, and more in- 
 dustrious, more capable, therefore, of holding his own against 
 external enemies or material disadvantages, so far is the 
 society composed of such individuals stronger. This, indeed, 
 must always be understood with reference to a tacit condition. 
 If an increase of courage necessarily involved an increase of 
 insubordination, we should have to ask whether the increased 
 power of the individual soldier was cheaply or dearly bought 
 by the weakened discipline of the army to which he belongs. 
 But there is no necessary divergence between the two qualities. 
 Increased energy may go along with increased power of co- 
 operation ; and therefore the rule, " Be strong," may be stated 
 without referring to a condition which is generally latent, 
 though, under particular circumstances, it may assert itself 
 and demand attention. Ceteris paribus, we may say the 
 increase of individual energy is an advantage to society 3 and.
 
 176 THE SCIENCE OF ETHICS. 
 
 as a matter of feet, we find that the civilised society differs 
 conspicuously from the ruder by stimulating more vigorously 
 and systematically the various energies of its members. The 
 most conspicuous virtue of this class is the virtue of courage. 
 The deduction of courage from the general condition of social 
 vitality is manifest. In cases where a society has to strucrirle 
 against external enemies, military excellence is the most obvious 
 guarantee for its security, and in rude societies, military ex- 
 cellence is proportional to courage. Savage tribes may often 
 be said to hold life at every moment upon the tenure of military 
 prowess. And, moreover, it is plain that the same principle 
 would hold good, not only where it is directly exemplified, 
 but where there are apparent deflections from the rule. That 
 is to say, that in cases where the external conditions are 
 such as to give less importance to the military energies, courage 
 is less highly estimated ; and so, again, the particular kind of 
 courage most in demand varies as fraud or force is the most 
 effective weapon for the particular race and purpose. The 
 most familiar instance is in the very different estimate which 
 is placed upon courage in the two sexes. Moralists would 
 hardly admit that the rule ought to be different; but they 
 admit that, in point of fact, want of courage in men provokes 
 contempt in most modern nations, and a want of chastity 
 is regarded with comparative indifference; whilst in the case 
 of women, the rule is altered : unchastity is held to be the 
 most unpardonable of crimes, whilst cowardice, in some rela- 
 tions at least, is thought to be rather graceful than otherwise. 
 Practical moralists lament these inconsistencies, and theorists 
 have invented more or less ingenious hypotheses to account 
 for them. The historical explanation is, within its own limits, 
 simple and obvious. We may state that in early social stages 
 fighting power was the critical or essential power for each race ; 
 that those in which it flourished most conquered, and often 
 exterminated the rest; and thus that a cultivation of the 
 military qualities, the most conspicuous of which was bravery, 
 was a characteristic of the dominant races. 7'he warrior was 
 the natural leader, and the best warrior had the first choice
 
 VIRTUE OF COURAGE. 177 
 
 of spoil, or the greatest chance of gratifying his passions. 
 Naturallv excellence in war was coveted and admired by every 
 one. The estimate once fixed tends to prolong itself even 
 when some of these conditions disappear. Every male child 
 in a certain rank in England is still brought up from its 
 cradle to value itself on being "a gentleman;" and to be a 
 gentleman is, amongst other things, to be ready to take one's 
 own part with sword or fist. To women, on the other hand, 
 has been assigned from the earliest period of the division of 
 labour the class of social functions for which military excel- 
 lence was not required. The savage acquired his wife by 
 knocking her down; to him, therefore, the ideal feminine 
 character must have included readiness to be knocked down, 
 or at least unreadiness to strike again ; and as some of the 
 forms of marriage recall the early system, so in the senti- 
 ments with which it is reirarded there mav still linger some- 
 thing of the early instinct associated with striking and being 
 struck. 
 
 V K. Thus we mav sav that courage is a necessary condition of 
 the vitality of a society so long as it depends upon militarv 
 activity; and this implies that every man is in such a 
 society trained to be brave in so far as his possession of that 
 quality entitles him to respect and the advantages of being 
 respected; and, again, in so far as he imbibes the current 
 opinion bv which the standard is fixed from his earhest period 
 of conscious thought. The development of society implies a 
 corresponding modification of this sentiment. The military 
 virtues become less prominent as war occupies a smaller part 
 of the total activities and is a less essential part of social 
 efficiency. But there is simultaneously a change in the whole 
 mode of thought. In an early social stage, we may suppose 
 that the warrior who shrinks from dano-er is reoarded with 
 contempt and dislike; since the life of every member of the 
 tribe may depend at any moment upon the prowess of his 
 fellows, the whole social group is closely interested in the 
 success of every one of its members. But, as society develops, 
 new cases present themselves for classification. For, in the 
 
 M
 
 \p:LMili-j 
 
 178 THE SCIENCE OF ETHICS. 
 
 first place, there are brave enemies as well as brave champions 
 of our own. What is the sentiment which they create? We 
 may conceive it possible that a brave enemy might be con- 
 sidered simply as a more dangerous antagonist. He might be 
 regarded with greater antipathy, just as a big wolf would be 
 worse than a little one. But in the more civilised race the 
 chivalrous sentiment begins to manifest itself in imperfect and 
 frao-mentary ways. In a comparatively civilised state, people 
 still hate enemies at a distance in proportion to their courage, 
 and set it down as more or less diabolical ; whilst at the same 
 time they are capable of a true chivalry towards those with 
 whom they are more nearly allied. The English and Scotch 
 Borderers might respect each other's courage, and be the better 
 friends when the fin-htino; was over. They mifjht, at the same 
 time, regard courage in a Saracen as a bad quality, demanding 
 a more undying antipathy. This growth of the chivalrous 
 feelino- implies, on the one side, a growth of sympathy, inas- 
 much as we are now capable of admiring the man who was 
 bevond the pale of any common feeling; and, on the other 
 hand, it may be regarded as involving an implicit generalisa- 
 ion. We say implicitly that we regard the brave enemy as 
 intrinsically admirable in so far as he has shown a good 
 quality, and objectionable only from the accidental circum- 
 stances which have made his interests incompatible with our 
 own. We thus have virtually reached the general principle 
 that courage in war is a valuable quality for its owner and 
 his side, and therefore one which we can admire, although it 
 may or may not be valuable to ourselves. 
 V 6. The increased intellio-ence or scnsibilitv which makes 
 such a judgment possible carries along with it other changes. 
 In the early state, attention is fixed exclusively upon the 
 simple case of military excellence. The warrior who runs 
 away is doing me an injury, for the tribe has its interests so 
 much in common that the bad conduct of one necessarilv 
 injures others. But as men become more intelligent and 
 society more complex, this simple observation requires to be 
 modified. For, in the first place, it must come to be per-
 
 VIRTUE OF COURAGE. 179 
 
 ceived that courage pei- se is not necessarily a good quality. 
 If, that is, courage be defined as simple insensibility to danofcr, 
 it may obviously be carried to excess; and we imply this by 
 the condemnatory phrase " rashness." Indeed it is so little 
 a virtue, that cases may certainly be imagined in which what 
 we call cowardice would be a virtue. There are races of 
 animals which owe their safety to a lively perception of danger. 
 The excellence of a hare consists in running away, and those 
 hares which were best at runnino- and also quickest at takino- 
 alarm would tend to survive, and set the standard of hare- 
 morality. If this is not the case with man, we can only 
 explain it by the fact that, in the conditions of human life, 
 inilitary excellence is the necessary condition of success, and 
 that military success requires courage. But courage, as 
 we see, requires to be more accurately defined. "Moral 
 courage," as it is called, tends to take the place of " physical 
 courage." By moral courage we must understand not simple 
 insensibility to danger, which is consistent with idiocy, but 
 a power, as we say, of " keeping our heads," or, in other 
 words, of reasonino: as deliberately and actino; as coolly under 
 danger as when there is no danfrer. This quality would be 
 as useful to hares as to men, and indeed is implied in the 
 intellectual development; for it is simply a statement that a 
 power of reasoning — that is, of consulting all relevant cir- 
 cumstances and actintr in accordance with a sound judgment 
 — is an essential part of practical reason. Courage, there- 
 fore, changes its quality to some extent, and we admire the 
 kind of courage which is manifested by the general com- 
 manding under stress of great danger and heavy responsibility 
 more than the simple courage of a soldier who Avalks up to a 
 battery, or of a hunter who confronts a tiger in his jungle. 
 X \^ 7. This, again, involves another process of implicit reason- 
 / ing; it becomes manifest, in short, that courage cannot be 
 considered by itself. We must determine its relations to the 
 whole character, or we shall be admiring it in cases where 
 it is absolutely prejudicial. Fighting, however, and so far 
 courage, has been from an early period an essential condition
 
 I So THE SCIENCE OF ETHICS. 
 
 of social vitality^ and therefore the internal relations of the 
 tribe and the various propensities of its members must have 
 been developed with reference to this condition. It has 
 always been a prominent condition, as in early times it was 
 the most conspicuous and predominant condition; but as 
 peaceful instincts have developed there must have been some 
 correlatinir and harmonisino; influence. The militarv instinct 
 is not necessarily incompatible with the industrial, but at 
 any given period the one may be developed to the prejudice of 
 the other. Where the race has been constantly, though in 
 part unconsciously, occupied upon the great problem how to 
 reconcile the two, and to secure at once efficiencv in war and 
 efficiency in peace, one quality or the other may be in excess ; 
 as there are races which are easily conquered, but which have 
 great capacity for thriving and extending themselves when 
 not encountered by enemies, and others in which the military 
 spirit is so strongly developed tliat, although they can resist 
 direct attacks, they are weak in adapting themselves to the 
 material conditions of life. The races which survive are 
 those in which there has been such a development that, on 
 the whole, a maximum of efficiency has been reached by the 
 best adaptation to divergent, though not naturally antagonistic, 
 conditions of development. The distinction appears in the 
 internal relations of any community; for as peaceful relations 
 become more prominent, it is evident that excellence of the 
 militarv kind may be combined with bad qualities of another 
 kind — that the bravest man or the best soldier may be lazy, 
 dissolute, or tyrannical in his other relations, and that we 
 must therefore substitute some more discriminating: mode of 
 judgment for that which was previously sufficient. 
 
 8. Hence arises the problem whether courage can be con- 
 sidered as a moral quality at all ; for it mav be as well to make 
 the remark, which will be frequently exemplified, that although 
 we speak of the moral law as though it corresponded to a 
 perfectly distinct mode of thinking and feeling, we are by no 
 means entitled to assume that the actual demarcation is so 
 sharp as our use of language suggests. In fact, we shall find
 
 VIRTUE OF COURAGE. iSi 
 
 that it is often exceedinglv difiicult to decide at what point 
 we are to trace that special shade of feehng which may be 
 called distinctly moral. The difficulty is the greater because 
 there is no reason to suppose that the same sentiment exists 
 in all members of a eiven society. The feelings with which 
 they regard an admittedly wrong action may vary greatly 
 according to individual idiosyncrasies, even though they agree 
 in condemning the same actions or admiring the same quali- 
 ties of character. This appears to be the case in the present 
 instance. When asked whether courage is or is not a moral 
 quality in the strictest sense, we must reply that at some 
 periods it has been considered as not only a virtue, but the 
 typical and cardinal virtue, whilst at others it begins to be 
 more or less doubtful whether it is, properly speaking, a 
 virtue at all; while, again, very different answers would pro- 
 bably be given upon this point by different classes of persons. 
 9. Remembering this, let us ask what are the facts as to 
 the existino; social code. If coura(i;e is intrinsically virtuous 
 — if, that is, the bare fact that a man is brave entitles him to 
 be called virtuous so far — we should have to admit that every 
 manifestation of courage was virtuous, and we should call a 
 man o;ood because he met a ticjer unfiinchingly when he was 
 simply engaged in sport. This, I think, would be rather a 
 strained use of lan^uacre, and we should decline to admit the 
 goodness unless the quality was exerted for benevolent pur- 
 poses; as, for example, in saving the life of one of the tiger's 
 victims. But in the latter case it seems that it is not in 
 respect of his courage that the man is called virtuous, but in 
 respect of his unselfishness or benevolence. Thus, to take a 
 further case, we feel it to be a perfectly justifiable form of 
 expression when Clarendon speaks of Cromwell as a " bold, 
 bad man." Courage, we admit, may be combined with 
 objectionable qualities, such as tyranny and hypocrisy. Shall 
 we then consider courage is to be in itself a morally 
 neutral qualitv, which is good or bad according to the other 
 qualities with which it is associated, and thus as simjily an 
 intensitive? In that case, we ouo:ht in consistency to dis-
 
 1 82 THE SCIENCE OF ETHICS. 
 
 like the bad man more in proportion to his courage, in so 
 far as it makes him a more mischievous person. This 
 certainly does not represent our ordinary mode of feeling. 
 The admission of courage qualifies our dislike. We respect 
 the brave man, so far as brave, not only if he is an enemy by 
 virtue of accidental circumstances, but even if he is a bad 
 man by the intrinsic quality of character. Thus, if our 
 judgment of Cromwell coincides with Clarendon's, we feel 
 that our moral disapproval is tempered by a certain admira- 
 tion; we feel that, as brave, he is less hateful, even though he 
 may be more mischievous than if he had combined his other 
 bad qualities with cowardice. We regret that he had not 
 better principles instead of regretting that he had so much 
 vio;our. This is virtually to admit that in so far as a man is 
 brave he is approved by the general feeling, although the 
 approval is not exactly of the kind which we call moral. If 
 we speak of some distinctly moral quality, the implicit 
 reasoning would be different. Charles I., let us assume, was 
 chaste but tyrannical. In so far as he was chaste he deserves 
 moral approval, and in so far as tyrannical, moral disapproval. 
 We have to settle the balance by conflicting considerations. 
 But in the case of a Cromwell, the respect which we feel for 
 his courage does not serve to qualify the moral verdict, but 
 to represent a feeling which is intermediate between that of 
 moral approbation and simple admiration for an endowment, 
 physical strength and beauty, for example, which has no 
 definite relation to moral judgment at all. It may be just 
 worth while to add, that the difference cannot be explained 
 by saying that courage is a simple property which does not 
 involve " free will," or is incapable of being modified by the 
 approval of others. How far this consideration affects our 
 moral judgment need not be considered here, inasmuch as it 
 obviously does not apply to this case. There is no quality 
 which is more imminently amenable to opinion. The special 
 characteristic of the warrior has always been taken to be his 
 thirst for <rlory, and the fact that courage can be generated 
 by discipline is one of the most familiar of observations.
 
 VIRTUE OF COURAGE. 183 
 
 10. It is, in factj easy to put a logical interpretation upon 
 the verdict of common sense; for it virtually asserts that 
 courage is a quality which is useful to society; that it there- 
 fore forms a part of the good man's character; and thus 
 that to call a man a good coward involves some inconsis- 
 tency, although it is not inconsistent to speak of a bad brave 
 man. Hence courage may be regarded as one of the neces- 
 sary conditions of ideal excellence, though its possession does 
 not necessarily imply the fulfilment of any other conditions. 
 To be good, you must be brave ; but you may be brave 
 without being good. And therefore when we see courage 
 united to wickedness, we reo;ard the a2:ent as in some degree 
 qualified for our approval, though for a kind of approval 
 differino; from that reserved for what we call virtue in the 
 most eminent degree. And the principle involved is ob- 
 vious, namely, that courage does not imply necessarily a 
 use of faculties for the good of others. In rude states of 
 society, as we have seen, this is necessarily (or almost neces- 
 sarily) the case, inasmuch as the individual is so closely iden- 
 tified with his tribe, the objects of his animosity are so certain 
 to be the objects dangerous to his fellows, that to be brave is 
 to be socially useful. But as society develops we see that 
 the necessity is external ; that is, dependent upon a state of 
 things which does not always exist in the higher stages ; and 
 hence we make a distinction, and whilst still admiring 
 courage, we scarcely regard it with that specific kind of 
 admiration reserved for qualities which, by their verv nature, 
 must be useful to others than the ag-ent. 
 
 11. Let us now consider for a moment the nature of the 
 process thus described. It supposes the development of a 
 complex organisation and the growth of correlative instincts. 
 The primitive society depends at every instant upon the 
 military prowess of its members. Respect for the good 
 fighter and contempt for the bad, respect, therefore, for 
 courage and contempt for cowardice, as the most obvious 
 conditions of good and bad fighting, appear as soon as the 
 simplest reflection is possible. And this sentiment tends to
 
 1 84 THE SCIENCE OF ETHICS. 
 
 persist, however formed originally, inasmuch as every one 
 from childhood imhibes the habitual modes of thought and 
 feeling of the social medium. But social development im- 
 plies a process of at least implicit reasoning, which modifies 
 the early assumptions. The chivalrous sentiment grows as 
 hostilities cease to be internecine and sympathy extends 
 beyond the limits of the community. Courage, again, takes 
 a different form as men come to reco2;nise the distinction 
 between the higher and lower forms of the quality, and feel 
 the advantage of discipline and subordination. Again, as 
 the demand for warlike excellence declines, courage is valued 
 so far as it can be combined with the full development of 
 peaceful eneroies and instincts. Finally, as men come to 
 distinguish between the qualities which are essentially useful 
 to society and those which may be combined with anti-social 
 qualities, they regard courage with a sentiment differing from 
 that which is reserved for the more directly social virtues. 
 The process may therefore be regarded as a prolonged induc- 
 tion, starting by a condemnation of certain actions plainly 
 mischievous to the society, and ending by the recognition of 
 a certain type of character as admirable, which includes 
 amongst its attributes a capacity for the conduct actually 
 admired, but which also includes fitness for conduct of a 
 different kind. Courage is now regarded as one manifestation 
 of a character which is fitted for all the requirements of social 
 existence. But as it may be turned against society, we scarcely 
 regard the brave man as therefore virtuous, but rather as 
 satisfying one of the conditions necessary to virtue. Thus 
 the process implies the elaboration of a more complete defini- 
 tion of the type necessary to the constitution of a vigorous 
 society, and thus the recognition of certain external rules of 
 conduct, — that, for example, which prohibits running away 
 in battle, as defining particular manifestations of the typical 
 character under specified conditions, and deriving their validity 
 from the more general condition of social fitness. 
 
 13. Hence this statement suggests a question which requires 
 a distinct answer. Courage in all its phases is a condition
 
 VIRTUE OF COURAGE. 185 
 
 of social vitality. The spirit in which a race confronts danger 
 must be always one criterion of its power of holding its own. 
 But we may still ask how far the approval is generated by a 
 perception of this fact. Is courage admired simply because 
 it is useful or because it is seen to be useful ? We have 
 obviously no right to assume a perception in all cases. The 
 average man accepts without reflection the standard of his race. 
 The warrior may hate the coward as he hates the harmless 
 necessary cat, without seeking a reason for his aversion. We 
 may therefore explain the sentiment in any given case by 
 savinor that the man has derived it from his neighbours. But 
 thisj of course, is no explanation of the whole phenomenon. 
 To explain this we must go back to the history of the race. 
 The courage of an animal is "explained" by the fact that it 
 is necessary under the conditions of his life. The courage 
 of the bulldog is a blind instinct; the blindness implying, 
 not that he is without feeling of pain, or even perception of 
 danger, for otherwise we could hardly speak of him as brave, 
 but that danger or pain does not affect him as it affects a 
 cur, and that he holds fast when the cur lets go. Such an 
 instinct may of course exist, and must exist, in the unreason- 
 ing animal in the absence of any reasoned calculation of 
 consequences. The creature survives by reason of his instinct, 
 but not because he forms any conscious judgment of the 
 advantages of this mode of conduct. And when we rise to 
 the reasoning agent, we may still ask whether the utility of a 
 given character implies, or in what sense it implies, a percep- 
 tion of the utility? The distinction requires to be noticed, 
 inasmuch as certain ambiguities result from a neglect to take 
 it into account. 
 
 13. Now it seems necessary to suppose that races owed 
 their survival to military prowess when reflection was still in 
 the most rudimentary stage. The utility of courage, indeed, 
 must have been a very obvious discovery as soon as any 
 reflection became possible. No condition of the preservation 
 of a community could be so palpable or pressed by such 
 constant and repeated experience upon the attention of its
 
 i86 THE SCIENCE OF ETHICS. 
 
 iiienibers. They could see every day that their existence 
 depended upon the readiness to confront danger. But the 
 quahty must have existed in some degree before it could be 
 discovered, and therefore it would be inaccurate to speak as 
 though the approved of courage were the result of an antecedent 
 perception that certain advantages were attainable by courage. 
 Men came to recognise the advantages already obtained by 
 the brave, and the recognition tended to intensify as also to 
 modify the sentiment already implicitly existing. Every 
 increase of reasoning power would bring out more fully the 
 importance to the race and to the individual of courage, 
 coolness, and resource in danger. The existence of 'any 
 distinct moral sentiment doubtless implies some reflection, 
 and there could hardly be a social approval or disapproval 
 until men were capable of observing consequences. The 
 approval may be, in that sense, the product of the per- 
 ception ; but, on the other hand, the perception implies 
 that the instinct approved was already in existence. The 
 race has a certain deg:ree of coura^'e, and must be so constituted 
 that, in point of fact, the courage of one is useful to others, 
 before the conscious recognition of the fact, and before the 
 definite emergence of a social approval. The difference is 
 that in the higher stages of development the fact becomes a 
 recotrnised fact. The other instincts are not only interested, 
 but their interests are perceived. Formerly they were interested 
 as slaves may be interested in the success of their owner ; now 
 as free men whose opinions have to be consulted. As reflec- 
 tion develops, men come to have a wider perception of the 
 bearing of their conduct in every direction, and this per- 
 ception modifies the actions and instincts ; but it must 
 always start from a pre-existing organisation of character 
 and society. 
 
 14. And hence we mav state more precisely the sense in 
 which the notion of courage is a statement of the general 
 conditions of social vitality. The problem is more complex 
 than appears at first sight. It has, we may say, a double 
 aspect. The " utility " of courage, meaning by utility the
 
 VIRTUE OF COURAGE. 187 
 
 fact that the braver race is so far the better qualified for the 
 struLi'o-le of existence, is in all cases the cause of the survival- 
 of the instinct. That is equally true whether we speak of a 
 blind or a reasoned instinct, of the bulldog or of the hero, 
 and gives the general principle applicable at every stage of 
 the process of evolution. The principle applies to the parti- 
 cular case of the reasoned instinct developed through the 
 social factor, and sioverns it as the general rule governs the 
 special application. The instincts of any race must comply 
 with the conditions of its existence, and this must be true 
 as well of the instincts which imply a perception of utility. 
 The approval of any conduct as useful must itself be useful, 
 if the race is to survive. 
 
 15. The necessity of calling attention to this distinction is 
 already manifest. For, in the first place, the conditions do 
 not at present appear to be necessarily coincident. Courage, 
 we say, is useful; or is, in other words, a condition of the 
 existence of the race. So far as the race becomes conscious 
 of this fact, the same condition applies to this consciousness. 
 The consciousness, that is, that courage is useful to the race 
 must stimulate courage. Now it may appear that this is 
 not a necessary result of .increased intelligence. If, in fact, 
 we suppose the intellect to become keener without any 
 increased desire for social welfare, which is clearly possible in 
 individual cases, the clearer perception of the social utility of 
 courage could onlv stimulate it in cases where the interest of 
 the individual was identical with that of his fellows. It 
 might then, on the average, produce cowardice, or a desire of 
 each to save himself whenever he could do so at the expense 
 of his fellows. In that case, increased intelligence would 
 operate so far to the disadvantage of the race, as the stupider 
 and therefore bolder races would have a better chance of 
 survival. Something of this kind seems occasionally to 
 happen. Looking, indeed, at the facts, and observing that 
 the more intelligent races have an advantage, and not least 
 in this quality, we must infer that the general principle 
 operates differently. Evolution affects the whole nature 3
 
 l88 THE SCIENCE OF ETHICS. 
 
 and we may suppose that those races are most successful 
 which are so constituted that a perception of the vitaHty of 
 courage goes along with an increase of courage. This, how- 
 ever, seems to imply an additional condition. The race 
 must not simply become more intelligent, but become more 
 intelligent in such a way that its social qualities improve 
 along with its intellectual. The bare extension of a percep- 
 tion of consequences might of itself tend only to more 
 calculated selfishness, and therefore a diminished disposition 
 to fulfil the conditions of social welfare. It may appear that 
 in the normal case, at least, the intellectual advance implies a 
 development of the social sympathies; but that cannot be 
 taken for granted at present. 
 
 i6. There is, however, in the second place, a further con- 
 sideration. It is clear from the nature of the process described 
 that there must be a close coincidence between the qualities 
 which are valued and those which are actually useful. Each 
 man and generation of men starts with a sentiment^ elabo- 
 rated bv the race and embodied in the accepted rules of 
 conduct. The most important are worked into the very struc- 
 ture of the lantruaofe and inG;rained in the organic customs 
 of society. Respect for courage, in particular, has become 
 part of the general inheritance and profoundly influences 
 the character of every member of a society which has passed 
 through a military stage. Now, on the one hand, it is clear 
 that any sentiment thus transmitted through the social factor, 
 must have a reference to the conceptions of social vitality. 
 All men could not agree to admire courage unless they thought 
 that courage was useful, — useful, that is, in the second or 
 subjective sense, namely, of generally tending to procure 
 happiness. The process by which any such sentiment is 
 transmitted and diffused, so as to be part of the general in- 
 heritance of the race, is a sufficient guarantee for this. Men 
 may agree to admire qualities which make their possessor 
 unhappy, but not at once to admire a quality and to think 
 that it diminishes happiness in general. The perception 
 that a quality is in this sense useful ov feUcjJic must nearly
 
 VIRTUE OF COURAGE. 189 
 
 coincide with, if it docs not generate, the admiration felt for 
 the quaUty; and, on the other hand, it is clear that the 
 actual utility of courao;e will n;enerate admiration as soon as 
 the fact is perceived. If the existence of a tribe really depends 
 on the courage of its members, it must become apparent that 
 in numberless special cases courage preserves the lives and 
 happiness of all, whilst cowardice is fatal. What is useful as 
 preservative must also be useful in the sense of happiness-giving. 
 Within certain limits men mav reason badlv, and even the 
 collective reasoning may be bad. Courage may be the object 
 of an extravagant admiration from the survival of prejudices 
 generated under special conditions, or a temporary absence of 
 danger may lower the estimate; but any wide deviation from 
 the correct judgment will be rectified, if not by the reason 
 of the society, then by the operation of the general principle 
 of utility, and the tendency of the erring society to disappear. 
 This principle, therefore, must be regarded as working, not 
 only through the blind and more or less unconscious instinct 
 of the lower races, but upon the conscious and reflective 
 judgments of the most highly organised society. In an ideal 
 state there would be an absolute coincidence. The society 
 would not only possess those instincts which are necessary to 
 its survival and vigour, but would understand precisely how, 
 in what respects, and under what conditions or limitations, 
 they were necessary; whilst its judgment of what was useful 
 as giving happiness would precisely coincide with its judgment 
 of what was useful as preservative of its existence. 
 
 17. These remarks are of course applicable, mutatis mutandis, 
 not only to courage, but to other qualities, such as industry, 
 energy, and so forth, which belong to the same class; and, 
 moreover, as we shall presently see, to the virtues in general. 
 And it may be as well to make one remark before finishing. 
 The ultimate and governing principle is in all cases the 
 utility of the quality, in the sense in which utility means fit- 
 ness for the conditions of life. The scientific explanation, at 
 any rate, cannot get beyond an exposition of the essential im- 
 plication of courage in the general conditions of social life.
 
 I90 THE SCIENCE OF ETHICS. 
 
 And this general principle may operate through an nnreason- 
 ing or a reasoned instinct; that is, through the instincts of a 
 beino- conscious or unconscious of their value. Now, although 
 morality proper is only possible when reflection and a power 
 of o;rasping general rules has been developed, this has still an 
 application even to the higher stages of social growth. For 
 a man, or even a race, may be content to accept the instinct as 
 justifying itself without seeking to discover its origin. And, 
 again, a moral development may be brought about either by 
 a reasoning or unreasoning process; either bv a process of 
 generalisation and the conscious aim at some social ideal 
 implying an accurate perception of the conditions of life ; or, 
 on the other hand, bv some " accidental " change, that is, by 
 a chano-e orio-inatino; in other conditions than those of in- 
 tellectual 'growth, as when the extension of an empire forces 
 peaceful habits upon previously hostile tribes, and so favours 
 the growth of tastes and pursuits previously impossible. I 
 need not now inquire which is the normal and most important 
 case. 
 
 III. Virtue of Temperance. 
 
 1 8. Proceeding to the virtues typified by chastitv and 
 temperance, we have first to remark that they occupy an 
 intermediate position between the virtues of strenjrth and the 
 directly social virtues. Some of them are a part of the 
 prudential, and others of the directly moral code. Thus, for 
 example, it is prudent to be temperate because temperance 
 is necessarv to health. Anv rules deduced from that con- 
 sideration are primarily concerned with the individual, and 
 interest the society through the individual. The primary 
 objection to drunkenness is that it injures the constitution, 
 and if I prove from purely medical considerations that certain 
 drinks are injurious and others innocuous, the rule deduced 
 is a law of prudence and consequently a part of morality. It 
 is my dutv, it mav be said, to maintain my health unless 
 some other moral duty accidentally conflicts, and thus the 
 rules for framing health may be taken up into the moral
 
 VIRTUE OF TEMPERANCE. 191 
 
 code ; but they are deduced without reference to the society 
 from the simple principle that whatever increases the health 
 of each man is so far a matter of prudence. The society 
 composed of healthy men is so far better^ but the rule is 
 deduced from a direct consideration of the individual, and 
 the sanitarv rule coincides with, because it is assumed by, 
 the moral rule. That is right which is health-preserving. 
 But another class of these virtues is directly social. In fact, 
 society is dependent from its very germ upon the sexual and 
 parental instincts j its most intimate structure depends at 
 every moment upon the way in which they are regulated ; 
 and conduct, which from the sanitary point of view may be 
 perfectly indifferent, may have the most important effects 
 upon the social organisation. Temperance in eating and 
 drinking is a dictate of simple prudence, which has no need 
 to take into account the consequence to others; but chastity 
 is a virtue which is only intelligible when we take into account 
 its bearing upon the vitality of the social organism. 
 
 ] 9. Hence we have a moral evolution similar to the former. 
 Chastitv, that is, some modification of the sexual instincts 
 imposed by the social sentiment, is implied in the earliest 
 sta<i;es of social growth. Society is constituted by the develop- 
 ment of an organic custom which defines, amongst other 
 things, the relations of the sexes. The marriage law of primi- 
 tive times differs in strange and unexpected ways from that of 
 civilised races, but it implies at every step the existence of 
 some restraining sentiment. Sexual rivalry is the great cause 
 of hostilitv araoncrst Yahoos; whilst the state and the familv 
 are barely distinguished, the disputes about women are a 
 main cause of war. The formation of the social bond implies 
 the formation of some modus vivendi in regard to such 
 matters, and a complex system of rules is developed at a very 
 early date. Hence the external rule is formed before a more 
 general theory of the passions has obtained acceptance. The 
 first law is, " Thou shalt not commit adultery." Little atten- 
 tion is aroused by conduct not so clearly injurious. The 
 savage sees no evil in gluttony, and intoxication appears to
 
 192 THE SCIENCE OF ETHICS. 
 
 him as a new pleasure revealed by the civilised man. In the 
 higher law the rule is extended, and applies to the motive as 
 well as to the specific action. Lust is condemned as well as 
 adultery. The man who would sin if he could is seen to be 
 as objectionable as the man who sins because he can. And 
 the condemnation extends itself to all forms of sensuality in 
 virtue of the general principle thus elaborated, as their effect 
 upon the character of the agent and their remoter conse- 
 quences to others are better understood. 
 
 20. To the savage, as to the animal, gluttony scarcely 
 appears to be objectionable. The conduct of either is regu- 
 lated by his appetite without any conscious reference to 
 general principles. The force and the nature of these appetites 
 are determined by the general principle of utility. The 
 savage, like the dog, eats when he is hungry and drinks when 
 he is thirsty without regard to consequences. If his appetites 
 were such as to be inconsistent with his permanent survival, 
 he would die out. Therefore his instincts are correlated by 
 a process which has nothing to do with conscious reflection. 
 But when the man acquires new powers by incipient civilisa- 
 tion, the conditions of his life are altered. He is able if he 
 pleases to eat and drink more than is good for him. If the 
 old instincts survived and were not restrained, he would 
 speedily eat and drink himself to death. They may grow 
 weaker in virtue of some physiological principle, but they 
 may also be held in check by his clearer perception of the evil 
 consequences of indulgence. In the latter case we have the 
 arowth of a moral and a prudential law. The reasoning man 
 does not require, we may sav, to lose the appetite because he 
 is able to counterbalance its action by calling other forces 
 into plav. He sees that the present gratification will have to 
 be paid for at a certain rate of future pain. But the moral 
 quality of the rule only becomes marked when there is 
 some recognition of the effects to others than himself. Now 
 it becomes evident that sensuality generally has a direct bear- 
 ing upon the social qualities, even in those forms which are 
 primarily objectionable on prudential grounds. Courage and
 
 VIRTUE OF TEMPERANCE. 193 
 
 enersrv crenerallv are of course clearly connected with tern- 
 perance; and so far as courage is regarded as a virtue, the 
 slothfulness and indifference which spring from all forms of 
 intemperance incur a share of the contempt bestowed upon 
 the quality which is their natural fruit. But, on the other 
 aspect, they have a more intrinsic bearing upon social vitality 
 than the virtues of streng-th. The brave man may still, as I 
 have said, be a thoroughly bad man ; he may use all his 
 energy to oppress his neighbours. To some extent this is 
 also true of the temperate man, but the connection with 
 the social qualities is more direct and invariable. So far 
 as a man is a drunkard, he is almost necessarily a bad 
 husband and father; so far as temperate, he is free 
 from some of the weaknesses which must generally dis- 
 qualify him from acting in that capacity. Courage and 
 energy may often be essential, and are generally useful, in the 
 discharge of the duties resulting from every social relation, 
 but they are not so closely implicated, and for a reason which 
 becomes obvious as men become more intelligent. Sensuality, 
 that is to say, implies selfishness. A man's love of his 
 bottle is so much deducted from his love of his wife and 
 children. So far as he is taken up with the gratification of 
 his appetites, there is less room for the development of his 
 affections. A coward and a sluggard may be affectionate, 
 though his affection will be comparatively useless to its 
 objects, but in a sensual character the affections are killed 
 at the root. He is incapable of really loving as well as of 
 beincj useful to those whom he loves. And thus we may ex- 
 plain the moral verdict passed upon certain kinds of conduct 
 which, on the ordinary judgment of common sense, are classi- 
 fied as " self-regarding." The most conspicuous evils caused 
 by some actions are produced upon the man himself. People 
 are inclined to regard them as breaches of prudence rather 
 than of morality. The intemperate man, according to the 
 common phrase, is an enemy to no one but himself. We 
 have, as we fancy, no right to object to him so long as he 
 only makes a beast of himself in private. In some cases, 
 
 N
 
 194 THE SCIENCE OF ETHICS. 
 
 again, this leads to the production of a kind of spurious 
 moral code. People admire the " good fellow " as thev 
 admire the spendthrift, because the immediate and obvious 
 consequences of his conduct are agreeable to his neio;hbours. 
 Convivial relationship generates a sort of law of its own, in 
 virtue of which the person who drinks fair and takes his 
 wine like a man is admired, partly because his prowess is taken 
 as a proof of vigour, and partly because he is " sociable " in 
 the lower sense of the word, and is therefore assumed to be 
 sociable in a higher sense. As the reason develops it be- 
 comes clear that both the spendthrift and the drunkard are 
 really mischievous, and that the prevalence of the conduct 
 which they practise is a mark of social decav. And, more 
 generally, it appears that we make a false estimate when we 
 confine our attention to the immediate consequences of a 
 man's actions. It is impossible really to maintain the dis- 
 tinction assumed. A man whose vice injures only himself 
 in the first place, becomes also by a necessarv consequence 
 incapable of benefiting others. If he is an enemy to himself 
 alone, he is also a friend to himself alone. The opium-eater, 
 for example, paralyses his will ; so far as he becomes in- 
 capable of energetic action, he is unfitted for every social 
 duty, and so far as he becomes the slave of his appetite, 
 becomes also unfitted for the social sentiments. Thus a 
 distinctly moral sentiment exists, and its verdict is justified 
 by logic in cases where the primary effect of conduct tells 
 upon the agent himself. In some cases, the feeling of 
 disgust and abhorrence is perhaps more strongly excited bv 
 such vices than by others more directly anti-social. And 
 this is justified by the consideration already explained in 
 a different connection. As we condemn the man whose 
 character is bad, whether external circumstances do or do not 
 give him an opportunity for displaying it, so we object 
 logically to the man who is destroying his social qualities, 
 whether the immediate effect of his conduct tells upon him- 
 self or upon others. He must be defective in characteristics 
 essential to the moral type.
 
 VIRTUE OF TEMPERANCE. 195 
 
 21. So far the case seems to be clear; that is, the moral 
 law does in fact come to condemn those qualities in respect 
 of which the individual deviates from the type prescribed 
 by the conditions of social welfare. But there is another 
 characteristic of the virtues of temperance which requires a 
 little more consideration. In the first place, the disgust 
 which we feel for some conduct seems to be inadequately 
 explained by the considerations just offered. Undoubtedly 
 the condemnation of sensuality goes along with a percep- 
 tion of the many and complicated social evils which spring 
 from it, and implies a recognition that a tendency to such 
 indulgence is a fatal symptom of social decay. But the 
 feeling of disgust seems to be a direct product of what we 
 may call physiological causes. As the intellectual sensibility 
 becomes keener, a pure-minded person is revolted by the 
 sensual indulgence of the grosser, and resents the attempt to 
 explain his disgust by a simple perception of the mischiefs 
 produced. The feeling which exists in a healthy state of 
 society is so strong and immediate that it seems to precede 
 any utilitarian calculation ; and I may add that, as a matter 
 of fact, utilitarians have found a special difficulty in account- 
 ing for or justifying the strength of the prevailing sentiment, 
 and are sometimes inclined to relax the severity of the code. 
 Many people object to any attempt even to discuss the 
 question, partly because they feel that there is a moral danger 
 in even talking about certain subjects, and partly, perhaps, 
 because they are afraid that their prepossessions may not be 
 fully justified. 
 
 22. This, again, seems to be connected with another pecu- 
 liarity of the moral judgment in this direction. I speak of 
 the ascetic theories which at different times acquire so much 
 strength, and which may in some forms appear to be a nega- 
 tion of the very principle of morality as I have stated it. 
 In some cases the natural morality of a race appears not only 
 to condemn intemperance — that is, an excessive addiction to 
 sensual pleasures — but condemns sensual pleasure in itself. 
 It regards every pleasure of the kind as in itself bad^ although
 
 196 THE SCIENCE OF ETHICS. 
 
 indulgence may be permitted under certain restrictions as a 
 safecvuard against worse evils. Now if this were to be taken 
 in the fullest sense, it is plain that it would condemn conduct 
 which is not only productive of pleasure, but which is abso- 
 lutely essential to the preservation of the species; and, in fact, 
 men are admired and regarded as entitled to special reverence 
 for a complete abstinence from some of the functions on 
 which society depends. The virgin and the hermit are re- 
 garded as admirable because they withdraw from direct parti- 
 cipation in social and domestic pleasures and duties; and 
 thus, though moralists have laid it down as a universal rule 
 of morality that every man should act so that his conduct 
 may be a rule for all, yet some moralists regard action as 
 specially admirable which cannot be a rule for all. It would 
 be easy, of course, to reply that such morality is really im- 
 moral ; that it implies erroneous reasoning, and would there- 
 fore be condemned by the right reason; or that, as it would 
 diminish happiness, it cannot be conducive to the greatest 
 happiness, and is therefore not moral according to the proper 
 definition of the word. But, from my point of view, this is 
 to evade the question, inasmuch as I find it impossible to 
 deny that such sentiments, whether right or wrong, mis- 
 chievous or beneficial, are a part of actual morality — that is, 
 of the fundamental rules of conduct impressed upon the 
 individual by the race — and therefore require to be explained 
 instead of being simply denounced. 
 
 • 23. There is this difficulty in every explanation, that it is 
 by no means easy to say what is the actual judgment in such 
 cases. The common opinions often differ widely from those 
 of authorised expositors, and there are innumerable devices 
 by which doctrines apparently absurd may be accommodated 
 to the dictates of common sense. This much, however, may 
 perhaps be taken for granted : the ascetic theory is the pro- 
 duct of a luxurious state of society — that is, of a state in 
 which the abuses of sensual indulgence have become specially 
 prominent. The growth of a rich and powerful class which 
 uses its power exclusively for its own enjoyment, in which
 
 VIRTUE OF TEMPERANCE. 197 
 
 great men plunge themselves into sensuality with indifll-rence 
 to the sufferings of their dependants, suggests the doctrine 
 that sensuality is the great enemy of mankind, and naturally 
 suggests an exaggerated statement of the advantages of the 
 opposite character. The hest teachers see that the passions 
 are strong enough, so to speak, to take care of themselves ; 
 there is not the least danger that men will be too little sen- 
 sitive to downright animal enjoyment; and therefore they 
 denounce that evil unsparingly, and think that they can cut 
 it up bv the roots by denouncing the source of mischief, 
 without supplying those qualifications which will be suffi- 
 ciently supplied by the facts. 
 
 ' 24. Further, it is to be remarked, there seems to be a kind 
 of tacit recognition of the flict that the principle has a limited 
 application. The virtues are divided into two classes — the 
 ordinary and the heroic, A man is held to have fulfilled the 
 law if he is chaste and temperate enough not to indulge in 
 adultery and intoxication ; but he has done something more 
 if he is so chaste and temperate as to suppress all indulgence 
 of one kind, and to confine the other within the limits neces- 
 sary for the support of life. He is then not only good, but 
 good enough for two. This, again, is implied in the theory 
 of merit. Whatever else may be intended by that word — 
 which will have to be considered hereafter — it recognises in 
 some shape a certain normal standard applicable to every 
 man, but which some men may transcend, and by transcend- 
 ing acquire merit. Now in the theological doctrines generally 
 associated with the theory it is supposed that the merit thus 
 acquired is in some way useful to the race. It propitiates 
 the gods, and makes them look more kindly upon the followers 
 of the saint. The man, therefore, who possesses heroic virtue 
 is not anti-social, althou<rh his conduct is such as, if universally 
 adopted, would destroy society. He is rather to be compared 
 to a man who sacrifices life and health in the pursuit of some 
 philanthropic purpose, and in this sense his conduct might be 
 approved by other moralists. We may, for example, think 
 that a Howard is admirable, although he neglects duties
 
 198 THE SCIENCE OF ETHICS. 
 
 which must be performed in the ordinary case. He does 
 more good in this particular instance by improving the con- 
 dition of prisoners than by attending, for example, to his own 
 affairs. This, however, merely admits that the general moral 
 rule of benevolence prescribes different conduct according to 
 a man's special opportunities and talents. So it might be 
 admitted by all moralists that in special cases a man of pecu- 
 liar temperament might find it right to suppress instincts 
 which should generally be allowed to regulate his conduct. 
 The difference would only be that such suppression would 
 not be re2;arded as unconditionally rioht, or as establishin<r 
 by itself, a title to superior respect. 
 
 25. It seems, therefore, to be quite possible to express the 
 ascetic theory in a form which is consistent with a reasonable 
 conception of social needs. We have, in fact, only to say 
 that, in a condition of society characterised by sensuality and 
 selfishness, the existence of men capable of refusing all sensual 
 gratification may be of the highest value, if only as a demon- 
 stration of the possibility of conquering the prevailing passions. 
 Further, it may well be that in such a state the best, and, 
 so far, the most meritorious men, and those who are most 
 sensitive to the social evils of the time, may show by their 
 lives the possibility of such conquest, and may, so far, deserve 
 the respect of their fellows. As the Red Indian proved his 
 courage by submitting to tortures, the saint may prove his 
 temperance and chastity by refusing all indulgence; and 
 such examples of transcendent virtue may tend to raise the 
 average standard. They are a forcible protest in flesh and 
 blood against the tyranny of the grosser natures. It is need- 
 less to dwell upon the dangers of the doctrine, upon the 
 erroneous assumptions with which it is associated, or upon the 
 limitations within which it should be confined. I only say 
 that it is so far perfectly compatible with a view deduced 
 from general considerations of social welfare, and that it 
 errs in attributing an absolute value to conduct valuable only 
 under certain conditions. So lono; as these conditions are 
 fulfilled in fact, the error of neglecting their importance will
 
 VIRTUE OF TEMPERANCE. 199 
 
 not lead to immediate error ; and thus an instinct which 
 seems to neglect all reference to the conditions of social wel- 
 fare mav, in fact, be the direct product of those conditions. 
 It corresponds to the disgust produced by certain evils which 
 arises spontaneously, and does not imply an accurate recog- 
 nition of the true nature of the evils, and thus condemns 
 absolutelv what is only relatively bad. 
 
 26, Hence follows a conclusion which has been already 
 noticed, namelv, that the general condition of the utility of a 
 moral sentiment cannot be identified, except under certain 
 limitations, with the condition of a perception of utility of 
 the corresponding rule. Thus, when we speak of the disgust 
 produced bv certain conduct, and proceed to ask whether the 
 distrust can be justified by the presumed consequences, we 
 sometimes fall into a perplexity. Are we or are we not to 
 take into account as amongst those consequences the disgust 
 itself? Is the offended faculty entitled to a seat on the bench, 
 or is it only entitled to a hearing as a witness ? In the former 
 case, we seem to be falling into a circular argument and 
 makino- the instinct its own justification. In the latter, we 
 may ask whv this particular kind of pain should be less re- 
 garded than any other in estimating the total consequences. 
 The answer, I think, is simple from a right point of view. 
 When we say that conduct or instinct is bad, we clearly make 
 one assumption, namely, that it can be eliminated, and that 
 the conduct or instinct which takes the place of that con- 
 demned "will be better. When I say that vice is mischievous, 
 I must mean that, on the whole, and taking in everything 
 implied in a change assumed to be possible, the temperate 
 man is better fitted for social duties, or, which is the same 
 thincr, that a society formed of the temperate is better adapted 
 for persistence than a society formed of the vicious. It is a 
 manifest fallacy if I compare the two things only in respect 
 of a particular set of differences which happen to be con- 
 spicuous on the surface, or if I fail to make a real comparison 
 between possible states, and therefore make a false assumption 
 as to the true alternatives.
 
 200 THE SCIENCE OF ETHICS. 
 
 27. Now, in the case before us, let us consider some vice, 
 sav gluttony, which is specially contemptible, but which 
 would perhaps generally be classed as self-regarding, inasmuch 
 as it does not imply the existence of malice and direct attacks 
 upon others. To estimate the evil of gluttony fairly, I must 
 compare the gluttonous man, including all that is implied in 
 glutton, with the temperate man. If I try to sum up the 
 consequences of gluttony, I shall probably think first of the 
 evils to health, of the consequences in the shape of gout, 
 indigestion, and so forth. But this gives a very imperfect 
 measure of the social evils of gluttony. The difference be- 
 tween the glutton and the temperate man is not that one is 
 more exposed than the other to certain diseases, or that in 
 consequence of the diseases he is less capable of strenuous 
 activity. It is also that the man who is a slave of his belly 
 is less capable of all the higher affections, of intellectual 
 pleasures or aesthetic and refined enjoyments, and presum- 
 ably selfish and incapable of extensive sympathies. If^ then, 
 my disapproval of gluttony is measured by the first set of 
 consequences alone, it is manifestly inadequate. It is as 
 though my estimate of the evils due to a disease were con- 
 fined to certain special symptoms, and not to the total con- 
 stitutional state implied. As a matter of fact, the senti- 
 ment actually entertained for the gluttonous man includes 
 an element which is not measurable by the estimate of the 
 imprudence implied in gluttony, nor of the general incapacita- 
 tion which may be the most conspicuous result. It is rather 
 an intrinsic feelino; due to the conflict between the hioher and 
 the lower type, a spasm of disgust which is produced as 
 directly as the rising of the gorge at any offensive object. The 
 sight of the human hog revolts us, as we should sav, simply 
 because he is a hog, and the smell of the stye turns our 
 stomachs. Wherever such an intrinsic feelino; exists, it 
 cannot be justified by summing up the " conse(juences" in 
 the manner described. It is a feeling as much as any other 
 which so far justifies itself, that the pains and pleasures due 
 to it must be reckoned in our calculation. But this does not
 
 VIRTUE OF TEMPERANCE. 201 
 
 imply that this intrinsic feeUng or instinct is not to be 
 measured bv a standard derived from the general principle of 
 utility. When we ask whether it is useful, we do not mean 
 to ask whether it implies a right estimate of the other conse- 
 quences of the conduct — that is, of the collateral results which 
 are offensive to other instincts ; for this is virtually to assume 
 that it has no independent existence, but is only a special 
 application of those instincts to the foreseen consequences. 
 We must also ask whether it is useful in the sense of being 
 a necessary part of the higher type of character. As men 
 become more intellectual, sympathetic, and so forth, they 
 gain fresh sensibilities, which are not simple judgments of 
 consequences hitherto improved, but as direct, imperative, 
 and substantial as anv of the primitive sensibilities. Hence 
 the justification of the instinct is not that it implies a judg- 
 ment of what is useful, but rather that it is a useful judgment. 
 To get rid of the sensibility you must lower the whole tone 
 of the character or destroy the perception of consequence. 
 Therefore some such sentiment is essential to social develop- 
 ment, though it is difficult or impossible to obtain any accurate 
 measure of the desirable strength. That is a problem which 
 can onlv be worked out by actual experiment. 
 
 28. This would appear to follow psychologically from the 
 fact that the excess in question (and this is of course equally 
 applicable to other sensual excesses) is itself prompted by a 
 direct instinct. It is a morbid development of a certain 
 appetite ; and it would seem that the disgust which we feel 
 for the excesses of others is a direct result of the correlative 
 impulse in ourselves. We are shocked by the excess of the 
 glutton, because our imagination is revolted when we put 
 ourselves in his place, and fancy ourselves consuming the 
 same monstrous masses of food. The question of the degree 
 in which it is desirable to cultivate this instinct must be diffi- 
 cult, because every direct instinct is in its nature incapable 
 of such measurement. When we are estimating the conse- 
 quences of conduct in the ordinary sense, we may sometimes 
 have an,available test. We are asking whether the pain given.
 
 202 THE SCIENCE OF ETHICS. 
 
 for example, by an attack of gout is greater or less than the 
 pleasure of a series of good dinners, and we may conceive at 
 least that in some cases the sum is fairly easy, and admits of 
 some approximate estimate. But when we ask whether a 
 certain instinct is on the whole useful, and in what degree, we 
 are really asking the enormously complex problem whether 
 the man who has it, together with all that is implied in it, is 
 on the whole a better or worse member of society than the 
 man who has it not, together with all that is implied in its 
 absence. It is clearly impossible to reduce the comparison 
 between two organisms to any direct process of calculation, 
 though we may at least attempt to calculate the quantities 
 of happiness in two different sets of actions ; and hence it 
 follows that the strenirth of the instinctive emotions of dis- 
 o-ust which we have been considering- mav varv widely with- 
 out being capable of immediate regulation by any reasoning 
 process. This may go to explain why the disgust excited by 
 particular vices, or, again, the admiration for their contraries 
 implied in the aesthetic theories of morality, is apt to deviate 
 very widely from any assignable measure of immediate utility. 
 
 IV. Virtue of Truth. 
 
 29. I may now proceed to the virtues which are in some 
 respects at the opposite pole. It needs no demonstration that 
 some regard for truth is implied in the simplest social state. 
 Lano-uao-e is at once the product of society, and essential to 
 anvthing that can be called society. No mutual understand- 
 ino- can exist without a communication of thought of which 
 language is the most perfect and the indispensable instru- 
 ment. To say that language is necessary is to say that truth 
 is necessarv, for otherwise we should speak of signs which 
 have no signification. Lving itself is only possible when 
 some degree of mutual understanding has been reached, and 
 truthfulness is therefore an essential condition of all social 
 development. Yet even the virtue of truth is often recog- 
 nised sraduallv and with many limitations. As tlie law of
 
 VIRTUE OF TRUTH. 203 
 
 chastity takes the form, " Thou shalt not commit adultery," 
 the law of truth appears in the form, " Thou shalt not bear 
 false witness." l^hose lies are forbidden which are palpably- 
 injurious to our neighbours. The imperfect respect enter- 
 tained for truth in general is indicated by the practice of 
 judicial oaths. This seems to imply a kind of tacit assump- 
 tion that the duty requires some extrinsic sanction. We 
 might suppose the wickedness of taking away a man's life by 
 a false statement to be so obvious that a man restrainable by 
 no other motive would hardly be influenced by the sanctity of 
 an oath. The fact, therefore, that oaths are regarded as effica- 
 cious in such cases shows the faintness of the average sen- 
 sibility to the virtue. Lying to an enemy, again, is often 
 regarded as at least pardonable, even when cruelty is seen to 
 be wrono-. The obliiration to truthfulness is limited to rela- 
 tions with members of the same tribe or state; and, more 
 generallv, it is curious to observe how a kind of local or 
 special morality is often developed in regard to this virtue. 
 The schoolboy thinks it a duty to his fellows to lie to his 
 master, the merchant to his customer, and the servant to his 
 employer; and, inversely, the duty is often recognised as 
 between members of some little clique or profession, as soon 
 as it is seen to be important for their corporate interest, even 
 at the expense of the wider social organisation. There is 
 honour among thieves, both of the respectable and other 
 varieties. So the agreement of gamblers to pay debts of 
 honour and the respect paid to hospitality in rude countries 
 are familiar instances of the growth of a partial morality, 
 wherever it is plainly the interests of a class to have a mutual 
 understanding. Finally, it may be noticed that truthfulness 
 in the highest sense must be of slow development because 
 only partially intelligible. The child cannot lie because it 
 cannot speak the truth; so long, that is, as it is unable to 
 distinguish bet\Veen the creations of its own fancy and the 
 facts perceived in the external world. Telling stories has 
 for it the ambio;uous sense of Ivino; or relatino; avowed fiction. 
 In the same way it appears that the infantile man produces
 
 204 THE SCIENCE OF ETHICS. 
 
 myths and fables, and all manner of superstitious beliefs, 
 without a distinct perception of the diiference between 
 imaci"ination, hypothesis, and historical statement. Fiction 
 in the modern sense and downriirht Ivino; have the same root, 
 and only come to be clearlv differenced at a comparatively 
 advanced stage. The inheritance of such figments, especially 
 in the matter of religious belief, continues to perplex men's 
 minds. The virtue of philosophical truthfulness is only now 
 obtaining recognition, and the process by which it has won 
 recoirnition is an interestino; illustration of the general 
 principle. 
 
 30. It is easily perceived that an erroneous belief is in- 
 jurious to society, and there is a tacit recognition of this in 
 the conviction that a man who insults the god of his tribe 
 mav bring down evil upon his tribesmen. As moral rules 
 are implicated with theological beliefs, indifference to either 
 implies indifference to the other; nothing, then, is more 
 natural than the hatred, and consequently the persecution, 
 of the heretic. The chan<re is chiefly brou2:ht about bv the 
 slow perception that a man's beliefs are accidental and 
 dependent upon his social surroundings, and that the man 
 himself, therefore, is estimable so far as he is sincere, not so 
 far as his conclusions are right. Even at a very late period 
 the enunciation of that simple principle was regarded with 
 horror, and it has come to be admitted only by the accunui- 
 latins: evidence which has demonstrated that the difference 
 between the orthodox and the heterodox does not coincide 
 with that between good and bad. Yet the process is still 
 so imperfect that the virtue of toleration is still considered 
 rather as an external than an internal rule. It is generally 
 admitted that persecution is a blunder, as it is admitted — 
 by many people, at least — that protection is a blunder; but 
 everybody allows that a good man may be a protectionist, 
 inasmuch as the evil consequences are not* so manifest to 
 all understandings that an advocacy of protection necessarily 
 implies indifference to the welfare of the race. In the same 
 way the persecutor may have been deluded as to the eflect
 
 VIRTUE OF TRUTH. 205 
 
 of his measures rather than deficient in sympathy with his 
 kind. The growth, however, of an adequate perception of 
 the value of truthfuhiess in its highest sense would make this 
 impossible. A thorough conviction that the welfare of the 
 race depends upon its intrinsic character, and not upon the 
 special results attained in any given stage of inquiry, would 
 show that persecution was necessarily injurious to the inte- 
 rests of the society, and therefore confine the practice to those 
 who were indifferent to such injurv. As the respect paid to 
 courajre comes to be sjiven equallv to the man who fi(::hts on 
 our side or against us, in consequence of the accidents of 
 position, so we should respect truthfulness whether the results 
 obtained were in asrreement or not in agreement with our 
 own opinions. 
 
 31. So f^ir, then, we may say of truthfulness, as we have 
 said of other classes of virtue, that the morality emerges in 
 the form of a condemnation of particular kinds of conduct 
 seen to be pernicious to the societv, and gradually passes into 
 the recognition of a certain quality as implied in the realisa- 
 tion of the highest social type. But we have now to consider 
 a respect in which it differs from the other virtues. We 
 may notice in the first place that the obligation to a certain 
 definite external mode of conduct is generally stated as abso- 
 lute. Philosophers have deduced all virtues from truth, and 
 this absoluteness of the statement is favourable to the method; 
 for though puritv and courage give rise to rules which are 
 almost invariable, such as fidelity in marriage or to military 
 obedience, still they seem to include an empirical element. 
 The particular marriage law, for example, may vary, and it 
 is conceivable at least that polygamy may be the rule in one 
 period and monogamy in another, whilst the decision as to 
 the superiority of either rule would depend upon variable 
 conditions of human life. The rule of truthfulness, on the 
 other hand, seems to possess the a priori quality of a mathe- 
 matical axiom. It seems possible to say that it is always 
 right to speak the truth, as it is always true that two and two 
 make four. Truth, in short, being always the same, truth-
 
 2o6 THE SCIENCE OE ETHICS. 
 
 fulness must be unvarvinc;. Thus "Be truthful" means, 
 " Speak the truth whatever the consequences, whether the 
 teller or the hearer receives benefit or injury." And hence, 
 it is inferred, truthfulness im))lies a quality independent of 
 the oro-anisation of the aijent or of society. The aa:ent 
 should act as if he were a pure intellect. Nothing is so 
 truthful as a calculating machine, which grinds out the same 
 results, whether they give pleasure or pain. The preceding 
 virtues imply a certain equilibrium between the passions; 
 we cannot define "excess" without takins; into account the 
 constitution of the persons affected ; but the rule of truth- 
 fulness is independent of any such considerations and under- 
 lies them all. A conception of truth is implied in all rea- 
 soning, for reasoning is nothing but a perception of truth 
 and error. Thus to convey truth to others is apparently the 
 simplest possible rule, and capable of the most absolute state- 
 ment; and thus, it is supposed, we may reach the rule, 
 " Be truthful," without more reference to society than we 
 make in asserting a geometrical or a logical axiom. 
 
 33. We must ask, then, whether this absoluteness of the 
 rule really affects the doctrine that it expresses a condition of 
 social welfare. Moralists agree approximately in the admis- 
 sion that truthfulness is an essential condition of the welfare 
 of society as known to us. This, according to me, is the 
 ultimate ground, in a scientific sense at least, of its moral 
 value. The attempt is still made to represent the principle 
 as possessing an authority independent of any social conside- 
 ration. A different rule, it is said, would not only be mis- 
 chievous — that is, inconsistent with social development — but 
 self-contradictory. So far as this asserts anything more than 
 this, that even a slight social advance implies some degree of 
 truthfulness, it appears to me to be erroneous, or at least 
 unprovable. If for the rule, " Speak the truth," we could 
 substitute the rule, " Tell lies," it is quite true that neither 
 language itself nor the simplest of existing social institutions 
 could have been developed. 'J'o invert the rule of truth 
 would imply such a change as would necessitate a modifica-
 
 VIRTUE OF TRUTH. 207 
 
 tioii of some of the apparently most essential attributes of 
 humanitv. Yet even in this case the rule, " Deceive," is not 
 in a formal sense more contradictory than the rule, " Do not 
 deceive," and perhaps it would be possible to imagine con- 
 ditions under which mutual deception might be profitable to 
 a race. Undoubtedly such a race would have to do without 
 lang^uao-e, and without many other thino;s which are essential 
 to society as we know it. The state of truthfulness is as 
 essential to society as you please, but it is illegitimate to 
 confound this doubtless important statement with the state- 
 ment that the opposite rule is logically impossible. Universal 
 Iving is not self-contradictory; it is only impossible within 
 the whole rano;e of our knowledge. 
 
 ^^. In the next place, the fact, if it were a fact, that the 
 direct inversion of the rule was contradictory, would not 
 prove the rule itself to be necessary. The moral rule might 
 conceivably be, " Speak truth to your friends and lie to your 
 cnemie?," or, "Speak truth when it is not plainly inconvenient;" 
 just as the moral rule may be, "Kill not, unless in war or bv 
 due form of law, and do not drink more than is good for you." 
 This is not only conceivable, but is a fact of actual morality. 
 The rule of truthfulness, as I have said, is understood, in fact, 
 with many limitations, and is still far from being universally 
 and thoroughly carried out ; and these limitations correspond 
 to the limitations, real or apparent, of its social value. 
 And, finally, even in the highest moral stages there remain 
 certain limitations. Just because the rule can be so absolutely 
 expressed, it gives rise to various casuistical problems when 
 applied to the varying facts of life. Exceptions are recognised, 
 and these exceptions still obey the general rule of conformity 
 to the conditions of social welfare. Such, for example, are 
 the familiar cases of tellino; the whereabouts of his victim to 
 a murderer, telling patients in certain cases of their true 
 condition of health, and revealing secrets which it is thouglit 
 right to preserve, even when Iving alone can preserve them. 
 These are generally cases in which the internal diverges from 
 the external law, and, however rare, arc therefore worth
 
 2o8 THE SCIENCE OF ETHICS. 
 
 notice. The rule, " Lie not," is the external rule, and corre- 
 sponds approximately to the internal rule, "Be trustworthy." 
 Cases occur where the rules diverge, and in such cases it is 
 the internal rule which is morally approved. Truthfulness 
 is the rule because in the vast majority of cases we trust a 
 man in so far as he speaks the truth ; in the exceptional cases, 
 the mutual confidence would be violated when the truth, not 
 when the lie, is spoken. In such cases, therefore, the moral 
 law admits of an apparent exception, though there are manv 
 difficulties npon which I must touch hereafter in the way of 
 definino; them. 
 
 34. Now, without going farther at present, we see that 
 truthfulness is a fresh exemplification of the principles already 
 laid down ; for truthfulness, so far as it is a virtue, implies 
 trustworthiness. Trustworthiness, again, is clearly a quality 
 the development of which is essential to all such social growth, 
 at -least, as we can conceive to exist. It is an essential, and 
 the most obviously essential, condition of social development. 
 The growth of a sentiment of mutual confidence is so clearly 
 necessary that it seems to have been virtually regarded bv 
 theorists who dealt in social contracts and so forth as the 
 essential element or ultimate basis of morality. The develop- 
 ment, however, of the sentiment implies a correlative develop- 
 ment of the whole nature of men and society — a development 
 so slow as to be very inadequately realised even in the highest 
 societies or existence, and generally limited by very incomplete 
 confidence between the members of different classes. So far 
 from implying any suppression of passions or of the emotions 
 as such, it implies the cultivation of all the qualities by which 
 society is held toother, so that the man who is trustworthy 
 in the highest sense, who has in all cases a self-respect which 
 makes it impossible for him to lie, is the final outcome, the 
 ripest fruit, of the healthiest social conditions. Nothing is 
 rarer, even in the best societies, than an exquisite sensitiveness 
 upon such points, and a morbid social condition which implies 
 a defect of the true social sentiment is marked by nothing 
 more clearly than the existence of partial codes of truthful-
 
 VIRTUE OF TRUTH. 209 
 
 ness and the small importance attributed to lying between 
 
 different classes. But this is to sav nothins; else than that 
 
 truthfulness or trustworthiness is one of the essential qualities 
 
 of the true social type, only to be generated, and indeed only 
 
 to be fully appreciated, as the result of a long elaboration and 
 
 a slow growth of harmonious relations. It implies, doubtless, 
 
 though it is not necessarily implied, by the narrower quality 
 
 which rather belongs to the prudential qualities of a hatred 
 
 for error. The man, indeed, who hates error will generally 
 
 hate falsehood. 
 
 "To thine own self be true, 
 And it must follow, as the day the night, 
 Thou canst not then be false to any man." 
 
 This kind of virtue would belong, however, in the highest 
 degree to the man who approximates as closely as possible to 
 the calculating machine. Such a man no doubt would have 
 what may be called a purely intellectual love of truth ; but'of 
 truth as opposed to error or equiv^alent to accurate knowledge, 
 not of truth as implied by trustworthiness. He would be 
 like the calculating machine, equally at the service of the 
 rogue or the honest man, unless he had also the social qualities 
 summed up by a high sense of honour. It would give him 
 no pain to know that he had deceived others, so long as his 
 knowledge of their error was accurate. To know is to avoid 
 error; a desire for knowledge and a love of being deceived 
 are doubtless incompatible; and the man who felt the desire 
 would be self-contradictory if he did not desire to be " true to 
 himself" in the sense of thinkino; and observino; accurately. 
 But he might be as great a liar as any one, as willing, that is, 
 to use knowledge selfishly, unless we suppose that he also 
 possesses the qualities implied in the social type. And there- 
 fore we may assume here, as before, that the moral law means 
 simply a statement of the conduct characteristic of the most 
 vigorous vitality.
 
 2IO THE SCIENCE OF ETHICS. 
 
 V. The Social Virtues. 
 
 ^^. We come, in the last place, to the directly social vir- 
 tues. It needs no demonstration that the existence of a 
 society and its maintenance at a certain stage of civilisation 
 are dependent upon the instincts which, whatever their ulti- 
 mate nature, imply the readiness of the individual to identify 
 himself with his fellows, and to seek his own happiness in or 
 through their happiness. The question is not whether such 
 qualities are virtuous, but whether they are not the sole vir- 
 tues. Accordingly, some moralists hold benevolence to be 
 the single virtue, as others deduce all the moral principles 
 from prudence, or purity, or truthfulness. It is equally clear, 
 again, that these virtues, like the others, are developed through 
 a gradual process of generalisation; that so long as each tribe 
 is an independent whole, the mutual goodwill of its members 
 is often limited to the little section which represents the 
 whole world for them, and that even in highly developed 
 communities the limitations of benevolence are such as to 
 show little recognition of external claims. The gradual 
 extension of the sphere of morality may be due to the gene- 
 ralising faculty itself — to any process by which the vmder- 
 standing becomes enlightened and the sympathies more sen- 
 sitive, or it may be the indirect result of such a change in 
 external conditions as is implied by the absorption of many 
 communities into a larger whole, with a consequent recog- 
 nition of a wider community of interest. We are still very 
 far from an unqualified recognition of the virtue. Patriotism 
 often takes the narrow and contemptible form of a desire for 
 the extension of our own political organisation, in complete 
 indifference to, or rather with a contemptuous refusal to 
 attach any importance to, the welfare of the outside world. 
 I will only observe, without dwelling farther upon sufficiently 
 obvious considerations, that the change may be regarded in 
 two aspects. The difference between the morality of military 
 savages and tliat of a civilised country is not necessarily so 
 great as it seems. The actual rule of conduct niav be, in one
 
 THE SOCIAL VIRTUES. 21 r 
 
 case, to knock a stranger on the head, and, in the other, to 
 ask him to dinner. But then it must be admitted that the 
 difference may correspond simply to a difference in the facts 
 to which the rule applies, not in the rule itself. Stranger, in 
 one sense, means a being who will presumably kill me if he 
 can; in the other, a being who will presumably be friendly. 
 Thus the most civilised beins; would have to sro armed in 
 a reo'ion where no mutual understandino; had been reached. 
 The difference would be that he would there anticipate from 
 every one the behaviour which in his own country he would 
 only anticipate from an exceptional person. Unless his 
 morality forbade self-defence — which is hardly the case in any 
 accepted code — it would therefore allow or prescribe hostility 
 towards strangers presumably hostile. Thus the growth of 
 morality implies the development of this mutual understand- 
 ing. It may be obvious that such an understanding would 
 be useful if attainable, and so far its establishment does not 
 imply a change in the moral sentiment, but only in the facts 
 to which it is applied. On the other hand, it may also be 
 said that, to attain the understanding, we require not simply 
 a perception of its utility, but a greater development of 
 emotional sensibility, and so a greater power of sympathy, 
 which must go along with the purely intellectual process. 
 
 'J^6. This remark suggests one other which must be taken 
 into account. The virtues of which I am speaking are often 
 reo-arded as fallinir under the two heads of justice and bene- 
 yolence. These two are sometimes described as so far from 
 coincident that they may occasionally come into conflict, 
 and morality is to prescribe a course which is a kind of dia- 
 gonal between the two diverging rules. The justice of the 
 Supreme Being, we are told, is tempered by his mercy, and 
 vice versa. It is plain that the distinction corresponds to 
 that already drawn between the intellect and the emotions. 
 Temperance, so far as it is strictly virtuous rather than pru- 
 dential, implies, as I have said, a restriction upon selfishness, 
 and the instincts by which the society is held together may 
 be regarded as a development from the sexual and parental
 
 212 THE SCIENCE OF ETHICS. 
 
 instincts. Truthfulness, on the other hand, so far as it too 
 is strictly virtuous, must be regarded as implying a distinc- 
 tively social quality, and so far a fitness to act in accordance 
 with social demands. Thus we may regard temperance as 
 included in the duty of benevolence, though it has also a 
 prudential aspect ; and truthfulness, with the same limita- 
 tion, as included under the virtue of justice. And hence, 
 again, it must be repeated that, although there may be a dis- 
 tinction, there can be no proper opposition. If benevolence 
 means a desire to see others happy, it defeats its own end 
 if it tries to promote happiness of some at the expense of 
 others; and justice ceases to be just when it implies the 
 observance of a rule imposed for the good of all in such a 
 way as to be injurious to all. 
 
 37. It seems desirable, however, to put this rather more 
 precisely. What, in fact, is meant by justice? The spe- 
 cial case in regard to which the virtue first emerges is that 
 of a partial judge, and the same principle will apply of 
 course to all other persons intrusted with power by the 
 organisation of the community. The judge, again, is unjust 
 so far as he acts from any other considerations than those 
 which are recognised as legitimate by the legal constitution. 
 He has to declare the law, and to apply it to particular cases 
 without fear or favour. He must therefore give the same 
 decision whether the persons interested be friends or foes, 
 relations or strangers, rich or poor. And so, extending 
 the same principle, we say that a minister is unjust who dis- 
 tributes offices from other considerations than the fitness of 
 the applicants, except in certain cases where it is understood 
 that the power of appointment is an indirect mode of re- 
 warding the appointee. A parent is unjust who does not 
 distribute his property to his children equally, whether they 
 are clever or stupid, and even, within wide limits, whether 
 good or bad. And, on the other hand, considerations which 
 are held to be irrelevant in one case are regarded as essen- 
 tially relevant in another. The man is unjust who, in his 
 judicial capacity, distinguishes between a friend and an
 
 THE SOCIAL VIRTUES. 213 
 
 enemy, a relation and a stranger; he is equally unjust if, in 
 his domestic capacity, he fails to distinguish. It would be 
 grossly unjust in a judge to favour a Tory as against a 
 Whig; in a minister, it might be grossly unjust not to con- 
 sider the claims of party. The judge would be unjust who 
 showed favour to his own son, and the parent would be 
 unjust who did not give his money to his own children. 
 The essence of justice, therefore, seems to be the uniform 
 application of rules according to relevant circumstances ; or, 
 as we may put it, it is an application to conduct of the prin- 
 ciple of sufficient reason. Every difference in my treatment 
 of others must be determined by some principle which is in 
 that case appropriate and sufficient. In this sense^ therefore, 
 justice means reasonableness. 
 
 38. From this, again, we see in what way justice is implied 
 in social development. How are we to say what, in any case, 
 is the sufficient reason ? What set of rules is relevant, and in 
 what relations of life are we or are we not to attend to par- 
 ticular kinds of claims? To this we must replv, in general 
 terms, that all social development implies, as we have seen, a 
 distribution of functions, and the possibility of any permanent 
 organisation depends upon a corresponding perception of the 
 implied duties. The royal power, for example, has been de- 
 veloped by certain social needs, such as the necessity of a 
 centralised military power. If a particular king uses the 
 power not so as to discharge his function in the most efficient 
 wav, but for some other purpose, such as the gratification of 
 his appetites, the social organ ceases to be efficient, and the 
 society is so far in a state of degeneration. The due perfor- 
 mance of the function is suspended, and the organ becomes 
 an encumbrance instead of a useful part of the system. The 
 function of the criminal judge is the suppression of certain 
 offences; if he only punishes criminals not related to him or 
 receives bribes, and even if, as a judge, he gives undue weight 
 to qualities which in another capacity he might properly 
 regard, the administration of justice is so far corrupted. The 
 same principle of course applies in every conceivable case, and
 
 214 THE SCIENCE OF ETHICS. 
 
 is analogous to the statement that in a living organism the 
 welfare of the animal depends upon his eyes seeing and his 
 ears hearing according to certain rules. So far as the king 
 is a tyrant or the judge corrupt, each is hindering the due 
 performance of an important function, and injures the country 
 as the distorting eye or ear injures the organism. 
 
 39. Hence, when we say that a judge is to be passionless, 
 we use the word in the same sense as in the case of truthful- 
 ness. He must be inaccessible to certain passions when they 
 are out of place ; the ruling passion which must govern him 
 in his judicial capacity is the passion for justice; and that, 
 as before, whether acting justly has become a habit or is 
 prompted by a constant sense of the vital importance of 
 judicial impartiality to society in general. He should apply 
 the rule with the precision of a calculating machine; but 
 he must have a motive for setting his intellect to work im- 
 partially; and, of course, such motives are sufficiently nume- 
 rous and powerful in a healthy social state to make judicial 
 purity a second nature. When, again, we speak of mercy 
 as conflicting with justice, we mean generally to imply that 
 either the benevolence is ill-guided, or that a rule founded 
 upon some important considerations is being applied in cases 
 when an exception should be made. Benevolence conflicts 
 with justice in the case of misguided charity or excessive leni- 
 ency, where the desire of giving pleasure to some one immedi- 
 ately before us causes us to overlook the indirect consequences 
 of demoralisation, and so forth, to the person benefited, and the 
 injury done to others who have equal claims, but who do not 
 happen to be in so conspicuous a station. There is a conflict, 
 again, when a rule is pedantically applied, so as to injure a 
 man whose conduct is forbidden because it belongs to a class 
 generally mischievous, though we are quite certain that in the 
 particular case it had not the objectionable characteristics of 
 that class; as, for example, if a man is punished for homicide 
 in a case where homicide was evidently conmiitted in self- 
 defence. But in all such cases it is not properly a conflict 
 between justice and benevolence, but between a benevolence
 
 THE SOCIAL DEFINITION OF VIRTUE. 215 
 
 which takes into account all the relevant circumstances, and 
 one which fails to do so ; between a justice which applies a rule 
 with, and one which applies it without due consideration. 
 There are, of course, many casuistical cases which may be 
 suggested, and which often present real difficulty — cases, for 
 example, such as have afforded materials for the most striking 
 tragical situations, where a man is distracted between demands 
 made upon him in two capacities, — between the claims of his 
 family, for example, and the claims of the state, as in the story 
 of Brutus and his sons. Of such cases it is enough here to 
 say that, on whatever principles they may be settled, they do 
 not involve any proper conflict between benevolence and 
 justice. The same criterion is ultimately implied in both 
 cases, though w^e may lay more stress upon one or other 
 aspect. The command, " Be benevolent," carries with it the 
 condition that your benevolence should be regulated by reason, 
 and therefore should not benefit one man at the expense of 
 the society. The command, "Be just," carries with it in 
 the same wav the condition that, whatever function you are 
 discharging or in whatever capacity you are acting, vou 
 must be animated bv public spirit, that is, by motives coin- 
 cident with the dictates of a reasonable benevolence ; and 
 thus, in any case, the conclusion is that the moral law 
 prescribes a type of character which includes amongst its 
 manifestations a desire to discharge all the social functions, 
 and which therefore implies fitness to form part of a sound 
 social state. 
 
 VI. The Social Definition of Virtue. 
 
 40. This brief survey of the main divisionsof morality, which, 
 as I have shown, run into each other, and to some extent 
 mutually imply each other, may justify the conclusion as to 
 the social definition of virtue. Society, I have said, may be 
 considered as an organism, implying what I have called a 
 social tissue, modified in various w^ays so as to form the 
 organs adapted to various specific purposes. The existence
 
 2i6 THE SCIENCE OF ETHICS. 
 
 of the social tissue at any stage of development^ and its 
 power of maintaining itself, either as a part of the special 
 order or as against other societies, depends essentially upon 
 the fulfilment of certain conditions. Since the qualities 
 by which societies differ do not depend upon the innate 
 qualities of its constituent members^ which remain constant 
 (or approximately constant) through long periods of social 
 development, but upon these qualities as modified and de- 
 veloped by means of the social factors, it follows again that 
 the society grows on condition of impressing a certain char- 
 acter upon its members. This takes place in the earlier 
 stages by the development of a social sentiment unfavour- 
 able to certain specific modes of conduct. As the society 
 becomes more reasonable, more capable of understanding 
 and applying general principles, the sentiment develops into 
 an approval of a certain type of character, the existence of 
 which fits the individual for membership of a thoroughly 
 eflBcient and healthy social tissue. To state the main qualities 
 thus impressed is to lay down what I have called the " law 
 of nature." This, as we see, has the two poles of prudence 
 and virtue, corresponding to the distinction of the qualities 
 which are primarily useful to the individual and those 
 which are primarily useful to the society. Thus, whatever 
 strengthens the individual, increases his courage, energy, 
 industry, and so forth, must, other things being equal, 
 strengthen the society. Hence the law of prudence, which 
 corresponds rather to a precedent condition of morality than 
 to morality itself. Then the individual is stronger and the 
 society is stronger so far as his passions are regulated in a 
 certain way, and including alike the passions which have a 
 direct bearing upon the individual life and those which have 
 a direct bearing upon the social bond ; whence the virtues of 
 temperance and chastity. Thirdly, the individual, so far as 
 reasonable, must avoid error on his own account, and must 
 avoid the propagation of error for the welfare of the society, 
 whence the virtues of wisdom and trustworthiness. And, 
 finally, since the social union implies a direct interest of the
 
 THE SOCIAL DEFINITION OF VIRTUE. 217 
 
 individual in the welfare of the society, wc have the directly 
 social virtues, which imply at once benevolence and justice, 
 accordino; as wc attend to the motive or to the rcirulatcd 
 action of motives. And these four classes of excellence, 
 which by the mode of development are necessarily recon- 
 cilable and mutually implicatory of each other, seem to 
 constitute all that is meant by the general moral law, 
 though admitting, of course, an indefinite variety of special 
 applications. 
 
 41. Briefly, then, we may say that morality is a statement 
 of the conditions of social welfare ; and morality, as dis- 
 tinguished from prudence, refers to those conditions which 
 imply a direct action upon the social union. In other words, 
 morality is the sum of the preservative instincts of a society, 
 and presumably of those which imply a desire for the good 
 of the society itself. And so far, as I may observe, moralists 
 w^ould c;enerallv asjree as to the fact, howev^er difFerentlv thev 
 might state it or account for it. That is to say^ no moralist 
 of any school would deny that the health of the society, its 
 power of persistence, and its harmony with the happiness 
 of its members, depends essentially and primarily upon its 
 condition in respect of morality and prudence. Lower the 
 average standard, and the society is so far in danger of stag- 
 nation, decay, and actual extermination ; raise it, and the 
 society is so far better calculated to preserve itself and to 
 develop new and richer life. But difiiculties and disputes 
 arise when we go farther. That which I am specially con- 
 cerned to meet concerns the mode in which virtue, the 
 nature of which can be clearly and simply expressed when 
 we speak of the social organism, presents itself when we 
 change our point of view, and ask how it comes to be im- 
 pressed upon the individual. Virtue is a condition of social 
 welfare; but why should I be virtuous, or what are the 
 motives by which the conformitv of the individual is or mav 
 be secured ? In order to answer this, and the many pro- 
 blems which arise from or are intimately connected with it, 
 it is necessary to consider what may be called the '^ theory of
 
 2i8 THE SCIENCE OF ETHICS. 
 
 obligation." This, accordingly, will form the topic of the 
 next three chapters. What is the quality in respect of 
 which the individual is susceptible to the social pressure? 
 What is the form taken by that pressure ? What is the 
 nature of the character which must be impressed upon the 
 individual ? These are the problems of altruism, of merit, 
 and of conscience, upon each of which I shall have something 
 to sav.
 
 ( 219 ) 
 
 CHAPTER VI. 
 
 ALTRUISM. 
 
 1. Egoistic Instincts. 
 
 I. The last chapter has brought us to the problem which 
 is at the root of all ethical discussion. The definition of 
 morality is perfectly simple so long as we fix our attention 
 upon the social organism ; the moral law is a statement of 
 certain essential conditions of the vitality of the society, and 
 specifically of those conditions which apply primarily to the 
 society, and only in a secondary sense to the individual. 
 The individual then must, so far as moral, be capable of 
 aiming at the social welfare before his own. In every moral 
 system, action for the good of others is regarded as virtuous, 
 if not as the only action entitled to be called virtuous. But 
 how is such action conceivable? Can any man really 
 sacrifice himself? Is not the admission of the possibility of 
 self-sacrifice inconsistent with the assertion that all conduct 
 is determined by the feelings of the agent, and therefore, as 
 it would seem, by his own pain or pleasure ? The egoistical 
 theory which accepts this conclusion is the favourite bug- 
 bear of ethical writers. By fixing upon an antagonist the 
 imputation of egoism you give him a bad name, and have 
 an easy opportunity for a rhetoric which reflects credit upon 
 the assailant. Only a cynic or a satirist here and there 
 maintains the thesis of universal selfishness, or a rare philo- 
 sopher propounds it in words, whilst carefully explaining 
 that he does not mean what he says. I am half ashamed to 
 join in the hue and cry against an outcast doctrine so often 
 repudiated, and so often pronounced to have been slain
 
 220 THE SCIENCE OF ETHICS. 
 
 outright. Yet Its singular vitality, in spite of repeated 
 assaults, shows it to possess some real plausibility. If it be 
 erroneous, we must be careful in laying down the positions 
 of which it is a perverted statement ; and as even the 
 egoist admits that altruism, though a mere mask, is yet a 
 mask invariably adopted by the virtuous, it is equally neces- 
 sary, upon his assumption, to examine its true nature, though 
 at the expense of painfully retracing some familiar ground. 
 
 3. Where does the puzzle arise? Common sense finds no 
 difficulty. A man is altruistic who loves his neighbour as 
 himself; who gives money to the poor which he might have 
 spent in luxury; who leaves house and home to convert 
 savages; who sacrifices health to comfort prisoners or sufTerers 
 in a plague-stricken city. Sir Philip Sidney was altruistic 
 when he gave the cup of water to the wounded soldier in- 
 stead of slaking his own dying thirst. Such deeds make our 
 nerves tingle at the hearing, and ennoble the dreary wastes of 
 folly and selfishness recorded in history. But the egoist has 
 an easy explanation. Sidney's conduct only proves that his 
 vanity was stronger than his thirst, and vanity is one of 
 the meanest of motives ; the charitable man is repaid by a 
 glow of self-complacency; the missionary hopes for a reward 
 in heaven; the physician in the plague-stricken city is eager 
 for praise and shrinks from general contempt. In all cases, 
 and however skilfully disguised, some personal gratification 
 supplies the cogent motive. Everywhere we find the con- 
 viction, tacit or express, that the conduct adopted will secure 
 the greatest happiness of the agent. Nobody, indeed, can 
 deny that men act so as to destroy their own happiness; 
 if vanitv causes a man to sacrifice his best chances, so mav 
 the desire of a brief sensual pleasure. One man goes to a 
 public-house, another leads a forlorn hope; each throws 
 away life for a brief enjoyment, whether mischievous or 
 beneficial to others. The egoist may explain both cases 
 with equal facility. Each act iniplies a temporary error of 
 judgment ; the love of gin or the love of glory overpowers 
 the reflective faculties. Let a man call up all the conse-
 
 EGOISTIC INSTINCTS. 221 
 
 quences of his action, and he will cease to be imprudent, 
 and therefore to throw away his chances of future happiness, 
 whether his happiness is bound up with or absolutely irre- 
 concilable with the happiness of others. A man may con- 
 ceivably be unselfish, but so far as really unselfish he is a 
 fool for his pains. He will only do good to others, if a wise 
 or a thoroughly enlightened man, so far as he expects to 
 derive some benefit for himself. Thus the statement that 
 altruism is impossible does not mean that it is impossible in 
 fact, but only in deliberate intention. No man, it is argued, 
 can sacrifice himself knowingly and intentionally. 
 
 3. To unravel these arguments as well as I can, I must begin 
 by recalling some considerations already noticed. All con- 
 duct has an indefinite series of consequences ; that is to say, 
 any act may be taken as the starting-point of a series of 
 events continuing for an indefinite period, all of which would 
 be different had the action been different. The agent can 
 of course contemplate only a small part of these conse- 
 quences. It is his intention to produce those events which 
 he sees to be dependent upon his actions, and from the 
 intention we may infer the motive — that is, the feeling 
 which is gratified by the realisation of the intention. Now 
 conduct may be beneficial to others without any benevolent 
 intention, and not only by accident, but from its intrinsic 
 nature. This, for example, is true of the sexual and parental 
 instincts of animals. The instinct which leads a stag to 
 form a harem and propagate his race is of essential utility 
 to the herd, but does not imply that the stag has the least 
 sympathy for the hinds or their fawns. The hen sacrifices 
 herself for her chicken, but it does not follow — perhaps 
 it is only our anthropomorphic tendencies which suggest the 
 inference — that she considers her chickens as anything more 
 than soft, warm lumps of down, which are comfortable 
 furniture in the nest. She sacrifices herself to save them, 
 it may be, just as she would run into equal risk if tempted 
 by a particular kind of food. The apparent love may be 
 (one hopes that it is not) simply a physical appetite.
 
 2'>'> 
 
 THE SCIENCE OF ETHICS. 
 
 4. It is, then, unsafe to infer altruistic intentions from 
 altruistic consequences. In human beings the sexual appetite 
 appears to be the most purely selfish of impulses, in so far as 
 it prompts to conduct often ruinous to its objects. On the 
 other hand, it is at the root of all the social virtues. The 
 passion cannot be gratified without important consequences 
 to others, and yet in its lower form implies no recognition 
 whatever of those consequences. The purely sensual appe- 
 tite remains in the reasonable beins; who can recoffnise the 
 consequences to others. If he still gratifies the passion 
 without reference to those consequences, he is prompted to 
 the grossest selfishness ; if, on the other hand, the sensual 
 impulse is so regulated that others are not injured bv its 
 gratification, it may become the nucleus of the most un- 
 selfish affection. Before the agent is enlightened by reflec- 
 tion it is hardly proper to call him either selfish or unselfish ; 
 he does not repudiate the claims of his fellows — he is inca- 
 pable of perceiving their existence. Selfishness or unselfish- 
 ness is developed as the intellect becomes capable of 
 contemplating the happiness of others besides the agent, 
 llie same change is manifest in other relations. The master 
 of slaves may regard them simply as convenient instruments ; 
 he may risk life and limb to defend them, as he would run 
 the same risk in defence of inanimate property. So far as 
 his interest is furthered by their health and safety, his rela- 
 tion to them may have beneficial consequences although 
 there is no benevolent intention ; but so far as he can 
 increase his own comfort by giving them pain, he may be 
 as willing to inflict pain as to give pleasure. He might, for 
 anything we can see, be as willing to feed his pigs on slaves 
 as to feed his slaves upon pork, if the price of the two com- 
 modities should vary. In this case the relation remains purely 
 external ; the slaves are considered simply as things, not as 
 human beings. But where external circumstances enforce a 
 certain identity of interest upon a particular group, there is 
 room for the development of genuine altruism. The mother 
 may be stimulated to actions which are de facto beneficent
 
 EGOISTIC INSTINCTS. 223 
 
 and self-sacrificing in their consequences by the pleasant 
 sensations connected with suckling and nursing; as soon as 
 she becomes aware that she is furthering the happiness of 
 her offspring, the happiness may itself become a motive 
 for conduct. There is already a framework provided within 
 which the affections have room to expand. The purely 
 sensual pleasure is now so blended with the pleasure derived 
 from a perception of the happiness conferred that it may 
 be impossible to discriminate between them. In the normal 
 case they operate in the same direction, and there is no 
 conflict so long as they do not dictate diverging lines of 
 conduct. The same change (as I have already argued) 
 takes place in regard to social relations generally. The 
 connection between husband and wife, which implied origi- 
 nally the subordination of one being to the sensual appetites 
 of the other, becomes the ground of the most perfect sym- 
 pathy and the strongest mutual affection. A slave-holding 
 community may develop into one in which the employer and 
 the employed have friendly domestic relations, and each 
 desires, or at least respects, the pleasure of the other ; and the 
 point at which this becomes possible must be at that stage 
 of intellectual development in which we are able to recognise 
 the happiness of others than ourselves. Till that is possible, 
 each beine; can be at most the instrument of the other's 
 pleasure, and regarded with feelings not differing in kind 
 from those excited by any lifeless object. So soon as we 
 realise the fact that we cause pain and pleasure to others, 
 their pain or pleasure may supply a motive. Till that 
 period, the agent is not so much selfish or unselfish in the 
 full sense as incapable of any feeling in the matter. There 
 must already be beneficent conduct, but there can be no 
 benevolent or malevolent intention. 
 
 5. This change is sometimes explained as a product of 
 association. The truth or falsehood of this doctrine in 
 general depends upon principles which lie beyond the present 
 inquiry; but it may be observed already that there is at 
 least a prima facie objection to the completeness of the
 
 2 24 THE SCIENCE OF ETHICS. 
 
 explanation. The simple association of a particular object, 
 material or animated, with certain pleasures, may doubtless 
 make us value it as a useful instrument, but we do not see 
 how it can change the instrument to an object of sympathy. 
 The child may regard its mother as a fountain of agreeable 
 drink, and the mother may regard the child as affording a 
 pleasant relief; but so far there is nothing in the associa- 
 tion which should lead the child to distinguish between the 
 mother and the bottle, or the mother to distino-uish between 
 the child and some mechanical contrivance, except in so far 
 as one may be more efficient than the other. Undoubtedly, 
 the pleasant association prepares the way for the higher 
 sentiment. The fact of the mutual convenience provides a 
 necessary condition for satisfaction in conferring mutual 
 pleasure. But the condition is obviously insufficient, for it 
 suggests no account of the distinction which arises between 
 the sentiment excited by a mother or that excited by a 
 comfortable garment. It might happen that, by throwing 
 away broken meat which was a nuisance to me, I was 
 contributing to the support of a poor family or to the 
 support of the crows. So long as I regard both simply 
 as conveniences for the removal of my refuse, I shall 
 simply prefer one or the other as it discharges that func- 
 tion most efficiently. If I am better pleased to benefit 
 the poor family than the crows, whilst the conduct in 
 other respects produces the same effect upon me, I am so 
 far "altruistic;" and this implies that I am capable of 
 sympathising with, and therefore of at least recognising, the 
 happiness conferred upon the human beings. The bare 
 convenience to me, being by hypothesis the same, would 
 not lead to any distinction by mere force of association. 
 
 6. I assume, then, that altruism, whatever its meaning or 
 analysis, begins at the point at which I am capable of 
 benevolent intentions ; or, in other words, where conferring 
 pleasure upon others becomes a possible motive. And here 
 the egoist meets us by denying that this can ever be an 
 ultimate motive. The desire to give happiness is always
 
 EGOISTIC INSTINCTS. 225 
 
 capable of a further analysis^ which shows it to inckide a 
 desire of happiness for ourselves. Nobody denies that the 
 wish to give happiness may be part of my motive^ and it 
 may be at a given moment the only part of which I am 
 distinctly conscious. I till a field in order that I may reap 
 the harvest, but whilst I am tilling I mav be thinkino; onlv 
 of the plough ; the means become a temporary or conditional 
 end. So I may be kind to you in order that you may here- 
 after be kind to me, and at a given instant of kindness I 
 may not be distinctly conscious of the ultimate end. But, 
 according to the egoist, such an end must always exist. The 
 goal of every conceivable desire is some state of agreeable 
 consciousness of my own. I may not look to the end of 
 the vista of intended consequences, but, if I look, I shall 
 always see my own reflection. This, again, is taken to be a 
 self-evident truth. To suppose genuine altruism, that is, a 
 desire of which the ultimate end is the happiness of some 
 other person, is to suppose a contradiction. 
 
 7. This statement appears to me to convey a palpable 
 truth or a great error according to our mode of interpreting 
 it. Every motive, as I have already said, may be described 
 either in objective or subjective language. I may either 
 state the external conditions of gratifying my desire or the 
 desire which is gratified. It is almost the same thing to say 
 that I desire food or that I am hungry, and it depends 
 upon the particular circumstances whether it is more con- 
 venient to use one or the other form of expression. Though 
 "almost," indeed, it is not quite the same thing, for the 
 reason that the external condition never corresponds abso- 
 lutely to the internal state. A given desire may be gratified 
 in various ways, and, again, a given set of conditions may 
 gratify various desires. The statement that I want a fire 
 expresses something more and also something less than the 
 statement that I want to be warmed ; for I may want a fire 
 in order to cook my dinner, and I may preserve warmth 
 by putting on a greatcoat. Hence it may not be simply 
 tautologous to say, " I want a fire because I want to be 
 
 p
 
 226 THE SCIENCE OF ETHICS. 
 
 warmed/' for the statement specifies the particular relation 
 out of several possible relations in which a fire is desirable ; 
 though it would be tautologous to say, '' I want the condi- 
 tions of warmth because I want to be warmed." I may, 
 indeed, give a reasonable meaning even to this latter state- 
 ment, if the question be whether the immediate cause of the 
 desire be a change in the external or the internal conditions. 
 I may, for example, have a desire to take food because I am 
 hungry, or be hungry because the food stimulates my appetite. 
 It may be said that in the first case the desire is an inde- 
 pendent cause, as it arises from organic changes which take 
 place independently of the internal changes, whilst in the 
 other case the desire itself arises from those changes. I may 
 ask how it comes to pass that I have certain feelings, in 
 which case I may of course trace back the series of events as 
 far as I please, and call any one " the cause," which being 
 altered, the subsequent events would be altered ; but at the 
 time of action the desire itself, to whatever it may be due, 
 must be the cause of conduct. 
 
 8. Now the assertion that there is both a subjective and an 
 objective condition of conduct tells us nothing whatever as 
 to the nature of the objective conditions of gratification. 
 We gain nothing by changing from one mode of statement 
 to the other. It is idle to say that I want a thing because I 
 want a thing, or to modify statements by emphasising in 
 one case the want and in the other the thing, or, again, by 
 laying a special stress upon the " I." Of course my con- 
 duct, whatever it is, must be conditioned by my desire ; but 
 the objective condition may be anything which can affect 
 my desire. There is, therefore, no prima facie objection to 
 the hypothesis that these objective conditions may include 
 the happiness of others. There is no more reason for denying 
 that we may receive pleasure from the pleasure of another 
 man than for doubting that we may receive it from the 
 combustion of coal. The only condition which we have so 
 far assumed is tiic obvious one that I can only desire that 
 which has some relation to me, for this is doubtless implied
 
 EGOISTIC INSTINCTS. 227 
 
 in saying that I desire it. A desire for warmth, for example, 
 could not prompt a desire for such a change in the atmos- 
 pheric conditions as would not affect my bodilv org-anisation. 
 The desire could not be gratified by a change in the tempera- 
 ture of Sirius or a fire in a desert island, for that would be 
 to desire a warmth which did not warm. In short, I must 
 not so state the objective conditions as to suppose that I can 
 desire them when they necessarily part company with the 
 subjective condition. To say that I desire something is to 
 say that the something has an influence upon me, since it has 
 an influence upon my happiness; and this condition must 
 be noticed, because, for the reasons already noticed, the 
 objective statement generally is too wide, and includes other 
 conditions besides those which gratify my desires. 
 
 9. Hence we reach the problem which has to be considered. 
 Conduct, I have said, is determined by feeling, or, in other 
 words, by happiness and unhappiness. My happiness, again, 
 depends at every moment upon my relation to the external 
 world, and this external world is constituted partly of thing^s 
 which I assume to possess, and partly of things which I do 
 not assume to possess a consciousness analogous to my own. 
 I may or may not take into account this external conscious- 
 ness. I may regard an oyster, as I regard a peach, simply as 
 a toothsome morsel, or I may suppose him to have a certain 
 capacity for pain or pleasure. I may regard my fellow-men 
 in either of these ways — as parts of a mechanism or as 
 sentient organisms. It may happen that, in the former 
 case, the conditions of my happiness are identical with the 
 conditions of yours. I may be unable to get my own 
 dinner without by the same action getting a dinner for you. 
 If so, there is, we may say, an external identity of interest 
 and my conduct may be beneficial to you without implyino- 
 the existence in me of any desire for your happiness as such. 
 If, however, the knowledge of your happiness has an essential 
 and unconditional tendency to promote my happiness, the 
 case is so far altered. I shall then make a distinction between 
 cases which previously appeared to be identical. I shall not
 
 228 THE SCIENCE OF ETHICS. 
 
 hold it to be the same thing whether I walk up a ladder of 
 wood or a ladder of sentient bodies. There is, as I have said, no 
 a priori objection to the hypothesis that one kind of feeling 
 may be as real as the other; that the object of desire may 
 include the feelings of others, as well as changes in external 
 objects which do not involve feeling. But there is the differ- 
 ence that the former kind of feeling admits of some further 
 analysis in a sense not possible in the other case, though not, 
 as I think, in the sense supposed by the egoist. Given the 
 feeling of hunger, for example, I can go no further in the 
 subjective analysis, though I may assign the correlative 
 physiological processes. It exists as an ultimate fact, and 
 so gives the only sufficient explanation of my conduct. But 
 if I sympathise with your hunger, there is another process 
 implied, of which it may be possible to give some account. I 
 suffer because you suffer ; and he may fairly be asked whether 
 this fact can be made more intelligible or the conditions of its 
 occurrence explained more precisely. Having done what we 
 can in that direction, we may be able to return to the original 
 problem. 
 
 II. Sympathy. 
 
 lo. Now, in the first place, the recognition that there are 
 other centres of consciousness besides my own is bound up in 
 the closest way with the recognition of what is called an 
 objective world. A thing is held to be objective — in one 
 sense, at least, of a most ambiguous word — when I hold that 
 it is perceptible by you as well as by me, and subjective when 
 I hold that it is perceptible by me alone. I do not assert or 
 deny that this is the sole meaning; but at least, in assert- 
 ing the objective existence of anything, I assert it to exist 
 for others as well as for me. The statement is bound up in 
 the process by which my world is constituted. It is the 
 power of so regarding the world which gives it, if I may say 
 so, a stereoscopic solidity. Each person sees only one aspect 
 of surrounding realities. He holds it to be real in so far as 
 he holds that other aspects are visible to his neighbours. The
 
 SYMPATHY. 229 
 
 actual sensations of every moment are completed and held 
 together in the mind by a whole system of ideal perceptions 
 more or less distinctly present in actual consciousness. The 
 room in which I sit is part of the house, the house of the city, 
 and so forth ; and such statements summon into comparative 
 vividness a set of perceptions not actually present at the 
 moment, but present to others, and which would be present 
 to me if I changed my position. 
 
 II. To think of anything as real is to call up a system 
 of such ideal perceptions. It is to rehearse a set of sensa- 
 tions which are somehow (the " how " is a problem of meta- 
 physics) regarded as representative of others not actually pre- 
 sent. If I have to do with simple relations of time and 
 space, no assignable emotion is produced. I complete my 
 picture of the exterior of my room by imagining what I should 
 see from outside, and so I may build up a picture of the whole 
 world. But the world is interesting to me so far as it is the 
 dwelling-place of myself and of beings analogous to myself. 
 The man as directly revealed by my senses is simply an object 
 of certain colours and dimensions, but the relations in which 
 he is really interesting to me are those in which he is moved 
 bv passions like my own. I do not really think of a man 
 till I have interpreted the external signs by the emotions 
 which they signify. Till I do that, he is for me merely a 
 coloured and moving statue. I know not whether he will 
 be a friend or an enemy, one who will save or destroy my 
 life. To complete the picture, I must therefore represent his 
 feelings. I must put myself in his place, feel what he feels, 
 and measure his conduct by the analogy of my own behaviour 
 under similar circumstances. The process is the same which 
 is implied in every intellectual process. I imagine a state of 
 consciousness not actually present, and besides imagining 
 mere sensations and perceptions of mechanical relations, I 
 imagine a set of emotions and reasoning processes analogous 
 to my own. I complete my picture of the house by putting 
 myself outside in imagination ; I add the imagined feelings 
 of standing in the rain and cold, and, in virtue of some in-
 
 230 THE SCIENCE OF ETHICS. 
 
 tellectual process not here to be discussed, I take those feel- 
 ings to be representative of those of the beggar at my door. 
 Till I have rehearsed those feelings I am not really thinking 
 of the beo-o-ar, but only of a lay-fig-ure of certain dimensions 
 
 CO ^ - y O 
 
 and colours. Through this mental operation alone is that 
 knowledge of the external world conceivable which is yet 
 necessary to give coherence to our own series of sensations 
 or to frame them into anything that can properly be called 
 knowledge. 
 
 12. Hence it would appear that sympathy is not an 
 additional instinct, a faculty which is added when the mind 
 has reached a certain stage of development, a mere incident 
 of intellectual growth, but something implied from the first 
 in the very structure of knowledge. I must be capable of 
 representative ideas in order to think coherently or to draw 
 the essential distinction between object and subject. I must 
 be able to regard certain modes of thought and feeling as 
 symbolic of modes present in other minds, and to my own in 
 other positions. To realise the world as a material whole, I 
 must have representative perceptions of time and space. To 
 realise the world of thought and feeling, that world upon 
 which my life and happiness depend at every instant, I must 
 have representative emotions. " Put yourself in his place " 
 is not merely a moral precept ; it is a logical rule implied in 
 the earliest germs of reason as a description of reasoning 
 itself, so far as it deals with other sentient beings. To know 
 that a man has certain feelings is to have representative feel- 
 ings, not equal in intensity, but identical in kind. Sympathy 
 and reason have so far an identical factor — each implies the 
 other. I cannot reason about another man except in so far 
 as I can rehearse his motives ; I cannot feel for him except 
 in so far as I can regard my feelings as representative. The 
 two processes are mutually involved, and^ whatever difficulties 
 may be suggested, it seems clear that I cannot properly know 
 what another man feels v^ithout in some degree feeling what 
 he feels, 
 
 13. Although I must take for granted the metaphysical
 
 SYMPATHY. 231 
 
 implications of this statement, whatever they may be, I must 
 dwell for a moment upon certain difTiculties which obscure 
 it even from the scientific point of view. The mechanism of 
 language tends to introduce certain perplexities ; for it is, 
 as I have said, one main use of language that it enables us to 
 reason by symbols without calling into distinct consciousness 
 all the feelings which are symbolised. The arithmetician per- 
 forms his processes without evoking a distinct vision of the 
 numbers with w'hich he deals, or recalling the primary intui- 
 tions which satisfied him of the truth of his rules. We say 
 " men " without attempting to call up more than a very small 
 part of all the thoughts which may at diflerent times be 
 suggested by the word. I may say, " My servant is ill, there- 
 fore I will give him a holiday," and may act accordingly, 
 though a vcrv faint image of illness, holidays, or servants 
 presents itself to my mind. When I say that a man has 
 been hanged, the expression is thus liable to many ambiguities. 
 It may suggest to me simply that a figure of a certain wxight 
 and shape has been suspended in a certain way. It may call 
 up merely certain affections of the senses of sight and touch ; 
 or, again, it may suggest certain visible signs of vital processes, 
 struggling limbs, the gradual cessation of motion, and the 
 conversion of a moving and coherent into a motionless and de- 
 caying body ; or it may further suggest the painful sensations, 
 the despair, horror, and remorse which I suppose the man to 
 have felt. I may stop at the external signs, and I may pass 
 beyond them to the emotions signified. And thus the same 
 words may call up the mental images which would be gene- 
 rated in the most and in the least sympathetic witness, and 
 serve equally to suggest certain mechanical relations as to 
 stimulate the deepest and most complex emotions. When, 
 therefore, I say that knowledge implies sympathy, I of course 
 do not mean to deny that we may have a knowledge of the 
 external fact, which is, for many purposes, all the knowledge 
 actually present to our minds, and which implies no sympathy 
 at all. I need not go beyond the feelings wdiich would be 
 called up by hanging a dead body, even when I am said to
 
 232 THE SCIENCE OF ETHICS. 
 
 know that a man has been put to death by hanging. In 
 every case, a large part of the possible emotion remains 
 unrealised. 
 
 14. It is more important, perhaps, to remark that I do not 
 profess to give a complete account of the process. Undoubt- 
 edly it must be held that the knowledge of the feeling is 
 something different from a simple rehearsal of the feeling. 
 The representative feeling may differ not only in intensity 
 but also in quality from that which it represents. The 
 knowledge that another man is suffering gives rise to com- 
 plex or varying emotions. Nothing, of course, is more com- 
 mon than to find that men take pleasure in humiliating 
 and mortifying their neighbours. The critic rejoices in tor- 
 menting a sensitive poet ; the child delights in teazing his 
 playfellows and his animals ; and it is an undeniable though 
 a hideous fact that there is such a thing as a voluptuous 
 pleasure in cruelty. Milton's "lust hard by hate" expresses 
 a profound psychological truth. And such facts demand a 
 brief consideration in order to show that they are not incon- 
 sistent with the theory just stated. They are rather, as I 
 think, examples of the danger which besets any one who 
 tries to translate emotional laws into logic, and to pronounce 
 any variety of human character impossible because it seems 
 to involve an implicit contradiction. 
 
 15. Much cruelty, in the first place, means simple insen- 
 sibilitv. The defect of sympathy is also an intellectual defect. 
 The child tormenting an insect or the savage abandoning 
 his infant is simply not capable (in the common phrase) of 
 enterino; into the feelins^s of his victim. The child is amused 
 by the spinning of the cockchafer as he is by the spinning of 
 a top ; it is simply a curious bit of mechanism. The savage 
 may throw away a baby when its cries are tiresome because 
 he does not think of its sufferings at all. Cruelty of this 
 kind is therefore nothing but intellectual torpor, an inca- 
 pacity for projecting oneself into the circumstances of others, 
 and therefore inability even to think about the most impor- 
 tant set of conditions of the happiness of more developed
 
 SYMPATHY. 233 
 
 beings. The dulness which incapacitates a l)Oor for appre- 
 ciating the feelings of the refined nature is so far a disquali- 
 fication for all the more complex social activities. And so 
 we may observe, that as a society becomes more civilised, as 
 the reasoning faculties become quicker and wider, and the 
 power of observing many relations between living beings 
 increases, there is an improvement in the virtue of humanity 
 if in nothing else. To think about other beings is to stimu- 
 late our sympathies, and our sensibility is quickened — to 
 the regret of some people — by the same power which implies 
 intellectual progress. Men may be as licentious, and in 
 some w^ays as selfish, in the most as in the least civilised 
 countries, but they also become more reluctant to inflict 
 pain, and open their ears to lamentations which were once 
 interpreted as idle sound. 
 
 16. Pleasure, again, in the sufferings of an enemy suggests 
 more complex considerations, but we may still distinguish 
 between taking pleasure in pain simply as pain, and that in 
 which pain is regarded as a necessary concomitant of some 
 other circumstances. When a man's interests are opposed 
 to my own, I wish for something which involves disappoint- 
 ment or vexation to him. Christians find pleasure in the 
 knowledge that their countrymen have killed, mangled, and 
 humiliated a large number of foreigners ; but the pain may 
 be imperfectly realised, or, if realised, realised as a drawback. 
 The generous enemy becomes capable of the true chivalrous 
 sentiment, and may desire a victory at the smallest possible 
 cost to his enemy. The barbarous sentiment implied in a 
 Roman triumph may have implied rather a want of the per- 
 ception that other people had feelings than a delight in 
 their suff'ering; and in an age when sympathy is wider this 
 delight becomes inexpressibly revolting. The most brutal 
 John Bull would hardly have wished to expose Napoleon to 
 insult as well as humiliation. It is simply reasonable in this 
 sense to love one's enemies. I might wish to prove a rival 
 controversialist to be a fool, because a conviction of his 
 folly is necessary to my vanity, but I should be sorry to
 
 234 THE SCIENCE OF ETHICS. 
 
 hear that my bitterest or most successful antagonist was 
 suffering from a toothache. Antagonism, of course, reconciles 
 us to the pain of our adversaries, and even allows the thought 
 of that pain to be part of a pleasurable emotion. The sym- 
 pathetic pang which it produces as represented by our 
 imagination is swallowed up by a multitude of associated 
 feelings. We desire, again, that a man should suffer when 
 we feel that our security is dependent upon his suffering ; 
 and this sentiment enters for something into the moral desire 
 of retribution. It is a part, at least, of that sentiment that 
 the moral order would be out of joint if wrong-doing did 
 not lead to pain. But this sentiment is compatible with, 
 though it is not necessarily combined with, a horror of inflict- 
 ing useless pain upon any one. Wherever the pain, that is, 
 is not essential as a deterrent, we have so far less desire for 
 its infliction. I should be heartily glad to know that the 
 most detestable criminal had, by some accident, become 
 insensible to the punishment which I think it right should 
 be inflicted. I do not assert that this is a necessary or even 
 the general feeling ; but so far as we regard punishment as 
 useful, that is, as having consequences productive of happi- 
 ness to others, we do not desire pain in itself. Whatever 
 other motives may be operative, this is one ingredient in 
 tbc desire for justice, and even to some extent in the desire 
 or vengeance ; and, so far as it exists, we must admit that 
 pain is only desired for extrinsic motives. It is not considered 
 as a good thing in itself, but as part of the conditions of 
 good to others than the sufferer. 
 
 17. Fresh complexities are introduced when we come to 
 the enmity which implies not a mere antagonism of interests 
 but a personal dislike to others. Sympathy, in the sense in 
 which I am using the w^ord, may give rise to antipathy. 
 We are led to detest a man's character because we can partly 
 share his feelings. The saint and the sensualist can each 
 enter into the motives of the other. The saint is still 
 accessible to the brutal passions which it has been the labour 
 of his life to master. It is the wild beast within him which
 
 SYMPATHY. 
 
 -JS 
 
 he sees incarnated in another agent, and which awakes his 
 horror and disirust. Conversely the sensuahst mav see in 
 the saint the triumphant conscience which can still inflict 
 pangs of remorse upon himself, though it cannot restrain his 
 conduct. There are conflictino; elements in the character 
 of every man, and parts of ourselves which we regard with 
 horror in memory, though under some special stimulus they 
 may overpower all restraining motives. We can sympathise 
 with other men, that is, realise their feelings in imagination, 
 because their character contains the same primary instinct ; 
 and this sympathy gives rise to admiration or contempt as 
 the consciousness of our own qualities gives rise to vanity or 
 humiliation, when the action shows that the dominating 
 motives differ in certain ways from those which, at the time 
 of reflection, appear to us to be natural and becoming. We 
 may thus come to regard a man as a mischievous agency in 
 the world, as predestined by his very constitution, and not 
 from mere accidents of circumstance, to thwart and humiliate 
 us, and as embodying those sentiments which we detest the 
 more heartily as we can realise them the more vividly. He is 
 a nuisance to be abated, a corrupting or discordant element 
 in the general system of things ; and therefore we must take 
 pleasure in conditions which necessarily involve his suffering 
 or destruction. 
 
 1 8. I do not attempt to give any analysis of such cases. 
 It is enough to say that in the complex mechanism of human 
 motives we may often come to results which apparently 
 conflict with the principles from which they are deduced. 
 Even in such personal antipathies the sympathy is the 
 fundamental fact. The hatred which is generated is always 
 a more or less painful emotion, because our spontaneous 
 sympathy leads in any case to some conflict of motives. We 
 cannot hate the man without feeling that some of our own 
 feelings are taking part against ourselves. And, further, the 
 feeling of hatred is perfectly compatible with an entire absence 
 of anything like delight in pain. We may simply desire to 
 keep the disagreeable person at a distance, to restrain his
 
 236 THE SCIENCE OF ETHICS. 
 
 activity or divert it into a harmless channel^ to convert him 
 to better modes of feeling and so forth. It is only when we 
 are so related that our satisfaction necessarily implies his 
 misery that we are tempted actually to desire his suffering. 
 The distinction between hating the sin and hating the sinner 
 is often hvpocritical enough^ but it also expresses the rational 
 conviction that all pain is in itself bad aiid painful to con- 
 template, though it may be inseparably connected with desir- 
 able results. 
 
 19, But it remains to be admitted that there is apparently 
 such a thing as pleasure in the pain of others — pure malignity, 
 which we call devilish, to mark that it is abnormal and signi- 
 ficant of a perverted nature. The existence of such a feeling 
 is a puzzle such as that which psychologists have discussed 
 under the name of the ^Muxury of grief." Sentimentalists 
 seem, at any rate, to delight in cultivating sorrow, which is 
 apparently a still more contradictory state of mind than 
 delight in the sufferings of others. The explanation, so far 
 as it need be considered here, seems to depend upon the fact 
 that there is a certain pleasure in almost every kind of excite- 
 ment. We like what relieves the dulness of our lives and 
 provides some channel for emotional discharge. The sluggish 
 and brutal nature delights in the stimulus of horror ; in spec- 
 tacles of blood and death, even though it would appear at 
 first siffht that heio-htened excitement meant necessarilv a 
 heightening of disgust. Men apparently humane and sensitive 
 have taken delight in executions. The Romans took pleasure 
 in the sight of dying gladiators ; Spaniards, in the sight of 
 mangled bulls and horses ; an English mob is fascinated by 
 a sickening accident in the streets ; and possibly we may 
 trace a remnant of the same feeling in the pleasure given by 
 the horrible in tragedies or by "sensational" incidents in 
 modern romance. We have, again, the more hideous cases 
 in which cruelty seems to afford a kind of voluptuous delight, 
 as in some historical monsters who made an art of torture. 
 Almost any pungent sensation, though provocative of sheer 
 disgust to the sensitive, seems to yield a kind of pleasure to
 
 SYMPATHY. 237 
 
 some natures. The problem must be left to the psychologist. 
 It is enough here to say^ first, that in all such cases the pain, 
 whether original or reflected, is but one strand in a highly 
 complex thread of feeling, and may produce its efl'ect as a 
 counter-irritant, or as heightening other sensations which are 
 in alliance with it ; and, secondly, that it is in any case a 
 comparatively rare and abnormal phenomenon, due to some 
 morbid condition of the faculties, or perhaps to the survival 
 of ferocious instincts from times when the intellect and 
 the sympathies were comparatively dormant. Sympathy is 
 the natural and fundamental fact. Even the most brutal 
 of mankind are generally sympathetic so far as to feel rather 
 pain than pleasure at the sight of suflering. The scum of a 
 civilised population gathered to pick pockets on a racecourse 
 would be pained at the sight of a child in danger of being 
 run over or being brutally assaulted by a ruffian, and would 
 be disposed to rescue it, or at least to cheer a rescuer, luiless 
 their spontaneous emotion were overpowered by some ex- 
 trinsic sentiment. 
 
 20. If this account of the sympathetic emotions be approxi- 
 mately accurate, we see that sympathy is implied in all 
 thoughts about others. Though it generates antipathy and 
 discord in numberless cases, the underlying and governing 
 process is sympathetic. We may say that we think about 
 other men by becoming other men. We appropriate pro- 
 visionally their circumstances and emotions. Metaphysicians 
 and mystics have expressed this by denying the ultimate 
 validity of individuality, and by saying that in some trans- 
 cendental sense a man is his neighbour, or that all men are 
 manifestations of one indivisible substance. The lantjuao'e, 
 though to my mind untenable, may serve to express the fact. 
 So far as I svmpathise with you I annex your conscious- 
 ness. I act as though my nerves could somehow be made 
 continuous with yours in such a w'ay that a blow which fell 
 upon your frame would convey a sensation to my brain. 
 Undoubtedly we must add that this current, so transmitted, 
 is greatly enfeebled in almost all cases. The reflex feeling is
 
 238 THE SCIENCE OF ETHICS. 
 
 normally far less acute than the direct. The thought of the 
 pains of starvation does not produce a pain at all comparable 
 to starvation itself. And as the represented object is distant 
 in time and space, the sensibility becomes rapidly dulled. 
 Most men have great difficulty in forming any vivid repre- 
 sentation of distant suffering. The actual sight of a stranger in 
 agony gives a keener feeling at the moment than the image 
 of a brother dying at the antipodes ; and the most bene- 
 volent of men hears with great composure of the destruction 
 of millions in China. It may, on the other hand, be remarked 
 in passing, that the suffering of another person may stimulate 
 a sympathetic person under certain conditions more forcibly 
 than similar sufferings of his own. The image may inci- 
 dentally set in motion a whole current of accumulated feeling. 
 A man who has been watching the sickbed of a wife may 
 be more moved by an accident to her than by one to himself, 
 not because the sight of her pain is keener than his own 
 pain in itself, but because it fires a whole train of anxieties, 
 hopes, and fears already prepared for explosion. So I may 
 take enormous trouble to give a very slight pleasure to a 
 man whom I like, not because I feel his pleasures more than 
 my own, but because the desire to do him honour is so strong 
 that I am glad to find any vent for it, however trifling in 
 itself. 
 
 21. Finally, if this be the true account of the process, the 
 difficulty is not to understand why the thought of your pain 
 !-hould give me pain, but to understand how it should ever 
 give me pleasure. It is not more true that to think of a fire 
 is to revive the sensations of warmth, than it is true that 
 to think of a man is to revive the emotions and thou2;ht 
 which we attribute to him. To think of him in any other 
 sense is to think of the mere doll or statue, the outside 
 framework, not of the organised mass of consciousness which 
 determines all the relations in which he is most deeply in- 
 teresting to us. llie primary sympathy is of course modified 
 in a thousand ways — by the ease or difficulty with which wc 
 can adopt his feelings ; by the attractiveness or repulsive-
 
 ALTRUISM. 239 
 
 ness of the feelings revealed ; by the degree \n which circum- 
 stances force us into co-operation or antagonism ; and by 
 innumerable incidental associations which make it pleasant 
 or painful to share his feelings. Tf by sympathy we mean 
 this power of vicarious emotion^ it may give rise to anti- 
 pathv, to hatred, rivalry, and jealousy, and even to the 
 diabolical perversion of pleasure in others' pain ; but the 
 direct and normal case is that in which sympathy leads to 
 genuine altruism_, or feeling in conformity with that which 
 it reflects. 
 
 III. Altruism. 
 
 22. We may now return to the original problem, what 
 is implied in unselfish or altruistic conduct? Sympathy, in 
 the sense explained, is not identical with altruism, but it is 
 the essential condition of altruism. I cannot be truly altru- 
 istic, that is, until the knowledge of another man's pain is 
 painfid to me. That is the groundwork of the more complex 
 sentiments which are involved in all truly moral conduct, 
 morality implying the existence of certain desires which have 
 for their immediate object the happiness of others. I have 
 tried to consider briefly the nature of this underlying senti- 
 ment. We have now to say precisely in what sense it leads to 
 self-sacrifice ; but we have still to get rid of certain ambigui- 
 ties which perplex the discussion before giving the answer, 
 which is in itself, as I hold, sufficiently clear. 
 
 23. How does altruistic conduct differ from that which is 
 not altruistic? Obviously (if my theory be sound) it does 
 not differ in any sense which would imply that my conduct 
 can ever spring from anything but my own feelings. So far as 
 my actions can be said to be determined by anything else, they 
 are not properly my actions. I am in such cases part of the 
 mechanism set in action by some external force, whether it 
 be the will of another agent or some mechanical circumstance. 
 Mv limbs arc for the time part of another man's limbs. 
 Voluntary action, or action determined by the motives of the 
 agent, is the definition of what is strictly conduct. I may be
 
 240 THE SCIENCE OF ETHICS. 
 
 prompted by pains and pleasures which represent those of 
 another man, but they must not the less be my pains and 
 pleasures. Hence we must exclude two alternative errors 
 resulting from the neglect either of the objective or the sub- 
 jective conditions of conduct, and therefore of the fact that 
 both must always be present. We may speak, for example, of 
 a man as preferring the pleasure of another to his own. We 
 must in such a case be understood to mean, not that the 
 motives of the other take the place of his own motives, 
 which is as absurd as to say that the food eaten by the other 
 nourishes his own organs, but that the sympathy is stronger 
 than other conflicting motives. When, for example, Sidney 
 gives the water to the soldier, it is not because Sidney 
 actually feels the soldier's thirst, but because Sidney's sym- 
 pathy for the soldier's sufferings is a stronger motive than his 
 own thirst. Sidney's conduct, as that of the most selfish 
 man, is always determined by his own feelings ; but in his 
 case the sympathetic feelings have so great a share in de- 
 termining conduct, that his compassion is stroiiger than his 
 thirst. Normally, indeed^ we may say that the reflected feel- 
 ing will be weaker than the original, the feeling produced 
 by the thirst of another than the feeling due to my own 
 thirst ; but this direct sympathy may be enforced by others, 
 by a sense of duty, justice, and so forth, so as to have a 
 greater effect upon the conduct. In any case, so far as it is 
 operative, it must be a feeling of the agent, and does not 
 imply that he acts without feeling or is moved by another 
 man's motives, but that the feeling which is due to his 
 knowledge of another man's feelings is abnormally strong. 
 
 24. On the other hand, it is equally erroneous to speak of 
 motives as being '' selfish" in any other sense than that 
 already implied, simply because they are my own. This fallacy 
 has already been noticed. If I say, " I dislike the taste of 
 wine because it is unpleasant to me/' I either say the same 
 thing twice or I use words in different senses. I may mean 
 to imply that I dislike the taste because the immediate sensa- 
 tion is disagreeable, and to exclude the hypothesis that I dis-
 
 ALTRUISM. 241 
 
 like the taste because I think it an indication of unwhole- 
 some qualities, or have some accidental association which 
 overpowers the pleasantness of the taste. In that case, I use 
 '^dislike" to include feeline^s different from the immediate sen- 
 sation, and '^ unpleasant ^^ to denote the sensation alone; and 
 so far my statement may be reasonable, though ambiguously 
 expressed. Now the egoist sometimes falls into a similar 
 ambiguity in discussing the question of altruism. To say, " I 
 dislike your pain," and to say, " Your pain is painful to me," 
 is to say the very same thing in different words ; but the 
 second statement is something given as an explanation of 
 the first, and not as a simple inversion. The inference thus 
 insinuated is that I dislike your pain because it is painful to 
 me in some special relation. I do not dislike it as your pain, 
 but in virtue of some particular consequence, such, for 
 example, as its making you less able to render me a service. 
 But this is really to assert that your pain does not give me 
 pain except as a link in a chain of events which brings about 
 some other disagreeable consequence. In that case, I do not 
 really object to your pain so far as it is your pain, but only 
 by some removable and accidental consequence. What I 
 really dislike is that consequence, whatever it may be; and 
 thus the statement that I dislike your pain becomes perverted 
 into the assertion that I dislike something else ; or, in other 
 words, it is inferred that sympathy is a mere delusion. 
 
 25. This, indeed, is expressly asserted by some psychologists, 
 who resolve sympathetic emotions into a product of associa- 
 tion, and explain regard and dislike to the suffering of others 
 as a case of dislike to the means which survives, when we 
 have forgotten, for a time at least, the ends to which they 
 originally owed their attractiveness. If I am right in the 
 foregoing argument, that is a totally inadequate explanation 
 of the phenomenon. The pain due to the pain of others is a 
 direct and necessary result of the very process of thinking 
 about others. A process of association is no doubt implied, 
 in so far as it is only by association (so at least I should say) that 
 we can learn to interpret certain sounds, sights, and so forth, 
 
 a
 
 242 THE SCIENCE OF ETHICS. 
 
 as indicative of the emotions of others. But we must so 
 interpret them in order to reason at all about the world of 
 thought and feelings and in so interpreting them we learn to 
 sympathise. Your pain is not painful to me because I infer 
 that some other consequence will result to me^ but because 
 the thought of your pain is itself painful. 
 
 26. This^ indeed, introduces another consideration which 
 must be taken into account. It is, in fact, true that vour 
 pain can only be painful to me under a certain condition ; 
 the condition, namely, that I must know of it or believ^e in 
 it. The sympathetic emotions, in other words, are clearly 
 dependent upon a reasoning process, which cannot be said 
 of some other feelings ; and we may ask how far this may 
 suggest any distinction between the altruistic and the non- 
 altruistic emotions. It does obviously suggest a distinction 
 which is ofttimes of great importance, and which shows that 
 there is a gap between simple sympathy and fully developed 
 altruism. We may, in fact, admit that your pain may be 
 intrinsically painful to me without admitting that I, there- 
 fore, become altruistic in the fullest sense. Sympathy may 
 establish only a temporary coincidence, not a permanent 
 identity of interest. You and I are at one so far as it is true 
 that the relief of your suffering would relieve me ; but we are 
 not really one, and therefore my suffering may be relievable 
 by means which would not relieve yours. If we were in- 
 separably united ; if, for example, we were confined in a 
 single cell, so that I necessarily had your sufferings con- 
 stantly before my eyes, and could not get rid of the svm- 
 pathetic pain without getting rid of its cause, the original 
 pain, our interests would be so far identical, and it would 
 seem to be an unimportant subtlety to consider whether my 
 desire for your comfort were properly to be called selfish or 
 altruistic. The same conduct would be dictated on either 
 hypothesis, as our interests would Ijc virtually identical. It 
 is clear, however, that this identity can never be perfectly 
 realised. The pain given by your pain may simply induce 
 me to shut my eyes. The Pharisee who passed by on the
 
 ALTRUISM. 243 
 
 other side may have disHked the sldit of the wounded 
 traveller as mueh as the good Samaritan. Indeed the sight 
 of suffering often directs irritation against the sufferer. 
 Dives is often angry with Lazarus for exposing his sores 
 before a respectable mansion ; and sometimes goes so far as 
 to think^ illogically perhaps^ that the beggar must have cul- 
 tivated his misery in order to irritate the nerves of his neigh- 
 bours. To give the order, "Take away that damned Lazarus/' 
 may be as natural an impulse as to say, '^Give him the means 
 of curing his ailments." 
 
 27. The fact thus stated is undeniable. It must be 
 observed that the limitation which it implies does not apply 
 to the sympathetic motives, but to every instinct of our 
 natures, in so far as they involve a belief in the distant or 
 the future. We do not wonder that a man should continue 
 to suffer from a disease because we see at once that he cannot 
 help it; but we think of a sympathetic emotion as some- 
 thing which can be helped. We can, we say, dismiss or 
 entertain a thought at our pleasure. It is painful to think 
 of a neighbour's disease. Then cease to think of it. The 
 remedy is in every one's hands. Why not adopt it ? To 
 this we may answer, in the first place, that, as a matter of 
 fact, it is very generally adopted. People reconcile them- 
 selves very quickly to the misfortunes of others, and pre- 
 cisely by ceasing to think about them. Not only so, but 
 in many cases the remedy is not only common, but often 
 irresistible, and often (though I am perhaps anticipating) 
 perfectly consistent with morality. I do not worry myself 
 about the bad government of Timbuctoo much more than I 
 worry mvself about the uninhabitable condition of the moon, 
 and for the same reason — that I can do nothing to improve 
 either. It would seem to be a general law that feelings 
 which do not or cannot produce any effect upon conduct 
 tend to become faint, and ultimately to disappear. And, 
 morally speaking, deliberate indulgence in emotions of the 
 painful kind at least, which bear no fruit in action, 
 is so much w-astc of power, and so far condemnable. If
 
 24+ THE SCIENCE OF ETHICS. 
 
 I am dreaming" about the millennium or fretting about the 
 evils of Chinese despotism, I am throwing away energies 
 which might improve the pauperism of London or contribute 
 to the social enjoyments of my next-door neighbours. But 
 to say generally that I can annihilate my sympathies because 
 they give me pain is clearly untrue. I can only abolish 
 thoughts when there is a sufficient motive to lead me into a 
 different train of thought. I can perhaps get rid of the 
 thought of my neighbour's suffering more easily than I can 
 get rid of certain material conditions. But the simple fact 
 that a particular emotion is dependent for its existence upon 
 an intellectual state does not enable men to suppress it. I 
 should perhaps be happier if I could forget that a surgeon 
 was in the next room ready to operate upon me in an hour, 
 but I cannot therefore fix ray mind upon a novel. The 
 general who broods over possible defeat after giving his 
 orders, the speculator who has a foretaste of ruin which he 
 cannot avoid, are on the highroad to suicide. Yet the 
 actual pain, added to the knowledge that the reflection only 
 aggravates the evil, leaves men unable to distract their 
 minds, or to refrain from drinking the bitter cup by antici- 
 pation as well as in reality. The sympathetic emotions are 
 equally potent. When a blow is hanging over my familv, 
 when I see symptoms of deadly and incurable disease in wife 
 and child, I cannot dispel my melancholy, however clearly I 
 know it to be useless. It must be added that, although I 
 have spoken of sympathy with pain, partly because it is the 
 keenest and most conspicuous phenomenon, it is also true 
 that great part of our pleasure is dependent upon sympathy, 
 and that the two are inseparably associated. If I am to live 
 with my friends, I must share both their joys and sorrows; 
 and the real question which I have to decide is not whether 
 I will drop a particular pain, but whether I will or will not 
 live the wider or the narrower existence. If I can abstract 
 my mind from thoughts of danger to my wife and child, I 
 also must give up all the enjoyment which is involved in the 
 close companionship.
 
 ALTRUISM. 245 
 
 28. The true statement would therefore recognise the fact 
 that emotions are inevitable, whether sympathetic or not, 
 in proportion not simply to the pain and pleasure at the 
 moment, but to the intensity and to the degree in which 
 they form part of my world — the world which is constituted 
 not by the mere sensations, but by the whole system of 
 thoughts and emotions sustained by the framework of per- 
 ception. I can no more strike out at will a fragment of the 
 world which is recos^nised throutrh the intellect, than of that 
 which is directly revealed through the sensations. The two 
 form a continuous whole, which is only modified in a sub- 
 ordinate degree by the shrinking from pain or the absorption 
 in pleasure. An emotion closely bound up with some vivid 
 sensation or perception from which I cannot free myself 
 is so far the more inevitable. This is equally true whether 
 sympathetic feeling is present or absent. As a fact, it is 
 generally easier to get rid of a sick friend than of a tooth- 
 ache; but the pain of suffering with him may generate the 
 desire to relieve his sufferings rather than the desire to forget 
 them, if his life is so bound up with my own that the selfish 
 remedy is in fact impossible under existing conditions, or if 
 the action of desertion appears to me as so repulsive in itself 
 that the pain of the sympathy is overpowered by the pain- 
 fulness of actino; badlv. 
 
 29. This consideration shows that the degree in which our 
 happiness is associated with the happiness of others is closer 
 than we might at first sight suppose. It may be necessary^ 
 to my happiness that I should relieve Lazarus, not only 
 when he is actually present, or when I foresee some ill con- 
 sequences to me from his misfortunes, but also when I am, 
 for any reason, unable to dismiss the thought of his suffer- 
 ings. In many cases this may be impossible without such 
 a dislocation of my whole system of thought and feeling as 
 may, for some reason, be impossible. But we cannot yet 
 say that my conduct is altruistic until wc know what is the 
 condition which makes it impossible for me to separate my 
 interests from those of the other. I am still, it mav be
 
 246 THE SCIENCE OF ETHICS. 
 
 ur2;ed, as selfish in desiring relief from the vicarious as from 
 the original pain. I desire a fire as warming and a friend's 
 happiness as cheering ; the desire for warmth could not 
 prompt a desire for a fire which could not warm — say, a 
 fire in the moon — except by a mistaken inference or an 
 arbitrary association of ideas ; and similarly it may be said 
 that sympathy can only prompt the desire for a friend's 
 happiness in so far as it cheers me. That is to say, it is still 
 my own happiness which I desire. The association of my 
 own happiness with that of my friend's is still extrinsic. I 
 am sympathetic but not truly altruistic. If and wherever 
 they can be separated, therefore, I shall be decided simply 
 by the consideration of the consequences to me. Wherever 
 it is possible to obtain relief from the sympathetic pain by 
 abolishing Lazarus instead of making him happy, I shall 
 abolish him. This, it may be said, remains equally true 
 whether the tie which binds me to Lazarus is such as does 
 or does not involve conditions dependent upon any intel- 
 lectual inferences and beliefs. I shall always, in any case, 
 prefer that course of conduct which is possible (and that of 
 course is always a condition), and which promises the greatest 
 happiness. 
 
 IV. The Rule of Cofidud. 
 
 30. This is the question, then, which must be discussed in 
 order to bring out the real meaning of the question at issue. 
 So far, in fact, as we have hitherto gone, we have not recog- 
 nised any difference between the conduct which does and that 
 which docs not imply the presence of sympathetic motive. 
 In both cases there is a subjective and an objective condition ; 
 in both cases I am prompted by my own feelings to do what 
 is pleasantest to myself. I do so equally whether I drink 
 the water myself or give it to the sufferer by my side. 
 Where, then, is the difference ? Since it cannot be in the 
 fact that in one case I have and in the other I have not 
 feelings, it can only be in the different law of the feeling.
 
 THE RULE OF CONDUCT. 247 
 
 In both cases I do what makes me happy^ but that which 
 makes me happy is very different in the two cases. A desire 
 for warmth can only prompt a desire for the fire which 
 warms, and a desire for the happiness of another only the 
 happiness which cheers. The statement is the precise equi- 
 valent of the other, if we keep strictly to the same meaning. 
 But when we ask what will be the law of the feeling, we see 
 at once that there is an important difference. What happi- 
 ness will cheer me ? Any happiness in which I believe and 
 which I can realise. Time and distance have no significance 
 to me except as diminishing the vividness of the impression. 
 If it is painful to me to realise your suffering when I see the 
 knife cutting your flesh, it is painful to me in a certain 
 proportion to think of the same torture to you in a distant 
 region. As non-sympathetic, I can desire only the fire 
 which warms me and which will warm me ; as sympathetic, 
 I desire the fire which warms you in the arctic regions or 
 provides warmth for a distant posterity. My feelings are still 
 my own in either case, but in the latter case they may prompt 
 me to conduct — such, for example, as economy in fuel for 
 the sake of my grandchildren's hearths — from which I shall 
 derive no benefit. Hence, so far as sympathy is real, it obeys 
 a law which has no necessary reference to any future state 
 of my own ; it may operate powerfully even in opposition to 
 any prospect of my happiness to come. The present pain is 
 the reflection of pain which depends upon conditions which 
 have no definite or uniform relation to my future happiness, 
 nor, therefore, to the total happiness which I contemplate 
 from a given course of conduct. 
 
 31. With these explanations we may come to the direct 
 issue. Admitting that my conduct must always be con- 
 ditioned by my feelings — by my aversion to painful and 
 attraction to pleasurable states — are my feelings necessarily 
 determined by the balance of anticipated pain and pleasure ? 
 Does the conviction that a certain course of conduct will 
 obtain for me the maximum of pleasure determine me neces- 
 sarily to adopt it ? Does that action always make me
 
 248 THE SCIENCE OF ETHICS. 
 
 happiest which promises most happiness to me? If these 
 statements are mutually equivalent^ it would seem that 
 sympathy must be an illusion, and that I can really desire 
 another man's happiness only so far as it is a means to my 
 own happiness. I hold that the inference is wrong, and that 
 the two statements which are regarded by the egoist as 
 identical are really incompatible ; and this, I think, is implied 
 in the foregoing arguments. But the point is of such critical 
 importance that I must try to bring out the contrast more 
 clearly. 
 
 33. The problem, as I have said, is shortly, What is the 
 " law " of motive ? May it always be described as a desire 
 for the greatest happiness of the agent? Let us ask first 
 whether it can ever be so described. The unreasoning 
 animal acts from blind instinct; his judgments, so far as he 
 judges, are limited to the immediate facts ; he judges or sees 
 this to be a fire, that to be a devouring animal, and so forth ; 
 but he has no prevision of the remoter consequences, and is 
 therefore neither selfish nor unselfish, for we can only predi- 
 cate selfishness where there is at once a knowledge and a 
 disregard of the feelings of others. A distinction, however, 
 may already be made in so far as he possesses instincts which 
 are in consequence, though not in intention, profitable to 
 others. Animals, as I have sufficiently said, possess instincts, 
 such as motherly love, non-essential, and in many cases, and 
 even in the average case, prejudicial to the individual, and 
 yet essential to the race. These instincts, therefore, must be 
 developed as the race thrives, and since the animal has no 
 prevision or only the most rudimentary prevision of conse- 
 quences, he will act without conscious regard to the con- 
 sequences. If, in the next place, we suppose the animal to 
 become enlightened so far as to be able to trace remoter 
 consequences — or, in other words, to contemplate the distant 
 and the future as well as the immediate — but without any 
 correlative extension of sympathies, the result would be a 
 limitation of these instincts. The instinct, indeed, would 
 not be abolished ; pleasure might still be derived, for example.
 
 THE RULE OF CONDUCT. 249 
 
 from the exercise of the maternal functions ; but such a 
 pleasure would be in the same category as any of the purely 
 sensual pleasures. A perception that drinking brandy is 
 mischievous — that is, productive of future suffering — does 
 not annihilate the pleasure of brandy-drinking, but it tends 
 to limit the indulgence by introducing the foretaste of misery 
 to come. The instincts of the non-sympathetic agent would 
 in the same way be limited so far as their operation was 
 normally productive of unpleasant consequences ; and, as 
 a matter of fact, we may observe frequent exemplifications 
 of this principle. In some societies the unwillingness -of 
 women to accept the burdens of maternity is proportioned 
 to their intelligence ; the more thoughtless continue to act 
 upon animal instincts which involve ultimate self-sacrifices, 
 \vhilst the more thoughtful restrain the instinct from purely 
 selfish considerations. 
 
 ^^. That this, indeed, cannot be the normal case follows 
 from what has been said ; for, if I am right, the intellectual 
 development must normally coincide with a development of 
 the sympathies, and whatever deflections from this coinci- 
 dence may be possible will be limited by the operation of 
 our general principle ; for if the increased reasoning power 
 meant a diminution of social qualities, the intellect would 
 exercise a disintegrating and enfeebling influence. The 
 more reasoning society would tend to be supplanted by that 
 which may be called lower so far as less intelligent, but 
 which would be superior in so far as better fitted for the 
 conditions of life. The perception of utility would in such 
 a case, as I have said, be in conflict with the general con- 
 ditions of utility. If cleverness carried with it, or so far as 
 it carries with it, inferior sociability, it is or would be a 
 mischievous factor, and would tend to be eliminated : the 
 world would be to the stupid. We may say, therefore, that 
 the intellectual development must at least carry with it 
 somethino; which counterbalances this anti-social tendency. 
 In some way, then, this anti-social tendency must be 
 counterbalanced, and this may be done to some extent
 
 250 THE SCIENCE OF ETHICS. 
 
 without assuming any increase of sympathy, for the enhght- 
 ened instinct would reveal not merely the disadvantages but 
 the advantages of acting for others, and, though it would 
 discourage self-sacrifice, would encourage such action for the 
 good of others as would bear fruit in good for the agent. 
 The difficulty of the question depends upon the intricate 
 intermixture of these intrinsic and extrinsic motives to 
 altruistic conduct, and we must of course admit with the 
 egoist that the extrinsic motives to social conduct exist, and 
 are of great importance, though we deny that they explain 
 the whole phenomenon ; and this is the point which has to 
 be made clear. 
 
 34. Now the general principle, which may be called the 
 fundamental axiom of prudence, the rule, namely, ''Act so as 
 to obtain the maximum of happiness," does not, as we see, 
 hold strictly even of those actions in which there is no admix- 
 ture of altruistic motive. It cannot, therefore, have that 
 absolute or a priori character which is sometimes claimed for 
 it. In order that it may be approximately verified, a con- 
 dition is requisite which may or may not be fulfilled. The 
 conduct must always be determined in this, as in all cases, by 
 the actually existing feelings. These feelings, again, may 
 include a foretaste of future pleasures and pains. But, as 
 a general rule, the influence of the future pleasure is less 
 than the influence of the immediate pleasure, the degree in 
 which it is less depending upon the constitution of the agent. 
 For the unreasoning agent the future is simply non-existent; 
 but even for the reasoning agent it does not necessarily 
 follow that conduct will correspond to calculation. If he 
 knows to a certainty that a present sacrifice of pleasure will 
 be repaid in kind by pleasures to come, he may still be 
 unwilling to make the sacrifice — a case which is daily ex- 
 emplified. He does not care, we may say, for the future self 
 as nmch as for the present self. It is true, however, that, so 
 far as he reasons, he tends to regard the future as well as the 
 present, and therefore, we may say, approximates to an adop- 
 tion of the prudential axiom. 71ie more reasonable he is, the
 
 THE RULE OF CONDUCT. 251 
 
 closer he comes to it ; but it still represents an ideal limit 
 never actually realised, and implying a corresponding 
 balance of passions which may be more or less perfectly 
 achieved. And even the approximation is possible in virtue 
 of the normal conditions of life. It is only in rare cases that 
 we have to make the kind of choice suggested between an 
 isolated pleasure in the future or one in the present. If I 
 had a fixed number of cakes, I might ask whether I should 
 obtain the maximum pleasure by eating all to-day or leaving 
 part for to-morrow. The answer would be determined by 
 the relative power of the immediate appetite and the fore- 
 taste ; and the stronger my reasoning, the greater the proba- 
 bility that the perception of the maximum happiness would 
 represent the governing motive. As a rule, however, the case 
 is simpler. What is pleasantest now is also most productive 
 of pleasure. To eat my dinner to-day is normally a condition 
 towards eating a dinner to-morrow. My appetites have 
 already been regulated when I set up as a reasoning being. 
 Each had a place in my system, and a force proportioned to 
 its normal utility. The appetite for food already approxi- 
 mately corresponds to the need for food ; that is, to the 
 importance of the function in the whole system of life. To 
 satisfy the appetite is therefore to satisfy the conditions of 
 health, and therefore of maximum enjoyment. The reason 
 finds the problem already approximately solved, and has 
 only to work out a closer approximation. It starts with 
 instincts already harmonised, not antagonistic; and there- 
 fore in fulfilling the dictates of pleasure I am already act- 
 ing with an unconscious reference to the needs of my whole 
 life. The instincts have been moulded by the conditions, 
 though they do not consciously attend to them. What is 
 true (as I may note in passing) of the balance between the 
 primary instincts which do not involve sympathy holds true 
 equally when sympathy is introduced. Normally it is prudent 
 to be virtuous, a point which will have to be considered more 
 fully hereafter. 
 
 ^^. Meanwhile we may say that, as a rule — leaving the
 
 252 THE SCIENCE OF ETHICS. 
 
 question of sympathy out of account — an increase of reason 
 implies an approximation to the prudential maxim ; that is 
 to say, that, as a fact, the working of the instincts or feelings 
 which dictates conduct approximately coincides with the 
 prevision as to the maximum of happiness obtainable by the 
 agent. This, however, is not an a priori principle, depen- 
 dent or incapable of being denied without contradiction, 
 but a deduction from the general conditions of life and the 
 mode of development of the faculties ; and the closeness of 
 the approximation depends upon extrinsic conditions. The 
 bare foresight that I am sacrificing the maximum of pleasure 
 to an immediate pleasure certainly does not make yielding 
 to temptation impossible. It only makes it unreasonable 
 in a sense in which unreason is thoroughly possible. I may 
 know that I should gain a greater amount of happiness if 
 the door of a public-house were locked ; but the knowledge 
 is not equivalent to locking the door. Still, so far — that is, 
 excluding all question of sympathy — we accept the maxim as 
 expressing the general law of the operative motive in reason- 
 able beings in proportion as they are reasonable. In all 
 cases the conduct depends upon the actual mechanism of 
 motive, but that mechanism is so arranged that it is possible 
 for the immediate conduct to act in conformity with the 
 formula. 
 
 ;^6. We now have to introduce the sympathetic motives, 
 and to inquire whether the same formula is applicable or 
 approximately applicable. The sympathetic feeling, again, 
 is one of which it is the law that your pain is therefore pain- 
 ful to me. The reasoning agent, in so far as not altruistic, 
 suffers from the knowledge that he will suffer hereafter, and, 
 so far as altruistic, he suffers from the knowledge that some 
 one else is suffering or is about to suffer. And, again, I have 
 already remarked that sympathetic motives are not necessarily 
 contrasted with others in respect cither of their consequences 
 or of the fact that they are still feelings of the agent. The 
 sympathetic motives, so to speak, always develop within the 
 framework already provided by the other motives^ and the
 
 THE RULE OF CONDUCT. 253 
 
 two arc often inextricably blended. The external bond of 
 maternity, implying no affection of the feelings of the 
 spring, develops into the closer union in which the sympa- 
 thetic feeling becomes predominant, and both forms may 
 persist at any stage of development. The mother may still 
 admire the child as she admires a flower, independently, that 
 is, of any recognition of its consciousness, and this sentiment 
 may blend indissolubly with the same maternal love ; and, 
 again, the non-sympathetic feeling may prompt to a virtual 
 self-sacrifice. A passion for beauty regarded as a purely 
 external quality may lead to imprudence as well as the 
 higher motive of personal affection ; and, in both cases, again, 
 self-sacrifice is impossible in the sense of action against the 
 predominant motive. Maternal love is still a feeling of the 
 mother, and therefore we have only to suppose it strong 
 enough to make self-sacrifice in a given case impossible. For 
 the pang of inflicting injury upon the child might be so great 
 that no other pleasure of which the mother is capable could 
 repay it. In this case, then, it might be urged that the 
 highest self-devotion was no devotion at all, for it is still 
 obedience to the mother's own feelings. In other words, if we 
 regard consequences, the least sympathetic action may imply 
 self-sacrifice, whilst the most sympathetic may apparently be 
 still selfish. This confusion follows if we attempt to base 
 our distinction upon the bare fact, common to all actions, 
 that they have both subjective and objective conditions. 
 
 37. Let us return, then, to the other problem, what is the 
 law of the two classes of motives ? Omitting the sympa- 
 thetic emotions, we have seen that the reasonable agent 
 approximates to an acceptance of the prudential axiom. 
 Supposing, then, a case in which the sympathetic feelings do 
 not come into play — as, for example, a case in which we are 
 concerned with lifeless objects alone — the maxim can be 
 applied without difficulty. No conflicting motive opposes 
 itself to the rejection of any course from which we antici- 
 pate a balance of unhappiness. We may calculate badly or 
 we may be misled by association. We sometimes acquire a
 
 254 THE SCIENCE OF ETHICS. 
 
 kind of sentiment in regard even to inanimate objects, mis- 
 led, it may be, by a kind of childish anthropomorphism which 
 leads us to treat them as though they had feehngs, and to 
 preserve them even at some cost of happiness to ourselves. 
 But so far as our action implies any assumption of this kind, 
 we feel it to be unreasonable, and it tends to disappear as we 
 become more reasonable. We suppress it if we wish to act 
 consistently. We approach more closely to the only assign- 
 able rule of conduct in such cases, that which prescribes action 
 for a maximum of happiness. Since by hypothesis we are 
 the only persons concerned, we are the only persons whose 
 happiness can be reasonably taken into account. If the 
 maxim applies unreasonably, we must treat every case 
 in the same way. The mother must be ready to abandon 
 her child whenever she anticipates more misery than 
 happiness from the connection. This is not only the sole 
 rule when we have to deal with material objects, but it 
 is also the rule in so far as we treat sentient beings 
 without regard to their feelings. We throw aside shoes 
 which hurt our feet as soon as we can get better shoes. 
 We may consider human beings as tools, and treat them in 
 the same way. We may discharge an old servant who can 
 no longer do his work without regard to his consequent 
 starvation. If we had no sympathy, this w^ould be the sole 
 rule of action. The same rule, again, is possible in regard 
 to objects (if we may call them objects) to which we do not 
 attribute actual existence. We derive much pleasure from 
 sharing in imagination the sorrows and joys of fictitious 
 persons. We may follow the histories of Juliet or of Jeanie 
 Deans with an interest not differing in kind from that which 
 we feel for real human beings ; and in this case (assuming this 
 pleasure to be our sole motive for reading) the only rational 
 principle will be to dismiss the fictitious persons from our 
 thoughts, so far as we are able to do it, directly the imagina- 
 tion gives us more pain than pleasure. The more reasonable 
 we are, the more consistently we shall obey this principle, 
 though, of course, habit and accidental associations often
 
 THE RULE OF CONDUCT. 
 
 -33 
 
 make our observance of it uncertain. And here, again, it is 
 possible and common to act in the same way in regard to 
 human beings — to treat them as mere dreams which pass 
 from existence as soon as they pass beyond our sphere of 
 observation. We read habitually of events in a distant 
 country as we read of events in a novel, with little more 
 belief in their objective existence, that is, in their existence 
 independently of our consciousness. Too often we treat even 
 our friends as old kings treated their fools — as sources of 
 amusement, to be annihilated for us as soon as they cease to 
 be amusing. To do so is to treat the object as though it 
 were unreal or existed solely in relation to us. 
 
 38. The difference, then, between the two cases is, in this 
 respect, sharply marked, and corresponds to the principles 
 already laid down. To believe in the objective existence of 
 anvthing is to believe that it exists independently of my 
 feelings — to believe that it is still there when I shut the eyes 
 of my body or of my mind. To believe in the existence of 
 a sentient being is to believe that it has feelings which may 
 persist when I am not aware of them. A real belief, again, 
 implies that at the moment of belief I have representative 
 sensations or emotions corresponding to those which imply 
 the actual presence of the object. Again, a material object 
 has an interest only so far as it is a condition of some kind 
 of feeling, and, when the sympathies are not concerned, of 
 some feeling of my own, whether implying or not implying 
 any foretaste of the future. To take any interest in any 
 material object, therefore, except in this relation, is unrea- 
 sonable, as it is unreasonable to desire food which cannot 
 nourish or fire which cannot warm. I want something 
 which has by hypothesis no relation to my wants. The same 
 is true of the sentient object so long, and only so long, as I 
 do not take its sentience into account. But to take the 
 sentience into account is to sympathise, or at least the 
 svmpathy is implied in the normal or only possible case. 
 The only condition necessary for the sympathy to exist, and 
 to be capable therefore of becoming a motive, is that I
 
 256 THE SCIENCE OF ETHICS. 
 
 should really believe in the object, and have, therefore, 
 representative feelings. To believe in it is to feel for it, to 
 have sympathies which correspond to my representations, 
 less vivid as the object is more distant and farther from the 
 sphere of my possible influence, but still real and therefore 
 effective motives. Systematically to ignore these relations, 
 then, is to act as I should act if I were an egoist in the ex- 
 tremest sense, and held that there was no consciousness in 
 the world except my own. But really to carry out this 
 principle is to be an idiot, for an essential part of the world 
 as interesting to me is constituted by the feelings of other 
 conscious agents, and I can only ignore their existence at 
 the cost of losing all the intelligence which distinguishes me 
 from the lower animal. 
 
 39. What, then, is the law of the motives when the in- 
 fluence of sympathy is admitted ? Can it still be said that 
 I shall always act for my own greatest happiness ? That, as 
 we have seen, must be approximately the law of motive for 
 the non-sympathetic animal, inasmuch as his conduct is 
 governed by his feelings, and those feelings can only be dis- 
 tino-uished bv includins:; or excluding; a foretaste of feelings 
 to come, but still of his own. The only elements in the 
 problem are therefore the feelings of the agent himself, 
 including the anticipations of his future state, and I have 
 suggested reasons for thinking that the correlation must be 
 such that his perceptions of maximum happiness will natur- 
 ally coincide with the strongest motive. When we intro- 
 duce the sympathetic feelings, it still remains as true as 
 before that the agent is governed by his own feelings, but 
 the law is no longer stateable in the same way. It is true, 
 in mathematical phrase, that the conduct of the agent is a 
 function of his feelings, but some of the feelings are them- 
 selves functions of independent variables, namely, the feelings 
 of others, and we therefore cannot deduce the law of conduct 
 from the agent considered by himself. The colour of a 
 reflecting body depends upon the intimate structure of the 
 body, as much as the colour of the non-reflecting body, but
 
 THE RULE OF CONDUCT. 257 
 
 the law of the colour will in one case necessarily include, 
 and in the other necessarily exclude, a reference to the sur- 
 rounding bodies. This, as I take it, gives the true and only 
 tenable line of distinction. The sympathetic being, that is, 
 becomes, in virtue of his sympathies, a constituent part of a 
 larger organisation. He is no more intelligible by himself 
 alone than the limb is in all its properties intelligible w^ith- 
 out reference to the body. Each part of the body must of 
 course be governed by its own properties, but they work in 
 such intimate connection with the whole organism, that they 
 are only intelligible, or, in other words, we can only obtain 
 the law of their action, when we take the whole body into 
 account. This is equally true of the being which has become 
 part of the social organism. It is true, we may say, in 
 respect of the direct sympathies which bind him to some 
 other person, so that his friend's joys and sorrows are also 
 his own. It is true, again, whenever such sympathies give 
 rise to a corporate spirit, to the domestic bonds which unite 
 families, the patriotism of states, or the military spirit of an 
 army. It is equally true of those instincts, the sense of hon- 
 our, and so forth, which are generated by the social factor, and 
 which, though they do not imply the presence of any special 
 organisation, are essential to the constitution of the social 
 tissue. All such instincts are products, we must suppose, of 
 sympathy; their growth and strength imply a capacity in 
 each of feeling for others ; and being accessible to impulses 
 not implying changes in the physical organisation, they are 
 so worked into the most essential modes of thought and 
 feeling that they must count as underlying and primary in- 
 stincts, and any personal element has been eliminated by the 
 very process of propagation. Though feelings of the indi- 
 vidual, their law can only be determined by reference to the 
 general social conditions. 
 
 40. This, again, enables us to state in what sense the 
 prudential axiom must now be limited. To become reason- 
 able is (to my mind, at least) to act on general principles, 
 and to act consistently; and this, as I have said, includes 
 
 R
 
 258 THE SCIENCE OF ETHICS. 
 
 the condition that a statement of the real cause of my action 
 should equally assign the reason of my actions. The law 
 which my feelings actually follow must coincide with the 
 principle which commends itself to my reason. In order, 
 then, that a being provided with social instincts should act 
 reasonably, it is necessary, not that he should take that 
 course of conduct which gives the greatest chances of happi- 
 ness, but that which gives the greatest chance of happiness 
 to that organisation of which he forms a constituent part. 
 Certain external conditions were necessary, as I have said, 
 to the adoption even of the prudential axiom — the condition, 
 namely, that the immediate pleasure should not be normally 
 inconsistent with the greatest sum of pleasure ; and it is of 
 course still more obvious that in the other case the weak 
 and intermittent sympathies of the less social man should 
 not be naturally out of harmony with the purely non- 
 sympathetic instincts. As a rule, the instincts of the social 
 organism must be closely coincident with those of the 
 individual ; in fighting for his tribe the savage must be 
 fighting for himself. Even in the highest societies both 
 reason and sympathy are feeble enough, but every extension 
 of reasoning power implies a wider and closer identification 
 of self with others, and therefore a greater tendency to merge 
 the prudential in the social axiom as a first principle of con- 
 duct. In the highest conceivable stage, a large part of con- 
 duct is still prompted by motives in which the sympathies are 
 not concerned, just as in the highest organisms each organ 
 has some properties which have no reference to the organic 
 union. But this, as has been sufficiently said, implies no in- 
 compatibility except in particular cases, and it is equally true 
 that so soon as I become sympathetic, even in the slightest 
 dco:ree, and thcrebv accessible to the social instincts, the 
 purely prudential maxim ceases to give the true law of 
 motive, and therefore of conduct, in all the cases in which 
 the sympathies or the derivative instincts are called into 
 action. 
 
 41. The explanation may be completed by considering the
 
 THE RULE OF CONDUCT. 259 
 
 prudential turn given to the facts upon the egoistic hypo- 
 thesis. That hypothesis starts from the theory, which I have 
 criticised upon different grounds, that reasoned action means 
 action for an end. I admit, of course, the truth impHed in 
 this statement, that in reasoned conduct every action is 
 regarded not merely in isolation but as part of a system, 
 and therefore includes more or less conscious reference to 
 the future as well as to the present ; but I have further 
 remarked that this must not be so interpreted as to imply 
 that the subjective condition of conduct can ever be any- 
 thing else than the present feeling. As soon as we slide 
 into that fallacious statement, we adopt the egoistic formula, 
 for in that case the dependent or identical proposition that 
 the conduct of an agent is determined by his own feelings, 
 since otherwise it would not be his own conduct, is expressed 
 by saying that his end must be his own happiness. Even 
 when we have to do with the sympathetic feelings, of which 
 it is the primary characteristic, from their most elementary 
 form, that they reflect the feelings of others, they have to be 
 forced into conformity with this formula ; and we therefore 
 have to assume that in all cases of sympathetic feeling there 
 is an egoistic end, which is dropped from consciousness 
 at the moment of action. This, upon my theory, is to 
 admit that all such conduct is unreasonable, or that it would 
 tend to become impossible in proportion as our reason, 
 and therefore our prevision of future consequences, became 
 stronger. This argument, however, is sometimes met by 
 accepting the apparent paradox, and declaring that a man 
 may be the happier by taking for his "end^' that which is 
 not the ultimate end. This, indeed, would still leave any 
 case of real self-sacrifice unreasonable ; for the advantage, 
 whatever it may be, of attending exclusively to the imme- 
 diate end cannot make it reasonable to pursue that end at 
 the expense of the ultimate end ; but it may be taken to 
 explain why, in point of fact, men may find pleasure in 
 pursuing the good of others when they anticipate no ultimate 
 good to themselves.
 
 26o THE SCIENCE OF ETHICS. 
 
 42. Now this statement certainly expresses an important 
 psychological truth, which may be worth considering in many 
 cases. It is no doubt true that the pleasure of any emo- 
 tional state is in proportion to its intensity, and therefore to 
 the exclusion of all other emotions for the time. We may 
 put this into a paradoxical shape if we say that our know- 
 ledge of any feeling is proportional to its intensity, and yet 
 that the greater the intensity the less we can know of it. 
 This is only a way of putting the fact that the presentative 
 knowledge excludes for the time the representative, I know 
 anything the more I know its relations to other things, and 
 I know a feeling as I know the conditions under which it 
 arises. I know it, again, the better as I have felt it more 
 keenly, but at the moment the intense feeling excludes all 
 reflection, and therefore its intensity suppresses knowledge 
 at the instant, though it is a condition of knowledge when 
 I come to reflect. In the same way we may hold that when 
 a man is acting for any end whatever, he may gain it more 
 effectually by not thinking about it at the time. If I aim 
 at a mark in order to win a prize, I must not think of the 
 prize whilst I am aiming, for to think of the prize is to allow 
 a number of distracting representations to interfere with my 
 absorption in the immediate action, and they may be equally 
 distracting at the moment, whether they are in some way 
 connected or not connected with the action, whether they 
 are thoughts of the cheers which are to greet my success, or 
 thoughts of some entirely different character. Exclusion of 
 everything irrelevant and extrinsic, absolute concentration 
 upon a single end, is a general condition of successful action, 
 even when that special end is part of a larger whole, and 
 would not be desired unless it contributed to something else. 
 43. This is clear enough, and it holds good of course of 
 many altruistic actions. I may do good to a man in order 
 to attain a reward ; I shoot at a mark to gain the prize ; 
 I cure a patient to get the fee. And, in point of fact, I 
 think that, as a rule, the mind generally "flickers" — that it 
 runs along a chain of consequences, stopping sometimes at
 
 THE RULE OF CONDUCT. 261 
 
 one point, sometimes at another, dwelling upon the final 
 success or the intermediate struggle, and therefore taking 
 various, though, so far as it is reasonable, consistent or 
 mutually dependent ends. But this statement does not in 
 the least aflfect the reality of each of the motives called into 
 play. I have no right to select the last state anticipated, 
 and to say that this alone is the essential motive, even though 
 other motives taken by themselves may be insufficient with- 
 out it. Every part of the foreseen consequences has its effect 
 as much as the ostensible end. I aim at the mark to get the 
 prize; that is, if it were not for the prize I should not 
 aim. But it is equally true that the desire for the prize 
 would not make me aim if the act of aiming were itself 
 disagreeable in a certain degree. Nor, again, could I dismiss 
 from my mind all thoughts of the end, and therefore I could 
 not fulfil the necessary condition of success unless the action 
 were agreeable up to a certain point. It must be in itself 
 tolerable, or I should have to call up a thought of extrinsic 
 consequences, and so far to distract my mind; and there- 
 fore the necessity of "disengagement" proves nothing against 
 the reality of each motive, which has for the moment to be 
 (if I may say so) self-supporting. On the contrary, in order 
 that it may be self-supporting, the motive must be real. 
 
 44. This is equally true in the case of the benevolent 
 action. The physician is not benevolent enough to cure me 
 unless he expects a fee ; but unless he is really kind, unless, 
 that is, he has a real sympathy for my suffering, he must be 
 always thinking of his fee, which is a very different thing. 
 He cannot be really benevolent so long as he regards his 
 patient simply as an instrument upon which he is to operate 
 for the sake of pay, without real interest in its feelings. I 
 need not ask whether he will or will not be a better physician 
 for being benevolent or not; though one may perhaps admit 
 that on some occasions he will do well to suspend both his 
 sympathies and his desire of fees, and to full back upon the 
 simple pleasure of skilful energy. But in any case, if he has 
 to forget his fee for the time, some other genuine motive
 
 262 THE SCIENCE OF ETHICS. 
 
 must take its place; and though the desire to reheve is only 
 one of the possible substitutes^ it must be real so far as it 
 has to produce any influence. And a real sympathy, so far 
 as it exists, is at once a feeling which does not conform to 
 the purely prudential axiom. He cannot be so good even 
 as a physician unless he is accessible to motives which may 
 carry him beyond the area of professional success. The 
 whole argument, in fact, merely comes to this, that in the 
 complex system of actions which constitute the active life of 
 any reasonable being, the suppression of any one motive would 
 clearly involve the alteration of others; our sympathies would 
 often be stifled if it were not for the co-operation of motives 
 of a diff'erent kind, and our non-sympathetic feelings would 
 be equally limited in their range if such modes of action 
 would not operate by motives which rest essentially upon 
 sympathetic feeling. 
 
 45. And this suggests a remark which will have to be 
 developed hereafter. The difference between the sympathetic 
 and the non-sympathetic feelings is a difference, as we have 
 seen, in their law or in the fundamental axiom which they 
 embody. When the egoist, therefore, maintains that it is 
 paradoxical to say that a man can be the happier for aiming 
 at something which is not his own happiness, he means that 
 a man cannot be the happier for sensibility to motives which 
 obey a different law from that of the simple desire for his 
 own happiness. Now, from what has gone before, it is plain 
 that this paradox has really no meaning for us. It is true 
 that the man acquires sympathies which may deviate from 
 the law of prudence, and which may therefore involve 
 self-sacrifice. Even the non- sympathetic instincts may, 
 as I have argued, involve self-sacrifice, and self-sacrifice is 
 clearly not essential to the sympathetic instincts ; it is only 
 an incident which has more or less importance according as 
 the interests of the society conflict more or less with those 
 of the individual. If and in so far as this conflict does not 
 exist, there is no paradox in supposing that the sympathetic 
 is happier than the non-sympathetic being. He differs in
 
 THE RULE OF CONDUCT. 263 
 
 having acquired new sensibilities ; he is not the same man 
 acting from different motives, which is, in fact, a contradic- 
 tory assumption, but a diflferent being with a difflirent set of 
 faculties ; he has gained a fresh capacity which has fresh 
 advantages as well as fresh dangers. It has indeed this plain 
 advantage, that he cannot develop as a reasonable agent 
 without it. To be reasonable, he must be sympathetic ; to 
 be thoroughly and systematically selfish, he must be an idiot; 
 or, in other words, we may say that he has made a bargain, 
 in virtue of which he makes a common stock of pains and 
 pleasures with the whole society to which he belongs, and 
 acquires all the new advantages which are dependent upon 
 the social union. We shall have to consider more fully 
 whether the bargain be a good one or a bad one ; but we 
 have at least no a prioi'i right to say that it is bad ; for if it 
 carries with it an obligation to occasional self-sacrifice, we 
 cannot tell whether the obligation is or is not oppressive on 
 the whole till we can tell how it operates in fact and what 
 are the correlative advantages which it implies.
 
 ( 264 ) 
 
 CHAPTER VII. 
 
 MERIT. 
 I. The Conception of Merit. 
 
 I. Altruism is, as I have argued, the faculty essentially 
 necessary to moral conduct. Were it not a reality, virtue 
 would be a name and society an impossibility. But, as I 
 have also said, the altruistic sentiment is not to be identified 
 with morality. I can only be an efficient member of any 
 society so far as I can identify myself with others. As altru- 
 istic, I can imbibe the corporate spirit of any social organism, 
 and become absorbed in my regiment, my church, my family, 
 or my club; but the sentiment itself thus generated is some- 
 thino- different from the altruism of which it is a product. 
 The elementarv sympathy must be regulated and disciplined 
 in order that it may give rise to the truly moral sentiments. 
 Virtues which belong to the type of truthfulness and justice 
 generally implv a severe restraint of the immediate sympa- 
 thetic impulses. A hatred of lying is a virtue, because the 
 typical character, as determined by the conditions of social 
 vitality, includes thorough trustworthiness. But at any given 
 moment the love of truth may dictate conduct which, at first 
 sight, at any rate, is contrary to that dictated by the love 
 of our neighbours. Hence virtue implies more than simple 
 altruism or benevolence, namelv, the elaboration and regula- 
 tion of the sympathetic character which takes place through 
 the social factor. 
 
 2. The recognition of this leads to a corresponding recogni- 
 tion of another aspect of the same process. As altruistic or 
 sympathetic, we arc not only sensitive to the pains and
 
 THE CONCEPTION OF MERIT. 265 
 
 pleasures of other?, but we catch the contagion of their 
 complex sentiments. We share their prejudices and passions, 
 their love?, hatreds, and modes of estimating men and things. 
 We come to love and hate our own qualities, with a love 
 and hatred reflected from the feelings of our neighbours, 
 and transmuted in this case into remorse or self-complacency. 
 As altruistic we are fitted into the social medium and inoc- 
 culated with its characteristic sentiments. Hence we have, 
 amongst other things, the complex sentiment of moral ap- 
 proval and disapproval. If virtue were identical with altru- 
 ism, we might identify moral approval with gratitude. It 
 would be simply a case of loving the man who does us a 
 good turn, because his action implies love for us or for our 
 fellows. But this seems to be an inadequate account of the 
 peculiar sentiment which is elaborated in any complex social 
 structure. The approval of which virtue is the object requires 
 to be explained, as well as the motives of which virtue is the 
 fruit. In both cases we have to consider sentiments which 
 imply the existence of a true altruism, but which also imply 
 some modification of the altruistic feelings. 
 
 3. Hence arise certain problems which require discussion, 
 and which take various forms according to the aspect under 
 which we regard them. The moral code itself, according to 
 the principles hitherto expounded, is briefly a statement of 
 the conditions of social vitality. A man is said to do his dufi/ 
 when he obeys this code. He has mei'lt in so far as he obeys 
 the law; or, according to some theories, he has merit if he 
 exceeds it, and demerit if he falls short of it. He is under an 
 obligation, again, to obey the law, as merit implies the fulfil- 
 ment of the obligation. He is virtuous so far as his character 
 secures that his conduct shall be invariably in' conformity 
 with the law; and the conscience is the feeling or group of 
 feelings which make conformity pleasant and a want of con- 
 formity painful to him. He is morally responsible for the 
 duties which he is able to perform. These various phrases 
 are of course closely connected. To explain one, therefore, 
 is, in fact, necessarily to explain the other. In some cases we
 
 266 THE SCIENCE OF ETHICS. 
 
 think primarily of the social sentiment of approval and dis- 
 approval, and in the others of the motives by which the agent 
 is actuated in virtuous conduct. Conduct is meritorious 
 when regarded as giving a claim upon the approval of others, 
 and is virtuous when we think of it as implying a disposition 
 of spontaneous conformity to the moral law. There is the 
 same kind of difference between the words virtue and merit 
 as there is between the words reason and argument. A reason 
 is an argument when it is applied to convince others, and 
 an argument is a reason when it supplies the ground of the 
 individual conduct. But there are certain difficulties which 
 are specially connected with each aspect; they are so far dis- 
 tinct that they can be separately discussed, and perhaps it does 
 not much matter in what order we take them. I propose 
 first to discuss the problems connected with the theory of 
 merit, and by showing how the conception of merit depends 
 upon that of virtue we shall clear the ground for the final 
 question as to the nature of virtuous motive and conscience. 
 I take for granted for the present, that intrinsically virtuous 
 motive is possible ; that is, that a man may be so constituted 
 as to obey the moral law unconditionally. I say, then, that 
 he is meritorious in so far as he is thus constituted, and I shall 
 try to explain certain fallacies which obscure this part of the 
 subject. 
 
 4. Merit, in the first place, clearly implies a close con- 
 nection with virtue. We may assume that, ceteris paribus, it 
 is proportioned to virtue. That man is the most meritorious 
 who, under the same conditions, is most virtuous, and that con- 
 duct the most meritorious which requires the greatest virtue for 
 its performance. Merit, in the next place, seems to carry a 
 reference to some reward. So far as meritorious, a man has 
 a claim upon the approval of his fellows or (upon some 
 systems) a claim upon the justice of his Maker. It is even 
 supposed, in some superstitions, that he can obtain a claim 
 which may be passed to the credit of others, llie genesis, 
 then, of the theory seems to be simple. So far as we share 
 the moral sentiment, we wish that virtue should be stimu-
 
 THE CONCEPTION OF MERIT. 267 
 
 lated, and therefore that it should be rewarded. The moral 
 rule begins, as I have argued, in the external form ; it is 
 stated, " Do this," instead of, " Be this." So long as it is in 
 this form we need not attend to the motives of the agent. 
 The conduct is approved simj)ly because it is useful, and it 
 is equally useful whatever his motives. I desire that a man 
 should not cut my throat, and may care little whether he is 
 restrained by a fear of the gallows or of hell, or by a desire 
 of pavment, or by sympathy for me. As the moral senti- 
 ment develops, I come to approve of the motives which imply 
 true morality, or of such a dislike to cutting my throat as 
 is founded not upon the extrinsic and separable motive, but 
 upon the intrinsic and inseparable motives of humanity and 
 good will. But this development does not imply that the old 
 motive is superseded, only that it is less prominent. The 
 fear of punishment may be operative or capable of being 
 called into activity. I still desire that my throat should not 
 be cut, and therefore that the gallows should remain appli- 
 cable in case of need, though I desire also that the case may 
 occur as seldom as possible, and that men may be actuated as 
 much as possible by the motives which are opposed to murder 
 as such, irrespectively of possible penal consequences, and so 
 far as I make this distinction, I recognise a difference in 
 the merit of the two classes of persons. The man, I say, is 
 meritorious who does from an intrinsic motive what another 
 man will do only from an extrinsic motive. The villain 
 onlv dislikes hantrino; and murder so far as it leads to hano:- 
 ing; the benevolent man objects to murder whether it has or 
 has not bad consequences to himself. I consider, therefore, 
 that he has a certain claim upon me and upon society at large, 
 inasmuch as he has done for nothino: what another man 
 will only do for pay, or has refrained spontaneously from 
 doing something from which another man can only be re- 
 strained by threats and coercion. 
 
 5. The principle so far seems to be simple enough, though, 
 like many simple principles, it leads to some intricate questions. 
 If we wish well to virtue, it is suggested, we must wish virtue
 
 268 THE SCIENCE OF ETHICS. 
 
 to be rewarded, and vet with the certaintv of reward virtue 
 disappears. A inau saves my life out of sheer benevolence, 
 and I reward him out of sheer gratitude. But if he had a 
 right to be rewarded, or could count upon reward as a 
 certainty, he would so far cease to be virtuous. He would 
 be saving my life from avarice instead of benevolence. So far 
 as I stimulate the extrinsic, I deduct from the intrinsic motive. 
 The contrast of course appears in many theological contro- 
 versies. If virtue is to be rewarded by heaven and vice by 
 hell, do they not, it has been asked, cease to be virtuous and 
 vicious? One difficulty which applies to human laws can of 
 course be avoided. A human legislator cannot secure the 
 coincidence of the extrinsic with the intrinsic motive. If he 
 pays for virtue, a love of pay takes the place of a love of 
 virtue, and love of pay may be pressed into the service 
 of vice ; a svstem of rewards may suggest a system of bribes, 
 and thus no external sanction can be uniformly annexed 
 to the moral law. The divine legislator is of course bound 
 by no such restrictions; he may secure the absolute coin- 
 cidence of the two classes of motive, and may affix to 
 virtuous and vicious conduct consequences which are not the 
 necessary outcome of the conduct itself. The purely self- 
 regarding motive may thus always operate in the same direc- 
 tion with the altruistic. The question remains (with which 
 I have nothing to do here), whether such a theory does not 
 destroy the essence of virtue by making the appearance of 
 altruism a mere illusion? In anv case, it illustrates the fact 
 that merit represents the claim of the virtuous person upon 
 the universe. In so far as he desires no reward here, he is 
 held to deserve a reward hereafter; and we need not here 
 inquire how, upon this hypothesis, a satisfactory distinction 
 can be drawn between prudence which aims at an immediate, 
 and virtue which aims at a remote advantage. 
 
 6. Merit, then, is a function of the social forces by which 
 our characters are moulded. It is attributed to any one in so 
 far as he dispenses with any extrinsic stimulus, or, in other 
 words, with motives which are equally available for other pur-
 
 THE CONCEPTION OF MERIT. 269 
 
 poses. Thus we find that the wants of mankind in particular 
 social stages generate a particular respect for certain virtues, 
 which under different conditions cease to be valued so highly, 
 because the wants can be supplied without calling the virtue 
 into plav. In sparsely settled countries, for example, hospi- 
 tality is stimulated bv its obvious convenience. A man is 
 forced, on penalty of forfeiting the esteem of himself and 
 his neio;hbours, to render services for nothing which would 
 elsewhere be rendered for pay. He is regarded as a churl 
 if he turns away a stranger from his door. The senti- 
 ment is developed wherever the conditions occur which 
 make the practice obviously convenient, and may be de- 
 scribed by saving that in such regions a man is induced to 
 do what an innkeeper does elsewhere, not by the prospect of a 
 bill, but bv dread of incurring contempt, or by the correspond- 
 ing sentiment which has become a part of his own character. 
 It does not of course follow that when the social demand is 
 lowered, the creneral level of virtue is lowered. A man who 
 lives in London may be called upon to approach a higher 
 standard of benevolence in general than an Arab in the desert 
 or a backwoodsman in America. But this particular kind of 
 benevolence is not demanded from him to the same extent, 
 and we do not censure him for a want of hospitality when 
 he sends a foreigner to a hotel or passes on a tramp to the 
 casual ward. A similar change takes place in regard to 
 many duties which in a rude state of society depend upon 
 the voluntary public spirit of individuals, and which are pro- 
 vided for in more civilised conditions by a regular part of the 
 social machinery. The demand for certain manifestations of 
 virtue becomes less when society is so constituted that the 
 corresponding kind of conduct can be commanded without 
 callimz; for self-sacrifice. Whenever society finds sacrifice of 
 the individual necessary, it pays for it, we may say, in terms 
 of merit. The deserving person has a blank form of credit 
 upon the world at large, not to be filled up in terms of hard 
 cash. The whole demand for benevolence may increase whilst
 
 270 THE SCIENCE OF ETHICS. 
 
 special modes of benevolence are less necessary, and therefore 
 regarded with less respect. 
 
 7. The conception of merit has thus a close analogy to 
 the economical conception of value. We may define merit 
 as the value set upon virtue. We have to distinguish between 
 the merit and the intrinsic virtue of an action as economists 
 distinguish between the value of any commodity as equiva- 
 lent to its intrinsic utility and what is called the value in 
 exchange. Water, as the economists tell us, has a certain 
 utility, which is, of course, independent of the abundance or 
 the scarcity of the supply. It has the same effect upon my 
 thirst whether I live upon the borders of a river or can only 
 obtain an occasional bucket from a well. But the value 
 in exchange depends upon the difficulty of attainment, and, 
 in the ordinary case, gravitates towards a certain average 
 standard, dependent upon the various processes which con- 
 stitute the industrial life of a community. The same state- 
 ments may be made in regard to virtue and merit. Benevo- 
 lence, we may say, is always benevolence, and truthfulness, 
 truthfulness ; but the estimate set upon these may vary within 
 wide limits. The moral law may remain in a sense unaltered, 
 whilst the price necessary to secure obedience may rise or 
 fall, the merit of obedience being greater in proportion to 
 the quantity of extrinsic motive necessary to enforce 
 obedience upon the average mind. An action is highly 
 meritorious in one country which in another is a mere 
 matter of course. That was regarded as an act of heroic 
 self-restraint in Scipio which would be so natural to a modern 
 general that to praise him for it would be an insult. We 
 scarcely thank a mother for a devotion to her child which, 
 if shown to a stranger, would imply the most unusual 
 benevolence, and therefore the highest merit. In all societies 
 some degree of maternal affection is necessary ; but in some, 
 a mother would be praiseworthy for attentions to her child, 
 the neglect of which in others, even for the sake of her own 
 health and comfort, would involve the severest censure. Thus, 
 we may suppose that whilst the scale of duty remains fixed,
 
 THE CONCEPTION OF MERIT. 271 
 
 the zero point of merit may shift upwards or downwards 
 according to circumstance. Of two courses of conduct, the 
 same may be regarded in all cases as the best, but the degree 
 of approval which it invites may change like the price of 
 a commodity. You are no more obliged to a man, it is 
 said, for being commonly honest or decently civil than you 
 have to pay for air in the open fields or for water on the 
 banks of the Nile. Merit thus carries with it a reference to 
 an assumed averao;e standard of conduct, and accrues to the 
 agent in proportion as he reaches that standard. Absolute 
 merit, if the phrase may be used, means a man's virtue, 
 considered abstractedly from the social state and the diffi- 
 culty of attaining it, whilst merit, in the more ordinary sense, 
 takes those conditions into account. The sober man is in the 
 first sense equally meritorious everywhere, because he every- 
 where shows the same quality ; but sobriety may be called 
 more meritorious in England than in a temperate country, 
 because the average standard of temperance is lower. 
 
 8. It is clear, again, from this that merit can only belong 
 to voluntary actions. A man is meritorious in so far as he 
 acts in a way which the average man will only act under from 
 the stimulus of some extrinsic motive. The act, therefore 
 must spring from his character ; it must be the fruit of some 
 motive which we regard as excellent ; and if it did not arise 
 from a motive — or, in other words, were not voluntary — it 
 would not, properly speaking, be his conduct at all. The 
 meritorious disposition must be capable of a stimulus from 
 the approval or disapproval of the society. There is no 
 price for commodities the supply of which is entirely beyond 
 the influence of demand, and we do not praise or blame a 
 man for qualities incapable of being altered by our praise 
 or blame. We may like or dislike a man for qualities which 
 w^e recognise as being entirely beyond control, but the senti- 
 ment only becomes praise or blame uhen we conceive it as 
 having a certain power of modifying its objects. Moral 
 approval is the name of the sentiment developed through 
 the social medium which modifies a man's character in such
 
 272 THE SCIENCE OF ETHICS. 
 
 a way as to fit him to be an efficient member of the social 
 '^tissue." It is the spiritual pressure which generates and 
 maintains morality. The whole man is moulded by the 
 beliefs and sentiments which he imbibes from the surround- 
 ing medium. He may be forced to obey the external law 
 both by intrinsic and extrinsic motives ; but so far as he is 
 really and intrinsically moral^ his character is regulated and 
 stimulated by the organised opinions of the society to which 
 he belongs. It is plain, therefore, that his merit, which 
 corresponds to the degree in which he has been thus regu- 
 lated, can only accrue in respect of the qualities capable of 
 being thus influenced, and these are the qualities implied in 
 all voluntary conduct. 
 
 9. In saying, then, that a man has merit, we mean that he 
 has virtue, whilst we implicitly recognise the fact that virtue 
 is the product of a certain social discipline. The individual 
 of course may be a more or less favourable subject of such 
 discipline ; his innate qualities may be such as spontaneously 
 mould themselves upon the moral code, or such as are only 
 forced into it with great difficulty. They are, in any case, 
 qualities which are modifiable, and susceptible of the social 
 discipline. When a man obeys the moral law from some 
 extrinsic motive, he is not, properly speaking, moral at all ; 
 so far as he can properly be called virtuous, it is because the 
 outward has become an inward law ; it is no longer a 
 law in the juridical but in a scientific sense ; it is not a rule 
 enforced by external sanctions, but the "law" of his charac- 
 ter, or the formula which expresses the way in which he 
 spontaneously acts. Society does not force him to act 
 against his will ; it has annexed and conquered his will 
 itself; the obligation is internal, and the action supplies its 
 own motive. The man, if we choose to say so, enforces 
 upon himself, which is the same thing as to say that he does 
 without force that which others can only be made to do by 
 some external force. We imply, therefore, that in this case 
 virtue is intrinsically desirable, or, in the common phrase, 
 becomes its own reward.
 
 THE CONCEPTION OF MERIT. 273 
 
 10. One other source of possible ambiguity must be 
 noticed. We speak, perhaps, more commonly of the merit 
 of an action than of the merit of the agent. What is meant 
 by such a phrase ? Obviously the moral quality, whatever 
 it may be, cannot be attributed to an action as distinguished 
 from the agent. By a kind action we mean the action 
 which is done by a man because of his kindness, or in so 
 far as he is kind. In other words, it is an action which 
 proves him to be kind, or which would not be done unless 
 he were kind. And in the same way, a meritorious action 
 is the action which proves a man to have merit, or, in other 
 words, to be virtuous. Upon this showing — and it is, I think, 
 the only consistent statement of the case — an action is meri- 
 torious in so far as it is a manifestation of certain qualities 
 already existing. But we speak of the action rather than 
 the agent as meritorious for obvious reasons. For, in the 
 first place, we can only know a man's character through his 
 actions. We could not know for certain that Leonidas was 
 a brave man until he had fallen at Thermopylae, for we 
 cannot see a man's bravery. We therefore may fall into the 
 confusion of speaking as though a quality were more real 
 because it is more clearly established. We are quite right 
 for honouring a man more who has given proofs of courage 
 by his behaviour under danger than one whose courage is 
 only inferred from more indirect inferences. If, then, we 
 mean by merit proved virtue, we may admit that of two 
 equally virtuous men one may have more merit than 
 another. We mean, not that he has more virtue, but that 
 he has shown more. The case, again, is complicated by the 
 reflection that action strengthens habit, and therefore that 
 a man may become actually more virtuous by giving greater 
 proofs of virtue. This being understood, however, it is only, 
 a question of words. A man may be equally virtuous whether 
 he has or has not had opportunities of showing his good 
 qualities; his intrinsic merit, therefore, is unaffected. But if 
 by merit we mean the established claim upon our respect, 
 his merit will, of course, be increased according to the 
 
 s
 
 274 
 
 THE SCIENCE OF ETHICS. 
 
 opportunities of manifestation. There may have been a 
 hundred men in the EngUsh fleet as brave as Nelson, but 
 honour could only be paid to the one who had shown his 
 valour. That only shows that honour in the world cannot 
 be proportioned to the merit absolutely, but only to the 
 merit which has become known. 
 
 II. The distinction sometimes gives rise to other difficulties. 
 Moralists have spoken of the goodness of an action as being 
 independent of the motive. Persecution, it has been said, 
 is equally bad, whether it proceeds from a religious motive 
 associated with a mistaken view of duty, or from some worse 
 motive— say, a simple love of despotic power. If we judge 
 morality by "consequences," it is equally wrong in both cases, 
 and therefore equally condemnable; and hence, if the merit 
 follows the morality, we must condemn the good man who 
 persecutes from misguided love of truth, and admire the bad 
 man who tolerates out of sheer indifference to truth. The 
 case is interesting because it suggests some troublesome 
 problems of actual occurrence, but we may answer it suffi- 
 ciently for our present purpose without much trouble. 
 According to the previous argument, it is the same thing 
 whether I say, "This is right," or, "This is commanded by the 
 moral law," or, "This is what all good men will do." The good 
 man is one who does what is right or what the moral law 
 prescribes, and the moral law prescribes that which is right 
 and which all good men do ; and therefore it is a contradic- 
 tion to say, " This is right," and to add that it may be done 
 cither by a good man or a bad man. But I have tried to show 
 how such an impression arises. The moral law, " Do not 
 persecute," is one of late growth, and for the simple reason 
 that the evil of persecution was not perceived until recently; 
 it therefore presented itself in the first instance in the shape 
 of an external law or a law of expediency; that is to say, 
 some people saw that persecution was mischievous but could 
 not convince others that it was mischievous. Whilst that 
 was the case, a man might persecute from a good motive, 
 say, a love of truth, not seeing that he was doing more harm
 
 THE CONCEPTION OF MERIT. 275 
 
 than good. Just so, as I have said, good men may still advocate 
 protection as well as free trade. If protection were recognised 
 as mischievous^ an advocacy of it would necessarily imply 
 selfishness, as at present it may imply either selfishness or 
 intellectual error. But whilst a proposed rule is in this state, 
 whilst its good or evil results are still disputed, and a con- 
 viction of its advantage has not forced itself into the accepted 
 moral standard, it is not properly a moral rule at all, or it is 
 a moral rule only for the more enlightened, who understand 
 its true nature. In saying, then, that it is wrong to persecute, 
 the early advocates of toleration meant that persecution was 
 mischievous, and therefore wrong for those who recognised 
 the mischief: but thev misfht admit that a "good" man 
 might still persecute if he did not see the mischief, meaning 
 by a good man a man of benevolence and love of truth, but 
 of a certain degree of stupidity. 
 
 12. Hence we may see what is the true criterion and the 
 cause of difficulty in its application. A given action, as 
 defined by its external relations, may always, as I have said, 
 be brought under various principles ; or, in other words, the 
 conduct may be the result of various motives. Heretics 
 may be burnt from religious or political motives, or spared 
 from religious indifference or out of respect for veracity ; 
 money may be given to the poor from ostentation or from 
 true charity. The inference, therefore, from the action to 
 the motive is always more or less precarious : it is precarious, 
 in particular, in the case of a growing morality, when the 
 true character of a given rule of conduct is not yet fully 
 recognised, and may be judged differently by different 
 people. Whatever we may think of the ultimate ground of 
 morality, we must all admit that the normal consequences 
 of any rule of conduct are relevant in determining its 
 moralitv. When we know that certain modes of indulo^ence 
 are injurious to health, the indulgence proves imprudence ; 
 and when we know that any mode of conduct is injurious 
 to our fellows, such conduct proves selfishness. So long as 
 the consequences remain uncertain, the implication as to
 
 276 THE SCIENCE OF ETHICS. 
 
 character depends upon the state of mind of the agent — that 
 is, upon his genuine belief as to the nature of his conduct. 
 Very difficult problems occur when '' material " is not iden- 
 tified with formal morality ; when, that is, we as lookers- 
 on are supposed to know that certain conduct is mischievous 
 in fact, but when at the same time the agent may be igno- 
 rant, or doubtful, or deluded by some sophistical argument. 
 The criterion, however, of merit seems to be clear, whatever 
 the difficulty of applying it. We assume, in the first place, 
 that the conduct springs from a certain motive; the man 
 gives money from charity or from ostentation, he is tolerant 
 from indifference to truth or from love of truthfulness. His 
 particular action, whatever it may be, is one case of a general 
 rule, which, again, expresses his character in a certain rela- 
 tion. In the next place, this rule coincides or diverges from 
 a given moral code, and the character of which it is a 
 partial expression is or is not virtuous as judged by that 
 code. Hence, again, follows the merit of the action. It is 
 more or less meritorious according to the degree of virtue 
 implied. Where the moral code is still doubtful, there may 
 of course be many difficulties in deciding the proper inference. 
 But, given the moral standard, we simply have to ask 
 wdiether the character implied does or does not correspond 
 to that type which spontaneously and invariably obeys the 
 moral law? The more closely it does so, the higher the 
 merit, which is, as I have said, nothing but the virtue 
 regarded from a particular point of view. 
 
 13. I have already argued that a true moral law can only 
 exist when it includes a definition of character. It is at most 
 an approximate statement of the moral law to say that we 
 should give money to the poor. The man who gives money 
 from ostentation is not really acting morally at all, though, 
 in the particular case, his conduct coincides with that which 
 morality prescribes. His conduct may be regarded as moral 
 so long as we attend only to the external rule; but it is a 
 sham morality which he obeys, for the character indicated 
 is not that which docs what really belongs to the best social
 
 THE CONCEPTION OF MERIT. 277 
 
 type, and therefore not that which is prescribed by a tenable 
 moral code. A i2:enuine moral law distino-uishes classes of con- 
 duct not according to external circumstances, but according to 
 the motives involved ; and therefore, when the conformity to 
 the law is only external, it is more proper to say that it is not 
 conformity at all. Vanity or avarice may often prompt the 
 actions which are equally commanded by a sense of duty or 
 by genuine love of my neighbours. The question whether 
 such actions are or are not virtuous is only intelligible as a 
 question as to the motives from which they spring. We may 
 or may not be able to answer that question decisively in any 
 particular case, but till it is answered we cannot say how 
 far the conduct is strictly meritorious. The test would be 
 given by placing a man in such a position that the only 
 motives operative are the intrinsic motives to virtue. If he 
 acts rightly when he can have no other motive for action 
 except those w^hich we hold to be virtuous, he is really 
 virtuous ; if not, we must suppose that his conformity to the 
 moral law was simply accidental. And this, it may be added, 
 is independent of any hypothesis as to the ultimate nature 
 of the intrinsically virtuous motive. If there be such a thing 
 as love of virtue for its own sake, the virtuous man must obey 
 the rule when all other motives, including those derived from 
 the happiness of his fellows, make the other way. If, on the 
 other hand, virtue is merely a form of selfishness, it is absurd 
 to suppose that virtue can ever command conduct which is 
 on the whole opposed to the interests of the individual. 
 But on that hypothesis we should consider a man to be in- 
 trinsically virtuous, and therefore meritorious, who, though 
 systematically selfish, never allowed immediate interests 
 to overpower a proper attention to his total interests. In 
 any case, his character must be such as to imply invariable 
 obedience to the moral code ; or, as we may safely say that 
 no one is virtuous up to this point, we should rather say 
 that he is virtuous or meritorious in proportion as he reaches 
 this ideal standard. 
 
 14. We have said, then, that a man's intrinsic merit is
 
 278 THE SCIENCE OF ETHICS. 
 
 not merely proportioned to his virtue, but is his virtue con- 
 sidered under a particular aspect, namely, as causing the 
 moral approval of his fellows, and that the merit of an 
 action means simply his proved virtue, that virtue, namely, 
 which he must possess in order to do the action in question ; 
 and, in saying this, we have assumed certain very simple 
 principles, which hav^e nevertheless produced libraries of con- 
 troversy. We assume, in fact, that merit can only attach 
 to voluntary conduct; for that is the same thing as to say 
 that it attaches to the character. Conduct which does not 
 spring from motives or from character is not, properly speak- 
 ing, conduct at all, A man is not truly an agent in matters 
 in which he is passive. In the next place, merit, as we have 
 seen, has a reference to a certain assumed standard : a man 
 is more or less meritorious as he is above or below the ordi- 
 nary standard in respect of virtue. Therefore conduct has 
 positive merit only in so far as it is more or less difficult for 
 the average man. Thirdly, the criterion of merit is that the 
 motive implied should be truly virtuous; that is, that its 
 agent is so far in conformity with the moral type. Now 
 these conditions are frequently expressed by saying that 
 merit implies free-will, that it implies effort, and that it 
 implies a love of right for the sake of right. A man can 
 have no merit so far as he acts under compulsion, or without 
 difficulty, or from some other motive than a love of virtue. 
 Other conceptions, especially that of ^' moral responsibility," 
 are equally involved in the controversies which have arisen 
 upon these points ; but it will be sufficient if I state the 
 bearing of my own theories upon the main points at issue. 
 
 II. Free- Will. 
 
 15. I have already said in a sunmiary way that I reject 
 the free-will theory, so far, at least, and only so far, as it 
 implies a negative of the "universal postulate" in regard to 
 human conduct. I would willingly pass by the whole con- 
 troversy with this statement. But it seems necessary to
 
 FREE-WILL. 279 
 
 traverse expressly the contention that a ^^ a determinist " 
 must logically be a disbeliever in merit. In one sense, in- 
 deed, that contention is admissible. I admit that there can 
 be no question of merit as between man and his Maker. 
 The potter has no right to be angry with his pots. If he 
 wanted them different, he should have made them differ- 
 ent. The consistent theologian must choose between the 
 Creator and the Judge. He must abandon the conception 
 of merit or the conception of absolute dependence. The 
 free-will argument, as understood by the school which seri- 
 ously maintains it, is an illogical attempt to reconcile two 
 conceptions which are radically contradictory by the device 
 of substituting the word '^mystery" for the plainer word 
 '' nonsense." Admitting the inherent difficulty of the ques- 
 tion, I must still admit frankly that to my mind the one 
 insuperable difficulty is the difficulty of reconciling deter- 
 minism with the ordinary theology. That difficulty, how- 
 ever, ceases to trouble us when we admit (with many theo- 
 logians) that the ordinary theology is erroneous. If any one 
 denies this, I must be content to refer him to the many 
 metaphysicians who, from the days of Hobbes and Jona- 
 than Edwards, have fully discussed the question. I proceed 
 to argue that not only is determinism consistent with a 
 belief in merit and moral responsibility, but that it is im- 
 plied at every step by that belief. 
 
 16. Let us start from a particular case. I sign what I 
 know to be a malicious libel. I am, then, a malevolent 
 liar. My conduct proves that I am neither benevolent nor 
 truthful. I deserve blame, and my conduct is de-meritorious. 
 But it is proved that my hand was held by overpowering 
 force. My action, then, was not wrong, or rather it was 
 not my action. My body was employed by somebody else, as 
 my pen was employed. My character, then, had no influence 
 upon the result. I may have been the most truthful and 
 benevolent of men. The moral law applies to my character, 
 and not to the mechanical movements of my limbs when 
 impelled by another man's will. Suppose it now proved
 
 2 So THE SCIENCE OF ETHICS. 
 
 that a pistol was held to my head or a bribe offered to me. 
 How am I now to be judged? From the whole operative 
 motive and the total implication as to character. The new 
 motives^ fear of death and love of money, are not in them- 
 selves bad^ for they may be shared by the best of men. The 
 implication as to my malevolence or falseness is not so strong 
 as before^ as the new motive counts for something. If the 
 temptation was very great and the injury to my victim very 
 trifling, it may perhaps be thought that my conduct was 
 such as an average man would adopt under the circum- 
 stances. If so^ I am not thought bad, though I should un- 
 doubtedly be better if I had enough courage, sense of hon- 
 our, and benevolence to resist the temptation. It may be 
 thought that resistance would have required heroic virtue, 
 or possibly that my yielding to a bribe implies a greed 
 which is still more contemptible than malevolence. Hence 
 arise many difficult psychological and moral problems : what 
 is the implication as to character ? what is the right morality? 
 and so forth. But the criterion remains the same, namely, 
 what was the quality of the motive indicated, and how far 
 is it indicative of a certain constitution of my character in 
 respect of morality ? 
 
 17. This is, I think, the argument sanctioned by common 
 sense, and to my mind it is perfectly sound and satisfactory. 
 The principle assumed is simple. I infer motive from con- 
 duct : so far as other causes do not account for the conduct, 
 my inference stands ; so far as other causes are assignable, 
 the inference must be modified accordingly; and the strength 
 of the motive is measured by the resistance which it over- 
 comes. So, in the case suggested, we only infer malevolence 
 in so far as the conduct is determined by the character of 
 the agent, and we infer that degree of malevolence which is 
 necessary to account for the action when the influence of 
 other motives is deducted. I reason precisely as I reason 
 in determining, for example, the motion of a body when I 
 set down, say, to the tension of a particular rope all the 
 share in supporting a given weight which is not otherwise
 
 FREE-WILL. 281 
 
 explained. At every step in tlic process I assume that there 
 is a causal connection between character and conduct, so 
 that I may infer motives from actions, and reciprocally 
 actions from motives. I assume freedom, in the sense of 
 freedom from external force, wherever I assume merit, because 
 the internal force accounts and must account for all that 
 part of the phenomenon for which the external force does 
 not account. Coercion or external force makes motive 
 irrelevant, and therefore annihilates the inference to motive. 
 But the inference would equally break down if I denied this 
 causal relation between action and motive. If, that is, 
 conduct did not imply motive when there was no coercion, 
 I could make no inference as to motive from the fact of 
 freedom. If, therefore, by assuming freedom, I mean to imply 
 that motive does not determine conduct, or that, given the 
 character, the man may either act or abstain from acting, I 
 so far destroy the inference as to virtue and merit. But, 
 upon my assumptions, this is to assume an absurdity, if not 
 a contradiction in terms. It is to destroy the sole postulate 
 in virtue of which reasoning is possible at all, or to make 
 the very essence of reasoning impossible. 
 
 18. Let us suppose, in fact, that my inference is uncertain, 
 or, as it is sometimes put, that the agent and all the relevant 
 circumstances being constant, the conduct varies. The same 
 man will in one case give and in another refrain from giving. 
 What is the legitimate inference ? Since the action varies, 
 I infer that some of the conditions must vary. Now in 
 many cases this is apparently an accurate statement of the 
 facts. The same man is liberal at one moment, stingy at 
 another. I infer that his character has varied, and that he 
 has seen one beggar before dinner and another after, or 
 that some accidental association of ideas {'' accidental " in 
 the sense that it is due to some combination not dependent 
 either upon his character or the assumed facts) has for the 
 moment modified his disposition. This, of course, is possible 
 upon any hypothesis. So far as this explanation is hypo- 
 thetic, the merit or the virtue is diminished in proportion
 
 282 THE SCIENCE OF ETHICS. 
 
 to the uncertainty. The man is less liberal than I supposed, 
 for his liberality is limited by a previously unknown con- 
 dition. He is only liberal when his temper is unruffled, or 
 when he is under the influence of a particular association. 
 But this is not enough for the advocate of free-will, for I 
 have still assumed that some variation of character or cir- 
 cumstance accounts for the varvino- conduct. He asserts 
 that, all conditions down to the minutest remaining constant, 
 there is still a possibility of variation in conduct. I deny 
 the possibility ; but, assuming it for the sake of argument, 
 I deny that the inference is legitimate. If the conduct 
 varies, and if no assignable change of conditions can account 
 for it, I cannot assume the intervention of some inscrutable 
 or unassignable condition which, as independent both of 
 character and circumstances, can only be described in nega- 
 tive terms. We are virtually postulating a blind fatality. 
 Some unknown and unknowable power must have governed 
 the action. But so far as that is the case, all inferences as 
 to merit and virtue are as illegitimate as in the case of 
 external coercion. Whatever determines conduct inde- 
 pendently of character so far destroys the moral value of 
 conduct. To say that a man is benevolent means that he 
 will always be benevolent ; to say that an action is bene- 
 volent means that it proves the man to be benevolent. What- 
 ever diminishes the certainty that a man is benevolent now 
 and will be benevolent hereafter must so far diminish his 
 virtue, and makes the whole theory contradictory. By con- 
 founding coercion with internal determination we fall into 
 endless perplexities ; for whereas we must admit that a man 
 is more truthful in proportion as he is certain always to 
 speak the truth, we make it an essential condition of his 
 virtue that it is intrinsically uncertain whether he will speak 
 the truth or lie. Indeed, if free-will be essential, then the 
 more free-will the better, and the smaller the certainty of 
 truth the greater the virtue. 
 
 19. The same argument of course applies to the correlative 
 theory of responsibility. A man is responsible for that alone
 
 FREE-WILL. 2S3 
 
 which he can do or leave undone. Obviously so if we mean 
 for that the doing or the not doing of which depends upon 
 his character ; for otherwise there is no inference as to 
 character. I am not responsible for dying when my throat 
 is cut, for I shall die equally whether I am a saint or a 
 sinner ; but I am responsible for cutting my throat, for I 
 shall not do so (assuming the immorality of suicide) if I am 
 a saint, but only if I am a sinner. If it is urged that, being 
 a saint, I am still free to cut my throat or leave it alone, 
 that is true in the sense in which it is true that, being a 
 saint, I may become a sinner. But the more saintly I am 
 the smaller is the possibility. If a fate called free-will or 
 anything else intervenes, and causes me to cut my throat 
 whilst I am still a saint, I am not the more but the less 
 responsible for an action which does not spring from my 
 character. 
 
 20. But, it is said, admitting the relation between char- 
 acter and conduct, it is true that each man can form his own 
 character. Undoubtedly every man is always forming his 
 own character. Every act tends to generate a habit or to 
 modify character, and consciously to form character is an 
 act like any other, and subject to the conditions already 
 stated. Nothing but fresh confusion is introduced by 
 attempting to draw the old distinction upon these lines. 
 We say, for example, that a man is less • responsible for 
 licentiousness who has been brought up in a corrupt society. 
 The argument is sound if reasonably interpreted. It is true 
 that a man who is now a drunkard may have been originally 
 as sober as another man who owes his sobriety to the absence 
 of temptations. The moral worth of the two men was 
 originally the same, and the difference is due to the differ- 
 ence of circumstance. Again, it is true that the whole in- 
 ference as to character is often very different according to 
 the different mode in which it has been formed. The man 
 who has been seduced to drinking by strong temptations is 
 as much a drunkard as the man who has taken to drink 
 without them^ but he has not given the same proof of
 
 284 THE SCIENCE OF ETHICS. 
 
 weakness of character, and may probably be more estimable 
 in other ways. When everybody drank, drunkenness was 
 more consistent with a sense of honour than it now is. 
 Such considerations show the necessity of guiding our judg- 
 ment by very complex inferences, and show that the merit 
 — in the sense of the proved virtue — of given conduct may 
 vary widely when the particular action alone is given. But 
 it is impossible to make this the ground for a distinction 
 between qualities due to circumstance and others due to the 
 man himself. In all actions, as in the whole of our lives, 
 there is a constant action and reaction between the external 
 and internal conditions. We cannot disentangle them into 
 two separate series of events, any more than we can say 
 whether breathing depends more upon the air or the lungs. 
 Every character is developed under circumstances, and the 
 development depends upon the continuous adjustment of 
 the relations. If we suppose that every man's "self" is a 
 separate entity of precisely the same qualities, then the 
 difference between the developed characters is due entirely 
 to circumstances, and therefore can have no merit on the 
 free-will theory, or to a mysterious act of choice which is due 
 neither to circumstance nor to a difference — for such differ- 
 ence is supposed not to exist — in the choosing subject. If, 
 again, the difference is due to some distinction between the 
 original selves, we come back to the determinist hypothesis; 
 for this difference is the condition of all subsequent differ- 
 ences, and must be itself due to previous growth, or to the 
 absolute will of the Creator. Or, finally, we come back to 
 the unintelligible theory of " accident" or "fatality" as the 
 foundation of merit, though merit is only intelligible as 
 excluding accident. 
 
 21. This argument has been so frequently and forcibly 
 stated that further insistence is needless. Identify free-will 
 with the occurrence of chance, and the conception of merit 
 becomes contradictory and repulsive. Exclude chance, and 
 you are virtually a determinist. From this dilemma I can 
 sec no escape, and I am not aware of a plausible answer.
 
 FREE-WILL. 285 
 
 The advocates of free-will theories are frequently content to 
 admit the force of the argument^ but retort it by suggesting 
 equal difficulties in the opposing theory, and asserting this 
 to be one of the speculations which lead to inevitable anti- 
 nomies. The practical reason is therefore left to choose the 
 most edifvinsf alternative. Though I am far from admitting 
 this assertion, I think that it is true in this as in many other 
 cases, that each party to the controversy is most effective 
 when assailing the position of its antagonists. I must 
 therefore give my reply to that which I take to be the most 
 telling argument of my opponents. The pith of it seems 
 to be as follows: — Moral responsibility, it is said, implies 
 freedom. A man is only responsible for that which he 
 causes. Now the cmisa causce is also the causa causaii. If 
 I am caused as well as cause, the cause of me is the cause of 
 my conduct; I am only a passive link in the chain which 
 transmits the force. Thus, as each individual is the product 
 of something external to himself, his responsibility is really 
 shifted to that something. The universe or the first cause 
 is alone responsible, and since it is responsible to itself alone, 
 responsibility becomes a mere illusion. 
 
 22. I admit, of course, the first statement. I am respon- 
 sible for that, and for that alone, which I cause. But does 
 the fact that I am also " caused " relieve me of responsibility ? 
 This I deny. A man's character is what it is ; it makes no 
 difference that, like everything in the universe, it has grown 
 according to assignable laws instead of springing into being 
 miraculously. Certain qualities of character are' virtuous, 
 and not the less so because their existence depends upon 
 conditions. The criterion of merit or responsibility is the 
 dependence of conduct upon character, and this remains 
 unaffected so long as- the character is the true proximate 
 cause of conduct. A man is not responsible when his hand 
 is another man's tool ; he is responsible whenever it is moved 
 by his will. I do not diminish a man's responsibility when 
 I '^ cause" him to act, but only when I cause him to '^act" 
 involuntarily. So far as I know a man's character, and
 
 286 THE SCIENCE OF ETHICS. 
 
 apply the motives which induce him to act^ I may be said 
 to cause his conduct, but I do not diminish his responsibihty. 
 If I give a man half-a-crown to shoot my enemy, he is not 
 the less a brutal murderer. His responsibility is measured 
 by the guilt of committing murder for half-a-crown. I have 
 only brought out the fact that he is a brute, and of course 
 encouraged the growth of his brutal habits. My guilt in the 
 murder is the same as if I had myself used the knife ; his 
 cruWt is the same as if his motive had been the plunder of 
 the victim instead of the bribe from me. 
 
 23. In this sense conduct may be '^caused" without 
 lessening the agent's responsibility; and this suggests the 
 inquiry which, if fully cleared up, would do more than any- 
 thing to remove the obscurities of these problems. I can 
 only touch upon it so far as it is relevant to the immediate 
 purpose. What is the meaning of the word '' cause " ? We 
 are apt to think of cause and effect as of two separate things, 
 one of which somehow governs or coerces the other. If we 
 carefully restrict ourselves to the necessary implications of the 
 word, and consider the cause as a constituent part of the total 
 process which is called the effect, many illusory associations 
 vanish. Thus, in the familiar phrase, a man is said to be 
 enslaved by his passions, as though he and his passions were 
 separable entities, and he could still be the same man without 
 them. All that can really be meant is that certain instincts 
 are unusually potent ; and to say what he would have done 
 without them is to say what a different man would have done. 
 They are not external fetters capable of being removed or 
 added without altering the man, but parts of the man him- 
 self. Yet this metaphorical phrase leads to a confusion 
 between strong will and absence of will, and we declare a 
 man incapable of choice just because he chooses so strongly. 
 In the same way we might speak of an assembly as being 
 enslaved by the majority, as though the assembly were an 
 entity separate from the majority ; and thus wc should con- 
 fuse a condition of energetic action with a condition of 
 impotence; such, for example, as one in which the assembly
 
 FREE-WILL. 287 
 
 is controlled by a foreign body. So^ if we state that a man's 
 conduct is determined by his character, we identify a state- 
 ment which implies the highest degree of volitional energy 
 with one in which volition has no influence. A lover is a 
 slave to his passion only in the sense that a part of himself 
 prompts him to vigorous activity, and his will is not sus- 
 pended but intensified. The case exemplifies the obvious 
 absurdity of confounding external with internal coercion, or 
 rather of using such a phrase as internal coercion at all. Self- 
 coercion can only mean determination in the logical sense. 
 Every conceivable object has certain qualities, and we are in- 
 dulging in a meaningless figure of speech if we speak of the 
 quality as something separate from the thing, and ^'forcing" 
 it to act in such and such a way. To say every material body 
 has weight is of course the same thing as to say that it is 
 heavy, and does not imply that weight is like a chain pulling 
 at the body and separable from it ; but some such feeling 
 seems to be always creeping in when we speak of the causa- 
 tion of conduct or of character. 
 
 24. Let us look at this a little more closely. What is 
 implied in the statements with which we are here concerned? 
 A man is an organism, and may be considered from without 
 as built up of mutually dependent organs, or from within 
 as consisting; of certain faculties or instincts. When we 
 say that his conduct is caused by one of those instincts, we 
 do not mean that there is a man plus the instinct, but that 
 the whole man, regarded as a unit, including this instinct, 
 acts in a certain way in which a man (if such a man be 
 possible) without the instinct would not act; or, again, if 
 the instinct be an essential part of the man, that the conduct 
 varies according to some variation in this instinct, or, in 
 other words, in the character considered in the corresponding 
 relation. "The cause of charity is benevolence," means 
 simply that benevolent men are charitable ; malevolent men 
 are not. ''The cause of eating is hunger," means that a 
 man eats more or less as he is more or less hun2:rv. If, 
 again, we speak of self-caused conduct, we use words inac-
 
 288 THE SCIENCE OF ETHICS. 
 
 curately^ as when we speak of self-coercion ; but the mean- 
 ing is clear enough. Briefly, we mean that the conduct in 
 question arises from an internal and not from an external 
 variation ; that it depends upon certain organic processes 
 which take place whilst the medium or external set of condi- 
 tions remains constant. I go to sleep because I am tired, 
 and my fatigue arises from my own activity, not from any 
 change in the surrounding conditions. A certain set of 
 external conditions is always necessary for my existence, and 
 therefore for my existence in any particular state, and I may, 
 if I please, consider them as a "cause" of my conduct. But 
 the cause is, so to speak, latent. It remains a constant, and 
 therefore is not relevant in determining the particular action 
 wdiich depends immediately upon processes taking place 
 within the sphere of my own organisation. In these cases, 
 then, to speak of my conduct as caused, is clearly not to 
 assert that there is something besides me and my surround- 
 ings which coerces me, but simply that I have certain 
 qualities which display themselves in certain ways, and are 
 the manifestation of my character. But we now have to 
 deal with external causes. My conduct at a given moment 
 is determined by my surroundings. I read because a book 
 is present, run away because there is danger, and so forth. 
 Am I therefore, as is sometimes said, the creature of circum- 
 stances ? Undoubtedly my conduct depends upon circum- 
 stances, but it is just equally true that it depends upon my 
 character. Both factors are essential, and neither at any 
 given moment is simply the product of the other. I am 
 now repeating the same set of intellectual operations as 
 before upon a larger whole. Instead of considering a man 
 as a separate organism, I include the whole set of processes 
 of which he forms a part, and, as before, there is a mutual 
 dependence, and no one part is to be regarded more than 
 another as determining or creating the other part. If, in 
 fact, I could analyse the whole universe (so far as is necessary 
 for the problem under consideration) into its constituent 
 factors, one of these factors would be the a<2;ent with certain
 
 FREE-WILL. 2S9 
 
 qualities. If the surroundings are all known, I can infer 
 various conditions by which the agent is limited, but there 
 would still remain an indeterminate factor, namely, the 
 character of the man himself. So, again, given the man, I 
 could infer a great part of the surroundings, namely, all the 
 conditions necessary for his existence in his actual state. This 
 merely states that wherever there is a complex whole with 
 mutually dependent parts, I can, so far as my knowledge 
 goes, infer either the parts from the whole or one part from 
 another; but it does not imply that besides all the constituent 
 parts there is a something else corresponding to a coercive 
 force. And this is equally true if I go back to what is called 
 the historical cause. The universe is a continuous system; no 
 abrupt changes suddenly take place. We could not suppose 
 them to take place without supposing that identical processes 
 might suddenly become different, which is like supposing that 
 a straight line may be produced in two different directions. 
 Hence every agent is a continuation of some preceding pro- 
 cess. He has not suddenly sprung into existence from 
 nowhere in particular ; the man has grown out of the 
 child. We might (though the language would be somewhat 
 strained) call the child in this sense the "cause" of the man. 
 But for the child the man would not exist. But there is not 
 a child pli/s a man, in which case there might be a coercion 
 of the man by the child. The child and man form a con- 
 tinuous whole, with properties slowly varying according to 
 its character and the external circumstances. A man, atrain. 
 has of course qualities which he has inherited; but this is 
 not to be understood as if there were a man plus inherited 
 qualities, which, therefore, somehow diminish his respon- 
 sibility. The whole man is inherited, if we may use such a 
 phrase. He starts at his birth with qualities dependent of 
 course upon the qualities of his parents, for their character- 
 istics and condition are the sole relevant conditions. The 
 fact that he inherits a particular temper no more implies that 
 he is one thing and the temper another thing superimposed, 
 than the fact that he inherits the general characteristics of 
 
 T
 
 290 THE SCIENCE OF ETHICS. 
 
 humanity would imply that the man is something in addi- 
 tion to all his essential qualities. From the child^ again^ we 
 may (within certain limits) infer the man. It is equally 
 true that from the man we may infer the child. It is true 
 that the man must have certain qualities which he would not 
 have had the child been different^ and it is equally true that 
 the child must have had certain qualities which he would 
 not have had if the man were different. In both cases we are 
 looking at a continuous process from varying points^ and we 
 can infer either backwards or forwards. The latter attitude is 
 more customary^ owing to the conditions of human conduct. 
 But it is not implied that something survives from the earlier 
 stage which is additional to^ and capable therefore of limit- 
 ing or coercing^ the later stage. Our power of inferring 
 simply expresses the fact of continuity and nothing more ; 
 and I may observe in passing that, even if we denied this 
 axiom, and supposed that men could spring into existence 
 out of nothing, it is impossible to see how their " responsi- 
 bility" or "free-will" would be affected. 
 
 25. The whole illusion in this part of the question seems 
 to rest upon a simple principle. Philosophers tell us (and 
 with undeniable truth) that "chance" is a mere name for 
 ignorance. It is a chance whether a penny will fall heads 
 or tails uppermost ; that is, we do not know which way it 
 will fall. I say equally when I am crossing a mountain 
 that it is a chance whether the other side of a rido-e is a 
 precipice or a slope. In this case, then, I assert nothing as 
 to the thing itself, for that is already there ; there is already 
 either a precipice or a slope ; but only assert that I do not 
 know which. And this, which is admittedly true of chance, 
 is equally true of the necessity which is the negation of 
 chance. It is simply a name for certainty, as chance for 
 absence of certainty. Thus, if I say you will "necessarily" 
 act in such a way, I mean that I know that you will act in 
 that way, and I mean nothing more. My knowledge may 
 be founded upon a knowledge of your character, and upon 
 a knowledge of certain physical limitations which make your
 
 FREE-WILL. 
 
 291 
 
 character irrelevant. In this sense, conduct or action de- 
 pendent upon character can rarely be certain or " necessary" 
 beyond a certain point, because we have rarely any accurate 
 knowledge of character. The data are more complex, but 
 they are not in themselves less determinate. An indeter- 
 minate thing is a non-existent thing, for it would be 
 something which is at once one thing and another thing. 
 The same applies to all such words as "necessary," "^^ pro- 
 bable," "potential," "possible,'* and so forth. They arc 
 simply names of the observer's state of mind, which, by 
 one of the most familiar of all fallacies, are supposed to 
 denote qualities of the thing observed. What I see to exist 
 " necessarily," may be only probable for you, but the thing 
 itself either exists or does not exist, and by saying "necessary" 
 I add nothing except a statement as to my knowledge. The 
 use of the word " necessary " in regard to conduct has given 
 offence from this common confusion of ideas. If I sav. for 
 example, that a man will necessarily commit murder under 
 given circumstances, I mean that I am quite certain that he 
 will. Why am I certain ? Because I know him to have the 
 murderous character. I do not assert the existence of a neces- 
 sity as of something external to his character, which will make 
 him commit murder whether he likes it or not, but simply 
 that I know that he will like it. And thus the only meaning\\ 
 of the necessity of conduct is that people have fixed qualities \\ 
 of character, which do, as a matter of fact (not determine ) 
 their conduct from without), but manifest themselves m// 
 conduct; in other words, that people have characters. If 
 they have not, I do not see how they can be virtuous, and 
 meritorious, and responsible. 
 
 26. Finally, let us return to the causa causce argument. 
 The statement is, that if action is caused, the original cause 
 alone can be responsible. If this is interpreted bv the 
 hypothesis of a Creator separate from the universe, who 
 has yet moulded it as the seal moulds the wax, and who at 
 every instant sustains and supports it, I admit that it is im- 
 possible to conceive of responsibility to such a being. It is
 
 292 THE SCIENCE OF ETHICS. 
 
 the case of a man bribing me to a crime and then punish- 
 ing me for committing it. I may be equally responsible^ as 
 I have said^ to others, but I do not see how I can be re- 
 sponsible to him. But the bare fact that an omniscient, or 
 even a highly intelligent being could foresee my actions with 
 certainty, would not destroy my responsibility. Such fore- 
 knowledge, it used to be argued, implied predestination, and 
 therefore an incapacity of the individual to act otherwise. 
 But the determination implied would not be an external 
 fate coercing the character, but simply the character itself. 
 I know that a man will remain motionless, not because I 
 know him to be in fetters, but because I know that he is 
 indolent. In one case, I should infer that he will be motion- 
 less whatever his character ; in the other, that he will be 
 motionless, assuming his character. The two cases are really 
 mutually exclusive, though they are so often confounded. 
 Prediction, in short, does not imply a man plus a fate, but a 
 man alone. 
 
 27. I am not, however, properly concerned with the so- 
 called transcendental causes. Within the sphere of scientific 
 thought, the case is as I have argued it. A man is respon- 
 sible, I have said, for those actions which are caused by his 
 moral character. The fact that he is himself " caused," or 
 that his actions are caused by circumstances, is only relevant 
 in so far as this statement means that his character ceases to 
 be a cause at all ; in other words, as the same phenomena 
 would result whatever his character might be. If my 
 benevolence causes my action, if, that is, a person not 
 benevolent would act differently, then I have the merit of 
 benevolence. The fact that my action depends upon circum- 
 stances — that is, would be different if circumstances differed 
 — is irrelevant if it also depends upon my character. My 
 character is still a true cause. The fact that my character 
 has been developed under the action of circumstances is 
 equally irrelevant so long as it is my character. The two 
 factors are implied at every stage of the process, and the 
 so-called " dependence " of one upon the other means simply
 
 FREE-WILL. 293 
 
 that T cannot infer the whole phenomenon from either taken 
 separately. Finally^ if I refer to the origin or the historical 
 cause of my character, it is of course true that the man's 
 character would be different if the character of the child's 
 character had been different, or the child's if the parents' had 
 been different. If the child had certain instincts, the man 
 will have corresponding iristincts ; as also if the parents 
 were monkeys instead of men, the child would be a monkey. 
 But there are not separate qualities which are left behind by 
 the parents or the child to hamper the man ; the whole of 
 the man's qualities are continuous with the qualities existing 
 in any previous stage of the same person. There is not a 
 common something which becomes either monkey or man 
 as a different form is imposed upon it. If we could suppose 
 a sudden appearance of the man out of nothing or by the 
 fiat of a Creator, we should still only shift the responsibility to 
 the Creator or to chance. We may infer the child from the 
 man as well as the man from the child, and there is no more 
 necessity in the one case than the other, except in the sense 
 which means certainty. When we know from one pheno- 
 menon that another exists, it is simply that we can (for some 
 reason) identify the two as parts of a whole of mutually 
 dependent parts. From an eye we infer an ear or a leg ; it is 
 not because the eye has a power to make ears and legs out of 
 formless matter, or because, besides eyes and ears and legs 
 and every part of the organism, there is some additional 
 coercive force which holds them together, but simply that 
 each part carries with it a reference to the rest. The diffi- 
 culty is dispelled so far as it can be dispelled when we have 
 got rid of the troublesome conception of necessity as a name 
 for something more than the certainty of the observer. 
 When we firmly grasp and push to its legitimate conse- 
 quences the truth that probability, chance, necessity, deter- 
 mination, and so forth, are simply names of our own states 
 of mind, or, in other words, have only a subjective validity; 
 that a thing either exists or does not exist, and that no 
 fresh quality is predicated when we say that it exists neces-
 
 294 THE SCIENCE OF ETHICS. 
 
 sarily; and that all dependence of one thing upon another 
 implies a mutual relation and not an abolition of one of the 
 things — we have got as far as we can towards removing the 
 perplexity now under consideration. But it is not likely to 
 disappear to-day or to-morrow. 
 
 TIL Efort. 
 
 28. I pass, therefore, to a further condition of morality, 
 which involves some similar ambiguities. The free-will 
 condition implies (upon my view) a misunderstanding of the 
 undeniable proposition that a man can be virtuous or meri- 
 torious in respect of those actions alon^ which are con- 
 ditioned by his character, or which are really his actions. 
 The agent must be really an agent^ not a bit of mechanism, 
 or a transmitter of internal force. But, in the next place, as 
 I have said, the qualities in respect of which a man is meri- 
 torious must be those which are amenable to discipline. 
 The moralised man is the trained savage ; he helps a stranger 
 instead of attacking him ; and this modified instinct which 
 he has acquired through the social factor distinguishes him 
 from the immoral man, in whom the instinct is deficient. 
 Thus, again, it seems that merit implies an effort. Those 
 actions, I have said, are regarded as meritorious which the 
 average man would not do without extrinsic reward, and 
 that character is meritorious which implies the modification 
 necessary to such conduct. Hence, again, it is often said, 
 with more or less propriety, that a man deserves nothing 
 for conduct which is pleasant or purely " natural." As these 
 statements involve various perplexities, I must seek to give 
 at least the clue by which they may be unravelled. 
 
 39. I observe, then, that (as already stated) a judgment 
 of merit implies reference to some standard ; and this implies, 
 again, that the person is supposed to belong to a certain 
 class, and to have the essential properties of that class. 
 When I say " a good " or " a bad " man, I of course use the 
 words in a diflcrcnt sense from those in which I should speak
 
 EFFORT. 295 
 
 of a good or bad horse. I mean that the man is better or 
 worse than the average man, and I take for granted that 
 he has all the qualities essential to humanity. This already 
 involves one ambiguity, for it becomes a question how far 
 certain qualities which may be absent should put a man 
 outside the class to which we are in all cases making a tacit 
 reference. What are the characteristics, one might ask, 
 which bring a man within the sphere of morality ? It is 
 not "fair," we often say, to judge a man in one age by the 
 moral standard of another age. It is not fair to judge a 
 man exposed to temptation by the standard applicable to 
 one who is not tempted. And with these difficulties may 
 be classed the more practical difficulties which so frequently 
 occur in criminal law, when we have to decide how far a 
 child or a madman can be considered as responsible for his 
 actions. The difficulty results to a great extent from a 
 confusion between different classes of questions. When we 
 speak of the merit of an action, we really speak, as I have 
 said, of the qualities which it implies. If we have solved 
 that question, which may be one of extreme difficulty, we 
 have next to ask how far and in what way the man who has 
 those qualities differs from the moral man, and how far, 
 therefore, they imply morality or the reverse? Keeping 
 these problems apart, we shall see, I think, that though they 
 involve great difficulties of application, the governing prin- 
 ciple is sufficiently clear. 
 
 30. A man tortures a prisoner, but torture is not con- 
 demned by the morality of the time. Is the torturer wicked? 
 Clearly not wicked according to one standard, and clearly 
 wicked accordino; to ours. Which standard should we 
 apply ? That, I should say, depends upon the problem 
 which we are considering. He is not proved to be a bad 
 man according to the morality of the day. The action, 
 therefore, does not imply the same degree of cruelty which 
 would be implied by a similar action now. We should not 
 be justified in inferring that he was cruel in other relations 
 of life, as we should now be justified in a similar inference.
 
 296 THE SCIENCE OF ETHICS. 
 
 Suppose that we have made our inferences correctly, and 
 discovered that the man was up to the average standard of 
 his time, our judgment is complete. He was not in advance 
 of his time, and we do not respect him so much as we should 
 respect one who was in advance. He is not so good a man, 
 again, as a man who is up to an improved moral standard; 
 that is, he is not so uniformly merciful or benevolent. 
 Possibly his innate qualities were as good ; and had he been 
 born under the improved standard, he would have been 
 as good. That, as it seems to me, is all that we can say. 
 "Merit" does not mean a separable quality which is the 
 same at all periods, but carries with it a statement of rela- 
 tion to a varying standard, and therefore cannot be definitely 
 valued apart from the circumstances without a sophistry. 
 Each of the two men, we must reply, is equally good, 
 measured by his own standard ; their innate qualities may 
 have been equally good also; but one man is absolutely 
 better in so far as he represents a more advanced type of 
 humanity. We cannot sum up these various statements by 
 a single statement about " merit." That can only be done 
 satisfactorily when we assume that both agents belong to 
 the same class, and are to be judged by a single standard. 
 We fall into the same kind of perplexity as if we were to 
 compare the wealth of a man of to-day with the wealth of 
 his ancestor. We may compare the amount of gold which 
 each man possesses, or the amount of actual enjoyment 
 which it purchases, or the position in the society which it 
 confers; but to answer definitely the question which is in- 
 volved, we must distinguish the sense in which the question 
 is asked. Then the ambiguity disappears, though the diffi- 
 culty of answering may be immense. 
 
 31. The question is more perplexing when we have to do 
 with cases such as the child or the madman, when the indi- 
 vidual has qualities (or an absence of qualities) which seem 
 to take him out of the class. But the same mode of answer- 
 ing applies equally. Thus, for example, children and some 
 adults are without the passions which are regulated by one
 
 EFFORT. 297 
 
 important part of the moral law. What are we to think of 
 them ? Simply this, that they are without these passions. 
 If^ then, they refrain from certain vices, their abstinence 
 does not prove chastity. It simply shows that in respect 
 of this quality they are neutral, neither good nor bad. It 
 would be a clear mistake in logic to identify the case of the 
 man who has certain passions under control and the man 
 who has not the passions at all. But a man is neither moral 
 nor immoral in so far as he has the passions, for in them- 
 selves they are neither good nor bad. The man who has 
 them not will indeed be so far disqualified for certain social 
 relations, as, on the other hand, he will be free from certain 
 temptations. The ordinary inferences as to character, and 
 therefore our judgment of his virtue or merit, will be falla- 
 cious. But he will still have a character, which may be 
 either virtuous or the reverse in the highest degree in all 
 other respects, and we shall judge him, if we judge rightly, 
 according to the conformity of his character to the type, so 
 far as it goes. He cannot be a perfect man in the full sense, 
 for he is without qualities which are essential to the race, 
 but then the defect is not such as to imply any moral 
 defect in any of the relations which he is capable of fulfilling. 
 It would therefore be absurd to call him either moral or 
 immoral in respect of this defect. We can only say that he 
 is defective, but that the defect is morally neutral. There 
 may be a liability to error in our ordinary inferences as to 
 the facts of character, and, when the facts are known, in 
 saying what degree of deflection is implied ; but the prin- 
 ciple seems to be simple. 
 
 33. Thus, again, we have the case of the madman, with 
 all the ever-recurring difficulties of deciding what is meant 
 by madness. It is agreed, of course, that a madman is not 
 responsible so far as he is under actual illusions, and supposes 
 himself to be beheading a cabbage when he is really killing 
 a man. The test and the ordinary inferences as to character 
 would fail. But madness shades into sanity by imperceptible 
 degrees, and we are asked to pronounce a moral judgment
 
 298 THE SCIENCE OF ETHICS. 
 
 upon such cases as that of the "homicidal mania." A mad- 
 man, it is suggested, is not responsible, because he cannot 
 help it or has lost his free-will ; the sane man can help 
 himself, and is therefore responsible. As I consider free-will 
 to be an illusion, I cannot accept this theory, nor can I see 
 why the madman should be supposed to be without it. The 
 more unaccountable a man's actions, the more one would 
 be inclined to admit the presence of some arbitrary agent. 
 A man who possessed free-will in a large degree, whose 
 actions obeyed no assignable rule, would certainly appear 
 to be mad. The argument, however, points to the obvious 
 explanation. No one would consider a man to be less 
 responsible for a murder in proportion to the strength of his 
 malice. The more malicious he is, the more certain he is to 
 commit murder ; the less is his malice restrainable by fear, 
 or conscience, or any other motive ; and therefore, in the 
 judgment of every man^ the greater is the crime. We hold 
 the madman to be not responsible for precisely the opposite 
 reason, namely, that in him murder does not imply malice, 
 but some different impulse; he is not accessible to the 
 ordinary motives. '' Homicidal mania," if used simply to 
 imply a high degree of cruelty or malice, would not take the 
 sufferer out of the ordinary moral code ; it can only do so 
 when it means that his mental machinery is out of gear, so 
 that the ordinary motives do not have their normal effect. 
 His mind, we say, is deranged. He deviates from, the type, 
 not as a man deviates in whom certain passions are unusually 
 strong, but by some undefined and undefinable organic dif- 
 ference, which prevents him from being amenable to the 
 natural motives. Murder is in him not the manifestation of 
 malice, but a proof that his brain is in an abnormal condition, 
 due perhaps to an accident or to some obscure constitutional 
 defect entirely unconnected with his moral character, and 
 therefore not justifying the ordinary inference. This, as it 
 seems to mc, is more or less explicitly assumed in all cases of 
 madness. We do not call a man mad uuless we assume a 
 deflection in his psychological organisation from that of the
 
 EFFORT. 299 
 
 type which prevents him from being amenable to the moral 
 motives, and such a deflection as implies disease and a dis- 
 integration of the faculties necessary for a due discharge of 
 the vital functions. The difficulty of saying in a given case 
 what constitutes this deflection is enormous; but in any case 
 we do not judge that the actions of the sane man are 
 independent of motives whilst those of the madman are de- 
 pendent, but that whilst both have motives, the madman's 
 motives are ^'irrational" or abnormal. 
 
 ^^. The perplexity which thus intrudes itself into our 
 conceptions of merit is a difficulty in judging of facts. 
 What character does a given action imply? What is the 
 value of a given character ? What is the class to which a 
 man is to be referred, and how far should our judgment be 
 affected by his passions or qualities which distinguish him 
 from other members of the class ? All such questions are 
 very difficult in themselves, and in our rough daily estimates 
 of merit we have to solve them after a very summary fashion, 
 and often with very inconsistent results. We apply one 
 measure of merit in particular, which, though perfectly 
 rational in itself, has led to various perplexities. Conduct, 
 I have said, is meritorious so far as it proves merit ; it can 
 only prove merit so far as it implies difficulty for the average 
 man. Leonidas was as brave before Thermopylae as after ; 
 he was only known to be brave when he had proved his 
 courage by accepting a fate from which cowards shrank. 
 Hence, it is argued, merit only accrues in respect of difficult 
 performance; whilst again it is said the greater the difficulty 
 the greater the merit, and thus that whatever a man does 
 simply to please himself is no credit to him. 
 
 34, The fallacy which is sometimes involved in this state- 
 ment is too simple to require a long examination ; the full 
 statement of the case is enough to expose it. The man is 
 most meritorious who has most virtue; consequently, if we 
 assume that a certain task has to be performed, the man 
 who performs it most easily is the most virtuous. Leonidas 
 was braver in proportion as he had no tendency to run
 
 300 THE SCIENCE OF ETHICS. 
 
 away ; I am honester in proportion as T feel less disposition 
 to pocket my neighbour's spoons. A man who felt no 
 disposition whatever to commit any sin would so far be 
 absolutely perfect, and such a character is attributed by 
 Christians to a divine man. Christ was not the less perfect 
 if he never felt the least velleity to do wrong; on the con- 
 trary, such a character represents the unattainable moral 
 ideal. For the same reason it is true that, if we suppose the 
 task to increase in difficulty, the man is most honest who 
 overcomes the greatest difficulty — that is, the greatest diffi- 
 culty for a given strength. The less the difficulty for him, 
 the greater the difficulty which he can overcome. The 
 greater the danger, the greater the bravery ; the heavier the 
 bribe offiired, the greater the honesty displayed in resisting 
 it, and so forth. The principle is precisely the same as in 
 the case of a mechanical exertion ; the man is the strongest 
 who can lift the heaviest weight, or who can lift a given 
 weight with the greatest ease. But (and it is a proof of the 
 loose argument which has often been accepted in ethical 
 disputes) the two cases have sometimes been confounded. 
 It would plainly be absurd to say, " The man is strongest who 
 lifts the greatest v»^eight ; therefore the man who makes the 
 greatest effort ; therefore the man who makes the greatest 
 struggle to lift a given weight." But it has occasionally 
 been said that the man is most virtuous who resists the 
 greatest temptation ; therefore the man who has the greatest 
 struggle; therefore the man who has the greatest difficulty 
 in resisting a given temptation. Though the fallacy 
 docs not occur in this bare form, it is not unfrequently 
 implied in the assumption that the effort, taken absolutelv, 
 is the measure of merit ; we are occasionally tempted, that 
 is, to confound the difficulty which arises from an extrinsic 
 or morally neutral motive with that which arises from the 
 moral (or immoral) impulse itself. We are thus led to 
 excuse a man for the very qualities which make him wicked. 
 True he committed a murder, but he was so spiteful that he 
 could not help it ; or he was exceedingly kind, but he is so
 
 EFFORT. 301 
 
 good-natured that it cost him no efTort. Obviously such 
 reasoning is absurd, but it suggests the necessity of guarding 
 our statement. 
 
 '^^. The man, I have said, is most virtuous who performs 
 a given virtuous act with the least effort ; but on looking 
 closer, we see that this statement might lead to the mis- 
 understanding already explained. We might argue, in fact, 
 that temperance showed greater virtue, when, in truth, it 
 merely shows a defect in some faculty of enjoyment. If a 
 man resists any inducement because it has no charms for 
 him, his act does not prove virtue, unless the inducement be 
 such as to appeal only to the wicked. Our formula really 
 involves an assumption which must be explicitly stated. The 
 man is most meritorious who is virtuous with least effort, 
 provided always that he has the normal passions of a man. 
 We virtually assume as the basis of our comparisons that two 
 men are constituted in the same way so far as their moral 
 qualities are not involved ; that each of them is equally sen- 
 sitive to certain modes of enjoyment so long as the enjoy- 
 ment is not opposed by any moral motive. Then we say, 
 and are justified in saying, that the man is most virtuous 
 who is so constituted that virtuous conduct is easiest to him. 
 If, for example, two men enjoy the taste of wine equally, 
 the man is most temperate to whom abstinence from exces- 
 sive use of wine costs the least exertion. But if one man 
 abstains because he dislikes wine, we cannot argue as to his 
 moral superiority; for, in the first place, his abstinence in 
 that case affords no presumption that he will refrain in other 
 cases ; and, in the next place, a man is so far the worse as 
 he is wanting in any capacity not intrinsically bad. A taste 
 for wine exposes a man to certain temptations, but it is also 
 presumably a symptom of certain organic advantages which 
 make him so far a more eflective being than the man in 
 whom it is deficient. 
 
 '^6. This brings us to a very important point. The infer- 
 ence from a particular act is always precarious, and part of 
 our perplexity arises from deciding what is or is not implied
 
 302 THE SCIENCE OF ETHICS. 
 
 in any given line of conduct. Conduct, which is the same 
 so far as the external circumstances are concerned, may be a 
 manifestation of very diiferent qualities of character acci- 
 dentally coinciding upon this particular point. The man 
 with strong but disciplined passions may act in the same 
 way upon a given occasion as the man without passions, 
 though he would on other occasions act differently. Hence 
 the "merit" of the action is indeterminate, inasmuch as 
 it may result from different motives. Putting this aside, or 
 assumins; the motive to be ascertained, we have therefore a 
 further question of facts. We may ask, in fact, whether the 
 primitive instincts of which a man's nature is composed are 
 to be regarded as morally indifferent, or whether there are 
 various primitive instincts, some of which are morally good, 
 whilst others are morally bad? In the latter case, the suppres- 
 sion of a bad instinct would of course imply a higher pitch 
 of virtue ; and this doctrine seems to be assumed in ascetic 
 systems, and is sometimes even pushed to the degree of 
 maintaining (in words at least) that all the natural instincts 
 are bad. It w^ould be superfluous for me to assign the 
 grounds upon which I reject this theory. The whole doc- 
 trine of evolution seems to imply that absolutely pernicious 
 instincts are eliminated in the struggle for existence, and to 
 fall in with the other assumption that virtue implies a cer- 
 tain organisation of the instincts, and not the extirpation of 
 any existing instincts. Assuming this for the present, the 
 inference as to the particular question before us seems to be 
 simple. 
 
 37. Every new sensibility or faculty is so far an advantage 
 to the agent. A man is a completer being or at a higher 
 stage of development in so far as he has acquired any faculty 
 not shared by his fellows. (I assume, of course, that it also 
 implies a total increase of power.) Again, every such faculty, 
 so far as it is morally indifferent, exposes its possessor to fresh 
 temptations, as well as gives him fresh capacities for virtue. 
 Cultivated tastes often encourage indolence, as they also 
 enable us to confer fresh services upon our fellows. If even
 
 EFFORT. 303 
 
 a sensual appetite, such as a love of wine, exposes a man 
 to temptation, it also helps materially to save him from a 
 sour and unsociable temper. A man so far as he possesses 
 greater faculties is not necessarily more moral or less moral ; 
 he is only capable of a more extensive virtue or a wider 
 deviation from virtue. He is moral in the highest sense 
 when he possesses the most richly endowed nature, and when 
 it is also so disciplined that he — that is, when his instincts — 
 are so correlated and organised that he spontaneouslyobeys the 
 moral law. To be defective in any faculty is to be on a lower 
 platform, and to come under the moral law at a smaller 
 number of points. An action which proves such a defect 
 proves nothing as to the morality of the agent in the cases 
 where morality is possible for him. But it is also true, of 
 course, that such a defect saves a man from certain tempta- 
 tions, as it also removes some opportunities for moral excel- 
 lence. He is without a faculty which may lead him into 
 moral ruin, as, if properly disciplined, it may enrich and 
 strengthen his morality. 
 
 38. Hence, therefore, we may say that, in one sense, effort 
 is essential to merit. A man, that is, must possess the 
 " natural" instincts in order to become a proper subject for 
 morality at all. If he possess them in any given degree, 
 he is the more moral in proportion as they are so disciplined 
 that moral action is easy for him. But inasmuch as the 
 instincts are themselves neutral, this always implies a certain 
 accessibility to temptation, and therefore a certain struggle. 
 He cannot be properly temperate unless he is capable of 
 enjoying the pleasures which tempt to excess. If he has no 
 such capacity, he is not in this respect a moral agent at all. 
 So far therefore as the effort is taken as a symptom that he 
 is accessible to the temptation, it is certainly essential to 
 virtue. But assuming a certain strength of the appetites, 
 the more virtuous he is, the less will be the effort in a given 
 case, for the more thoroughly they will be harmonised and 
 disciplined. No human being, we may add, can be absolutely 
 or infinitely virtuous. Every motive in a finite being must
 
 304 ' THE SCIENCE OF ETHICS. 
 
 have a finite strength. We may therefore say with the cynic 
 that every man has his price. Some conceivable strength of 
 temptation would overpower any human virtue. Thus a 
 capacity for sinning is implied in a capacity for doing right, 
 for both imply the existence of passions which may be 
 enlisted on either side of the struggle. We may even say in 
 this sense that the more a man is capable of sinning, the 
 more he is capable of virtue, for the virtuous and the vicious 
 character are different modifications of the same primitive 
 instincts, and the man is on a higher stas;e the more thev 
 are developed. But given the instincts, the temptation to 
 sin and the disposition to sin can no more prove a man to be 
 virtuous than the actual sin. 
 
 IV. Knowledge. 
 
 39. This brings us to the third consideration. The doctrine 
 that virtue implies effort is associated with the doctrine that 
 the primary instincts are either corrupt or neutral. Some 
 theological systems, starting from the dogma of the corruption 
 of human nature, are logically drawn to the conclusion that 
 virtue represents a supernatural influence governing and 
 subduing the natural instincts. The philosopher who rejects 
 this domia and denies the validitv of this division between 
 the natural and the supernatural so far admits the method 
 as to assign to reason the functions of divine grace. Reason 
 thus dominates and controls the passions, though they are 
 not regarded as positively bad, but rather as neutral or 
 morallv indifferent. So far as a man acts from instinct he acts 
 " irrationally,^' and therefore not morally. Every instinct, 
 as I have said, may be the ally or the opponent of the moral 
 motives. A mother's love makes her sacrifice herself to her 
 children, but it may also lead her to cruelty to others for 
 their fjood. It is onlv virtuous then, or meritorious, insofar 
 as it implies a love of virtue and a principle of action which 
 will guide her when her passions are on the wrong side. 
 Conduct, it is inferred, is truly virtuous when, and only
 
 KNOWLEDGE. ' 305 
 
 whcn^ the action is done because it is virtuous. The motive 
 must be the pure love of virtue or of virtue for its own sake. 
 Otherwise, since conduct dictated by the emotions alone is 
 morally neutral, or even, as it is sometimes said, arbitrary, 
 it cannot be regarded as properly speaking virtuous. 
 
 40. The principle involved in this reasoning is implied in 
 all that I have said. To say that an action is virtuous is to 
 say that it is a manifestation of a virtuous character. It is 
 perfectly true, therefore, that we cannot infer virtue from 
 an action which springs from a motive conmion to all men 
 or to all animals, such as hunger or fear. An action which 
 only proves that a man has an appetite or dislikes pain 
 does not prove him to be either good or bad. A man's 
 virtue or intrinsic merit depends upon his whole character, 
 and his character is a complex of many instincts connected 
 by laws inscrutable to our imperfect methods, and acting 
 and reacting in a vast variety of most delicate and intricate 
 ways. The question as to virtue or merit is the question as 
 to whether the whole character does or does not imply action 
 in conformity with the moral law. The question, then, 
 may be asked, how far such conformity is possible without 
 an explicit recognition of the law. If it is possible for a 
 man to act as the law directs without so acting because the 
 law directs, then his conduct is said to be "materially" but 
 not "formally" moral. Now there can be no a priori objec- 
 tion to such a conformity. It is not true (as some language 
 seems to imply) that conduct dictated by instinct or emo- 
 tion is arbitrary. Nothing is arbitrary. Every instinet 
 has its own law, though it may not consciously recognise 
 the law. And though each instinct, taken separately, may 
 be morally indifierent, it does not follow that the complex 
 character built up of those instincts is morally indifferent. 
 An instinct is not a separate entity, but expresses the reac- 
 tion of the whole character under a special stimulus; and 
 the whole must have laws of its own, which express more or 
 less approximation to, or deviation from, the moral type. 
 
 How far that conformity involves an explicit recognition of 
 
 u
 
 3o6 THE SCIENCE OF ETHICS. 
 
 the law is a question of facts^ of great complexity doubtless, 
 upon which a few words must here be sufficient. 
 
 41. It is clear^ in the -first place^ that reason is in some 
 sense essential to virtue. I deny the possibility of separating 
 the reason from the feelings as two distinct faculties^ and 
 hold, on the contrary, that in every act of an intelligent 
 being both reason and emotion are necessarily involved. 
 Morality proper becomes possible at the point at which 
 sympathy is possible ; and sympathy involves reason, for it 
 involves the recognition of other centres of consciousness. 
 The germs of morality may be present at an earlier stage, 
 but the germ is not the developed product. The moral law, 
 again, involves the knowledge of many facts, and often of 
 very complex facts. Cruelty and kindness cannot emerge 
 until we recognise the existence of other sentient beings, 
 nor truthfulness or justice till we can understand general 
 rules, nor chastity and temperance till we can distinguish 
 between the cases of legitimate and illegitimate indulgence. 
 The growth of any of the social relationships, of the state or 
 the family, imposes obligations which cannot be appreciated 
 until those relationships are intelligible; and in the higher 
 stages of life this involves very complex conceptions. Every 
 development of intelligence implies an extension of the 
 emotional range, and consequently of the sphere of duty. 
 Conversely, every extension of the moral code implies a deve- 
 lopment both of intelligence and feeling. A man who could 
 feel without thinking or think without feeling is an incon- 
 ceivable being, and we have no more to do with him than 
 an ordinary geometer with space of four dimensions. 
 
 42. Consistently with this, it still remains such that a man 
 may (and indeed must) act regularly though he does not act 
 by rule ; or again, that he may act in conformity with a 
 moral rule without an explicit regard to its moral character. 
 He may become aware that, in fact, it expresses his character, 
 though he has not referred to it as the reason of his actions. 
 The moral law is the ^'objective" law of his conduct, but 
 not the '''subjective" law or " reason." I eat when I am
 
 KNOWLEDGE. 307 
 
 hungry ; so far I act temperately, though I may not think of 
 the rule. I help a man in distress because distress always 
 affects me, though I never think of the general principle ; 
 and so far I am charitable without recoo-nisino; the dutv of 
 charity. The more spontaneous the kindly feeling, the less 
 the need for reference to a moral rule; and accordingly we 
 regard a man with some suspicion who has always to remind 
 himself of the moral obligation in order that he may do what 
 others do spontaneously. This indeed follows if we regard 
 morality as the product of a character which allows the 
 various instincts to operate in their proper places instead of 
 superseding them by some different faculty, and it suggests 
 the answer to our difficulty. Maternal love, it is said, is 
 not virtuous so far as it is a mere natural instinct. This 
 (upon our hypothesis) means that it is not virtuous because 
 it is consistent with many bad qualities, and may prompt to 
 injustice on behalf of the children; but no one can deny 
 that it is so far virtuous that it is implied in a virtuous 
 character. We should condemn a mother in whom it was 
 deficient, and condemn her expressly on the ground that she 
 would be ''wanting in natural feeling." The feeling does 
 not constitute positive merit, because it is assumed in the 
 average standard ; but for the same reason its absence is a 
 positive demerit, or a conclusive proof of inferiority to the 
 moral type. The mother who had no affection for her child 
 would be so far a bad woman ; she would be without the 
 emotions which are the very groundwork of all the virtues. 
 It is of course true that the affection is not sufficient bv 
 itself to make her virtuous, and that even then, in many 
 cases, it may lead her astray. Still it is essential to virtue, 
 and, in the vast majority of cases, favourable to virtue; 
 whilst, in the rare cases in which it leads to wrongdoino-, 
 it may be held in check by other sympathetic feelings which 
 spring from the same root without any conscious reference 
 to a moral law. It is true that so long as that conscious 
 reference is absent we are without the full guarantee for 
 a regular observance of the moral law. A recognition of
 
 3o8 THE SCIENCE OF ETHICS. 
 
 that law is, on every hypothesis, the crown and final outcome 
 of the moralised character ; but a close approximation to 
 morality may exist previously, just because it is a product, 
 and not a precedent condition. If this be the true state of 
 the case or an approximation to it, if morality may exist 
 in fact without a previous reference to the moral law, we 
 are quarrelling over words if we deny the name " virtuous " 
 to such characters. 
 
 43. What is true of the maternal affections is true of all 
 the instincts by which conduct is determined. A man is 
 virtuous, not so far as he is without passions, nor so far as 
 they are dominated by some external force, but so far as they 
 constitute a harmonised whole, determining the approxima- 
 tion of character to the best social type. The main and 
 essential condition of morality is the altruism which enables 
 a man to appropriate the feelings of others^ and so to acquire 
 instincts with a reference to the social good. That quality, 
 which is not a separate organ, but an inseparable incident of 
 the development of the reasoning and feeling agent, supplies 
 the necessary leverage upon which the social sentiments 
 operate. Its bare existence, even in a high degree, does not 
 suffice to make a man thoroughly moral, but it capacitates him 
 for morality, and renders immorality proportionately difficult. 
 A man is moral because and in so far as his instincts are 
 correlated according to a certain type. He cannot become 
 thoroughly moral, especially in a society in which the moral 
 law has been distinctly formulated, without becoming con- 
 scious of the law of his own action, and the recognition be- 
 comes of great importance as a guide for his conduct in cases 
 where, without such a rule, he might fail to perceive the true 
 nature of his conduct. He may be brave, temperate, and 
 truthful without ever reflecting upon the law which com- 
 mands the corresponding actions; his sympathies may be so 
 strong as to guard him against most kinds of wrongdoing ; 
 and if so, I should certainly call him virtuous, though he 
 should never think of right and wrong as such, but be 
 always guided by his immediate feelings. Undoubtedly such
 
 KNOWLEDGE. 309 
 
 a man would still be defective, as a man would be defective 
 who trusted to his immediate perceptions for the knowledge 
 that two sides of a triangle were greater than the third in 
 every particular case, instead of recognising the general 
 principle. But upon my theory, the recognition of the 
 general rule follows from the specific intuitions instead of 
 preceding them. Upon a different philosophical theory the 
 facts would be differently expressed. Meanwhile I can only 
 say, that, upon my showing, it is simply a question of fact 
 how far the actual observance of the law involves a recogni- 
 tion of the law, and one which cannot be solved by a priori 
 considerations. 
 
 44. This brings us to the question which must be more 
 fully considered in the next chapter. What is the nature of 
 conscience, or the intrinsic motives to right-doing? The 
 question is obviously of the highest importance, and the 
 answer to it will clench this part of the inquiry. At pre- 
 sent I have only tried to show that, whatever it may be, 
 intrinsic merit is properly a name for the virtuous character 
 considered in a special relation, and that the perplexities 
 arise in great measure from a confusion between merit in 
 this sense and that merely relative merit which means not 
 the actual virtue, but the virtue proved by any given action; 
 or, again, between the qualities considered in themselves and 
 the estimate placed upon them at different periods. Clear- 
 ing away these difficulties, there remain difficulties which I 
 have considered, and of which I need only remark in con- 
 clusion, that they are all more or less associated with the 
 familiar illusion sanctioned by the double sense of the word 
 'Maw." A scientific law, as has been so often said, is nothing 
 but a generalised statement of facts ; but, in spite of all that 
 has been said, it is difficult to avoid the impression that it 
 corresponds in some way to an external something impressed 
 upon the facts. In this way true if not self-evident state- 
 ments lead to inextricable labyrinths of confusion. If we 
 speak of a man's character as obeying a law, we mean only 
 that he has a fixed character. We are taken to mean that
 
 3IO THE SCIENCE OF ETHICS. 
 
 besides this character there is a law which governs the cliar- 
 acter^ and thus external coercion is confounded with internal 
 determination, and freedom, in the sense of freedom of the 
 character from itself as well as freedom from an outward 
 restraint, is made a condition of moral conduct. Again, if 
 we take the rule as a something forced upon a set of bad or 
 neutral instincts, we assume that an effort corresponding to 
 this process is essential to virtue. Virtue is not simply the 
 expression of a certain harmony between the instincts, but 
 a constraining power opposed to instinct and emotion in 
 general. And so, finally, this abstraction becomes the sole 
 principle of virtuous conduct, and virtue is only possible in 
 so far as it is recognised. Thus, according to my theory, 
 the various conditions of merit which we have been con- 
 sidering are distortions of some perfectly simple principles. 
 The simple fact is, that when we speak of a man's conduct 
 as virtuous, we assume that it is really conduct, that is, really 
 determined by his character; that it is the conduct of a real 
 man, that is, of one who has the normal instincts, appetites, and 
 emotions ; and really virtuous, that is, the manifestation of a 
 character which conforms to the type and implies a uniform 
 obedience to the law. This is distorted into the assumption 
 that a man must be free, not only from outward restraint, 
 but from his own character ; that his feelings must not only 
 be regulated by each other, but entirely suppressed by some 
 external power; and, finally, that he must not only act 
 rationally, but act from abstract principles of reason instead 
 of regulated emotion.
 
 ( 311 ) 
 
 CHAPTER VIII. 
 
 THE CONSCIENCE. 
 
 I. Theories of Conscience. 
 
 1. Having thus attempted to clear away the ambiguities 
 connected with the word " merit/' and to show that merit 
 means simply the value set upon virtue^ we have still to ask, 
 What is the quality valued ? A man, I have said, can only 
 be virtuous when he obeys the moral law "spontaneously," 
 " unconditionally," or from the intrinsic motives implied in 
 the law itself. A man gives money to the poor ; is he 
 charitable ? The question can only be answered if I know 
 the motive, or, which is the same thing, know how the man 
 will act when circumstances vary. If he acts from direct 
 sympathy, he is charitable ; if from ostentation, he is only 
 acting from a desire of praise. His action conforms externally 
 in this particular case to the rule which would be dictated by 
 charity, but it is not therefore charitable. The truly chari- 
 table man would give wherever he could relieve distress, 
 whether he received praise or failed to receive it ; the osten- 
 tatious man would do whatever gained praise, whether his 
 action did or did not relieve his nei2;hbour. If we make a 
 similar statement in regard to virtuous conduct generally, 
 we must say that a man is truly virtuous only when he acts 
 from an intrinsically virtuous motive — when his action is 
 therefore a guarantee that he will always be virtuous, or 
 when the simple fact that conduct of a certain kind is com- 
 manded by the moral law is a sufficient motive with him for 
 adopting that mode of conduct. 
 
 2. What, then, is the intrinsic motive to virtue ?
 
 312 THE SCIENCE OF ETHICS. 
 
 Various explanations may be adopted according to our 
 moral theory. In any case, it may be admitted that con- 
 duct is only virtuous in so far as it is the manifestation 
 of a truly virtuous character — that is, of a character 
 such that the agent will always act in conformity to the 
 moral law. Obedience to a law is sometimes explained 
 most simply from the analogy of the positive law as im- 
 plying respect for some external authority. The motive, 
 then, common to all virtuous conduct is the motive (what- 
 ever it may be) which prompts obedience to this rule. 
 With theological utilitarians, for example, it is taken to be 
 the fear of supernatural penalties. The difficulty which 
 then occurs is that the motive appears to be extrinsic ; it is 
 fear of a god or devil which makes us moral ; and though 
 we may say that, as a fact, the Deity affixes penalties to 
 conduct of a certain kind, the connection seems to be arbi- 
 trary. We cannot say why this or that particular kind of 
 conduct should be punished or rewarded, and morality is 
 thus explained by explaining it away. This, again, may be 
 avoided by saying, with the ordinary utilitarians, that moralitv 
 means that conduct which produces a maximum of happi- 
 ness. The difficulty presents itself that the test seems to put 
 all kinds of happiness on a level, and therefore to affiard no 
 apparent means of explaining the specific feeling attached 
 to moral conduct, or the difference between the propositions 
 "This is right" and "This is useful." Another answer is 
 therefore adopted, and we are told that the essential motive 
 to morality is not the fear of a deity simply, but the fear of 
 a good deity, or that there is a specific feeling, different 
 from all others, to which we give the name of Conscience, 
 and which supplies its own credentials. The obvious diffi- 
 culty is that such an explanation explains nothing. We 
 cannot tell what a good deity will approve unless we know 
 what is meant by goodness ; and if goodness is explained to 
 us by a faculty of which it is the sole function to declare 
 what is good, we fall into a vicious circle. Another answer 
 is, therefore, to say that the law gives its own authority in
 
 THEORIES OF CONSCIENCE. 313 
 
 the same sense as a logical proposition ; it is ])incling because 
 reasonable; we cannot deny its validity without falling into 
 a contradiction in terms. Conduct is virtuous only when it 
 implies a love of virtue "for its own sake;" and this is 
 interpreted to mean that the rule, when apprehended by 
 the reason, compels our assent by its inherent reasonable- 
 ness, like the rule that things which are equal to a third are 
 equal to each other. But here we assume that conduct can 
 be determined without reference to feeling, and we explain 
 the uniformity of action by an assumption which makes 
 action unintelligible, 
 
 3. How are these various difficulties to be met? The 
 method which resolves morality into reason is, from my 
 point of view, unacceptable upon grounds sufficiently indi- 
 cated in the whole course of my argument. I have, in fact, 
 assumed all along that conduct is determined by feeling, and 
 this, which appears to me to be true, so far as we can push 
 our analysis, is certainly true, I think, within the sphere of 
 science. It is true, that is, that we are determined to act 
 by our desires, appetites, or emotions, even if the metaphy- 
 sician and the outologist can by some means explain a desire 
 as a process of the individual or the universal reason. The 
 statement that I eat because I am hungry expresses a fact 
 with which we are concerned, even if it be not the ultimate 
 form of expression. Any rule, again, which is deduced from 
 the pure reason seems to me to have too wide an application. 
 Formal rationality has no special relation to ethics. No 
 moralist has succeeded in any plausible deduction of the 
 moral code without tacitly introducing an appeal to specific 
 facts properly irrelevant to his doctrine. If general logical 
 rules can be deduced in this way, moral rules can only be 
 deduced by some dexterous sleight of hand. " Ought" and 
 "ought not" (as Hume somewhere says) are suddenly in- 
 serted in the place of " is " and " is not," Nor, again, can 
 such a method decide between different codes of morality so 
 long as they satisfy the general condition of logical coher- 
 ence. " Let every one care for one," is as good a rule in
 
 314 THE SCIENCE OF ETHICS. 
 
 the sense of formal logic as the rule, " Let every one love 
 his neighbour as himself." We can only decide which is 
 the best rule by appealing to facts given by experience, but 
 not deducible from any d priori canon of logic. Not only 
 so, but the law, as generally stated, seems to me to give a 
 wrong rule in many cases. In some senses it tends (as I 
 shall have to observe hereafter) to make the external rule 
 unconditional instead of the internal, and to substitute 
 " Do this " for " Be this." And, finallv, I think that it is 
 at bottom an inadequate statement of the undeniable truth 
 that not only is logical consistence implied in the moral 
 law, but that the emotional development implies at every 
 step a corresponding development of the reason. It may be 
 maintained that as men are, to be perfectly reasonable is 
 also to be perfectly moral. But we cannot omit the con- 
 dition "as men are," or infer morality, without taking into 
 account the specific constitution of men and human society, 
 and we shall have perhaps further to affix a special sense to 
 the words " perfectly reasonable." 
 
 4. The theory, again, of an autonomous or independent 
 conscience, of a faculty which exists as a primitive and ele- 
 mentary instinct, and which is therefore incapable of further 
 analysis, appears to be equally untenable. I agree, indeed, 
 that here too we have an inaccurate statement of a highly 
 important truth. The theory needs the less discussion be- 
 cause it is part of an obsolete form of speculation. Nothing 
 is easier than to make out a list of separate faculties and to 
 call it a psychology. The plan had its negative advantages 
 so far as it was in useful antithesis to an easy-going analysis, 
 which was too quickly satisfied with explanations of complex 
 mental phenomena. At the present day no one will deny 
 the propriety of rigidly cross-examining the claims of anv 
 instinct to be an ultimate factor in the orcjanisation. The 
 difliculties which apply to all such speculations (as, for ex- 
 ample, to the phrenological theory of separate organs) are 
 not diminished in the case of conscience. When we take 
 into account any theory of evolution they arc greatly
 
 THEORIES OF CONSCIENCE. 315 
 
 increased. The conscience appears historically as a develop- 
 ment of simpler instincts. The broad fact must be admitted 
 that "material morality" makes its appearance long before 
 any conscious recognition of a moral law. We may pro- 
 bably trace the germs of the moral instincts down to the 
 associations of animals ; and we may at least assume that 
 men live together and obey certain rules, of which the exist- 
 ing moral law is a continuous development, long before they 
 have any distinct conception of such a law, as distinguished 
 from other rules with which it is originally identified, and 
 from which it is slowly disengaged as civilisation advances. 
 It would require a very forced interpretation of the pro- 
 cess to see in it the introduction of an entirely new factor 
 instead of a gradual development of previously existing 
 sensibilities. 
 
 5. Without appealing to the evolutionist, we need not 
 hesitate to say that the theory of conscience as an elementary 
 faculty is untenable or superfluous. Conscience in any case 
 means the pain felt by the wrongdoer, or rather the sen- 
 sibility implied by that pain. It is exerted when we judge 
 that we have deserved blame, and we deserve blame when 
 we display some moral deficiency. Now a separate instinct 
 — a physical appetite, for example, such as hunger or lust — 
 mav give us pain when its dictates are suppressed by some 
 conflicting impulse. It corresponds to a particular function 
 of the organism ; it is excited by the appropriate stimulus, 
 and is the sole instinct directly interested in a given class of 
 actions. It is supreme within its own province, but it has 
 to struggle because it is part of a complex whole which can 
 only act in one way at once, though accessible to a variety 
 of stimuli. But it is impossible to conceive of the conscience, 
 in accordance with this analogy, as a particular faculty 
 co-ordinate with others, or as possessing a separate province 
 within which alone it is applicable. We may indeed say, 
 in a sense already explained, that some conduct is morally 
 indiflerent, and therefore, if we will, outside the sway of the 
 conscience. That is merely to say that many actions do
 
 3i6 THE SCIENCE OF ETHICS. 
 
 not justify any inference as to the moral character^ and are 
 therefore neither meritorious nor demeritorious in the sense 
 of proving virtue or vice. From the fact that a man is 
 hungry or obeys hunger we can only infer that he has a 
 quality common to the virtuous and the vicious. No moral 
 judgment arises. But this does not mean that the instinct 
 itself has no moral quality; only that the quality cannot be 
 inferred from the particular action. A man's appetite for 
 food has certain determinate relations to his other instincts 
 and affections ; and according as it exerts a certain influence^ 
 the man is either greedy or temperate, and so far vicious 
 or virtuous. Hence a man's intrinsic merit or his virtue is 
 always a function of his character considered as a whole, 
 and is affected by every variation of the elementary instincts, 
 though his merit, in the sense of his manifested or proved 
 virtue, is not inferable from the simple fact that he has 
 some instinct which may be exerted either for or against the 
 moral law. Conscience, in short, always implies a judgment 
 of the whole character, although, as a rule, it considers only 
 some special manifestation, and a defect or fitness in some 
 special relation. If it were an instinct co-ordinate with 
 others, we should then require a further judgment to say 
 whether it was in excess or defect relatively to the whole 
 of which it forms a part ; and this other faculty would have 
 a better claim to be called conscience. 
 
 6. In this sense, it may be remarked, the love of virtue 
 " for its own sake " is sometimes used so as to convey 
 an absurdity. I may love eating for its own sake, because 
 there is a specific appetite which corresponds to a particular 
 stimulus, and I mean simply to say that the act of eating 
 is pleasant because it gratifies this appetite, whilst no other 
 appetite is directly interested. But it is impossible to con- 
 ceive of a love of virtue for its own sake as implying a state 
 of mind in which the conscience was gratified whilst no 
 other instinct was interested. If conduct is such as to give 
 no direct stimulus to anv of the other instincts, to my 
 love or hate, or hope or fear, or my physical appetites or
 
 THEORIES OF CONSCIENCE. 317 
 
 feelings, then the conscience cannot be stimulated. Tt is 
 awakened whenever the agent perceives that his appetites 
 are excessive or his emotions distorted ; but it has nothing 
 to act upon when none of his appetites or emotions are 
 concerned. By the love of virtue for its own sake, we 
 can only mean that the fact that an action is moral, with 
 all the necessary implications of that statement, whatever 
 they may be, is enough to determine to that action ; and not 
 that it has absolutely no other implications, which would, 
 in fact, be to make it inconceivable as conduct at all. 
 
 7. If we still speak of the conscience as a separate faculty, 
 but admit that it is stimulated only when other emotions 
 are stimulated, the proposition seems to evade our grasp. 
 If we consider it as a simple feeling, excited through the 
 perception of other emotions, its law must be dependent 
 upon them. The specific emotion will be produced when 
 certain conditions arise which can be stated independently 
 or in terms of the other instincts. It is a kind of parasi- 
 tical or dependent sensibility, which gives its responses in 
 obedience to the impinging forces. It varies as they vary; 
 and therefore, whether we admit or reject the hypothesis, it 
 cannot give its own law directly. We must inquire else- 
 where what are the conditions under which it produces a 
 painful or a pleasurable emotion. If, on the other hand, we 
 conceive of conscience as a something which judges of 
 an action by some inherent power, we see that we really 
 have to attribute to it reasoning and feeling, a power of 
 estimating an action, comparing it with a rule, of weighing 
 and sympathising with all the various passions concerned; and, 
 in short, we find (as is generally the case with the separate 
 faculties of psychologists) that we are not really speaking of 
 a conscience as one amongst various powers of an individual, 
 but of a conscientious man, who is somehow a spectator of 
 the agent from within, 
 
 8. Without further expanding considerations which are 
 sufficiently obvious, and which refer to a rather anti- 
 quated form of theory, I must come to the question,
 
 3i8 THE SCIENCE OF ETHICS. 
 
 What, then, Is the conscience? If it be neither a distinct 
 emotion nor a purely intellectual judgment, how are we to 
 account for it? I admit, of course, that there is such a 
 feeling — or rather, my theory rests upon the admission that 
 it exists. Conduct is determined by feeling, and virtuous 
 conduct by the particular kind of feeling which we call the 
 conscience. I consider it, indeed, to be not a primary attri- 
 bute of the agent (to borrow Spinoza's language), but a mode 
 of the attributes. It is not the less important. Remorse 
 for crime is clearly amongst the most poignant of emotions. 
 It has driven some men mad ; it has blighted the whole 
 lives of others ; it has in all ages been one great source of 
 the power of the classes who were able to regulate its action 
 and alleviate its pangs. Its existence is as undeniable as the 
 existence of hunger or cold or heat. Yet I feel bound to 
 add that well-meaning moralists are much given to exaggerate 
 the sorrow which it actually excites. In almost every case 
 the pain which we feel for a bad act is complex, and due 
 only in part to our conviction that we have broken the 
 moral law. The sorrow which I feel for having injured a 
 friend is made up in part, but only in part, of the sorrow 
 which I feel for having injured him wrongfully. We may 
 frequently observe how faint is the purely conscientious 
 emotion. I kill a man, let us suppose, by an accident — by 
 carelessly handling a loaded gun, or when really trying to 
 do him a service, as by mistaking him in the dark for a wild 
 beast, whose attack might be fatal to us both. Or, again, I 
 mean to kill him, and take every step in my power, but he 
 escapes by some entirely unforeseen circumstance. I give 
 him what I believe to be poison, and it turns out to be a 
 wholesome drink ; or, vice versa, I poison him meaning to 
 do him good. Now my guilt is obviously proportioned in 
 such cases to my intention : the carelessness is precisely the 
 same whether it produces fatal effects or none; the malice 
 is equal whether the object docs or docs not happen to 
 escape. But it cannot be doubted, I think, that as a rule 
 the remorse felt by most men depends almost entirely upon
 
 THE SENSE OF SHAME. 319 
 
 the event. Men arc made wretched for life when they have 
 killed a friend by pure accident^ even where the carelessness 
 has been most trifling; they speedily acquit themselves for 
 the ffuilt of an action dictated bv a malevolent intention, 
 and completed so far as depended upon them, when by some 
 accident the intention has not been realised. These facts 
 may show, as indeed I think they do show, that men reason 
 very loosely in such matters, and often receive pain or 
 pleasure from mere illusions of the imagination. But this 
 comes to the same thing — namely, that, as a matter of 
 fact, the purely conscientious feeling, the pain resulting 
 from the consciousness of my wickedness, is often very 
 feeble, and that much of what we call by that name is 
 the simpler feeling of regret for the mischief caused, irre- 
 spectively of the wickedness of causing it. Deduct from 
 repentance all that is not purely moral, and we must admit 
 that conscience is not so strong de facto as perhaps it ought 
 to be de jure. Indeed I should say that most men find 
 nothing easier than to suppress its stings, when some imme- 
 diately bad consequence, or the contempt and abhorrence 
 of their neighbours, does not constantly instil the venom. 
 This is as far as possible from proving that an increased 
 strength of conscience is not highly desirable, and that, even 
 in the existing state of things, its influence is not of the last 
 importance. The force of gravitation, as physicists tell us, 
 is intrinsically very feeble compared with many others, but 
 it keeps the planets in their orbits ; and so the sense of duty, 
 faint and flickering as it is in the great mass of men, is suffi- 
 cient to keep the social order from disruption. 
 
 II. The Sense of Shame. 
 
 9. To explain fully what is meant by conscience, or by 
 any other mode of feeling, would require a complete psycho- 
 logy, such as is not at present in existence. It is enough for 
 my purpose if I can show that it is explicable in conformity 
 with the theory already laid down. We have to ask how
 
 320 THE SCIENCE OF ETHICS. 
 
 the conduct forbidden by the moral law comes to excite a 
 specific sentiment^ which we interpret as disapproval when the 
 agent is some one else^ as a pang of conscience when he is our- 
 self ? The difficulty seems to be, that if we interpret it as a 
 simple emotion, it appears to be more or less arbitrary ; if as 
 an intellectual perception, it is difficult to see how it can affect 
 conduct ; whilst the theory which makes conscience an inde- 
 pendent faculty, invested both with intellectual and emotional 
 attributes, seems only to evade the difficulty by help of an 
 unjustifiable assumption. In any case, it seems clear, how- 
 ever, that there must be both an intellectual and an emotional 
 side to the process. The old moralists distinguished between 
 the conscience which declared the law and that which 
 punished a breach of the law. Each process, in fact, seems 
 to imply the other. If I perceive that conduct is wrong 
 because forbidden by law, the sense that I am breaking the 
 law must be painful, in order that the law may have any 
 binding force. If, again, we suppose that every wrong act is 
 attended by a specific feeling, I could construct the law by 
 generalising from my experience of the feeling. This or that 
 act causes the specific pain called remorse; these actions, 
 therefore, are forbidden by the moral law; or the moral law 
 is a statement of the actions which cause remorse. A wrong 
 action would be definable as a remorse-causing action. The 
 function of conscience would be similar to that of hearing 
 in regard to music. The ear decides authoritatively that 
 certain sounds are discordant and others harmonious. The 
 scientific observer notes these cases, and proceeds to determine, 
 with the help of his reason and his other senses, what are the 
 conditions which produce disagreeable sounds. If the con- 
 science were in fact as distinct and separable a faculty as the 
 hearing, we should be able to decide by a similar process what 
 were the conditions which caused painful and pleasurable 
 emotions of this particular class. It might, of course, turn 
 out that the conscience varied indefinitely from one man to 
 another, or that it gave uniform decisions in all men. In 
 the former case, we should again have to inquire what were
 
 THE SENSE OF SHAME. 321 
 
 the laws of its variation, and then to inquire whether, and if 
 so, in what sense, one conscience could be regarded as better 
 than another. We have, however, the prehminary difficulty 
 already stated, that the conscience is not in this way marked 
 off from all other modes of feeling or reasoning, and that 
 the law is given much more distinctly than the feelino; by 
 which it is enforced. 
 
 10. There is, indeed, a sensibility which seems to have 
 as good a claim as any other to be regarded as elementary, 
 and which is clearly concerned in most of our moral judg- 
 ments. The sense of shame appears to me, so far as one 
 can judge by the direct introspective method, to be one of the 
 most distinctive of our feelings, and the presumption seems 
 to be confirmed by its having a distinct physical manifestation 
 of blushing. If we assume that this emotion is really some- 
 thing distinct in itself, we may ask, as we ask in the case of 
 music, what are the conditions under which it arises? It is 
 clearly excited by breaches of the moral law, and especially 
 by detected breaches. A man is ashamed of himself for con- 
 duct which is actually condemned by the moral judgment of 
 his neighbours, so far at least as he sympathises with the 
 general morality, and he is ashamed of conduct which would 
 be condemned if known. In most men, however, and indeed 
 in all but exceptionally sensitive men, the shame is enor- 
 mously increased by the actual condemnation, and, in many 
 cases, seems to be exclusively due to the consciousness of this 
 condemnation. Again, so far as I can guess, it does not appear 
 to me that the sense of shame is proportioned to the moral 
 gravity of the offence. There is a difficulty in speaking posi- 
 tively upon such a matter, because the relative importance of 
 different kinds of offences is very differently estimated by 
 different moralists, and it is hard to sunu'est any assionable 
 measure of the gravity. Some moralists, for example, attach a 
 preponderating importance to veracity and others to chastity ; 
 some think more of the virtue of justice, and others of the 
 virtue of benevolence ; but it is not possible to define in any 
 way the weight assigned to difierent considerations, especially 
 
 X
 
 322 THE SCIENCE OF ETHICS. 
 
 as there is no agreement as to the irrelevancy of particular 
 considerations. Speaking roughly, however, one would say 
 that a sense of shame is more excited bv offences of sensuality 
 than by offences of cruelty. We say, indeed, "What a shame ! " 
 when we hear of a gross act of oppression ; but a man con- 
 victed of tvrannv seems hardly to be liable to shame in the 
 same way as a man convicted of some offence against purity. 
 This tvrant may excite more abhorrence but less disgust; he 
 is not regarded with contempt, and the sense of being con- 
 temptible is peculiarly connected with shame. 
 
 II. Thus, again, we find that the conduct enforced by 
 the sense of shame seems to extend beyond the sphere of 
 morality proper. We may perhaps say, in a general way, 
 that indecency is wrong; but there are a great many acts 
 which we call indecorous, if not actually indecent, to which 
 we should scruple to give so grave a name. A want of com- 
 pliance with the regulations of the society in which we live 
 excites shame, often very acutely and painfully, and yet we 
 should hesitate to describe it as immoral. This, of course, 
 comes in partly under a familiar principle. If it is right to 
 obey a ruler, it is right to obey his particular commands in 
 cases where the conduct would otherwise be indifferent ; and 
 so, in many cases, it is right to do at Rome as Rome does ; 
 it is right to cover parts of the body in England which it 
 may be right to leave bare in Abyssinia. But it seems also 
 to be true that the emotion of shame extends beyond the 
 actions which would be regarded as in any sense wicked. 
 Nobody would call a man immoral for appearing at a 
 dinner-party in a shooting-coat, but a young man would 
 probably feel more ashamed of himself for such an action 
 than for many offences which he would admit to be far 
 graver. Many sensitive people would feel far less shame 
 if detected in a crime than in connnitting an indecent 
 action, even though the breach of decency were involuntary, 
 and therefore not in any sense immoral. So women may 
 cease to be virtuous without ceasing to be modest, and some 
 might possibly prefer a loss of virtue to a loss of modesty.
 
 THE SENSE OF SHAME. 323 
 
 12. The same may be said in cases which arc less ambi- 
 guous. Decency may always be regarded as a kind of minor 
 morality; but the sense of shame is often most powerfully 
 excited in cases where there is no question of morality at 
 all. There are few things which a man remembers with a 
 more hearty and ineffaceable sense of shame than his having 
 — in the vulgar phrase — made a fool of himself. A youth 
 who has tried to say something witty, and whose luckless 
 joke has fallen flat or provoked ridicule, has a memory 
 which will revisit him in dreams, and make him blush in 
 private whenever it recurs to him. Even a grave moralist 
 may often, I suspect, suffer more pain for such slips than is 
 visited upon him for indisputable moral offences. And not 
 only thus, but in many cases the sense of shame operates 
 against the moral code. The boy who thinks it right to 
 say his prayers is ashamed when his practice is detected by 
 unsympathetic observers. In some cases a young man is 
 ashamed of chastity and sobriety when his companions act 
 upon different principles, even though they do not explicitly 
 deny the validity of the moral code. We call this, of course, a 
 false shame, and doubtless it is " false " in the sense that it 
 can only be justified by false reasoning; but it is, whilst it 
 lasts, as real as any other feeling, and therefore, whatever 
 the cause of its being enlisted in the service of the enemy, 
 it is clear that it is not invariably to be found as an ally of 
 conscience. 
 
 13. Now the general principle which appears to be in- 
 volved in this case is simply that a sense of shame is somehow 
 involved in a state of heightened self-consciousness. The 
 bare fact that we are the objects of attention is sufficient to 
 produce the painful seiisation of shyness, and that even when 
 we are the objects of admiring attention. We see ourselves 
 through the eyes of others; we attend to ourselves out of 
 sympathy with their attention to us, and are at once object 
 and subject of our own feelings. When, in addition to this, 
 the attention is of a hostile kind, or the self-consciousness a 
 consciousness of defects; when we are the objects of our
 
 324 THE SCIENCE OF ETHICS. 
 
 own hatred^ contempt^ disgust, or ridicule, we have some of 
 the most disagreeable emotions of which our nature is cap- 
 able. If this be the general law of the feeling, it is intelli- 
 gible that it should be closely connected with, and yet in 
 many respects diverge from, the conscientious feeling. So 
 far as we break the moral law we are acting in opposition 
 to the general sentiment, and therefore incurring disapproval. 
 But the law of shame does not coincide precisely with the 
 moral code; for, in the first place, our feeling in regard to 
 different moral offences seems to differ in quality, and our 
 abhorrence of crueltv, though it may be in some senses 
 stronger, is less keen, and therefore less provocative of the 
 sense of shame, than the spasm of physical disgust which we 
 feel for some kinds of sensuality ; and, in the next place, 
 the code extends beyond the moral code, inasmuch as many 
 things are exquisitely ridiculous which are not immoral ; 
 and, finally, it may even conflict with morality, in that case, 
 at least, in which a kind of spurious moral code is formed by 
 a special section of society, or in cases where the conduct 
 deflects from the average standard, not by falling short, but 
 by exceeding it in a way which seems to imply a tacit 
 reproach to others. 
 
 14. The sense of shame, then, is in some sense implicated 
 in conscientious feeling. It is clearly a part of the emotion 
 which restrains me from wrongdoing. I shrink from detec- 
 tion in shameful conduct, and from conduct which would be 
 shameful if detected. The motive, whether I call it con- 
 scientiousness or not, acts on behalf of any accepted moral 
 code ; but we cannot identify it with the conscience, because 
 it operates fitfully, affects conduct which is not moral, and is 
 sometimes even opposed to morality; as also because it con- 
 demns crimes which are found out much more emphatically 
 than the same crimes when they are not found out. The 
 question therefore arises how such an emotion can supply an 
 intrinsic motive to virtue. Has it not rather a dangerous and 
 disturbino; influence ? So far as I am accessible to shame, shall 
 I not be inclined to over-estimate the judgment of the special
 
 THE SENSE OF SHAME. 325 
 
 class in which T live, to regard decorum as of more importance 
 than real virtue, to make respectability the measure of my 
 conduct, to prefer the infliction of a real injury upon one 
 who cannot complain to showing the least disregard in public 
 to the slightest fancies of a conspicuous ruler, to obey codes 
 which I disapprove in my heart, such as that which enforces 
 duellin(r, and to break throujrh moral laws which are {generally 
 neo-lected, such as that which condemns bribery ; and, above 
 all, shall I not feel a much greater fear of being found out 
 than of beinc- g-uJltv? To this I should replv: Undoubtedlv 
 shame often acts in this undesirable way. It is not identical 
 with conscience, and when badly informed or regulated it 
 may even put obstacles in the way of moral progress, and 
 account for the fitful and apparently arbitrary fluctuations in 
 the moral standard. But I should also find it impossible to 
 say that the shame felt by a sinner is not part of the con- 
 scientious feeling or of the intrinsic sanction of morality. 
 The difficulty is, that if we admit it as part of the conscien- 
 tious feeling, we admit an element of feeling which may vary 
 from one man to another, and which does not account for 
 the specific character of conscience. This requires a little 
 further consideration. 
 
 15. Shame, in the first place, is the name of a certain state 
 of consciousness. I know that I am ashamed as I know that 
 I have the toothache, or as I know that I am amused — by 
 direct feeling. It is so far an ultimate fact, which cannot 
 be explained any more than any other immediate feeling or 
 emotion ; we feci, and therefore we know that wc feel, and 
 no more can be said. But we may, of course, proceed to 
 ask, What are the conditions under which this feeling is 
 generated? It involves, we say, a certain intellectual per- 
 ception ; we have only, then, to say what is this perception 
 in order to assign the law of the feeling. So, for example, it 
 would be easy to suggest that shame arises when we perceive 
 our inferiority to others, and then to infer that shame is the 
 perception of inferiority. This method, however tempting it 
 may be, leads in fact to endless confusion. Nothing is really
 
 326 THE SCIENCE OF ETHICS. 
 
 more difficult than to discover the law of a feeling. Philo- 
 sophers dispute endlessly as to the definitions of such words 
 as " beautiful," " ludicrous/' or " shameful." Although 
 every one may understand what is the emotion which arises 
 when we see a beautiful object, or are moved to laughter by 
 the ludicrous, or to blushing by the shameful, it is at least as 
 difficult to discover the law of these emotions as to discover 
 the law of any physical phenomena, and the only satisfactory 
 method is in either case by a systematic interrogation of 
 experience; we should, in fact, have to discover all the cases 
 in which we do, in fact, feel the emotion in question. When 
 we are ashamed we are ashamed, and we are often ashamed 
 when our preconceived formula seems to be inapplicable. 
 We feel, and in each particular case we know our feeling, or 
 in other words we are conscious ; but we are not by any means 
 directly conscious of the law of the feeling, or able to say in 
 what other cases it will arise. So soon as we begin to gene- 
 ralise we are liable to indefinite error, and we are just as apt 
 to go wrong about the phenomena of our own consciousness 
 as about any other phenomena. The reason is, as I should 
 say, that the perception, for example, of our inferiority is not 
 a bare logical process, but is also a mode of feeling: it involves 
 a comparison of immediate and represented states of feeling; 
 and the compound emotion which arises will vary as they 
 vary, and cannot be deduced from the simple statement of 
 the identity or diversity of two conceptions. In fact, 
 nothing is more common than to find that the philosophical 
 definition hopelessly breaks down, and that we do not feel 
 wdiat our theorists tell us that we shall feel when the case 
 actually occurs. In short, to find the law of any of these 
 modes of feeling requires, not a simple generalisation of 
 a formal loi2;ical statement, but a verifiable and scientific 
 psychology. 
 
 i6. The feeling of shame, then, appears as a positive datum 
 in our theories. It represents a matter of fact which can 
 only be explained and its laws determined through careful 
 observation. In this sense it seems to irive rise to direct and
 
 THE SENSE OF SHAME. 327 
 
 unassailable intuitions. In any given case, at least, I am 
 conscious of feeling or of not feeling shame, and I may, if I 
 please, state this as an intuitive perception, that the condition 
 in which I am placed is shameful. Nor, again, am I much 
 further advanced if I admit that shame arises whenever I am 
 conscious of doimr something disgustins;, ridiculous, or even 
 strongly provocative of the attention of others; for I must 
 ask what are the conditions which make conduct ridiculous, 
 disgusting, and so forth. This is no easier than to say what 
 makes it shameful. And not only so, but it would seem 
 that the relation is reciprocal ; that is to say, that a man in 
 whom a sensibility to shame is keen will find things ludicrous 
 or disgusting which are not so to his less sensitive neighbour; 
 and thus, again, we seem to be falling into a vicious circle, or 
 to have an insoluble problem. 
 
 17. The case, however, so far as we are concerned, may 
 be sufficiently solved. In the first, shame, considered as a 
 mode of feeling, has a certain and conceivably determinable 
 function in the moral constitution. It is not arbitrary except 
 in the sense that it is a datum which is given by experience, 
 but cannot be deduced by any a priori process. It is not 
 arbitrary in the sense of being susceptible of indefinite varia- 
 tion, which seems to be the impression of those who would 
 say that to admit such a feeling as determining the conscience 
 is to make morality a mere "matter of taste." Even taste 
 or fashion is not " arbitrary " in the full sense of the word, 
 for in that sense nothing is arbitrary; it has its laws, though 
 thev are laws consistent with a very wide variation from one 
 individual to another. The emotion is of course limited by 
 the physiological and psychological laws of the individual 
 structure. It may, as I have said, vary very widely in some 
 respects. A degree of nudity which is excusable in one 
 region is disgusting in another ; and, generally speaking, 
 the emotion — as follows from its general nature — is of course 
 very amenable to custom. As a medical student soon loses 
 the sense of horror at surgical operations, a child brought up 
 in a degraded social state feels no instinctive revolt against
 
 328 THE SCIENCE OF ETHICS. 
 
 impurity, dishonesty, falsehood, or cruelty. But the variation, 
 though it is impossible to assign its limits, because we have 
 not the necessary experience for deciding such a problem, is 
 clearly not indefinite, llie instinctive repugnance to sights 
 of blood and mangled flesh may be suppressed, but it could 
 not be generated at pleasure in regard to any other objects. 
 So, as I have suggested, the feeling of shame is specially 
 stimulated where certain passions are concerned, and this 
 implies the existence of physiological laws which cannot be 
 altered so long as the most fundamental properties of human 
 nature remain. We may doubt how far it would be possible, 
 by any course of training, to invert the sense of shame in 
 particular cases, and to make those actions disgraceful which 
 are now honourable or vice versa. But in any case, it is 
 clear that the variation is limited and forced to take place 
 along certain lines, if we may so speak, by the constitution 
 of our nature. It can no more be arbitrarily changed at the 
 will of some other person than a legislator could order men 
 to be seasick on land and comfortable at sea, or force our 
 gorge to rise at wholesome food and make carrion appetising. 
 The organism is not the less subject to precise laws because it 
 is capable of responding in a vast variety of ways to different 
 stimuli. 
 
 1 8. Undoubtedly, however, the possible variation is great 
 enough to conform to the observed variability of moral 
 codes. If men are, in fact, shameless in certain respects in 
 one country and ashamed in another ; if the standards of 
 courage, purity, truthfulness, and benevolence vary widely, 
 and without any corresponding variation in the innate 
 powers of the individual, it follows of course that the deter- 
 mining condition must be the social medium. What is not 
 explained by the individual organism must be explained by 
 the social organism. If modern Englishmen are disgusted 
 by conduct which did not disgust Socrates, we must explain 
 the difference by the whole social development which has 
 taken place in the interval ; and, if my theory of morality 
 be correct, this shows in what sense variation is possible.
 
 THE SENSE OF SHAME. 329 
 
 The existence of a social order or a certain stage of develop- 
 ment implies a corresponding development in the individual, 
 considered as a constituent part of the society. This develop- 
 ment implies on the one hand the attainment of a certain 
 moral standard, which, again, implies obedience to and 
 respect for the primary moral laws; and, on the other, it 
 implies that the whole character of the individual, including 
 the sense of shame, which is one of the most powerful factors 
 in determining his conduct, must be so modified as to imply 
 an acceptance of the standard. The fixed element which 
 causes the sense of shame to develop, so to speak, along 
 certain lines, is simply the social necessity. I have tried to 
 show in what sense morality is essential to social vitality, 
 and in that sense and within these limits the sense of shame 
 must be so moulded as to maintain the moral standard, 
 unless the society is not to deteriorate, and to raise it if the 
 society is to make progress. 
 
 19. This is enough to show that there are narrow though 
 not easily assignable limits to a possible variation of the 
 instinctive feeling. Social development implies moral deve- 
 lopment, but it of course implies much more. It implies the 
 development of a certain type of character, which includes 
 as essential certain moral qualities. It implies, that is, the 
 growth of mutual confidence, peacefulness, the restraint of 
 antisocial passions, and so forth. But though a certain 
 moral conformation is implied, and is essential to the social 
 efficiency of the agent, his possession of these qualities does 
 not define his character. We may have an indefinite variety 
 of talents and sensibilities not directly concerned in this 
 morality. Consistently with moral excellence, he mav be 
 an artist, a philosopher or a poet, a statesman or an artisan, 
 a priest or a lawyer, and in each case his character and his 
 intellect will be stimulated in different ways and have differ- 
 ent excellences. The morality is an attribute of the core or 
 nucleus of character, separable only by an abstraction, not 
 as a concrete entity : it defines the qualities necessary in 
 every relation ; it imposes certain limits upon every instinct,
 
 330 THE SCIENCE OF ETHICS. 
 
 and is itself strengthened or weakened by the reaction of 
 every part of our natures. There is no reason therefore to 
 suppose that it has any faculty peculiar to itself. A given 
 instinctj such as the sense of shame^ may be so developed 
 that the whole character conforms to the moral type; but it 
 may be called into play in many cases where morality is not 
 immediately concerned^ and every other instinct must also 
 have a part along with it in enforcing the moral law. The 
 qualities implied by morality do not correspond to any 
 separate instinct any more than moral actions constitute a 
 separate class amongst other qualities. They only define cer- 
 tain modes of reaction of the whole organism in particular 
 relations. 
 
 20. It follows that the dictates of an instinct such as 
 shame are not arbitrary in the sense sometimes assumed. 
 If, in fact, some conduct were condemned simply as offensive 
 to a separate instinct which might or might not be present, 
 there would be a difficulty in obtaining a fixed rule. This 
 disgusts me and that disgusts you; who is to say which is in 
 the right? To this it may be answered that he is right who 
 is disgusted by really mischievous things. The answer is 
 satisfactory as far as it goes, but it leaves a difficulty un- 
 solved. If the instinct which forbids certain conduct, say a 
 sense of decency, were something separate and unsociable, 
 we must ask whether it has a rio;ht to be considered in the 
 matter. Indecent conduct is no doubt mischievous so far 
 as it gives pain, but we can get rid of the sense of decency, 
 and then the conduct will not give that kind of pain. The 
 objection, then, to the indecent action is extrinsic; it is not 
 bad simply because indecent, but because it has some other 
 ill effect; and if the sense of decency could be abolished 
 without any other alteration in character, we might propose 
 to get rid of it as, on the whole, a troublesome and illusory 
 quality. We should then object only to those indecent 
 actions which appeared to be mischievous upon other grounds, 
 and judge directly by their utility as so measured, without 
 considering their superfluous or supplementary inconvenience.
 
 THE SENSE OF SHAME. 331 
 
 When, however, we consider the whole problem, this is 
 obviously inadequate. Undoubtedly the evil of indecent 
 actions is not confined to the simple pain inflicted upon that 
 particular sensibility. The sense of decency is closely con- 
 nected with the virtue of purity, and it is desirable to main- 
 tain it as a kind of outpost which defends a virtue most 
 essential to society. But this, as we have seen, does not 
 exhaust the utility of the instinct, when we consider it not as 
 a separate mode of feeling, but as a sensibility implied in the 
 whole development of character. A man, that is, who is not 
 endowed with a capacity for such feeling, must be made 
 throutrhout of coarser fibre. A society in which some code 
 of decency is not developed must be entirely wanting in a 
 refinement and delicacy which is an essential symptom, and 
 reciprocally an essential condition, of any but the lowest stages 
 of barbarism. The penal code may vary widely; some code 
 is necessary if men are to be more than a herd of brutes. 
 We must, in short, as I have previously said, consider the 
 whole oriranisation of man in society. We cannot measure 
 the value of a sense of decency by simply considering a 
 particular set of bad consequences resulting from indecent 
 actions other than the shock to decency, but the whole 
 difference between a state of society which possesses and one 
 which does not possess such a code. That difference would 
 probably turn out to be the difference between a thoroughly 
 brutalised and a really refined and intelligent social state, in 
 which the moral qualities were in harmony with the general 
 advance in the scale of civilisation. To ask what is the use 
 of a sense of decency is to ask how far any tolerable social 
 order could be constructed without stimulating the various 
 emotive sensibilities which are concerned in generating that 
 sense and forming a corresponding code, and no mode of 
 inquiry can be satisfactory which omits a full consideration 
 of all the implied conditions.
 
 332 THE SCIENCE OF ETHICS. 
 
 III. Esthetic Judgments. 
 
 2T. The value, therefore, of any particular instinct depends 
 upon its place in the whole character, and is thus not arbi- 
 trary in the sense in which it would be arbitrary if we could 
 speak intelligibly of men as made up of different faculties, 
 any one of which could be removed without implying a 
 chano-e of the whole organisation. The question remains 
 how it comes to pass that the conscientious feeling, which is 
 thus a function of the whole character and not a specific 
 faculty, comes to have so distinctive a quality as is at least 
 frequently attributed to it? This difficulty, as I may observe 
 in the first place, is not peculiar to ethical judgments. When 
 we say, " This is beautiful or ugly," we refer to a set of feel- 
 ings which, in the judgment of common sense, are quite as 
 distinctive in respect of quality as those to which we refer 
 when we say, "This is right or wrong," if by those words 
 we mean agreeable or offensive to the moral feeling. In the 
 latter case, it is true, there is also a reference to a fixed rule, 
 which is supposed to be the same for every one, whilst the 
 aesthetic judgment has little or no reference to any such rule. 
 This, however, only means that the laws of the aesthetic 
 feelings are much more dependent upon the idiosyncrasies of 
 the individual than the laws of the ethical feelings. I will 
 consider directly what is implied in this. It is not incon- 
 sistent with the statement, which seems in any case to be 
 true, that we know as well what is meant by the sense of the 
 beautiful as by the sense of the morally right, just as we know 
 what is meant by the sensual appetite for food, although one 
 man likes one food and one another. The class to which the 
 feelings belong is as distinctly marked off from other classes, 
 thou2:h it is more variable in its dictates. 
 
 23. Now the aesthetic feelings, whatever else they may 
 be, are not a set of separate emotions, distinct from others 
 as one physical appetite is distinct from another. On the 
 contrary, it is the prerogative of art to call into play every
 
 ESTHETIC JUDGMENTS. 333 
 
 possible variety of emotion. We say that a woman, a sun- 
 set, a picture, a tunc, or a simple colour or form, is beautiful. 
 We mean that some of our passions are agreeably stimulated, 
 not that a set of passions distinct from those which affect us 
 in other relations is stimulated. In one case, it is the direct 
 pleasure of the senses; in others, moods which call into play 
 an indefinite variety of intellectual processes; as in reading 
 a fine poem, our pleasure may be derived from the tender 
 melancholy of old associations or from thoughts of the sorrows 
 and joys of the whole human race. To decide what consti- 
 tutes the aesthetic mood would be to enter upon a very thorny 
 problem, for which I am not competent, and which does not 
 appear to be relevant to the present question. One point, 
 however, is sufBcient for my purpose. We may assume, that 
 is, the truth of the general statement that the end of all truly 
 ffisthetic indulgence is the immediate pleasure ; and this state- 
 ment would be sufficient if it could be made quite accurate 
 for my present purpose. 
 
 23. I find, indeed, a certain difficulty in stating this 
 criterion to my satisfaction, and the difficulty arises, I think, 
 from the fact that the distinction is not so absolute as is 
 frequently assumed. All conduct whatever is determined 
 in the sense already explained by the pleasantness or painful- 
 ness of the corresponding feeling; and conduct in this sense 
 must include both aesthetic and other modes of activity. The 
 proposition, for example, must be equally true whether I am 
 composing or listening to music, or painting pictures, or fight- 
 ing, or labouring in the fields. It must be equally true, again, 
 that the immediate pleasure is the sole determining condition, 
 whether I am drinking a glass of wine or listening to a song; 
 and we should be stretching the definition of aesthetic enjoy- 
 ment too widelv if we used it to include the former kind of 
 pleasure. The distinction seems to correspond to the distinc- 
 tion between play and work. Certain activities are con- 
 sciously intended to procure some future pleasure and to 
 modify the conditions of my existence ; in others, I have no 
 thouo-hts of anything beyond the present, and I find pleasure
 
 334 THE SCIENCE OF ETHICS. 
 
 in the simple utterance of my emotions and their direct 
 stimulation without reference to future consequences. If I 
 hunt to get food, I am at work ; if I hunt without caring for 
 any ulterior results, I am at play. When I read a storv or 
 see a play, the same emotions are stimulated which would be 
 affected if I were reading history or witnessing a scene in 
 real life : the sense of the unreality of the objects presented 
 to my eyes or my imagination converts my feelings into the 
 aesthetic state as it deprives them of the normal influence 
 upon conduct. In Dryden's Ode, Alexander passes through 
 all the emotions in turn of pride, ambition, love, pitv, and 
 anger, till one happens to coincide with an opportunity for 
 action, and the emotional force which had previously expended 
 itself in vague excitement suddenly discharges along a fixed 
 channel and propels him to energetic action. In one case, it 
 seems, we are, so to speak, merely discharo^ino- feelino- or blow- 
 ing off steam ; in the other, applying it to the purpose of 
 doing mechanical work. 
 
 24. The difference, then, between aesthetic and other 
 pleasures depends upon the form of the gratification, not 
 upon the instincts gratified. The poet or the artist appeals 
 to my love, hate, or sympathy as much as the preacher or 
 the philosopher, though he does not direct them to any 
 specific practical end ; and hence we have the variability of 
 the aesthetic canons of taste which seems to distincruish them 
 from the rule of conscience. Whatever gives pleasure may 
 give aesthetic pleasure, and there seems to be in some direc- 
 tions hardly any limits to the divergence of individual tastes. 
 If I like this colour or taste and you like that, we are equally 
 pleased, and there is no criterion for deciding our difference. 
 So treacherous and fluctuatino- a mode of feelino- should there- 
 fore, it is said, be excluded from moral judgments if morality 
 is not to be regarded as a mere fashion. The answer, so far 
 as the moralist is concerned, is not difficult. However vari- 
 able the taste may be, it is not — for nothing is — absolutely 
 variable. Even in the simple case of a direct pleasure of the 
 senses — a love of bright colours, for example — the scientific
 
 ESTHETIC JUDGMENTS. 335 
 
 observer may show that certaui sensibiHties are essentially 
 connected with certain organic conditions, and thus that 
 some imply healthy and others morbid conditions. The 
 sense of the beautiful, again, implies the presence of an 
 intellectual element, and the development of the intellect 
 thus imposes certain conditions on the development of the 
 taste. So, for example, the pleasure which we derive from 
 the sight of a fine figure or a perfect statue implies a power 
 of judging of certain relations of form, a capacity for recog- 
 nising that a given conformation corresponds to the best 
 possible combination of strength and activity. It is, indeed, 
 another question to ask why such a combination should give 
 us pleasure; and we need not attempt to decide whether the 
 recognition is explicit or implicit, whether we feel or reason, 
 or judge by an instinct, or in any degree by a recognised 
 formula. In any case the taste must conform to the facts, 
 and will be more or less the same whenever the conditions of 
 strength and activity are the same, and a perception that they 
 are fulfilled gives pleasure; and this is equally true where 
 the perception which gives pleasure involves some condition 
 of the intellectual emotions. 
 
 25. This, I think, explains the sense in which we must 
 admit that the conscience includes an aesthetic element, and 
 in which we may properly speak of a moral sense. Any 
 pleasurable emotion whatever may be involved in what is 
 called aesthetic pleasure. We derive pleasure, therefore, from 
 a vast variety of perceptions which have no assignable rela- 
 tion to moral feeling. Anything which stimulates the emo- 
 tions agreeably may give rise to an aesthetic pleasure, the 
 only conditions being that the mode of feeling must be 
 agreeable, and must not be expended on what we call prac- 
 tical effort. One kind of aesthetic pleasure, therefore, is that 
 which we derive from the contemplation of certain characters 
 or the play of certain emotions in our fellows. Moral 
 approval includes the pleasure derived from the contemplation 
 of virtuous character, and may therefore give rise to an 
 aesthetic pleasure. If we admire heroism, unselfishness.
 
 33^ ' THE SCIENCE OF ETHICS. 
 
 simplicity, and other moral qualities iu real life, the artist 
 may appeal to that emotion as well as to any other: he may 
 set before us imaginary ideals, from the contemplation of 
 which we may derive a very keen as well as very elevating 
 pleasure, or he may provide us with a means of uttering the 
 emotions which are habitually stimulated by such contem- 
 plation^ or by the directly moral emotions themselves. So 
 far as there is any emotional element in our approval of 
 moral conduct, it may take the aesthetic form. According 
 to me, it would be impossible that morality itself should be 
 maintained if it did not excite these pleasurable emotions. 
 The conception of merit implies, as I have argued, the social 
 pressure which consists in a general approval and admiration 
 of certain qualities and a disapproval and contempt of their 
 opposites. The emotion may pass from one phase to the 
 other according to circumstances. It is in the aesthetic phase 
 when we simply enjoy the contemplation of some beautiful 
 moral tvpe set before us in history or fiction, and passes^into 
 the practical phase as soon as it begins to have a definite 
 relation to the conduct of our lives. If^ for example, I admire 
 the simplicity and tenderness of the " Vicar of Wakefield " 
 whilst I am reading Goldsmith's masterpiece, I am simply in 
 the aesthetic frame of mind. If an analogous case presents 
 itself at the moment, the emotion kindled by the artist may 
 prompt me to charitable and so far moral action. The strains 
 of military music may produce aii ardent excitement which 
 expends itself in mere feeling, or may stimulate a regiment to 
 mount a dangerous breach. 
 
 26. But admitting the existence of the feeling, how far 
 does it bring any guarantee of conformity to the moral law ? 
 How far are the dictates of the moral sense infallible or 
 authoritative? The bare fact that the contemplation of 
 certain emotions or types of character is pleasurable does not 
 prove that they are morally good or bad, for, in the first 
 place, no moral cjuestion mav be raised. I may simply be 
 sympathising with emotions common to the good and the 
 bad ; but if a moral sentiment be involved, the bare fact that
 
 mSTIIETIC JUDGMENTS. 337 
 
 you and I approve is clearly not conclusive as to its Tightness. 
 Nor does it always follow that a conviction of the moral 
 excellence of certain character or conduct will suffice to make 
 the contemplation agreeable. This, indeed, is obvious from 
 a consideration of the way in which we suppose the moral 
 law to arise. When I say that I know an action to be right, 
 I mav mean simplv that it is in conformity with a law which, 
 for whatever reason, I respect. So long as the moral law is 
 stated in the external form, and is obeved from some extrinsic 
 motive, such as respect for the authority of the supposed 
 legislator, I may approve obedience without feeling any 
 vivid and spontaneous pleasure. I admit this or that to be 
 right — that is, commanded — and on the whole, therefore, I 
 wish to see it done; but I do not feel that spontaneous and 
 instinctive sympathy which is necessary to generate aesthetic 
 pleasure. It is only in so far as the moral law has become 
 the law of my character, and expresses the way in which my 
 emotions act previously to reflection upon any abstract prin- 
 ciple, that I can be said to have a moral sense. Undoubtedly 
 this moral instinct may be exceedingly powerful and sensitive 
 in some characters, and give indications more delicate than 
 anv process of deliberate calculation. A man of fine moral 
 sensibility mav perceive emotional discords as a man of fine 
 musical ear mav be sensitive to differences of sound too slight 
 to be measured by scientific observers. So we may in certain 
 cases accept the judgment of a man who is remarkable for a 
 keen sense of honour. The fact that certain conduct does 
 not shock him is evidence that it cannot be dishonourable : 
 we hold that his instinctive perceptions supply a more delicate 
 test than can be embodied in any cut-and-dried formula. 
 
 37. We may admit, then, that the thorough assimilation 
 of the moral law implies the growth of a sensibility which 
 may be called aesthetic — a capacity for receiving delight from 
 the bare contemplation of high moral qualities, abstractedly 
 from any special advantages expected from them, or from 
 any extrinsic consequences. We may agree that to some 
 extent this sensibility must be developed in every truly virtu-
 
 338 THE SCIENCE OF ETHICS. 
 
 ous person^ that is, in any person in whom the instincts 
 which dictate obedience to the moral law have become de- 
 finitely organised; and, further, that this aesthetic pleasure 
 implies a corresponding sentiment as governing practice, and 
 that in the more finely constituted natures it implies a deli- 
 cacy of discrimination beyond that which can be formulated 
 in any of the accepted moral commonplaces. Unless, in- 
 deed, the moral sense is cultivated up to this pitch, so that 
 we take a spontaneous delight in the spectacle of heroic or 
 philanthropic energy, we can hardly say that a society has 
 become tolerably moralised. But we still have to admit 
 that there is an apparently arbitrary or indeterminate ele- 
 ment in such feeling. It varies from one man to another, 
 and it is difficult to suggest any test for distinguishing'be- 
 tween the healthy and the morbid sensibility ; between that 
 exquisite perception which outruns the more tangible tests 
 and the perverted perception which is biassed by some in- 
 dividual peculiarity. A man of a high sense of honour, for 
 example, may be Quixotic ; he may attach undue value to 
 certain considerations, and sanction a retro2;rade instead of 
 an advanced moral code. 
 
 28. What then is the fixed element? The instinctive 
 feeling always includes, we may say, an intellectual judg- 
 ment. But what is this implicit judgment? It is prima 
 facie a judgment that this or that conduct gives pleasure. 
 But as different things may please different people, two diver- 
 gent judgments may be equally right. It is still, after all, a 
 question of taste, and therefore comes within the proverbial 
 exclusion from a possibility of logical decision. It is clear, 
 however, that even a question of taste may often admit of 
 being brought to an issue of facts. So, in the analogy 
 already suggested, two sculptors may prefer two different 
 ideals, yet both of them may agree in preferring that form 
 which represents, say, the most perfect combination of 
 strength and agility. There is therefore a clear objective 
 test. One of the two forms represented will in fact perform 
 a given feat with the least effort, or with a given effort pro-
 
 THE CONSCIENCE. 339 
 
 dace the greatest results. The process of educating the 
 taste is virtually a process of learning to solve such problems 
 instinctively, and we may say that the taste is best which 
 solves them most accurately. The question is therefore a 
 question of fact, though it may be one of such complexity 
 that it is impossible to obtain more than an approximate 
 solution. 
 
 29. Now, if we apply this analogy, we have to say what 
 is the problem presented to the moralist. Every moral 
 judgment, as I have argued, is an implicit (if not an explicit) 
 approval of a certain type of character. It includes, there- 
 fore, an assertion that the highest type includes certain 
 qualities of character, which, of course, imply corresponding 
 modes of conduct. The highest type, again, must, according 
 to our theory, be that which is on the whole best fitted for 
 the conditions of social welfare. The problem is just as 
 precise as the problem which physical conformation is best 
 adapted to satisfy the conditions of health, strength, and 
 activity. The fact, so far as it is a fact, that we cannot 
 obtain an accurate solution does not prove that there is no 
 such solution to be found, but only that the solution requires 
 longer observation and a more elaborate set of experiments 
 before we can hit upon it. The experiment, in fact, is that 
 which is being always carried on by the collective experience 
 of the race; and though we have established beyond a 
 possibility of doubt certain general principles which are the 
 basis of the accepted moral code, there is still a considerable 
 margin of uncertainty in details. Upon this assumption, 
 the problem for the moralist is analogous to the problem for 
 the artist ; each is virtually trying to discover a certain type 
 which has definite conditions to satisfy — briefly spcakino-, 
 that of bodily vigour in one case and of social vitality in the 
 other. 
 
 IV, The Consciejice. 
 
 30. But it must also be admitted that, although I have 
 argued that this gives a correct definition of morality, or a
 
 340 THE SCIENCE OF ETHICS. 
 
 description of the function actually discharged by the moral 
 sentiments, I have not maintained^ nor does it seem to be 
 in fact maintainable, that it is a description of the explicit 
 aim of moral conduct. I have considered what is.the ground 
 or cause of morality, what is the explanation of its existence 
 to an independent observer ; but I have not considered what 
 is the reason which is consciously admitted by moral agents. 
 The society is moral up to a certain standard " because " the 
 society could not reach a certain stage of development without 
 being so far moral. But this does not imply that the end of 
 every man, so far as moral, is the elevation of society or 
 the preservation of its vigour. On the contrary, it may be 
 true that though the moral man contributes to that end, he 
 may never think about it. So far as the units are moral, the 
 social organism which they constitute is healthy ; but a man's 
 reason for being moral need not include any reference to that 
 fact, nor need the reasons always be identical. A man may 
 act from a given instinct without asking how he comes to 
 have that instinct, nor whether it is original or derivative, 
 permanent or destructible, useful in any sense to himself or to 
 others. Indeed, the plausibility of rival systems of ethics is 
 partly due to the fact that men may be moral, that is, may 
 obey certain external rules of conduct, from different motives ; 
 with some the motive may be fear of hell, an association 
 with certain accidental circumstances, or some instinct which 
 has been generated thev know not how; or, again, the im- 
 mediate motive may be a desire of consistency, or direct 
 sympathy with their surrounding neighbours; and although 
 it is true that, in each of these cases, there .corresponds a 
 certain type of character, and the difference might be revealed 
 by the occurrence of special conditions, still, under the average 
 circumstances of life, the resulting rule of conduct may be 
 approximately the same. How far one motive or system of 
 motives is more strictly entitled to be called conscientious than 
 another is a question to be considered directly; and thus 
 we have to ask how the different modes of reorardinir morality 
 can all lead to the same result, or how men should obtain the
 
 THE CONSCIENCE. 341 
 
 answer to one problem when they are apparently aiming at 
 another ? 
 
 31. The answer, as I take it, follows from the facts alreadv 
 stated. Tq any particular association of human beings there 
 must correspond a certain corporate sentiment. A state 
 implies the existence of feelings of loyalty or patriotism; a 
 church, a certain religious sentiment which carries with it 
 attachment to the ecclesiastical order; an armv implies dis- 
 cipline ; a college, a school, and a club imply certain sentiments 
 of mutual goodwill and readiness to accejit the conditions of 
 common action for the purposes of the particular association. 
 Whatever ma.y be the cause of the existing sentiments, what- 
 ever may be the end of the association or the function of the 
 social organ, the corporate sentiment which holds it together 
 must always imply conformity to certain rules necessary to 
 the welfare of the body ; and as the sentiment is vigorous or 
 feeble, the body will so far tend to flourish or decay. More- 
 over, the sentiment which springs up and binds men together 
 implies something which it is often very hard to distinguish 
 from a moral sentiment. The spirit of loyalty to some special 
 association sometimes conflicts with the ordinary moral law, 
 and is often strong; enoutrh to overcome it. A thief is 
 bound to his gang by a sentiment which we call immoral, 
 because it implies conduct condemned by the prevailing moral 
 code ; but so far as it implies a genuine identification of him- 
 self with the gang, and a sacrifice of his private interests to 
 those of the community, it is rather a kind of spurious or 
 class morality, implying obedience to a rival moral code. So 
 the member of a particular class acquires a sense of oblifration 
 to the class even where its interests conflict with those of the 
 organism at large. The proverbial noblesse oblige may imply 
 a more refined sense of honour, but it also frequently implies 
 a regard for the privileges of the class abstractedly from the 
 question of their utility to the whole communitv. When we 
 take the wider associations, the state, for example, it becomes 
 diflicult to distinguish between the sphere of conscience and 
 that of the more specific sentiment. Patriotism, we sav, is
 
 342 THE SCIENCE OF ETHICS. 
 
 a dutv; and of course It is a duty in every man to promote 
 the welfare of the nation to which he belongs. But there 
 are cases — unfortunately of not infrequent occurrence — in 
 which the duty of patriotism seems to diverge from the duty 
 in a wider sense. Even intelligent people are not ashamed 
 to limit their obligations by the supposed interests of their 
 country — to declare that it is a man's " duty" to be for his 
 country "right or wrong" — to promote the happiness of 
 Englishmen even at the expense of the general welfare of the 
 world, or even to extend the British empire at the expense of 
 the happiness of all its constituent members. The sense of 
 duty or obligation to any class to which we may happen to 
 belong seems to have very much the same quality, so to 
 speak, as the moral sense, though we only use the word " duty " 
 when we are not considering the class, or when it is so large 
 that our intellectual horizon does not practically extend 
 beyond it. 
 
 32. Now, upon my showing, the sense of duty or the purely 
 moral oblio-ation has the same relation to the " social tissue " 
 as the various special sentiments corresponding to each organ 
 or association have to the body to which they correspond. 
 I am patriotic so far as Englishman, and moral so far as 
 human being, or rather as a constituent member of a certain 
 social order. The difficulty, then, with which we are now con- 
 cerned is simply that in the one case there is, and in the other 
 there is not, a certain definite and rounded body which may 
 serve as the concrete object of my devotion. It would be 
 hopeless to attempt an analysis of all the sentiments which 
 go to form patriotism, and it is enough to say that at times 
 they include the most unselfish emotions and the widest 
 intellectual culture, as in the case of our best statesmen, 
 whilst sometimes the sentiment may exhibit itself in the 
 grotesque form of the aversion of a clown for a race of which 
 he only knows that it uses strange words, and includes such 
 a sentiment in its composition many unamiable and selfish 
 feelings. And yet there is so much similarity that the clown 
 and the statesman may be equally ready to die for the flag, or
 
 THE CONSCIENCE. 343 
 
 uphold what is generally supposed to be the honour of their 
 nation. Now if the nation were something formless and 
 indefinite, something which had no definite internal symbols 
 and could issue no definite orders, how could any devotion 
 be either generated or displayed ? And this seems to be the 
 case with that extremely indefinite entity the " social tissue," 
 which can neither give rules nor be regarded as an object of 
 devotion. 
 
 ^^. The answer seems to bs twofold. In the first place, 
 every possible form of association implies some moral training. 
 So far as a man is a member of any larger organisation, the 
 qualities which fit him for social action are stimulated and 
 disciplined. As a citizen, as a member of a church, and in 
 everv other capacity in life, he learns subordination, self- 
 restraint, consideration for others, and so forth. The quality 
 of every particular organ depends upon the qualities of the 
 tissue from which it is constituted, and have invariably an 
 influence upon it. The character of each unit is affected 
 in some degree by his position as a member of the larger 
 body, and the modification thus impressed affects him in all 
 other capacities. If I learn obedience as a soldier or self- 
 restraint as a member of a club, I shall be so far adapted 
 to display the same characteristics in any other relation in 
 which I may be placed. And further, as a general rule, the 
 conditions of vitality of the society at large must be also 
 conditions for the vigour of any particular association formed 
 from it. Anvthino: which contributes to a man's health in 
 general is also useful in any particular employment. There 
 is not one set of sanitary rules for a peasant and another for 
 an artisan, though each occupation has its special conditions, 
 which may be regarded as modifications of the more general. 
 This is equally true of the conditions of social health. So 
 far as a man is morally better, he is better fitted for the 
 particular duties imposed upon him by his special position 
 in the general organisation. In fact, if an association were 
 so constituted as to require a set of rules different from the 
 general rules, or, in other words, if immorality were a condi-
 
 344 THE SCIENCE OF ETHICS. 
 
 tion of membership^ it would correspond to a morbid growth 
 instead of a normal organ. The society would so far be in 
 a condition of decay, and the elimination of the association 
 in question would be necessary for its vitality. The general 
 condition, in short, of social vigour implies an approximate 
 identity of interests between the whole and every constituent 
 part. It is, of course, true indeed that the special interests 
 of a particular part of the community may conflict with 
 those of the whole, and it is the chief duty of a statesman 
 to guard against such deviations on pain of revolution or 
 social decay. The devotion of the soldier may tell in favour 
 of despotism, and patriotic spirit may lead to the most 
 atrocious conduct towards outsiders. In such cases, of 
 course, we have imperfect or one-sided moral systems, but 
 they are still moral. That is to say, qualities are stimulated 
 which are so far moral as they imply an identity between 
 the individual and some larger organism, although the 
 interests of that organism diverge from those of society at 
 large. In order that a man may be an effective member of 
 any society, he must have certain moral qualifications, and 
 what is required is an enlargement of his perceptions which 
 shall force him to take into account wider considerations of 
 a similar kind ; to sympathise with men even though they 
 use different symbols for communication, and to respect other 
 claims upon his loyalty than those which are associated with 
 military leadership. The morality impressed upon a man is 
 not always, perhaps it is never, absolutely right — that is, it is 
 never an absolutely correct impression of the ideal qualities; 
 but it must be almost always an approximate expression, 
 and capable generally of reconciliation by a simple widen- 
 ing of the field of view. 
 
 34. In the next place, the true school of morality is the 
 family, which represents a mode of association altogether 
 closer, more intimate than any other, and in which there is 
 not the same possibility of deviation from the moral code. 
 The moral quality of every man is determined to a very great 
 extent in his infancy. We learn our morality, not from
 
 THE CONSCIENCE. 345 
 
 books and lessons, but in the nursery, or at our mother's 
 knee, or from intercourse with our brothers and sisters. 
 There it is that the core of character is fixed, and that the 
 deepest organic quaHties are permanently stamped upon us. 
 The essential part of our education, we may say, is that 
 which we receive in the stage of absolute dependence upon 
 others. The adult, it is true, acquires passions of which 
 the infant was incapable, but it is still through the family rela- 
 tions that they are principally disciplined. The purity of the 
 domestic relations is the essential guarantee for a most im- 
 portant class of virtues, and the family is moulded and deter- 
 mined by the discipline to which thev are subject. The 
 sympathies, again, receive their chief stimulus through the 
 domestic relations, and it is in the sphere of the family that 
 we normally find a degree of altruism which is scarcely to be 
 expected elsewhere. The love of a mother is the typical and 
 central virtue of which all others seem to be faint reflections. 
 It is, as I have said, in the family that the binding forces 
 which hold society together come as it were to the surface, 
 and are directly visible without admixture of any ties of an 
 inferior order of intimacy ; and therefore the family is the 
 main organ of morality. 
 
 ^^. This familiar fact gives the historical explanation of 
 the moral sense in so far as it shows how the development of 
 the conscience naturally takes place. It shows further, I 
 think, what is to be regarded as the true form of moralitv. 
 The child has become a moral agent as it has learned self- 
 restraint, sympathy, truthfulness in the special concrete case. 
 It knows nothing whatever of society at large, or of the 
 formulas sanctioned by moral philosophers. It is a good 
 child because it loves its parents and has imbibed certain 
 organic instincts around which all later developments of 
 feeling group themselves spontaneously. The assumed end 
 of moral conduct has never been presented to it, and cannot 
 supply it with a motive. It has never thought of the greatest 
 happiness of the greatest number, nor is it capable of a priori 
 deductions or categorical imperatives. If the morality taught
 
 346 THE SCIENCE OF ETHICS. 
 
 is avowedly based upon some theological dogma, the child 
 can only conceive of the Deity as an invisible father, or per- 
 haps as an invisible being who always approves the paternal 
 commands. What jxives the real force to the moral teachino- 
 which it may receive is the stimulus given to its affections 
 by its actual intercourse with the little microcosm which 
 bounds its intellectual vision. In what sense, then, is the 
 morality of the child determined by conditions of social 
 welfare? Obviously not by any conscious appreciation of 
 those conditions. As obviously the conditions may be called 
 operative in this sense, that as the social tissue is composed 
 of human beings, whose most intimate bond is the family 
 relation, the conditions of social welfare are necessarily coin- 
 cident with the general conditions of family welfare. So far 
 as men are better husbands, fathers, and sons, and women 
 better wives, mothers, and daughters, society is in a more 
 wholesome condition. Every variation in the strength and 
 purity of the family relations implies a corresponding variation 
 in the stability of the society at large. As the cohesion of a 
 whole tissue depends upon the cohesion of the compound 
 molecules of which it is built up, so the society depends upon 
 the family ; whatever qualities are useful in one relation are 
 useful in the other. The good child becomes the good man 
 by simply widening its sphere of svmpathy, exerting the same 
 qualities under a new stimulus, and generalising in its other 
 relations the same principles which it has applied in the 
 nurserv. 
 
 36. This statement, indeed, requires to be guarded. It is 
 obvious, and I certainly should not seek to evade the admission, 
 that the family affections, like those which are generated in 
 any wider association, may lead to a kind of compound selfish- 
 ness. As a man's patriotism may make him a bad cosmo- 
 politan, or his attachment to a particular church may make 
 him intolerant of other religions, so his love to his family 
 may make him a corrupt judge or an avaricious tradesman. 
 Fathers of families are capable, it is said, of anything; that is, 
 of any amount of injustice to others. And, again, it is of
 
 THE CONSCIENCE. 347 
 
 course a commonplace that the domestic relations incapacitate 
 a man for heroic action either of the good or bad. A man's 
 fondness for his children may make him less disposed to run 
 risks for the good of his race. It follows that the domestic 
 virtues are not a sufficient condition of virtue in general. 
 This, of course, is undeniable, but it does not conflict with 
 the principle as properly understood. For, in the first place, 
 though domestic virtue does not of necessity imply public 
 virtue in a corresponding degree, it implies some moral 
 qualities. What is essentially bad in other social relations is 
 bad, if not permanently bad, in the family relations. To be 
 a good member of a family, a man must practise the duties 
 of chastity, temperance, truthfulness, and of kindliness and 
 justice to at least his immediate surroundings. He must have 
 taken the great step of crossing the tremendous gulf which 
 separates each man from the rest of the universe ; and that 
 gulf once crossed, all further advance is a question of time and 
 cultivation of the sympathies. The question remains, What 
 will be the law of the sympathies? The purely egoistic form 
 may of course observe the external law of domestic virtue as 
 of other virtues ; he may be kind to his wife as to his horse, 
 with an exclusive view to his own comfort; but in that case 
 his virtue is an outside sham which will disappear when 
 accident divides the interests of the persons related to liini. 
 But admitting the reality of altruism, we must also admit 
 that it is naturally stimulated in the family case at the earliest 
 period of life, and that family affections are both the type and 
 the root of all truly altruistic feeling. As soon as we are 
 affected by the sorrow of our brothers, we can be really moved 
 by the sorrow of any other human being who comes into 
 any relation to us. We have the raw material of the moral 
 sense, which will afterwards be developed and regulated by 
 our position in the whole social organism. The development 
 in different cases of course varies greatly. Some men have 
 very strong affections for those whom they see, and very weak 
 sympathies for the distant, because, perhaps, the imaginative 
 faculty is weak ; whilst in other cases the relation is reversed.
 
 348 THE SCIENCE OF ETHICS. 
 
 The affection In the particular case does not of itself deter- 
 mine the law of the affection in other cases^ and that will 
 vary according to individual idiosyncrasies. If my affection 
 for my family rests upon genuine altruism, then so far as I 
 am reasonable I shall sympathise with others in so far as I 
 am able to realise their feelings, and to my power of being 
 useful to them, that is, of giving a practical effect to mv feel- 
 ings. That is, as I love my brother I shall love others, and 
 discharge whatever duties arise from other relations. If I 
 love him only in name, and am kind to him only to gain some 
 advantage to myself, then, of course, I am a good brother and 
 a good friend only in name. To ask how far the domestic 
 affection will prompt to others is to ask how far it is real. 
 
 37. Thus we may say that, if we start from the general 
 point of view, the organisation of society implies a certain 
 distribution of functions which has been gradually elaborated 
 in obedience to the general conditions of welfare. As a cer- 
 tain constitution of the family is bound up in the very struc- 
 ture of the social tissue, all other relations have been developed 
 with reference to these primary relations. And therefore, 
 if a man acted with an explicit reference to the welfare of 
 the whole organisation, he would necessarily acquire the 
 characteristics essential to the good son, husband, and father, 
 as well as those essential to the good citizen, soldier, or 
 craftsman. The whole problem has been worked out by a 
 single process, and therefore every part of the organism has 
 been developed with reference to the rest. But if we start 
 from the opposite or individual point of view, which neces- 
 sarily corresponds to the historical development in each par- 
 ticular case, the character is developed through the immediate 
 surroundings; the man learns to be affectionate, truthful, 
 and so forth through his relation to his own little world, 
 without being even able for a long time to apprehend the 
 general principle, and so acquires the qualities which fit him 
 for other relations as he comes to be sensible of them and 
 required to act on them. Whether we start from the whole 
 and argue to the constituent part or reverse the process, we
 
 THE CONSCIENCE. 349 
 
 come to the same conclusion, though what is laid down as 
 an explicit principle in one case is implicitly assumed in the 
 other, and gradually ev^olved in the process of life. 
 
 38. And further, it must of course be admitted that in 
 many particular cases the duties may conflict. The general 
 qualities which fit a man for excellence in the v^arious rela- 
 tions of life are identical, so far as the moralist can take 
 notice of them, but he cannot say what weight is to be 
 given to special considerations in given cases. It is plain 
 that so far as a man acts as a judge, he must not be moved 
 by the interests of his family ; and the same character which 
 causes a man to act justly as a father will make him act 
 justly as a judge, for he will feel that such conduct is im- 
 posed upon him by the claims of the wider organism, and 
 he will not wish that his children more than himself should 
 be helped by corrupt actions. But it is not possible to lay 
 down a general moral principle which shall decide whether 
 a man should devote himself to a domestic or a public career. 
 For, in the first place, it involves the difficult question of 
 facts as to which career will be the most useful, and, in the 
 next place, the answer must partly depend upon the indivi- 
 dual character. If a man's affections are strong but narrow, 
 he will be better fitted for one position; if diffusive and 
 lively, better fitted for another; and we cannot say which 
 man is the most moral. Such discussions really take us 
 beyond the sphere of morality into questions of prudence, 
 which can generally be decided by nothing but instinctive 
 tact. 
 
 39. For my immediate purpose I have gone far enough. 
 The moral law being, in brief, conformity to the conditions 
 of social welfare, conscience is the name of the intrinsic 
 motives to such conformity. So far as we feel ourselves to 
 be members of any social organisation and identify ourselves 
 with it, we are, in virtue of that sentiment, prompted to this 
 conformity and feel a sense of obligation. In this way a 
 kind of subordinate conscience is formed in regard to even 
 the more cursory forms of association, and still more in re-
 
 350 THE SCIENCE OF ETHICS. 
 
 gard to the more permanent and conspicuous, through which 
 our stronger instincts are gratified. When pubHc spirit im- 
 poses upon us sacrifices to our country, we are actuated by 
 a feeUng which is of the same kind as conscience, and is 
 often indistinguishably blended with it. But it is the 
 pecuHarity of the moral law that it belongs to us, not in 
 any special capacity, but as belonging to the indefinite and 
 formless organisation of the race at large. On the other 
 hand, although this organisation has no form or definite 
 limits, we are in the closest possible contact with it, and it is 
 the underlying substance of all other associations, and 
 especially as the main bonds which hold it together are 
 those of the family relations. Our whole character is 
 moulded from our earliest infancy by the family tie, and 
 the conformation of character so impressed upon us carries 
 with it the wider moral sensibilities. The conscience, there- 
 fore, is not a separate faculty which responds only to a 
 special set of stimuli, but is a compound feeling to which all 
 the strongest instincts of our nature contribute. Through 
 our affections for our friends and our brothers our feelings 
 are stamped and moulded, and prepared to be developed 
 under the action of all the other relations into which we are 
 brought, as our intellects and sympathies expand and our 
 passions come into play. In this way the primary instincts 
 undergo modifications, causing them to act in certain ways, 
 and to obey certain rules which have necessarily a moral 
 quality, or, in other words, a definite relation to the condi- 
 tions of social welfare. The perception that this rule is 
 formed bv somethino; outside us, that we imbibe it from the 
 medium in which we live, gives the sense of obligation, 
 though we may become conscious of it as the expression of 
 instincts which have grown up before distinct reflection, 
 and are involved in all our modes of thought and feeling. 
 And as the process of working it into our character is always 
 more or less imperfect, we have, as a rule, plenty of oppor- 
 tunities for finding that obedience costs an eflbrt, though 
 disobedience may bring with it a pang. The conscience is
 
 THE CONSCIENCE. 351 
 
 the utterance of the public spirit of the race, ordering us to 
 obey the primary conditions of its welfare, and it acts not 
 the less forcibly though we may not understand the source 
 of its authority or the end at which it is aiming. 
 
 40. I will conclude by applying this to the particular case 
 of maternal love, which seems to be as well the purest 
 type as an original germ of virtue. It has been denied by 
 sage philosophers that it is a virtue at all, and this because 
 it is an instinct, which therefore implies at most a compound 
 selfishness, and involves no explicit recognition of the general 
 principle of morality. A mother is not good because she loves 
 her children, but would be virtuous if she deduced the dutv of 
 doing good to them from some abstract principle — from the 
 doctrine that you must act so that your rule may be a rule 
 for all. or again from the belief that by doing them good she 
 was actino- from a wise calculation of her own greatest inte- 
 rest, or of the greatest happiness of the greatest number. I 
 hold, on the contrary, that a mother who loves her children 
 is so far good, though happily she has not much merit, as 
 the virtue is a very common one, and therefore that it is 
 more accurate to say that a mother who does not love her 
 children is very bad. But how is this admission of the in- 
 stinctive character of the virtue to be reconciled with the 
 doctrine that it is reasonable as the embodiment of a general 
 principle in a particular case? My answer is that we must 
 distinguish two closely allied questions. If you ask, " Why 
 is maternal love a virtue?" the answer is, "Because it is 
 essential to social virtue ; " because, in other words, the 
 vitality of every society from the earliest period is dependent 
 upon the vigorous action of this instinct. The same con- 
 sideration shows that, though an essential, it is not a sufficient 
 condition of virtue. The love for infants must be controlled 
 by some interest in others. It can be controlled by the 
 wider instinct generally because the instinct springs from the 
 same root, and the sympathy excited in the mother by the 
 dependent infant is homogeneous with her sympathy for other 
 infants, and for all to whom she can render services. The
 
 352 THE SCIENCE OF ETHICS. 
 
 altruism, again, which is thus generated becomes, in a mind 
 capable of reflection, the conscious acceptance of the general 
 principle, which of course receives additional strength when 
 it is explicitly announced as part of the fundamental convic- 
 tion of the society. It is reinforced by all the other motives, 
 which are enabled to co-operate with it because society is so 
 developed as to secure their normal coincidence. In this 
 sense, then, the general condition may be stated as deter- 
 minins; the moral character of the instinct : it is essential, 
 and perceived to be essential, to social welfare, and therefore 
 (for this is the only reason we can give) it is a virtue, and a 
 recoc^nised virtue. But if we look at the case from the 
 opposite side, and ask for the mother's reason of action, we 
 must invert the order of the deduction. The mother loves 
 because she is so constituted as to be capable of loving, and 
 because she is part of a society in which the instinct is stimu- 
 lated and fostered. For her the love is its own justification; 
 she has the sentiment, and need look no further. The wider 
 love which she comes to feel for others is not the cause of 
 the narrower instinct, but the product whea it comes to be 
 enlin-htened and extended ; and the conscientious feeling itself 
 which sanctions and strengthens the primary instinct is not 
 something which exists independently, but which springs 
 from the instinct as developed through the emotions and the 
 imbibation of the social instincts. The cause, in other words, 
 must be found in the social utility ; the reason in the indi- 
 vidual constitution as developed by the social life. 
 
 41. And this leads to a new division of the subject. I 
 have started from the condition of social welfare, and tried to 
 show how this implies the growth of a sense of obligation in 
 its constituent members; but now we must start from the 
 opposite pole; each man has to be regarded as acting from his 
 instincts, however it may have come to pass that he has those 
 instincts, and therefore as acting with an immediate reference 
 to his happiness. Whence we have to inquire what is the 
 relation between morality and happiness, which I shall proceed 
 to do in the next two chapters.
 
 ( 353 ) 
 
 CHAPTER IX. 
 
 HAPPINESS AS A CRITERION. 
 
 I. Utilitarianism. 
 
 I. The phrase "utility," I have remarked, has a double 
 sense. Conduct mav be called " useful " as it contributes to 
 the preservation of the agent, or as it contributes to his 
 happiness ; and it is an essential part of the evolution theory 
 that these two characteristics must approximately coincide ; 
 that is, that there must be a correlation between the per- 
 nicious and the painful on the one hand, and on the other 
 between the beneficial and the agreeable. By applying this 
 principle to the social organism, we have come to the 
 conclusion that the development of the society implies the 
 development of certain moral instincts in the individual, or 
 that the individual must be so constituted as to be capable 
 of identifying himself with the society, and of finding his 
 pleasure and pain in conduct which is socially beneficial or 
 pernicious. The necessary condition for morality is altruism. 
 The altruistic person is moulded and modifiejd by the society 
 of which he forms a part, and acquires the moral sense which 
 implies a certain intellectual and emotional constitution. But 
 the facts may be equally regarded from the opposite point 
 of view. Conduct, we say, is a function of character and 
 circumstance. If we ask, " Why does a man act in such a way 
 under given circumstances?" the immediate answer must 
 always be in the form, " Because it is pleasant," or, in other 
 words, because it gratifies some of the instincts which toiiether 
 form his character. This must in all cases give the reason 
 
 of his conduct. But if we choose to go further and ask, " Whv 
 
 z
 
 354 THE SCIENCE OF ETHICS. 
 
 is it pleasant ? " the answer must be given by showing how his 
 character comes to be constituted in this particular way; and 
 the only explanation that can be given is the exposition of 
 the relations which the agent bears to the whole system of 
 which he forms a part. " Why do I eat ? " To satisfy hunger. 
 "Why do I hunger?" Because I am constituted in such a 
 way that the consumption of, and, therefore, the desire for, 
 food is essential to my life. In this case I assign (so far as 
 it can be assigned) the cause of my conduct. The problem 
 hitherto considered has corresponded to this last question. We 
 have asked what is the cause of the development of morality, 
 and we have answered by reference to the social organism. 
 We have still to ask what is the reason of morality, or what 
 are the motives which operate upon the individual. 
 
 2. A moral agent must have a reason for moral action, 
 and the reason must clearly have some relation to his happi- 
 ness. The conduct must be always that in which he finds 
 happiness at the time, and the " end " must always be 
 either his own happiness or the happiness of others. If, in 
 fact, the preservation of the race meant the continuance of 
 misery; if, like Milton's devils, we were kept in existence in 
 order — 
 
 " Strongly to suffer and endure our woes," 
 
 we could not reasonably desire existence. We have, therefore, 
 to justify morality both as happiness-giving and as life-pre- 
 serving. If the ends necessarily diverged, we should get into 
 considerable difficulties; but, as I have urged, the very principle 
 of evolution implies that there must be at least an approximate 
 coincidence, and there is no apparent a priori reason why 
 the coincidence should not be indefinitely close. The pessi- 
 mist indeed may regard life as essentially miserable. He 
 escapes from the conclusion that annihilation is desirable 
 by declaring that it is impossible. All that is left for us 
 tipon his showing is to minimise the misery which cannot 
 be annulled. His morality, therefore, aims at what is 
 equivalent to a maximum of happiness, although he states it
 
 UTILITARIANISM. 355 
 
 in the opposite form of a minimum of misery. With this 
 question, at any rate, I am not for the present concerned. I 
 assume that, in any case, we are invariably determined by 
 pain and pleasure, though it is equally true that our pain 
 and pleasure have a necessary relation to the tendency of 
 the corresponding conduct to our preservation or destruction. 
 We have now to ask how the moral rule can be constructed 
 from this secondary point of view, always bearing in mind 
 the condition defined by the primary. The rules which 
 formerly appeared as conditions of maintaining the vigour of 
 the race will now appear as conditions of securing its happi- 
 ness. We have to inquire how the two are related; whether 
 the rule constructed on the one principle coincides necessarilv, 
 and how far it coincides, with that constructed on the other; 
 and under what condition that which is the cause of moral 
 conduct will or will not supply an adequate reason or motive 
 for such conduct to the agent. This, in the utilitarian phraseo- 
 logy which now becomes appropriate, is to inquire into the 
 criterion and the sanction of morality ; and in this chapter 
 I shall speak of the first of these problems. 
 
 3. Utilitarianism is the system which endeavours to con- 
 struct the moral rule exclusively from the principle of happi- 
 ness, and I propose to ask briefly what modification must 
 be imposed upon this system in order to make it square with 
 the theory here adopted. The general assumption upon which 
 it proceeds may be easily laid down. Happiness is the sole 
 end of conduct; the '^utility" of an action is its tendency to 
 produce happiness; its morality is measured by its utility; 
 that conduct is right which produces most happiness, and 
 by this we must be understood to mean which produces 
 most happiness on the average; for since we can seldom 
 calculate more than a small part of the consequences of any 
 action, we are forced to act upon rules correspondino- to the 
 general limits of observation. We find that, on an average, 
 certain kinds of conduct increase and others diminish happi- 
 ness. We have to act upon this probability, and thus we 
 attain the moral law. " This action is wrong," means that.
 
 356 THE SCIENCE OF ETHICS. 
 
 on an average, this action causes a balance of misery. Further, 
 the motive to morality must be the motive, whatever it may 
 be, which makes us desire to promote the general happi- 
 ness. Here utilitarians divide, according as they do or do not 
 admit the reality of unselfish impulses. The egoistic utili- 
 tarian holds that we desire to promote the happiness of others 
 because we shall in some way promote our own happiness; 
 the altruistic holds that the desire of happiness to others may 
 be an ultimate motive. 
 
 4. Although this doctrine is, as I shall presently argue, 
 unsatisfactory as a complete account of morality, it contains, 
 as I think, a core of inexpugnable truth. A great deal is 
 said about the vagueness of the word " happiness " and the 
 impossibility of devising a calculus for determining the effect 
 of conduct upon happiness. The criticism would be con- 
 clusive if utilitarianism required any such calculus. If I 
 attempted to lay down rules for the whole conduct of life, 
 and to say whether in any given case this or that course will 
 give a maximum of pleasure, I should be hopelessly at a loss. 
 On the one side the vast complexity of consequences, on the 
 other the vast variety of tastes, would make it impossible to 
 give trustworthy rules. There is no hope that we shall ever 
 construct a pocket calculating machine which will tell us by 
 a short and easy method what is the path to happiness. But, 
 then, this very uncertainty is an essential part of the utili- 
 tarian contention. It is just because the calculation is so 
 hopelessly intricate that we are forced to trust to rules formed 
 from an average, and that we can obtain so very few of such 
 rules. The moral law can only give us a few very simple in- 
 dications, because our powers of calculating happiness are so 
 limited. In order, therefore, to make the objection valid, it 
 must be shown that our uncertainty is so great as to extend 
 to the consequences of moral behaviour. We must show 
 that there is such a thing as a law, admitted to be moral, and 
 yet of doubtful effects upon happiness. If there be such a 
 thing, we shall hardly hear of it from the other schools of 
 morality, for they are at least as anxious as their opponents
 
 UTILITARIANISM. 357 
 
 to show that morality produces happiness. Nobody, indeed, 
 will seriously profess any doubt that cruelty, lying, sensuality, 
 and so forth, do diminish the stock of happiness. Many 
 people deny that the mischief is the ground or the sole ground 
 of our condemnation, but they do not denv, or rather they 
 solemnly assert, the reality of the mischief. If, then, they admit 
 the fact that wickedness causes misery and virtue happiness, 
 thev cannot attack the utilitarian for holdino; that the fact is as- 
 certainable. If we can know for certain that morality produces 
 happiness, the utilitarian who makes it consist in producing 
 happiness cannot be accused of placing morality upon an un- 
 certain base. The truth upon which he rests is admitted by 
 his antagonist, and they cannot consistently argue that it is 
 a truth which cannot be known. Yet more, if it can be 
 ascertained that any class of conduct increases or diminishes 
 the general sum of happiness, all moralists admit that it is 
 so far right. If it were proved that certain conduct did no 
 harm to anybody, that conduct could hardly be wrong. The 
 duty of benevolence orders us to increase happiness, and happi- 
 ness is per se a good thing, though there may be contingent 
 objections upon other grounds to particular kinds of happiness. 
 The question, therefore, of the tendency of actions to pro- 
 duce happiness cannot be irrelevant to its morality, nor can 
 we deny that moral conduct has that tendency, or that con- 
 duct proved to possess it thereby becomes moral. So far as 
 this is the substance of a good many attacks upon the utili- 
 tarian, I think that he is perfectly capable of holding his 
 own, and has a good solid basis of fact from which it would 
 be rash to attempt to dislodge him. Crime is mischievous; 
 it causes bodily and mental agony; it is the great source of 
 all human suffering, and it is bad for that plain and undeni- 
 able reason. If you could get rid of the reason you would 
 find it very hard to substitute any other of equal cogency. 
 And, indeed, the utilitarian argument appears from a certain 
 point of view to be so cogent, that one is half disposed to 
 regard all the argumentation about morality as grotesque. 
 Can it be necessary to go into such elaborate reasoning to
 
 358 THE SCIENCE OF ETHICS. 
 
 account for the fact that men have generally agreed to con- 
 demn the practice of cutting each other's throats ? Why- 
 should not they ? 
 
 5. Yet, when we try to answer more explicitly the various 
 criticisms that have been so frequently and forcibly expressed, 
 we become sensible that the utilitarian position requires at 
 least re-statement or reconstruction. The system has been 
 attacked as giving an inadequate account of all the most 
 essential characteristics of the moral law. It is said, in the 
 first place, that since morality depends upon the calculus of 
 happiness, since men's conceptions of happiness vary within 
 almost indefinite limits, and since the tendency of actions 
 to produce particular kinds of happiness is only to be dis- 
 covered by examining a vast variety of complex phenomena 
 which elude all scientifie inquiry, the rules which result must 
 necessarily be arbitrary or indefinitely fluctuating. If at a 
 given moment they take one shape, there is no assignable 
 reason why they should not take another at any other time 
 or place. Since, again, we start from individual conceptions 
 of happiness, and we have no more reason for assigning 
 special importance to the judgment of one man than to that 
 of any other, or of preferring the estimate of the saint to the 
 estimate of the sinner, the standard which results from the 
 average judgment must be an inferior or debasing standard. 
 Further, since on this hypothesis the morality of conduct is 
 essentially dependent upon its consequences, that is, upon 
 something different from the action itself, we must always be 
 led to an external moral code. Evil cannot be objectionable 
 as evil or good desirable as good, but we must always con- 
 sider morality as a means to some ulterior end; and thus the 
 very essence of virtue is destroyed ; a conscience becomes 
 superfluous; and hence, finally, the moral law of the utili- 
 tarian can never get beyond expediency. There is always 
 some other condition by reference to which we must decide 
 upon any particular line of conduct, and therefore the moral 
 rule, though it may serve as a useful indication of what is to 
 be done in average cases, cannot be a supreme and absolute
 
 THE EVOLUTIONIST CRITERION. 359 
 
 rule, deciding what is to be done in all cases. Hence all 
 the specific characteristics of which we took account in 
 framing our theory of morality are more or less destroyed ; 
 for though the utilitarian can provide a kind of substitute for 
 the various qualities described, he can only make an outward 
 show of morality, and run up an edifice which looks like the 
 everlasting structure, but falls to pieces at the first touch. 
 He may call his code moral, but in fact it is a code which 
 has neither permanence nor supremacy, nor uniformity nor 
 unconditional validity. 
 
 H. The Evolutionist Criterion. 
 
 6. I have already given by anticipation my answers to 
 these charges, if applied to the moral system which I am 
 defending. The svstem, however, according to many thinkers, 
 is simply the old dog in a new doublet. I propose, therefore, 
 to consider more precisely how far the evolutionist morality 
 can meet the theories which have some cogency as against the 
 older utilitarianism. Utilitarianism, let us note in the first 
 place, springs from the mode of speculation which renounces 
 as much as possible every a priori method, and rejects all 
 "intuitions^' or supposed logical necessities of thought, in 
 order to base morality upon pure experience. The tendency 
 of the utilitarian is therefore to consider knowledge in general 
 as conforming to the type of that purely empirical know- 
 ledge in which the experience of a former coincidence of two 
 distinct phenomena is the sole basis for our expectation of a 
 future coincidence. Carrying out this principle as far as 
 possible, reasoning is essentially a process of associating ideas, 
 and the association, though practically indissoluble in some 
 cases, is regarded as always potentially dissoluble. The logical 
 result is atomism, or the reduction of every kind of organised 
 system, whether of " ideas," regarded as existing in the 
 mind, or of the objects external to the mind and represented 
 by the ideas, to an aggregate of independent luiits, capable 
 of indefinite analysis in the mind, or of being taken to pieces
 
 36o THE SCIENCE OF ETHICS. 
 
 and reconstructed in reality. All a priori truths, therefore, 
 disappear; and as so-called a priori truths may be unverified 
 and erroneous assumptions, the application of a thoroughgoing 
 analysis is at least useful provisionally, even if the scepticism 
 to which it leads should not be ultimately justifiable. Further, 
 as it is the tendency of thinkers of this class to account for 
 all differences between two organisms as in some sense due 
 to " circumstance," they are forced by a logical necessity to 
 assume the existence of uniform atoms upon which the 
 circumstances operate. The difference, for example, between 
 two men being due to the various associations, and not to 
 those innate tendencies of character which are suspected of an 
 affinity to " innate ideas," we must suppose that there is a 
 uniform man — a colourless sheet of paper or primitive atom 
 — upon whom all qualities are imposed by the circumstances 
 under which he is placed. This assumption, in fact, plays a 
 considerable part in some utilitarian theories. 
 
 7. The existence of assumptions more or less explicitly 
 accepted explains the general tendency of the school, as it 
 may help to render intelligible some of their shortcomings. 
 Society, according to that doctrine, is an aggregate built 
 up of the uniform atoms called men. The only primitive 
 property which can be attributed to man is the desire for 
 happiness ; and we must conceive of happiness as a kind of 
 emotional currency, capable of being calculated and distri- 
 buted in " lots," which have a certain definite value indepen- 
 dent of any special taste of the individual. Conduct, then, is 
 moral or immoral according as it tends to swell or diminish 
 the volume of this hypothetical currency. Pains and pleasures 
 can be handed about like pieces of money, and we have simply 
 to calculate how to gain a maximum of pleasure and a mini- 
 mum of pain. I have certainly no desire to fix down the 
 utilitarian to any extreme form of his theory, or to pursue 
 some grotesque consequences to which it is occasionally 
 applied. The criticisms which I shall consider are those 
 which seem to me to be applicable to the essence of the 
 doctrine.
 
 THE EVOLUTIONIST CRITERION. 361 
 
 8. The criterion thus suggested is, in fact, Hable to one 
 criticism which appears to me to be decisive so far as it 
 appHes, and to show the real Hmitations of the method; I 
 mean that it fails to take into account an essential condition 
 of any tenable theory, and that precisely because it refuses 
 to take into account the true nature of the social organism, 
 and considers it a simple combination of independent atoms. 
 The utilitarian argument would be perfectly relevant if we 
 could take each action by itself, sum up its consequences, and 
 then generalise as to the actions of the class. So, for example, 
 I find that eating green fruit is always followed by a painful 
 sensation. I resolve to abstain; and if all other men made 
 the same remark, they would, if wise, follow my resolution ; 
 assuming, of course, that the subsequent pain was clearly 
 greater than the immediate pleasure, and that no counter- 
 balancing advantages were observable. VVe may argue in 
 the same way about murder, stealing, lying, drunkenness, 
 and so forth. They do infinite mischief, and mischief which 
 clearly overbalances the pleasure. We judge that thev 
 diminish the sum of happiness, and I have no doubt that our 
 judgment of utility is so implicated in the moral judgment 
 that one could not change without a corresponding change in 
 the other. But it is also true that our judgment as to the 
 effects of immoral conduct are very inadequately represented 
 by this simple and direct process. The primary evil of murder 
 thus estimated is the pain suffered by the victim, against 
 which, if we take happiness as good per se, we must set off 
 the pleasure of the murderer. If morality is to be defined by 
 happiness, we must of course allow all kinds of happiness to 
 count, and to count equally so far as they are actually equal. 
 We must reckon the pleasures of malevolence as well as those 
 of benevolence. Allowing that the balance inclines, even 
 upon this showing, against the practice, the calculation seems 
 insufScient to justify the strength of the general prejudice. 
 We amend our argument, therefore, by taking into account 
 the secondary or ulterior consequences, and especially the 
 shock to the general sense of security. We may possibly
 
 ^62 THE SCIENCE OF ETHICS. 
 
 object, again, to allowing the murderer's pleasure to count, 
 because a motive which implies pleasure in the infliction of 
 pain is a mischievous motive, and therefore whatever pleasure 
 it may produce is bought at a ])rice to society at large. But it 
 is now evident that we must take into account a consideration 
 hitherto neglected, namely, the existence of a certain social 
 order, and of a corresponding character in the individual 
 constituents; for as the shock to the sense of security is 
 undoubtedly an important item in the account, the shock is 
 proportioned to the existence of a certain standard of mutual 
 confidence. Murder means, speaking brieflv, killing by 
 private persons. The executioner and the soldier may kill 
 under certain circumstances; and though war may be de- 
 nounced by hasty theorists as wholesale murder, the distinction 
 is important, for war does not in the same way imply a disin- 
 tegration of a social order, but is, on the contrary, an essential 
 part of the process by which that order has been built up. 
 It is easy to propose the summary abolition of war, and we 
 all hope that it may be abolished ; but all men, except a few 
 enthusiasts, can see that to propose its abolition is to propose 
 a complete social and moral reconstruction. It is not an 
 excrescence which can be simply dropped, but the result of 
 processes essential to the growth of society in certain stages. 
 What is true of war is equally true of murder. At an early 
 period the distinction between public and private killing is 
 unintelligible, and for want of an organisation fitted to 
 suppress individual conflicts, men must be allowed to fight 
 out their own quarrels, and to act in the way which we should 
 afterwards (that is, when a law has been developed) describe 
 as " taking the law into their own hands." Again, it is quite 
 true that the murderer in the present day is a malevolent and 
 therefore mischievous individual, whose gratification is not 
 desirable because it inflicts more evils than are compensated 
 by his pleasure. But this, again, is virtually to assert that a 
 social development has been evoked in which the pugnacious 
 instincts are so mischievous that they can be placed under 
 certain restraints. At an earlier period, when there was not
 
 THE EVOLUTIONIST CRITERION. 363 
 
 a residentiary police, the calculus of happiness would be 
 materially different. The murderer would not be injuring 
 a sense of security which did not in fact exist — perhaps 
 would even be dischartrino; a necessary function : nor, 
 again, would the pain of the sufferer be the same, nor the 
 pleasure derived from the killing be indicative of so mis- 
 chievous a character. 
 
 9. It follows, then, that the direct application of the 
 calculus omits a most essential element in the calculation. 
 Let us note — for the point is one of vital importance — what 
 are the tacit assumptions involved. The utilitarian or indi- 
 vidualist considers society to be formed of an aggregate of 
 similar human beings. The character of each molecule is 
 regarded as constant. The only difference which he considers 
 to be relevant in a moral sense consists in a more or less 
 exhaustive and accurate calculation of the consequences of 
 actions. That society, therefore, will enjoy the greatest 
 happiness which has the clearest perception of these conse- 
 quences,' and consequently enforces the corresponding rule; 
 for we at present assume that a perception of the evil leads 
 to its suppression. Now at any given moment, as character 
 varies slowly and the social relations may be taken as approxi- 
 mately fixed, this gives an approximately accurate test; that 
 is to say, the consequences of immoral conduct generally 
 involve misery, and the further we trace them the more 
 evident is the fact. But when we try to frame something 
 like a scientific criterion from such considerations, w^e become 
 sensible of the inadequacy of the statement. For the conse- 
 quences can only be traced when we recognise the nature of 
 the social structure, which again implies the existence of a 
 certain stage of individual development, and neither of these 
 is deducible from the properties of the assumed unit. Human 
 nature is not a constant, but, on the contrary, a variable, and 
 the aim of the moralist is precisely to modify it. The problem 
 changes in our hands as we consider it. If, in f^ict, we ask 
 what are to be considered as the consequences, it is plain at 
 once that we cannot make an arbitrary selection of the most
 
 364 THE SCIENCE OF ETHICS. 
 
 obvious and prominent. What are the consequences of a 
 murder? Evervthing which isimphed in the murder not hap- 
 pening, including the consequences to the victim, to societv 
 at larsie, and all that follows from the implied change in 
 the character of the murderer. What are the consequences 
 of a certain frequency of murder? All that is implied in the 
 difference between a society where murder is frequent and 
 one in which it is not frequent. This, again, implies a com- 
 plete structural change, the consequences of which reach inde- 
 finitely beyond this particular mischief. To suppress murder 
 is to civilise a societv ; and unless we take into account the 
 laws of social growth, it is impossible to say what is included 
 in civilisation. 
 
 10. The utilitarian criterion, again, is frequently presented 
 in the form of a maximum. Morality is conduct tendinc; to 
 produce the greatest happiness of the greatest number. But 
 a maximum is meaningless unless we assume certain fixed 
 conditions. Here we must mean the greatest happiness 
 possible, and therefore possible on a certain assumption. But 
 on what assumption? The assumption of the fixity ofhuman 
 nature. But that alone is insufficient. It would lead to 
 such conclusions as this: that the society was happiest in 
 which there were fewest murders, whatever the cause of their 
 rarity. But this is at least a doubtful truth, since violence 
 may be diminished as well by diminution of energy as by an 
 increase of peaceful ness, and the bare elimination of particular 
 practices gives a totally inadequate measure of social welfare. 
 Doubtless a diminution of certain evils will be a symptom of 
 social progress after a certain stage, but it is not a measure 
 of the whole complex process. If, then, we suppose that a 
 given stage has in fact been reached, and that on comparing 
 two societies at that stage, that will be the happiest in which 
 there are the fewest murders, we are making a tenable pro- 
 position, and one which is undoubtedly of vast importance. 
 But it has an appearance of being arbitrary, because we take 
 for granted the existence of certain instincts as an ultimate 
 fact; and therefore, though excluding intuitions, we are vir-
 
 THE EVOLUTIONIST CRITERION. 365 
 
 tuallv assuming the existence of a certain disposition, whilst 
 we make no attempt to justify our assumption. We prove 
 that, under given circumstances, murder is on the average 
 objectionable, but we do not attempt to state what is the 
 cause of its badness, or to state the general principle upon 
 which we are proceeding. It may always be argued that we 
 are biassed by certain prejudices which are more or less arbi- 
 trary, or which, in other words, might be changed without 
 injury to the general happiness. We object to murder be- 
 cause in the existing state of society it does more harm than 
 good. But suppose we get rid of some of the feelings con- 
 cerned, might we not be the happier on the whole ? To 
 answer this we are thrown back upon the previous case, and 
 have to compare the amount of happiness in two societies 
 agreeing only in the circumstance that both are composed of 
 men, which seems to render the whole problem too intricate 
 and indeterminate for practical application. 
 
 II. Consider for a moment what is perhaps something 
 more than an analogous case. So far as our physical con- 
 stitution is concerned, the only conceivable motive is the 
 attainment of pleasure. We may say, therefore, that a man 
 acts most wisely — considered simply as an animal — who acts 
 so as to obtain the maximum of pleasure. But if we should 
 seek to frame a rule of life directly from this consideration, 
 we should fall into infinite perplexity. We must obviously 
 say, in fact, that he must act with reference to his constitu- 
 tion. Pleasure is not a separate thing independently of his 
 special organisation. The bare rule, " Get as much pleasure as 
 you can," is unintelligible unless we proceed further and point 
 out some of the conditions which must be observed. Each 
 instinct, for example, must have its turn, and their respective 
 provinces must be determined by the general organic balance. 
 We can undoubtedly point out that certain modes of con- 
 duct produce pain and others pleasure, and this is a primd 
 facie reason, at least, for avoiding one and accepting the 
 other. But, again, some pains imply a remedial process, 
 whilst others imply disease; and the conduct which increases
 
 366 THE SCIENCE OF ETHICS. 
 
 them may therefore be either wise or foolish in the highest de- 
 gree. For the simple rule, therefore, " Get the most pleasure," 
 we must substitute the general rule, " Preserve health." The 
 two rules plainly coincide very closely. The healthiest man 
 is generally the happiest, and therefore the best and only 
 general rule for securing physical pleasure is, "Be healthy." 
 No doubt some kind of rule might be constructed by aimino- 
 at happiness. We may recommend temperance, for ex- 
 ample, because we observed as a truth that intemperance is 
 generally followed by a headache, and the practice of intem- 
 perance by all the pains of various diseases. But we are then 
 tacitly referring to the organic conditions which are summed 
 up by saying that intemperance is inconsistent with health. 
 This, therefore, whether explicitly or implicitly stated, is a 
 necessary element in our statement, and gives the only o-eneral 
 criterion. We wish for health, it may be, only as a condition 
 of pleasure, but the only general rule for obtaining pleasure 
 is to secure this general condition. If we tried directlv to sum 
 up the various kinds of pleasure, to compare the value, to 
 discover how far they were compatible, and to decide how 
 much we should take of each, we should embark upon a 
 hopelessly complicated undertaking, which is made needless 
 by the single consideration that the process is in all cases 
 conditioned by the maintenance of the organic condition 
 called health. The organism has solved the problem for us 
 approximately. It has come to be so constituted that what 
 is pleasant is approximately wholesome. We start with that 
 assumption, and correct the errors by the inverse conclusion 
 that what is wholesome is in the long run also productive of 
 most pleasure. 
 
 13. This, as it seems to me, represents the real difference 
 between the utilitarian and the evolutionist criterion. The one 
 lays down as a criterion the happiness, the other the health of 
 the society. The two are not really divergent; on the con- 
 trary, they necessarily tend to coincide; but the latter satisfies 
 the conditions of a scientific criterion in a sense in which 
 the former fails. I desire happiness. I discover by experience
 
 THE EVOLUTIONIST CRITERION. 367 
 
 that this or that particular set of conditions makes me happy, 
 and by degrees I learn how to divide myself amongst different 
 impulses, and so to obtain certain general rules of happiness. 
 But to obtain any germ of a scientific theory I must turn 
 from without to within ; the law of happiness will appear as 
 simple only when it is regarded as a law in terms of the state 
 of the single individual who goes through all these v^arious 
 experiences. The external conditions of happiness are multi- 
 tudinous and incapable of summation, but they must all agree 
 in this, that they stimulate me in a manner consistent with 
 the laws of my being, in the unity of which they are com- 
 bined and correlated. Therefore the general rule must be a 
 rule relative to my state, and briefly the law of my health. 
 What is true of the individual is true in proportion of the 
 much valuer and less coherent social oro;anism. We obtain 
 unity of principle when we consider, not the various external 
 relations, but the internal condition of the organism. The 
 conditions might conceivably be laid down either by saying 
 that the various social functions are discharged and the rela- 
 tion between the social organs maintained in a certain equili- 
 brium, or by trying to sum up all the various modes of conduct 
 which produce happiness to its various members ; but we only 
 get a tenable and simple law when we start from the structure, 
 which is itself a unit. 
 
 13. Hence, again, we obtain a rule which is fixed and 
 elastic in the right place. An organism may be healthy or 
 diseased at any period of its growth, and the laws of health 
 or disease will be continuous, although varying within limits 
 as the organism itself varies. But if we take the direct 
 utilitarian criterion, it seems to be rigid and yet indefinitely 
 variable in different directions. For as human nature is taken 
 as a constant, we should always have the same rule of 
 conduct; and yet as men's thoughts and feelings are sup- 
 posed to be indefinitely variable in obedience to accidental 
 association, we seem to have no guarantee for the permanence 
 of any moral criterion. In fact, utilitarian moralists have 
 dwelt upon the variations of the moral standard in order to
 
 368 THE SCIENCE OF ETHICS. 
 
 prove the necessity of resting morality upon experience and 
 to get rid of a priori intuitions. They have dwelt upon the 
 same facts in order to justify a belief in human perfectibility. 
 Yet surely if human nature is in this sense so modifiable, 
 we have no guarantee for expecting amelioration rather than 
 deterioration. And if, as I have said, human nature is re- 
 garded as in some sense a constant, the science of morality, 
 which should be rigidly deducible from its properties, can 
 hardly be realised when the human material is capable of 
 being worked up into indefinitely varying forms. It is in 
 substituting for these contradictory examples the conception 
 of a slowly developing social organism that the evolution 
 philosophy has rendered the greatest service to ethics, as the 
 variations become themselves reducible under a fixed rule, and 
 the necessity of recognising the social organism as something 
 not formed by simple mechanical combination restores the 
 due authority to social instincts without elevating them into 
 transcendental intuitions. 
 
 14. Briefly, then, I . regard utilitarianism as giving what 
 may be called instantaneous morality. It corresponds to the 
 way in which men actually reason and are justified in reasoning 
 provisionally as to moral questions. We see that a certain 
 social arrangement or regulation produces bad and good 
 effects. We try roughly to sum them up, and to regulate or 
 repeal accordingly. Our moral judgments are in all cases de- 
 termined by these observations, or must be in conformity with 
 them. Any class of conduct which clearly produces a balance 
 of misery is so far bad, and that which produces a balance of 
 happiness is so far good. A constant and steady attempt to 
 get rid of the misery-causing, and to encourage the happiness- 
 causing activities is the condition of all moral progress. In 
 order to modify any moral law or any social arrangement, we 
 try to show that it actually causes some misery, and that its 
 modification would produce more happiness. I do not see 
 that any other mode of argument has ever any real efficiency. 
 The actual progress in morality is always determined at every 
 point by utilitarian considerations. But when we try to
 
 THE EVOLUTIONIST CRITERION. 369 
 
 generalise from this, and to sav that the form of morality or 
 the criterion of moral conduct is the tendency to produce 
 happiness, we get into difficulties. The reason is that already 
 civen. We are generalising in such a wav as to omit an 
 essential condition of an accurate statement. We are taking 
 constants for variables and variables for constants. If v^'e 
 compare anv two organisms, we assume a certain organic 
 identity. Both of them, we suppose, have a certain fixed core 
 of instincts, and corresponding habits which define their 
 essential character. It is also capable of varying in a sub- 
 ordinate degree compatibly with the uniformity of these 
 fundamental organic relations. Now we may say that any 
 conduct which produces pain is so far bad. So far as the 
 being is capable of intelligent observation, it may classify the 
 various kinds of conduct as they lead to painful or pleasurable 
 conditions, and its action will be determined accordingly. On 
 the other side, pain implies a certain condition of the organ- 
 ism itself which we may call morbid as depending upon a 
 deviation from the normal equilibrium. The being itself 
 must be so constituted that the conditions of pleasure coin- 
 cide closely with the conditions of health. So far, therefore, 
 that is, whilst we assume an identity of organic relations, 
 the two rules will coincide ; those which cause pain are bad, 
 and those which cause morbid action are bad. The internal 
 and the external law approximately coincide. The external 
 rule assumes indeed the existence of a certain internal con- 
 stitution, but as that is fixed it does not require to be ex- 
 pressly stated. But now we suppose a new instinct to be 
 required ; we perceive that certain conduct produces pain 
 on an average, and we therefore propose to eliminate it. It 
 cannot be simply left out, because we are dealing with an 
 organism, and every such change involves an action upon the 
 whole constitution. The organic relations, therefore, are 
 themselves changed, and we substitute virtually a new organ- 
 ism for the old. The new, we may assume, is on the whole 
 more efficient than its predecessor, and represents a more 
 
 complete solution of the general problem of life. As it 
 
 2 A
 
 370 THE SCIENCE OF ETHICS. 
 
 has new sensibilities, it no longer estimates happiness in the 
 same way as the old. If we compare the two, we must not 
 suppose that we have an identical being placed in different 
 circumstances and making different estimates of happiness, 
 but two different beings, with different measures of happiness. 
 The difference is not represented by a more complete attain- 
 ment of the same ends, but by a change in the end itself, and 
 a greater total efficiency of the whole new system. The 
 common rule is that each oro-anism is better as it obeys the 
 conditions of health, but we cannot found any common rule 
 upon the happiness, the standard of which changes as the 
 organism itself changes. 
 
 15. In this sense the growth of the social organism is 
 precisely analogous to that of the individual. The devel- 
 opment of the animal implies the slow acquisition of new 
 instincts, which in time become part of its organic constitu- 
 tion. Whilst they are not fully organised they determine its 
 conduct more or less by the pain and pleasure with which 
 they are associated, and they tend to become fixed as they 
 imply on the whole a superior or more efficient form of 
 organisation. The moral instincts of the society correspond 
 in the same way to the social development, and express at 
 every instant the judgment formed of the happiness and 
 misery caused by corresponding modes of conduct. As they 
 become organised the whole society becomes more efficiently 
 constituted, and its standard of happiness is also modified. 
 We may therefore say that at any period the vitilitariau 
 judgment must be satisfied. Given a certain stage of social 
 development, the society will be in a healthier state and the 
 general happiness greater in proportion as it is moral. But 
 since the happiness itself changes as the society develops, we 
 cannot compare the two societies at different stages as if they 
 were more or less efficient machines for obtaining an identical 
 product. 
 
 16. The importance of the distinction is illustrated in 
 almost every important social discussion. We notice certain 
 bad results from a particular economical or social arrange-
 
 VARIABILITY OF MORALITY. 371 
 
 ment. The indissolubility of marriage inflicts hardship 
 upon many individuals ; let it be dissoluble in those cases. 
 The importation of foreign products ruins certain manu- 
 facturers ; let it be prohibited. We remedy the immediate 
 evil by suppressing more obvious symptoms ; but we often 
 forget that we are dealing with a complex organism, and 
 that the real problem involves innumerable and far-reaching 
 actions and motives due to its constitution. We may be 
 remedying the grievances of individual husbands and wives 
 l)y lowering the general sanctity of family relations, and 
 helping a particular class at the expense of the general 
 efficiency of the nation. I need not say how constantly 
 such omissions vitiate plausible arguments and involve the 
 accomplishment of the chimerical hopes of reformers. A 
 similar practical conclusion may be drawn from this part of 
 the argument, namely, that to prove that a rule of conduct 
 involves pain in many cases is not a sufficient reason for 
 abolishing it, though it is a presumption for modifying it. 
 We are also bound to show that all the consequences involved 
 in the change, including the changes in the individual 
 character and in the social structure, will produce a more 
 efficient social organism. It is true that, as a rule, we have 
 to work out such changes by actual experiment ; but it is well 
 to note beforehand the need of conforming our expectations 
 to the complexity of the problem. 
 
 III. Variahillty of Morality. 
 
 17. The evolution theory necessarily assumes a variation 
 of morality, but not an indefinite or arbitrary variability. 
 And this may lead us to a further question. We must 
 admit, of course, that the calculus of happiness will give 
 different results at different periods. Qualities will be re- 
 garded as virtuous amongst savages which cease to be virtu- 
 ous amongst civilised men. Revenge is sometimes regarded 
 as a duty, and in a rude social state we may say that it 
 may conceivably cause more happiness than misery. As,
 
 372 THE SCIENCE OF ETHICS. 
 
 according to Bacon, revenge is a kind of wild justice, it may 
 to a certain extent discharge the functions of justice before 
 a settled system of law has been developed. To weaken 
 the motive without corresponding development of the 
 virtues of order, would be to remove a penalty upon wrong- 
 doing, and might imply rather a deficiency in energy of 
 feeling than a defect of sympathy. Of two societies at the 
 corresponding stage of development, the one which had it 
 least might be in a morbid condition, and therefore in one 
 less favourable to the average happiness than the other in 
 which the sentiment was vigorous. This inconclusiveness of 
 the utilitarian standard if no reference be made to the social 
 state follows from the previous argument, and it presents 
 another difficulty which frequently occurs. The variability 
 of different estimates of happiness is pointed out as one 
 of the main difficulties of the utilitarian creed. One man 
 prefers sensual, another intellectual pleasures. Which is 
 right, and why ? The question is one of importance in 
 regard to the sanction of morality — that is, in assigning the 
 general motive to which moralists may appeal — and I shall 
 return to it in the next chapter. Meanwhile let us ask 
 how far it affects the criterion of morality. If the moral 
 criterion resulted from taking an average of different esti- 
 mates of happiness, there would be, doubtless, a difficulty. 
 But then I deny that this gives in any case the true theory 
 of morality. The moral sense is, indeed, according to me, 
 a product of the social factor. It is the sum of certain 
 instincts which have come to be imperfectly organised in 
 the race, and which are vigorous in proportion as the society 
 is healthy and vigorous. Undoubtedly, again, they fall in 
 with the general belief as to the effects upon happiness of 
 certain modes of conduct. This, again, is equally true 
 whether we suppose that the society has thriven because it 
 had useful instincts — that is, because its judgments of happi- 
 ness were in fact such as to make it thrive — or whether 
 we suppose that the instinct has been consequent upon an 
 observation of certain bad consequences. In the earlier stages
 
 Variability of morality. 373 
 
 of development the first method will presumably be the 
 dominant, and in later stages the second, whieh implies 
 a certain power of distinct reasoning. But in any case, there 
 can be no doubt that the judgment that conduct is im- 
 moral will coincide with the judgment that it diminishes 
 happiness ; and therefore a race which has low concep- 
 tions of happiness will have a low view of morality. If the 
 sensual pleasures play a great part in the general estimate of 
 happiness, it is probable that the virtue of temperance will 
 not be hi2;hlv estimated. 
 
 18. This, I imagine, can hardly be disputed; but, on the 
 other hand, it is not true that the moral judgment preva- 
 lent in a race amounts implicitly or explicitly to the asser- 
 tion that the average standard of happiness is that which a 
 moral agent will desire, or which it is desirable to secure for 
 society at large. In fact, as I have argued, the general pro- 
 cess in any progressive state of society implies something 
 materially different. The moral law first appears, as I have 
 said, in the external form. People come to recognise, that 
 is, that certain lines of conduct do harm irrespectively of 
 the motives of the agent. They observe, more or less clearly, 
 that certain breaches of the law of temperance produce inju- 
 ries to men in various relations; that the attempt to enforce 
 particular beliefs causes war, misery, endless discord, and 
 the suppression of intellectual activity. When this opinion 
 has been established we have what may be called a purely 
 utilitarian law of morals. There is a general conviction 
 that a certain class of conduct causes suffering, but it is not 
 yet clearly realised that such conduct is in itself wicked. 
 It may be adopted by good men or from good motives — if 
 goodness is measured by conformity to the existing standard. 
 But if society improves, the external law must become an 
 internal law; the conduct which produces the bad effects 
 must be intrinsically repugnant, and there is therefore a 
 demand for the true virtues of temperance and toleration, 
 with all the necessary implications as to the whole char- 
 acter. The purely utilitarian rule proposes a problem
 
 374 THE SCIENCE OF ETHICS. 
 
 which Is solved by generating a type of character so con- 
 stituted that the evils perceived are intrinsically hateful to it. 
 19. It is, in fact, plain that the moral judgment does not 
 correspond exactly to the individual estimate of happiness. 
 A man, for example, may be sensual: w^hen a gross pleasure 
 is within his grasp, he may prefer it to a refined pleasure; 
 but in his judgment of his neighbours he sees that the sensual 
 person incurs diseases and inflicts injuries upon his neighbour. 
 Even the most sensual of mankind wishes his daughters to 
 grow up chaste, and would try to prevent his sons from be- 
 coming drunkards. A man may be intolerant in the sense of 
 desiring to impose his own opinions by downright force, or by 
 other modes than pure reason, but he may recognise the fact 
 that the general admission of this principle leads to innume- 
 rable evils. It is a common observation that persecution has 
 been attacked by men who had a full share of the persecuting 
 spirit. They objected to being persecuted themselves, not to 
 persecution in general ; but in order to secure themselves they 
 were forced to appeal to principles common to all, and there- 
 fore to insist upon the bad consequences of persecution in 
 general, and to become, in spite of themselves, advocates of 
 the genuine virtue of toleration. This process is the typical 
 one. In attacking particular evils we become sensible to the 
 general underlying principle, and by pointing out the mischie- 
 vous results of certain kinds of conduct we are virtually prepar- 
 ing a higher type of character, to which not only that kind 
 of conduct is repugnant, but all such conduct as springs from 
 similar qualities of character. We object to the sensual con- 
 duct, not primarily as sensual, but as mischievous in some 
 other way; we come to object to the sensuality itself when 
 we recoi^nise it as the source of these and of other evils. The 
 process is possible because at every instant we start from an 
 approximate solution of the problem, and with instincts so 
 far balanced and correlated as to be consistent with the con- 
 ditions of existence : the process is slow because the redress 
 of any evil involves a readjustment of the whole character; 
 and it is endless, because from each point we have reached we
 
 VARIABILITY OF MORALITY. 375 
 
 develop new faculties, and have to attack new and wider 
 problems. The whole process corresponds, not to the sum- 
 mary solution of a problem from fixed data, but to an inces- 
 sant series of approximations, in which we start from one 
 organisation which works tolerablv to another which will on 
 the whole work better. 
 
 20. In this sense it is true, as I should say, that the actual 
 character of men, and therefore their estimate of happiness, 
 must always provide the basis for every further improvement. 
 The new rule of morality can only be introduced by making 
 them sensible to evils appreciable at the lower stage. But at 
 any given point of the process the moral law implicitly com- 
 mands conduct for the realisation of which an improvement of 
 the whole order, and therefore an elevation of the moral sense 
 itself, is commanded. The lower natures can only be reached 
 by motives suitable to their application ; but in commanding 
 conduct upon the lower ground the moralist is already favour- 
 ing the introduction of the higher motive, and the actual 
 moral sense is thus, in a progressive society, always in advance 
 of the actual standard of the average individual. He can 
 see, even from his own point of view, the advantages of a 
 better morality, though it has not yet become a principle of 
 his own character. 
 
 21. The general condemnation of utilitarian morality as 
 degrading springs partly from another cause. I mean that 
 the utilitarian is naturally the man who is beyond all things 
 anxious to have his feet on solid earth, and to assign definite 
 and tangible grounds for every conclusion. He is a realist 
 as opposed to an idealist, prosaic rather than poetical, or 
 belongs to the school which has more affinity for the mater- 
 ialist than the idealist conclusions. This is, of course, unde- 
 niable, and utilitarian codes of morality are spun of coarser 
 if more enduring materials than those of the antagonistic 
 systems. But this follows from the temperament rather 
 than from the principles of the moralist. The same disposi- 
 tion which makes him a utilitarian leads him to assign com- 
 paratively little importance to the kinds of happiness which
 
 376 THE SCIENCE OF ETHICS. 
 
 imply a poetic imagination or a delight in the ideal world. 
 But though he is liable to this error^ it is an error upon his 
 own principles. In whatever degree the poetic faculties give 
 real pleasure^ he must admit that pleasure into his calculus ; 
 in whatever degree they are really conducive to the eleva- 
 tion or development of the race, he must admit that they 
 are as "useful" as the humbler instincts. The facts must 
 decide, and there is no a priori reason for assuming that 
 they will give a humiliating decision, though it is easy enough 
 to see why the intellectual temperament, favourable, as specu- 
 lation has hitherto been conducted, to the less idealist view, 
 should also incline a man to put what we may simply call a 
 lower interpretation upon the facts. But this is parenthetical. 
 
 IV. Extrinsic Morality. 
 
 32. The argument thus brings us to another set of criti- 
 cisms upon utilitarian theories. The utilitarian takes what 
 I have called the external view of morality; he judges 
 from consequences conclusively, and says that conduct is 
 good or bad conclusively as it produces a balance of happi- 
 ness or misery, and this irrespectively of the motives of the 
 agent. The motive, therefore, to moral conduct is always 
 extrinsic. It is not in itself bad, but bad as producing some 
 other effect. This doctrine takes various shapes with differ- 
 ent utilitarian moralists, and if it were my purpose to criti- 
 cise their doctrine exhaustively, I should have to consider 
 precisely the meaning of some of the terms employed — as, 
 for example, of the distinction between an act and its con- 
 sequences — for the meaning seems to fluctuate considerably. 
 I have, however, virtually given my answer in defending mv 
 own view, and I will therefore be content with indicating 
 the general nature of the divergence. I have said, in the 
 first place, that a law becomes truly moral when it can be 
 stated in the internal form. Morality is the conduct of the 
 truly moral man, and immorality the conduct which is 
 intrinsically repugnant to him. But here is the ambiguity
 
 EXTRINSIC MORALITY. 377 
 
 which I have already tried to explain. The utilitarian 
 asserts that there is no such thing as a love of virtue "for 
 its own sake." In one sense of the words I entirely agree 
 with him ; in another, I should say that there is no real 
 virtue at all unless it imply a love of virtue for its own sake. 
 I agree with the utilitarian in so far as I deny that con- 
 science is a separate facultv, instead of a mode of reaction of 
 the whole character. If, therefore, there is any class of con- 
 duct which has no relation whatever to happiness of any 
 other kind, which does not gratify or repel other instincts 
 than the conscience, it can have no interest for the con- 
 science either. Virtuous actions are not a separate class of 
 actions, and actions which have no effect upon happiness 
 are for that reason morally indifferent. On the other hand, 
 it is equally true that a man is virtuous only if the bare fact 
 tnat an action is right is to him a motive for acting in that 
 way, and he is virtuous in proportion as it is always a suffi- 
 cient motive. This, again, is possible because the virtuous 
 man means simply the man who corresponds to the best 
 social type, and will therefore act on all occasions in con- 
 formity with the character so defined. It is not because all 
 virtuous actions have one definite end, which end is itself 
 something different from virtue, but because all virtuous 
 action implies action in conformity with certain instincts 
 which have become organic in the virtuous man. To say, 
 then, to such a man, " This is right," is the same thing as to 
 say, "This satisfies your instincts;" or, in other words, it 
 states a sufficient inducement to the corresponding action. 
 
 23. The utilitarian statement represents this by defining 
 moral conduct as that which has for its end the production 
 of happiness, irrespectively of its quality. To love virtue 
 for its own sake would be to love the means irrespectively 
 of the end, which would be clearly irrational, though not 
 impossible. But upon the view of the egoistic utilitarian, 
 it seems that all truly virtuous conduct must be placed in 
 this predicament. He explains the origin of the benevolent 
 impulses through association in such a way as to destroy
 
 378 THE SCIENCE OF ETHICS. 
 
 their reality. The typical example is that of the miser, who, 
 from loving money as a means for promoting pleasure, comes 
 to love it as an end, and even when possession implies a 
 sacrifice of pleasure. Similarly, it is suggested, we associate 
 pleasant sensations with certain persons, and we love them 
 as the cause of those sensations. If, however, love thus 
 explained should prompt us to act in such a way as to 
 sacrifice our pleasure for the good of others, we should be 
 unreasonable in the same sense as the miser. We should be 
 applying a rule in a case where it was plainly inapplicable, 
 and using means for an end in a case where we knew that 
 they would not produce that end. Association in this sense 
 implies illusion ; and the more reasonable we become, the 
 more we should deliver ourselves from the bondage of such 
 errors. 
 
 24. This view seems to include an imperfect statement of 
 one side of the truth. I cannot admit its accuracy, indeed, 
 even in regard to our old friend the miser. As a rule, it 
 seems to me, the man who really desires money as a means 
 to enjoyment is more likely to become a spendthrift than a 
 miser. He will be so eager for present enjoyment as to 
 neglect provisions for future enjoyment. The miser, as I 
 should say, is in the normal case a man who desires money 
 in order to guard against the misery of poverty and depend- 
 ence. His motive is not the positive desire for pleasure, but 
 its negative side, the dread of actual suffering. When his 
 desire for money becomes excessive, he is still guarding 
 against the same evil, though he is taking exaggerated pre- 
 cautions. He resembles a man who is afraid of falling over a 
 cliff', and who therefore will not go within a hundred yards, 
 though in fact he would be equally safe within a yard of the 
 edge. He reasons badly, therefore, in so far as his terrors are 
 extravagant. But the motive is not necessarily changed in 
 character. There is always a danger of loss and poverty, though 
 the danger may be so small that a wise man would not consider 
 it. The motives of the miser may be changed in other direc- 
 tions. He loses all relish for pleasures which he has never
 
 EXTRINSIC MORALITY. 379 
 
 allowed himself to ciijoy,and many tastes havetbus been stifled 
 in the germ. And, on the other side, the various activities 
 necessary for the acquisition and preservation of money 
 have become pleasant to him, as any mode of activity v^hich 
 gives room for skill, forethought, and a discharge of energy 
 may be really pleasant. There is, for example, a pleasure in 
 skilful speculation, as there is a pleasure in the game of chess, 
 abstractedly from any consideration whatever of an ^^end." 
 It is only when we assume that all activity is conditioned by 
 the prospect of the pleasure attainable at the end that we 
 are forced into the supposition of a confusion between means 
 and ends. Every kind of activity has its own pleasure, as 
 affording a means for the discharge of energy or escape from 
 ennui. The miser who finds pleasure in the act of money- 
 making need not be unreasonable even if he proposes to 
 make no use of the money. The money-making is in itself 
 pleasant, and pleasant because it is a regulated mode of 
 activity which gives occupation to a variety of faculties. 
 
 35. Now if we compare this case with that of benevolent 
 action, we may see what is really implied. It is doubtless 
 true that we may learn to love our neighbours because they 
 have contributed to our pleasures. The child, in whom sympa- 
 thy and intelligence is still dormant, may thus regard its mother 
 as the source of a2:reeable sensations. More ffenerallv, as I 
 have argued, moral progress may proceed in an analogous way. 
 We mav become interested in the welfare of our neio-hbours 
 
 O 
 
 because our interests are identical ; and this observation mav 
 convince us that the identity extends to cases in which the 
 direct sympathy would not dictate benevolent action. When 
 this community of action has once been established in any 
 degree, there is room for the play of sympathy. The happi- 
 ness of others, so soon as we can appreciate it, becomes an 
 end in itself; but this sympathy is enabled to expand because, 
 previously to its existence, the rule of conduct which it pre- 
 scribes has already been adopted from different motives. The 
 external rule then passes into the internal, and genuine and 
 intrinsic moral motives become possible. Hence the associa-
 
 38o THE SCIENCE OF ETHICS. 
 
 tion theory has a certain truth within its province, but it 
 does not by any means account for the whole phenomenon, 
 and it fails just at the point where true morality begins: it 
 may explain, that is, why we first take a pleasure in the 
 welfare of others; it may in some cases even explain why, 
 though still selfish, we act unselfishly, and so far, in this case, 
 unreasonably. It is possible, too, that, as in the case of avarice, 
 the various kinds of activity called forth by conduct which 
 does good to others becomes in itself pleasant. A man, for 
 example, may learn to take pleasure in surgical operations, 
 though without any genuine sympathy for his patients; he 
 may even be externally unselfish to the point of heroism, and 
 sacrifice his life to his pursuit, as another man might sacrifice 
 his in the pursuit of some purely selfish and even degradino; 
 pleasure. The conditions of life, which force every member 
 of a society to conform in some degree to the interests 
 of others, and which therefore involve a considerable con- 
 formity between the egoistic and altruistic sentiments, may 
 for that reason generate a kind of fictitious benevolence, a 
 pleasure in conduct which has in fact good results to others, 
 even though the contemplation of those results affords no 
 pleasure. Upon this view, however, sympathy is so far 
 unreasonable. Directly it prompts to self-sacrifice we are the 
 slaves of sophistry or a misleading association. Upon my 
 theory, on the contrary, sympathy is a real motive implied in 
 all true morality, though it can only operate as some germ of 
 reason becomes developed, and begins to operate within the 
 line already laid down by the non-sympathetic customs. 
 
 V. Expedienaj. 
 
 26. The utilitarian, therefore, who is also an egoist, docs, as 
 I hold, deprive morality of its essential meaning, and the as- 
 sumption that the principle of association explains all modes of 
 thought and feeling lends itself to that mode of regardino; the 
 facts. The utilitarian, however, is more naturally a genuine 
 believer in altruism : that belief fulls in with the theory that
 
 EXPEDIENCY. 381 
 
 the criterion of morality is the tendency of conduct to promote 
 happiness, for this tendency then corresponds to a genuine 
 motive. We have, however, still to inquire whether the 
 morality thus constructed has sufficient stability, or whether 
 it may not be rijihtlv condemned as encourao"ina: mere "ex- 
 pediency." This is perhaps the most frequent line of attack 
 upon utilitarian systems. Tiie force, indeed, of the criticism 
 depends to a great extent upon the assumption that expe- 
 diency is another name for selfishness. When a man breaks 
 a particular moral rule — that of truthfulness, for examj)le — 
 because he thinks that a lie will do more good than harm, we 
 accuse him of acting upon principles of expediency. But 
 there is certainly a wide moral difference between the cases 
 in which the lie is prompted simply by the consideration that 
 a lie will be useful to the agent, or that in which it is 
 prompted by the genuine belief that it will increase the 
 general happiness. We are apt to confound the two cases, 
 and to saddle the man who breaks the rule upon any con- 
 sideration with all the blame due to the selfish consideration. 
 The inference may be frequently correct in fact; expediency 
 may, in point of fact, be generally used as a cloak for selfish- 
 ness; or, which is probably nearer the truth, a man who 
 allows himself to break rules for the good of others may be 
 strongly tempted to break them for his own private good at 
 the expense of others. Still, a man who should really act 
 upon the altruistic version of the utilitarian criterion might 
 be shifty and unreliable in his conduct, but could not pro- 
 perly be called selfish. His conduct would generally be 
 moral, thou<rh he mioht diverge in certain cases from the 
 moral law. 
 
 27. What, then, is implied by the supremacy of the 
 moral law, or autonomy of the conscience? If there is any 
 rule of conduct which can be laid down as absolute, it must 
 at once decide the morality of the conduct affected in every 
 conceivable case. If we admit of any criterion outside of 
 morality, it must limit the sway of the moral law. If the 
 true test be the utilitarian, then we nmst either admit that
 
 382 THE SCIENCE OF ETHICS. 
 
 lying, for example, is right when it increases happiness, or we 
 must prove that lying never increases happiness. The utili- 
 tarian admits that any such proof is impossible. There 
 may be cases in which a lie will do more good than harm ; 
 therefore " Lie not" cannot be an absolute law. He says, 
 indeed^ that we are forced to act for the most part upon 
 probabilities, and that experience shows probability to be 
 enormously in favour of truth-telling. But the question occurs, 
 Why should we act upon this general probability without 
 reference to the particular circumstances ? The general rule 
 is not properly " supreme" over the particular, and it is hard to 
 see how it can be said to override it. Take any ordinary case 
 of prudence. I know that, as a rule, it is imprudent, sav, to 
 trust monev to a stranger. That is a sufficient guide for me 
 if the only thing that I know about a man is the bare fact 
 that he is a stranger. It holds, ao-ain, so lonf as I know 
 nothing to take him out of the class as an exceptional person. 
 But directly I do know something which takes him out of 
 that class, I must, if I am reasonable, modify my judgment 
 accordingly. The only rule that can be given is to act upon 
 the whole probabilities of the case. This, of course, includes 
 a reference to the general probability which arises from the 
 fact that the man is a stranger, but it does not give any 
 absolute predominance to the rule. I must take it into 
 council, but not obey it as a master. My ground of action in 
 any particular case must be my judgment of the probability, 
 which will be guided partly by reference to the general rule 
 that strangers are not to be trusted, but also by reference 
 to any other rule which may tend to show that this particular 
 strano-er is trustworthy. I have then said evervthino; when 
 I have said, "Act upon the probabilities," for that of course 
 includes a reference to every applicable principle of judgment. 
 Why should I not say the same in moral questions ? Should 
 not the rule be, " Act in that way which will probably pro- 
 duce the greatest happiness," where the " probably " of course 
 includes a reference to general induction from similar cases 
 already observed, but regards it only as an induction, not
 
 EXPEDIENCY. 383 
 
 as an infalible test? And if this be so, how can the moral law- 
 be regarded as absolute and supreme? It is nothing but a 
 useful collection of precedents tending to guide our judgment 
 in particular cases. 
 
 28. This, it is felt, is a dangerous principle. 7"he utilitarian, 
 however, diminishes the importance of his attack upon the 
 supremacy of the moral rule by obvious considerations; that 
 is to say, he insists upon the extent of our ignorance as to the 
 consequences, and upon the danger of trusting ourselves. 
 The pleas must be admitted to have considerable force ; but 
 they seem to leave the difficulty in some undefined degree. 
 For, in the first place, however great the importance of trust- 
 ing to the general rule, it cannot be denied that we are able to 
 make some estimate of the consequences, or otherwise we 
 could not make the necessary induction ; and thus, again, it 
 must be admitted that, after making all allowance for our 
 ignorance, the question is still essentially one of probabilities, 
 and the presumption in favour of the beneficial consequences 
 of certain particular lies may be enormous. And, in the next 
 place, the danger of trusting ourselves is undoubtedly very 
 great; and yet it is a danger from which we cannot shrink. 
 The danger is, in fact, a danger of misinterpreting the prece- 
 dents which we have set. The general rule must be, not that 
 lies are immoral, but that mischievous lies are immoral. Why, 
 then, should it be supposed that we should be more in danger of 
 tellins; a mischievous lie because we have told a beneficial lie? 
 If we do so, we are fallina; into an error of reasoninc;. We 
 are holding fast to the accident and dropping the substance. 
 Why should not the fact that we have acted for the best in 
 one case determine us to act for the best in another, although 
 in one case the best action accidentally involved a lie, and in 
 the other accidentally involved truth-speaking? Generally, if 
 the promotion of the general happiness is the sole criterion, 
 why should not the desire for general happiness be the sole 
 motive for moral conduct? If so, can we properly consider 
 moral rules as anything more than formulae, useful in calcula- 
 ting the probable consequence of conduct to happiness, but
 
 384 THE SCIENCE OF ETHICS. 
 
 always liable — like any other empirical formulae — to be over- 
 ridden by special circumstances ? 
 
 29. I cannot, upon my own hypothesis, meet this by de- 
 claring the particular moral laws to be absolute and supreme. 
 The attempt to do so only leads to casuistry of the objection- 
 able kind. Make the rule against Iving absolute, and vou 
 find yourself obliged to bring the particular cases under a 
 different rule by some device analogous to a legal fiction, and 
 more demoralising than an open declaration that the moral 
 law is not to be always observed. Nor can I accept the 
 statement in the form that the contingent consequences of 
 an action must always be set aside. It seems to me that the 
 known consequences of an action must always be relevant to 
 its morality. If I were absolutely certain that a lie would do 
 goodj I should certainly hesitate before speaking the truth, 
 and the certainty might be of such a kind as to make me 
 think it a dutv to lie. It has been said bv moralists that a 
 good man would not commit the most venial sin, even though 
 the consequences of his virtuous action were the perishing of 
 all mankind in torture. The statement is either shockino- or 
 meanino-less. I can conceive of no action which would not 
 become an abominable crime by the simple fact of its entail- 
 ing such consequences, and of no crime which would not 
 become an imperative dutv if it evaded them. A man who 
 saved his dirty soul at such a price would deserve damnation 
 ten times over; and if he did not get it, I should not regard 
 his deity as a truly moral agent. If, on the other hand, the 
 act still remained virtuous, it could only be because we 
 assume, for some reason or other, that the moral action has 
 other good consequences — say the eternal happiness of the 
 agent — which overbalance the evil consequences. In that 
 case the utilitarian position is not really denied. 
 
 30. The assertion that the particular moral rules are not 
 absolute is a repetition of a principle already laid down. It 
 is impossible to secure more than an approximate coincidence 
 between the external and the internal law. However closclv 
 they may, and frequently do, coincide, it is always (or almost
 
 EXPEDIENCY. 385 
 
 always) possible to suggest cases in which they diverge. 
 Wherever such a deviation occurs, the true moral rule must 
 be the internal rule. Morality is essentially a determination 
 of character, and hence wherever adherence to a fixed rule 
 implies a change of character, or, inversely, the persistency 
 of character implies a breach of the rule, the virtuous man 
 will break the rule. Hence, again, the attempt to secure an 
 absolute and immutable moral law in its external shape 
 must be illusory. The moral law can be stated uncondition- 
 ally when it is stated in the form "Be this," but not when 
 stated in the form "Do this." We may say without any 
 qualification whatever that the good man must be merciful, 
 just, truthful, temperate, courageous, and so forth ; but for 
 the very reason that the law is in this sense absolute, it can- 
 not be absolute as prescribing external modes of conduct. 
 Hence, again, whatever the difficulty of deciding in particular 
 cases what is the rio;ht sense of conduct, we have a definite 
 principle, and therefore a fixed rule, though not always an 
 easily discoverable rule. The truly virtuous man is the typical 
 man whose character conforms to the conditions of social 
 vitality. The question what is right is in all cases equivalent 
 to the question what the right-minded man would do, sup- 
 posing him of course to be fully informed, and to reason cor- 
 rectly from the facts. In nine hundred and ninety-nine cases 
 out of a thousand he will act in conformity with the ordinary 
 moral rule, say, of speaking the truth. We may say absolutely 
 that the moral man must be thoroughly trustworthy. Wher- 
 ever trustworthiness implies truth-telling, he will tell the truth ; 
 but if, in any particular case, the same character implies 
 lying, he will lie, and will lie like a man, that is, without 
 trying to conceal from himself by any shuffle of casuistry 
 the true nature of his conduct. If in this sense morality in- 
 cludes an element of expediency, not, that is, in the sense of 
 denying that there is a rule of conduct in this case as in all 
 others, but in the sense that the rule may fail in certain 
 cases definable in principle, but not precisely ascertainable 
 
 by any definite external indications. Normally, any deflec- 
 
 a B
 
 386 THE SCIENCE OF ETHICS. 
 
 tion from the rule implies a deficiency in some good quality. 
 In certain very rare cases it implies the presence of good 
 qualities. In the latter case the conduct is right, as in the 
 former it is wrong. 
 
 31. To apply this more specifically to the problems which 
 actually present themselves would be to write a treatise upon 
 casuistry ; but one or two remarks may clear the general 
 principle. It must be observed, in the first place, that many 
 so-called problems of casuistry are not properly moral pro- 
 blems at all. In many cases we must admit that the mor- 
 ality of a given course of conduct depends upon its tendency 
 to produce happiness. Ceteris paribus, I must do that which 
 will give most pleasure to others. But, again, to decide 
 what will give most pleasure to others is often a very diffi- 
 cult quci^tion, which admits of being decided only by tact 
 or instinct. We must judge in most practical cases of any 
 difficulty by probability, or, in other words, by guessing. 
 When we really believe that we shall do most good by a 
 certain course of conduct, that conduct is so far right ; and 
 here the moral rule is plain, however difficult the application. 
 Again, many casuistical problems turn upon such questions 
 as the nature of some implied contract. It is a question of 
 law what rights I have created by marrying, or accepting 
 an office, or making some commercial bargain. Morality in 
 such cases may simply order me to keep my contract, what- 
 ever it may be, and the " whatever " is to be decided by the 
 lawyers. And this, again, leads to another distinction often, 
 as I have said, overlooked. The problem is frequently stated 
 ambiguously, and may either be, " Does the actually existing 
 standard of morality forbid a certain action? " or it maybe, "Is 
 it desirable that the action should be forbidden by morality? " 
 The moralist, like the lawyer, is apt to conceal the fact that 
 he is reallv making new laws under the pretext of enlarging 
 the old. In reality, he is trying to frame a more precise rule 
 than the one which has hitherto been accepted. It would 
 be better frankly to admit the clear nature of the case, be- 
 cause it would then be easier to argue upon the principles to
 
 EXPEDIENCY. 387 
 
 be applied. Upon my theory, the method is dcfmcd by the 
 doctrine already laid down. A certain kind of conduct, we 
 suppose, has hitherto been regarded as indifferent. There is 
 no established rule for assuming whether a good man would 
 or would not act in that way. To decide, then, whether the 
 conduct is right or wrong is virtually to add a new moral 
 precept; and that is implicitly to inquire whether the adop- 
 tion of the new principle would, on the utilitarian showing, 
 do more good or harm ; or, more accurately, according to me, 
 to ask whether the incorporation of this rule in the moral 
 system would imply a closer approximation to the solution 
 of the great problem of social utility. Of course, the ques- 
 tion of the affinity of this proposed rule to previously accepted 
 principles will throw light upon the whole problem; but we 
 need not conceal the fact that we are really trvin"- to extend 
 and amend these principles, and not sim pi v judging new cases 
 by the old principle. Here, therefore, the ultimate difficulty 
 is still a question of fact — whether, namely, social develop- 
 ment implies the adoption of the new rule sugcrested ? 
 
 32. We may now come to the proper casuistical questions 
 where there is a real difficulty in interpreting the law. If it 
 is ever right to tell a lie, when is it right? What are the 
 principles by which we must judge? The utilitarian prin- 
 ciple seems to suggest the answer, Whenever the probabilities 
 are in favour of a lie doing more good than harm. The 
 morality of this answer depends partly upon the question, 
 What are the "consequences" to be admitted as relevant? 
 May I take into account in my calculation the pain which 
 the bare telling of a lie gives to a man accustomed to speak 
 the truth ? Or shall we say that as this pain depends upon 
 a fallacy of association, we ought to accustom ourselves to lie, 
 and so to find lying painless, whenever we see a balance of 
 good consequences in its favour? There can be no doubt 
 that such a principle strikes most people as immoral. The 
 actual standard of morality, whatever its justification, con- 
 demns lying in almost all cases, and only admits of doubtful 
 exceptions w^here the balance is seen to be overpoweringly in
 
 388 THE SCIENCE OF ETHICS. 
 
 favour of a lie, such as the famous case of giving informa- 
 tion to a murderer. It seems, then, that in this case the 
 average moral judgment does not conform precisely to the 
 utilitarian standard. It condemns useful lies, though we 
 may admit that there is a distinction between the cases of a 
 lie which is useful to the ao-ent and that which is useful to 
 
 O 
 
 the world at large. The former kind of lie would, of course, 
 be condemned by the utilitarians, in cases at least where the 
 good to the agent is counterbalanced by evil to others; 
 whilst, in regard to the latter, the moral standard would seem 
 to be rather unsettled, and people are generally inclined to 
 evade a solution by denying that the case ever occurs, or by 
 saying that it occurs so rarely as not to be worth ' consi- 
 deration. There is a strons; feeling that it is dancrerous to 
 consider such facts closely, as an admission of a possibility 
 of exceptions tends to increase the freedom of making 
 exceptions, 
 
 ^^. Now I certainly think that the utilitarian principle 
 does not explain the whole case. The moral rule is formed 
 in a great degree, as I have argued, by a utilitarian method; 
 that is, by observing the bad and good consequences of certain 
 kinds of conduct. But morality includes more than this, as 
 the moral sense is the product of the whole social development, 
 and therefore of the development of social instincts by other 
 processes than that of direct calculation. The whole system 
 of instincts must be such as are implied in a healthful social 
 growth, but the sentiment may justify itself by being actually 
 useful instead of being generated from a perception of utility. 
 Hence, as I think, the true analogv must be found elsewhere. 
 A kind of moral sentiment, as I have said, grows up in every 
 social organisation. The patriotic spirit of a citizen or the 
 military spirit of the soldier is closely related to the moral 
 sentiments which discharge a similar function in tlie " social 
 tissue." Let us take, for example, the striking case of the 
 soldier's devotion to his regiment or army. The essential 
 condition of military excellence is the spirit of subordination. 
 The ideal soldier — the man who realises most fully the condi-
 
 EXPEDIENCY. 389 
 
 tions of military life — is one who obeys orders implicitly. The 
 code of mllltarv duty includes, therefore, as Its first principle, 
 the great rule, "Do as you are bid." In almost every case, 
 disobedience to orders proceeds from laziness, cowardice, want 
 of patriotic spirit, or some quality which so far makes a man 
 a worse soldier. And hence, again, the rule is put forth 
 absolutely; and no commander would endeavour to Impress 
 upon the mind of a subordinate the importance of sometimes 
 disobeying orders. And yet it is plain that questions may 
 arise in this capacity which are precisely analogous to the 
 ordinary problems of casuistry. No one will deny that the 
 importance of implicit obedience is founded upon the directly 
 utilitarian consideration that obedient soldiers make a better 
 armv ; but cases occur in which the implicit obedience 
 produces a bad effect. When, for example, the Roman 
 soldier stood at his post at Pompeii to be overwhelmed by 
 the eruption, his heroic devotion was so far a source of weak- 
 ness to the armv. Had he been less devoted, the army would 
 have had one brave soldier (though not quite so brave a 
 soldier) the more. Or if we call him stupid, we must then 
 admit that superior Intelligence would lead him In this instance 
 to break a law unconditionally laid down. Take, again, the 
 opposite case. Nelson won several victories by direct insub- 
 ordination, and is therefore a rather awkward precedent for 
 the people who like to tell us that to learn to command 
 we should first learn to obey. Nelson clearly failed in his 
 dut)'-, and had he failed in his venture would probably have 
 been punished. Was he right or wrong? If wrong, it is 
 plain that he would in fiict have been a less efficient com- 
 mander for having a stronger sense of duty; and if right, 
 we must admit that a man may be right sometimes to take 
 the law into his own hands. The most reasonable view 
 is probably that Nelson, being really a man of genius and 
 conscious of his genius, was justified in acting as he did. He 
 was "justified," that is, in this sense, that his action proved 
 him to be a more efficient commander on the whole, although 
 his case cannot be made into a precedent, because we cannot
 
 390 THE SCIENCE OF ETHICS. 
 
 say what constitutes a man of genius or suggest any satis- 
 factory test of the presence of genius. 
 
 34. Such cases may of course be multiphed indefinitely, 
 and they pass imperceptibly into moral problems. The prin- 
 ciple upon which they rest is apparently simple. It is essen- 
 tial, that is, to the welfare of the army, that a certain spirit 
 of discipline should be generated. The stronger it is, within 
 certain limits, the more efficient is the army. An army 
 usually composed of such men as the Pompeian sentinel 
 would be invincible by any ordinary foe. In certain special 
 cases the sentiment thus stimulated produces the very result 
 against which it is a safeguard in the normal case. A good 
 soldier sticks to his post or his colours when the post is useless 
 and the colours represent nothing but so much rag. Still, 
 as a rule, the danger is entirely on the other side, and we 
 strengthen the motives to obedience as much as possible, 
 leaving it to the occasional man of genius to break rules on 
 his own responsibility. We tacitly admit that a breach of 
 rules mav be right in certain very rare cases, but they are 
 cases which it is impossible to define by any sufficient criterion. 
 They must spring from zeal, not from the absence of zeal, 
 but they are only justifiable by self-conscious genius, w^hich 
 unfortunately has a strong outward resemblance to self- 
 deceiving folly. Finally, if we can reckon upon more intel- 
 ligence In the man, we may allow a greater latitude. We 
 recoo-nlse the danger of maklmr the soldier too much of a 
 machine, and we try to allow for a greater latitude of indi- 
 vidual action. This, indeed, does not alter the general prin- 
 ciple of obedience, but only makes the particular rules less 
 precise; still, as stimulating the tendency of independent 
 judgment, it probably lowers the absolute character of the 
 disciplinary rules; the obedience is no longer blind, but 
 includes a more or less explicit reference to the end for which 
 it was imposed. 
 
 ^^. It seems to me that the same principle really governs 
 the moral case. The absoluteness of the moral law does in 
 fact rest upon the principle alleged by the utilitarian, namely,
 
 EXPEDIENCY. 391 
 
 that men cannot be trusted, and should not trust themselves. 
 They could trust themselves, perhaps, if thev always reasoned 
 rightly; but then people do not reason rightly, and still less, 
 if we may use the phrase, do our emotions reason rightly. 
 The fact may be regretted, but it is a fact which it is idle to 
 neglect in morality, which is meant to bind men as they are, 
 not for men of some ideal but hitherto unapproachable state. 
 The utilitarian principle of acting in every case from a judg- 
 ment of the probable consequences might answer at a time 
 when we could also allow every soldier in our army to act 
 upon his own judgment as to running away and fighting, 
 because we could trust each man to see when the good of 
 the army required him to run, and to run only when required 
 by the good of the army. Till that time arrives we cannot 
 grant a liberty which is liable to abuse. There are moral 
 precepts which clearly await such a process. Why, for 
 example, do we not allow life to be taken when life means 
 continued agony? We cannot allow it because we cannot 
 trust the persons who would have to apply it. The physician 
 makes it an absolute rule to preserve life as much as he can, 
 because, if he once admitted any other consideration, he would 
 open the door to innumerable abuses. That is a plain, and, 
 as I think, an amply sufficient reason for maintaining the 
 rule, and so far deviating from the utilitarian method, whicli 
 would prescribe calculation of the probabilities for each par- 
 ticular case. We cannot, in fact, keep up the value of 
 human life without fostering what is in some sense a super- 
 stitious regard for life — a tenderness for life when the desira- 
 bility of life has disappeared, and so far a kind of conduct 
 which may be called unreasonable. I do not inquire whether, 
 as a matter of fact, this sentiment is not encouraged to excess 
 at the present day; but the justification depends upon this 
 principle, and, so far as I can see, upon this principle alone, 
 unless we resort to arbitrary hypotheses to cover an unrea- 
 soned prejudice. 
 
 ;^6. If, now, we suppose that a man, knowing that life 
 meant for him nothing but agony, and that moreover his life
 
 392 THE SCIENCE OF ETHICS. 
 
 could not serve others, and was only giving useless pain to 
 his attendants^ and perhaps involved the sacrifice of health to 
 his wife and children, should commit suicide, what ought we 
 to think of him ? He would, no doubt, be breaking the 
 accepted moral code; but why should he not break it? 
 Because he is setting a bad example? Undoubtedly that 
 must be allowed to have some weight. If his conduct tended 
 to depress the general estimate of the value of life below the 
 point which is necessary to social welfare, he is so far setting 
 a bad precedent — bad, that is, in so far as it is likely to be 
 abused. In conforming to the law he is making himself a 
 martyr to the general welfare, and deserves the praise of 
 unselfishness. But if, again, he knew that his conduct 
 could not have this effect, as in a case where he might 
 be quite certain that his action would remain unknown, 
 what harm would he be doing? May we not say that he is 
 acting on a superior moral principle, and that because he is 
 clearly diminishing the sum of human misery? It is impos- 
 sible to settle the case in concrete instances, because there is 
 no fixed external test. The conduct may spring either from 
 cowardice or from a loftier motive than the ordinary, and the 
 merit of the action is therefore not determinable ; but, assu- 
 ming the loftier motive, I can see no ground for disapproving 
 the action which flows from it. 
 
 37. Briefly, then, if conduct be such as to increase the 
 actual sum of happiness, if it does not imply a defect of 
 altruism in the agent, and if it be such as not to set a bad 
 precedent either to ourselves or to others, I do not see in 
 what sense it is morally blameworthy. To adhere to the 
 rule, when the rule clearly does not apply, is not to be moral, 
 but to be a moral pedant. The truth is simply that the race 
 must form its moral code in the same way as we have to 
 stimulate the sense of discipline in an army; that is, we have 
 to deal with human beings who cannot be trusted ; we have 
 to encourage modes of feeling which, though generally of the 
 highest value, do not point infallibly to that kind of conduct 
 which is most productive of happiness; we have to take
 
 EXPEDIENCY. 393 
 
 human nature in the rough, and to give external rules which 
 do not formally and at all points correspond to the judgment 
 which would be passed by the most intelligent and sensitive 
 of the species. So far as the utilitarian mode of reasoning 
 tends to obscure this essential fact, it may be rightly con- 
 demned because liable to abuse; but it is possible to interpret 
 the utilitarian principle differently, and I regard it as in any 
 case an approximation to the truth, though tending perhaps 
 to a certain laxity in practice, in so far as it assumes an im- 
 practicable nicety in the application of the criterion. 
 
 38. In what sense, however, can the theory here stated be 
 properly stigmatised as a doctrine of expediency ? To act 
 rightly, I say, is to act as the truly moral man would act; 
 the truly moral man being defined by his fulfilment of the 
 conditions of social welfare. He has, therefore, certain 
 qualities which may be called absolutely good ; that is, that 
 they are good on the sole assumption of a certain social and 
 individual organisation. They imply, again, obedience to 
 the established moral standard in almost every case. The 
 thoroughly trustworthy man will speak the truth all but in- 
 variably. But in certain rare cases, the internal and external 
 laws cease to coincide absolutelv. Now in such cases the 
 ideal or thoroughly moral man will break the external law, 
 and break it in obedience to his instinct. He must trust his 
 instincts, for he has nothing else to which he can trust. If, 
 in fact, the exceptions to the law admitted of being definitelv 
 formulated, we should have the case of a new moral law. The 
 rule would no longer be, " Speak the truth," but " Speak the 
 truth under such and such circumstances." We miirht then 
 have some difficult questions as to the rightness of obeying 
 the recognised law, or obeying the law which is still strug- 
 gling for recognition. On our hypothesis, the difficulty is 
 that the problem has to be solved without any fixed test to 
 go by. Still the difficulty is essentially the same. The man 
 has to act upon his instinctive perception that in this case 
 the moral law is defective. But to say that he must trust 
 his instinct is not to say that whatever his instinct tells
 
 394 THE SCIENCE OF ETHICS. 
 
 him will be right. It is simply that in certain cases the 
 refined and well-informed instinct outruns the definite and 
 assignable formula. He is the final court of appeal, but not, 
 therefore, above the law. The ear of the musician may- 
 judge of discords and harmonies too fine to be measured by 
 mathematical instruments. He will accept the general rules, 
 but will apply them with greater severity than any mechani- 
 cal contrivance can sufficiently test. In the same way, the 
 man of fine moral taste has to solve problems too delicate 
 for the coarser rules which serve for everyday life ; not be- 
 cause he despises those rules, but because he has a keener 
 sense of their true value. Some of the noblest actions re- 
 corded have generally been where a man has broken with 
 the accepted code of steadygoing respectable moralists ; but 
 the Tightness or wrongness of his conduct is still amenable 
 to a test, which, though only implicit, is as valid and fixed 
 as any other. The exception to the law must always itself 
 imply a law. It must embody some general principle, though 
 the principle cannot be absolutely laid down in particular 
 cases j for the general principle, briefly stated, must be this, 
 namely, that the conduct in question would be commanded 
 by a higher code of morality, and by '' higher" I mean such 
 a code as would be obtained by a general increase of social 
 efficiency, and a closer approximation to an adequate solu- 
 tion of the great problem of life. The problem, therefore, 
 just noted must always occur, namely, whether I ought to 
 act upon a code which is not yet recognised by the accepted 
 morality, and which cannot be recognised on account of 
 human stupidity and insensibility. That question, again, 
 can only be answered by taking into account the total 
 effect, and asking whether I am doing more good by my 
 action or more harm by setting a precedent capable of 
 misinterpretation. 
 
 39. There is, then, only one test for such problems. Does 
 a given deviation from the law imply an advance or a decline 
 in the stage of moral development? Does it mean that the 
 qualities which imply conformity to the moral law have not
 
 EXPEDIENCY. 395 
 
 been sufficiently organised in the individual, or that they 
 have become a part of his nature so thoroughly that a still 
 finer instinct has been generated which transcends him ? 
 That is a question of fact, which it is generally pretty easy 
 to answer. Most conduct which offends the average standard 
 proceeds from some vicious motive. But we cannot deny 
 that there are examples to the contrary; and we must admit 
 this freely in spite of any taunts as to expediency, which 
 really confound the two cases or overlook the undoubted 
 importance of the great fact upon which the importance of 
 every kind of general rule of conduct really rests, namely, 
 that men are in fact too stupid, too wilful, and too skilful 
 in self-deception to be trusted to act from more refined 
 considerations. 
 
 40. The answer suggests one other difficulty. Let us 
 grant that the moral man will sometimes break a law 
 which it is vitally important to impress upon the ordinary 
 man. Is not this to make two rules, one for the good and 
 one for the bad ? Rather it is to assert that the ordinary man 
 cannot perform the same action as the good. If the same 
 internally, it is not really the same because it does not imply 
 the same motive. But to make a general rule of conduct is to 
 make a law which can be brought to bear upon everybodv. 
 A moral law, as here stated, seems to be simply a description 
 of the way in which certain people will act, and therefore 
 to have no meaning for others. This introduces, therefore, 
 the problem which we now have to consider. What is the 
 relation between morality and happiness? What, in the 
 utilitarian language, is the general sanction of morality ? A 
 man may say, "That is doubtless 'right,' or, in other words, it 
 is what a virtuous man would do ; but I happen not to be a 
 virtuous man; why should I do it? " How are we to answer 
 him ?
 
 ( 396 ) 
 
 CHAPTER X. 
 
 MORALITY AND HAPPINESS. 
 
 I. The Sanction. 
 
 I. I HAVE thus considered the problem, What is the cri- 
 terion which, upon the evolutionist theory, must be substi- 
 tuted for the greatest-happiness criterion of the utilitarian ? 
 And this leads naturally to the problem how, upon the same 
 theory, must the utilitarian sanction be modified? Why 
 should a man be virtuous? The answer depends upon the 
 answer to the previous question, What is it to be virtuous? 
 If, for example, virtue means all such conduct as promotes 
 happiness, the motives to virtuous conduct must be all such 
 motives as impel a man to aim at increasing the sum of 
 happiness. These motives constitute the sanction, and the 
 sanction may be defined either as an intrinsic or an extrinsic 
 sanction; it may, that is, be argued either that virtuous conduct 
 invariably leads to consequencee which are desirable to every 
 man, whether he be or be not virtuous ; or, on the other 
 hand, that virtuous conduct as such, and irrespectively of any 
 future consequences, makes the agent happier. Some mo- 
 ralists say that a good man will go to heaven and a bad man 
 to hell. Others, that virtue is itself heaven and vice hell. 
 Now, I have already said that an answer which assigns a 
 really extrinsic motive is, to my mind, a virtual evasion of 
 the question. Yet the line between extrinsic and intrinsic 
 motives is perhaps not clearly drawn, and there are diffi- 
 culties in the assignment of a purely intrinsic motive which 
 require explicit consideration. The criterion which I have 
 accepted is, briefly, that a man is virtuous or the reverse so
 
 THE SANCTION. 397 
 
 far as he does or does not conform to the type defined by the 
 healthy condition of the social organism. We have, there- 
 fore, to consider what advantages are implied in the possession 
 or the acquirement of such a character, or the observance of 
 the corresponding rules of action. 
 
 2. The problem is thus to find a scientific basis for the art 
 of conduct. The "sanction" must supply the motive power 
 by which individuals are to be made virtuous. It is for the 
 practical moralist the culminating point of all ethical theory. 
 Now, according to my argument, the primary and direct incid- 
 ence, if I may say so, of moral sanctions is upon the social 
 organism, whilst the individual is only indirectly and second- 
 arily affected. There is (as I hold) a necessary and immediate 
 relation between social vitality and morality. We may say 
 unconditionally that healthy development implies an efficient 
 moral code, and that social degeneration implies the reverse. 
 But it does not follow that there is the same intimate con- 
 nection in the individual case. As, indeed, the society has 
 no existence apart from the individual by which it is con- 
 stituted, there must be a close connection ; but it must be 
 subject to various limitations, in so far as the conditions of 
 health and happiness of any particular person are not ex- 
 haustively defined by those of the society to which he 
 belongs. We cannot transfer to each member of the society 
 what we can say of the society regarded as a whole. It is 
 at least conceivable that the sacrifice of some of its members 
 may be essential to the welfare of the society itself. The 
 virtuous men may be the very salt of the earth, and yet the 
 discharge of a function socially necessary may involve their 
 own misery. A great moral and religious teacher has often 
 been a martyr, and we are certainly not entitled to assume 
 either that he was a fool for his pains, or, on the other hand, 
 that the highest conceivable degree of virtue can make 
 martyrdom agreeable. 
 
 3. Here, in short, we come to one of the multiform and 
 profound problems which has tortured men in all ages. 
 Virtue — no one denies it — does good to somebody, but how
 
 398 THE SCIENCE OF ETHICS. 
 
 often to the agent? A belief in justice as regulating the 
 universe has been held to imply (I do not ask whether 
 rightly held) that happiness should somehow go along with 
 virtue. To give up the belief in such a supreme regulation 
 seemed, again^ to be an admission that virtue was folly. 
 Yet how can this doctrine be reconciled to the plainest facts 
 of experience ? The lightning strikes the good and the bad ; 
 the hero dies in the ruin of his cause ; the highest self-denial 
 is repaid by the blackest ingratitude ; the keenest sympathy 
 with our fellows implies the greatest liability to suffering; 
 the cold, the sensual, and the systematically selfish often 
 seem to have the pleasantest lots in life. Great men in 
 despair have pronounced virtue to be but a name ; philoso- 
 phers have evaded the difficulty by a verbal denial of the 
 plainest truths ; theologians have tried to console their 
 disciples by constructing ideal worlds which have served 
 for little more than a recognition of the unsatisfactory 
 state of the actual world. The problem, so often attacked, 
 will perhaps be solved when we know the origin of evil. 
 Meanwhile, we have only to consider in what way it is 
 related to ethical theories. 
 
 4. One preliminary remark may help to clear the way. 
 Any sound proposition whatever about happiness must in- 
 clude some reference, tacit or explicit, to the constitution of 
 the agent. Happiness, whatever else it may imply, implies 
 a state of some conscious being; it must be conditioned by, 
 or, in the mathematical phrase, be a function of, the organic 
 state. We may, of course, discover conditions of happiness 
 common to large classes of organised beings, or even to 
 organised beings in general. Such a condition, for example, 
 would be asserted in the proposition (true or false) that 
 happiness implies a state of heightened vitality. Other 
 more special conditions apply only to particular classes. The 
 intellectual pleasures imply a certain degree of mental culture; 
 the pleasure of music implies a certain specific sensibility. 
 Such pleasures imply a faculty which is not common to all 
 men, but to all intellectual or to all musical men. Unless
 
 THE SANCTION. 399 
 
 therefore^ virtue cand happiness are related as merely different 
 modes of regarding the same quality, it is idle to attempt to 
 assign their relations by a simple inspection of the concepts 
 themselves. Whatever relation they may have must be 
 dependent upon the constitution of the agent. And though 
 it might conceivably turn out that the relation was such as 
 to be independent of any possible variations in the individual 
 constitution, we cannot make any such assumption at start- 
 ing. The presumption is in this, as in every other case, that 
 the relation of happiness and virtue will be dependent upon 
 the character of the agent. It may be true that the happi- 
 ness of a given class depends upon virtue, and that class may 
 be a very large one ; but we have no right to assume that 
 the class is identical with the whole human race; nor can 
 any such theory be established without an investigation suffi- 
 cient to rebut the natural presumption that some of the 
 class qualities are necessary for its existence. In fact, the 
 most obvious answer to the question, "^Does virtue coincide 
 with happiness ? " would probably be, "Yes, for the virtuous 
 man." As music gives pleasure only to men who have a 
 M musical ear, so it may be virtue gives pleasure to men with 
 . a conscience. If the answer is not generally given, it is not 
 because it is destitute of considerable plausibility, and, as I 
 think, of a great degree of accuracy ; but because it is calcu- 
 lated to shock many respectable people. But this is not a 
 philosophical consideration. 
 
 5. The remark is enough to exhibit the fallacy of a familiar 
 short cut to the desired conclusion. When attempting to 
 estimate the happiness of the virtuous, we may, it has been 
 said, confine ourselves to the judgment of the virtuous. The 
 intellectual man alone can judge of the value of intellectual 
 pleasures, whilst he can judge as well as the sensual of the 
 value of sensual pleasures ; he alone, therefore, can make a 
 comparison ; and as every one which can thus compare does 
 in fact (so it is asserted) prefer the intellectual, we are justi- 
 fied in accepting this authoritative decision. The argument, 
 like many other formally unsound arguments, has, I think,
 
 400 THE SCIENCE OF ETHICS. 
 
 some force in a shape to be presently noticed ; but in this 
 shape it appears to be hopelessly untenable. For, in the 
 first place, no judgment of pleasure proceeding by this 
 method of direct inspection alone can have much authority. 
 We are very bad judges even of our own pleasures, and we 
 have innumerable temptations to give a coloured judgment. 
 We may, therefore, always appeal from a man's avowed 
 sentiments to his practice ; and it can hardly be said that 
 men of the highest intellectual qualities always display a 
 relative indifference to the pleasures of the senses. Solomon 
 the preacher must be compared with Solomon the king. 
 Nor does it follow that if I can judge of my own pleasures 
 I can therefore judge of other men's pleasures. The fallacy 
 inherent in all such inferences is one of the most familiar 
 topics of experience. If I prefer Shakespeare to a mutton-chop, 
 I may say that I so far judge the pleasures of imagination 
 to be preferable for me to those of the senses. But how can 
 I leap from that proposition to the proposition that they are 
 preferable for others ? They are clearly not preferable for 
 the pig, or to the Patagonian, or even to those civilised men 
 who are in this matter of the pig's way of thinking. At 
 most, I may infer that certain cultivated minds find m.ore 
 pleasure in poetry than in eating, but still it does not 
 follow that the cultivated man finds more pleasure in poetry 
 than the sensual man finds in eating. The two men are 
 differently constituted throughout, and it may be that the 
 intellectual man has lost in one kind of sensibility what he 
 has gained in another. To assert positively that he gains on 
 the whole is to make an assumption often disputed, and of 
 which either side may be taken without self-contradiction, — 
 the assumption, namely, that an increased intellectual sensi- 
 bility necessarily carries with it increased power of enjoy- 
 ment on the whole; or, in other words, that it always answers 
 to cultivate your brain at the expense of your stomach. 
 
 6. There is yet another decisive objection. It is not even 
 true that any man absolutely prefers Shakespeare to a mutton- 
 chop. Rather, the phrase is nonsensical. Only an infant
 
 THE SANCTION. 401 
 
 compares his lov^e for his cousin with his love for jam-tart. 
 Shakespeare himself would at the right moment have pre- 
 ferred a cup of sack to the sweetest music or the loftiest 
 poetry. A starving saint might choose to eat a crust of 
 bread rather than listen to the most edifying sermon. Briefly, 
 a pleasure is not a separate thing, which has a certain con- 
 stant value, but a state of feeling which varies both accord- 
 ing to the permanent constitution and the varying condition 
 of that constitution. Each desire has a certain force, de- 
 pending upon the circumstances and character of the agent, 
 and there are conditions under which any particular desire 
 may, for the moment at least, become predominant. We 
 can only obtain absurd propositions when we attempt, in 
 discussing pains and pleasures, to abstract from the agent 
 who feels them. Nor can we evade the force of this argu- 
 ment by making a distinction between the quantity and 
 quality of pleasure. It is doubtless possible to give some 
 meaning to such phrases. We may, for example, distinguish 
 between pleasures as they affect different sensibilities, and 
 call those pleasures highest which require activity of the 
 intellect or the sympathetic emotions, and those lower which 
 excite only the senses. Or, again, it may be true that some 
 pleasures are massive and others acute; that some imply a 
 state of the whole organism, and others only the excitement 
 of a particular organ. But though this and much more 
 may be true, and for some purposes important, it does not 
 affect our present argument ; it does not enable us to say, 
 that is, that one pleasure is absolutely preferable to another. 
 Pleasures may indeed be compared in respect of pleasantness; 
 that is, we may say that, under given conditions, the desire 
 for one will overpower the desire for another. But each will 
 be alternately strongest as the conditions vary, and therefore 
 the formula which expresses their relations must always take 
 those conditions into account, and can never yield the state- 
 ment that one is always preferable to the other. This is 
 equally true whether we mean by preferable that which is 
 
 actually preferred, or that which ought to be preferred ; for, 
 
 3 c
 
 402 THE SCIENCE OF ETHICS. 
 
 upon any moral theory worth discussing, every natural 
 desire has its turn not only of preference but of rightful 
 preference. 
 
 7. I return, then, to the problem, which cannot be solved 
 by any such summary method, What is the relation between 
 virtue and happiness ? and I have to admit that it may be 
 a relation dependent upon the character of the agent. And 
 here we have to notice at once that there is an apparent 
 difficulty in our assumptions which necessitates a re-statement 
 of the question. Given the character, I have said, the 
 conduct follows ; the virtuous man acts virtuously and 
 the vicious man viciously. If we suppose, then, that the 
 character remains constant, there is apparently no sense in 
 asking whether it would make the vicious man happier to 
 act virtuously, for by acting virtuously he would become 
 virtuous, and this, by the hypothesis, does not happen. So 
 long, therefore, as we assume character to be fixed, the 
 question resolves itself into this, whether the virtuous man 
 is happier than the vicious man ? Assuming the external 
 circumstances to be the same for both, we have to ask how 
 far virtue is under all or under what circumstances a guaran- 
 tee for happiness ? But, in the next place, the assumption 
 of the fixity of character is of course arbitrary. A man's 
 character is in all cases the product of all the influences to 
 which he has been subjected from his infancy acting upon 
 his previously existing character ; and though more variable 
 in early life than afterwards, it is in every case undergoing 
 a continuous process of development. We have therefore 
 to ask whether the acquisition of a virtuous character — of 
 those instincts and modes of conduct which are prescribed 
 by the moral law — be in all cases conducive to happiness^ 
 or whether (as is conceivable) there are natures which are 
 made happier and others which arc not made happier by 
 such discipline ? We have, on the former hypothesis, to ask 
 whether a man with eyes is happier than a man without ; 
 and upon this, whether, supposing a man to have eyes, it is 
 always worth his while to cultivate the art of seeing. And,
 
 THE SANCTION. 403 
 
 thirdly^ we have to observe that conduct is not fixed, even 
 when character and circumstances are givcn^ unless we in- 
 chide under "circumstances" the particular set of thoughts 
 and feelings which are present in the agent's mind. Hence 
 we have to consider the problem (which^ indeed, seems to 
 be the most frequently discussed) whether we can discover 
 any sanction for the observance of the moral law as such 
 which would be equally operative upon all men irrespectively 
 of their moral character. To exhibit to a selfish man the 
 pleasures of sympathy is to suggest to him a motive for be- 
 coming virtuous ; we are virtually telling him to cultivate a 
 taste which he has neglected ; and this falls under the pre- 
 vious case. But we may also try to persuade him that even 
 upon purely selfish or prudential grounds he ought to do 
 good to his neighbour. In short, we may try to prove that 
 obedience to the outward rule answers, whatever the internal 
 motive. We have thus to ask three questions : first, whether 
 the virtuous man as such is happier than the vicious ; 
 secondly, whether it is worth while on prudential grounds 
 for the vicious to acquire the virtuous character; thirdly, 
 whether it can be worth while in the same sense for the 
 vicious man to observe the external moral law ? 
 
 8. Before applying myself to the argument directly, I will 
 make one remark. I think it altogether superfluous to dilate 
 upon the old test that honesty is the best policy. The point 
 has been so much laboured, that, although there is abundant 
 room for rhetoric, there is little need of argument. Con- 
 sidering how closely we are dependent upon our neighbours 
 and upon their good opinion, and the enormous difficulty of 
 retaining that good opinion without deserving it, there can be 
 no serious doubt that, on the average, every man will find his 
 account in observing the accepted moral code and acquiring 
 the corresponding instincts. To be on good terms with vour 
 family, to avoid picking and stealing, to be decently sober, in- 
 dustrious, and good-natured, are rules of conduct so obviously 
 expedient upon all grounds, that I will not burn daylight by 
 insisting upon them. I assume a general coincident between
 
 404 THE SCIENCE OF ETHICS. 
 
 the dictates of morality and of prudence, and refer readers in 
 need of further argument to the Hbraries of excellent treatises 
 extant upon this familiar topic, I will add that in the follow- 
 ing remarks I shall content myself, for the same reasons, with 
 pointing out what I take to be the true issues, without seeking 
 to estimate the weight of the appropriate evidence which may, 
 as a rule, be taken for granted. 
 
 II. Happiness and Virtue. 
 
 9. Assuming, then, that virtue and prudence approxi- 
 mately coincide, whilst, on the other hand, we cannot assume 
 that they are absolutely coincident, how much farther can 
 we go ? Can we fix with any greater precision the limits 
 of possible duration? Does the new form into which the 
 old aro;uments must be cast — for a change of form is all that 
 can be regarded as possible — help us to discuss the question 
 more effectively? Every writer is tempted to recommend 
 his own scheme by showing that it establishes a closer con- 
 nection between virtue and morality than the scheme of his 
 rivals, and there is therefore a very natural tendency to pass 
 too lightly over the less solid parts of ethical theory. We 
 feel that what is wanting in reason will be made up by the 
 goodwill of our audience. But if our aim is the discovery 
 of the truth, we must be more upon our guard against the 
 plausibilities of writers who wish to combine edification with 
 argument. It would be highly agreeable to find sound reasons 
 for holding virtue and prudence to be identical ; it would be 
 both wrong and foolish to make sham reasons where we have 
 not got real ones. 
 
 10. If any man outside of a pulpit were to ask himself 
 frankly what were the main conditions of human happiness, 
 the answer would certainly include one proposition. The 
 first, most essential, and most sufficient condition of happi- 
 ness is health — health, of course, inclusive of cerebral, and so 
 far of mental health ; or, in the opposite phrase, an absence 
 of every disease of mind or body. This, again, remains
 
 HAPPINESS AND VIRTUE. 405 
 
 equally certain^ whatever the difficulty of giving a satisfactory 
 definition of health or disease. We can only say vaguely 
 that health corresponds to the maintenance, and disease to 
 the disturbance of a certain organic balance. But though 
 we may not know precisely what they are, we cannot doubt 
 their importance. A sound digestion, an active liver, 
 strong nerves — briefly, all the qualities that go to a perfect 
 physical machinery — form the best of all outfits for the 
 voyage of life, so far as "best" means most productive of 
 happiness. 
 
 II. The question then arises. Can we push this theory any 
 farther ? Can we use the w^ord "health" in such a way as to 
 include the right working of all the functions — including 
 the intellectual and the emotional, as well as the purely 
 animal — and then assert that health in this wide acceptation, 
 including the normal constitution of the "moral sense," is 
 the most general and essential condition of happiness ? 
 Common sense seems to sanction some such doctrine. The 
 viejis Sana in corpore sano is still the truest possible definition 
 of the general condition of happiness, and we have no hesita- 
 tion in applying the words "healthy" and "morbid" to purely 
 mental phenomena. But it may be urged that the phrase 
 implies a metaphor too vague to bear much argumentative 
 stress. We are making an awkward transition from ob- 
 jective to subjective considerations, from the comparatively 
 firm land of physiology to the treacherous morasses of 
 psychology. We may assume, in virtue of the general 
 principle, that health of brain is a condition of happiness 
 at least as much as health of stomach ; and more vaguely 
 we may assume that health of brain has some correlation 
 with health of mind. In extreme cases it is plain enough 
 that a diseased brain implies madness and misery ; but we 
 are certainly unable to say what state of brain is correlated 
 with those good and bad qualities which constitute the 
 virtuous or vicious character, but which may vary from one 
 pole to the other without exceeding the limits of perfect 
 sanity. It is possible that we may be justified in inferring
 
 4o6 THE SCIENCE OF ETHICS. 
 
 that some correlation exists, but our inference, even if well 
 established, gives us next to no help. It asserts a relation 
 between two things, one of which, the state of the brain, 
 is hopelessly inaccessible, and can therefore afford no inde- 
 pendent information. If vice, like some physical diseases, 
 were due to the presence of a parasite in the tissues, the 
 fact might help us to define its effects upon the organic state 
 generally, and therefore upon the happiness of the agent. 
 But as the existence of any physical condition is a hypothesis 
 which cannot be confirmed by any direct test, and the con- 
 dition is not likely to be of this definite and assignable kind, 
 any such hypothesis would seem to be a mere curiosity. 
 And yet I think that the analogy has a real force, and may 
 be of use — not so much in suggesting new arguments as 
 in giving more coherence and fulness to the old familiar 
 argument. And this is as much as we can expect in ethical 
 inquiries. 
 
 12. That health implies happiness may be asserted on 
 purely empirical grounds. A simple induction may con- 
 vince us that all states which imply the disturbance of the 
 normal equilibrium are also states of misery. The convic- 
 tion of this truth may not be much intensified, but it seems 
 to be more or less " explained," — that is, shown to be not 
 merely a fact, but a necessary consequence of a wider prin- 
 ciple, — when we take into account the evolutionist doctrine, 
 which I have everywhere assumed. The whole process of 
 nature, upon that doctrine, implies, speaking briefly, a 
 correlation between the painful and the pernicious, and 
 thus (as I have argued at sufficient length) the elaboration 
 of types in which this problem is solved by an ever-increas- 
 ing efficiency and complexity of organisation. This holds 
 equally from the simplest up to the highest species. And 
 hence we may infer that the typical or ideal character at 
 any given stage of development, the organisation which, as 
 we may say, represents the true line of advance, corresponds 
 to a maximum of vitality ; to a maximum power of pre- 
 serving the equilibrium or resisting morbid conditions, and
 
 HAPPINESS AND VIRTUE. 407 
 
 therefore of a maximum accuracy in the correlation between 
 painful and pernicious states. It seems, again, that this 
 typical form, as it is the healthiest, must represent not only 
 the strongest — that is, the most capable of resisting un- 
 favourable influences — but also the happiest ; for every de- 
 viation from it affords a strong presumption, not merely of 
 liability to the destructive processes which are distinctly 
 morbid, but also to a diminished efficiency under normal 
 conditions. The man who has the weakness which predis- 
 poses to disease may not suff'er so long as the disease is not 
 actuallv generated, but he has presumably less power of 
 enjovment under any circumstances, A defect in the 
 machinery will imply defective working, though actual decay 
 may not take place in the absence of a special strain. What 
 is true of the elaborate machinery of the physical organisa- 
 tion is also presumably true of the less accessible and definable 
 machinery of that which is called the spiritual organisation. 
 13. The argument, again, whatever its weight, is not 
 affected so far by the consideration that the phrase "per- 
 nicious " must be supposed to include conditions primarily 
 injurious to the society instead of the individual. So long 
 at least as we are speaking of the type, such an argument is 
 irrelevant. For, as I have said at sufficient length, the type 
 always and from the earliest period is moulded with an equally 
 essential, though not an equally permanent, reference to the 
 needs of the race. The typical or most effective form is the 
 most effective relatively to the whole conditions, and the so- 
 called social qualities are no less essential than those which 
 immediately refer to the fixed material conditions. The 
 reproductive and maternal functions of the animal are just as 
 much a part of its nature as the digestive or the respiratory, 
 and they must, therefore, have their proper place in the 
 oro-anisation. The same is equally true of the affections 
 which become prominent in the rational animal capable of 
 true social life. To the type, though not so fully to every 
 individual, capacity for society is just as essential as capacity 
 for breathing air. The whole organisation is moulded, we
 
 4o8 THE SCIENCE OF ETHICS. 
 
 may say, with a view to this social function, and a defective 
 constitution in this respect carries with it an impHcation of 
 unfitness for the presumed conditions of hfe just as much as 
 defects in any other direction. 
 
 14. This may suggest the bare outline of an argument 
 w^iich might be carried out and illustrated in much greater 
 detail by those who have the necessary knowledge. How- 
 ever fully set forth, it would still, I suspect, remain exceed- 
 ingly vague, and probably would have some logical gaps. It 
 is scarcely to be regarded as an independent argument, but 
 as affording a presumption that some generally admitted 
 truths may be manifestations of a general scientific principle, 
 and thus necessary under the known conditions of life. 
 Whatever its vagueness, I have briefly stated it, because I 
 think that it will enable us to suggest the best method of 
 viewing the ordinary reasonings as to the relation between 
 virtue and happiness. 
 
 15. And, in the first place, it is to be noted that upon this 
 showing — as indeed upon any intelligible ground — the con- 
 nection of morality and virtue is conditional rather than 
 absolute. For the true statement is, so far as we have gone, 
 that the typical man is so far the happiest. But the typical 
 man, though he is, on my theory, the virtuous man, is also 
 much more than is generally understood by that name. 
 Happiness is the reward offered, not for virtue alone, but for 
 conformity to what I have called the law of nature ; that 
 law, namely, of which it is the great commandment, "Be 
 strong." If the problem at which the whole race is perpe- 
 tually, even when unconsciously, labouring, is the production 
 of the most vigorous type, and if pain and pleasure be the 
 great incentives to labour, then we must admit that the prizes 
 must be won by those who, on the whole, and considering 
 all their relations, are the most efficiently constituted. As 
 individual excellence assumes fitness for the social functions, 
 so the social excellence clearly assumes fitness for the non- 
 social functions ; and therefore we must expect to find that 
 happiness is correlated not only with the qualities which
 
 HAPPINESS AND VIRTUE. 409 
 
 involve approval, but those also which excite admiration. 
 Beauty, strength, intellectual vigour, aesthetic sensibility, 
 prudence, industry, and so forth, are all implied in the best 
 type, and are so far conducive to happiness. 
 
 16. This remark, obvious enough, seems to be implicitly 
 disregarded in some of the attempts to prove that virtue 
 brings happiness. For if virtue be taken in its narrower 
 sense, and as implying chiefly the negative quality of habi- 
 tual abstinence from forbidden actions, there is no reason to 
 suppose that it coincides with happiness. It marks only a 
 partial, and perhaps a subordinate, conformity to the essential 
 conditions of life. Nature — if I may use that convenient 
 personification for things considered as part of a continuous 
 system — wants big, strong, hearty, eupeptic, shrewd, sensible 
 human beings ; and would be grossly inconsistent if she 
 bestowed her highest reward of happiness upon a billions, 
 scrofulous, knockkneed saint, merely because he had a strong 
 objection to adultery, drunkenness, murder, and robbery, or an 
 utter absence of malice, or even highly cultivated sympathies. 
 You can only raise a presumption that moral excellence 
 coincides closely with a happy nature if you extend "moral " 
 to include all admirable qualities, whether they are or are 
 not the specifically moral products of altruistic feeling. The 
 attempt to evade this conclusion is, of course, exceedingly 
 natural. The practical moralist holds that the non-social 
 qualities may be left to take care of themselves. The moral 
 progress is dependent upon the extension of the sympathies, 
 and, at any given period, the huge dead weight of selfishness 
 is the great obstructive force. This is reflected in the theory 
 of desert assumed by the ordinary morality of common sense. 
 No man acquires a claim upon his neighbours by attending 
 to his own wants. If, indeed, men were differently consti- 
 tuted — if, as is sometimes the case in individual cases, sym- 
 pathy meant a morbid state of mind, a neglect of a man's 
 own affairs, and a useless expenditure of emotion upon affairs 
 which he could not control; — if, in short, men erred in the 
 direction of being sentimental busybodics, it might become
 
 4IO THE SCIENCE OF ETHICS. 
 
 necessary to preach the opposite doctrine. We should then 
 feel '^ obliged " to the man who strenuously minded his own 
 business and cultivated his own corner of the world, and 
 might even come to think him the most '^meritorious/' 
 For the present, there is no appreciable risk in this direction. 
 There is an inexhaustible supply of selfishness and a most 
 plentiful lack of public spirit; and therefore the aim of all 
 moralists — rather of all moral human beings — will be to 
 develop the side of human nature now most defective. Only 
 when this essentially relative conception of merit is made 
 absolute we get into all the perplexities which warn us that 
 we are flying in the face of facts. If from the doctrine that 
 virtue is meritorious we infer, not that the development of 
 certain qualities is a necessary element of social progress, 
 but that the ultimate end of the universe is to stimulate, 
 and therefore to reward virtue, we get into the region of 
 unrealities. If we can speak of an " end " at all in that 
 sense, it must, so long as we are in the scientific region, be 
 the end towards which we can perceive some progress, 
 namely, the development of a higher type — higher in respect 
 of the general efficiency of the whole organism, and certainly 
 not the production of virtuous beings simply ; especially if by 
 virtue we mean conformity to that narrower moral code which 
 is only a part of the " law of nature." To make any such 
 theory square with the facts, we must either overlook the 
 most palpable and undeniable phenomena of life, or we 
 must take refuge in a world of arbitrary fancies. 
 
 17. We have still, however, to consider the bearing of 
 this doctrine upon the relations between virtue and happi- 
 ness; for it is possible, consistently with the theory, to 
 regard that relation as approaching absolute coincidence. 
 If, in fact, we are justified in assuming that the typical man 
 is the happiest, and further that he is also the most virtuous, 
 we may say that the connection is of a very intimate kind. 
 It is true, indeed, that a limit is suggested, for it is only upon 
 condition of an alliance with the other useful qualities that 
 we suppose virtue to imply happiness. Now as most men
 
 HAPPINESS AND VIRTUE. 41 1 
 
 are very far from possessing the other qualities in perfection, 
 it may be that their virtues should be diminished in propor- 
 tion in order to produce a maximum of happiness. This, 
 for example, is sometimes insinuated in the cynical maxim 
 which recommends us , to keep our hearts cold and our 
 stomachs warm, for this seems to be an epigrammatic asser- 
 tion of the theory that warm affections are apt to be injurious 
 to the digestion. Perhaps, indeed, its author meant to go 
 further; he meant, it may be, to assert that the affections 
 were, on the whole, sources of more misery than happiness, 
 and he was probably one of the people who consider that 
 your heart, like vour liver, must be out of order whenever 
 YOU become distinctly conscious of its existence. But omitting 
 this, which will be presently noticed, the doctrine may be 
 interpreted as an assertion that the pleasures of sympathy are 
 worth having, but only on condition of their not interfering 
 with sanitary considerations. And this, true or not, leaves 
 a vague but very wide sphere for their possibly agreeable 
 activity; and, in fact, it seems impossible to deny that a 
 man's capacity for emotional excitement may be too great 
 for his constitutional power ; that he carries too m.uch sail for 
 his prudential ballast, and that he has, in this sense, too much 
 virtue for happiness. 
 
 18. The same thing indeed may be stated in a less offensive 
 way. We may either say that the man has too much virtue 
 for his prudence, or that he has too little prudence for his 
 virtue. We mean in either case that he deviates from the 
 type by a relative excess of those qualities which are, on an 
 average of cases, defective. He is so far a less efficient being 
 on the whole, and therefore less calculated for happiness, as 
 his nature is ill balanced ; and the balance might be redressed 
 either by strengthening the weaker or weakening the stronger 
 side. But it is not true in an absolute sense that he is too 
 virtuous for happiness; for by our assumption the virtue is 
 the typical character considered under one aspect or in 
 certain definite relations. The typical man represents the 
 best kind of man who can be made out of given materials ;
 
 412 THE SCIENCE OF ETHICS. 
 
 and this Involves both the quahties which excite admiration 
 and those which excite moral approval. It is possible, indeed, 
 that the supposed man may represent an altogether higher 
 type than that implied in the actual moral standard ; and 
 we shall ask presently whether we have any right to suppose 
 that such a being is presumably happier. In the ordinary 
 case, the answer already given is sufficient. The assertion 
 that a man has too fine feelings means equaliv he has too 
 weak a fibre. He is deficient in one kind of excellence, 
 though not in the moral kind ; and any deficiency so far tells 
 against his happiness. This, in fact, seems to be true of many 
 of the men who have been made miserable by qualities which 
 w^e rightly call good, because they are implied in the highest 
 type, but which, taken by themselves, are an insufficient 
 guarantee for happiness. Virtue, in the sense of benevolence 
 or quick sympathies, will not supply the place of strong 
 nerves and high vitality. The whole man does not represent 
 the best type, for a race which developed its sympathies at 
 the expense of its vitality as a whole would be, for that reason, 
 an inferior race, and represent not progress but deterioration. 
 19. The general argument might be confirmed by going 
 again over the chief virtues, considered in their individual 
 instead of their social aspect, and showing how they implied 
 greater fitness for the normal conditions of life. To do so 
 would be to repeat much that I have said, and to insist upon 
 some of the familiar arguments, which I am content rather 
 to take for granted ; and it will be sufficient to indicate here 
 a few obvious points, w^ithout fuller elaboration. Thus, in 
 the first place, we might dwell upon the direct connection 
 between the prudential and the moral aspect of sanitary 
 rules. Excesses of sensuality may be forbidden for purely 
 prudential reasons ; a man may be temperate to avoid the 
 pains of delirium tremens. The higher character is certainly 
 not insensible to this motive, but he has the additional 
 motives that by injuring his health he is injuring his power 
 of discharging every social function, and that sensuality 
 implies the dulling of his intellect and the deadening of his
 
 HAPPINESS AND VIRTUE, 413 
 
 emotions; and thus, as I have said, it matters not whether 
 some rules forbidding conduct which in its first incidence 
 is "self-regarding" be regarded as prudential or as moral. 
 They are properly both. The same rule, " Be temperate," is 
 applicable from the lowest to the highest stage ,of develop- 
 ment, though the details of the code would vary as the con- 
 stitution. The typical being at each stage would be tempe- 
 rate, whether from mere instinct, or from a reasoned obser- 
 vation of consequences, or from the wider considerations 
 operative upon the being capable of genuine morality. It is 
 not less important at the highest than at the lowest stages 
 that the bodily health should be preserved ; and a regard for 
 sanitary rules is therefore implied in the typical being, and 
 in the consistent code of rules expressing his mode of conduct, 
 and embodying both the moral and the prudential law. Of 
 course, there are special occasions on which the virtuous man 
 will have to sacrifice health to other considerations ; but the 
 whole law must be so constructed as to harmonise with the 
 sanitary code under normal conditions ; for qualities which 
 could only be acquired at the expense of health would not 
 be useful to the society, and would therefore not represent 
 the true line of moral advance. Saints, like poets, are subject 
 to certain morbid tendencies; but the saintliness which 
 necessarily promotes sickliness is the kind of saintliness 
 which diverges from sound morality. 
 
 20. From this one might proceed to show how the qualities 
 more specifically moral fall in with the same principle. So, 
 for example, the so-called intellectual virtues, the virtues of 
 which truth and justice are the specifically moral expression, 
 are also the product of what we naturally call a thoroughly 
 sound mind ; that is to say, a mind which works normally 
 undisturbed by prejudice and passion, which has that couraQ-e to 
 face disagreeable facts and to draw painful conclusions which 
 does not imply absence of the emotions, but a power of re- 
 straining that irregular action which distorts and colours the 
 intellectual vision. This would be followed, ao:ain, bv reckon- 
 ing up the vast internal advantages which result to the man
 
 414 THE SCIENCE OF ETHICS. 
 
 whose intellect is thus healthy ; for health implies a fitness for 
 the normal conditions of life. And here I might expatiate upon 
 the immense advantages to every man of imperturbable sound- 
 ness of sense, and the undeviating honesty, justice, and trust- 
 worthiness which are its natural fruit in a man's intercourse 
 with his neighbours. It would, again, be equally easy to 
 dilate upon the advantage of a healthy emotional nature, one 
 which is neither morbidly overwrought nor deficient in strength 
 and sensibility; for, in spite of the advantages attributed to 
 a cold heart, or, more moderately, to a calculated selfishness, 
 it must be admitted that some warmth of affection is neces- 
 sary to happiness. So it may be doubted whether any consider- 
 able intellectual development is possible without a correspond- 
 ing emotional development. Without it a man is a hog in 
 mind as well as in heart. Of all conditions of health, again, 
 one of the most obvious is steady and regulated exercise of 
 the various faculties. To weaken the affections is to weaken 
 all the motives to action which are capable of dominating and 
 giving continuous interest to life. These supply the springs 
 of all actions which can really satisfy a man's whole nature. 
 Nobody will doubt that domestic peace is the most important 
 of all conditions of happiness after bodily health; and next 
 to it we may place the possession of interests which give 
 motives for continuous though not exhausting labour. The 
 great bulk of mankind is absorbed in bodily or other labour, 
 which has little interest except as a means of support to 
 themselves and their families; and its pleasantness or other- 
 wise must depend chiefly upon its healthiness or the reverse. 
 The happiness of the whole life in such cases depends gene- 
 rally upon the domestic interests round which it centres; but 
 beyond this, a man's happiness may be said to depend chiefly 
 upon his power of associating agreeably with his fellow-men, 
 and of devoting himself to some of those pursuits which imply 
 public spirit, and therefore a capacity for sympathy in some 
 of its multitudinous forms. 
 
 21. This hint at a famiHar line of argument must suffice, 
 except that I must once more take brief notice of an opinion
 
 HAPPINESS AND VIRTUE. 415 
 
 which has encountered us in various forms. Here it declares 
 that there is something of paradox in our fundamental 
 theory. That theory asserts that a man is the happiest who 
 does not aim at his own happiness. In order, then, to be 
 happy we must not aim at happiness. We have, as it were, 
 to keep a secret from ourselves, and to hit the mark by pre- 
 tending to look in the opposite direction. I deny, however 
 (not to go into other questions), that the argument is here 
 relevant. Sympathetic feelings are just as much feelings, 
 and, of course, feelings of the agent himself, as any other 
 feelings. They cause real pains and pleasures; and there 
 can be no a priori reason for supposing that a man endowed 
 with sensibility to such pains and pleasures may have, 
 on the whole, more happiness than the man without. If, 
 indeed, a man retaining precisely the same character could 
 act either selfishly or unselfishly, it might seem strange to 
 assert that he would get more happiness for himself when 
 pursuing the happiness of others. But this involves a con- 
 tradictory assumption. Unless a man be really altruistic 
 his actions are really selfish, even when they aim incidentally 
 at good to others. If he be altruistic, he is a different 
 man from the purely selfish. The comparison, therefore, is 
 not really between the happiness of the same man acting 
 from one or the other motive, but between the happiness of 
 two differently constituted people, one of whom has the 
 capacity for a whole set of pains and pleasures which are 
 denied to the other. Experience alone can tell us which 
 constitution is the best suited for happiness ; and I have 
 indicated the line of argument which, as I think, is 
 conclusive in favour of the sympathetic character. The 
 fact that the sympathetic man aims at the happiness of 
 others can only appear, at first sight, to be an argument 
 against him. For, in the first place, it makes him happv to 
 aim at the happiness of others; and if this happiness has the 
 awkward quality that it may lead to an actual sacrifice of his 
 own future happiness, that is not less true of the selfish forms 
 of enjoyment. Love of drink causes a good deal more "self-
 
 4i6 THE SCIENCE OF ETHICS. 
 
 sacrifice" in this sense than love of family. Gin is a more 
 potent source of imprudence,- even in the moderate sense, 
 than family affection. The happiness which results from 
 aiming at something different from our own future happiness 
 calls no doubt for conspicuous calls of self-abnegation, where- 
 as the purely selfish motives always promise payment in 
 kind to the agent himself. But the sympathetic motives 
 have on their side the far greater intrinsic advantage that they 
 promote ends more permanent, far richer in interest, and giving 
 a proper employment to all the faculties of our nature, besides 
 the intrinsic advantages which spring from friendly relations 
 with the society of which we form a part. Nobody can doubt 
 but a man would on an average, that is, if he has the necessary 
 faculties, have an incomparably greater guarantee for happiness 
 who was a devoted member, say of some political or religious 
 body, than if he were exclusively bent upon satisfying one of 
 the sensual appetites. 
 
 32. The arguments upon which I have touched, when 
 duly expanded and enforced by appropriate evidence, may 
 convince us of the truth of the proposition laid down. 
 There is, namely, a necessary connection between virtue and 
 happiness, inasmuch as there is a necessary connection be- 
 tween virtue and total efficiency (to put it briefly), and as 
 virtue forms a necessary part of efficiency. Therefore in so 
 far as a man is not virtuous he deflects from the type, and 
 is less calculated for happiness. And this is quite consistent 
 with the admission that he may deflect from the type in 
 other ways, which do not entitle us to call him vicious, but 
 which are equally fatal to happiness. But there is an obvious 
 gap in the argument, for at every step of the argument it is 
 clear that we virtually assume a certain state of social deve- 
 lopment; for the " typical " man is, roughly speaking, the 
 best that can be made out of existing materials, and by the 
 " best," again, we have meant the best raw material for the 
 formation of a vigorous society. But, again, it is plain that 
 the type gradually changes, and though the change is for the 
 most part slow, it becomes considerable by accumulation.
 
 HAPPINESS AND VIRTUE. 417 
 
 Now if we suppose a man to be distinctly in advance of his 
 age, we must say that he is more moral, or represents a 
 superior type; but for that reason he is out of harmony 
 with his social medium, and loses, therefore, all that part of 
 the advantages of virtue which depends upon such harmony. 
 In fact, it seems that the argument tends to show the advan- 
 tage of the '^type," considered as representing the best 
 instantaneous development. Happiness will be the reward 
 of fitness for the actual conditions, and therefore of an em- 
 bodiment of the code of virtues actually recognised and 
 possessing authority. Admitting that it is worth while, say, 
 to be respectable, we may doubt whether it answers to be a 
 moral hero. Or, again, we may say that as the advantages 
 depend upon a certain agreement amongst men, I have com- 
 paratively small inducement to respect the agreement until 
 other people can be brought to respect it also. In fact, it is 
 plain that in many ways this is a very important consideration. 
 A fine ear for music is said to be a source of torture to a man 
 condemned to live amongst organ-grinders; delicate tastes un- 
 fit us for living amongst the coarse ; even intellectual activity 
 exposes a man to much discomfort if he has to live amongst 
 the stupid. We may say summarily that, though the intrinsic 
 advantages of the higher organisation are not changed, the 
 extrinsic disadvantages due to life in an uncongenial medium 
 go far to neutralise them. 
 
 33. It seems impossible to deny that this holds good of the 
 moral qualities. Temperance is no doubt directly useful in 
 so far as it is sanitary, and this will be equally true wherever 
 man lives; but in a gross society, where the temperate man 
 is an object of ridicule, and necessarily cut off from participa- 
 tion in the ordinary pleasures of life, he may find his moral 
 squeamishness conducive to misery. The just and honour- 
 able man is made miserable in a corrupt society where the 
 social combinations are simply bands of thieves and his high 
 spirit only awakens hatred ; and the benevolent is tortured 
 in proportion to the strength of his sympathies in a society 
 where they meet with no return, and where he has to witness 
 
 a D
 
 41 8 THE SCIENCE OF ETHICS. 
 
 cruelty triumphant and mercy ridiculed as weakness. The 
 domestic affections are the greatest sources of happiness, but 
 if a man's wife be faithless and his children " unnatural/' 
 they are so many possibilities of exquisite torture ; and similar 
 truths hold of the wider social circles. All this tends to show, 
 not only that a civilised man among savages, a gentleman 
 amongst rogues, and so forth, is exposed to misery by reason 
 of his superiority, but even that, every reformer who breaks 
 with the world, though for the world's good, must naturally 
 expect much pain, and must be often tempted to think that 
 peace and harmony is worth buying even at the price of 
 condoning evil. '' Be good if you would be happy," seems to 
 be the verdict even of worldly prudence ; but it adds in an 
 emphatic aside, ^^Be not too good." 
 
 III. Moral Discipline. 
 
 24. If we admit (as I think that we must admit) this limi- 
 tation upon the general principle, we virtually admit that an 
 excessive virtue cannot be recommended to the selfish person 
 upon grounds intelligible to him. And there is another, and, 
 for most purposes, more important exception. Saints are 
 rare, and will probably be little affected by prudential argu- 
 ments. But there are plenty of selfish and sensual people 
 to whom we may be unwilling to admit that they will not 
 find it worth their while to be virtuous; and yet, may they 
 not uree very forcibly that a hio;her strain of virtue is a 
 questionable advantage to the lofty nature, and even an aver- 
 age degree of virtue may be a doubtful gain for lower natures ? 
 Granting, they may say, for the sake of argument, that the 
 virtuous are generally the happier for their virtue, is that 
 equally true for people with small aptitudes in that direction ? 
 Why should they undergo the necessary discipline ? You 
 would not advise a man to try to learn music unless he had 
 a rudimentary perception of the difference between harmony 
 and discord ; and if a man's appetite be gross and sensual
 
 MORAL DISCIPLINE. 419 
 
 and his sympathies slow, can you suggest to him any adequate 
 motives for cultivating his moral sense ? 
 
 25. This brings us to the second of the problems which 
 I proposed to consider ; and the general remark may be 
 made, that the argument of a coincidence between happiness 
 and virtue (so far as it exists) docs not depend upon the 
 mode in which a man becomes virtuous. I have assumed 
 that- moral progress (speaking roughly) does not imply an 
 advance in the innate qualities, but a development through 
 the social factor. Children, no doubt, start with infinitely 
 varying aptitudes for moral culture, as they start with 
 stomachs of varying strength of digestion ; but in every 
 case the action of the social medium is an essential factor 
 in the result; and therefore the bare fact that the qualities 
 have to be developed is not more peculiar to one individual 
 than another, and can afford no presumption as to their 
 connection with his happiness. If a man has the normal 
 constitution, he will presumably be the happier for a moral 
 development, as, if he has the normal intellect, he will 
 derive the normal benefits from education, or, if he has a 
 normal stomach, he will derive the normal benefits from 
 observance of sanitary rules. The foregoing reasons are 
 therefore equally applicable, though in particular cases moral 
 culture may be thrown away upon uncongenial natures. 
 Briefly^ so far as a man is the happier for being virtuous, he 
 is presumably the happier for a virtuous training. Still some- 
 thing too must presumably depend upon his aptitude for receiv- 
 ing it ; and we may therefore ask the question, What are the 
 inducements to a cultivation of virtue which exist indepen- 
 dently of and antecedently to the virtue itself? What pur- 
 chase have we in the list of the character for stimulating a 
 man to the development of his latent good qualities ? To 
 answer that would be to say whether it is worth while for 
 the man — reckoning the ''worth while ^' in terms already 
 intelligible to him — to become a new man. 
 
 26. We may still rely to some extent upon the old prin- 
 ciple. The process of moralisation is part of a complex pro-
 
 420 THE SCIENCE OF ETHICS. 
 
 cess ; and whether we consider the race^ or the individual 
 who repeats the race-experience^ it depends upon the total 
 intellectual^ emotional, and physical growth. It follows 
 that, although the undeveloped man may be blind to some 
 of the advantages of the moral character, he may have an 
 eye for the advantages of development in general, or of some 
 particular kind of development with which the moral is 
 necessarily bound up. He is accessible to all the ordinary 
 prudential arguments as to the superior fitness for society 
 of the moral character, even if he has himself a very slight 
 appreciation of the more intellectual pleasures. And here 
 comes in the true bearing of an argument which we have 
 had to reject in a different shape. I see no use in asking 
 whether Shakespeare or a pig is the best judge of the relative 
 merits of pigwash and poetry, but it may be worth the pig's 
 W'hile to aim at making some approach to Shakespeare. The 
 stupidest lad may see some of the advantages of sharpening 
 his faculties ; he will not see — for it is not true — that poetry 
 is absolutely pleasanter than dinner, but he may easily see 
 that he would be the happier if he had such an education 
 as would make intellectual pleasure determine a much 
 greater share of his total activities. He may see that 
 stupidity is in a thousand ways a disadvantage ; that it 
 makes him inferior to competitors, despised by his equals, 
 incapable of many enjoyments, and so forth. Even a prize- 
 fighter or a foxhunter finds a good brain useful up to a 
 certain point, and has sometimes capacity enough to infer 
 that a more vigorous brain would be still more useful. And 
 putting the same remark generally, we may say that there 
 is something like a general (though clearly not a universal) 
 consent as to the advantage of a full development of our 
 faculties. It is, I think, hopeless to produce a balance-sheet 
 of pains and pleasures which would prove that the virtuous 
 man gets a greater sum of pleasant emotions or a sum of 
 emotions superior in (juality ; but it may be proved that a 
 man gains by growing as much as it is in him to grow, and 
 that this necessarily involves moral growth.
 
 MORAL DISCIPLINE. 421 
 
 37. The argument thus sug:2;cstccl dilTcrs from the ancient 
 and famihar arguments of all practical moralists only by 
 laying rather more stress than they generally do upon the 
 intrinsic as distinguished from the extrinsic considerations. 
 The ordinary arguments, "Be industrious, and you may come 
 to be Lord Mayor ; " "^Ee sober and steady, and your master 
 may allow you to marry his daughter," are very good argu- 
 ments, so far as they go ; and the practical moralist who 
 scorned to make use of them would, I think, be very ill 
 advised. I am only pointing out that such a statement 
 does not exhaust the purely prudential argument. It is 
 equally possible to prove to a man upon purely prudential 
 grounds the advantage of cultivating such germs of good 
 feeling as he may possess, and make him see that besides the 
 direct pleasure derivable from their exercise and the inci- 
 dental advantages derivable from the respect of his fellows, 
 such culture is a necessary element in the full development 
 of his nature, and therefore in his excellence in other facul- 
 ties. The general statement has already been sufficiently 
 indicated to show the advantages of a moralised nature. 
 I have had to speak of this essential connection between the 
 qualities embodied in the type, and what I am now saying is 
 merely an application of the same principle. A one-sided deve- 
 lopment is always a disadvantage, as a defect in one faculty 
 has some reaction upon those which form part of the same 
 nature. We may undoubtedly find cases enougli of an 
 apparent divergence between different excellences. There 
 are men of vast intellect who have been thoroughly selfish 
 and unscrupulous ; or, again, men of fine sensibilities and 
 rich emotional natures who have become prophets of the 
 baser elements in our nature; and, in such cases, it may 
 seem as if eminence were the reward, not simply of indifl:er- 
 ence to, but of downright contempt for, moral considerations. 
 Yet a fair examination would rather prove, I think, that 
 this is an erroneous view. The statesman who is wanting 
 in genuine moral feeling or sympathy for his kind is under 
 one of the greatest disadvantages possible ; and a great
 
 422 THE SCIENCE OF ETHICS. 
 
 man, even though the moraUst may feel bound to condemn 
 him, is generally in a kind of tacit alliance with morality, 
 simply because the intellect which makes him perceive the 
 realities of life reveals to him the advantage of being on the 
 side of the strongest forces, and therefore, even in his own 
 despite, of progress. But he is more powerful if he is impelled 
 by more spontaneous sympathies. So, on the other side, we 
 may be inclined to say that the highly emotional nature 
 is at once a source of power over the sympathies of others 
 and of many moral weaknesses ; but it is as true to say 
 that a development of the counterbalancing faculties will 
 improve his morality without weakening his power over 
 others. Incontinence of feeling is a source of weakness, 
 though it may at times cause a superficial appearance of 
 vigour. In proportion to the development of a man's intellect 
 is not only his power but his need of sympathy. As he is 
 able to identify himself with a larger part of the social 
 organism, an identity between his own interests and the 
 deepest interests of his fellows becomes more essential to 
 his life. 
 
 28. This, again, falls in with the doctrine, which may be 
 presented in various lights, of the unity of virtue, or of the 
 statement that virtue has a positive, and vice only a negative 
 meaning. This, in fact, is one way of putting the truth that 
 virtue is one aspect of the development of a single type, 
 whereas vice means any tendency towards disintegration. 
 Consistency and harmony of conduct is therefore one 
 characteristic of the virtuous, whilst vice is by its nature 
 heterogeneous and leads to discord. If, for example, a man 
 has a lofty intellect, and has yet overpowering sensual 
 impulses, his nature is, as we say, at war with itself. For as, 
 on the one hand, he is fitted for the pursuit of abstract truth 
 and capable of wide sympathies and lofty ideals, he is, on 
 the other hand, attracted by the lower kind of pleasure 
 which can only be attained at the expense of the higher. 
 He is a great scholar or statesman by fits, and at other 
 times the slave of the bottle or of women. This kind of
 
 MORAL DISCIPLINE. 423 
 
 inconsistency or irrationality is in fact the expression of an 
 ill-balanced character^ and by " ill-balanced " wc mean that 
 which deflects from the type actually worked out by con- 
 formity to the condition of maximum efficiency. 
 
 29. And here we come upon a fresh form of the old diffi- 
 culty ; for we see that the discord of which we are speak- 
 ing is the mark, as it were, of a hybrid nature — of a man 
 who belongs in some respects to the higher, and in others to 
 a lower type. And thus we have still to ask whether a man 
 may not be also harmonious with himself if he belongs to a 
 lower type, by some innate defect which unfits him for 
 receiving the normal development, and whether in that case 
 we can suppose that he will be the happier for moral dis- 
 cipline? Are there not, in fact, men whose sympathies are 
 so dull and whose intellects are so torpid that a pursuit of 
 sensual pleasure will always be more congenial to them, and 
 that they will only derive misery for themselves, if not for 
 others, by being forced into conformity with moral rules? 
 Are there not idiots, and beings at every point of the scale 
 between absolute idiocy and the loftiest genius ? and if so, is it 
 possible to say that all of them will be the happier for accept- 
 ing the same law ? A man with a capacity for sympathy may 
 be the happier for having it fully developed. But suppose 
 that he has none ? If by any process you can persuade him 
 to act like a benevolent man, will this fictitious benevolence 
 repay him for abstinence from gin ? 
 
 30. I can certainly see no reason to doubt the existence of 
 human brutes — of men who not only do nOt possess, but can- 
 not be made to possess, the instincts characteristic of the 
 higher type. There are, I fully believe, men capable of intense 
 pleasure from purely sensual gratification, and incapable of 
 really enjoying any of the pleasures which imply public 
 spirit or private affection or vivid imagination. It is true, 
 generally, that each man has certain capacities for moral as 
 for every other kind of development, and capacities which 
 vary from the top to the bottom of the scale. No process 
 of education or discipline whatever would convert a Judas
 
 424 THE SCIENCE OF ETHICS. 
 
 Iscariot into a Paul or a John. And therefore, too, as a 
 plain matter, we must admit that the ends which men pursue 
 vary indefinitely, and that some men, possibly the mass of 
 men, are fitted for those positions in the social organism 
 which do not demand any great activity of the higher 
 faculties, or make any strain upon a man's devotion to the 
 race or to truth. So far, again, as a man cannot be made 
 moral, there is very little use in asking whether he would be 
 the happier for being made moral. That is really to ask, 
 as before, whether a different kind of man is happier. And 
 thus, when we speak of the morality of the lower type, we 
 must mean that the habit of obedience to the moral law may 
 be impressed upon it, although the normal instincts which 
 make such obedience the spontaneous fruit of his charac- 
 ter may be very imperfectly developed, and therefore, as a 
 general rule, that to some extent other instincts, such as the 
 fear of punishment or of the contempt of his fellows, have 
 been called into play, so as to make, as it were, a substitute 
 for genuine morality. Is such a process likely to make him 
 happier ? 
 
 31. To this we must reply, that the external and prudential 
 sanctions must generally remain in force for all men inde- 
 pendently of their character. The evils of drunkenness, so 
 far as they consist in disease or poverty, are applicable to 
 and appreciable by the most sensual man, though he is not 
 equally sensitive to the injury to his intellect. The objec- 
 tions to a want of honesty which depend upon the action 
 of the policeman or the victims of fraud are applicable to 
 every one in a civilised country. And generally the human 
 hog as well as the genuine man will find his account in 
 being on tolerable terms with the society in which he lives. 
 But it certainly does not appear that this is equally true 
 of the intrinsic sanctions. The sensualist who has enough 
 prudence to avoid disease or general disgrace may get a 
 quantity of pleasure from immoral practices. If by any 
 device, by persuading him that he will go to hell for such 
 indulgence, or by so strictly drilling him till abstinence has
 
 MORAL DISCIPLINE. 425 
 
 become a custom, and the sensuality has thus lost some of 
 its power without making room for the development of the 
 nobler feelings, it is hard to say why such restraints should 
 make a man happier any more than physical restraints. The 
 drunkard in a jail who cannot get his bottle is probably less 
 happy than when he is at large ; and if his jail be made of 
 prejudices forced upon his nature, and justifiable by motives 
 which he cannot understand, it does not seem to make much 
 difference. So, again, when a man has no real benevolence, 
 but is forced into the outward practice of benevolence by 
 custom and dread of public opinion, he is probably less at 
 ease than if he were systematically cold-blooded. So the 
 man who goes to church because he has always been taught 
 to do so, without possessing the instincts which are gratified 
 by religious worship, is probably more uncomfortable, even 
 though he may not be a conscious hypocrite, than if he were 
 frankly a " worldling." And thus, though it is very hard 
 to sum up such considerations, our argument would seem to 
 point to the conclusion, that, as a very general rule at least, 
 obedience to the external moral law is a matter of prudence 
 for everybody ; that it can be proved to almost any man 
 that it is safer for him not to be at war with his fellows or 
 indulge his appetites to excess; but that, on the other hand, 
 it cannot be said with any confidence that if we were to 
 consult the happiness of the agent exclusively, we should 
 always try to infuse into him habits of virtue which tran- 
 scend this rather moderate limit. It is possible to make a 
 man less fitted for enjoyment under normal circumstances 
 by trying to put too high a polish upon his moral nature, as 
 it is possible to achieve the same result by cultivating tastes 
 for art or intellectual study in those who have no natural 
 aptitude. 
 
 32. It may be necessary to add, that this is no sufficient 
 reason for not trying to do it. Happily, as I have already 
 said, the moral standard of society does not depend upon 
 each man's estimate of what will be most for his own happi- 
 ness, but much more upon the general conviction of every
 
 426 THE SCIENCE OF ETHICS. 
 
 man that it is his interest to Hve in a moral world. We 
 may, indeed, desire to see this conviction a great deal 
 stronger than it is. Yet even the sensualist, who, for his 
 own part, expects to get the greatest amount of enjoyment 
 out of life by indulging his appetites, is generally anxious 
 that the world at large should be guided by different prin- 
 ciples. Doubtless the estimate which every man forms of 
 the general conditions of happiness is greatly biassed by his 
 character, and the Utopia of the drunkard is very different 
 from that of the philosopher. Still the moral standard is 
 not formed by a simple generalisation of each man from his 
 own experience, and a desire for the greatest possible divi- 
 dend of such pleasure as he can appreciate, but is worked 
 out by the general social process, which virtually decides the 
 observance of certain rules to be essential to the general 
 welfare, which, again, may imply conduct and character very 
 different from the private standard of the man who honestly 
 accepts the general estimate. And, finally, whatever doubts 
 we may have as to the possibility of making any given person 
 moral, and of contributing to his private happiness by doing 
 so, we can have no scruples in making him as moral as we 
 can. For we cannot know till we have tried what capa- 
 bilities of development there may be in him, and the general 
 principle that moral development involves good to him and 
 still more to the society, is sufficiently demonstrated to com- 
 pel us to do what we can. The danger of failing on the 
 other side is the only one worth notice in any given case. 
 
 IV. Self- Sacrifice. 
 
 33. This brings us to the last question proposed, which 
 has indeed been more or less answered by anticipation. 
 Have we any right to say, not merely that the good are 
 happier than the bad, or that moral development is calcu- 
 lated to make a man happy, but that in every particular case 
 a man, whether good or bad, will be the happier for acting 
 rightly ? After the conditional answer which we have been
 
 SELF-SACRIFICE. 4^7 
 
 forced to give even to the more general questions, it is 
 impossible that we should expect to answer this more con- 
 fidentlv. I am, for my part, convinced that there are 
 occasions upon which we have to choose between two 
 masters. This way is the path of duty; that is the path of 
 happiness. We shall at times have to choose, and to 
 choose wuth our eyes open. Let us take as illustration any 
 of the famous cases of moralists. Regulus preferred death 
 by torture to dishonour. Was he acting for his own happi- 
 ness ? To make the question intelligible, w^e must suppose 
 that Regulus could have acted differently ; and this may be 
 taken to mean either that he could have declined martyr- 
 dom if his character had been different, or that he could 
 have declined it, his character remaining unaltered. On 
 the first hypothesis I must first remark, that the greater the 
 virtue the less the sacrifice in acting virtuously — a doctrine 
 which has a paradoxical sound only if we understand the 
 ''self" in self-sacrifice to be the man without his virtue, 
 and his virtue to be something external. The stronger the 
 virtuous impulse, the greater the force of other motives 
 which it can override, and so far the greater the "sacrifice" 
 which it can impose. But the less, for the same reason, is 
 the sacrifice of the whole nature implied in a given act, for 
 the gratification of that impulse counts for more in the 
 whole nature. Now, we may suppose a man so constituted 
 that death by torture would imply less misery than life in 
 disgrace ; and still more easily, so constituted that a certainty 
 of speedy death would imply a less prospect of evil than the 
 certainty, not of life, for life is always precarious, but of an 
 escape from death with disgrace for the indefinite remainder 
 of life. Would, then, a man in the position of Regulus have 
 greater chance of happiness for possessing such a sense of 
 honour as would determine him to martyrdom ? I think that 
 it is impossible to answer in the affirmative. Many men live 
 " infamous and contented " after saving life at the expense of 
 honour. Even if we suppose Regulus to have had so strong 
 a sense of honour as to make his martyrdom a " bed of roses,"
 
 428 THE SCIENCE OF ETHICS. 
 
 T do not see my way to deny that a man of less virtue might 
 have easily complied^ and have passed a very agreeable old 
 age at Capua as a retired general officer, deriving an amount 
 of pleasure out of life greater than any which fell to Regulus. 
 In this sense, Regulus's virtue was, under the given circum- 
 stances, a disadvantage to him personally. A less virtuous 
 man would have been in that sense better adapted to the 
 position. But the question generally has a different sense. 
 We assume in asking it that Regulus might have acted in 
 either way consistently with his character. His state of 
 health, the excitement of seeing his home or his family, the 
 presence of some symbol associated with patriotic sentiment 
 or vividly recalling the pangs of torture — briefly, that intri- 
 cate play of varying and conflicting emotions which gives 
 rise to the illusion of '^ free-will " — mi2;ht cause him to act in 
 either way. Again, I do not see my way to any definite 
 answer. The answer would really depend upon qualities of 
 character not assigned in our assumed data. A lapse from 
 a high standard may embitter the whole life of an honour- 
 able man. It may embitter it so keenly as to make such a 
 life worse than death ; and in that case, a choice of death 
 would be a choice of happiness. But the opposite hypothesis 
 is certainly conceivable. I see nothing contrary to the recog- 
 nised laws of human nature in supposing that Regulus 
 might commit himself to martyrdom in a moment of en- 
 thusiasm, and be afterwards overcome by a weakness which 
 would double the pangs of death ; whilst, on the other hand, 
 had he given way, he might have made the discovery — not a 
 very rare one — that remorse is amongst the passions most 
 easily lived down. Briefly, then, I admit in both senses that 
 Regulus may have acted in defiance of a calculation of happi- 
 ness ; and I may add that upon my principles the fullest 
 recognition of the fact might be quite compatible with such 
 action. It may be true both that a less honourable man 
 would have had a happier life, and that a temporary fall 
 below the hitrhest strain of heroism would have secured for 
 him a greater chance of happiness.
 
 SELF-SACRIFICE. 429 
 
 34. All this, again, may be applied to much more ordinary 
 cases. It is true, so far as I can see, that not merely heroic 
 virtue, but even virtue of the ordinary kind, demands real 
 sacrifices on some occasions. When we say to a man, " This 
 is right," we cannot also say invariably and unhesitatingly, 
 " This will be for your happiness. " The cold-hearted and 
 grovelling nature has an argument which, from its own 
 point of view, is not only victorious in practice, but logically 
 unanswerable. Not only is it impossible to persuade people 
 to do right always — a matter of fact as to which there is not 
 likely to be much dispute— but there is no argument in exis- 
 tence which, if exhibited to them, would always appear to 
 be conclusive. A thoroughly selfish man prefers to spend 
 money on gratifying his own senses which might save some 
 family from misery and starvation. He prefers to do so, let 
 us sav, even at the cost of breaking some recognised obli- 
 gation — of telling a lie or stealing. How can we argue with 
 him ? By pointing out the misery which he causes ? If to 
 point it out were the same thing as to make him feel it, the 
 method might be successful, and we may hold that there is 
 no reasonable being who has not at least the germs of sym- 
 pathetic feeling, and therefore no one who is absolutely 
 inaccessible to such appeals. But neither can we deny, with- 
 out flying in the face of all experience, that in a vast number 
 of cases the sympathies are so feeble and intermittent as to 
 supply no motive capable of encountering the tremendous 
 force of downright selfishness in a torpid nature. Shall we 
 then appeal to some extrinsic motive, to the danger of being 
 found out, despised, and punished ? Undoubtedly that will 
 be effective as far as it goes. But if, for any reason, the 
 man is beyond the reach of such dangers ; if he is certain of 
 escaping detection, or so certain that the chance of punish- 
 ment does not outweigh the chance of impunity, he may 
 despise our arguments, and we have no more to offer. He 
 may say — and, as it appears to me, may say with truth — '' I 
 shall personally get more pleasure from doing wrong than 
 from doing rioht, and I care for nothing but my personal
 
 430 THE SCIENCE OF ETHICS. 
 
 pleasure." The first statement may be — it often is — un- 
 deniably true. Of the second he is the only judge, and 
 though we may prove that he ought to be differently con- 
 stituted, or that a virtuous man would be differently con- 
 stituted, we do not thereby alter his nature, or even prove 
 by any arguments accessible to him that it is worth his 
 while to alter it. Against some people, in short, the only 
 effective arguments are the gallows or the prison. Unluckily 
 they are arguments which cannot be brought to bear with all 
 the readiness desirable, and therefore I think it highly pro- 
 bable that there will be bad men for a long time to come. 
 
 ^;;^. I see no use in shutting or trying to shut our eyes to 
 so plain a truth. As regards the world with which alone 
 scientific reasoning can have any concern, it is a simple 
 statement of undeniable facts, or of facts which can only be 
 denied in some potential sense, that is to say, not really 
 denied at all. The theologian and the philosopher are, in- 
 deed, willing to admit it so far as the world of experience 
 goes, and sometimes inclined to insist upon it as affording a 
 presumption in favour of a better constituted world beyond 
 or underlying; and I have here nothing to say to their solu- 
 tions. But the attempt to evade the truth as regards the 
 existing state of things is often made, and leads, as I fancy, 
 to a weary waste of sophistry. The attempt to establish an 
 absolute coincidence between virtue and happiness is in 
 ethics what the attempting to square the circle or to dis- 
 cover perpetual motion are in geometry and mechanics. I 
 think it better frankly to abandon the hopeless endeavour. 
 Indeed, this admission, true or false, would seem to be but 
 the inverse side of a doctrine which most moralists have 
 laboured, and, as I think, successfully laboured to establish. 
 When we listen to the careful demonstrations of the reality 
 of benevolence, when we are told aijain and aeain that a 
 man may, and in fact does, sacrifice his own happiness to 
 the good of his fellows, we are edified and convinced. But 
 we receive something of a shock when the edifying moralist 
 suddenly turns round and tells us that the sacrifice is only
 
 SELF-SACRIFICE. 431 
 
 temporary, that is to say, that it is after all unreal. It is still 
 more surprising when this is presented, and precisely by the 
 moralists who profess to take the loftiest theory, not merely as 
 expressing the fact, but as an a priori truth deducible from the 
 very nature of things. For what can this be but to fall back 
 upon the purely egoistic doctrine ? Self-sacrifice, says the 
 egoist, is impossible. Yes, agrees his opponent, it is impos- 
 sible. The only distinction is that the egoist denies that a 
 man can ever take anything but his own happiness for an 
 ultimate reason. The lofty moralist denies that anything 
 but his own happiness can ever be the ultimate result. He 
 thus comesto recommend a game of self-deceit like that already 
 noticed — selfishness, not abolished, but banished to the uncon- 
 scious motives. " You are to act," he says, " at every given 
 moment from a genuine desire for the good of others, but 
 vou are also to be intimatelv convinced that whatever is for 
 their good is also for your own happiness. When acting, 
 you are to ignore this esoteric doctrine; when philosophising, 
 vou are to hold to it as a necessarv truth." Nothino- can 
 exhibit the plausibility of the egoistic theory more forcibly 
 than this process, by which its professed antagonists manage 
 quickly to sidle round into adopting its fundamental tenet. 
 
 ^6. For my part, I accept the altruist theory, and I accept 
 what I hold to be its legitimate and inseparable conclusion — 
 the conclusion, namely, that the path of duty does not 
 coincide with the path of happiness. I should meet the 
 objection (so far as it can be met at all) that this is an 
 immoral doctrine, not by trying to prove that in some way 
 the immediate loss will be repaid, but by reasserting the 
 doctrine of altruism. By acting rightly, I admit, even the 
 virtuous man will sometimes be making a sacrifice; and I 
 do not denv it to be a real sacrifice ; I onlv denv that such 
 a statement will be conclusive for the virtuous man. His 
 own happiness is not his sole ultimate aim, and the clearest 
 proof that a given action v ill not contribute to it will there- 
 fore not deter him from the action. To a great extent this 
 is true of everv man who has the normal faculties. There
 
 432 THE SCIENCE OF ETHICS. 
 
 is scarcely any man, I believe, at all capable of sympatby or 
 reason, who would not in many cases unhesitatingly sacrifice 
 his own happiness for a sufficient advantage to others. Almost 
 every mother would die or expose herself to sufferings which 
 can never be repaid for the good of her infant ; and though 
 maternal love affords the most perfect example of devotion 
 to others, and is of course much stronger than most other 
 benevolent feelings, I think that the same principle is illus- 
 trated even in those commonplace acts of good nature of 
 which almost every man is capable. 
 
 37. I do not wish to exaggerate any more than to extenuate 
 the extent of this fundamental discord, I believe it to 
 exist, but I do not believe that it materially modifies the 
 ordinary statement. I take for granted that as a rule it is 
 prudent to be moral, and still more unequivocally that it is 
 prudent to encourage the morality of our neighbours. But 
 I also admit that this argument in favour of morality cannot 
 be rightly put in the form. Morality is always and necessarily 
 coincident with prudence. In exhorting a man to be virtuous, 
 we really exhort him to develop his nature upon the lines 
 which the experience of the race has conclusively proved to 
 coincide with the general conditions both of social and indi- 
 vidual welfare. This is to exhort him to acquire a quality 
 of character which, under normal conditions, and in the vast 
 majority of particular cases, will make him the happier be- 
 cause better fitted for the world in which he lives, capable of 
 wider and more enduring aims, and susceptible to motives 
 which will call out the fullest and most harmonious play of 
 all the faculties of his nature ; but it is also to exhort him 
 to acquire a quality which will in many cases make him less 
 fit than the less moral man for getting the greatest amount 
 of happiness from a given combination of circumstances. I 
 advise a man to acquire habits of temperance on a simple 
 calculation of pleasure, from wider prudential considerations, 
 and upon purely moral grounds. In each case the argument 
 is conclusive, but in each case it admits of certain exceptions. 
 Temperance will, as a rule, procure him most pleasure,
 
 SELF-SA CRIFICE. 433 
 
 because it will make him healthy ; but if he were certain to 
 die to-morrow, he might get most pleasure by being drunk 
 to-night. It will make him fitter for work, and therefore, 
 as a rule, secure him a more comfortable position ; but, in 
 particular cases, it might lose him the favour of some immoral 
 person who could do him a service. It will, again, make 
 him more virtuous, and so far a better husband and father ; 
 but it is still as before easy to imagine particular cases 
 in which the very strength of the feelings which form the 
 best guarantee for happiness may cause the most exquisite 
 pain, and make him miserable in proportion to their strength. 
 If, indeed, life were — as seems to be implied in the theories 
 of some moralists — a series of detached acts, in each of which 
 a man could calculate the sum of happiness or misery attain- 
 able by different courses, and calculate them without reference 
 to his character, the whole argument would be different. 
 But this is precisely what it is not. Every man starts with 
 an inborn set of qualities which are gradually moulded, 
 developed, or suppressed by the circumstances in which he 
 is placed, and by the inherent processes of growth and decay. 
 The happiness or misery due to any set of external condi- 
 tions depends essentially upon the disposition upon which 
 they operate. Therefore it may be, or rather it plainly is, 
 necessary for a man to acquire certain instincts, amongst 
 them the altruistic instincts, which fit him for the general 
 conditions of life, though in particular cases they may cause 
 him to be more miserable than if he w-ere without them. 
 And thus, again, the acquisition of altruistic feeling may 
 be recommended on purely prudential grounds, although 
 these grounds can never supply an exhaustive statement of 
 the motives ; and some power of altruistic feeling is pre- 
 supposed in the very capacity to become moral. But it does 
 not follow that on special occasions prudence and virtue 
 will coincide ; and, as a matter of fact, I think that they 
 often emphatically differ. 
 
 a E
 
 ( 434 ) 
 
 CHAPTER XI. 
 
 CONCLUSION. 
 
 1. I HAVE reached a conclusion which suggests further 
 controversy. What is it that I have done, supposing my 
 arguments to be satisfactory ; and what is it that, even upon 
 that hypothesis, still remains to be done ? These are questions 
 to which, for reasons to be immediately assigned, I cannot 
 give a full answer. Yet some answer is required, so far at 
 least as an indication of my own view of the case can be 
 regarded as an answer. 
 
 2. I concluded my statement by expressly admitting that 
 one great difficulty must remain unsolved. Rather, I assert 
 that it is intrinsically insoluble. There is no absolute coin- 
 cidence between virtue and happiness. I cannot prove that 
 it is always prudent to act rightly or that it is always happiest 
 to be virtuous. My inability to prove those propositions 
 arises, as I hold, from the fact that they are not true. This 
 admission does nothing to diminish our belief in the sur- 
 passing importance of morality and of its essential connec- 
 tion with social welfare ; and further, it does not diminish 
 the intrinsic motives to virtue, inasmuch as those motives 
 are not really based upon prudence. But I cannot go further. 
 I do not think that any one has really gone further except 
 in words ; nor do I see any reason to expect that any one 
 will go further. The problem is bound up in the old one of 
 the origin of evil. Given the existence of misery or vice, 
 you may conceivably ^^ explain" them by showing what are 
 the general conditions under which they exist ; but you 
 cannot, without contradiction, explain them away. Yet^
 
 CONCLUSION. 435 
 
 until you have explained them away^ your explanation is 
 pronounced to be insufficient. If so, all explanation will be 
 insufficient, so long as vice and misery exist, or rather so 
 long as it can be truly said that they ever have existed. The 
 problem of the association between misery and virtue is 
 really part of this ancient puzzle ; and when one enigma is 
 solved, perhaps the other may be clear. This, however, will 
 appear more clearly as we proceed ; for we may still ask 
 whether we cannot attempt some solution or modifica- 
 tion of our difficulties by going beyond the limits hitherto 
 accepted. 
 
 3. One distinction must be made at starting. Whatever 
 more may be said, it is perfectly clear that there is a great 
 deal more to be done. To expound the nature of any evil 
 is not the same thing as to cure it. A physiologist has ful- 
 filled his duty when he has stated what are the actual pro- 
 cesses of life, and therefore what are the conditions of 
 healthy or diseased life. The physician applies these laws 
 when he cures diseases, and the sanitary reformer when he 
 secures the observation of healthful modes of life. It would 
 be manifestly absurd to condemn the physiologist because 
 he does not make disease impossible or induce men to live 
 healthily. He has done all that is proper to him in his 
 scientific capacity when he has said, for example, "Such and 
 such are the processes implied by drunkenness, and such 
 and such are the necessary and probable consequences." He 
 shows what are the appetites gratified by drunkenness as 
 well as the evils which it causes. So long as he is a man of 
 science, he has simply to observe facts and formulate laws; 
 he is successful in so far as his observations and generalisa- 
 tions correspond to the fact, and he has nothing to do with 
 consequences or applications. If, after hearing all that he 
 has to say, I prefer drink and delirium tremens, that is not 
 his concern in his scientific capacity. His aim is the dis- 
 covery of truth, and his success is proportionate to the truths 
 which he discovers. With the physician or reformer the 
 case is of course differentk One has to cure disease and the
 
 436 THE SCIENCE OF ETHICS. 
 
 Other to suppress bad customs. The end of each is to pro- 
 duce a given result^ and his success is measured by the attain- 
 ment. And therefore, though one chief aim of physiology 
 is to furnish means for suppressing disease, yet the theory is 
 " right " as it most exactly reflects the truth, as the conduct 
 is "right" which most accurately attains the end. The 
 theory would be useless unless it led to conduct, and the 
 conduct could not succeed unless it were guided by the 
 theory; but we do not expect the theory to cure diseases by 
 itself, and therefore the correctness of his theories does not 
 excuse the physician who fails to effect a cure. 
 
 4. The same, as I take it, is the relation between the scien- 
 tific and the practical moralist. The scientific moralist has 
 fulfilled his task when he has explained what virtue and vice 
 actually are, what are their normal consequences to society 
 and the individual, and what are the conditions under which 
 they are generated. He has done all that he can do if he 
 has laid down true propositions upon such matters. He 
 tries to discover what is the " form " of morality, what is its 
 intrinsic or essential character, how good or bad men act, 
 what are their motives, and what the normal consequences 
 of their actions. The physiologist, as I have said, may ask 
 similar questions in regard to health and disease. But the 
 effect of a belief in conclusions attained, whether in ethics 
 or in physiology, wdll depend upon the character of the 
 believer. " This is wrong, and I choose to act wrongly ; " 
 " This is unwholesome, and I choose to risk my health," 
 represent states of mind not only possible but common. It 
 is part of the ethical problem to determine what character is 
 implied in them. We should contradict the commonest ex- 
 perience if we denied their possibility. The practical moralist 
 who tries to raise the standard of morals or to influence a 
 particular man must start from the science ; and his success 
 will be measured by the degree in which he affects conduct. 
 But it is an error to try the scientific moralist by the test 
 applicable to the practical moralist. His theory is sound, 
 like every other theory, so far as it explains the facts ; and
 
 CONCLUSION. 437 
 
 it must explain, and therefore admit, the existence of vice as 
 well as virtue. And this seems to l^e overlooked when an 
 ethical theory is condemned because it does not of itself 
 constrain the will as well as convince the intellect. That is 
 to confound the art with the science, or practice with theory. 
 A theory is a systematic statement of belief, and the only 
 question about a belief is in any and every case whether it 
 is true or fiilse, not whether it does or does not produce any 
 assumed effect upon conduct. In this respect the analogy 
 is complete between the scientific and practical moralist and 
 the scientific and practical physiologist. It is as idle to 
 suppose that an ethical theory will show vice to be impossible 
 as to suppose that a physiological theory will show disease 
 to be impossible. If that were the case, we should happily 
 be able to dispense with theories altogether. 
 
 5. Philosophers have attempted to evade this diflficulty in 
 many ways. They have laboured indefatigably so to state 
 the ethical principle that disobedience may be " unreason- 
 able " in the same sense as refusal to believe in a mathe- 
 matical demonstration. Every such attempt is doomed to 
 failure in a world which is not made up of working syllogisms 
 and where the w^ill and the reason mean different things. But 
 the difHculty recurs so often, that I may attempt to explain 
 my opinion by some further considerations. It is some- 
 times said that the moralist has to treat of what ^"^ ought" 
 to be, not of what "is." To which I might reply that 
 there is no real disjunction, and that it would be just as 
 reasonable to say that the medical writer treats of the health- 
 ful and the morbid, and not of the actual. The moral in- 
 stincts are realities, and in treating of them, I treat of 
 morality. But the intention of the remark seems to be 
 different. The implied assertion (if I understand it rightly) 
 is to this effect : The man of science aims in all cases at the 
 discovery of " laws," where by " law " we mean nothing but 
 generalised statements of fact. But the moralist has to do 
 something different from this; he has to establish " laws" in 
 the juridical sense ; meaning, therefore, commands, or at
 
 438 THE SCIENCE OF ETHICS. 
 
 least formulce of some kind which have a "coercive" action 
 upon the will. Hence there is, it is said^ a real difference 
 between the cases. The physicist proves (say) that the 
 planets attract each other ; and the fact so summarised takes 
 place by a mechanical necessity, or independently of the 
 human rule. The sociologist again (if the possibility of such 
 a science be admitted) obtains, or would obtain, formulae of 
 the same kind ; that is to say, " All men do so and so," 
 "Good men do this," "Bad men do that," and so forth. 
 This can never amount to more than a statement of facts. 
 The forces at play now include desires and volitions ; but the 
 man of science lays down social rules as the physicist lays 
 down his formulae, as statements of what will invariably or 
 " necessarily " happen. Assuming, therefore, that he can 
 obtain such formulae, they cannot be moral laws ; for the 
 moral law says, not, " This and this is true," but, " This and 
 this is to be done." It is a command or a law proper, not a 
 law in the derivative or scientific sense. But the moralist 
 has to discover laws proper, not laws metaphysical, and 
 therefore has to do with the " ought," not with the " is." 
 
 6. To this I must reply, so far as I dissent, by giving 
 my own view. The statement just given seems to me to 
 be accurate, but leads to an apparent difficulty only when 
 we confound the spheres of practice and theory, or confound 
 the meanino;s of " fact " and " truth." When dealing with 
 beliefs, we can only ask whether they are true or false. 
 When, on the contrary, we are dealing with commands, it 
 is nonsense to ask whether they are true or false ; we can 
 only ask whether they do or do not operate, that is, compel 
 obedience. So I do not properly "believe" or "disbelieve" 
 in a command : I simply obey or disobey it ; though I may, 
 of course, believe or disbelieve that the command has been 
 given or that certain consequences will follow disobedience. 
 But although it is not proper to use the words "fact" and 
 " truth " as though they were synonymous, every truth is, 
 of course, concerned with facts ; and there are truths and 
 beliefs which are concerned with commands as with facts of
 
 CONCLUSION. 439 
 
 every kind. And further, a belief may be necessarily im- 
 plied in a command, and in such a way that a command 
 may be the necessary result of a belief, which, indeed, must 
 generally be the case where the belief is a belief about con- 
 duct. The " law," in the case of physical inquiries, may be 
 accepted without any effect upon the character. The accept- 
 ance of the law of universal gravitation has no bearing what- 
 ever upon the ends of the believer. It will sometimes affect 
 his means, as in the case of calculations founded upon the 
 assumption of its truth; it will not affect his principles. 
 But the case is different when conduct is the subject- 
 matter of the science. The belief, then, is inseparably 
 united with a feeling. The belief that all good men act in 
 a certain w^ay must have a reaction upon my character if I 
 have any feeling in regard to goodness and to good men. 
 The effect will depend upon my previous character, but in 
 really accepting it my character will also be modified. 
 
 7. Let us apply this more closely to the case. There is 
 a law, using the word in the strict sense of " positive law," 
 against murder. From a scientific point of view this is 
 simply a statement of facts. The sociologist asserts the 
 existence of a certain social order with appropriate customs 
 and instincts, one result of which is that detected murderers 
 are liable to hanging. The psychologist recognises and tries 
 to explain the effect produced by this knowledge upon the 
 mind of an intending murderer. Their conclusions are 
 expressible as statements of general fact or scientific " laws ; " 
 murderers are hanged at certain stages of social develop- 
 ment ; all men of a certain character are deterred by the 
 fear of hanging, and so forth. The effect, again, which is 
 produced in a given case is a special instance of these laws. 
 When the scientific observer has laid them down, he has 
 done all that he can do. The man, again, who is about to 
 commit a murder knows the facts and has to choose. The 
 choice is determined not merely by the recognition of the 
 scientific truth but by the man's character. He hates his 
 enemy, and is aware that by gratifying his hatred he runs
 
 440 THE SCIENCE OF ETHICS. 
 
 a risk of the gallows. His action is the correlative of the 
 struggle between fear and hatred. His conduct is one of 
 the facts with which the man of science has to deal. But 
 in any case^ the bare intellectual recognition of the general 
 proposition, ^'Murderers are liable to hanging," is not 
 sufficient to determine them. The recognition will operate 
 differently according to his character. If it is a genuine 
 recognition, involving a realisation of all the facts implied, it 
 is but the logical aspect of an emotional process, which is 
 symbolised by the words. He must have a foretaste of the 
 various horrors connected with hanging; and the painful 
 foretaste of an appearance on the gallows will strug-gle with 
 the pleasurable foretaste of gratified revenge and determine 
 his conduct. For him, the terror, the hatred, are facts ; 
 actual forces which move him one way or the other, and 
 which are rendered possible by means of the intellectual 
 foresight, as it, again, is only possible through them; that 
 is, neither of them could exist separately. 
 
 8. We may say the same of the moral law. The socio- 
 logist and the psychologist have to describe the nature and 
 laws of action of the moral instincts, as in the other case of 
 the instincts which support the political order. And in the 
 same way, the agent who recognises the fact that murder is 
 wicked as well as criminal will act according to the resul- 
 tant of the various emotions set in play through his recog- 
 nition of that accepted principle. These emotions, again, it 
 must be repeated, may vary widely. The moral sense may, 
 perhaps generally does, correspond simply to an unreasoning 
 instinct, and the recognition that murder is wrong may 
 involve nothing but the action of this unexplained feeling. 
 It means that the hatred of the victim is checked by the 
 awakening of the acquired instinct which makes bloodshed 
 hateful. The thought, again, that murder is wrong may 
 call into play other emotions, according to the moral theories 
 accepted by the agent. It may bring to his mind the wrath 
 of his God, or the injury to society, or the hatred which 
 others would feel for him, or the prospect of lasting remorse,
 
 CONCLUSION. 441 
 
 or various other associated emotions. In any case, it means 
 the perception of other modes of contemplating the pro- 
 posed action besides that which suggested the murderous 
 impulse, and the bringing into play of the various emotions 
 bound up with these perceptions. The command is only a 
 moral command so far as it appeals to the intrinsically moral 
 motives, whatever they may be. It represents a force in so 
 far as the recoirnition of its truth involves the action of the 
 strictly moral instincts. 
 
 9. We may now return to the difficulty. It is quite true 
 that the simple scientific statement does not, upon my show- 
 ing, necessarily carry with it a governing principle. That 
 is to say that a theory of motives is not itself a motive. The 
 admission that the moral laws are statements of essential 
 conditions of social welfare will not even tend to make me 
 moral, if I care nothing for society. If I am incapable of 
 sympathy, no proof of the adv^antages of good actions to 
 others will induce me to sacrifice myself. This, however, is 
 merely to say, with every moralist who ever wrote, that the 
 bare moral maxims will do nothing without a thorough 
 training of the emotional nature. We must not merely 
 learn that our actions affect the happiness of others, but 
 must acquire the habit of feeling for them. It may, indeed, 
 be said that a nominal acceptance of the formula can 
 scarcely be called a real belief in it, unless the meaning of 
 the symbols is realised ; and so far to teach me that my 
 conduct hurts others is to make me feel for others if I am 
 capable of the sympathy, whilst it has no proper meaning 
 for me unless I am capable of sympathy. And this is so far 
 true that a genuine assimilation of the moral precepts may 
 be assumed to imply a growth of moral sentiment. To learn 
 really to appreciate the general bearings of moral conduct 
 is to learn to be moral in the normally constituted man, 
 though we must always make the condition that a certain 
 aptitude of character exists. But in saying this I have only 
 recognised the fact that beyond the science of morality there 
 is the art which depends upon it.
 
 442 THE SCIENCE OF ETHICS. 
 
 lo. This, however, is regarded as an insufficient statement. 
 After all, it is said, the moral system remains insufficiently 
 clenched. All that it comes to is, " If I am wicked, I shall 
 commit murder; if I am not wicked, I shall not." The 
 agent, therefore, simply recognises his motives, but cannot 
 alter them or properly approve or condemn them. The 
 judge can say, " I will force you to refrain from crime." The 
 moralist can only say, " I will give you reasons for refraining 
 ifyovL are already good." Hence arises the desire of moral- 
 ists to prove that there is some reason binding every man 
 simply as reasonable. The search, if I am right, is hopeless ; 
 as all motives in every case involve another element. In any 
 and every case, the compulsion, legal and moral, must 
 depend upon the character. The legal sanction appeals to 
 feelings more universal than the moral, though the moral 
 has the advantage of covering actions which the legal can- 
 not touch; but even the legal sanction may cease to be 
 effective in conceivable cases. Its highest penalty depends 
 for efficacy upon the love of life ; and there are many cir- 
 cumstances under which a man may cease to care for life, 
 and so far be beyond the power of the legislator. The 
 moralist, undoubtedly, can only appeal to people who have 
 certain instincts ; for an appeal to extrinsic motives is merely 
 a cutting of the knot. Unless a man has certain sensibilities 
 the moralist has absolutely no leverage. This I take to be 
 a simple fact, undeniable in practice, and, so long at least 
 as we have to deal with science, a fact which must be recog- 
 nised like any other. But I must guard against one misinter- 
 pretation which seems to be implied in the above statement. 
 It is true (as I hold) that if a man is without any conscience 
 you cannot move him by speaking of right and wrong ; 
 but it seems to be inferred that this is in some way a justi- 
 fication of his acting wrongly. This is a sophistry. It is 
 capable of proof — scientific proof, if you will — that murder 
 is wrong — " wrong" as opposed to the rules actually accepted 
 and regarded with reverence in every civilised society, and 
 wrong, again, as being opposed to that underlying moral
 
 CONCLUSION. 443 
 
 code to which actual morality is an approximation, and 
 which expresses the conditions of social welfare. I may, 
 therefore, prove to the murderer that he is acting wickedly. 
 But he says, '^ I have no conscience." If he adds, '^ There- 
 fore I shall commit murder," his reasoning is perfectly 
 sound. If he goes on, "Therefore I am right in committing 
 murder," he is contradicting himself. The true inference 
 is, therefore, " I am wrong by my own confession." Right 
 conduct is the conduct dictated by a conscience, or the 
 conduct (as I have put it) of the ideal man. To allege the 
 absence of good motive as a justification is nonsense, because 
 justification means the proof of good motives. It is, there- 
 fore, a simple "objective" fact that a man acts rightly or 
 wrongly in a given case, and a fact which may be proved 
 to him ; and, further, though the proof will be thrown away 
 if he is a moral idiot, that is, entirely without the capacities 
 upon which morality is founded, the proof is one which 
 must always affect his character, if we suppose the truth 
 to be assimilated, and not the verbal formula to be merely 
 learnt by rote. 
 
 II. Here we are approaching the inevitable free-will 
 puzzle. I am content to say once more that, so long as we 
 are in the region of science, it is idle to evade the plain 
 facts. Many men are wicked ; no man is perfect. No dex- 
 terous logic will make the wicked man good, nor, so long 
 as it appeals to motives which they cannot appreciate, will it 
 convince them that it is well to be good. The man of 
 science, indeed, sometimes tries to evade this painful con- 
 clusion. He points out that although men are now so 
 constituted that right conduct is often painful and virtue 
 itself not unconditionally desirable, this discord may be re- 
 solved in some better state. Progress means approximation 
 to some Utopia in which our natures will be so improved 
 that we shall always sympathise with each other, and society 
 be so happily constituted that the conduct which gives to 
 any man the greatest chance of happiness will give it an 
 equal advantage to his neighbours. Such speculations are
 
 444 THE SCIENCE OF ETHICS. 
 
 legitimate. They may be useful in defining an end towards 
 which all w^ell-wishers to their fellows may desire to act, 
 and they may serve also to bring out the fact that altruistic 
 emotion does not involve self-sacrifice in essence, but only by 
 accident. Yet it does not solve the difficulty for us. Specu- 
 lations about the future of society are rash ; and the know- 
 ledge — if we could attain the knowledge — that our descend- 
 ants would be better off than ourselves would not disprove 
 the existence of the present evil. Moreover, various doubts 
 suggest themselves. We cannot tell that progress will be 
 indefinite. It seems rather that science points to a time at 
 which all life upon the planet must become extinct ; and 
 the social organism may, according to the familiar analogy, 
 have its natural old age and death. In any case there is 
 one obvious difficulty. Progress means a stage of evolution. 
 Evolution from the earliest to the latest sta2;es means a con- 
 tinuous process of adjustment, which is always determined 
 by the fact that at any existing stage the adjustment is 
 imperfect. Complete equilibrium, or an elimination of 
 this discordant element, would therefore mean, ndt perhaps 
 stagnation, but a cessation of progress, an attainment of the 
 highest arc of the curve, after which we would only expect 
 descent. 
 
 12. Consider, in fact, what is meant by moral progress. 
 It is an essential part of a development in virtue of which 
 the society is more vigorous, the individual is bound by 
 further-reaching sympathies with his fellows, and thus a 
 more comprehensive code of conduct is recognised. The 
 member of the higher society is more fully conscious of the 
 nature and consequences of his actions than the member of 
 the lower society. In this sense, therefore, he is more moral. 
 The higher man is affected by sufferings to which the lower 
 man is indifferent. And thus, again, conduct which would 
 require abnormal excellence at the lower stage has become 
 normal as the corresponding instincts have become definitely 
 organised in the higher. But, in another sense, the moral 
 progress is less clear; that is to say, the savage deviates
 
 CONCLUSION. 445 
 
 less frequently than the civilised man from the code recog- 
 nised in each case. The savage law is lower, but it is more 
 regularly observed. So, if we go back to the animals, in 
 whom morality proper does not exist, the obedience to instinct 
 is more regular still. Sin comes throuu;h the law, as it is only 
 when the agent is capable of laying down general rules that 
 be begins to be sensible of deviations from them. If we 
 measure both stages of development by the same law, we 
 may say that the higher is more moral than the lower. If 
 we measure each by its own law, the observance is more 
 uniform in the lower than the higher; as, in fact, a greater 
 mobility and flexibility of character seems to be necessarily 
 implied in the higher type. Moral progress involves a con- 
 stant laying down of new problems. Old evils are avoided, 
 old hostilities reconciled, the whole life is fuller and more 
 vigorous ; but the process implies at the same time that the 
 new capacities and sensibilities developed constantly bring 
 with them new evils or difficulties which again require to be 
 reconsidered. A nation acquires new mechanical powers, 
 but it has not yet the habits necessary to turn them to 
 account, and they are so many temptations to immoral in- 
 dulgence. Sympathy expands, but, as not guided by know- 
 ledge, leads to rash changes productive of evil as well as good. 
 To improve, whether for the race or the individual, whether 
 in knowledge or sympathy, is to be put in a position where a 
 new set of experiments has to be tried, and experience to 
 be bought at the price of pain. And as this seems to be 
 esssentially implied in all progress that we can imagine, I 
 see no reason to suppose that pain will be eliminated, or that 
 it will be so distributed that there shall never be a diver2;ence 
 between the painful and the pernicious either to man or 
 society. From the scientific point of view we may hold that 
 evolution implies progress — progress at any rate to a point 
 beyond our present achievements ; and, further, progress im- 
 plies a solution of many discords, and an extirpation of many 
 evils J but I can at least see no reason for supposing that it
 
 446 THE SCIENCE OF ETHICS. 
 
 implies the extirpation of evil in general or the definitive 
 substitution of harmony for discord. 
 
 13. We may, however, and, according to many thinkers, 
 we must, go beyond this point of view. We may find in 
 metaphysical speculation or in the transcendental world 
 which it reveals to us an answer to the doubt and a solution 
 of the discord. Here, indeed, I might stop; for I started at 
 the outset with a disavowal of any desire to go beyond the 
 scientific boundary. I should have to conclude by challenging 
 the metaphysician to dispute the truth of my conclusions, 
 and should urge that they would have to be recognised upon 
 any metaphysical theory, although different schools would 
 interpret them in different ways. And I might further 
 decline to follow the metaphysician into his own ground. 
 This, I might say, is the meaning to be attached to morality 
 so long as we remain in the world of experience ; and if, in 
 the transcendental world, you can find a deeper foundation 
 for morality, that does not concern me. I am content to 
 build upon the solid earth. You may, if you please, go down 
 to the elephant or the tortoise. I think it better, however, 
 to indicate my own view in the briefest way possible. For, 
 in the first place, many thinkers consider that it is necessary 
 to begin at the very beginning ; and to solve the whole 
 problem of the universe before they get down to morality. 
 A discussion which does not try to settle the issues between 
 materialism and idealism, and between theism, pantheism, 
 and atheism, and investigate the nature of the ego, is from their 
 point of view simply frivolous. It touches the barest outside 
 of the question. And, in the next place, T feel that I may 
 be accused of unfair reticence; unless I proceed further, my 
 antagonists will hold that my dismissal of certain problems 
 to the metaphysician is a juggle, and that my real meaning 
 is that the problems in question are barren logomachies, 
 which may be left to the manufacturers of intellectual cob- 
 webs for their amusement, but not for the instruction of 
 mankind. In which, indeed, there seems to me to be a great
 
 CONCLUSION. 447 
 
 deal of truth ; only I do not wish to conceal my opinions, so 
 far as I have any settled opinions. 
 
 14. To ffive a satisfactory answer, I should have to write 
 a metaphysical treatise; for without such a discussion I could 
 not define accurately the boundaries between metaphysics 
 and science. And I must add, that one difficulty which I 
 feel is that mv own mind is not as clear as I could wish upon 
 these questions. I can, however — or so I hope — indicate 
 briefly certain crucial points. And first of all, I will say that I 
 utterly disbelieve in any so-called ontology. I regard it as a 
 barren region haunted by shadowy chimeras, mere spectres, 
 which have not life enough in them even to be wrong, non- 
 entities veiled under dexterously woven masses of verbiage. 
 I don't believe that any one can, by any device whatever, 
 spin out of his own mind a demonstration of the ultimate 
 nature of things in general; and that is what the ontologist 
 substantially tries to do. The direct arguments against such 
 systems are not more conclusive than the futility of the 
 supposed conclusions. The ontologist proves something with 
 infinite display of logic; and if you admit it to be proved, 
 you find that the admission leaves you exactly where you 
 were before. And if this has an arrogant sound, I can only 
 say that it is the only conclusion compatible with a due re- 
 spect for philosophers. For when 1 look at the vast systems set 
 forth and supported by arguments devised by the most famous 
 thinkers, and consider their instability, their helpless incapa- 
 city to withstand assaults or to solve real difficulties, I can 
 only infer either that their authors were fools — which would be 
 the height of arrogance — or that they were attempting pro- 
 blems beyond the reach of reason. The greatest athlete can- 
 not get off his own shadow. 
 
 15. The desired consummation of some approximation to 
 unity is, therefore, as I hold, to be sought by distinguishing 
 the metaphysical from the scientific problem. Ethical in- 
 vestio-ations, like others, will have some definite results 
 when we turn to what are called historical methods of 
 inquiry. This is true, I think, of the allied inquiries into
 
 448 THE SCIENCE OF ETHICS. 
 
 social, and political, and aesthetic inquiry. The tendency of 
 modern speculation to take that form, and to look into the 
 history of the past for an answer to problems which were 
 once attacked by looking simply into our own minds, im- 
 plies a recognition of this principle. And I believe that we 
 may look for some approximation to agreement as the 
 method is more generally adopted and more systematically 
 carried out. I need not dwell upon the nature of the 
 advantage thus gained. It is enough to say that when we 
 cease to ask '^What is the beautiful or the moral?" and ask 
 " What have men actually admired ? " we get rid of the many 
 illusions generated by mistaking our own special tastes for 
 universal tastes ; that we take into account that " social 
 factor," without a recognition of which no tenable theory 
 can be put together, and that we have an advantage com- 
 parable to that gained when we see a large arc of a curve 
 instead of the infinitesimal fraction perceptible at a given 
 moment. To this it is often replied that an account of how 
 things come to be is after all radically different from an 
 account of what they are. I do not quite admit the radical 
 difference, for I think that the true nature of a thing may 
 reveal itself only when we see it on a large scale. But I 
 admit that there is something more to be done, and that, 
 after all that can be said by the man of science, we must 
 still come upon the old metaphysical problems. In what 
 form, then, do they enter our investigation ? 
 
 1 6. When we have taken leave of ontology, is there any- 
 thing left for the metaphysician? Metaphysics has been 
 described as a theory of knowledge, a systematic account of 
 all that can be said about knowledge as knowledge. This 
 must not be understood as though we could explain what is 
 meant by knowledge in terms of something different from 
 knowledge. To attempt to give an answer of that kind 
 would be to attempt to get outside ourselves, and to fall into 
 the sham philosophy of the ontologists. We may still, it is 
 said, be able to understand the intimate nature of knowledge, 
 to regard it as a vast process in which every particular de-
 
 CONCLUSION. 449 
 
 partment is organically connected with the rest, as the 
 special manifestation of some universal principles. To try to 
 define this vague suggestion would be to discuss the most 
 thorny metaphysical problems. It is enough in this place 
 if I endeavour to suggest certain conclusions. It is much 
 easier to say what metaphysics are not than to say what they 
 are. And I say, in the first place, that though it may be 
 conceivably possible to exhibit the most general forms of 
 knowledge and understand their relations, no such theory 
 would enable us to get from theory of knowledge to theory 
 of facts. We may twist and turn as long as we please, but 
 the theory of knowledge will not in the least help us to dis- 
 cover the actual constitution of the world. We cannot 
 make the slightest approach to explaining why human beings 
 should be constituted as they are, why they should have 
 such and such senses and faculties, and no more, or why 
 matter should be distributed in the actual combinations ; and 
 therefore, as it seems to me, wherever there is a question of 
 facts, we must find some ground of knowledge outside of 
 metaphysical inquiries. And if, in the next place, we ask 
 What can metaphysics do for us? I can only reply indirectly: 
 A man is a sound reasoner when his thoughts accurately 
 reflect the external world. And this, as we come to see, 
 implies not merely correct performance of logical opera- 
 tions, but the elimination of certain fundamental illusions 
 generated in earlier stages of thought. Sound metaphysics 
 consist, in Berkeley's phrase, in the laying of the dust which 
 we have ourselves raised, in the removal of certain scaffold- 
 ings which have been useful in the building up of the edifice, 
 and the perception that they are not essential parts of the 
 structure; briefly, in becoming distinctly conscious of our 
 own mental operations, and therefore destroying fallacies 
 which lie at the base of all logical procedure. If this be any- 
 thing like the accurate statement — and I can only speak in 
 the roughest way — we may say that metaphysical inquiry is 
 of the highest importance in this sense, that our reasoning 
 cannot be thoroughly sound unless we are sound metaphysi- 
 
 2 F
 
 450 THE SCIENCE OF ETHICS. 
 
 cians^ but that we must not expect from metaphysics any 
 system of definite statement of fact. And further^ we must 
 infer that legitimate metaphysical investigations affect know- 
 ledge in general, but have no special relation to any par- 
 ticular department of knowledge. 
 
 17. My view, therefore, is that the science of ethics deals 
 with realities ; that metaphvsical speculation does not help us 
 to ascertain the relevant facts; and, therefore, that it has no 
 more relation to ethical science than to any other branch of 
 knowledge. Our statement of the facts will be modified as 
 our metaphysical creed changes; the approximation to precise 
 and lasting truth will be made closer, and, so far, the science of 
 ethics will be improved by metaphysical progress. But ethical 
 principles are onlv affected as the principles of all other 
 sciences are affected. Though I cannot offer any proof of 
 this doctrine, I shall try briefly to show how it bears upon 
 certain cardinal points; and why, in my opinion, it is useless 
 to look for any further light from metaphysical inquiries. 
 This is virtually to challenge the metaphysician to show that 
 he is of anv use in the matter. 
 
 18. The first principle which I have sought to establish is, 
 briefly, that a moral rule is a statement of a condition o^ 
 social welfare. Temperance, according to me, is proved to 
 be immoral by the same methods which prove it to be un- 
 wholesome. Scientific observation shows that it is productive 
 of diseased states either of the individual or of society. In great 
 part the arguments are identical, and though the moralist 
 has to go into questions not considered by the physioloo-ist, 
 he still uses the same methods. Moreover, it is by scientific 
 observation alone that the facts can be proved ; for by the 
 purely metaphysical method you cannot even approach the 
 relevant questions, or prove even the advantages of drinkinsr, 
 much less of moderate drinking. What is true in this case 
 is true, according to me, of all moral problems. They can 
 only be examined when we have some knowledge of the 
 organisation of man and of society which is unattainable by 
 any other than the scientific method. And I have further
 
 CONCLUSION. 451 
 
 urged that moral systems have actually grown up through a 
 recognition, more or less explicit, of these truths. Morality 
 has been regarded as imposed by the will of a deity; but then 
 it is taken to be the will of a good deity; that is, of a deity who 
 wishes well to mankind. From the purely utilitarian point 
 of view, it is regarded as a statement of the conditions of 
 happiness; but the happiness is that of beings whose character 
 is already determined at all points by their being part of the 
 social organism, and the sentiment can only become part of 
 the social instinct when it is purified from any individual 
 aberration. 
 
 19. If this, which it has been my purpose to prove, be 
 admitted, what is the relation of the ethical to the meta- 
 physical theory? The metaphysician, as I take it, has a 
 locus standi only so far as he can throw light upon the general 
 conditions of belief, or, again, upon the nature of certain ulti- 
 mate distinctions, such as that of object and subject, taken 
 for granted in the scientific statement. The ontologist, 
 indeed, and the theologian claim to do more; for ontology, 
 as I believe, is but the spectre of theology, the caput mortuum 
 to which it is reduced when the attempt is made to purify it 
 from all sensuous and anthropomorphic elements. But what 
 more is required ? If you assert that morality does not coin- 
 cide with the conditions of social welfare, you have not only 
 to meet the direct aro;umentsalle2;ed in favour of the identitv, 
 but to give some other criterion of morality. If, for ex- 
 ample, the criterion be conformity to the will of a deitv 
 who commands men to transgress the conditions of social 
 welfare, we have to meet the reply that the deity is malevolent 
 and commands a false morality, or rather imposes an im- 
 moral rule. Even the most ascetic system, according to me, 
 carries with it a tacit reference to social welfare, and the 
 deviations are due to errors of judgment as to the facts. In 
 any case, if the criterion be rejected, morality is arbitrary 
 until some further criterion be suggested. The impossibility 
 of any such attempt is so clear, that it may be almost taken 
 for granted that the coincidence between morality and the
 
 452 THE SCIENCE OF ETHICS. 
 
 conditions of social welfare is admitted. But if so, the rule is 
 given by the conditions. If it be admitted tliat this coin- 
 cidence exists in fact ; if, again, it be admitted that the 
 social utility of certain kinds of conduct and character can 
 be proved, and that, if proved, they become for that reason 
 "right" and "virtuous," then it seems to follow that the 
 moral law is actuallv established in the way suggested, namelv, 
 by observation of the facts. You may perhaps prove that 
 the deity approves obedience to the law thus established ; 
 but for the mere fact of ascertaining the law, the observation 
 is the sole and sufficient method. Nor can vou know that 
 the deity approves otherwise than by knowing that he is good, 
 that is, desires the welfare of mankind. 
 
 20. The same arguments take a different form in the case 
 of metaphysical discussion. The ontologist insists that 
 morality cannot be established until "materialism" be con- 
 futed. To this I should answer, that for scientific purposes 
 the discussion is irrelevant, I can prove that unselfishness 
 and all the qualities which require intellectual development 
 are essential conditions of social welfare. If the materialist 
 can prove that the emotions and the intellect are in some 
 sense mere dances of atoms, I can only reply that in that 
 case atoms have more in them than I should have supposed, 
 but that the phenomena which I take into account remain 
 the same. And meanwhile, from the scientific point of 
 view, I can obtain a solution of the real problem, which the 
 ontologist cannot approach. For as the organism has both 
 physical appetites and intellectual emotions, and as the condi- 
 tions of social welfare require a certain balance between the 
 faculties, and not the abolition of either class of faculties, 
 the study of those conditions gives a measure of the degree 
 in which the intellect should be developed; so far, namely, as 
 is consistent with the health of the whole organism ; whereas 
 any absolute conclusion as to the merit of intellectual 
 and physical parts of our nature abstractedly from the 
 conditions of life, would be futile because absolute. The 
 metaphysician, however, more frequently takes a position
 
 CONCLUSION. 453 
 
 corresponding to that of the theologian who allows the 
 coincidence between morality and social welfare, and only 
 disputes as to the mode of proof. The metaphysician dis- 
 tinguishes between the form and the contents of duty. He 
 does not deny that in any special case we must discover 
 what is right by observing the facts, and that the fact that 
 a given action will make men happier is so far a proof that 
 it is a right action ; but he argues that the conception of 
 " duty/' of " ought," or of " right and wrong," must be 
 independently obtainable. The question, which involves the 
 basis of a whole metaphysical system, cannot be adequately 
 argued here. But I may answer briefly, first, that even 
 upon this assumption, the science, if there be a science, of 
 the conditions of social welfare must be of essential impor- 
 tance to morality; for, if it leads to any conclusions, those 
 conclusions will immediately determine the morality of some 
 classes of conduct. If I can prove drunkenness to be socially 
 mischievous, I shall certainly prove it to be wicked. The blank 
 concept " right and wrong" will be of no use till I can pro- 
 vide it with contents upon which to operate ; and therefore 
 the science of ethics will still be of paramount importance 
 if it can be established. And, secondly, I answer that, upon 
 my showing, this is really a perversion of a plain principle. 
 In the metaphysical region you may come across ultimate 
 canons of truth, but by no conceivable ingenuity upon prin- 
 ciples of action. And, in fact, we find accordingly that 
 thinkers who accept this position have to make the most 
 desperate efforts to twist the " ought " out of the " is." The 
 ultimate principles which they propose are simply logical 
 principles in a thin disguise. So, for example, the famous 
 theorem that you are to act so that your rule of conduct may 
 be a rule for all men is a transfigured bit of logic. To think 
 so that my thoughts may be true for everybody, is perhaps a 
 description of thinking that is objectively true. But, as I 
 have already argued, the theorem as stated is strictly true of 
 all conduct whatever. I cannot help acting in the way in 
 which every one would act with my character and under
 
 454 THE SCIENCE OF ETHICS. 
 
 the same circumstances ; nor, upon my showing, has this 
 proposition more to do with moral conduct than with any 
 other kind of conduct ; nor has it more to do with the 
 narrowest and most prudential than with the most sym- 
 pathetic and self-sacrificing principles. And this is only 
 what we must expect, if, in fact, metaphysical inquiry has, 
 as I maintain, no special relation to ethics, and can only be 
 forced into relation with it by ingenious sophistry. 
 
 21. Let us pass to the next head under which I have con- 
 sidered morality. It is argued that without some transcen- 
 dental principle there is no adequate foundation for morality; 
 and this, not in the sense just considered, that the moral law 
 remains indeterminate, but in the sense of our having no 
 security for its permanence, I have already explained at 
 full length my own theory. Morality, I have said, is a. 
 product of the social factor; the individual is moralise d 
 through his identification with the social organism ; the 
 conditions, therefore, of the security of morality are the 
 conditions of the persistence of society ; and if we ask from 
 the scientific point of view what these conditions are, we can 
 only reply by stating that the race is dependent upon the 
 environment; by tracing, so far as we are able, the conditions 
 under which it has been developed, and trying to foresee the 
 future from the past. We may, again, appeal to the meta- 
 physician if we want to examine into the validity and the 
 various implications of the ultimate principles involved in our 
 argument. Only from science, if from science, can we learn 
 anything, however trifling, as to the order of actual existence 
 in the present or the future. Our hopes and our fears must 
 be alike based upon our experience, and upon nothing else. 
 The theologian who pronounces this procedure to be in- 
 sufficient has to meet the dif^culty after his own fashion. 
 We can only feel confidence in the existence of that which 
 rests upon a transcendental basis. Morality has a pre- 
 carious position, therefore, unless we believe in a moral 
 governor of the universe. But here immediately arise the 
 whole array of insuperable difficulties over which theologians
 
 CONCLUSION. 455 
 
 have vainly struggled since theology existed. It is enough 
 to hint at them. What is a moral governor? A governor 
 who makes his subjects moral ? No, for many men are 
 wicked. Could he prevent their being wicked ? If he could 
 and did not, we have no security for his willing them to be 
 moral. If he did not because he could not, his willing them to 
 be moral will not secure the existence of morality. He can, 
 however, secure the punishment of the wicked; and therefore, 
 if moral government does not mean the immediate morality 
 of the universe, it means the ultimate rule of justice. What, 
 then, is justice? It means, in brief, that punishment should 
 be impartial, and should be proportioned to the offence. 
 Now, if the wicked are to be punished impartially, they must 
 be punished in proportion to their wickedness; that is, for 
 the wickedness due to their own character, not to the acci- 
 dental circumstances. But their own character means their 
 innate qualities, that is, the qualities with which they 
 were created ; or, in other words, the Creator as governor 
 punishes them exactly for being what he made them. Again, 
 the punishment is " proportioned " to the offence. How is 
 the proportion to be determined? Surely the perfection of 
 human justice is measured by its efficiency. That system is 
 best which most diminishes crime. But if we apply this rule 
 to divine justice we get into hopeless difficulties. We must 
 suppose that the Creator wishes to diminish wickedness as 
 much as possible, for otherwise he would inflict useless suf- 
 fering. Yet we have to suppose that he inflicts punishments 
 — infinite and eternal, according to the most logical theo- 
 logians — in such a wav that the reforming influence is a 
 minimum and the sufferino- a maximum. If a human ruler 
 admitted that the punishments inflicted by his laws had very 
 little deterrent effi^ct, but argued as a set-off that he kept the 
 greatest part of his subjects in perpetual confinement and 
 incessant torture, we should certainly say that, whether 
 by his misfortune or fault, he had a very ill-regulated king- 
 dom. Yet, when we try to reconcile ourselves to the exist- 
 ing evils by assuming the existence of this supernatural
 
 456 THE SCIENCE OF ETHICS. 
 
 balance, we necessarily present the universe after this 
 fashion. Whether it is an edifying theory or not I can- 
 not say. I do not see how it helps to strengthen our 
 belief in the safeguards of morality. The explanation is 
 simple enough. The world is what we see it, abounding in 
 misery and wickedness. If you believe in a moral governor, 
 you are bound to put extraordinary limitations upon his 
 power to vindicate his benevolence, or to limit his benevo- 
 lence in order to vindicate his power; and, in either case, 
 you take away with one hand that safeguard to morality 
 which you give with the other. Meanwhile^ in any case, 
 you have to stop all logical gaps by talking about mystery. 
 It is simpler to admit that the whole is a mystery, and to 
 cease the effort to play ourselves with words. 
 
 23. The ontolon;ist sees the weakness of the theolocrical arsxu- 
 ment and tries to remedy it. The puzzle is that virtue may be 
 painful and vice pleasant. You cannot really get rid of this 
 difficulty by adding extrinsic considerations, by assuming that 
 virtue will always meet with a reward either from man or 
 God, and so the initial error be made up by a subsequent 
 compensation. The hypothesis cannot be made to work, and 
 makes a fresh breach with every attempted repair. The 
 ontologist therefore cuts at the very root of the difficulty, 
 not by saying that the balance will be redressed, or by admit- 
 ting the mystery — for to him there is no mystery — but by boldly 
 asserting that evil is merely a negative term, a privation, not 
 a positive existence, or, in short, that in some way or other it 
 does not exist at all. If it is any satisfaction to anybody to 
 repeat such phrases, it would be a pity to deprive them of it, 
 
 23. I come to the final question about morality, and the 
 most difficult. Must we not go to transcendental considera- 
 tions in order to find a sufficient motive for moral conduct ? 
 This brings us back to the previous argument. To say what 
 morality is — I repeat the statement once more — is by no 
 means the same thing as to enforce morality. It is not 
 when one is approaching the last page of an ethical essay 
 that one can have any illusions upon that point. Un-
 
 CONCLUSION. 457 
 
 doubtedly a man may have very clear conceptions about 
 the nature of morahty, and yet may receive from that 
 cireumstance a very sHght impulse towards being himself 
 moral. Beyond the science, I have said, lies the art ; when 
 we have admitted the truth, it has still to become an operative 
 influence in moulding our characters. The abstract formula 
 may be admitted without producing any vivid representa- 
 tion of concrete facts ; the imagination has to be stimu- 
 lated as well as the intellect to be convinced ; and there is 
 still much to be done before the character is moulded, so that 
 the actual impulse shall in each case represent the general 
 principle. A man may be forced to comply with external 
 morality by some appeal to extrinsic motives ; but to make 
 him really moral we must stimulate the intrinsic motives; 
 and this must be in all cases the product of the pressure put 
 upon him in gradually imbibing the principles developed by 
 the social factor. This is recognised in the statement that a 
 religion is always more than a morality. It is not a mere 
 statement that certain rules of conduct are desirable, but it 
 is such an embodiment of some theory of the universe as 
 may impress the imagination and govern the emotions. 
 
 24. Here, in fact, arises a vast problem or series of pro- 
 blems at which only the briefest glance is possible. The rela- 
 tion between morality and religion suggests at once whole 
 libraries of controversy and lifetimes of investigation. A 
 religion implies a theory of the universe. It rests upon some 
 doctrine as to the ultimate facts ascertainable about human 
 life and the world in which we live ; and therefore, of 
 course, every moral theory is based upon, or at least closely 
 implicated in, the religious doctrines of the persons who hold 
 it. Whatever is meant by " best," we can only say what is 
 best for man as man by considering him as part of the 
 general system. But many relations are equally possible. 
 The theory, for example, that the universe is governed by 
 some inscrutable and invisible power may be worked out 
 with little reference to morality j the gods may be conceived
 
 438 THE SCIENCE OF ETHICS. 
 
 as indifferent to moral considerations, and when a moral 
 system is evolved they may take up very different attitudes 
 in regard to it ; whilst, again, in some systems the supreme 
 power may be regarded as essentially moral, and perhaps as 
 revealed to us through the conscience. In any case, where 
 there is a philosophical religion the ethical doctrine will 
 doubtless be vitally connected with the religious, and will 
 be stated partly in religious language ; and therefore every 
 srreat relio-ious crisis has a moral reaction. Sometimes it is 
 due to a moral revolt: the gods are opposed to the interests 
 of morality, and are punished by being abolished. In other 
 cases, they may be on the side of morality, and a disbelief in 
 their existence, due to some other cause, may weaken the 
 respect entertained for the moral rule. To examine into the 
 complex processes of thought and feeling thus set up must 
 always be a task of immense difficulty. I only hint at such 
 questions to draw one conclusion. A religion, so far as it 
 is moral (for, of course, many religions have a very question- 
 able relation to morality), must act by stimulating the intrin- 
 sic motives to morality. Further, it can act only through 
 the genuine belief which it embodies ; and thus when we 
 wish to estimate the effect of a religion upon morality, we 
 come at once to a further problem. 
 
 25. In what sense, we have to ask, is a belief the ultimate 
 condition of conduct? We may, as it is easy to see, fall 
 into great diff culties, unless we can say also what governs 
 the belief. So long, indeed, as a man has a certain belief, 
 we may say that it does not matter what was the origin of 
 the belief, nor whether it were true or false. A man will 
 refrain from conduct if he fears to be punished, whether the 
 punishment dreaded is to come from a real or an imaginary 
 being. The fear of hell, so long as people believe in hell, 
 may be a genuine restraint just as much as the fear of the 
 o-allows. But there is an obvious difference between the 
 cases. The fear of a real enemy may be due to actual ex- 
 perience. We explain the belief sufficiently by supposing
 
 CONCLUSION. 459 
 
 the enemy to exist and to be perceived. But when wc 
 have to deal with the imae:inarv enemy, we have alwavs a 
 possible explanation of a different kind. There must have 
 been some cause for the imagination ; and that cause may 
 be something which is not removed, and which is still opera- 
 tive when the existence of the imaginary enemy is disproved. 
 If the fear of the enemy be the sole cause of abstinence from 
 the conduct, the conduct will cease to be disagreeable when 
 the enemy is no longer feared. But the case may be entirely 
 diflferent. Perhaps the conduct was found to produce dis- 
 agreeable consequences ; they were erroneously explained as 
 due to the hostility of an imaginary being; and though the 
 being is proved not to exist, the same consequences will 
 result, and will, as soon as recognised, still dictate abstinence 
 from the conduct. 
 
 26. This, of course, applies to the case of so-called religious 
 sanctions. Is the dread of hell the cause of abstinence from 
 vice ? It may no doubt be the immediate cause for the in- 
 dividual. If a man believe that drunkards will be damned, 
 that is for him a sufficient cause for abstainino- • and if vou 
 prove to him that hell is a fiction, he may become a drunkard. 
 But this is an obviously inadequate account of the whole 
 phenomenon. If this belief in hell were the result of scientific 
 inquiry, if the tendency of drunkenness to produce damna- 
 tion were proved like its tendency to produce delirium 
 tremens, the explanation would be sufficient. We should 
 say simply that the existence of a known place of torment 
 was one of the causes which limited drunkenness. But if 
 hell be an imaginary place we must necessarily go further. 
 People are afraid of being damned for drunkenness, and 
 therefore they do not drink. But why do they anticipate 
 damnation as a consequence of drunkenness? Obviously 
 they must think it hateful for some independent reason. 
 They think that drunkards will be damned because they 
 think drunkenness hateful ; or at any rate, the belief in the 
 damnation of drunkards has arisen from a perception of its
 
 46o THE SCIENCE OF ETHICS. 
 
 other evil consequences. The supposed ultimate ground, 
 therefore, of the dislike is itself a corollary from the dis- 
 like. We must distinguish between the social and the 
 individual creed. A given person may be influenced solely 
 by the belief which he has accepted from his neighbours, 
 whether it be true or false. But the belief has been developed 
 in the society from a perception of the evil, and is a product 
 of that perception, not the determining cause. 
 
 27. Thus the true statement of the case will be, upon my 
 theory, that the limiting and determining cause of the moral . 
 objection to vice is in all cases measured by the percept i^ ui 
 of the social evils which it causes. Whilst the society is 
 permeated by a belief in the supernatural, this perception 
 has to express itself in terms of the supernatural sanction ; 
 and to a man who is a member of such a society, who really 
 explains the phenomena in general as significant of the action 
 of a supernatural being, it is the same thing to say that 
 conduct is harmless, and to say that it is not punished super- 
 naturally. But as the belief in such interference decays, 
 the perception of the pernicious consequences which expressed 
 itself in terms of hell may use a difterent language without 
 being therefore less efficacious. The motive alleged in the 
 old dialect was nominally indeed of infinite weight, but the 
 effect upon conduct was entirely disproportionate, because, 
 in the first place, it did not correspond to a genuine belief, 
 and, in the second place, the imaginary penalty would always 
 be supposed capable of evasion by imaginary remedies. The 
 difference upon this showing is not in the strength of the 
 motive, but in the dialect which has to be used to accommodate 
 it to the prevailing system of belief. The ethical sentiment 
 becomes, however, stable and demonstrable when that which 
 is the real cause of its development is recognised as being 
 also its sufficient reason, and when people find in the various 
 motives which command a conformity to the social interest 
 
 a sufficient ground of conduct. 
 
 28. I cannot expand this further at present. I conclude by
 
 CONCLUSION. 461 
 
 one remark which it suggests, and which seems to require 
 explicit statement. It is sometimes said that science cannot 
 provide a new basis of morality j and this is urged as though 
 it were an objection. I at least must thoroughly accept the 
 statement. What science proves, according to me, is pre- 
 cisely that the only basis of morality is the old basis ; it 
 shows that one and the same principle has always determined 
 the development of morality, although it has been stated in 
 different phraseology. And, moreover, this principle is not 
 the suggestion of any end distinct from all others. The 
 great forces which govern human conduct are the same that 
 they always have been and always will be. The dread of 
 hunger, thirst, and cold ; the desire to gratify the passions ; 
 the love of wife and child or friend ; sympathy with the 
 sufferings of our neighbours; resentment of injury inflicted 
 upon ourselves — these and such as these are the great forces 
 which govern mankind. When a moralist tries to assign 
 anything else as an ultimate motive, he is getting beyond 
 the world of realities. If a theologian tells me to love my 
 mother because God commands me to love her, he is invert- 
 ing the true order of thought. My love of those who are 
 nearest to my sympathies must be the ultimate ground of 
 any love that I can have for anybody else. My desire for 
 the welfare of my race grows out of my desire for the wel- 
 fare of my own intimates; and that exists independently 
 of any ethical theory whatever. A theoloirical basis of 
 morality is conceivable so far as the supreme being is repre- 
 sented as knowable and lovable; but to order morality in 
 the name of logical consistency is reasonable only when 
 I can stir men's blood by assuming that two and two make 
 four. On my theory, then, the moralist assigns no new 
 motives ; he accepts human nature as it is, and he tries to 
 show how it may maintain and improve the adv^antages 
 already acquired. His influence is little enough ; but, 
 such as it is, it depends upon the fact that a certain har- 
 mony has already come into existence ; and that men are
 
 462 THE SCIENCE OF ETHICS. 
 
 therefore so constituted that they desire a more thorougli 
 solution of existing discords. A sound moral system is 
 desirable in order to give greater definiteness to the aims 
 and methods ; and it is doubtless important to obtain one 
 in a period of rapid decay of old systems. But it is happy 
 for the world that moral progress has not to wait till an 
 unimpeachable system of ethics has been elaborated. 
 
 THE END. 
 
 I'RINTED BV BALLANTYNE, HANSON AND CO. 
 EDINbURGH ANU LONDON 
 
 ^ 

 
 university ol California ^ 
 
 SOUTHERN REGIONAL L1BRABV_FAC,L^33, 
 
 ^°= tls'ANGELls, c'uf'oRNIA 90095-1388 
 
 L 
 
 Form L9-11
 
 UC SOUTHERN REGIONAL LIBRARY FACILITY 
 
 / 
 
 t 
 
 PLEA?5: DO NOT REMOVE 
 THIS BOOK CARDz!] 
 
 ^ILIBRARY(9/C 
 
 ^ 
 
 lo - 
 
 University Research Library 
 
 "^8 01205 2428 
 
 _;3 
 
 _J 
 
 .J 
 
 J 
 
 J 
 
 1 
 
 J3 
 
 C 
 
 o 
 o 
 
 0) 
 
 CD 
 
 1\J 
 
 ;t 
 
 > 
 
 c 
 -1 
 
 X 
 
 o 
 
 70 
 
 V
 
 ■1'^' ■''■!!»'' Will 1 "I ft#K'nTO'P'»«T^^