{"collection":"philosophy-ethics","full_text_available":true,"graph_url":"https://unchartedknowledge.5gfusion.net/v2/api/page/gutenberg_53430/graph","html_url":"https://unchartedknowledge.5gfusion.net/v2/page/gutenberg_53430","id":"gutenberg_53430","license":"Public Domain (US)","rights_basis":"gutenberg_metadata_copyright_false","source":"gutenberg","source_url":"https://www.gutenberg.org/ebooks/53430","summary":null,"tags":["Category: Philosophy & Ethics","Ethics"],"text":"TRANSCRIBER\u2019S NOTE\n\n  * Italics are denoted by underscores as in _italics_.\n\n  * Bold text is denoted by equals as in =bold=.\n\n  * Small caps are represented in upper case as in SMALL CAPS.\n\n  * Obvious printer errors have been silently corrected.\n\n  * Original spelling was kept, but variant spellings were made\n    consistent when a predominant usage was found.\n\n  * Footnotes have been renumbered into a single series. Each footnote\n    is placed at the end of the paragraph which includes its anchor.\n\n\n\n\nPRINCIPIA ETHICA\n\n\n\n\n  CAMBRIDGE UNIVERSITY PRESS\n  C. F. CLAY, MANAGER\n  LONDON: FETTER LANE, E.C. 4\n\n  [Illustration]\n\n  NEW YORK: THE MACMILLAN CO.\n  BOMBAY   }\n  CALCUTTA } MACMILLAN AND CO., LTD.\n  MADRAS   }\n  TORONTO: THE MACMILLAN CO. OF CANADA, LTD.\n  TOKYO: MARUZEN-KABUSHIKI-KAISHA\n\n  ALL RIGHTS RESERVED\n\n\n\n\n  PRINCIPIA ETHICA\n\n  BY\n  GEORGE EDWARD MOORE\n\n  LITT.D. CAMBRIDGE, HON. LL.D. ST ANDREWS,\n  FELLOW OF THE BRITISH ACADEMY,\n  LECTURER IN MORAL SCIENCE IN THE UNIVERSITY OF CAMBRIDGE\n\n\n    \u201cEverything is what it is,\n    and not another thing\u201d\n      BISHOP BUTLER\n\n\n  CAMBRIDGE\n  AT THE UNIVERSITY PRESS\n  1922\n\n\n\n\n  _First Edition_ 1903\n  _Reprinted_ 1922\n\n\n\n\n  DOCTORIBUS AMICISQUE CANTABRIGIENSIBUS\n  DISCIPULUS AMICUS CANTABRIGIENSIS\n  PRIMITIAS\n  D. D. D.\n  AUCTOR\n\n\n\n\nPREFACE.\n\n\nIt appears to me that in Ethics, as in all other philosophical studies,\nthe difficulties and disagreements, of which its history is full, are\nmainly due to a very simple cause: namely to the attempt to answer\nquestions, without first discovering precisely _what_ question it\nis which you desire to answer. I do not know how far this source of\nerror would be done away, if philosophers would _try_ to discover what\nquestion they were asking, before they set about to answer it; for the\nwork of analysis and distinction is often very difficult: we may often\nfail to make the necessary discovery, even though we make a definite\nattempt to do so. But I am inclined to think that in many cases a\nresolute attempt would be sufficient to ensure success; so that, if\nonly this attempt were made, many of the most glaring difficulties\nand disagreements in philosophy would disappear. At all events,\nphilosophers seem, in general, not to make the attempt; and, whether in\nconsequence of this omission or not, they are constantly endeavouring\nto prove that \u2018Yes\u2019 or \u2018No\u2019 will answer questions, to which _neither_\nanswer is correct, owing to the fact that what they have before their\nminds is not one question, but several, to some of which the true\nanswer is \u2018No,\u2019 to others \u2018Yes.\u2019\n\nI have tried in this book to distinguish clearly two kinds of question,\nwhich moral philosophers have always professed to answer, but which,\nas I have tried to shew, they have almost always confused both with one\nanother and with other questions. These two questions may be expressed,\nthe first in the form: What kind of things ought to exist for their\nown sakes? the second in the form: What kind of actions ought we to\nperform? I have tried to shew exactly what it is that we ask about a\nthing, when we ask whether it ought to exist for its own sake, is good\nin itself or has intrinsic value; and exactly what it is that we ask\nabout an action, when we ask whether we ought to do it, whether it is a\nright action or a duty.\n\nBut from a clear insight into the nature of these two questions, there\nappears to me to follow a second most important result: namely, what\nis the nature of the evidence, by which alone any ethical proposition\ncan be proved or disproved, confirmed or rendered doubtful. Once we\nrecognise the exact meaning of the two questions, I think it also\nbecomes plain exactly what kind of reasons are relevant as arguments\nfor or against any particular answer to them. It becomes plain that,\nfor answers to the _first_ question, no relevant evidence whatever\ncan be adduced: from no other truth, except themselves alone, can it\nbe inferred that they are either true or false. We can guard against\nerror only by taking care, that, when we try to answer a question of\nthis kind, we have before our minds that question only, and not some\nother or others; but that there is great danger of such errors of\nconfusion I have tried to shew, and also what are the chief precautions\nby the use of which we may guard against them. As for the _second_\nquestion, it becomes equally plain, that any answer to it _is_ capable\nof proof or disproof--that, indeed, so many different considerations\nare relevant to its truth or falsehood, as to make the attainment of\nprobability very difficult, and the attainment of certainty impossible.\nNevertheless the _kind_ of evidence, which is both necessary and alone\nrelevant to such proof and disproof, is capable of exact definition.\nSuch evidence must contain propositions of two kinds and of two kinds\nonly: it must consist, in the first place, of truths with regard to\nthe results of the action in question--of _causal_ truths--but it must\n_also_ contain ethical truths of our first or self-evident class. Many\ntruths of both kinds are necessary to the proof that any action ought\nto be done; and any other kind of evidence is wholly irrelevant. It\nfollows that, if any ethical philosopher offers for propositions of\nthe first kind any evidence whatever, or if, for propositions of the\nsecond kind, he either fails to adduce both causal and ethical truths,\nor adduces truths that are neither, his reasoning has not the least\ntendency to establish his conclusions. But not only are his conclusions\ntotally devoid of weight: we have, moreover, reason to suspect him\nof the error of confusion; since the offering of irrelevant evidence\ngenerally indicates that the philosopher who offers it has had before\nhis mind, not the question which he professes to answer, but some other\nentirely different one. Ethical discussion, hitherto, has perhaps\nconsisted chiefly in reasoning of this totally irrelevant kind.\n\nOne main object of this book may, then, be expressed by slightly\nchanging one of Kant\u2019s famous titles. I have endeavoured to write\n\u2018Prolegomena to any future Ethics that can possibly pretend to be\nscientific.\u2019 In other words, I have endeavoured to discover what are\nthe fundamental principles of ethical reasoning; and the establishment\nof these principles, rather than of any conclusions which may be\nattained by their use, may be regarded as my main object. I have,\nhowever, also attempted, in Chapter VI, to present some conclusions,\nwith regard to the proper answer of the question \u2018What is good in\nitself?\u2019 which are very different from any which have commonly been\nadvocated by philosophers. I have tried to define the classes within\nwhich all great goods and evils fall; and I have maintained that very\nmany different things are good and evil in themselves, and that\nneither class of things possesses any other property which is both\ncommon to all its members and peculiar to them.\n\nIn order to express the fact that ethical propositions of my _first_\nclass are incapable of proof or disproof, I have sometimes followed\nSidgwick\u2019s usage in calling them \u2018Intuitions.\u2019 But I beg it may be\nnoticed that I am not an \u2018Intuitionist,\u2019 in the ordinary sense of\nthe term. Sidgwick himself seems never to have been clearly aware\nof the immense importance of the difference which distinguishes his\nIntuitionism from the common doctrine, which has generally been called\nby that name. The Intuitionist proper is distinguished by maintaining\nthat propositions of my _second_ class--propositions which assert that\na certain action is _right_ or a _duty_--are incapable of proof or\ndisproof by any enquiry into the results of such actions. I, on the\ncontrary, am no less anxious to maintain that propositions of _this_\nkind are _not_ \u2018Intuitions,\u2019 than to maintain that propositions of my\n_first_ class _are_ Intuitions.\n\nAgain, I would wish it observed that, when I call such propositions\n\u2018Intuitions,\u2019 I mean _merely_ to assert that they are incapable of\nproof; I imply nothing whatever as to the manner or origin of our\ncognition of them. Still less do I imply (as most Intuitionists have\ndone) that any proposition whatever is true, _because_ we cognise it in\na particular way or by the exercise of any particular faculty: I hold,\non the contrary, that in every way in which it is possible to cognise a\ntrue proposition, it is also possible to cognise a false one.\n\nWhen this book had been already completed, I found, in Brentano\u2019s\n\u2018Origin of the Knowledge of Right and Wrong[1],\u2019 opinions far more\nclosely resembling my own, than those of any other ethical writer with\nwhom I am acquainted. Brentano appears to agree with me completely\n(1) in regarding all ethical propositions as defined by the fact that\nthey predicate a single unique objective concept; (2) in dividing such\npropositions sharply into the same two kinds; (3) in holding that the\nfirst kind are incapable of proof; and (4) with regard to the kind of\nevidence which is necessary and relevant to the proof of the second\nkind. But he regards the fundamental ethical concept as being, not\nthe simple one which I denote by \u2018good,\u2019 but the complex one which I\nhave taken to define \u2018beautiful\u2019; and he does not recognise, but even\ndenies by implication, the principle which I have called _the principle\nof organic unities_. In consequence of these two differences, his\nconclusions as to what things are good in themselves, also differ very\nmaterially from mine. He agrees, however, that there are many different\ngoods, and that the love of good and beautiful objects constitutes an\nimportant class among them.\n\n  [1] \u2018The Origin of the Knowledge of Right and Wrong.\u2019 By Franz\n  Brentano. English Translation by Cecil Hague. Constable, 1902.--I\n  have written a review of this book, which will, I hope, appear\n  in the _International Journal of Ethics_ for October, 1903. I\n  may refer to this review for a fuller account of my reasons for\n  disagreeing with Brentano.\n\nI wish to refer to one oversight, of which I became aware only when\nit was too late to correct it, and which may, I am afraid, cause\nunnecessary trouble to some readers. I have omitted to discuss directly\nthe mutual relations of the several different notions, which are\nall expressed by the word \u2018end.\u2019 The consequences of this omission\nmay perhaps be partially avoided by a reference to my article on\n\u2018Teleology\u2019 in Baldwin\u2019s _Dictionary of Philosophy and Psychology_.\n\nIf I were to rewrite my work now, I should make a very different, and\nI believe that I could make a much better book. But it may be doubted\nwhether, in attempting to satisfy myself, I might not merely render\nmore obscure the ideas which I am most anxious to convey, without a\ncorresponding gain in completeness and accuracy. However that may be,\nmy belief that to publish the book as it stands was probably the best\nthing I could do, does not prevent me from being painfully aware that\nit is full of defects.\n\nTRINITY COLLEGE, CAMBRIDGE.\n_August_, 1903.\n\n\n[This book is now reprinted without any alteration whatever, except\nthat a few misprints and grammatical mistakes have been corrected. It\nis reprinted, because I am still in agreement with its main tendency\nand conclusions; and it is reprinted without alteration, because I\nfound that, if I were to begin correcting what in it seemed to me to\nneed correction, I could not stop short of rewriting the whole book.\n\n  G. E. M.]\n\nCAMBRIDGE, 1922.\n\n\n\n\nTABLE OF CONTENTS.\n\n\n  CHAPTER I.\n\n  THE SUBJECT-MATTER OF ETHICS.\n\n  A.\n\n  SECTION.                                                          PAGE\n\n  =1.= In order to define Ethics, we must discover what is both\n  common and peculiar to all undoubted ethical judgments;              1\n\n  =2.= but this is not that they are concerned with human\n  conduct, but that they are concerned with a certain predicate\n  \u2018good,\u2019 and its converse \u2018bad,\u2019 which may be applied both to\n  conduct and to other things.                                         1\n\n  =3.= The subjects of the judgments of a scientific Ethics are\n  not, like those of some studies, \u2018particular things\u2019;                3\n\n  =4.= but it includes all _universal_ judgments which assert\n  the relation of \u2018goodness\u2019 to any subject, and hence includes\n  Casuistry.                                                           3\n\n  B.\n\n  =5.= It must, however, enquire not only what things are\n  universally related to goodness, but also, what this predicate,\n  to which they are related, is:                                       5\n\n  =6.= and the answer to this question is that it is indefinable       6\n\n  =7.= or simple: for if by definition be meant the analysis of\n  an object of thought, only complex objects can be defined;           7\n\n  =8.= and of the three senses in which \u2018definition\u2019 can be used,\n  this is the most important.                                          8\n\n  =9.= What is thus indefinable is not \u2018the good,\u2019 or the whole\n  of that which always possesses the predicate \u2018good,\u2019 but this\n  predicate itself.                                                    8\n\n  =10.= \u2018Good,\u2019 then, denotes one unique simple object of thought\n  among innumerable others; but this object has very commonly\n  been identified with some other--a fallacy which may be called\n  \u2018the naturalistic fallacy\u2019                                           9\n\n  =11.= and which reduces what is used as a fundamental principle\n  of Ethics either to a tautology or to a statement about the\n  meaning of a word.                                                  10\n\n  =12.= The nature of this fallacy is easily recognised;              12\n\n  =13.= and if it were avoided, it would be plain that the only\n  alternatives to the admission that \u2018good\u2019 is indefinable, are\n  either that it is complex or that there is no notion at all\n  peculiar to Ethics--alternatives which can only be refuted by\n  an appeal to inspection, but which can be so refuted.               15\n\n  =14.= The \u2018naturalistic fallacy\u2019 illustrated by Bentham; and\n  the importance of avoiding it pointed out.                          17\n\n  C.\n\n  =15.= The relations which ethical judgments assert to hold\n  universally between \u2018goodness\u2019 and other things are of two\n  kinds: a thing may be asserted either to _be_ good itself or to\n  be causally related to something else which is itself good--to\n  be \u2018good as a means.\u2019                                               21\n\n  =16.= Our investigations of the latter kind of relation cannot\n  hope to establish more than that a certain kind of action will\n  _generally_ be followed by the best possible results;               22\n\n  =17.= but a relation of the former kind, if true at all,\n  will be true of all cases. All ordinary ethical judgments\n  assert _causal_ relations, but they are commonly treated as\n  if they did not, because the two kinds of relation are not\n  distinguished.                                                      23\n\n  D.\n\n  =18.= The investigation of intrinsic values is complicated by\n  the fact that the value of a whole may be different from the\n  sum of the values of its parts,                                     27\n\n  =19.= in which case the part has to the whole a relation, which\n  exhibits an equally important difference from and resemblance\n  to that of means to end.                                            29\n\n  =20.= The term \u2018organic whole\u2019 might well be used to denote\n  that a whole has this property, since, of the two other\n  properties which it is commonly used to imply,                      30\n\n  =21.= one that of reciprocal causal dependence between parts,\n  has no necessary relation to this one,                              31\n\n  =22.= and the other, upon which most stress has been laid, can\n  be true of no whole whatsoever, being a self-contradictory\n  conception due to confusion.                                        33\n\n  =23.= Summary of chapter.                                           36\n\n\n  CHAPTER II.\n\n  NATURALISTIC ETHICS.\n\n  =24.= This and the two following chapters will consider certain\n  proposed answers to the second of ethical questions: What is\n  _good in itself_? These proposed answers are characterised by\n  the facts (1) that they declare some _one_ kind of thing to be\n  alone good in itself; and (2) that they do so, because they\n  suppose this _one_ thing to define the meaning of \u2018good.\u2019           37\n\n  =25.= Such theories may be divided into two groups (1)\n  Metaphysical, (2) Naturalistic: and the second group may be\n  subdivided into two others, (_a_) theories which declare some\n  natural object, other than pleasure, to be sole good, (_b_)\n  Hedonism. The present chapter will deal with (_a_).                 38\n\n  =26.= Definition of what is meant by \u2018Naturalism.\u2019                  39\n\n  =27.= The common argument that things are good, because they\n  are \u2018natural,\u2019 may involve either (1) the false proposition\n  that the \u2018normal,\u2019 as such, is good;                                41\n\n  =28.= or (2) the false proposition that the \u2018necessary,\u2019 as\n  such, is good.                                                      44\n\n  =29.= But a _systematised_ appeal to Nature is now most\n  prevalent in connection with the term \u2018Evolution.\u2019 An\n  examination of Mr Herbert Spencer\u2019s Ethics will illustrate this\n  form of Naturalism.                                                 45\n\n  =30.= Darwin\u2019s scientific theory of \u2018natural selection,\u2019 which\n  has mainly caused the modern vogue of the term \u2018Evolution,\u2019\n  must be carefully distinguished from certain ideas which are\n  commonly associated with the latter term.                           47\n\n  =31.= Mr Spencer\u2019s connection of Evolution with Ethics seems to\n  shew the influence of the naturalistic fallacy;                     48\n\n  =32.= but Mr Spencer is vague as to the ethical relations of\n  \u2018pleasure\u2019 and \u2018evolution,\u2019 and his Naturalism may be mainly\n  Naturalistic Hedonism.                                              49\n\n  =33.= A discussion of the third chapter of the _Data of Ethics_\n  serves to illustrate these two points and to shew that Mr\n  Spencer is in utter confusion with regard to the fundamental\n  principles of Ethics.                                               51\n\n  =34.= Three possible views as to the relation of Evolution to\n  Ethics are distinguished from the naturalistic view to which\n  it is proposed to confine the name \u2018Evolutionistic Ethics.\u2019 On\n  any of these three views the relation would be unimportant, and\n  the \u2018Evolutionistic\u2019 view, which makes it important, involves a\n  double fallacy.                                                     54\n\n  =35.= Summary of chapter.                                           58\n\n\n  CHAPTER III.\n\n  HEDONISM.\n\n  =36.= The prevalence of Hedonism is mainly due to the\n  naturalistic fallacy.                                               59\n\n  =37.= Hedonism may be defined as the doctrine that \u2018Pleasure is\n  the sole good\u2019: this doctrine has always been held by Hedonists\n  and used by them as a fundamental ethical principle, although\n  it has commonly been confused with others.                          61\n\n  =38.= The method pursued in this chapter will consist in\n  exposing the reasons commonly offered for the truth of Hedonism\n  and in bringing out the reasons, which suffice to shew it\n  untrue, by a criticism of J. S. Mill & H. Sidgwick.                 63\n\n  A.\n\n  =39.= Mill declares that \u2018Happiness is the only thing desirable\n  as an end,\u2019 and insists that \u2018Questions of ultimate ends are\n  not amenable to direct proof\u2019;                                      64\n\n  =40.= yet he gives a proof of the first proposition, which\n  consists in (1) the fallacious confusion of \u2018desirable\u2019 with\n  \u2018desired,\u2019                                                          66\n\n  =41.= (2) an attempt to shew that nothing but pleasure is\n  desired.                                                            67\n\n  =42.= The theory that nothing but pleasure is desired seems\n  largely due to a confusion between the _cause_ and the _object_\n  of desire: pleasure is certainly not the sole _object_ of\n  desire, and, even if it is always among the _causes_ of desire,\n  that fact would not tempt anyone to think it a good.                68\n\n  =43.= Mill attempts to reconcile his doctrine that pleasure is\n  the sole object of desire with his admission that other things\n  are desired, by the absurd declaration that what is a means to\n  happiness is \u2018part\u2019 of happiness.                                   71\n\n  =44.= Summary of Mill\u2019s argument and of my criticism.               72\n\n  B.\n\n  =45.= We must now proceed to consider the principle of Hedonism\n  as an \u2018Intuition,\u2019 as which it has been clearly recognised\n  by Prof. Sidgwick alone. That it should be thus incapable of\n  _proof_ is not, in itself, any reason for dissatisfaction.          74\n\n  =46.= In thus beginning to consider what things are good in\n  themselves, we leave the refutation of Naturalism behind, and\n  enter on the second division of ethical questions.                  76\n\n  =47.= Mill\u2019s doctrine that some pleasures are superior \u2018in\n  quality\u2019 to others implies both (1) that judgments of ends must\n  be \u2018intuitions\u2019;                                                    77\n\n  =48.= and (2) that pleasure is _not_ the sole good.                 79\n\n  =49.= Prof. Sidgwick has avoided these confusions made by\n  Mill: in considering his arguments we shall, therefore, merely\n  consider the question \u2018Is pleasure the sole good?\u2019                  81\n\n  =50.= Prof. Sidgwick first tries to shew that nothing outside\n  of Human Existence can be good. Reasons are given for doubting\n  this.                                                               81\n\n  =51.= He then goes on to the far more important proposition\n  that no part of Human Existence, except pleasure, is desirable.     85\n\n  =52.= But _pleasure_ must be distinguished from _consciousness\n  of pleasure_, and (1) it is plain that, when so distinguished,\n  _pleasure_ is not the sole good;                                    87\n\n  =53.= and (2) it may be made equally plain that _consciousness\n  of pleasure_ is not the sole good, if we are equally careful to\n  distinguish it from its usual accompaniments.                       90\n\n  =54.= Of Prof. Sidgwick\u2019s two arguments for the contrary view,\n  the second is equally compatible with the supposition that\n  pleasure is a mere _criterion_ of what is _right_;                  91\n\n  =55.= and in his first, the appeal to reflective intuition,\n  he fails to put the question clearly (1) in that he does not\n  recognise the principle of _organic unities_;                       92\n\n  =56.= and (2) in that he fails to emphasize that the agreement,\n  which he has tried to shew, between hedonistic judgments and\n  those of Common Sense, only holds of _judgments of means_:\n  hedonistic judgments of _ends_ are flagrantly paradoxical.          94\n\n  =57.= I conclude, then, that a reflective intuition, if proper\n  precautions are taken, will agree with Common Sense that it is\n  absurd to regard mere consciousness of pleasure as the sole\n  good.                                                               95\n\n  C.\n\n  =58.= It remains to consider Egoism and Utilitarianism. It\n  is important to distinguish the former, as the doctrine that\n  \u2018my own pleasure is sole good,\u2019 from the doctrine, opposed to\n  Altruism, that to pursue my own pleasure exclusively is right\n  _as a means_.                                                       96\n\n  =59.= Egoism proper is utterly untenable, being\n  self-contradictory: it fails to perceive that when I declare\n  a thing to be my own good, I must be declaring it to be _good\n  absolutely_ or else not good at all.                                97\n\n  =60.= This confusion is further brought out by an examination\n  of Prof. Sidgwick\u2019s contrary view;                                  99\n\n  =61.= and it is shewn that, in consequence of this confusion,\n  his representation of \u2018the relation of Rational Egoism to\n  Rational Benevolence\u2019 as \u2018the profoundest problem of Ethics,\u2019\n  and his view that a certain hypothesis is required to \u2018make\n  Ethics rational,\u2019 are grossly erroneous.                           102\n\n  =62.= The same confusion is involved in the attempt to infer\n  Utilitarianism from Psychological Hedonism, as commonly held,\n  _e.g._ by Mill.                                                    104\n\n  =63.= Egoism proper seems also to owe its plausibility to its\n  confusion with Egoism, as a doctrine of means.                     105\n\n  =64.= Certain ambiguities in the conception of Utilitarianism\n  are noticed; and it is pointed out (1) that, as a doctrine\n  of the end to be pursued, it is finally refuted by the\n  refutation of Hedonism, and (2) that, while the arguments most\n  commonly urged in its favour could, at most, only shew it to\n  offer a correct _criterion_ of right action, they are quite\n  insufficient even for this purpose.                                105\n\n  =65.= Summary of chapter.                                          108\n\n\n  CHAPTER IV.\n\n  METAPHYSICAL ETHICS.\n\n  A.\n\n  =66.= The term \u2018metaphysical\u2019 is defined as having reference\n  primarily to any object of knowledge which is not a part of\n  Nature--does not exist in time, as an object of perception; but\n  since metaphysicians, not content with pointing out the truth\n  about such entities, have always supposed that what does not\n  exist in Nature, must, at least, _exist_, the term also has\n  reference to a supposed \u2018supersensible reality\u2019:                   110\n\n  =67.= and by \u2018metaphysical Ethics\u2019 I mean those systems which\n  maintain or imply that the answer to the question \u2018What is\n  good?\u2019 _logically depends_ upon the answer to the question\n  \u2018What is the nature of supersensible reality?.\u2019 All such\n  systems obviously involve the same fallacy--the \u2018naturalistic\n  fallacy\u2019--by the use of which Naturalism was also defined.         113\n\n  =68.= Metaphysics, as dealing with a \u2018supersensible reality,\u2019\n  may have a bearing upon _practical_ Ethics (1) if its\n  supersensible reality is conceived as something future, which\n  our actions can affect; and (2) since it will prove that\n  _every_ proposition of practical Ethics is false, if it can\n  shew that an eternal reality is either the only real thing\n  or the only good thing. Most metaphysical writers, believing\n  in a reality of the latter kind, do thus imply the complete\n  falsehood of every practical proposition, although they fail to\n  see that their Metaphysics thus contradicts their Ethics.          115\n\n  B.\n\n  =69.= But the theory, by which I have defined Metaphysical\n  Ethics, is _not_ that Metaphysics has a logical bearing upon\n  the question involved in _practical_ Ethics \u2018What effects will\n  my action produce?,\u2019 but that it has such a bearing upon the\n  fundamental ethical question \u2018What is good in itself?.\u2019 This\n  theory has been refuted by the proof, in Chap. I, that the\n  naturalistic fallacy is a fallacy: it only remains to discuss\n  certain confusions which seem to have lent it plausibility.        118\n\n  =70.= One such source of confusion seems to lie in the failure\n  to distinguish between the proposition \u2018This is good,\u2019 when\n  it means \u2018This _existing_ thing is good,\u2019 and the same\n  proposition, when it means \u2018The existence of this _kind_ of\n  thing would be good\u2019;                                              118\n\n  =71.= and another seems to lie in the failure to distinguish\n  between that which _suggests_ a truth, or is a _cause_ of\n  our knowing it, and that upon which it _logically_ depends,\n  or which is a _reason_ for believing it: in the former\n  sense fiction has a more important bearing upon Ethics than\n  Metaphysics can have.                                              121\n\n  C.\n\n  =72.= But a more important source of confusion seems to lie\n  in the supposition that \u2018to be good\u2019 is _identical_ with the\n  possession of some supersensible property, which is also\n  involved in the definition of \u2018reality.\u2019                           122\n\n  =73.= One cause of this supposition seems to be the logical\n  prejudice that all propositions are of the most familiar\n  type--that in which subject and predicate are both existents.      123\n\n  =74.= But ethical propositions cannot be reduced to this type:\n  in particular, they are obviously to be distinguished              125\n\n  =75.= (1) from Natural Laws; with which one of Kant\u2019s most\n  famous doctrines confuses them,                                    126\n\n  =76.= and (2) from Commands; with which they are confused both\n  by Kant and by others.                                             127\n\n  D.\n\n  =77.= This latter confusion is one of the sources of the\n  prevalent modern doctrine that \u2018being good\u2019 is _identical_ with\n  \u2018being willed\u2019; but the prevalence of this doctrine seems to be\n  chiefly due to other causes. I shall try to shew with regard\n  to it (1) what are the chief errors which seem to have led to\n  its adoption; and (2) that, apart from it, the Metaphysics of\n  Volition can hardly have the smallest logical bearing upon\n  Ethics.                                                            128\n\n  =78.= (1) It has been commonly held, since Kant, that\n  \u2018goodness\u2019 has the same relation to Will or Feeling, which\n  \u2018truth\u2019 or \u2018reality\u2019 has to Cognition: that the proper method\n  for Ethics is to discover what is _implied_ in Will or Feeling,\n  just as, according to Kant, the proper method for Metaphysics\n  was to discover what is _implied_ in Cognition.                    129\n\n  =79.= The actual relations between \u2018goodness\u2019 and Will or\n  Feeling, from which this false doctrine is inferred, seem to\n  be mainly (_a_) the _causal_ relation consisting in the fact\n  that it is only by reflection upon the experiences of Will and\n  Feeling that we become aware of ethical distinctions; (_b_) the\n  facts that a cognition of goodness is perhaps _always_ included\n  in certain kinds of Willing and Feeling, and is _generally_\n  accompanied by them:                                               130\n\n  =80.= but from neither of these _psychological_ facts does it\n  follow that \u2018to be good\u2019 is identical with being willed or\n  felt in a certain way: the supposition that it does follow\n  is an instance of the fundamental contradiction of modern\n  Epistemology--the contradiction involved in both distinguishing\n  and identifying the _object_ and the _act_ of Thought, \u2018truth\u2019\n  itself and its supposed _criterion_:                               131\n\n  =81.= and, once this analogy between Volition and Cognition is\n  accepted, the view that ethical propositions have an essential\n  reference to Will or Feeling, is strengthened by another error\n  with regard to the nature of Cognition--the error of supposing\n  that \u2018perception\u2019 denotes _merely_ a certain way of cognising\n  an object, whereas it actually includes the assertion that the\n  object is also _true_.                                             133\n\n  =82.= The argument of the last three \u00a7\u00a7 is recapitulated; and\n  it is pointed out (1) that Volition and Feeling are _not_\n  analogous to Cognition, (2) that, even if they were, still \u2018to\n  be good\u2019 could not _mean_ \u2018to be willed or felt in a certain\n  way.\u2019                                                              135\n\n  =83.= (2) If \u2018being good\u2019 and \u2018being willed\u2019 are not\n  _identical_, then the latter could only be a _criterion_ of the\n  former; and, in order to shew that it was so, we should have\n  to establish _independently_ that many things were good--that\n  is to say, we should have to establish most of our ethical\n  conclusions, before the Metaphysics of Volition could possibly\n  give us the smallest assistance.                                   137\n\n  =84.= The fact that the metaphysical writers who, like Green,\n  attempt to _base_ Ethics on Volition, do not even attempt this\n  independent investigation, shews that they start from the false\n  assumption that goodness is _identical_ with being willed, and\n  hence that their ethical reasonings have no value whatsoever.      138\n\n  =85.= Summary of chapter.                                          139\n\n\n  CHAPTER V.\n\n  ETHICS IN RELATION TO CONDUCT.\n\n  =86.= The question to be discussed in this chapter must\n  be clearly distinguished from the two questions hitherto\n  discussed, namely (1) What is the nature of the proposition:\n  \u2018This is good in itself\u2019?                                          142\n\n  =87.= and (2) What things are good in themselves? to which we\n  gave one answer in deciding that pleasure was not the only\n  thing good in itself.                                              144\n\n  =88.= In this chapter we shall deal with the _third_ object of\n  ethical enquiry: namely answers to the question \u2018What conduct\n  is a _means_ to good results?\u2019 or \u2018What ought we to do?\u2019 This\n  is the question of _Practical_ Ethics, and its answer involves\n  an assertion of _causal_ connection.                               146\n\n  =89.= It is shewn that the assertions \u2018This action is right\u2019\n  or \u2018is my duty\u2019 are equivalent to the assertion that the total\n  results of the action in question will be the best possible;       146\n\n  =90.= and the rest of the chapter will deal with certain\n  conclusions, upon which light is thrown by this fact. Of which\n  the first is (1) that Intuitionism is mistaken; since no\n  proposition with regard to duty can be self-evident.               148\n\n  =91.= (2) It is plain that we cannot hope to prove which\n  among all the actions, which it is possible for us to perform\n  on every occasion, will produce the best total results:\n  to discover what is our \u2018duty,\u2019 in this strict sense, is\n  impossible. It may, however, be possible to shew which among\n  the actions, which we are _likely_ to perform, will produce\n  the best results.                                                  149\n\n  =92.= The distinction made in the last \u00a7 is further explained;\n  and it is insisted that all that Ethics has done or can do, is,\n  _not_ to determine absolute duties, but to point out which,\n  among a _few_ of the alternatives, possible under _certain_\n  circumstances, will have the better results.                       150\n\n  =93.= (3) Even this latter task is immensely difficult, and\n  no adequate proof that the total results of one action are\n  superior to those of another, has ever been given. For (_a_)\n  we can only calculate actual results within a comparatively\n  near future: we must, therefore, assume that no results of the\n  same action in the infinite future beyond, will reverse the\n  balance--an assumption which perhaps _can_ be, but certainly\n  has not been, justified;                                           152\n\n  =94.= and (_b_) even to decide that, of any two actions, one\n  has a better total result than the other _in the immediate\n  future_, is very difficult; and it is very improbable, and\n  quite impossible to prove, that any single action is _in all\n  cases_ better as means than its probable alternative. Rules\n  of duty, even in this restricted sense, can only, at most, be\n  _general_ truths.                                                  154\n\n  =95.= But (_c_) most of the actions, most universally approved\n  by Common Sense, may perhaps be shewn to be _generally_ better\n  as means than any probable alternative, on the following\n  principles. (1) With regard to some rules it may be shewn\n  that their general observance would be useful in any state of\n  society, where the instincts to preserve and propagate life and\n  to possess property were as strong as they seem always to be;\n  and this utility may be shewn, independently of a right view as\n  to what is good in itself, since the observance is a means to\n  things which are a necessary condition for the attainment of\n  _any_ great goods in considerable quantities.                      155\n\n  =96.= (2) Other rules are such that their general observance\n  can only be shewn to be useful, as means to the preservation\n  of society, under more or less temporary conditions: if any of\n  these are to be proved useful in _all_ societies, this can only\n  be done by shewing their causal relation to things good or evil\n  in themselves, which are not generally recognised to be such.      158\n\n  =97.= It is plain that rules of class (1) may _also_ be\n  justified by the existence of such temporary conditions as\n  justify those of class (2); and among such temporary conditions\n  must be reckoned the so-called _sanctions_.                        159\n\n  =98.= In this way, then, it may be possible to prove the\n  _general_ utility, for the present, of those actions, which in\n  our society are _both_ generally recognised as duties _and_\n  generally practised; but it seems very doubtful whether a\n  conclusive case can be established for any proposed change in\n  social custom, without an independent investigation of what\n  things are good or bad in themselves.                              159\n\n  =99.= And (_d_) if we consider the distinct question of how a\n  single individual should decide to act (\u03b1) in cases where the\n  _general_ utility of the action in question is certain, (\u03b2) in\n  other cases: there seems reason for thinking that, with regard\n  to (\u03b1), where the generally useful rule is also generally\n  observed, he should _always_ conform to it; but these reasons\n  are not conclusive, if _either_ the general observance _or_ the\n  general utility is wanting:                                        162\n\n  =100.= and that (\u03b2) in all other cases, _rules of action_\n  should not be followed at all, but the individual should\n  consider what positive goods, _he_, in his particular\n  circumstances, seems likely to be able to effect, and what\n  evils to avoid.                                                    164\n\n  =101.= (4) It follows further that the distinction denoted by\n  the terms \u2018duty\u2019 and \u2018expediency\u2019 is not primarily ethical:\n  when we ask \u2018Is this really expedient?\u2019 we are asking precisely\n  the same question as when we ask \u2018Is this my duty?,\u2019 viz.\n  \u2018Is this a means to the best possible?.\u2019 \u2018Duties\u2019 are mainly\n  distinguished by the non-ethical marks (1) that many people\n  are often tempted to avoid them, (2) that their most prominent\n  effects are on others than the agent, (3) that they excite\n  the moral sentiments: so far as they _are_ distinguished by\n  an ethical peculiarity, this is not that they are peculiarly\n  useful to perform, but that they are peculiarly useful to\n  sanction.                                                          167\n\n  =102.= The distinction between \u2018duty\u2019 and \u2018interest\u2019 is\n  also, in the main, the same non-ethical distinction: but\n  the term \u2018interested\u2019 does also refer to a distinct ethical\n  predicate--that an action is to \u2018my interest\u2019 asserts only that\n  it will have the best possible effects of one particular kind,\n  not that its total effects will be the best possible.              170\n\n  =103.= (5) We may further see that \u2018virtues\u2019 are not to be\n  defined as dispositions that are good in themselves: they are\n  not necessarily more than dispositions to perform actions\n  generally good as means, and of these, for the most part,\n  only those classed as \u2018duties\u2019 in accordance with section\n  (4). It follows that to decide whether a disposition is or is\n  not \u2018virtuous\u2019 involves the difficult causal investigation\n  discussed in section (3); and that what is a virtue in one\n  state of society may not be so in another.                         171\n\n  =104.= It follows also that we have no reason to presume, as\n  has commonly been done, that the exercise of virtue in the\n  performance of \u2018duties\u2019 is ever good in itself--far less, that\n  it is the sole good:                                               173\n\n  =105.= and, if we consider the intrinsic value of such\n  exercise, it will appear (1) that, in most cases, it has no\n  value, and (2) that even the cases, where it has some value,\n  are far from constituting the sole good. The truth of the\n  latter proposition is generally inconsistently implied, even by\n  those who deny it;                                                 174\n\n  =106.= but in order fairly to decide upon the intrinsic value\n  of virtue, we must distinguish three different kinds of\n  disposition, each of which is commonly so called and has been\n  maintained to be the only kind deserving the name. Thus (_a_)\n  the mere unconscious \u2018habit\u2019 of performing duties, which is the\n  commonest type, has no intrinsic value whatsoever; Christian\n  moralists are right in implying that mere \u2018external rightness\u2019\n  has no intrinsic value, though they are wrong in saying that it\n  is therefore not \u2018virtuous,\u2019 since this implies that it has no\n  value even as a means:                                             175\n\n  =107.= (_b_) where virtue consists in a disposition to have,\n  and be moved by, a sentiment of love towards really good\n  consequences of an action and of hatred towards really evil\n  ones, it has some intrinsic value, but its value may vary\n  greatly in degree:                                                 177\n\n  =108.= finally (_c_) where virtue consists in\n  \u2018conscientiousness,\u2019 _i.e._ the disposition not to act, in\n  certain cases, until we believe and feel that our action is\n  right, it seems to have some intrinsic value: the value of this\n  feeling has been peculiarly emphasized by Christian Ethics, but\n  it certainly is not, as Kant would lead us to think, either the\n  sole thing of value, or always good even as a means.               178\n\n  =109.= Summary of chapter.                                         180\n\n\n  CHAPTER VI.\n\n  THE IDEAL.\n\n  =110.= By an \u2018ideal\u2019 state of things may be meant either (1)\n  the Summum Bonum or absolutely best, or (2) the best which the\n  laws of nature allow to exist in this world, or (3) anything\n  greatly good in itself: this chapter will be principally\n  occupied with what is ideal in sense (3)--with answering the\n  fundamental question of Ethics;                                    183\n\n  =111.= but a correct answer to this question is an essential\n  step towards a correct view as to what is \u2018ideal\u2019 in senses (1)\n  and (2).                                                           184\n\n  =112.= In order to obtain a correct answer to the question\n  \u2018What is good in itself?\u2019 we must consider what value things\n  would have if they existed absolutely by themselves;               187\n\n  =113.= and, if we use this method, it is obvious that personal\n  affection and aesthetic enjoyments include by far the greatest\n  goods with which we are acquainted.                                188\n\n  =114.= If we begin by considering I. _Aesthetic Enjoyments_, it\n  is plain (1) that there is always _essential_ to these some one\n  of a great variety of different emotions, though these emotions\n  may have little value _by themselves_:                             189\n\n  =115.= and (2) that a cognition of really beautiful qualities\n  is equally essential, and has equally little value by itself.      190\n\n  =116.= But (3) granted that the appropriate combination of\n  these two elements is always a considerable good and may be\n  a very great one, we may ask whether, where there is _added_\n  to this _a true belief in the existence of the object of the\n  cognition_, the whole thus formed is not much more valuable\n  still.                                                             192\n\n  =117.= I think that this question should be answered in the\n  affirmative; but in order to ensure that this judgment is\n  correct, we must carefully distinguish it                          194\n\n  =118.= from the two judgments (_a_) that knowledge is valuable\n  _as a means_, (_b_) that, where the object of the cognition\n  is itself a good thing, its existence, of course, adds to the\n  value of the whole state of things:                                195\n\n  =119.= if, however, we attempt to avoid being biassed by these\n  two facts, it still seems that mere true belief may be a\n  condition essential to great value.                                197\n\n  =120.= We thus get a _third_ essential constituent of many\n  great goods; and in this way we are able to justify (1) the\n  attribution of value to _knowledge_, over and above its value\n  as a means, and (2) the intrinsic superiority of the proper\n  appreciation of a _real_ object over the appreciation of an\n  equally valuable object of mere imagination: emotions directed\n  towards real objects may thus, even if the object be inferior,\n  claim equality with the highest imaginative pleasures.             198\n\n  =121.= Finally (4) with regard to the _objects_ of the\n  cognition which is essential to these good wholes, it is the\n  business of Aesthetics to analyse their nature: it need only be\n  here remarked (1) that, by calling them \u2018beautiful,\u2019 we mean\n  that they have this relation to a _good_ whole; and (2) that\n  they are, for the most part, themselves complex wholes, such\n  that the admiring contemplation of the whole greatly exceeds\n  in value the sum of the values of the admiring contemplation of\n  the parts.                                                         200\n\n  =122.= With regard to II. _Personal Affection_, the object is\n  here not merely beautiful but also good in itself; it appears,\n  however, that the appreciation of what is thus good in itself,\n  viz. the mental qualities of a person, is certainly, by itself,\n  not so great a good as the whole formed by the combination\n  with it of an appreciation of corporeal beauty; it is doubtful\n  whether it is even so great a good as the mere appreciation of\n  corporeal beauty; but it is certain that the combination of\n  both is a far greater good than either singly.                     203\n\n  =123.= It follows from what has been said that we have every\n  reason to suppose that a cognition of _material qualities_, and\n  even their existence, is an essential constituent of the Ideal\n  or Summum Bonum: there is only a bare possibility that they are\n  not included in it.                                                205\n\n  =124.= It remains to consider _positive evils_ and _mixed\n  goods_. I. _Evils_ may be divided into three classes, namely       207\n\n  =125.= (1) evils which consist in the love, or admiration, or\n  enjoyment of what is evil or ugly                                  208\n\n  =126.= (2) evils which consist in the hatred or contempt of\n  what is good or beautiful                                          211\n\n  =127.= and (3) the consciousness of intense pain: this appears\n  to be the only thing, either greatly good or greatly evil,\n  which does not involve _both_ a cognition _and_ an emotion\n  directed towards its object; and hence it is not analogous to\n  pleasure in respect of its intrinsic value, while it also seems\n  not to add to the vileness of a whole, _as a whole_, in which\n  it is combined with another bad thing, whereas pleasure does\n  add to the goodness of a whole, in which it is combined with\n  another good thing;                                                212\n\n  =128.= but pleasure and pain are completely analogous in this,\n  that pleasure by no means always increases, and pain by no\n  means always decreases, the total value of a whole in which it\n  is included: the converse is often true.                           213\n\n  =129.= In order to consider II. _Mixed Goods_, we must first\n  distinguish between (1) the value of a whole _as a whole_,\n  and (2) its value _on the whole_ or total value: (1) = the\n  difference between (2) and the sum of the values of the parts.\n  In view of this distinction, it then appears:                      214\n\n  =130.= (1) That the mere combination of two or more evils is\n  never positively good _on the whole_, although it may certainly\n  have great intrinsic value _as a whole_;                           216\n\n  =131.= but (2) That a whole which includes a cognition of\n  something evil or ugly may yet be a great positive good _on the\n  whole_: most virtues, which have any intrinsic value whatever,\n  seem to be of this kind, _e.g._ (_a_) courage and compassion,\n  and (_b_) moral goodness; all these are instances of the hatred\n  or contempt of what is evil or ugly;                               216\n\n  =132.= but there seems no reason to think that, where the evil\n  object _exists_, the total state of things is ever positively\n  good _on the whole_, although the existence of the evil may add\n  to its value _as a whole_.                                         219\n\n  =133.= Hence (1) no actually existing evil is necessary to the\n  Ideal, (2) the contemplation of imaginary evils is necessary to\n  it, and (3) where evils already exist, the existence of mixed\n  virtues has a value independent both of its consequences and of\n  the value which it has in common with the proper appreciation\n  of imaginary evils.                                                220\n\n  =134.= Concluding remarks.                                         222\n\n  =135.= Summary of chapter.                                         224\n\n\n\n\nCHAPTER I.\n\nTHE SUBJECT-MATTER OF ETHICS.\n\n\n=1.= It is very easy to point out some among our every-day judgments,\nwith the truth of which Ethics is undoubtedly concerned. Whenever\nwe say, \u2018So and so is a good man,\u2019 or \u2018That fellow is a villain\u2019;\nwhenever we ask, \u2018What ought I to do?\u2019 or \u2018Is it wrong for me to do\nlike this?\u2019; whenever we hazard such remarks as \u2018Temperance is a virtue\nand drunkenness a vice\u2019--it is undoubtedly the business of Ethics\nto discuss such questions and such statements; to argue what is the\ntrue answer when we ask what it is right to do, and to give reasons\nfor thinking that our statements about the character of persons or\nthe morality of actions are true or false. In the vast majority of\ncases, where we make statements involving any of the terms \u2018virtue,\u2019\n\u2018vice,\u2019 \u2018duty,\u2019 \u2018right,\u2019 \u2018ought,\u2019 \u2018good,\u2019 \u2018bad,\u2019 we are making\nethical judgments; and if we wish to discuss their truth, we shall be\ndiscussing a point of Ethics.\n\nSo much as this is not disputed; but it falls very far short of\ndefining the province of Ethics. That province may indeed be defined as\nthe whole truth about that which is at the same time common to all such\njudgments and peculiar to them. But we have still to ask the question:\nWhat is it that is thus common and peculiar? And this is a question to\nwhich very different answers have been given by ethical philosophers\nof acknowledged reputation, and none of them, perhaps, completely\nsatisfactory.\n\n\n=2.= If we take such examples as those given above, we shall not be\nfar wrong in saying that they are all of them concerned with the\nquestion of \u2018conduct\u2019--with the question, what, in the conduct of us,\nhuman beings, is good, and what is bad, what is right, and what is\nwrong. For when we say that a man is good, we commonly mean that he\nacts rightly; when we say that drunkenness is a vice, we commonly mean\nthat to get drunk is a wrong or wicked action. And this discussion of\nhuman conduct is, in fact, that with which the name \u2018Ethics\u2019 is most\nintimately associated. It is so associated by derivation; and conduct\nis undoubtedly by far the commonest and most generally interesting\nobject of ethical judgments.\n\nAccordingly, we find that many ethical philosophers are disposed to\naccept as an adequate definition of \u2018Ethics\u2019 the statement that it\ndeals with the question what is good or bad in human conduct. They hold\nthat its enquiries are properly confined to \u2018conduct\u2019 or to \u2018practice\u2019;\nthey hold that the name \u2018practical philosophy\u2019 covers all the matter\nwith which it has to do. Now, without discussing the proper meaning\nof the word (for verbal questions are properly left to the writers of\ndictionaries and other persons interested in literature; philosophy,\nas we shall see, has no concern with them), I may say that I intend to\nuse \u2018Ethics\u2019 to cover more than this--a usage, for which there is, I\nthink, quite sufficient authority. I am using it to cover an enquiry\nfor which, at all events, there is no other word: the general enquiry\ninto what is good.\n\nEthics is undoubtedly concerned with the question what good conduct\nis; but, being concerned with this, it obviously does not start at the\nbeginning, unless it is prepared to tell us what is good as well as\nwhat is conduct. For \u2018good conduct\u2019 is a complex notion: all conduct is\nnot good; for some is certainly bad and some may be indifferent. And on\nthe other hand, other things, beside conduct, may be good; and if they\nare so, then, \u2018good\u2019 denotes some property, that is common to them and\nconduct; and if we examine good conduct alone of all good things, then\nwe shall be in danger of mistaking for this property, some property\nwhich is not shared by those other things: and thus we shall have made\na mistake about Ethics even in this limited sense; for we shall not\nknow what good conduct really is. This is a mistake which many writers\nhave actually made, from limiting their enquiry to conduct. And hence\nI shall try to avoid it by considering first what is good in general;\nhoping, that if we can arrive at any certainty about this, it will be\nmuch easier to settle the question of good conduct: for we all know\npretty well what \u2018conduct\u2019 is. This, then, is our first question: What\nis good? and What is bad? and to the discussion of this question (of\nthese questions) I give the name of Ethics, since that science must, at\nall events, include it.\n\n\n=3.= But this is a question which may have many meanings. If, for\nexample, each of us were to say \u2018I am doing good now\u2019 or \u2018I had a\ngood dinner yesterday,\u2019 these statements would each of them be some\nsort of answer to our question, although perhaps a false one. So, too,\nwhen A asks B what school he ought to send his son to, B\u2019s answer\nwill certainly be an ethical judgment. And similarly all distribution\nof praise or blame to any personage or thing that has existed, now\nexists, or will exist, does give some answer to the question \u2018What is\ngood?\u2019 In all such cases some particular thing is judged to be good or\nbad: the question \u2018What?\u2019 is answered by \u2018This.\u2019 But this is not the\nsense in which a scientific Ethics asks the question. Not one, of all\nthe many million answers of this kind, which must be true, can form a\npart of an ethical system; although that science must contain reasons\nand principles sufficient for deciding on the truth of all of them.\nThere are far too many persons, things and events in the world, past,\npresent, or to come, for a discussion of their individual merits to be\nembraced in any science. Ethics, therefore, does not deal at all with\nfacts of this nature, facts that are unique, individual, absolutely\nparticular; facts with which such studies as history, geography,\nastronomy, are compelled, in part at least, to deal. And, for this\nreason, it is not the business of the ethical philosopher to give\npersonal advice or exhortation.\n\n\n=4.= But there is another meaning which may be given to the question\n\u2018What is good?\u2019 \u2018Books are good\u2019 would be an answer to it, though\nan answer obviously false; for some books are very bad indeed. And\nethical judgments of this kind do indeed belong to Ethics; though\nI shall not deal with many of them. Such is the judgment \u2018Pleasure\nis good\u2019--a judgment, of which Ethics should discuss the truth,\nalthough it is not nearly as important as that other judgment, with\nwhich we shall be much occupied presently--\u2018Pleasure _alone_ is\ngood.\u2019 It is judgments of this sort, which are made in such books on\nEthics as contain a list of \u2018virtues\u2019--in Aristotle\u2019s \u2018Ethics\u2019 for\nexample. But it is judgments of precisely the same kind, which form\nthe substance of what is commonly supposed to be a study different\nfrom Ethics, and one much less respectable--the study of Casuistry.\nWe may be told that Casuistry differs from Ethics, in that it is much\nmore detailed and particular, Ethics much more general. But it is\nmost important to notice that Casuistry does not deal with anything\nthat is absolutely particular--particular in the only sense in which\na perfectly precise line can be drawn between it and what is general.\nIt is not particular in the sense just noticed, the sense in which\nthis book is a particular book, and A\u2019s friend\u2019s advice particular\nadvice. Casuistry may indeed be _more_ particular and Ethics _more_\ngeneral; but that means that they differ only in degree and not in\nkind. And this is universally true of \u2018particular\u2019 and \u2018general,\u2019 when\nused in this common, but inaccurate, sense. So far as Ethics allows\nitself to give lists of virtues or even to name constituents of the\nIdeal, it is indistinguishable from Casuistry. Both alike deal with\nwhat is general, in the sense in which physics and chemistry deal with\nwhat is general. Just as chemistry aims at discovering what are the\nproperties of oxygen, _wherever it occurs_, and not only of this or\nthat particular specimen of oxygen; so Casuistry aims at discovering\nwhat actions are good, _whenever they occur_. In this respect Ethics\nand Casuistry alike are to be classed with such sciences as physics,\nchemistry and physiology, in their absolute distinction from those of\nwhich history and geography are instances. And it is to be noted that,\nowing to their detailed nature, casuistical investigations are actually\nnearer to physics and to chemistry than are the investigations usually\nassigned to Ethics. For just as physics cannot rest content with the\ndiscovery that light is propagated by waves of ether, but must go on\nto discover the particular nature of the ether-waves corresponding to\neach several colour; so Casuistry, not content with the general law\nthat charity is a virtue, must attempt to discover the relative merits\nof every different form of charity. Casuistry forms, therefore, part\nof the ideal of ethical science: Ethics cannot be complete without it.\nThe defects of Casuistry are not defects of principle; no objection can\nbe taken to its aim and object. It has failed only because it is far\ntoo difficult a subject to be treated adequately in our present state\nof knowledge. The casuist has been unable to distinguish, in the cases\nwhich he treats, those elements upon which their value depends. Hence\nhe often thinks two cases to be alike in respect of value, when in\nreality they are alike only in some other respect. It is to mistakes of\nthis kind that the pernicious influence of such investigations has been\ndue. For Casuistry is the goal of ethical investigation. It cannot be\nsafely attempted at the beginning of our studies, but only at the end.\n\n\n=5.= But our question \u2018What is good?\u2019 may have still another meaning.\nWe may, in the third place, mean to ask, not what thing or things are\ngood, but how \u2018good\u2019 is to be defined. This is an enquiry which belongs\nonly to Ethics, not to Casuistry; and this is the enquiry which will\noccupy us first.\n\nIt is an enquiry to which most special attention should be directed;\nsince this question, how \u2018good\u2019 is to be defined, is the most\nfundamental question in all Ethics. That which is meant by \u2018good\u2019\nis, in fact, except its converse \u2018bad,\u2019 the _only_ simple object of\nthought which is peculiar to Ethics. Its definition is, therefore,\nthe most essential point in the definition of Ethics; and moreover a\nmistake with regard to it entails a far larger number of erroneous\nethical judgments than any other. Unless this first question be fully\nunderstood, and its true answer clearly recognised, the rest of Ethics\nis as good as useless from the point of view of systematic knowledge.\nTrue ethical judgments, of the two kinds last dealt with, may indeed\nbe made by those who do not know the answer to this question as well\nas by those who do; and it goes without saying that the two classes\nof people may lead equally good lives. But it is extremely unlikely\nthat the _most general_ ethical judgments will be equally valid, in\nthe absence of a true answer to this question: I shall presently try\nto shew that the gravest errors have been largely due to beliefs in a\nfalse answer. And, in any case, it is impossible that, till the answer\nto this question be known, any one should know _what is the evidence_\nfor any ethical judgment whatsoever. But the main object of Ethics,\nas a systematic science, is to give correct _reasons_ for thinking\nthat this or that is good; and, unless this question be answered, such\nreasons cannot be given. Even, therefore, apart from the fact that a\nfalse answer leads to false conclusions, the present enquiry is a most\nnecessary and important part of the science of Ethics.\n\n\n=6.= What, then, is good? How is good to be defined? Now, it may be\nthought that this is a verbal question. A definition does indeed often\nmean the expressing of one word\u2019s meaning in other words. But this\nis not the sort of definition I am asking for. Such a definition can\nnever be of ultimate importance in any study except lexicography. If I\nwanted that kind of definition I should have to consider in the first\nplace how people generally used the word \u2018good\u2019; but my business is not\nwith its proper usage, as established by custom. I should, indeed, be\nfoolish, if I tried to use it for something which it did not usually\ndenote: if, for instance, I were to announce that, whenever I used the\nword \u2018good,\u2019 I must be understood to be thinking of that object which\nis usually denoted by the word \u2018table.\u2019 I shall, therefore, use the\nword in the sense in which I think it is ordinarily used; but at the\nsame time I am not anxious to discuss whether I am right in thinking\nthat it is so used. My business is solely with that object or idea,\nwhich I hold, rightly or wrongly, that the word is generally used to\nstand for. What I want to discover is the nature of that object or\nidea, and about this I am extremely anxious to arrive at an agreement.\n\nBut, if we understand the question in this sense, my answer to it\nmay seem a very disappointing one. If I am asked \u2018What is good?\u2019 my\nanswer is that good is good, and that is the end of the matter. Or if\nI am asked \u2018How is good to be defined?\u2019 my answer is that it cannot\nbe defined, and that is all I have to say about it. But disappointing\nas these answers may appear, they are of the very last importance. To\nreaders who are familiar with philosophic terminology, I can express\ntheir importance by saying that they amount to this: That propositions\nabout the good are all of them synthetic and never analytic; and that\nis plainly no trivial matter. And the same thing may be expressed more\npopularly, by saying that, if I am right, then nobody can foist upon\nus such an axiom as that \u2018Pleasure is the only good\u2019 or that \u2018The good\nis the desired\u2019 on the pretence that this is \u2018the very meaning of the\nword.\u2019\n\n\n=7.= Let us, then, consider this position. My point is that \u2018good\u2019\nis a simple notion, just as \u2018yellow\u2019 is a simple notion; that, just\nas you cannot, by any manner of means, explain to any one who does\nnot already know it, what yellow is, so you cannot explain what good\nis. Definitions of the kind that I was asking for, definitions which\ndescribe the real nature of the object or notion denoted by a word,\nand which do not merely tell us what the word is used to mean, are\nonly possible when the object or notion in question is something\ncomplex. You can give a definition of a horse, because a horse has many\ndifferent properties and qualities, all of which you can enumerate.\nBut when you have enumerated them all, when you have reduced a horse\nto his simplest terms, then you can no longer define those terms. They\nare simply something which you think of or perceive, and to any one who\ncannot think of or perceive them, you can never, by any definition,\nmake their nature known. It may perhaps be objected to this that we\nare able to describe to others, objects which they have never seen or\nthought of. We can, for instance, make a man understand what a chimaera\nis, although he has never heard of one or seen one. You can tell him\nthat it is an animal with a lioness\u2019s head and body, with a goat\u2019s head\ngrowing from the middle of its back, and with a snake in place of a\ntail. But here the object which you are describing is a complex object;\nit is entirely composed of parts, with which we are all perfectly\nfamiliar--a snake, a goat, a lioness; and we know, too, the manner\nin which those parts are to be put together, because we know what is\nmeant by the middle of a lioness\u2019s back, and where her tail is wont to\ngrow. And so it is with all objects, not previously known, which we are\nable to define: they are all complex; all composed of parts, which may\nthemselves, in the first instance, be capable of similar definition,\nbut which must in the end be reducible to simplest parts, which can no\nlonger be defined. But yellow and good, we say, are not complex: they\nare notions of that simple kind, out of which definitions are composed\nand with which the power of further defining ceases.\n\n\n=8.= When we say, as Webster says, \u2018The definition of horse is \u201cA\nhoofed quadruped of the genus Equus,\u201d\u2019 we may, in fact, mean three\ndifferent things. (1) We may mean merely: \u2018When I say \u201chorse,\u201d you are\nto understand that I am talking about a hoofed quadruped of the genus\nEquus.\u2019 This might be called the arbitrary verbal definition: and I\ndo not mean that good is indefinable in that sense. (2) We may mean,\nas Webster ought to mean: \u2018When most English people say \u201chorse,\u201d they\nmean a hoofed quadruped of the genus Equus.\u2019 This may be called the\nverbal definition proper, and I do not say that good is indefinable\nin this sense either; for it is certainly possible to discover how\npeople use a word: otherwise, we could never have known that \u2018good\u2019 may\nbe translated by \u2018gut\u2019 in German and by \u2018bon\u2019 in French. But (3) we\nmay, when we define horse, mean something much more important. We may\nmean that a certain object, which we all of us know, is composed in a\ncertain manner: that it has four legs, a head, a heart, a liver, etc.,\netc., all of them arranged in definite relations to one another. It is\nin this sense that I deny good to be definable. I say that it is not\ncomposed of any parts, which we can substitute for it in our minds when\nwe are thinking of it. We might think just as clearly and correctly\nabout a horse, if we thought of all its parts and their arrangement\ninstead of thinking of the whole: we could, I say, think how a horse\ndiffered from a donkey just as well, just as truly, in this way, as now\nwe do, only not so easily; but there is nothing whatsoever which we\ncould so substitute for good; and that is what I mean, when I say that\ngood is indefinable.\n\n\n=9.= But I am afraid I have still not removed the chief difficulty\nwhich may prevent acceptance of the proposition that good is\nindefinable. I do not mean to say that _the_ good, that which is good,\nis thus indefinable; if I did think so, I should not be writing\non Ethics, for my main object is to help towards discovering that\ndefinition. It is just because I think there will be less risk of error\nin our search for a definition of \u2018the good,\u2019 that I am now insisting\nthat _good_ is indefinable. I must try to explain the difference\nbetween these two. I suppose it may be granted that \u2018good\u2019 is an\nadjective. Well \u2018the good,\u2019 \u2018that which is good,\u2019 must therefore be the\nsubstantive to which the adjective \u2018good\u2019 will apply: it must be the\nwhole of that to which the adjective will apply, and the adjective must\n_always_ truly apply to it. But if it is that to which the adjective\nwill apply, it must be something different from that adjective itself;\nand the whole of that something different, whatever it is, will be\nour definition of _the_ good. Now it may be that this something will\nhave other adjectives, beside \u2018good,\u2019 that will apply to it. It may be\nfull of pleasure, for example; it may be intelligent: and if these two\nadjectives are really part of its definition, then it will certainly be\ntrue, that pleasure and intelligence are good. And many people appear\nto think that, if we say \u2018Pleasure and intelligence are good,\u2019 or if we\nsay \u2018Only pleasure and intelligence are good,\u2019 we are defining \u2018good.\u2019\nWell, I cannot deny that propositions of this nature may sometimes be\ncalled definitions; I do not know well enough how the word is generally\nused to decide upon this point. I only wish it to be understood that\nthat is not what I mean when I say there is no possible definition of\ngood, and that I shall not mean this if I use the word again. I do most\nfully believe that some true proposition of the form \u2018Intelligence is\ngood and intelligence alone is good\u2019 can be found; if none could be\nfound, our definition of _the_ good would be impossible. As it is,\nI believe _the_ good to be definable; and yet I still say that good\nitself is indefinable.\n\n\n=10.= \u2018Good,\u2019 then, if we mean by it that quality which we assert to\nbelong to a thing, when we say that the thing is good, is incapable\nof any definition, in the most important sense of that word. The most\nimportant sense of \u2018definition\u2019 is that in which a definition states\nwhat are the parts which invariably compose a certain whole; and in\nthis sense \u2018good\u2019 has no definition because it is simple and has no\nparts. It is one of those innumerable objects of thought which are\nthemselves incapable of definition, because they are the ultimate\nterms by reference to which whatever _is_ capable of definition must\nbe defined. That there must be an indefinite number of such terms is\nobvious, on reflection; since we cannot define anything except by\nan analysis, which, when carried as far as it will go, refers us to\nsomething, which is simply different from anything else, and which by\nthat ultimate difference explains the peculiarity of the whole which we\nare defining: for every whole contains some parts which are common to\nother wholes also. There is, therefore, no intrinsic difficulty in the\ncontention that \u2018good\u2019 denotes a simple and indefinable quality. There\nare many other instances of such qualities.\n\nConsider yellow, for example. We may try to define it, by describing\nits physical equivalent; we may state what kind of light-vibrations\nmust stimulate the normal eye, in order that we may perceive it. But a\nmoment\u2019s reflection is sufficient to shew that those light-vibrations\nare not themselves what we mean by yellow. _They_ are not what we\nperceive. Indeed we should never have been able to discover their\nexistence, unless we had first been struck by the patent difference of\nquality between the different colours. The most we can be entitled to\nsay of those vibrations is that they are what corresponds in space to\nthe yellow which we actually perceive.\n\nYet a mistake of this simple kind has commonly been made about \u2018good.\u2019\nIt may be true that all things which are good are _also_ something\nelse, just as it is true that all things which are yellow produce a\ncertain kind of vibration in the light. And it is a fact, that Ethics\naims at discovering what are those other properties belonging to all\nthings which are good. But far too many philosophers have thought that\nwhen they named those other properties they were actually defining\ngood; that these properties, in fact, were simply not \u2018other,\u2019 but\nabsolutely and entirely the same with goodness. This view I propose\nto call the \u2018naturalistic fallacy\u2019 and of it I shall now endeavour to\ndispose.\n\n\n=11.= Let us consider what it is such philosophers say. And first it is\nto be noticed that they do not agree among themselves. They not only\nsay that they are right as to what good is, but they endeavour to prove\nthat other people who say that it is something else, are wrong. One,\nfor instance, will affirm that good is pleasure, another, perhaps, that\ngood is that which is desired; and each of these will argue eagerly\nto prove that the other is wrong. But how is that possible? One of\nthem says that good is nothing but the object of desire, and at the\nsame time tries to prove that it is not pleasure. But from his first\nassertion, that good just means the object of desire, one of two things\nmust follow as regards his proof:\n\n(1) He may be trying to prove that the object of desire is not\npleasure. But, if this be all, where is his Ethics? The position he is\nmaintaining is merely a psychological one. Desire is something which\noccurs in our minds, and pleasure is something else which so occurs;\nand our would-be ethical philosopher is merely holding that the latter\nis not the object of the former. But what has that to do with the\nquestion in dispute? His opponent held the ethical proposition that\npleasure was the good, and although he should prove a million times\nover the psychological proposition that pleasure is not the object of\ndesire, he is no nearer proving his opponent to be wrong. The position\nis like this. One man says a triangle is a circle: another replies\n\u2018A triangle is a straight line, and I will prove to you that I am\nright: _for_\u2019 (this is the only argument) \u2018a straight line is not a\ncircle.\u2019 \u2018That is quite true,\u2019 the other may reply; \u2018but nevertheless\na triangle is a circle, and you have said nothing whatever to prove\nthe contrary. What is proved is that one of us is wrong, for we agree\nthat a triangle cannot be both a straight line and a circle: but which\nis wrong, there can be no earthly means of proving, since you define\ntriangle as straight line and I define it as circle.\u2019--Well, that is\none alternative which any naturalistic Ethics has to face; if good is\n_defined_ as something else, it is then impossible either to prove that\nany other definition is wrong or even to deny such definition.\n\n(2) The other alternative will scarcely be more welcome. It is that the\ndiscussion is after all a verbal one. When A says \u2018Good means pleasant\u2019\nand B says \u2018Good means desired,\u2019 they may merely wish to assert that\nmost people have used the word for what is pleasant and for what is\ndesired respectively. And this is quite an interesting subject for\ndiscussion: only it is not a whit more an ethical discussion than the\nlast was. Nor do I think that any exponent of naturalistic Ethics\nwould be willing to allow that this was all he meant. They are all so\nanxious to persuade us that what they call the good is what we really\nought to do. \u2018Do, pray, act so, because the word \u201cgood\u201d is generally\nused to denote actions of this nature\u2019: such, on this view, would be\nthe substance of their teaching. And in so far as they tell us how we\nought to act, their teaching is truly ethical, as they mean it to be.\nBut how perfectly absurd is the reason they would give for it! \u2018You are\nto do this, because most people use a certain word to denote conduct\nsuch as this.\u2019 \u2018You are to say the thing which is not, because most\npeople call it lying.\u2019 That is an argument just as good!--My dear sirs,\nwhat we want to know from you as ethical teachers, is not how people\nuse a word; it is not even, what kind of actions they approve, which\nthe use of this word \u2018good\u2019 may certainly imply: what we want to know\nis simply what _is_ good. We may indeed agree that what most people\ndo think good, is actually so; we shall at all events be glad to know\ntheir opinions: but when we say their opinions about what _is_ good, we\ndo mean what we say; we do not care whether they call that thing which\nthey mean \u2018horse\u2019 or \u2018table\u2019 or \u2018chair,\u2019 \u2018gut\u2019 or \u2018bon\u2019 or \u2018\u1f00\u03b3\u03b1\u03b8\u03cc\u03c2\u2019; we\nwant to know what it is that they so call. When they say \u2018Pleasure is\ngood,\u2019 we cannot believe that they merely mean \u2018Pleasure is pleasure\u2019\nand nothing more than that.\n\n\n=12.= Suppose a man says \u2018I am pleased\u2019; and suppose that is not a lie\nor a mistake but the truth. Well, if it is true, what does that mean?\nIt means that his mind, a certain definite mind, distinguished by\ncertain definite marks from all others, has at this moment a certain\ndefinite feeling called pleasure. \u2018Pleased\u2019 _means_ nothing but having\npleasure, and though we may be more pleased or less pleased, and even,\nwe may admit for the present, have one or another kind of pleasure;\nyet in so far as it is pleasure we have, whether there be more or less\nof it, and whether it be of one kind or another, what we have is one\ndefinite thing, absolutely indefinable, some one thing that is the\nsame in all the various degrees and in all the various kinds of it\nthat there may be. We may be able to say how it is related to other\nthings: that, for example, it is in the mind, that it causes desire,\nthat we are conscious of it, etc., etc. We can, I say, describe its\nrelations to other things, but define it we can _not_. And if anybody\ntried to define pleasure for us as being any other natural object; if\nanybody were to say, for instance, that pleasure _means_ the sensation\nof red, and were to proceed to deduce from that that pleasure is a\ncolour, we should be entitled to laugh at him and to distrust his\nfuture statements about pleasure. Well, that would be the same fallacy\nwhich I have called the naturalistic fallacy. That \u2018pleased\u2019 does not\nmean \u2018having the sensation of red,\u2019 or anything else whatever, does not\nprevent us from understanding what it does mean. It is enough for us to\nknow that \u2018pleased\u2019 does mean \u2018having the sensation of pleasure,\u2019 and\nthough pleasure is absolutely indefinable, though pleasure is pleasure\nand nothing else whatever, yet we feel no difficulty in saying that we\nare pleased. The reason is, of course, that when I say \u2018I am pleased,\u2019\nI do _not_ mean that \u2018I\u2019 am the same thing as \u2018having pleasure.\u2019 And\nsimilarly no difficulty need be found in my saying that \u2018pleasure\nis good\u2019 and yet not meaning that \u2018pleasure\u2019 is the same thing as\n\u2018good,\u2019 that pleasure _means_ good, and that good _means_ pleasure.\nIf I were to imagine that when I said \u2018I am pleased,\u2019 I meant that\nI was exactly the same thing as \u2018pleased,\u2019 I should not indeed call\nthat a naturalistic fallacy, although it would be the same fallacy\nas I have called naturalistic with reference to Ethics. The reason\nof this is obvious enough. When a man confuses two natural objects\nwith one another, defining the one, by the other, if for instance,\nhe confuses himself, who is one natural object, with \u2018pleased\u2019 or\nwith \u2018pleasure\u2019 which are others, then there is no reason to call the\nfallacy naturalistic. But if he confuses \u2018good,\u2019 which is not in the\nsame sense a natural object, with any natural object whatever, then\nthere is a reason for calling that a naturalistic fallacy; its being\nmade with regard to \u2018good\u2019 marks it as something quite specific, and\nthis specific mistake deserves a name because it is so common. As for\nthe reasons why good is not to be considered a natural object, they may\nbe reserved for discussion in another place. But, for the present, it\nis sufficient to notice this: Even if it were a natural object, that\nwould not alter the nature of the fallacy nor diminish its importance\none whit. All that I have said about it would remain quite equally\ntrue: only the name which I have called it would not be so appropriate\nas I think it is. And I do not care about the name: what I do care\nabout is the fallacy. It does not matter what we call it, provided we\nrecognise it when we meet with it. It is to be met with in almost every\nbook on Ethics; and yet it is not recognised: and that is why it is\nnecessary to multiply illustrations of it, and convenient to give it a\nname. It is a very simple fallacy indeed. When we say that an orange is\nyellow, we do not think our statement binds us to hold that \u2018orange\u2019\nmeans nothing else than \u2018yellow,\u2019 or that nothing can be yellow but an\norange. Supposing the orange is also sweet! Does that bind us to say\nthat \u2018sweet\u2019 is exactly the same thing as \u2018yellow,\u2019 that \u2018sweet\u2019 must\nbe defined as \u2018yellow\u2019? And supposing it be recognised that \u2018yellow\u2019\njust means \u2018yellow\u2019 and nothing else whatever, does that make it any\nmore difficult to hold that oranges are yellow? Most certainly it does\nnot: on the contrary, it would be absolutely meaningless to say that\noranges were yellow, unless yellow did in the end mean just \u2018yellow\u2019\nand nothing else whatever--unless it was absolutely indefinable. We\nshould not get any very clear notion about things, which are yellow--we\nshould not get very far with our science, if we were bound to hold that\neverything which was yellow, _meant_ exactly the same thing as yellow.\nWe should find we had to hold that an orange was exactly the same thing\nas a stool, a piece of paper, a lemon, anything you like. We could\nprove any number of absurdities; but should we be the nearer to the\ntruth? Why, then, should it be different with \u2018good\u2019? Why, if good is\ngood and indefinable, should I be held to deny that pleasure is good?\nIs there any difficulty in holding both to be true at once? On the\ncontrary, there is no meaning in saying that pleasure is good, unless\ngood is something different from pleasure. It is absolutely useless,\nso far as Ethics is concerned, to prove, as Mr Spencer tries to do,\nthat increase of pleasure coincides with increase of life, unless good\n_means_ something different from either life or pleasure. He might just\nas well try to prove that an orange is yellow by shewing that it always\nis wrapped up in paper.\n\n\n=13.= In fact, if it is not the case that \u2018good\u2019 denotes something\nsimple and indefinable, only two alternatives are possible: either it\nis a complex, a given whole, about the correct analysis of which there\nmay be disagreement; or else it means nothing at all, and there is no\nsuch subject as Ethics. In general, however, ethical philosophers have\nattempted to define good, without recognising what such an attempt\nmust mean. They actually use arguments which involve one or both of\nthe absurdities considered in \u00a7 11. We are, therefore, justified in\nconcluding that the attempt to define good is chiefly due to want of\nclearness as to the possible nature of definition. There are, in fact,\nonly two serious alternatives to be considered, in order to establish\nthe conclusion that \u2018good\u2019 does denote a simple and indefinable notion.\nIt might possibly denote a complex, as \u2018horse\u2019 does; or it might have\nno meaning at all. Neither of these possibilities has, however, been\nclearly conceived and seriously maintained, as such, by those who\npresume to define good; and both may be dismissed by a simple appeal to\nfacts.\n\n(1) The hypothesis that disagreement about the meaning of good is\ndisagreement with regard to the correct analysis of a given whole,\nmay be most plainly seen to be incorrect by consideration of the fact\nthat, whatever definition be offered, it may be always asked, with\nsignificance, of the complex so defined, whether it is itself good. To\ntake, for instance, one of the more plausible, because one of the more\ncomplicated, of such proposed definitions, it may easily be thought,\nat first sight, that to be good may mean to be that which we desire\nto desire. Thus if we apply this definition to a particular instance\nand say \u2018When we think that A is good, we are thinking that A is one\nof the things which we desire to desire,\u2019 our proposition may seem\nquite plausible. But, if we carry the investigation further, and ask\nourselves \u2018Is it good to desire to desire A?\u2019 it is apparent, on a\nlittle reflection, that this question is itself as intelligible, as the\noriginal question \u2018Is A good?\u2019--that we are, in fact, now asking for\nexactly the same information about the desire to desire A, for which\nwe formerly asked with regard to A itself. But it is also apparent\nthat the meaning of this second question cannot be correctly analysed\ninto \u2018Is the desire to desire A one of the things which we desire to\ndesire?\u2019: we have not before our minds anything so complicated as the\nquestion \u2018Do we desire to desire to desire to desire A?\u2019 Moreover any\none can easily convince himself by inspection that the predicate of\nthis proposition--\u2018good\u2019--is positively different from the notion of\n\u2018desiring to desire\u2019 which enters into its subject: \u2018That we should\ndesire to desire A is good\u2019 is _not_ merely equivalent to \u2018That A\nshould be good is good.\u2019 It may indeed be true that what we desire to\ndesire is always also good; perhaps, even the converse may be true: but\nit is very doubtful whether this is the case, and the mere fact that we\nunderstand very well what is meant by doubting it, shews clearly that\nwe have two different notions before our minds.\n\n(2) And the same consideration is sufficient to dismiss the hypothesis\nthat \u2018good\u2019 has no meaning whatsoever. It is very natural to make the\nmistake of supposing that what is universally true is of such a nature\nthat its negation would be self-contradictory: the importance which has\nbeen assigned to analytic propositions in the history of philosophy\nshews how easy such a mistake is. And thus it is very easy to conclude\nthat what seems to be a universal ethical principle is in fact an\nidentical proposition; that, if, for example, whatever is called \u2018good\u2019\nseems to be pleasant, the proposition \u2018Pleasure is the good\u2019 does not\nassert a connection between two different notions, but involves only\none, that of pleasure, which is easily recognised as a distinct entity.\nBut whoever will attentively consider with himself what is actually\nbefore his mind when he asks the question \u2018Is pleasure (or whatever\nit may be) after all good?\u2019 can easily satisfy himself that he is not\nmerely wondering whether pleasure is pleasant. And if he will try\nthis experiment with each suggested definition in succession, he may\nbecome expert enough to recognise that in every case he has before his\nmind a unique object, with regard to the connection of which with any\nother object, a distinct question may be asked. Every one does in\nfact understand the question \u2018Is this good?\u2019 When he thinks of it, his\nstate of mind is different from what it would be, were he asked \u2018Is\nthis pleasant, or desired, or approved?\u2019 It has a distinct meaning for\nhim, even though he may not recognise in what respect it is distinct.\nWhenever he thinks of \u2018intrinsic value,\u2019 or \u2018intrinsic worth,\u2019 or\nsays that a thing \u2018ought to exist,\u2019 he has before his mind the unique\nobject--the unique property of things--which I mean by \u2018good.\u2019\nEverybody is constantly aware of this notion, although he may never\nbecome aware at all that it is different from other notions of which\nhe is also aware. But, for correct ethical reasoning, it is extremely\nimportant that he should become aware of this fact; and, as soon as the\nnature of the problem is clearly understood, there should be little\ndifficulty in advancing so far in analysis.\n\n\n=14.= \u2018Good,\u2019 then, is indefinable; and yet, so far as I know, there\nis only one ethical writer, Prof. Henry Sidgwick, who has clearly\nrecognised and stated this fact. We shall see, indeed, how far many of\nthe most reputed ethical systems fall short of drawing the conclusions\nwhich follow from such a recognition. At present I will only quote one\ninstance, which will serve to illustrate the meaning and importance of\nthis principle that \u2018good\u2019 is indefinable, or, as Prof. Sidgwick says,\nan \u2018unanalysable notion.\u2019 It is an instance to which Prof. Sidgwick\nhimself refers in a note on the passage, in which he argues that\n\u2018ought\u2019 is unanalysable[2].\n\n  [2] _Methods of Ethics_, Bk. I, Chap. iii, \u00a7 1 (6th edition).\n\n\u2018Bentham,\u2019 says Sidgwick, \u2018explains that his fundamental principle\n\u201cstates the greatest happiness of all those whose interest is in\nquestion as being the right and proper end of human action\u201d\u2019; and yet\n\u2018his language in other passages of the same chapter would seem to\nimply\u2019 that he _means_ by the word \u201cright\u201d \u201cconducive to the general\nhappiness.\u201d Prof. Sidgwick sees that, if you take these two statements\ntogether, you get the absurd result that \u2018greatest happiness is the\nend of human action, which is conducive to the general happiness\u2019; and\nso absurd does it seem to him to call this result, as Bentham calls\nit, \u2018the fundamental principle of a moral system,\u2019 that he suggests\nthat Bentham cannot have meant it. Yet Prof. Sidgwick himself states\nelsewhere[3] that Psychological Hedonism is \u2018not seldom confounded with\nEgoistic Hedonism\u2019; and that confusion, as we shall see, rests chiefly\non that same fallacy, the naturalistic fallacy, which is implied in\nBentham\u2019s statements. Prof. Sidgwick admits therefore that this fallacy\nis sometimes committed, absurd as it is; and I am inclined to think\nthat Bentham may really have been one of those who committed it. Mill,\nas we shall see, certainly did commit it. In any case, whether Bentham\ncommitted it or not, his doctrine, as above quoted, will serve as a\nvery good illustration of this fallacy, and of the importance of the\ncontrary proposition that good is indefinable.\n\n  [3] _Methods of Ethics_, Bk. I, Chap. iv, \u00a7 1.\n\nLet us consider this doctrine. Bentham seems to imply, so Prof.\nSidgwick says, that the word \u2018right\u2019 _means_ \u2018conducive to general\nhappiness.\u2019 Now this, by itself, need not necessarily involve\nthe naturalistic fallacy. For the word \u2018right\u2019 is very commonly\nappropriated to actions which lead to the attainment of what\nis good; which are regarded as _means_ to the ideal and not as\nends-in-themselves. This use of \u2018right,\u2019 as denoting what is good as\na means, whether or not it be also good as an end, is indeed the use\nto which I shall confine the word. Had Bentham been using \u2018right\u2019 in\nthis sense, it might be perfectly consistent for him to _define_ right\nas \u2018conducive to the general happiness,\u2019 _provided only_ (and notice\nthis proviso) he had already proved, or laid down as an axiom, that\ngeneral happiness was _the_ good, or (what is equivalent to this)\nthat general happiness alone was good. For in that case he would have\nalready defined _the_ good as general happiness (a position perfectly\nconsistent, as we have seen, with the contention that \u2018good\u2019 is\nindefinable), and, since right was to be defined as \u2018conducive to _the_\ngood,\u2019 it would actually _mean_ \u2018conducive to general happiness.\u2019\nBut this method of escape from the charge of having committed the\nnaturalistic fallacy has been closed by Bentham himself. For his\nfundamental principle is, we see, that the greatest happiness of all\nconcerned is the _right_ and proper _end_ of human action. He applies\nthe word \u2018right,\u2019 therefore, to the end, as such, not only to the means\nwhich are conducive to it; and, that being so, right can no longer be\ndefined as \u2018conducive to the general happiness,\u2019 without involving the\nfallacy in question. For now it is obvious that the definition of right\nas conducive to general happiness can be used by him in support of the\nfundamental principle that general happiness is the right end; instead\nof being itself derived from that principle. If right, by definition,\nmeans conducive to general happiness, then it is obvious that general\nhappiness is the right end. It is not necessary now first to prove or\nassert that general happiness is the right end, before right is defined\nas conducive to general happiness--a perfectly valid procedure; but on\nthe contrary the definition of right as conducive to general happiness\nproves general happiness to be the right end--a perfectly invalid\nprocedure, since in this case the statement that \u2018general happiness is\nthe right end of human action\u2019 is not an ethical principle at all, but\neither, as we have seen, a proposition about the meaning of words, or\nelse a proposition about the _nature_ of general happiness, not about\nits rightness or goodness.\n\nNow, I do not wish the importance I assign to this fallacy to be\nmisunderstood. The discovery of it does not at all refute Bentham\u2019s\ncontention that greatest happiness is the proper end of human action,\nif that be understood as an ethical proposition, as he undoubtedly\nintended it. That principle may be true all the same; we shall consider\nwhether it is so in succeeding chapters. Bentham might have maintained\nit, as Professor Sidgwick does, even if the fallacy had been pointed\nout to him. What I am maintaining is that the _reasons_ which he\nactually gives for his ethical proposition are fallacious ones so\nfar as they consist in a definition of right. What I suggest is that\nhe did not perceive them to be fallacious; that, if he had done so,\nhe would have been led to seek for other reasons in support of his\nUtilitarianism; and that, had he sought for other reasons, he _might_\nhave found none which he thought to be sufficient. In that case he\nwould have changed his whole system--a most important consequence. It\nis undoubtedly also possible that he would have thought other reasons\nto be sufficient, and in that case his ethical system, in its main\nresults, would still have stood. But, even in this latter case, his\nuse of the fallacy would be a serious objection to him as an ethical\nphilosopher. For it is the business of Ethics, I must insist, not only\nto obtain true results, but also to find valid reasons for them. The\ndirect object of Ethics is knowledge and not practice; and any one who\nuses the naturalistic fallacy has certainly not fulfilled this first\nobject, however correct his practical principles may be.\n\nMy objections to Naturalism are then, in the first place, that it\noffers no reason at all, far less any valid reason, for any ethical\nprinciple whatever; and in this it already fails to satisfy the\nrequirements of Ethics, as a scientific study. But in the second place\nI contend that, though it gives a reason for no ethical principle, it\nis a _cause_ of the acceptance of false principles--it deludes the mind\ninto accepting ethical principles, which are false; and in this it is\ncontrary to every aim of Ethics. It is easy to see that if we start\nwith a definition of right conduct as conduct conducive to general\nhappiness; then, knowing that right conduct is universally conduct\nconducive to the good, we very easily arrive at the result that the\ngood is general happiness. If, on the other hand, we once recognise\nthat we must start our Ethics without a definition, we shall be much\nmore apt to look about us, before we adopt any ethical principle\nwhatever; and the more we look about us, the less likely are we to\nadopt a false one. It may be replied to this: Yes, but we shall look\nabout us just as much, before we settle on our definition, and are\ntherefore just as likely to be right. But I will try to shew that this\nis not the case. If we start with the conviction that a definition of\ngood can be found, we start with the conviction that good _can mean_\nnothing else than some one property of things; and our only business\nwill then be to discover what that property is. But if we recognise\nthat, so far as the meaning of good goes, anything whatever may be\ngood, we start with a much more open mind. Moreover, apart from the\nfact that, when we think we have a definition, we cannot logically\ndefend our ethical principles in any way whatever, we shall also be\nmuch less apt to defend them well, even if illogically. For we shall\nstart with the conviction that good must mean so and so, and shall\ntherefore be inclined either to misunderstand our opponent\u2019s arguments\nor to cut them short with the reply, \u2018This is not an open question: the\nvery meaning of the word decides it; no one can think otherwise except\nthrough confusion.\u2019\n\n\n=15.= Our first conclusion as to the subject-matter of Ethics is, then,\nthat there is a simple, indefinable, unanalysable object of thought by\nreference to which it must be defined. By what name we call this unique\nobject is a matter of indifference, so long as we clearly recognise\nwhat it is and that it does differ from other objects. The words which\nare commonly taken as the signs of ethical judgments all do refer to\nit; and they are expressions of ethical judgments solely because they\ndo so refer. But they may refer to it in two different ways, which\nit is very important to distinguish, if we are to have a complete\ndefinition of the range of ethical judgments. Before I proceeded to\nargue that there was such an indefinable notion involved in ethical\nnotions, I stated (\u00a7 4) that it was necessary for Ethics to enumerate\nall true universal judgments, asserting that such and such a thing was\ngood, whenever it occurred. But, although all such judgments do refer\nto that unique notion which I have called \u2018good,\u2019 they do not all refer\nto it in the same way. They may either assert that this unique property\ndoes always attach to the thing in question, or else they may assert\nonly that the thing in question is _a cause or necessary condition_\nfor the existence of other things to which this unique property does\nattach. The nature of these two species of universal ethical judgments\nis extremely different; and a great part of the difficulties, which\nare met with in ordinary ethical speculation, are due to the failure\nto distinguish them clearly. Their difference has, indeed, received\nexpression in ordinary language by the contrast between the terms \u2018good\nas means\u2019 and \u2018good in itself,\u2019 \u2018value as a means\u2019 and \u2018intrinsic\nvalue.\u2019 But these terms are apt to be applied correctly only in the\nmore obvious instances; and this seems to be due to the fact that the\ndistinction between the conceptions which they denote has not been made\na separate object of investigation. This distinction may be briefly\npointed out as follows.\n\n\n=16.= Whenever we judge that a thing is \u2018good as a means,\u2019 we are\nmaking a judgment with regard to its causal relations: we judge\n_both_ that it will have a particular kind of effect, _and_ that that\neffect will be good in itself. But to find causal judgments that\nare universally true is notoriously a matter of extreme difficulty.\nThe late date at which most of the physical sciences became exact,\nand the comparative fewness of the laws which they have succeeded\nin establishing even now, are sufficient proofs of this difficulty.\nWith regard, then, to what are the most frequent objects of ethical\njudgments, namely actions, it is obvious that we cannot be satisfied\nthat any of our universal causal judgments are true, even in the\nsense in which scientific laws are so. We cannot even discover\nhypothetical laws of the form \u2018Exactly this action will always, under\nthese conditions, produce exactly that effect.\u2019 But for a correct\nethical judgment with regard to the effects of certain actions we\nrequire more than this in two respects. (1) We require to know\nthat a given action will produce a certain effect, _under whatever\ncircumstances it occurs_. But this is certainly impossible. It is\ncertain that in different circumstances the same action may produce\neffects which are utterly different in all respects upon which the\nvalue of the effects depends. Hence we can never be entitled to more\nthan a _generalisation_--to a proposition of the form \u2018This result\n_generally_ follows this kind of action\u2019; and even this generalisation\nwill only be true, if the circumstances under which the action occurs\nare generally the same. This is in fact the case, to a great extent,\nwithin any one particular age and state of society. But, when we take\nother ages into account, in many most important cases the normal\ncircumstances of a given kind of action will be so different, that the\ngeneralisation which is true for one will not be true for another.\nWith regard then to ethical judgments which assert that a certain\nkind of action is good as a means to a certain kind of effect, none\nwill be _universally_ true; and many, though _generally_ true at one\nperiod, will be generally false at others. But (2) we require to know\nnot only that _one_ good effect will be produced, but that, among all\nsubsequent events affected by the action in question, the balance\nof good will be greater than if any other possible action had been\nperformed. In other words, to judge that an action is generally a means\nto good is to judge not only that it generally does _some_ good, but\nthat it generally does the greatest good of which the circumstances\nadmit. In this respect ethical judgments about the effects of action\ninvolve a difficulty and a complication far greater than that involved\nin the establishment of scientific laws. For the latter we need only\nconsider a single effect; for the former it is essential to consider\nnot only this, but the effects of that effect, and so on as far as our\nview into the future can reach. It is, indeed, obvious that our view\ncan never reach far enough for us to be certain that any action will\nproduce the best possible effects. We must be content, if the greatest\npossible balance of good seems to be produced within a limited period.\nBut it is important to notice that the whole series of effects within\na period of considerable length is actually taken account of in our\ncommon judgments that an action is good as a means; and that hence\nthis additional complication, which makes ethical generalisations so\nfar more difficult to establish than scientific laws, is one which is\ninvolved in actual ethical discussions, and is of practical importance.\nThe commonest rules of conduct involve such considerations as the\nbalancing of future bad health against immediate gains; and even if we\ncan never settle with any certainty how we shall secure the greatest\npossible total of good, we try at least to assure ourselves that\nprobable future evils will not be greater than the immediate good.\n\n\n=17.= There are, then, judgments which state that certain kinds of\nthings have good effects; and such judgments, for the reasons just\ngiven, have the important characteristics (1) that they are unlikely to\nbe true, if they state that the kind of thing in question _always_ has\ngood effects, and (2) that, even if they only state that it _generally_\nhas good effects, many of them will only be true of certain periods\nin the world\u2019s history. On the other hand there are judgments which\nstate that certain kinds of things are themselves good; and these\ndiffer from the last in that, if true at all, they are all of them\nuniversally true. It is, therefore, extremely important to distinguish\nthese two kinds of possible judgments. Both may be expressed in the\nsame language: in both cases we commonly say \u2018Such and such a thing is\ngood.\u2019 But in the one case \u2018good\u2019 will mean \u2018good as means,\u2019 _i.e._\nmerely that the thing is a means to good--will have good effects:\nin the other case it will mean \u2018good as end\u2019--we shall be judging\nthat the thing itself has the property which, in the first case, we\nasserted only to belong to its effects. It is plain that these are very\ndifferent assertions to make about a thing; it is plain that either or\nboth of them may be made, both truly and falsely, about all manner of\nthings; and it is certain that unless we are clear as to which of the\ntwo we mean to assert, we shall have a very poor chance of deciding\nrightly whether our assertion is true or false. It is precisely this\nclearness as to the meaning of the question asked which has hitherto\nbeen almost entirely lacking in ethical speculation. Ethics has always\nbeen predominantly concerned with the investigation of a limited\nclass of actions. With regard to these we may ask _both_ how far they\nare good in themselves _and_ how far they have a general tendency to\nproduce good results. And the arguments brought forward in ethical\ndiscussion have always been of both classes--both such as would prove\nthe conduct in question to be good in itself and such as would prove\nit to be good as a means. But that these are the only questions which\nany ethical discussion can have to settle, and that to settle the one\nis _not_ the same thing as to settle the other--these two fundamental\nfacts have in general escaped the notice of ethical philosophers.\nEthical questions are commonly asked in an ambiguous form. It is asked\n\u2018What is a man\u2019s duty under these circumstances?\u2019 or \u2018Is it right to\nact in this way?\u2019 or \u2018What ought we to aim at securing?\u2019 But all these\nquestions are capable of further analysis; a correct answer to any\nof them involves both judgments of what is good in itself and causal\njudgments. This is implied even by those who maintain that we have a\ndirect and immediate judgment of absolute rights and duties. Such a\njudgment can only mean that the course of action in question is _the_\nbest thing to do; that, by acting so, every good that _can_ be secured\nwill have been secured. Now we are not concerned with the question\nwhether such a judgment will ever be true. The question is: What\ndoes it imply, if it is true? And the only possible answer is that,\nwhether true or false, it implies both a proposition as to the degree\nof goodness of the action in question, as compared with other things,\nand a number of causal propositions. For it cannot be denied that the\naction will have consequences: and to deny that the consequences matter\nis to make a judgment of their intrinsic value, as compared with the\naction itself. In asserting that the action is _the_ best thing to do,\nwe assert that it together with its consequences presents a greater sum\nof intrinsic value than any possible alternative. And this condition\nmay be realised by any of the three cases:--(_a_) If the action itself\nhas greater intrinsic value than any alternative, whereas both its\nconsequences and those of the alternatives are absolutely devoid\neither of intrinsic merit or intrinsic demerit; or (_b_) if, though\nits consequences are intrinsically bad, the balance of intrinsic value\nis greater than would be produced by any alternative; or (_c_) if, its\nconsequences being intrinsically good, the degree of value belonging\nto them and it conjointly is greater than that of any alternative\nseries. In short, to assert that a certain line of conduct is, at a\ngiven time, absolutely right or obligatory, is obviously to assert\nthat more good or less evil will exist in the world, if it be adopted,\nthan if anything else be done instead. But this implies a judgment as\nto the value both of its own consequences and of those of any possible\nalternative. And that an action will have such and such consequences\ninvolves a number of causal judgments.\n\nSimilarly, in answering the question \u2018What ought we to aim at\nsecuring?\u2019 causal judgments are again involved, but in a somewhat\ndifferent way. We are liable to forget, because it is so obvious,\nthat this question can never be answered correctly except by naming\nsomething which _can_ be secured. Not everything can be secured; and,\neven if we judge that nothing which cannot be obtained would be of\nequal value with that which can, the possibility of the latter, as\nwell as its value, is essential to its being a proper end of action.\nAccordingly neither our judgments as to what actions we ought to\nperform, nor even our judgments as to the ends which they ought to\nproduce, are pure judgments of intrinsic value. With regard to\nthe former, an action which is absolutely obligatory _may_ have no\nintrinsic value whatsoever; that it is perfectly virtuous may mean\nmerely that it causes the best possible effects. And with regard to the\nlatter, these best possible results which justify our action can, in\nany case, have only so much of intrinsic value as the laws of nature\nallow us to secure; and they in their turn _may_ have no intrinsic\nvalue whatsoever, but may merely be a means to the attainment (in a\nstill further future) of something that has such value. Whenever,\ntherefore, we ask \u2018What ought we to do?\u2019 or \u2018What ought we to try to\nget?\u2019 we are asking questions which involve a correct answer to two\nothers, completely different in kind from one another. We must know\n_both_ what degree of intrinsic value different things have, _and_\nhow these different things may be obtained. But the vast majority of\nquestions which have actually been discussed in Ethics--_all_ practical\nquestions, indeed--involve this double knowledge; and they have been\ndiscussed without any clear separation of the two distinct questions\ninvolved. A great part of the vast disagreements prevalent in Ethics is\nto be attributed to this failure in analysis. By the use of conceptions\nwhich involve both that of intrinsic value and that of causal relation,\nas if they involved intrinsic value only, two different errors have\nbeen rendered almost universal. Either it is assumed that nothing has\nintrinsic value which is not possible, or else it is assumed that what\nis necessary must have intrinsic value. Hence the primary and peculiar\nbusiness of Ethics, the determination what things have intrinsic\nvalue and in what degrees, has received no adequate treatment at all.\nAnd on the other hand a _thorough_ discussion of means has been also\nlargely neglected, owing to an obscure perception of the truth that\nit is perfectly irrelevant to the question of intrinsic values. But\nhowever this may be, and however strongly any particular reader may be\nconvinced that some one of the mutually contradictory systems which\nhold the field has given a correct answer either to the question what\nhas intrinsic value, or to the question what we ought to do, or to\nboth, it must at least be admitted that the questions what is best\nin itself and what will bring about the best possible, are utterly\ndistinct; that both belong to the actual subject-matter of Ethics; and\nthat the more clearly distinct questions are distinguished, the better\nis our chance of answering both correctly.\n\n\n=18.= There remains one point which must not be omitted in a complete\ndescription of the kind of questions which Ethics has to answer. The\nmain division of those questions is, as I have said, into two; the\nquestion what things are good in themselves, and the question to what\nother things these are related as effects. The first of these, which is\nthe primary ethical question and is presupposed by the other, includes\na correct comparison of the various things which have intrinsic value\n(if there are many such) in respect of the degree of value which they\nhave; and such comparison involves a difficulty of principle which has\ngreatly aided the confusion of intrinsic value with mere \u2018goodness\nas a means.\u2019 It has been pointed out that one difference between a\njudgment which asserts that a thing is good in itself, and a judgment\nwhich asserts that it is a means to good, consists in the fact that\nthe first, if true of one instance of the thing in question, is\nnecessarily true of all; whereas a thing which has good effects under\nsome circumstances may have bad ones under others. Now it is certainly\ntrue that all judgments of intrinsic value are in this sense universal;\nbut the principle which I have now to enunciate may easily make it\nappear as if they were not so but resembled the judgment of means in\nbeing merely general. There is, as will presently be maintained, a\nvast number of different things, each of which has intrinsic value;\nthere are also very many which are positively bad; and there is a still\nlarger class of things, which appear to be indifferent. But a thing\nbelonging to any of these three classes may occur as part of a whole,\nwhich includes among its other parts other things belonging both to\nthe same and to the other two classes; and these wholes, as such, may\nalso have intrinsic value. The paradox, to which it is necessary to\ncall attention, is that _the value of such a whole bears no regular\nproportion to the sum of the values of its parts_. It is certain that\na good thing may exist in such a relation to another good thing that\nthe value of the whole thus formed is immensely greater than the sum\nof the values of the two good things. It is certain that a whole formed\nof a good thing and an indifferent thing may have immensely greater\nvalue than that good thing itself possesses. It is certain that two\nbad things or a bad thing and an indifferent thing may form a whole\nmuch worse than the sum of badness of its parts. And it seems as if\nindifferent things may also be the sole constituents of a whole which\nhas great value, either positive or negative. Whether the addition of\na bad thing to a good whole may increase the positive value of the\nwhole, or the addition of a bad thing to a bad may produce a whole\nhaving positive value, may seem more doubtful; but it is, at least,\npossible, and this possibility must be taken into account in our\nethical investigations. However we may decide particular questions, the\nprinciple is clear. _The value of a whole must not be assumed to be the\nsame as the sum of the values of its parts._\n\nA single instance will suffice to illustrate the kind of relation in\nquestion. It seems to be true that to be conscious of a beautiful\nobject is a thing of great intrinsic value; whereas the same object, if\nno one be conscious of it, has certainly comparatively little value,\nand is commonly held to have none at all. But the consciousness of a\nbeautiful object is certainly a whole of some sort in which we can\ndistinguish as parts the object on the one hand and the being conscious\non the other. Now this latter factor occurs as part of a different\nwhole, whenever we are conscious of anything; and it would seem that\nsome of these wholes have at all events very little value, and may\neven be indifferent or positively bad. Yet we cannot always attribute\nthe slightness of their value to any positive demerit in the object\nwhich differentiates them from the consciousness of beauty; the object\nitself may approach as near as possible to absolute neutrality. Since,\ntherefore, mere consciousness does not always confer great value upon\nthe whole of which it forms a part, even though its object may have\nno great demerit, we cannot attribute the great superiority of the\nconsciousness of a beautiful thing over the beautiful thing itself\nto the mere addition of the value of consciousness to that of the\nbeautiful thing. Whatever the intrinsic value of consciousness may\nbe, it does not give to the whole of which it forms a part a value\nproportioned to the sum of its value and that of its object. If this\nbe so, we have here an instance of a whole possessing a different\nintrinsic value from the sum of that of its parts; and whether it be so\nor not, what is meant by such a difference is illustrated by this case.\n\n\n=19.= There are, then, wholes which possess the property that their\nvalue is different from the sum of the values of their parts; and the\nrelations which subsist between such parts and the whole of which they\nform a part have not hitherto been distinctly recognised or received\na separate name. Two points are especially worthy of notice. (1) It\nis plain that the existence of any such part is a necessary condition\nfor the existence of that good which is constituted by the whole. And\nexactly the same language will also express the relation between a\nmeans and the good thing which is its effect. But yet there is a most\nimportant difference between the two cases, constituted by the fact\nthat the part is, whereas the means is not, a part of the good thing\nfor the existence of which its existence is a necessary condition. The\nnecessity by which, if the good in question is to exist, the means to\nit must exist is merely a natural or causal necessity. If the laws of\nnature were different, exactly the same good might exist, although\nwhat is now a necessary condition of its existence did not exist.\nThe existence of the means has no intrinsic value; and its utter\nannihilation would leave the value of that which it is now necessary\nto secure entirely unchanged. But in the case of a part of such a\nwhole as we are now considering, it is otherwise. In this case the\ngood in question cannot conceivably exist, unless the part exist also.\nThe necessity which connects the two is quite independent of natural\nlaw. What is asserted to have intrinsic value is the existence of the\nwhole; and the existence of the whole includes the existence of its\npart. Suppose the part removed, and what remains is _not_ what was\nasserted to have intrinsic value; but if we suppose a means removed,\nwhat remains is just what _was_ asserted to have intrinsic value. And\nyet (2) the existence of the part may _itself_ have no more intrinsic\nvalue than that of the means. It is this fact which constitutes the\nparadox of the relation which we are discussing. It has just been\nsaid that what has intrinsic value is the existence of the whole, and\nthat this includes the existence of the part; and from this it would\nseem a natural inference that the existence of the part has intrinsic\nvalue. But the inference would be as false as if we were to conclude\nthat, because the number of two stones was two, each of the stones was\nalso two. The part of a valuable whole retains exactly the same value\nwhen it is, as when it is not, a part of that whole. If it had value\nunder other circumstances, its value is not any greater, when it is\npart of a far more valuable whole; and if it had no value by itself,\nit has none still, however great be that of the whole of which it now\nforms a part. We are not then justified in asserting that one and the\nsame thing is under some circumstances intrinsically good, and under\nothers not so; as we are justified in asserting of a means that it\nsometimes does and sometimes does not produce good results. And yet we\nare justified in asserting that it is far more desirable that a certain\nthing should exist under some circumstances than under others; namely\nwhen other things will exist in such relations to it as to form a more\nvaluable whole. _It_ will not have more intrinsic value under these\ncircumstances than under others; _it_ will not necessarily even be a\nmeans to the existence of things having more intrinsic value: but it\nwill, like a means, be a necessary condition for the existence of that\nwhich _has_ greater intrinsic value, although, unlike a means, it will\nitself form a part of this more valuable existent.\n\n\n=20.= I have said that the peculiar relation between part and whole\nwhich I have just been trying to define is one which has received no\nseparate name. It would, however, be useful that it should have one;\nand there is a name, which might well be appropriated to it, if only\nit could be divorced from its present unfortunate usage. Philosophers,\nespecially those who profess to have derived great benefit from the\nwritings of Hegel, have latterly made much use of the terms \u2018organic\nwhole,\u2019 \u2018organic unity,\u2019 \u2018organic relation.\u2019 The reason why these terms\nmight well be appropriated to the use suggested is that the peculiar\nrelation of parts to whole, just defined, is one of the properties\nwhich distinguishes the wholes to which they are actually applied with\nthe greatest frequency. And the reason why it is desirable that they\nshould be divorced from their present usage is that, as at present\nused, they have no distinct sense and, on the contrary, both imply and\npropagate errors of confusion.\n\nTo say that a thing is an \u2018organic whole\u2019 is generally understood\nto imply that its parts are related to one another and to itself\nas means to end; it is also understood to imply that they have a\nproperty described in some such phrase as that they have \u2018no meaning\nor significance apart from the whole\u2019; and finally such a whole is\nalso treated as if it had the property to which I am proposing that\nthe name should be confined. But those who use the term give us, in\ngeneral, no hint as to how they suppose these three properties to be\nrelated to one another. It seems generally to be assumed that they are\nidentical; and always, at least, that they are necessarily connected\nwith one another. That they are not identical I have already tried to\nshew; to suppose them so is to neglect the very distinctions pointed\nout in the last paragraph; and the usage might well be discontinued\nmerely because it encourages such neglect. But a still more cogent\nreason for its discontinuance is that, so far from being necessarily\nconnected, the second is a property which can attach to nothing, being\na self-contradictory conception; whereas the first, if we insist on its\nmost important sense, applies to many cases, to which we have no reason\nto think that the third applies also, and the third certainly applies\nto many to which the first does not apply.\n\n\n=21.= These relations between the three properties just distinguished\nmay be illustrated by reference to a whole of the kind from which\nthe name \u2018organic\u2019 was derived--a whole which is an organism in the\nscientific sense--namely the human body.\n\n(1) There exists between many parts of our body (though not between\nall) a relation which has been familiarised by the fable, attributed\nto Menenius Agrippa, concerning the belly and its members. We can\nfind in it parts such that the continued existence of the one is a\nnecessary condition for the continued existence of the other; while the\ncontinued existence of this latter is also a necessary condition for\nthe continued existence of the former. This amounts to no more than\nsaying that in the body we have instances of two things, both enduring\nfor some time, which have a relation of mutual causal dependence on one\nanother--a relation of \u2018reciprocity.\u2019 Frequently no more than this is\nmeant by saying that the parts of the body form an \u2018organic unity,\u2019 or\nthat they are mutually means and ends to one another. And we certainly\nhave here a striking characteristic of living things. But it would be\nextremely rash to assert that this relation of mutual causal dependence\nwas only exhibited by living things and hence was sufficient to define\ntheir peculiarity. And it is obvious that of two things which have this\nrelation of mutual dependence, neither may have intrinsic value, or\none may have it and the other lack it. They are not necessarily \u2018ends\u2019\nto one another in any sense except that in which \u2018end\u2019 means \u2018effect.\u2019\nAnd moreover it is plain that in this sense the whole cannot be an end\nto any of its parts. We are apt to talk of \u2018the whole\u2019 in contrast to\none of its parts, when in fact we mean only _the rest_ of the parts.\nBut strictly the whole must include all its parts and no part can\nbe a cause of the whole, because it cannot be a cause of itself. It\nis plain, therefore, that this relation of mutual causal dependence\nimplies nothing with regard to the value of either of the objects which\nhave it; and that, even if both of them happen also to have value, this\nrelation between them is one which cannot hold between part and whole.\n\nBut (2) it may also be the case that our body as a whole has a value\ngreater than the sum of values of its parts; and this may be what is\nmeant when it is said that the parts are means to the whole. It is\nobvious that if we ask the question \u2018Why _should_ the parts be such as\nthey are?\u2019 a proper answer may be \u2018Because the whole they form has so\nmuch value.\u2019 But it is equally obvious that the relation which we thus\nassert to exist between part and whole is quite different from that\nwhich we assert to exist between part and part when we say \u2018This part\nexists, because that one could not exist without it.\u2019 In the latter\ncase we assert the two parts to be causally connected; but, in the\nformer, part and whole cannot be causally connected, and the relation\nwhich we assert to exist between them may exist even though the parts\nare not causally connected either. All the parts of a picture do not\nhave that relation of mutual causal dependence, which certain parts of\nthe body have, and yet the existence of those which do not have it may\nbe absolutely essential to the value of the whole. The two relations\nare quite distinct in kind, and we cannot infer the existence of the\none from that of the other. It can, therefore, serve no useful purpose\nto include them both under the same name; and if we are to say that\na whole is organic because its parts are (in this sense) \u2018means\u2019 to\nthe whole, we must _not_ say that it is organic because its parts are\ncausally dependent on one another.\n\n\n=22.= But finally (3) the sense which has been most prominent in\nrecent uses of the term \u2018organic whole\u2019 is one whereby it asserts the\nparts of such a whole to have a property which the parts of no whole\ncan possibly have. It is supposed that just as the whole would not\nbe what it is but for the existence of the parts, so the parts would\nnot be what they are but for the existence of the whole; and this is\nunderstood to mean not merely that any particular part could not exist\nunless the others existed too (which is the case where relation (1)\nexists between the parts), but actually that the part is no distinct\nobject of thought--that the whole, of which it is a part, is in its\nturn a part of it. That this supposition is self-contradictory a very\nlittle reflection should be sufficient to shew. We may admit, indeed,\nthat when a particular thing is a part of a whole, it does possess a\npredicate which it would not otherwise possess--namely that it is a\npart of that whole. But what cannot be admitted is that this predicate\nalters the nature or enters into the definition of the thing which has\nit. When we think of the part _itself_, we mean just _that which_ we\nassert, in this case, to _have_ the predicate that it is part of the\nwhole; and the mere assertion that _it_ is a part of the whole involves\nthat it should itself be distinct from that which we assert of it.\nOtherwise we contradict ourselves since we assert that, not _it_, but\nsomething else--namely it together with that which we assert of it--has\nthe predicate which we assert of it. In short, it is obvious that no\npart contains analytically the whole to which it belongs, or any other\nparts of that whole. The relation of part to whole is _not_ the same\nas that of whole to part; and the very definition of the latter is\nthat it does contain analytically that which is said to be its part.\nAnd yet this very self-contradictory doctrine is the chief mark which\nshews the influence of Hegel upon modern philosophy--an influence which\npervades almost the whole of orthodox philosophy. This is what is\ngenerally implied by the cry against falsification by abstraction: that\na whole is always a part of its part! \u2018If you want to know the truth\nabout a part,\u2019 we are told, \u2018you must consider _not_ that part, but\nsomething else--namely the whole: _nothing_ is true of the part, but\nonly of the whole.\u2019 Yet plainly it must be true of the part at least\nthat it is a part of the whole; and it is obvious that when we say it\nis, we do _not_ mean merely that the whole is a part of itself. This\ndoctrine, therefore, that a part can have \u2018no meaning or significance\napart from its whole\u2019 must be utterly rejected. It implies itself that\nthe statement \u2018This is a part of that whole\u2019 has a meaning; and in\norder that this may have one, both subject and predicate must have a\ndistinct meaning. And it is easy to see how this false doctrine has\narisen by confusion with the two relations (1) and (2) which may really\nbe properties of wholes.\n\n(_a_) The _existence_ of a part may be connected by a natural or\ncausal necessity with the existence of the other parts of its whole;\nand further what is a part of a whole and what has ceased to be such a\npart, although differing intrinsically from one another, may be called\nby one and the same name. Thus, to take a typical example, if an arm\nbe cut off from the human body, we still call it an arm. Yet an arm,\nwhen it is a part of the body, undoubtedly differs from a dead arm:\nand hence we may easily be led to say \u2018The arm which is a part of the\nbody would not be what it is, if it were not such a part,\u2019 and to think\nthat the contradiction thus expressed is in reality a characteristic\nof things. But, in fact, the dead arm never was a part of the body;\nit is only _partially_ identical with the living arm. Those parts of\nit which are identical with parts of the living arm are exactly the\nsame, whether they belong to the body or not; and in them we have an\nundeniable instance of one and the same thing at one time forming\na part, and at another not forming a part of the presumed \u2018organic\nwhole.\u2019 On the other hand those properties which _are_ possessed by\nthe living, and _not_ by the dead, arm, do not exist in a changed form\nin the latter: they simply do not exist there _at all_. By a causal\nnecessity their existence depends on their having that relation to the\nother parts of the body which we express by saying that they form part\nof it. Yet, most certainly, _if_ they ever did not form part of the\nbody, they _would_ be exactly what they are when they do. That they\ndiffer intrinsically from the properties of the dead arm and that they\nform part of the body are propositions not analytically related to one\nanother. There is no contradiction in supposing them to retain such\nintrinsic differences and yet not to form part of the body.\n\nBut (_b_) when we are told that a living arm has no _meaning_ or\n_significance_ apart from the body to which it belongs, a different\nfallacy is also suggested. \u2018To have meaning or significance\u2019 is\ncommonly used in the sense of \u2018to have importance\u2019; and this again\nmeans \u2018to have value either as a means or as an end.\u2019 Now it is quite\npossible that even a living arm, apart from its body, would have no\nintrinsic value whatever; although the whole of which it is a part\nhas great intrinsic value owing to its presence. Thus we may easily\ncome to say that, _as_ a part of the body, it has great value, whereas\n_by itself_ it would have none; and thus that its whole \u2018meaning\u2019\nlies in its relation to the body. But in fact the value in question\nobviously does not belong to _it_ at all. To have value merely as a\npart is equivalent to having no value at all, but merely being a part\nof that which has it. Owing, however, to neglect of this distinction,\nthe assertion that a part has value, _as a part_, which it would\nnot otherwise have, easily leads to the assumption that it is also\ndifferent, as a part, from what it would otherwise be; for it is, in\nfact, true that two things which have a different value must also\ndiffer in other respects. Hence the assumption that one and the same\nthing, because it is a part of a more valuable whole at one time than\nat another, therefore has more intrinsic value at one time than at\nanother, has encouraged the self-contradictory belief that one and the\nsame thing may be two different things, and that only in one of its\nforms is it truly what it is.\n\nFor these reasons, I shall, where it seems convenient, take the\nliberty to use the term \u2018organic\u2019 with a special sense. I shall use\nit to denote the fact that a whole has an intrinsic value different\nin amount from the sum of the values of its parts. I shall use it to\ndenote this and only this. The term will not imply any causal relation\nwhatever between the parts of the whole in question. And it will not\nimply either, that the parts are inconceivable except as parts of that\nwhole, or that, when they form parts of such a whole, they have a value\ndifferent from that which they would have if they did not. Understood\nin this special and perfectly definite sense the relation of an organic\nwhole to its parts is one of the most important which Ethics has to\nrecognise. A chief part of that science should be occupied in comparing\nthe relative values of various goods; and the grossest errors will be\ncommitted in such comparison if it be assumed that wherever two things\nform a whole, the value of that whole is merely the sum of the values\nof those two things. With this question of \u2018organic wholes,\u2019 then, we\ncomplete the enumeration of the kind of problems, with which it is the\nbusiness of Ethics to deal.\n\n\n=23.= In this chapter I have endeavoured to enforce the following\nconclusions. (1) The peculiarity of Ethics is not that it investigates\nassertions about human conduct, but that it investigates assertions\nabout that property of things which is denoted by the term \u2018good,\u2019 and\nthe converse property denoted by the term \u2018bad.\u2019 It must, in order\nto establish its conclusions, investigate the truth of _all_ such\nassertions, _except_ those which assert the relation of this property\nonly to a single existent (1-4). (2) This property, by reference to\nwhich the subject-matter of Ethics must be defined, is itself simple\nand indefinable (5-14). And (3) all assertions about its relation to\nother things are of two, and only two, kinds: they either assert in\nwhat degree things themselves possess this property, or else they\nassert causal relations between other things and those which possess\nit (15-17). Finally, (4) in considering the different degrees in which\nthings themselves possess this property, we have to take account of\nthe fact that a whole may possess it in a degree different from that\nwhich is obtained by summing the degrees in which its parts possess it\n(18-22).\n\n\n\n\nCHAPTER II.\n\nNATURALISTIC ETHICS.\n\n\n=24.= It results from the conclusions of Chapter I, that all ethical\nquestions fall under one or other of three classes. The first class\ncontains but one question--the question What is the nature of that\npeculiar predicate, the relation of which to other things constitutes\nthe object of all other ethical investigations? or, in other words,\nWhat is _meant_ by good? This first question I have already attempted\nto answer. The peculiar predicate, by reference to which the sphere of\nEthics must be defined, is simple, unanalysable, indefinable. There\nremain two classes of questions with regard to the relation of this\npredicate to other things. We may ask either (1) To what things and\nin what degree does this predicate directly attach? What things are\ngood in themselves? or (2) By what means shall we be able to make what\nexists in the world as good as possible? What causal relations hold\nbetween what is best in itself and other things?\n\nIn this and the two following chapters, I propose to discuss certain\ntheories, which offer us an answer to the question What is good in\nitself? I say advisedly--_an_ answer: for these theories are all\ncharacterised by the fact that, if true, they would simplify the study\nof Ethics very much. They all hold that there is only _one_ kind of\nfact, of which the existence has any value at all. But they all also\npossess another characteristic, which is my reason for grouping them\ntogether and treating them first: namely that the main reason why the\nsingle kind of fact they name has been held to define the sole good,\nis that it has been held to define what is meant by \u2018good\u2019 itself. In\nother words they are all theories of the end or ideal, the adoption of\nwhich has been chiefly caused by the commission of what I have called\nthe naturalistic fallacy: they all confuse the first and second of the\nthree possible questions which Ethics can ask. It is, indeed, this fact\nwhich explains their contention that only a single kind of thing is\ngood. That a thing should be good, it has been thought, _means_ that\nit possesses this single property: and hence (it is thought) only what\npossesses this property is good. The inference seems very natural; and\nyet what is meant by it is self-contradictory. For those who make it\nfail to perceive that their conclusion \u2018what possesses this property is\ngood\u2019 is a significant proposition: that it does not mean either \u2018what\npossesses this property, possesses this property\u2019 or \u2018the word \u201cgood\u201d\ndenotes that a thing possesses this property.\u2019 And yet, if it does\n_not_ mean one or other of these two things, the inference contradicts\nits own premise.\n\nI propose, therefore, to discuss certain theories of what is good in\nitself, which are _based_ on the naturalistic fallacy, in the sense\nthat the commission of this fallacy has been the main cause of their\nwide acceptance. The discussion will be designed both (1) further to\nillustrate the fact that the naturalistic fallacy is a fallacy, or, in\nother words, that we are all aware of a certain simple quality, which\n(and not anything else) is what we mainly mean by the term \u2018good\u2019;\nand (2) to shew that not one, but many different things, possess this\nproperty. For I cannot hope to recommend the doctrine that things which\nare good do not owe their goodness to their common possession of any\nother property, without a criticism of the main doctrines, opposed\nto this, whose power to recommend themselves is proved by their wide\nprevalence.\n\n\n=25.= The theories I propose to discuss may be conveniently divided\ninto two groups. The naturalistic fallacy always implies that when\nwe think \u2018This is good,\u2019 what we are thinking is that the thing in\nquestion bears a definite relation to some one other thing. But this\none thing, by reference to which good is defined, may be either\nwhat I may call a natural object--something of which the existence\nis admittedly an object of experience--or else it may be an object\nwhich is only inferred to exist in a supersensible real world. These\ntwo types of ethical theory I propose to treat separately. Theories\nof the second type may conveniently be called \u2018metaphysical,\u2019 and I\nshall postpone consideration of them till Chapter IV. In this and\nthe following chapter, on the other hand, I shall deal with theories\nwhich owe their prevalence to the supposition that good can be defined\nby reference to a _natural object_; and these are what I mean by the\nname, which gives the title to this chapter, \u2018Naturalistic Ethics.\u2019 It\nshould be observed that the fallacy, by reference to which I define\n\u2018Metaphysical Ethics,\u2019 is the same in kind; and I give it but one name,\nthe naturalistic fallacy. But when we regard the ethical theories\nrecommended by this fallacy, it seems convenient to distinguish those\nwhich consider goodness to consist in a relation to something which\nexists here and now, from those which do not. According to the former,\nEthics is an empirical or positive science: its conclusions could be\nall established by means of empirical observation and induction. But\nthis is not the case with Metaphysical Ethics. There is, therefore,\na marked distinction between these two groups of ethical theories\nbased on the same fallacy. And within Naturalistic theories, too, a\nconvenient division may also be made. There is one natural object,\nnamely pleasure, which has perhaps been as frequently held to be the\nsole good as all the rest put together. And there is, moreover, a\nfurther reason for treating Hedonism separately. That doctrine has, I\nthink, as plainly as any other, owed its prevalence to the naturalistic\nfallacy; but it has had a singular fate in that the writer, who first\nclearly exposed the fallacy of the naturalistic arguments by which\nit had been attempted to _prove_ that pleasure was the sole good,\nhas maintained that nevertheless it _is_ the sole good. I propose,\ntherefore, to divide my discussion of Hedonism from that of other\nNaturalistic theories; treating of Naturalistic Ethics in general in\nthis chapter, and of Hedonism, in particular, in the next.\n\n\n=26.= The subject of the present chapter is, then, ethical theories\nwhich declare that no intrinsic value is to be found except in the\npossession of some one _natural_ property, other than pleasure; and\nwhich declare this because it is supposed that to be \u2018good\u2019 _means_ to\npossess the property in question. Such theories I call \u2018Naturalistic.\u2019\nI have thus appropriated the name Naturalism to a particular method\nof approaching Ethics--a method which, strictly understood, is\ninconsistent with the possibility of any Ethics whatsoever. This\nmethod consists in substituting for \u2018good\u2019 some one property of a\nnatural object or of a collection of natural objects; and in thus\nreplacing Ethics by some one of the natural sciences. In general, the\nscience thus substituted is one of the sciences specially concerned\nwith man, owing to the general mistake (for such I hold it to be)\nof regarding the matter of Ethics as confined to human conduct. In\ngeneral, Psychology has been the science substituted, as by J. S. Mill;\nor Sociology, as by Professor Clifford, and other modern writers.\nBut any other science might equally well be substituted. It is the\nsame fallacy which is implied, when Professor Tyndall recommends us\nto \u2018conform to the laws of matter\u2019: and here the science which it is\nproposed to substitute for Ethics is simply Physics. The name then is\nperfectly general; for, no matter what the something is that good is\nheld to mean, the theory is still Naturalism. Whether good be defined\nas yellow or green or blue, as loud or soft, as round or square, as\nsweet or bitter, as productive of life or productive of pleasure, as\nwilled or desired or felt: whichever of these or of any other object in\nthe world, good may be held to _mean_, the theory, which holds it to\n_mean_ them, will be a naturalistic theory. I have called such theories\nnaturalistic because all of these terms denote properties, simple\nor complex, of some simple or complex natural object; and, before I\nproceed to consider them, it will be well to define what is meant by\n\u2018nature\u2019 and by \u2018natural objects.\u2019\n\nBy \u2018nature,\u2019 then, I do mean and have meant that which is the\nsubject-matter of the natural sciences and also of psychology. It may\nbe said to include all that has existed, does exist, or will exist in\ntime. If we consider whether any object is of such a nature that it may\nbe said to exist now, to have existed, or to be about to exist, then\nwe may know that that object is a natural object, and that nothing, of\nwhich this is not true, is a natural object. Thus for instance, of our\nminds we should say that they did exist yesterday, that they do exist\nto-day, and probably will exist in a minute or two. We shall say that\nwe had thoughts yesterday, which have ceased to exist now, although\ntheir effects may remain: and in so far as those thoughts did exist,\nthey too are natural objects.\n\nThere is, indeed, no difficulty about the \u2018objects\u2019 themselves, in\nthe sense in which I have just used the term. It is easy to say which\nof them are natural, and which (if any) are not natural. But when we\nbegin to consider the properties of objects, then I fear the problem\nis more difficult. Which among the properties of natural objects are\nnatural properties and which are not? For I do not deny that good is a\nproperty of certain natural objects: certain of them, I think, _are_\ngood; and yet I have said that \u2018good\u2019 itself is not a natural property.\nWell, my test for these too also concerns their existence in time.\nCan we imagine \u2018good\u2019 as existing _by itself_ in time, and not merely\nas a property of some natural object? For myself, I cannot so imagine\nit, whereas with the greater number of properties of objects--those\nwhich I call the natural properties--their existence does seem to me to\nbe independent of the existence of those objects. They are, in fact,\nrather parts of which the object is made up than mere predicates which\nattach to it. If they were all taken away, no object would be left, not\neven a bare substance: for they are in themselves substantial and give\nto the object all the substance that it has. But this is not so with\ngood. If indeed good were a feeling, as some would have us believe,\nthen it would exist in time. But that is why to call it so is to commit\nthe naturalistic fallacy. It will always remain pertinent to ask,\nwhether the feeling itself is good; and if so, then good cannot itself\nbe identical with any feeling.\n\n\n=27.= Those theories of Ethics, then, are \u2018naturalistic\u2019 which declare\nthe sole good to consist in some one property of things, which exists\nin time; and which do so because they suppose that \u2018good\u2019 itself can\nbe defined by reference to such a property. And we may now proceed to\nconsider such theories.\n\nAnd, first of all, one of the most famous of ethical maxims is that\nwhich recommends a \u2018life according to nature.\u2019 That was the principle\nof the Stoic Ethics; but, since their Ethics has some claim to be\ncalled metaphysical, I shall not attempt to deal with it here. But\nthe same phrase reappears in Rousseau; and it is not unfrequently\nmaintained even now that what we ought to do is to live naturally. Now\nlet us examine this contention in its general form. It is obvious, in\nthe first place, that we cannot say that everything natural is good,\nexcept perhaps in virtue of some metaphysical theory, such as I shall\ndeal with later. If everything natural is equally good, then certainly\nEthics, as it is ordinarily understood, disappears: for nothing is\nmore certain, from an ethical point of view, than that some things\nare bad and others good; the object of Ethics is, indeed, in chief\npart, to give you general rules whereby you may avoid the one and\nsecure the other. What, then, does \u2018natural\u2019 mean, in this advice to\nlive naturally, since it obviously cannot apply to everything that is\nnatural?\n\nThe phrase seems to point to a vague notion that there is some such\nthing as natural good; to a belief that Nature may be said to fix and\ndecide what shall be good, just as she fixes and decides what shall\nexist. For instance, it may be supposed that \u2018health\u2019 is susceptible\nof a natural definition, that Nature has fixed what health shall be:\nand health, it may be said, is obviously good; hence in this case\nNature has decided the matter; we have only to go to her and ask her\nwhat health is, and we shall know what is good: we shall have based an\nethics upon science. But what is this natural definition of health? I\ncan only conceive that health should be defined in natural terms as\nthe _normal_ state of an organism for undoubtedly disease is also a\nnatural product. To say that health is what is preserved by evolution,\nand what itself tends to preserve, in the struggle for existence, the\norganism which possesses it, comes to the same thing: for the point\nof evolution is that it pretends to give a causal explanation of why\nsome forms of life are normal and others are abnormal; it explains the\norigin of species. When therefore we are told that health is natural,\nwe may presume that what is meant is that it is normal; and that when\nwe are told to pursue health as a natural end, what is implied is that\nthe normal must be good. But is it so obvious that the normal must\nbe good? Is it really obvious that health, for instance, is good?\nWas the excellence of Socrates or of Shakespeare normal? Was it not\nrather abnormal, extraordinary? It is, I think, obvious in the first\nplace, that not all that is good is normal; that, on the contrary, the\nabnormal is often better than the normal: peculiar excellence, as well\nas peculiar viciousness, must obviously be not normal but abnormal.\nYet it may be said that nevertheless the normal is good; and I myself\nam not prepared to dispute that health is good. What I contend is\nthat this must not be taken to be obvious; that it must be regarded\nas an open question. To declare it to be obvious is to suggest the\nnaturalistic fallacy: just as, in some recent books, a proof that\ngenius is diseased, abnormal, has been used in order to suggest that\ngenius ought not to be encouraged. Such reasoning is fallacious, and\ndangerously fallacious. The fact is that in the very words \u2018health\u2019\nand \u2018disease\u2019 we do commonly include the notion that the one is good\nand the other bad. But, when a so-called scientific definition of them\nis attempted, a definition in natural terms, the only one possible is\nthat by way of \u2018normal\u2019 and \u2018abnormal.\u2019 Now, it is easy to prove that\nsome things commonly thought excellent are abnormal; and it follows\nthat they are diseased. But it does not follow, except by virtue of the\nnaturalistic fallacy, that those things, commonly thought good, are\ntherefore bad. All that has really been shewn is that in some cases\nthere is a conflict between the common judgment that genius is good,\nand the common judgment that health is good. It is not sufficiently\nrecognised that the latter judgment has not a whit more warrant for\nits truth than the former; that both are perfectly open questions. It\nmay be true, indeed, that by \u2018healthy\u2019 we do commonly imply \u2018good\u2019;\nbut that only shews that when we so use the word, we do not mean the\nsame thing by it as the thing which is meant in medical science.\nThat health, _when_ the word is used to denote something good, is\ngood, goes no way at all to shew that health, when the word is used\nto denote something normal, is also good. We might as well say that,\nbecause \u2018bull\u2019 denotes an Irish joke and also a certain animal, the\njoke and the animal must be the same thing. We must not, therefore, be\nfrightened by the assertion that a thing is natural into the admission\nthat it is good; good does not, by definition, mean anything that is\nnatural; and it is therefore always an open question whether anything\nthat is natural is good.\n\n\n=28.= But there is another slightly different sense in which the\nword \u2018natural\u2019 is used with an implication that it denotes something\ngood. This is when we speak of natural affections, or unnatural\ncrimes and vices. Here the meaning seems to be, not so much that the\naction or feeling in question is normal or abnormal, as that it is\nnecessary. It is in this connection that we are advised to imitate\nsavages and beasts. Curious advice certainly; but, of course, there\nmay be something in it. I am not here concerned to enquire under what\ncircumstances some of us might with advantage take a lesson from the\ncow. I have really no doubt that such exist. What I am concerned with\nis a certain kind of reason, which I think is sometimes used to support\nthis doctrine--a naturalistic reason. The notion sometimes lying at\nthe bottom of the minds of preachers of this gospel is that we cannot\nimprove on nature. This notion is certainly true, in the sense that\nanything we can do, that may be better than the present state of\nthings, will be a natural product. But that is not what is meant by\nthis phrase; nature is again used to mean a mere part of nature; only\nthis time the part meant is not so much the normal as an arbitrary\nminimum of what is necessary for life. And when this minimum is\nrecommended as \u2018natural\u2019--as the way of life to which Nature points her\nfinger--then the naturalistic fallacy is used. Against this position\nI wish only to point out that though the performance of certain acts,\nnot in themselves desirable, may be _excused_ as necessary means to\nthe preservation of life, that is no reason for _praising_ them, or\nadvising us to limit ourselves to those simple actions which are\nnecessary, if it is possible for us to improve our condition even at\nthe expense of doing what is in this sense unnecessary. Nature does\nindeed set limits to what is possible; she does control the means we\nhave at our disposal for obtaining what is good; and of this fact,\npractical Ethics, as we shall see later, must certainly take account:\nbut when she is supposed to have a preference for what is necessary,\nwhat is necessary means only what is necessary to obtain a certain end,\npresupposed as the highest good; and what the highest good is Nature\ncannot determine. Why should we suppose that what is merely necessary\nto life is _ipso facto_ better than what is necessary to the study of\nmetaphysics, useless as that study may appear? It may be that life\nis only worth living, because it enables us to study metaphysics--is\na necessary means thereto. The fallacy of this argument from nature\nhas been discovered as long ago as Lucian. \u2018I was almost inclined to\nlaugh,\u2019 says Callicratidas, in one of the dialogues imputed to him[4],\n\u2018just now, when Charicles was praising irrational brutes and the\nsavagery of the Scythians: in the heat of his argument he was almost\nrepenting that he was born a Greek. What wonder if lions and bears and\npigs do not act as I was proposing? That which reasoning would fairly\nlead a man to choose, cannot be had by creatures that do not reason,\nsimply because they are so stupid. If Prometheus or some other god had\ngiven each of them the intelligence of a man, then they would not have\nlived in deserts and mountains nor fed on one another. They would have\nbuilt temples just as we do, each would have lived in the centre of\nhis family, and they would have formed a nation bound by mutual laws.\nIs it anything surprising that brutes, who have had the misfortune\nto be unable to obtain by forethought any of the goods, with which\nreasoning provides us, should have missed love too? Lions do not love;\nbut neither do they philosophise; bears do not love; but the reason\nis they do not know the sweets of friendship. It is only men, who, by\ntheir wisdom and their knowledge, after many trials, have chosen what\nis best.\u2019\n\n  [4] \u1f1c\u03c1\u03c9\u03c4\u03b5\u03c2, 436-7.\n\n\n=29.= To argue that a thing is good _because_ it is \u2018natural,\u2019 or\nbad _because_ it is \u2018unnatural,\u2019 in these common senses of the term,\nis therefore certainly fallacious: and yet such arguments are very\nfrequently used. But they do not commonly pretend to give a systematic\ntheory of Ethics. Among attempts to _systematise_ an appeal to nature,\nthat which is now most prevalent is to be found in the application to\nethical questions of the term \u2018Evolution\u2019--in the ethical doctrines\nwhich have been called \u2018Evolutionistic.\u2019 These doctrines are those\nwhich maintain that the course of \u2018evolution,\u2019 while it shews us the\ndirection in which we _are_ developing, thereby and for that reason\nshews us the direction in which we _ought_ to develop. Writers, who\nmaintain such a doctrine, are at present very numerous and very\npopular; and I propose to take as my example the writer, who is\nperhaps the best known of them all--Mr Herbert Spencer. Mr Spencer\u2019s\ndoctrine, it must be owned, does not offer the _clearest_ example of\nthe naturalistic fallacy as used in support of Evolutionistic Ethics.\nA clearer example might be found in the doctrine of Guyau[5], a writer\nwho has lately had considerable vogue in France, but who is not so well\nknown as Spencer. Guyau might almost be called a disciple of Spencer;\nhe is frankly evolutionistic, and frankly naturalistic; and I may\nmention that he does not seem to think that he differs from Spencer by\nreason of his naturalism. The point in which he has criticised Spencer\nconcerns the question how far the ends of \u2018pleasure\u2019 and of \u2018increased\nlife\u2019 coincide as motives and means to the attainment of the ideal: he\ndoes not seem to think that he differs from Spencer in the fundamental\nprinciple that the ideal is \u2018Quantity of life, measured in breadth\nas well as in length,\u2019 or, as Guyau says, \u2018Expansion and intensity\nof life\u2019; nor in the naturalistic reason which he gives for this\nprinciple. And I am not sure that he does differ from Spencer in these\npoints. Spencer does, as I shall shew, use the naturalistic fallacy in\ndetails; but with regard to his fundamental principles, the following\ndoubts occur: Is he fundamentally a Hedonist? And, if so, is he a\nnaturalistic Hedonist? In that case he would better have been treated\nin my next chapter. Does he hold that a tendency to increase quantity\nof life is merely a _criterion_ of good conduct? Or does he hold that\nsuch increase of life is marked out by nature as an end at which we\nought to aim?\n\n  [5] See _Esquisse d\u2019une Morale sans Obligation ni Sanction_, par\n  M. Guyau. 4me \u00e9dition. Paris: F. Alcan, 1896.\n\nI think his language in various places would give colour to all these\nhypotheses; though some of them are mutually inconsistent. I will try\nto discuss the main points.\n\n\n=30.= The modern vogue of \u2018Evolution\u2019 is chiefly owing to Darwin\u2019s\ninvestigations as to the origin of species. Darwin formed a strictly\nbiological hypothesis as to the manner in which certain forms of animal\nlife became established, while others died out and disappeared. His\ntheory was that this might be accounted for, partly at least, in the\nfollowing way. When certain varieties occurred (the cause of their\noccurrence is still, in the main, unknown), it might be that some of\nthe points, in which they varied from their parent species or from\nother species then existing, made them better able to persist in the\nenvironment in which they found themselves--less liable to be killed\noff. They might, for instance, be better able to endure the cold or\nheat or changes of the climate; better able to find nourishment from\nwhat surrounded them; better able to escape from or resist other\nspecies which fed upon them; better fitted to attract or to master the\nother sex. Being thus less liable to die, their numbers relatively to\nother species would increase; and that very increase in their numbers\nmight tend towards the extinction of those other species. This theory,\nto which Darwin gave the name \u2018Natural Selection,\u2019 was also called the\ntheory of survival of the fittest. The natural process which it thus\ndescribed was called evolution. It was very natural to suppose that\nevolution meant evolution from what was lower into what was higher;\nin fact it was observed that at least one species, commonly called\nhigher--the species man--had so survived, and among men again it was\nsupposed that the higher races, ourselves for example, had shewn a\ntendency to survive the lower, such as the North American Indians.\nWe can kill them more easily than they can kill us. The doctrine of\nevolution was then represented as an explanation of how the higher\nspecies survives the lower. Spencer, for example, constantly uses\n\u2018more evolved\u2019 as equivalent to \u2018higher.\u2019 But it is to be noted that\nthis forms no part of Darwin\u2019s scientific theory. That theory will\nexplain, equally well, how by an alteration in the environment (the\ngradual cooling of the earth, for example) quite a different species\nfrom man, a species which we think infinitely lower, might survive us.\nThe survival of the fittest does _not_ mean, as one might suppose, the\nsurvival of what is fittest to fulfil a good purpose best adapted to a\ngood end: at the last, it means merely the survival of the fittest to\nsurvive; and the value of the scientific theory, and it is a theory of\ngreat value, just consists in shewing what are the causes which produce\ncertain biological effects. Whether these effects are good or bad, it\ncannot pretend to judge.\n\n\n=31.= But now let us hear what Mr Spencer says about the application of\nEvolution to Ethics.\n\n\u2018I recur,\u2019 he says[6], \u2018to the main proposition set forth in these\ntwo chapters, which has, I think, been fully justified. Guided by\nthe truth that as the conduct with which Ethics deals is part of\nconduct at large, conduct at large must be generally understood\nbefore this part can be specially understood; and guided by the\nfurther truth that to understand conduct at large we must understand\nthe evolution of conduct; we have been led to see that Ethics has\nfor its subject-matter, that form which universal conduct assumes\nduring the last stages of its evolution. We have also concluded that\nthese last stages in the evolution of conduct are those displayed\nby the _highest_[7] type of being when he is forced, by increase of\nnumbers, to live more and more in presence of his fellows. And there\nhas followed _the corollary that conduct gains ethical sanction_[7]\nin proportion as the activities, becoming less and less militant and\nmore and more industrial, are such as do not necessitate mutual injury\nor hindrance, but consist with, and are furthered by, co-operation and\nmutual aid.\n\n  [6] _Data of Ethics_, Chap. II, \u00a7 7, _ad fin._\n\n  [7] The italics are mine.\n\n\u2018These implications of the Evolution-Hypothesis, we shall now see\nharmonize with the leading moral ideas men have otherwise reached.\u2019\n\nNow, if we are to take the last sentence strictly--if the propositions\nwhich precede it are really thought by Mr Spencer to be _implications_\nof the Evolution-Hypothesis--there can be no doubt that Mr Spencer has\ncommitted the naturalistic fallacy. All that the Evolution-Hypothesis\ntells us is that certain kinds of conduct are more evolved than others;\nand this is, in fact, all that Mr Spencer has attempted to prove in\nthe two chapters concerned. Yet he tells us that one of the things it\nhas proved is that _conduct gains ethical sanction_ in proportion as it\ndisplays certain characteristics. What he has tried to prove is only\nthat, in proportion as it displays those characteristics, it is _more\nevolved_. It is plain, then, that Mr Spencer _identifies_ the gaining\nof ethical sanction with the being more evolved: this follows strictly\nfrom his words. But Mr Spencer\u2019s language is extremely loose; and we\nshall presently see that he seems to regard the view it here implies\nas false. We cannot, therefore, take it as Mr Spencer\u2019s definite view\nthat \u2018better\u2019 means nothing but \u2018more evolved\u2019; or even that what is\n\u2018more evolved\u2019 is _therefore_ \u2018better.\u2019 But we are entitled to urge\nthat he is influenced by these views, and therefore by the naturalistic\nfallacy. It is only by the assumption of such influence that we can\nexplain his confusion as to what he has really proved, and the absence\nof any attempt to prove, what he says he has proved, that conduct\nwhich is more evolved is better. We shall look in vain for any attempt\nto shew that \u2018ethical sanction\u2019 is in proportion to \u2018evolution,\u2019 or\nthat it is the \u2018highest\u2019 type of being which displays the most evolved\nconduct; yet Mr Spencer concludes that this is the case. It is only\nfair to assume that he is not sufficiently conscious how much these\npropositions stand in need of proof--what a very different thing is\nbeing \u2018more evolved\u2019 from being \u2018higher\u2019 or \u2018better.\u2019 It may, of\ncourse, be true that what is more evolved is also higher and better.\nBut Mr Spencer does not seem aware that to assert the one is in any\ncase not the same thing as to assert the other. He argues at length\nthat certain kinds of conduct are \u2018more evolved,\u2019 and then informs\nus that he has proved them to gain ethical sanction in proportion,\nwithout any warning that he has omitted the most essential step in such\na proof. Surely this is sufficient evidence that he does not see how\nessential that step is.\n\n\n=32.= Whatever be the degree of Mr Spencer\u2019s own guilt, what has\njust been said will serve to illustrate the kind of fallacy which\nis constantly committed by those who profess to \u2018base\u2019 Ethics on\nEvolution. But we must hasten to add that the view which Mr Spencer\nelsewhere most emphatically recommends is an utterly different one. It\nwill be useful briefly to deal with this, in order that no injustice\nmay be done to Mr Spencer. The discussion will be instructive partly\nfrom the lack of clearness, which Mr Spencer displays, as to the\nrelation of this view to the \u2018evolutionistic\u2019 one just described; and\npartly because there is reason to suspect that in this view also he is\ninfluenced by the naturalistic fallacy.\n\nWe have seen that, at the end of his second chapter, Mr Spencer seems\nto announce that he has already proved certain characteristics of\nconduct to be a measure of its ethical value. He seems to think that\nhe has proved this merely by considering the evolution of conduct; and\nhe has certainly not given any such proof, unless we are to understand\nthat \u2018more evolved\u2019 is a mere synonym for \u2018ethically better.\u2019 He\nnow promises merely to _confirm_ this certain conclusion by shewing\nthat it \u2018harmonizes with the leading moral ideas men have otherwise\nreached.\u2019 But, when we turn to his third chapter, we find that what\nhe actually does is something quite different. He here asserts that\nto establish the conclusion \u2018Conduct is better in proportion as it\nis more evolved\u2019 an entirely new proof is necessary. That conclusion\nwill be _false_, unless a certain proposition, of which we have heard\nnothing so far, is true--unless it be true that life is _pleasant_ on\nthe whole. And the ethical proposition, for which he claims the support\nof the \u2018leading moral ideas\u2019 of mankind, turns out to be that \u2018life\nis good or bad, according as it does, or does not, bring a surplus\nof agreeable feeling\u2019 (\u00a7 10). Here, then, Mr Spencer appears, not as\nan Evolutionist, but as a Hedonist, in Ethics. No conduct is better,\n_because_ it is more evolved. Degree of evolution can at most be a\n_criterion_ of ethical value; and it will only be that, if we can prove\nthe extremely difficult generalisation that the more evolved is always,\non the whole, the pleasanter. It is plain that Mr Spencer here rejects\nthe naturalistic identification of \u2018better\u2019 with \u2018more evolved\u2019;\nbut it is possible that he is influenced by another naturalistic\nidentification--that of \u2018good\u2019 with \u2018pleasant.\u2019 It is possible that Mr\nSpencer is a naturalistic Hedonist.\n\n\n=33.= Let us examine Mr Spencer\u2019s own words. He begins this third\nchapter by an attempt to shew that _we call_ \u2018good the acts conducive\nto life, in self or others, and bad those which directly or indirectly\ntend towards death, special or general\u2019 (\u00a7 9). And then he asks: \u2018Is\nthere any assumption made\u2019 in so calling them? \u2018Yes\u2019; he answers,\n\u2018an assumption of extreme significance has been made--an assumption\nunderlying all moral estimates. The question to be definitely raised\nand answered before entering on any ethical discussion, is the question\nof late much agitated--Is life worth living? Shall we take the\npessimist view? or shall we take the optimist view?... On the answer\nto this question depends every decision concerning the goodness or\nbadness of conduct.\u2019 But Mr Spencer does not immediately proceed to\ngive the answer. Instead of this, he asks another question: \u2018But now,\nhave these irreconcilable opinions [pessimist and optimist] anything\nin common?\u2019 And this question he immediately answers by the statement:\n\u2018Yes, there is one postulate in which pessimists and optimists agree.\nBoth their arguments assume it to be self-evident that life is good or\nbad, according as it does, or does not, bring a surplus of agreeable\nfeeling\u2019 (\u00a7 10). It is to the defence of this statement that the rest\nof the chapter is devoted; and at the end Mr Spencer formulates his\nconclusion in the following words: \u2018No school can avoid taking for the\nultimate moral aim a desirable state of feeling called by whatever\nname--gratification, enjoyment, happiness. Pleasure somewhere, at\nsome time, to some being or beings, is an inexpugnable element of the\nconception\u2019 (\u00a7 16 _ad fin._).\n\nNow in all this, there are two points to which I wish to call\nattention. The first is that Mr Spencer does not, after all, tell us\nclearly what he takes to be the relation of Pleasure and Evolution in\nethical theory. Obviously he should mean that pleasure is the _only_\nintrinsically desirable thing; that other good things are \u2018good\u2019 only\nin the sense that they are means to its existence. Nothing but this\ncan properly be meant by asserting it to be \u2018_the_ ultimate moral\naim,\u2019 or, as he subsequently says (\u00a7 62 _ad fin._), \u2018_the_ ultimately\nsupreme end.\u2019 And, if this were so, it would follow that the more\nevolved conduct was better than the less evolved, only because, and in\nproportion as, it gave more pleasure. But Mr Spencer tells us that two\nconditions are, taken together, _sufficient_ to prove the more evolved\nconduct better: (1) That it should tend to produce more life; (2) That\nlife should be worth living or contain a balance of pleasure. And the\npoint I wish to emphasise is that if these conditions are sufficient,\nthen pleasure cannot be the sole good. For though to produce more life\nis, if the second of Mr Spencer\u2019s propositions be correct, _one way_ of\nproducing more pleasure, it is not the only way. It is quite possible\nthat a small quantity of life, which was more intensely and uniformly\npresent, should give a greater quantity of pleasure than the greatest\npossible quantity of life that was only just \u2018worth living.\u2019 And in\nthat case, on the hedonistic supposition that pleasure is the only\nthing worth having, we should have to prefer the smaller quantity of\nlife and therefore, according to Mr Spencer, the less evolved conduct.\nAccordingly, if Mr Spencer is a true Hedonist, the fact that life gives\na balance of pleasure is _not_, as he seems to think, sufficient to\nprove that the more evolved conduct is the better. If Mr Spencer means\nus to understand that it _is_ sufficient, then his view about pleasure\ncan only be, not that it is the sole good or \u2018ultimately supreme end,\u2019\nbut that a balance of it is a necessary constituent of the supreme end.\nIn short, Mr Spencer seems to maintain that more life is decidedly\nbetter than less, _if only_ it give a balance of pleasure: and that\ncontention is inconsistent with the position that pleasure is \u2018_the_\nultimate moral aim.\u2019 Mr Spencer implies that of two quantities of life,\nwhich gave an equal amount of pleasure, the larger would nevertheless\nbe preferable to the less. And if this be so, then he must maintain\nthat quantity of life or degree of evolution is itself an ultimate\ncondition of value. He leaves us, therefore, in doubt whether he is\nnot still retaining the Evolutionistic proposition, that the more\nevolved is better, simply because it is more evolved, alongside of\nthe Hedonistic proposition, that the more pleasant is better, simply\nbecause it is more pleasant.\n\nBut the second question which we have to ask is: What reasons has Mr\nSpencer for assigning to pleasure the position which he does assign\nto it? He tells us, we saw, that the \u2018arguments\u2019 both of pessimists\nand of optimists \u2018assume it to be self-evident that life is good or\nbad, according as it does, or does not, bring a surplus of agreeable\nfeeling\u2019; and he betters this later by telling us that \u2018since avowed or\nimplied pessimists, and optimists of one or other shade, taken together\nconstitute all men, it results that this postulate is universally\naccepted\u2019 (\u00a7 16). That these statements are absolutely false is,\nof course, quite obvious: but why does Mr Spencer think them true?\nand, what is more important (a question which Mr Spencer does not\ndistinguish too clearly from the last), why does he think the postulate\nitself to be true? Mr Spencer himself tells us his \u2018proof is\u2019 that\n\u2018reversing the application of the words\u2019 good and bad--applying the\nword \u2018good\u2019 to conduct, the \u2018aggregate results\u2019 of which are painful,\nand the word \u2018bad\u2019 to conduct, of which the \u2018aggregate results\u2019 are\npleasurable--\u2018creates absurdities\u2019 (\u00a7 16). He does not say whether this\nis because it is absurd to think that the quality, which we _mean by\nthe word_ \u2018good,\u2019 really applies to what is painful. Even, however,\nif we assume him to mean this, and if we assume that absurdities are\nthus created, it is plain he would only prove that what is painful is\nproperly thought to be _so far_ bad, and what is pleasant to be _so\nfar_ good: it would not prove at all that pleasure is \u2018_the_ supreme\nend.\u2019 There is, however, reason to think that part of what Mr Spencer\nmeans is the naturalistic fallacy: that he imagines \u2018pleasant\u2019 or\n\u2018productive of pleasure\u2019 is the very meaning of the word \u2018good,\u2019 and\nthat \u2018the absurdity\u2019 is due to this. It is at all events certain that\nhe does not distinguish this possible meaning from that which would\nadmit that \u2018good\u2019 denotes an unique indefinable quality. The doctrine\nof naturalistic Hedonism is, indeed, quite strictly implied in his\nstatement that \u2018virtue\u2019 cannot \u2018_be defined_ otherwise than in terms of\nhappiness\u2019 (\u00a7 13); and, though, as I remarked above, we cannot insist\nupon Mr Spencer\u2019s words as a certain clue to any definite meaning, that\nis only because he generally expresses by them several inconsistent\nalternatives--the naturalistic fallacy being, in this case, one such\nalternative. It is certainly impossible to find any further reasons\ngiven by Mr Spencer for his conviction that pleasure both is the\nsupreme end, and is universally admitted to be so. He seems to assume\nthroughout that we _must_ mean by good conduct what is productive of\npleasure, and by bad what is productive of pain. So far, then, as he is\na Hedonist, he would seem to be a naturalistic Hedonist.\n\nSo much for Mr Spencer. It is, of course, quite possible that his\ntreatment of Ethics contains many interesting and instructive remarks.\nIt would seem, indeed, that Mr Spencer\u2019s main view, that of which he\nis most clearly and most often conscious, is that pleasure is the\nsole good, and that to consider the direction of evolution is by far\nthe best _criterion_ of the way in which we shall get most of it:\nand this theory, _if_ he could establish that amount of pleasure is\nalways in direct proportion to amount of evolution _and also_ that it\nwas plain what conduct was more evolved, _would_ be a very valuable\ncontribution to the science of Sociology; it would even, if pleasure\nwere the sole good, be a valuable contribution to Ethics. But the above\ndiscussion should have made it plain that, if what we want from an\nethical philosopher is a scientific and systematic Ethics, not merely\nan Ethics professedly \u2018based on science\u2019; if what we want is a clear\ndiscussion of the fundamental principles of Ethics, and a statement of\nthe ultimate reasons why one way of acting should be considered better\nthan another--then Mr Spencer\u2019s \u2018Data of Ethics\u2019 is immeasurably far\nfrom satisfying these demands.\n\n\n=34.= It remains only to state clearly what is definitely fallacious\nin prevalent views as to the relation of Evolution to Ethics--in\nthose views with regard to which it seems so uncertain how far Mr\nSpencer intends to encourage them. I proposed to confine the term\n\u2018Evolutionistic Ethics\u2019 to the view that we need only to consider the\ntendency of \u2018evolution\u2019 in order to discover the direction in which\nwe _ought_ to go. This view must be carefully distinguished from\ncertain others, which may be commonly confused with it. (1) It might,\nfor instance, be held that the direction in which living things have\nhitherto developed is, as a matter of fact, the direction of progress.\nIt might be held that the \u2018more evolved\u2019 is, as a matter of fact, also\nbetter. And in such a view no fallacy is involved. But, if it is to\ngive us any guidance as to how we ought to act in the future, it does\ninvolve a long and painful investigation of the exact points in which\nthe superiority of the more evolved consists. We cannot assume that,\nbecause evolution is progress _on the whole_, therefore every point\nin which the more evolved differs from the less is a point in which\nit is better than the less. A simple consideration of the course of\nevolution will therefore, on this view, by no means suffice to inform\nus of the course we ought to pursue. We shall have to employ all the\nresources of a strictly ethical discussion in order to arrive at a\ncorrect valuation of the different results of evolution--to distinguish\nthe more valuable from the less valuable, and both from those which\nare no better than their causes, or perhaps even worse. In fact it\nis difficult to see how, on this view--if all that be meant is that\nevolution has _on the whole_ been a progress--the theory of evolution\ncan give any assistance to Ethics at all. The judgment that evolution\nhas been a progress is itself an independent ethical judgment; and even\nif we take it to be more certain and obvious than any of the detailed\njudgments upon which it must logically depend for confirmation, we\ncertainly cannot use it as a datum from which to infer details. It is,\nat all events, certain that, if this had been the _only_ relation held\nto exist between Evolution and Ethics, no such importance would have\nbeen attached to the bearing of Evolution on Ethics as we actually find\nclaimed for it. (2) The view, which, as I have said, seems to be Mr\nSpencer\u2019s main view, may also be held without fallacy. It may be held\nthat the more evolved, though not itself the better, is a _criterion_,\nbecause a concomitant, of the better. But this view also obviously\ninvolves an exhaustive preliminary discussion of the fundamental\nethical question what, after all, is better. That Mr Spencer entirely\ndispenses with such a discussion in support of his contention that\npleasure is the sole good, I have pointed out; and that, if we attempt\nsuch a discussion, we shall arrive at no such simple result, I shall\npresently try to shew. If however the good is not simple, it is by\nno means likely that we shall be able to discover Evolution to be a\ncriterion of it. We shall have to establish a relation between two\nhighly complicated sets of data; and, moreover, if we had once settled\nwhat were goods, and what their comparative values, it is extremely\nunlikely that we should need to call in the aid of Evolution as a\ncriterion of how to get the most. It is plain, then, again, that if\nthis were the only relation imagined to exist between Evolution and\nEthics, it could hardly have been thought to justify the assignment\nof any importance in Ethics to the theory of Evolution. Finally, (3)\nit may be held that, though Evolution gives us no help in discovering\nwhat results of our efforts will be best, it does give some help in\ndiscovering what it is _possible_ to attain and what are the means to\nits attainment. That the theory really may be of service to Ethics in\nthis way cannot be denied. But it is certainly not common to find this\nhumble, ancillary bearing clearly and exclusively assigned to it. In\nthe mere fact, then, that these non-fallacious views of the relation\nof Evolution to Ethics would give so very little importance to that\nrelation, we have evidence that what is typical in the coupling of the\ntwo names is the fallacious view to which I propose to restrict the\nname \u2018Evolutionistic Ethics.\u2019 This is the view that we ought to move\nin the direction of evolution simply _because_ it is the direction\nof evolution. That the forces of Nature are working on that side is\ntaken as a presumption that it is the right side. That such a view,\napart from metaphysical presuppositions, with which I shall presently\ndeal, is simply fallacious, I have tried to shew. It can only rest on\na confused belief that somehow the good simply _means_ the side on\nwhich Nature is working. And it thus involves another confused belief\nwhich is very marked in Mr Spencer\u2019s whole treatment of Evolution.\nFor, after all, is Evolution the side on which Nature is working? In\nthe sense, which Mr Spencer gives to the term, and in any sense in\nwhich it can be regarded as a fact that the more evolved is higher,\nEvolution denotes only a _temporary_ historical process. That things\nwill permanently continue to evolve in the future, or that they have\nalways evolved in the past, we have not the smallest reason to believe.\nFor Evolution does not, in this sense, denote a natural _law_, like\nthe law of gravity. Darwin\u2019s theory of natural selection does indeed\nstate a natural law: it states that, given certain conditions, certain\nresults will always happen. But Evolution, as Mr Spencer understands\nit and as it is commonly understood, denotes something very different.\nIt denotes only a process which has actually occurred at a given time,\nbecause the conditions at the beginning of that time happened to be of\na certain nature. That such conditions will always be given, or have\nalways been given, cannot be assumed; and it is only the process which,\naccording to natural law, must follow from _these_ conditions and no\nothers, that appears to be also on the whole a progress. Precisely the\nsame natural laws--Darwin\u2019s, for instance--would under other conditions\nrender inevitable not Evolution--not a development from lower to\nhigher--but the converse process, which has been called Involution. Yet\nMr Spencer constantly speaks of the process which is exemplified in\nthe development of man as if it had all the augustness of a universal\nLaw of Nature: whereas we have no reason to believe it other than a\ntemporary accident, requiring not only certain universal natural laws,\nbut also the existence of a certain state of things at a certain time.\nThe only _laws_ concerned in the matter are certainly such as, under\nother circumstances, would allow us to infer, not the development,\nbut the extinction of man. And that circumstances will always be\nfavourable to further development, that Nature will always work on the\nside of Evolution, we have no reason whatever to believe. Thus the\nidea that Evolution throws important light on Ethics seems to be due\nto a double confusion. Our respect for the process is enlisted by the\nrepresentation of it as the Law of Nature. But, on the other hand, our\nrespect for Laws of Nature would be speedily diminished, did we not\nimagine that this desirable process was one of them. To suppose that a\nLaw of Nature is _therefore_ respectable, is to commit the naturalistic\nfallacy; but no one, probably, would be tempted to commit it, unless\nsomething which _is_ respectable, were represented as a Law of Nature.\nIf it were clearly recognised that there is no evidence for supposing\nNature to be on the side of the Good, there would probably be less\ntendency to hold the opinion, which on other grounds is demonstrably\nfalse, that no such evidence is required. And if both false opinions\nwere clearly seen to be false, it would be plain that Evolution has\nvery little indeed to say to Ethics.\n\n\n=35.= In this chapter I have begun the criticism of certain ethical\nviews, which seem to owe their influence mainly to the naturalistic\nfallacy--the fallacy which consists in identifying the simple notion\nwhich we mean by \u2018good\u2019 with some other notion. They are views which\nprofess to tell us what is good in itself; and my criticism of them\nis mainly directed (1) to bring out the negative result, that we have\nno reason to suppose that which they declare to be the sole good,\nreally to be so, (2) to illustrate further the positive result, already\nestablished in Chapter I, that the fundamental principles of Ethics\nmust be _synthetic_ propositions, declaring what things, and in what\ndegree, possess a simple and unanalysable property which may be called\n\u2018intrinsic value\u2019 or \u2018goodness.\u2019 The chapter began (1) by dividing the\nviews to be criticised into (_a_) those which, supposing \u2018good\u2019 to\nbe defined by reference to some supersensible reality, conclude that\nthe sole good is to be found in such a reality, and may therefore be\ncalled \u2018Metaphysical,\u2019 (_b_) those which assign a similar position to\nsome natural object, and may therefore be called \u2018Naturalistic.\u2019 Of\nnaturalistic views, that which regards \u2018pleasure\u2019 as the sole good has\nreceived far the fullest and most serious treatment and was therefore\nreserved for Chapter III: all other forms of Naturalism may be first\ndismissed, by taking typical examples (24-26). (2) As typical of\nnaturalistic views, other than Hedonism, there was first taken the\npopular commendation of what is \u2018natural\u2019: it was pointed out that by\n\u2018natural\u2019 there might here be meant either \u2018normal\u2019 or \u2018necessary,\u2019\nand that neither the \u2018normal\u2019 nor the \u2018necessary\u2019 could be seriously\nsupposed to be either always good or the only good things (27-28). (3)\nBut a more important type, because one which claims to be capable of\nsystem, is to be found in \u2018Evolutionistic Ethics.\u2019 The influence of the\nfallacious opinion that to be \u2018better\u2019 _means_ to be \u2018more evolved\u2019 was\nillustrated by an examination of Mr Herbert Spencer\u2019s Ethics; and it\nwas pointed out that, but for the influence of this opinion, Evolution\ncould hardly have been supposed to have any important bearing upon\nEthics (29-34).\n\n\n\n\nCHAPTER III.\n\nHEDONISM.\n\n\n=36.= In this chapter we have to deal with what is perhaps the most\nfamous and the most widely held of all ethical principles--the\nprinciple that nothing is good but pleasure. My chief reason for\ntreating of this principle in this place is, as I said, that Hedonism\nappears in the main to be a form of Naturalistic Ethics: in other\nwords, that pleasure has been so generally held to be the sole good,\nis almost entirely due to the fact that it has seemed to be somehow\ninvolved in the _definition_ of \u2018good\u2019--to be pointed out by the very\nmeaning of the word. If this is so, then the prevalence of Hedonism has\nbeen mainly due to what I have called the naturalistic fallacy--the\nfailure to distinguish clearly that unique and indefinable quality\nwhich we mean by good. And that it is so, we have very strong evidence\nin the fact that, of all hedonistic writers, Prof. Sidgwick alone has\nclearly recognised that by \u2018good\u2019 we do mean something unanalysable,\nand has alone been led thereby to emphasise the fact that, if\nHedonism be true, its claims to be so must be rested solely on its\nself-evidence--that we must maintain \u2018Pleasure is the sole good\u2019 to be\na mere _intuition_. It appeared to Prof. Sidgwick as a new discovery\nthat what he calls the \u2018method\u2019 of Intuitionism must be retained as\nvalid alongside of, and indeed as the foundation of, what he calls the\nalternative \u2018methods\u2019 of Utilitarianism and Egoism. And that it was a\nnew discovery can hardly be doubted. In previous Hedonists we find no\nclear and consistent recognition of the fact that their fundamental\nproposition involves the assumption that a certain unique predicate\ncan be directly seen to belong to pleasure alone among existents: they\ndo not emphasise, as they could hardly have failed to have done had\nthey perceived it, how utterly independent of all other truths this\ntruth must be.\n\nMoreover it is easy to see how this unique position should have\nbeen assigned to pleasure without any clear consciousness of the\nassumption involved. Hedonism is, for a sufficiently obvious reason,\nthe first conclusion at which any one who begins to reflect upon\nEthics naturally arrives. It is very easy to notice the fact that we\nare pleased with things. The things we enjoy and the things we do not,\nform two unmistakable classes, to which our attention is constantly\ndirected. But it is comparatively difficult to distinguish the fact\nthat we _approve_ a thing from the fact that we are pleased with it.\nAlthough, if we look at the two states of mind, we must see that they\nare different, even though they generally go together, it is very\ndifficult to see in _what respect_ they are different, or that the\ndifference can in any connection be of more importance than the many\nother differences, which are so patent and yet so difficult to analyse,\nbetween one _kind_ of enjoyment and another. It is very difficult to\nsee that by \u2018approving\u2019 of a thing we mean _feeling that it has a\ncertain predicate_--the predicate, namely, which defines the peculiar\nsphere of Ethics; whereas in the enjoyment of a thing no such unique\nobject of thought is involved. Nothing is more natural than the vulgar\nmistake, which we find expressed in a recent book on Ethics[8]: \u2018The\nprimary ethical fact is, we have said, that something is approved\nor disapproved: that is, in other words, the ideal representation\nof certain events in the way of sensation, perception, or idea, is\nattended with a feeling of pleasure or of pain.\u2019 In ordinary speech,\n\u2018I want this,\u2019 \u2018I like this,\u2019 \u2018I care about this\u2019 are constantly used\nas equivalents for \u2018I think this good.\u2019 And in this way it is very\nnatural to be led to suppose that there is no distinct class of ethical\njudgments, but only the class \u2018things enjoyed\u2019; in spite of the fact,\nwhich is very clear, if not very common, that we do not always approve\nwhat we enjoy. It is of course, very obvious that from the supposition\nthat \u2018I think this good\u2019 is identical with \u2018I am pleased with this,\u2019\nit cannot be _logically_ inferred that pleasure alone is good. But, on\nthe other hand, it is very difficult to see what could be logically\ninferred from such a supposition; and it seems _natural_ enough that\nsuch an inference should suggest itself. A very little examination\nof what is commonly written on the subject will suffice to shew that\na logical confusion of this nature is very common. Moreover the very\ncommission of the naturalistic fallacy involves that those who commit\nit should not recognise clearly the meaning of the proposition \u2018This\nis good\u2019--that they should not be able to distinguish this from other\npropositions which seem to resemble it; and, where this is so, it is,\nof course, impossible that its logical relations should be clearly\nperceived.\n\n  [8] A. E. Taylor\u2019s _Problem of Conduct_, p. 120.\n\n\n=37.= There is, therefore, ample reason to suppose that Hedonism is in\ngeneral a form of Naturalism--that its acceptance is generally due to\nthe naturalistic fallacy. It is, indeed, only when we have detected\nthis fallacy, when we have become clearly aware of the unique object\nwhich is meant by \u2018good,\u2019 that we are able to give to Hedonism the\nprecise definition used above, \u2018Nothing is good but pleasure\u2019: and it\nmay, therefore, be objected that, in attacking this doctrine under the\nname of Hedonism, I am attacking a doctrine which has never really been\nheld. But it is very common to hold a doctrine, without being clearly\naware what it is you hold; and though, when Hedonists argue in favour\nof what they call Hedonism, I admit that, in order to suppose their\narguments valid, they must have before their minds something _other_\nthan the doctrine I have defined, yet, in order to draw the conclusions\nthat they draw, it is necessary that they should _also_ have before\ntheir minds this doctrine. In fact, my justification for supposing that\nI shall have refuted _historical_ Hedonism, if I refute the proposition\n\u2018Nothing is good but pleasure,\u2019 is, that although Hedonists have\nrarely stated their principle in this form and though its truth, in\nthis form, will certainly not follow from their arguments, yet their\nethical _method_ will follow logically from nothing else. Any pretence\nof the hedonistic method, to discover to us practical truths which\nwe should not otherwise have known, is founded on the principle that\nthe course of action which will bring the greatest balance of pleasure\nis certainly the right one; and, failing an absolute proof that the\ngreatest balance of pleasure _always_ coincides with the greatest\nbalance of other goods, which it is not generally attempted to give,\nthis principle can only be justified if pleasure be the sole good.\nIndeed it can hardly be doubted that Hedonists are distinguished by\narguing, in disputed practical questions, _as if_ pleasure were the\nsole good; and that it is justifiable, for this among other reasons, to\ntake this as _the_ ethical principle of Hedonism will, I hope, be made\nfurther evident by the whole discussion of this chapter.\n\nBy Hedonism, then, I mean the doctrine that pleasure _alone_ is good\nas an end--\u2018good\u2019 in the sense which I have tried to point out as\nindefinable. The doctrine that pleasure, _among other things_, is good\nas an end, is not Hedonism; and I shall not dispute its truth. Nor\nagain is the doctrine that other things, beside pleasure, are good as\nmeans, at all inconsistent with Hedonism: the Hedonist is not bound\nto maintain that \u2018Pleasure alone is good,\u2019 if under good he includes,\nas we generally do, what is good as means to an end, _as well as_ the\nend itself. In attacking Hedonism, I am therefore simply and solely\nattacking the doctrine that \u2018Pleasure _alone_ is good as an end or in\nitself\u2019: I am not attacking the doctrine that \u2018Pleasure _is_ good as an\nend or in itself,\u2019 nor am I attacking any doctrine whatever as to what\nare the best means we can take in order to obtain pleasure or any other\nend. Hedonists do, in general, recommend a course of conduct which is\nvery similar to that which I should recommend. I do not quarrel with\nthem about most of their practical conclusions, I quarrel only with the\nreasons by which they seem to think their conclusions can be supported;\nand I do emphatically deny that the correctness of their conclusions is\nany ground for inferring the correctness of their principles. A correct\nconclusion may always be obtained by fallacious reasoning; and the good\nlife or virtuous maxims of a Hedonist afford absolutely no presumption\nthat his ethical philosophy is also good. It is his ethical philosophy\nalone with which I am concerned: what I dispute is the excellence of\nhis reasoning, not the excellence of his character as a man or even as\nmoral teacher. It may be thought that my contention is unimportant, but\nthat is no ground for thinking that I am not in the right. What I am\nconcerned with is knowledge only--that we should think correctly and so\nfar arrive at some truth, however unimportant: I do not say that such\nknowledge will make us more useful members of society. If any one does\nnot care for knowledge for its own sake, then I have nothing to say to\nhim; only it should not be thought that a lack of interest in what I\nhave to say is any ground for holding it untrue.\n\n\n=38.= Hedonists, then, hold that all other things but pleasure, whether\nconduct or virtue or knowledge, whether life or nature or beauty, are\nonly good as means to pleasure or for the sake of pleasure, never\nfor their own sakes or as ends in themselves. This view was held by\nAristippus, the disciple of Socrates, and by the Cyrenaic school which\nhe founded; it is associated with Epicurus and the Epicureans; and\nit has been held in modern times, chiefly by those philosophers who\ncall themselves \u2018Utilitarians\u2019--by Bentham, and by Mill, for instance.\nHerbert Spencer, as we have seen, also says he holds it; and Professor\nSidgwick, as we shall see, holds it too.\n\nYet all these philosophers, as has been said, differ from one\nanother more or less, both as to what they mean by Hedonism, and as\nto the reasons for which it is to be accepted as a true doctrine.\nThe matter is therefore obviously not quite so simple as it might\nat first appear. My own object will be to shew quite clearly what\nthe theory must imply, if it is made precise, if all confusions and\ninconsistencies are removed from the conception of it; and, when\nthis is done, I think it will appear that all the various reasons\ngiven for holding it to be true, are really quite inadequate; that\nthey are not reasons for holding Hedonism, but only for holding some\nother doctrine which is confused therewith. In order to attain this\nobject I propose to take first Mill\u2019s doctrine, as set forth in his\nbook called _Utilitarianism_: we shall find in Mill a conception of\nHedonism, and arguments in its favour, which fairly represent those\nof a large class of hedonistic writers. To these representative\nconceptions and arguments grave objections, objections which appear to\nme to be conclusive, have been urged by Professor Sidgwick. These I\nshall try to give in my own words; and shall then proceed to consider\nand refute Professor Sidgwick\u2019s own much more precise conceptions and\narguments. With this, I think, we shall have traversed the whole field\nof Hedonistic doctrine. It will appear, from the discussion, that the\ntask of deciding what is or is not good in itself is by no means an\neasy one; and in this way the discussion will afford a good example\nof the method which it is necessary to pursue in attempting to arrive\nat the truth with regard to this primary class of ethical principles.\nIn particular it will appear that two principles of method must be\nconstantly kept in mind: (1) that the naturalistic fallacy must not\nbe committed; (2) that the distinction between means and ends must be\nobserved.\n\n\n=39.= I propose, then, to begin by an examination of Mill\u2019s\n_Utilitarianism_. That is a book which contains an admirably clear and\nfair discussion of many ethical principles and methods. Mill exposes\nnot a few simple mistakes which are very likely to be made by those\nwho approach ethical problems without much previous reflection. But\nwhat I am concerned with is the mistakes which Mill himself appears\nto have made, and these only so far as they concern the Hedonistic\nprinciple. Let me repeat what that principle is. It is, I said, that\npleasure is the only thing at which we ought to aim, the only thing\nthat is good as an end and for its own sake. And now let us turn to\nMill and see whether he accepts this description of the question at\nissue. \u2018Pleasure,\u2019 he says at the outset, \u2018and freedom from pain, are\nthe only things desirable as ends\u2019 (p. 10[9]); and again, at the end\nof his argument, \u2018To think of an object as desirable (unless for the\nsake of its consequences) and to think of it as pleasant are one and\nthe same thing\u2019 (p. 58). These statements, taken together, and apart\nfrom certain confusions which are obvious in them, seem to imply the\nprinciple I have stated; and if I succeed in shewing that Mill\u2019s\nreasons for them do not prove them, it must at least be admitted that I\nhave not been fighting with shadows or demolishing a man of straw.\n\n  [9] My references are to the 13th edition, 1897.\n\nIt will be observed that Mill adds \u2018absence of pain\u2019 to \u2018pleasure\u2019\nin his first statement, though not in his second. There is, in this,\na confusion, with which, however, we need not deal. I shall talk of\n\u2018pleasure\u2019 alone, for the sake of conciseness; but all my arguments\nwill apply _\u00e0 fortiori_ to \u2018absence of pain\u2019: it is easy to make the\nnecessary substitutions.\n\nMill holds, then, that \u2018happiness is desirable, and _the only thing\ndesirable_[10], as an end; all other things being only desirable\nas means to that end\u2019 (p. 52). Happiness he has already defined as\n\u2018pleasure, and the absence of pain\u2019 (p. 10); he does not pretend that\nthis is more than an arbitrary verbal definition; and, as _such_, I\nhave not a word to say against it. His principle, then, is \u2018pleasure is\nthe only thing desirable,\u2019 if I may be allowed, when I say \u2018pleasure,\u2019\nto include in that word (so far as necessary) absence of pain. And\nnow what are his reasons for holding that principle to be true? He\nhas already told us (p. 6) that \u2018Questions of ultimate ends are not\namenable to direct proof. Whatever can be proved to be good, must be so\nby being shewn to be a means to something _admitted to be good without\nproof_.\u2019 With this, I perfectly agree: indeed the chief object of my\nfirst chapter was to shew that this is so. Anything which is good as an\nend must be admitted to be good without proof. We are agreed so far.\nMill even uses the same examples which I used in my second chapter.\n\u2018How,\u2019 he says, \u2018is it possible to prove that health is good?\u2019 \u2018What\nproof is it possible to give that pleasure is good?\u2019 Well, in Chapter\nIV, in which he deals with the proof of his Utilitarian principle, Mill\nrepeats the above statement in these words: \u2018It has already,\u2019 he says,\n\u2018been remarked, that questions of ultimate ends do not admit of proof,\nin the ordinary acceptation of the term\u2019 (p. 52). \u2018Questions about\nends,\u2019 he goes on in this same passage, \u2018are, in other words, questions\nwhat things are desirable.\u2019 I am quoting these repetitions, because\nthey make it plain what otherwise might have been doubted, that Mill is\nusing the words \u2018desirable\u2019 or \u2018desirable as an end\u2019 as absolutely and\nprecisely equivalent to the words \u2018good as an end.\u2019 We are, then, now\nto hear, what reasons he advances for this doctrine that pleasure alone\nis good as an end.\n\n  [10] My italics.\n\n\n=40.= \u2018Questions about ends,\u2019 he says (pp. 52-3), \u2018are, in other words,\nquestions what things are desirable. The utilitarian doctrine is, that\nhappiness is desirable, and the only thing desirable, as an end; all\nother things being only desirable as means to that end. What ought to\nbe required of this doctrine--what conditions is it requisite that the\ndoctrine should fulfil--to make good its claim to be believed?\n\n\u2018The only proof capable of being given that a thing is visible, is that\npeople actually see it. The only proof that a sound is audible, is\nthat people hear it; and so of the other sources of our experience. In\nlike manner, I apprehend, the sole evidence it is possible to produce\nthat anything is desirable, is that people do actually desire it. If\nthe end which the utilitarian doctrine proposes to itself were not,\nin theory and in practice, acknowledged to be an end, nothing could\never convince any person that it was so. No reason can be given why\nthe general happiness is desirable, except that each person, so far\nas he believes it to be attainable, desires his own happiness. This,\nhowever, being the fact, we have not only all the proof which the case\nadmits of, but all which it is possible to require, that happiness is\na good: that each person\u2019s happiness is a good to that person, and the\ngeneral happiness, therefore, a good to the aggregate of all persons.\nHappiness has made out its title as _one_ of the ends of conduct, and\nconsequently one of the criteria of morality.\u2019\n\nThere, that is enough. That is my first point. Mill has made as na\u00efve\nand artless a use of the naturalistic fallacy as anybody could desire.\n\u2018Good,\u2019 he tells us, means \u2018desirable,\u2019 and you can only find out what\nis desirable by seeking to find out what is actually desired. This is,\nof course, only one step towards the proof of Hedonism; for it may be,\nas Mill goes on to say, that other things beside pleasure are desired.\nWhether or not pleasure is the only thing desired is, as Mill himself\nadmits (p. 58), a psychological question, to which we shall presently\nproceed. The important step for Ethics is this one just taken, the\nstep which pretends to prove that \u2018good\u2019 means \u2018desired.\u2019 Well, the\nfallacy in this step is so obvious, that it is quite wonderful how Mill\nfailed to see it. The fact is that \u2018desirable\u2019 does not mean \u2018able to\nbe desired\u2019 as \u2018visible\u2019 means \u2018able to be seen.\u2019 The desirable means\nsimply what _ought_ to be desired or _deserves_ to be desired; just\nas the detestable means not what can be but what ought to be detested\nand the damnable what deserves to be damned. Mill has, then, smuggled\nin, under cover of the word \u2018desirable,\u2019 the very notion about which\nhe ought to be quite clear. \u2018Desirable\u2019 does indeed mean \u2018what it is\ngood to desire\u2019; but when this is understood, it is no longer plausible\nto say that our only test of _that_, is what is actually desired. Is\nit merely a tautology when the Prayer Book talks of _good_ desires?\nAre not _bad_ desires also possible? Nay, we find Mill himself talking\nof a \u2018better and nobler object of desire\u2019 (p. 10), as if, after all,\nwhat is desired were not _ipso facto_ good, and good in proportion to\nthe amount it is desired. Moreover, if the desired is _ipso facto_ the\ngood; then the good is _ipso facto_ the motive of our actions, and\nthere can be no question of finding motives for doing it, as Mill is\nat such pains to do. If Mill\u2019s explanation of \u2018desirable\u2019 be _true_,\nthen his statement (p. 26) that the rule of action may be _confounded_\nwith the motive of it is untrue: for the motive of action will then be\naccording to him _ipso facto_ its rule; there can be no distinction\nbetween the two, and therefore no confusion, and thus he has\ncontradicted himself flatly. These are specimens of the contradictions,\nwhich, as I have tried to shew, must always follow from the use of\nthe naturalistic fallacy; and I hope I need now say no more about the\nmatter.\n\n\n=41.= Well, then, the first step by which Mill has attempted to\nestablish his Hedonism is simply fallacious. He has attempted to\nestablish the identity of the good with the desired, by confusing the\nproper sense of \u2018desirable,\u2019 in which it denotes that which it is good\nto desire, with the sense which it would bear, if it were analogous\nto such words as \u2018visible.\u2019 If \u2018desirable\u2019 is to be identical with\n\u2018good,\u2019 then it must bear one sense; and if it is to be identical with\n\u2018desired,\u2019 then it must bear quite another sense. And yet to Mill\u2019s\ncontention that the desired is necessarily good, it is quite essential\nthat these two senses of \u2018desirable\u2019 should be the same. If he holds\nthey are the same, then he has contradicted himself elsewhere; if\nhe holds they are not the same, then the first step in his proof of\nHedonism is absolutely worthless.\n\nBut now we must deal with the second step. Having proved, as he thinks,\nthat the good means the desired, Mill recognises that, if he is further\nto maintain that pleasure alone is good, he must prove that pleasure\nalone is really desired. This doctrine that \u2018pleasure alone is the\nobject of all our desires\u2019 is the doctrine which Prof. Sidgwick has\ncalled Psychological Hedonism: and it is a doctrine which most eminent\npsychologists are now agreed in rejecting. But it is a necessary step\nin the proof of any such Naturalistic Hedonism as Mill\u2019s; and it is\nso commonly held, by people not expert either in psychology or in\nphilosophy, that I wish to treat it at some length. It will be seen\nthat Mill does not hold it in this bare form. He admits that other\nthings than pleasure are desired; and this admission is at once a\ncontradiction of his Hedonism. One of the shifts by which he seeks\nto evade this contradiction we shall afterwards consider. But some\nmay think that no such shifts are needed: they may say of Mill, what\nCallicles says of Polus in the _Gorgias_[11], that he has made this\nfatal admission through a most unworthy fear of appearing paradoxical;\nthat they, on the other hand, will have the courage of their\nconvictions, and will not be ashamed to go to any lengths of paradox,\nin defence of what they hold to be the truth.\n\n  [11] 481 C-487 B.\n\n\n=42.= Well, then, we are supposing it held that pleasure is the object\nof all desire, that it is the universal end of all human activity.\nNow I suppose it will not be denied that people are commonly said to\ndesire other things: for instance, we usually talk of desiring food\nand drink, of desiring money, approbation, fame. The question, then,\nmust be of what is meant by desire, and by the object of desire. There\nis obviously asserted some sort of necessary or universal relation\nbetween something which is called desire, and another thing which is\ncalled pleasure. The question is of what sort this relation is; whether\nin conjunction with the naturalistic fallacy above mentioned, it will\njustify Hedonism. Now I am not prepared to deny that there is some\nuniversal relation between pleasure and desire; but I hope to shew,\nthat, if there is, it is of such sort as will rather make against than\nfor Hedonism. It is urged that pleasure is always the object of desire,\nand I am ready to admit that pleasure is always, in part at least, the\n_cause_ of desire. But this distinction is very important. Both views\nmight be expressed in the same language; both might be said to hold\nthat whenever we desire, we always desire _because of_ some pleasure:\nif I asked my supposed Hedonist, \u2018Why do you desire that?\u2019 he might\nanswer, quite consistently with his contention, \u2018Because there is\npleasure there,\u2019 and if he asked me the same question, I might answer,\nequally consistently with my contention, \u2018Because there is pleasure\nhere.\u2019 Only our two answers would not mean the same thing. It is this\nuse of the same language to denote quite different facts, which I\nbelieve to be the chief cause why Psychological Hedonism is so often\nheld, just as it was also the cause of Mill\u2019s naturalistic fallacy.\n\nLet us try to analyse the psychological state which is called \u2018desire.\u2019\nThat name is usually confined to a state of mind in which the idea of\nsome object or event, not yet existing, is present to us. Suppose,\nfor instance, I am desiring a glass of port wine. I have the idea of\ndrinking such a glass before my mind, although I am not yet drinking\nit. Well, how does pleasure enter in to this relation? My theory is\nthat it enters in, in this way. The _idea_ of the drinking causes a\nfeeling of pleasure in my mind, which helps to produce that state\nof incipient activity, which is called \u2018desire.\u2019 It is, therefore,\nbecause of a pleasure, which I already have--the pleasure excited\nby a mere idea--that I desire the wine, which I have not. And I am\nready to admit that a pleasure of this kind, an actual pleasure, is\nalways among the causes of every desire, and not only of every desire,\nbut of every mental activity, whether conscious or sub-conscious. I\nam ready to _admit_ this, I say: I cannot vouch that it is the true\npsychological doctrine; but, at all events, it is not _prim\u00e2 facie_\nquite absurd. And now, what is the other doctrine, the doctrine which\nI am supposing held, and which is at all events essential to Mill\u2019s\nargument? It is this. That when I desire the wine, it is not the wine\nwhich I desire but the pleasure which I expect to get from it. In\nother words, the doctrine is that the idea of a pleasure _not actual_\nis always necessary to cause desire; whereas my doctrine was that the\n_actual_ pleasure caused by the idea of something else was always\nnecessary to cause desire. It is these two different theories which I\nsuppose the Psychological Hedonists to confuse: the confusion is, as\nMr Bradley puts it[12], between \u2018a pleasant thought\u2019 and \u2018the thought\nof a pleasure.\u2019 It is in fact only where the latter, the \u2018thought of\na pleasure,\u2019 is present, that pleasure can be said to be the _object_\nof desire, or the _motive_ to action. On the other hand, when only a\npleasant thought is present, as, I admit, _may_ always be the case,\nthen it is the object of the thought--that which we are thinking\nabout--which is the object of desire and the motive to action; and the\npleasure, which that thought excites, may, indeed, cause our desire or\nmove us to action, but it is not our end or object nor our motive.\n\n  [12] _Ethical Studies_, p. 232.\n\nWell, I hope this distinction is sufficiently clear. Now let us see how\nit bears upon Ethical Hedonism. I assume it to be perfectly obvious\nthat the idea of the object of desire is not always and only the idea\nof a pleasure. In the first place, plainly, we are not always conscious\nof expecting pleasure, when we desire a thing. We may be only conscious\nof the thing which we desire, and may be impelled to make for it at\nonce, without any calculation as to whether it will bring us pleasure\nor pain. And, in the second place, even when we do expect pleasure,\nit can certainly be very rarely pleasure _only_ which we desire. For\ninstance, granted that, when I desire my glass of port wine, I have\nalso an idea of the pleasure I expect from it, plainly that pleasure\ncannot be the only object of my desire; the port wine must be included\nin my object, else I might be led by my desire to take wormwood instead\nof wine. If the desire were directed _solely_ towards the pleasure,\nit could not lead me to take the wine; if it is to take a definite\ndirection, it is absolutely necessary that the idea of the object, from\nwhich the pleasure is expected, should also be present and should\ncontrol my activity. The theory then that what is desired is always and\nonly pleasure must break down: it is impossible to prove that pleasure\nalone is good, by that line of argument. But, if we substitute for this\ntheory, that other, possibly true, theory, that pleasure is always the\ncause of desire, then all the plausibility of our ethical doctrine\nthat pleasure alone is good straightway disappears. For in this case,\npleasure is not what I desire, it is not what I want: it is something\nwhich I already have, before I can want anything. And can any one feel\ninclined to maintain, that that which I already have, while I am still\ndesiring something else, is always and alone the good?\n\n\n=43.= But now let us return to consider another of Mill\u2019s arguments\nfor his position that \u2018happiness is the sole end of human action.\u2019\nMill admits, as I have said, that pleasure is not the only thing we\nactually desire. \u2018The desire of virtue,\u2019 he says, \u2018is not as universal,\nbut it is as authentic a fact, as the desire of happiness[13].\u2019 And\nagain, \u2018Money is, in many cases, desired in and for itself[14].\u2019 These\nadmissions are, of course, in naked and glaring contradiction with\nhis argument that pleasure is the only thing desirable, because it\nis the only thing desired. How then does Mill even attempt to avoid\nthis contradiction? His chief argument seems to be that \u2018virtue,\u2019\n\u2018money\u2019 and other such objects, when they are thus desired in and\nfor themselves, are desired only as \u2018a part of happiness[15].\u2019 Now\nwhat does this mean? Happiness, as we saw, has been defined by Mill,\nas \u2018pleasure and the absence of pain.\u2019 Does Mill mean to say that\n\u2018money,\u2019 these actual coins, which he admits to be desired in and\nfor themselves, are a part either of pleasure or of the absence of\npain? Will he maintain that those coins themselves are in my mind,\nand actually a part of my pleasant feelings? If this is to be said,\nall words are useless: nothing can possibly be distinguished from\nanything else; if these two things are not distinct, what on earth is?\nWe shall hear next that this table is really and truly the same thing\nas this room; that a cab-horse is in fact indistinguishable from St\nPaul\u2019s Cathedral; that this book of Mill\u2019s which I hold in my hand,\nbecause it was his pleasure to produce it, is now and at this moment\na part of the happiness which he felt many years ago and which has\nso long ceased to be. Pray consider a moment what this contemptible\nnonsense really means. \u2018Money,\u2019 says Mill, \u2018is only desirable as a\nmeans to happiness.\u2019 Perhaps so; but what then? \u2018Why,\u2019 says Mill,\n\u2018money is undoubtedly desired for its own sake.\u2019 \u2018Yes, go on,\u2019 say we.\n\u2018Well,\u2019 says Mill, \u2018if money is desired for its own sake, it must be\ndesirable as an end-in-itself: I have said so myself.\u2019 \u2018Oh,\u2019 say we,\n\u2018but you also said just now that it was only desirable as a means.\u2019 \u2018I\nown I did,\u2019 says Mill, \u2018but I will try to patch up matters, by saying\nthat what is only a means to an end, is the same thing as a part of\nthat end. I daresay the public won\u2019t notice.\u2019 And the public haven\u2019t\nnoticed. Yet this is certainly what Mill has done. He has broken down\nthe distinction between means and ends, upon the precise observance of\nwhich his Hedonism rests. And he has been compelled to do this, because\nhe has failed to distinguish \u2018end\u2019 in the sense of what is desirable,\nfrom \u2018end\u2019 in the sense of what is desired: a distinction which,\nnevertheless, both the present argument and his whole book presupposes.\nThis is a consequence of the naturalistic fallacy.\n\n  [13] p. 53.\n\n  [14] p. 55.\n\n  [15] pp. 56-7.\n\n\n=44.= Mill, then, has nothing better to say for himself than this. His\ntwo fundamental propositions are, in his own words, \u2018that to think of\nan object as desirable (unless for the sake of its consequences), and\nto think of it as pleasant, are one and the same thing; and that to\ndesire anything except in proportion as the idea of it is pleasant,\nis a physical and metaphysical impossibility[16].\u2019 Both of these\nstatements are, we have seen, merely supported by fallacies. The first\nseems to rest on the naturalistic fallacy; the second rests partly on\nthis, partly on the fallacy of confusing ends and means, and partly\non the fallacy of confusing a pleasant thought with the thought of a\npleasure. His very language shews this. For that the idea of a thing is\npleasant, in his second clause, is obviously meant to be the same fact\nwhich he denotes by \u2018thinking of it as pleasant,\u2019 in his first.\n\n  [16] p. 58.\n\nAccordingly, Mill\u2019s arguments for the proposition that pleasure is the\nsole good, and our refutation of those arguments, may be summed up as\nfollows:\n\nFirst of all, he takes \u2018the desirable,\u2019 which he uses as a synonym\nfor \u2018the good,\u2019 to _mean_ what _can_ be desired. The test, again, of\nwhat can be desired, is, according to him, what actually is desired:\nif, therefore, he says, we can find some one thing which is always and\nalone desired, that thing will necessarily be the only thing that is\ndesirable, the only thing that is good as an end. In this argument the\nnaturalistic fallacy is plainly involved. That fallacy, I explained,\nconsists in the contention that good _means_ nothing but some simple or\ncomplex notion, that can be defined in terms of natural qualities. In\nMill\u2019s case, good is thus supposed to _mean_ simply what is desired;\nand what is desired is something which can thus be defined in natural\nterms. Mill tells us that we ought to desire something (an ethical\nproposition), because we actually do desire it; but if his contention\nthat \u2018I ought to desire\u2019 means nothing but \u2018I do desire\u2019 were true,\nthen he is only entitled to say, \u2018We do desire so and so, because we\ndo desire it\u2019; and that is not an ethical proposition at all; it is\na mere tautology. The whole object of Mill\u2019s book is to help us to\ndiscover what we ought to do; but, in fact, by attempting to define the\nmeaning of this \u2018ought,\u2019 he has completely debarred himself from ever\nfulfilling that object: he has confined himself to telling us what we\ndo do.\n\nMill\u2019s first argument then is that, because good means desired,\ntherefore the desired is good; but having thus arrived at an ethical\nconclusion, by denying that any ethical conclusion is possible, he\nstill needs another argument to make his conclusion a basis for\nHedonism. He has to prove that we always do desire pleasure or freedom\nfrom pain, and that we never desire anything else whatever. This second\ndoctrine, which Professor Sidgwick has called Psychological Hedonism,\nI accordingly discussed. I pointed out how obviously untrue it is that\nwe never desire anything but pleasure; and how there is not a shadow\nof ground for saying even that, whenever we desire anything, we always\ndesire pleasure _as well as_ that thing. I attributed the obstinate\nbelief in these untruths partly to a confusion between the cause of\ndesire and the object of desire. It may, I said, be true that desire\ncan never occur unless it be preceded by some _actual_ pleasure; but\neven if this is true, it obviously gives no ground for saying that the\nobject of desire is always some _future_ pleasure. By the object of\ndesire is meant that, of which the idea causes desire in us; it is some\npleasure, which we anticipate, some pleasure which we have not got,\nwhich is the object of desire, whenever we do desire pleasure. And any\nactual pleasure, which may be excited by the idea of this anticipated\npleasure, is obviously not the same pleasure as that anticipated\npleasure, of which only the idea is actual. This actual pleasure is not\nwhat we want; what we want is always something which we have not got;\nand to say that pleasure always causes us to want is quite a different\nthing from saying that what we want is always pleasure.\n\nFinally, we saw, Mill admits all this. He insists that we do _actually_\ndesire other things than pleasure, and yet he says we do _really_\ndesire nothing else. He tries to explain away this contradiction,\nby confusing together two notions, which he has before carefully\ndistinguished--the notions of means and of end. He now says that a\nmeans to an end is the same thing as a part of that end. To this last\nfallacy special attention should be given, as our ultimate decision\nwith regard to Hedonism will largely turn upon it.\n\n\n=45.= It is this ultimate decision with regard to Hedonism at which\nwe must now try to arrive. So far I have been only occupied with\nrefuting Mill\u2019s naturalistic arguments for Hedonism; but the doctrine\nthat pleasure alone is desirable may still be true, although Mill\u2019s\nfallacies cannot prove it so. This is the question which we have now\nto face. This proposition, \u2018pleasure alone is good or desirable,\u2019\nbelongs undoubtedly to that class of propositions, to which Mill at\nfirst rightly pretended it belonged, the class of first principles,\nwhich are not amenable to direct proof. But in this case, as he also\nrightly says, \u2018considerations may be presented capable of determining\nthe intellect either to give or withhold its assent to the doctrine\u2019\n(p. 7). It is such considerations that Professor Sidgwick presents,\nand such also that I shall try to present for the opposite view. This\nproposition that \u2018pleasure alone is good as an end,\u2019 the fundamental\nproposition of Ethical Hedonism, will then appear, in Professor\nSidgwick\u2019s language, as an object of intuition. I shall try to shew you\nwhy my intuition denies it, just as his intuition affirms it. It _may_\nalways be true notwithstanding; neither intuition can _prove_ whether\nit is true or not; I am bound to be satisfied, if I can \u2018present\nconsiderations capable of determining the intellect\u2019 to reject it.\n\nNow it may be said that this is a very unsatisfactory state of things.\nIt is indeed; but it is important to make a distinction between two\ndifferent reasons, which may be given for calling it unsatisfactory.\nIs it unsatisfactory because our principle cannot be proved? or is\nit unsatisfactory merely because we do not agree with one another\nabout it? I am inclined to think that the latter is the chief reason.\nFor the mere fact that in certain cases proof is impossible does not\nusually give us the least uneasiness. For instance, nobody can prove\nthat this is a chair beside me; yet I do not suppose that any one is\nmuch dissatisfied for that reason. We all agree that it is a chair,\nand that is enough to content us, although it is quite possible we may\nbe wrong. A madman, of course, might come in and say that it is not a\nchair but an elephant. We could not prove that he was wrong, and the\nfact that he did not agree with us might then begin to make us uneasy.\nMuch more, then, shall we be uneasy, if some one, whom we do not think\nto be mad, disagrees with us. We shall try to argue with him, and we\nshall probably be content if we lead him to agree with us, although we\nshall not have proved our point. We can only persuade him by shewing\nhim that our view is consistent with something else which he holds to\nbe true, whereas his original view is contradictory to it. But it will\nbe impossible to prove that that something else, which we both agree\nto be true, is really so; we shall be satisfied to have settled the\nmatter in dispute by means of it, merely because we are agreed on it.\nIn short, our dissatisfaction in these cases is almost always of the\ntype felt by the poor lunatic in the story. \u2018I said the world was\nmad,\u2019 says he, \u2018and the world said that I was mad; and, confound it,\nthey outvoted me.\u2019 It is, I say, almost always such a disagreement, and\nnot the impossibility of proof, which makes us call the state of things\nunsatisfactory. For, indeed, who can prove that proof itself is a\nwarrant of truth? We are all agreed that the laws of logic are true and\ntherefore we accept a result which is proved by their means; but such\na proof is satisfactory to us only because we are all so fully agreed\nthat it is a warrant of truth. And yet we cannot, by the nature of the\ncase, prove that we are right in being so agreed.\n\nAccordingly, I do not think we need be much distressed by our admission\nthat we cannot prove whether pleasure alone is good or not. We may be\nable to arrive at an agreement notwithstanding; and if so, I think\nit will be satisfactory. And yet I am not very sanguine about our\nprospects of such satisfaction. Ethics, and philosophy in general,\nhave always been in a peculiarly unsatisfactory state. There has been\nno agreement about them, as there is about the existence of chairs and\nlights and benches. I should therefore be a fool if I hoped to settle\none great point of controversy, now and once for all. It is extremely\nimprobable I shall convince. It would be highly presumptuous even to\nhope that in the end, say two or three centuries hence, it will be\nagreed that pleasure is not the sole good. Philosophical questions\nare so difficult, the problems they raise are so complex, that no one\ncan fairly expect, now, any more than in the past, to win more than a\nvery limited assent. And yet I confess that the considerations which\nI am about to present appear to me to be absolutely convincing. I do\nthink that they _ought_ to convince, if only I can put them well.\nIn any case, I can but try. I _shall_ try now to put an end to that\nunsatisfactory state of things, of which I have been speaking. I shall\ntry to produce an agreement that the fundamental principle of Hedonism\nis very like an absurdity, by shewing what it must mean, if it is\nclearly thought out, and how that clear meaning is in conflict with\nother beliefs, which will, I hope, not be so easily given up.\n\n\n=46.= Well, then, we now proceed to discuss Intuitionistic Hedonism.\nAnd the beginning of this discussion marks, it is to be observed, a\nturning-point in my ethical method. The point I have been labouring\nhitherto, the point that \u2018good is indefinable,\u2019 and that to deny this\ninvolves a fallacy, is a point capable of strict proof: for to deny it\ninvolves contradictions. But now we are coming to the question, for\nthe sake of answering which Ethics exists, the question what things or\nqualities are good. Of any answer to _this_ question no direct proof\nis possible, and that, just because of our former answer, as to the\nmeaning of good, direct proof _was_ possible. We are now confined to\nthe hope of what Mill calls \u2018indirect proof,\u2019 the hope of determining\none another\u2019s intellect; and we are now so confined, just because, in\nthe matter of the former question we are not so confined. Here, then,\nis an intuition to be submitted to our verdict--the intuition that\n\u2018pleasure alone is good as an end--good in and for itself.\u2019\n\n\n=47.= Well, in this connection it seems first desirable to touch on\nanother doctrine of Mill\u2019s--another doctrine which, in the interest\nof Hedonism, Professor Sidgwick has done very wisely to reject.\nThis is the doctrine of \u2018difference of quality in pleasures.\u2019 \u2018If I\nam asked,\u2019 says Mill[17], \u2018what I mean by difference of quality in\npleasures, or what makes one pleasure more valuable than another,\nmerely as a pleasure, except its being greater in amount, there is but\none possible answer. Of two pleasures, if there be one to which all\nor almost all who have experience of both give a decided preference,\nirrespective of any feeling of moral obligation to prefer it, that is\nthe more desirable pleasure. If one of the two is, by those who are\ncompetently acquainted with both, placed so far above the other that\nthey prefer it, even though knowing it to be attended with a greater\namount of discontent, and would not resign it for any quantity of the\nother pleasure which their nature is capable of, we are justified in\nascribing to the preferred enjoyment a superiority in quality, so far\noutweighing quantity as to render it, in comparison, of small account.\u2019\n\n  [17] p. 12.\n\nNow it is well known that Bentham rested his case for Hedonism on\n\u2018quantity of pleasure\u2019 alone. It was his maxim, that \u2018quantity\nof pleasure being equal, pushpin is as good as poetry.\u2019 And Mill\napparently considers Bentham to have proved that nevertheless poetry\nis better than pushpin; that poetry does produce a greater quantity of\npleasure. But yet, says Mill, the Utilitarians \u2018might have taken the\nother and, as it may be called, higher ground, with entire consistency\u2019\n(p. 11). Now we see from this that Mill acknowledges \u2018quality of\npleasure\u2019 to be another or different ground for estimating pleasures,\nthan Bentham\u2019s quantity; and moreover, by that question-begging\n\u2018higher,\u2019 which he afterwards translates into \u2018superior,\u2019 he seems to\nbetray an uncomfortable feeling, that, after all, if you take quantity\nof pleasure for your only standard, something may be wrong and you\nmay deserve to be called a pig. And it may presently appear that you\nvery likely would deserve that name. But, meanwhile, I only wish to\nshew that Mill\u2019s admissions as to quality of pleasure are either\ninconsistent with his Hedonism, or else afford no other ground for it\nthan would be given by mere quantity of pleasure.\n\nIt will be seen that Mill\u2019s test for one pleasure\u2019s superiority\nin quality over another is the preference of most people who have\nexperienced both. A pleasure so preferred, he holds, is more desirable.\nBut then, as we have seen, he holds that \u2018to think of an object as\ndesirable and to think of it as pleasant are one and the same thing\u2019\n(p. 58). He holds, therefore, that the preference of experts merely\nproves that one pleasure is pleasanter than another. But if that is\nso, how can he distinguish this standard from the standard of quantity\nof pleasure? Can one pleasure be pleasanter than another, except in\nthe sense that it gives _more_ pleasure? \u2018Pleasant\u2019 must, if words are\nto have any meaning at all, denote some one quality common to all the\nthings that are pleasant; and, if so, then one thing can only be more\npleasant than another, according as it has more or less of this one\nquality. But, then, let us try the other alternative, and suppose that\nMill does not seriously mean that this preference of experts merely\nproves one pleasure to be pleasanter than another. Well, in this case\nwhat does \u2018preferred\u2019 mean? It cannot mean \u2018more desired,\u2019 since, as\nwe know, the degree of desire is always, according to Mill, in exact\nproportion to the degree of pleasantness. But, in that case, the basis\nof Mill\u2019s Hedonism collapses, for he is admitting that one thing may\nbe preferred over another, and thus proved more desirable, although\nit is not more desired. In this case Mill\u2019s judgment of preference is\njust a judgment of that intuitional kind which I have been contending\nto be necessary to establish the hedonistic or any other principle. It\nis a direct judgment that one thing is more desirable, or better than\nanother; a judgment utterly independent of all considerations as to\nwhether one thing is more desired or pleasanter than another. This is\nto admit that good is good and indefinable.\n\n\n=48.= And note another point that is brought out by this discussion.\nMill\u2019s judgment of preference, so far from establishing the principle\nthat pleasure alone is good, is obviously inconsistent with it. He\nadmits that experts can judge whether one pleasure is more desirable\nthan another, because pleasures differ in quality. But what does this\nmean? If one pleasure can differ from another in quality, that means,\nthat _a_ pleasure is something complex, something composed, in fact, of\npleasure _in addition to_ that which produces pleasure. For instance,\nMill speaks of \u2018sensual indulgences\u2019 as \u2018lower pleasures.\u2019 But what\nis a sensual indulgence? It is surely a certain excitement of some\nsense _together with_ the pleasure caused by such excitement. Mill,\ntherefore, in admitting that a sensual indulgence can be directly\njudged to be lower than another pleasure, in which the degree of\npleasure involved may be the same, is admitting that other things may\nbe good, or bad, quite independently of the pleasure which accompanies\nthem. _A_ pleasure is, in fact, merely a misleading term which\nconceals the fact that what we are dealing with is not pleasure but\nsomething else, which may indeed necessarily produce pleasure, but is\nnevertheless quite distinct from it.\n\nMill, therefore, in thinking that to estimate quality of pleasure\nis quite consistent with his hedonistic principle that pleasure and\nabsence of pain alone are desirable as ends, has again committed the\nfallacy of confusing ends and means. For take even the most favourable\nsupposition of his meaning; let us suppose that by a pleasure he does\nnot mean, as his words imply, that which produces pleasure and the\npleasure produced. Let us suppose him to mean that there are various\nkinds of pleasure, in the sense in which there are various kinds of\ncolour--blue, red, green, etc. Even in this case, if we are to say that\nour end is colour alone, then, although it is impossible we should\nhave colour without having some particular colour, yet the particular\ncolour we must have, is only a _means_ to our having colour, if colour\nis really our end. And if colour is our only possible end, as Mill says\npleasure is, then there can be no possible reason for preferring one\ncolour to another, red, for instance, to blue, except that the one is\nmore of a colour than the other. Yet the opposite of this is what Mill\nis attempting to hold with regard to pleasures.\n\nAccordingly a consideration of Mill\u2019s view that some pleasures are\nsuperior to others _in quality_ brings out one point which may \u2018help\nto determine the intellect\u2019 with regard to the intuition \u2018Pleasure is\nthe only good.\u2019 For it brings out the fact that if you say \u2018pleasure,\u2019\nyou must mean \u2018pleasure\u2019: you must mean some one thing common to all\ndifferent \u2018pleasures,\u2019 some one thing, which may exist in different\ndegrees, but which cannot differ in _kind_. I have pointed out that,\nif you say, as Mill does, that quality of pleasure is to be taken\ninto account, then you are no longer holding that pleasure _alone_ is\ngood as an end, since you imply that something else, something which\nis _not_ present in all pleasures, is _also_ good as an end. The\nillustration I have given from colour expresses this point in its most\nacute form. It is plain that if you say \u2018Colour alone is good as an\nend,\u2019 then you can give no possible reason for preferring one colour\nto another. Your only standard of good and bad will then be \u2018colour\u2019;\nand since red and blue both conform equally to this, the only standard,\nyou can have no other whereby to judge whether red is better than blue.\nIt is true that you cannot have colour unless you also have one or\nall of the particular colours: they, therefore, if colour is the end,\nwill all be good as means, but none of them can be better than another\neven as a means, far less can any one of them be regarded as an end in\nitself. Just so with pleasure: If we do really mean \u2018Pleasure alone\nis good as an end,\u2019 then we must agree with Bentham that \u2018Quantity\nof pleasure being equal, pushpin is as good as poetry.\u2019 To have thus\ndismissed Mill\u2019s reference to quality of pleasure, is therefore to have\nmade one step in the desired direction. The reader will now no longer\nbe prevented from agreeing with me, by any idea that the hedonistic\nprinciple \u2018Pleasure alone is good as an end\u2019 is consistent with the\nview that one pleasure may be of a better quality than another. These\ntwo views, we have seen, are contradictory to one another. We must\nchoose between them: and if we choose the latter, then we must give up\nthe principle of Hedonism.\n\n\n=49.= But, as I said, Professor Sidgwick has seen that they are\ninconsistent. He has seen that he must choose between them. He has\nchosen. He has rejected the test by quality of pleasure, and has\naccepted the hedonistic principle. He still maintains that \u2018Pleasure\nalone is good as an end.\u2019 I propose therefore to discuss the\nconsiderations which he has offered in order to convince us. I shall\nhope by that discussion to remove some more of such prejudices and\nmisunderstandings as might prevent agreement with me. If I can shew\nthat some of the considerations which Professor Sidgwick urges are such\nas we need by no means agree with, and that others are actually rather\nin my favour than in his, we may have again advanced a few steps nearer\nto the unanimity which we desire.\n\n\n=50.= The passages in the _Methods of Ethics_ to which I shall now\ninvite attention are to be found in I. IX. 4 and in III. XIV. 4-5.\n\nThe first of these two passages runs as follows:\n\n\u201cI think that if we consider carefully such permanent results as are\ncommonly judged to be good, other than qualities of human beings, we\ncan find nothing that, on reflection, appears to possess this quality\nof goodness out of relation to human existence, or at least to some\nconsciousness or feeling.\n\n\u201cFor example, we commonly judge some inanimate objects, scenes, etc. to\nbe good as possessing beauty, and others bad from ugliness: still no\none would consider it rational to aim at the production of beauty in\nexternal nature, apart from any possible contemplation of it by human\nbeings. In fact when beauty is maintained to be objective, it is not\ncommonly meant that it exists as beauty out of relation to any mind\nwhatsoever: but only that there is some standard of beauty valid for\nall minds.\n\n\u201cIt may, however, be said that beauty and other results commonly judged\nto be good, though we do not conceive them to exist out of relation to\nhuman beings (or at least minds of some kind), are yet so far separable\nas ends from the human beings on whom their existence depends, that\ntheir realization may conceivably come into competition with the\nperfection or happiness of these beings. Thus, though beautiful\nthings cannot be thought worth producing except as possible objects\nof contemplation, still a man may devote himself to their production\nwithout any consideration of the persons who are to contemplate them.\nSimilarly knowledge is a good which cannot exist except in minds; and\nyet one may be more interested in the development of knowledge than in\nits possession by any particular minds; and may take the former as an\nultimate end without regarding the latter.\n\n\u201cStill, as soon as the alternatives are clearly apprehended, it will, I\nthink, be generally held that beauty, knowledge, and other ideal goods,\nas well as all external material things, are only reasonably to be\nsought by men in so far as they conduce (1) to Happiness or (2) to the\nPerfection or Excellence of human existence. I say \u2018human,\u2019 for though\nmost utilitarians consider the pleasure (and freedom from pain) of the\ninferior animals to be included in the Happiness which they take as the\nright and proper end of conduct, no one seems to contend that we ought\nto aim at perfecting brutes except as a means to our ends, or at least\nas objects of scientific or \u00e6sthetic contemplation for us. Nor, again,\ncan we include, as a practical end, the existence of beings above the\nhuman. We certainly apply the idea of Good to the Divine Existence,\njust as we do to His work, and indeed in a preeminent manner: and\nwhen it is said that, \u2018we should do all things to the glory of God,\u2019\nit may seem to be implied that the existence of God is made better by\nour glorifying Him. Still this inference when explicitly drawn appears\nsomewhat impious; and theologians generally recoil from it, and\nrefrain from using the notion of a possible addition to the Goodness of\nthe Divine Existence as a ground of human duty. Nor can the influence\nof our actions on other extra-human intelligences besides the Divine be\nat present made matter of scientific discussion.\n\n\u201cI shall therefore confidently lay down, that if there be any Good\nother than Happiness to be sought by man, as an ultimate practical\nend, it can only be the Goodness, Perfection, or Excellence of Human\nExistence. How far this notion includes more than Virtue, what its\nprecise relation to Pleasure is, and to what method we shall be\nlogically led if we accept it as fundamental, are questions which we\nshall more conveniently discuss after the detailed examination of these\ntwo other notions, Pleasure and Virtue, in which we shall be engaged in\nthe two following Books.\u201d\n\nIt will be observed that in this passage Prof. Sidgwick tries to limit\nthe range of objects among which the ultimate end may be found. He does\nnot yet say what that end is, but he does exclude from it everything\nbut certain characters of Human Existence. And the possible ends, which\nhe thus excludes, do not again come up for consideration. They are put\nout of court once for all by this passage and by this passage only. Now\nis this exclusion justified?\n\nI cannot think it is. \u2018No one,\u2019 says Prof. Sidgwick, \u2018would consider\nit rational to aim at the production of beauty in external nature,\napart from any possible contemplation of it by human beings.\u2019 Well, I\nmay say at once, that I, for one, do consider this rational; and let\nus see if I cannot get any one to agree with me. Consider what this\nadmission really means. It entitles us to put the following case. Let\nus imagine one world exceedingly beautiful. Imagine it as beautiful as\nyou can; put into it whatever on this earth you most admire--mountains,\nrivers, the sea; trees, and sunsets, stars and moon. Imagine these all\ncombined in the most exquisite proportions, so that no one thing jars\nagainst another, but each contributes to increase the beauty of the\nwhole. And then imagine the ugliest world you can possibly conceive.\nImagine it simply one heap of filth, containing everything that is\nmost disgusting to us, for whatever reason, and the whole, as far\nas may be, without one redeeming feature. Such a pair of worlds we\nare entitled to compare: they fall within Prof. Sidgwick\u2019s meaning,\nand the comparison is highly relevant to it. The only thing we are\nnot entitled to imagine is that any human being ever has or ever, by\nany possibility, _can_, live in either, can ever see and enjoy the\nbeauty of the one or hate the foulness of the other. Well, even so,\nsupposing them quite apart from any possible contemplation by human\nbeings; still, is it irrational to hold that it is better that the\nbeautiful world should exist, than the one which is ugly? Would it not\nbe well, in any case, to do what we could to produce it rather than\nthe other? Certainly I cannot help thinking that it would; and I hope\nthat some may agree with me in this extreme instance. The instance is\nextreme. It is highly improbable, not to say, impossible, we should\never have such a choice before us. In any actual choice we should have\nto consider the possible effects of our action upon conscious beings,\nand among these possible effects there are always some, I think, which\nought to be preferred to the existence of mere beauty. But this only\nmeans that in our present state, in which but a very small portion of\nthe good is attainable, the pursuit of beauty for its own sake must\nalways be postponed to the pursuit of some greater good, which is\nequally attainable. But it is enough for my purpose, if it be admitted\nthat, _supposing_ no greater good were at all attainable, then beauty\nmust in itself be regarded as a greater good than ugliness; if it be\nadmitted that, in that case, we should not be left without any reason\nfor preferring one course of action to another, we should not be left\nwithout any duty whatever, but that it would then be our positive duty\nto make the world more beautiful, so far as we were able, since nothing\nbetter than beauty could then result from our efforts. If this be once\nadmitted, if in any imaginable case you do admit that the existence\nof a more beautiful thing is better in itself than that of one more\nugly, quite apart from its effects on any human feeling, then Prof.\nSidgwick\u2019s principle has broken down. Then we shall have to include\nin our ultimate end something beyond the limits of human existence. I\nadmit, of course, that our beautiful world would be better still, if\nthere were human beings in it to contemplate and enjoy its beauty. But\nthat admission makes nothing against my point. If it be once admitted\nthat the beautiful world _in itself_ is better than the ugly, then it\nfollows, that however many beings may enjoy it, and however much better\ntheir enjoyment may be than it is itself, yet its mere existence adds\n_something_ to the goodness of the whole: it is not only a means to our\nend, but also itself a part thereof.\n\n\n=51.= In the second passage to which I referred above, Prof. Sidgwick\nreturns from the discussion of Virtue and Pleasure, with which he has\nmeanwhile been engaged, to consider what among the parts of Human\nExistence to which, as we saw, he has limited the ultimate end, can\nreally be considered as such end. What I have just said, of course,\nappears to me to destroy the force of this part of his argument too.\nIf, as I think, other things than any part of Human Existence can be\nends-in-themselves, then Prof. Sidgwick cannot claim to have discovered\nthe Summum Bonum, when he has merely determined what parts of Human\nExistence are in themselves desirable. But this error may be admitted\nto be utterly insignificant in comparison with that which we are now\nabout to discuss.\n\n\u201cIt may be said,\u201d says Prof. Sidgwick (III. XIV. \u00a7\u00a7 4-5), \u201cthat we\nmay ... regard cognition of Truth, contemplation of Beauty, Free or\nVirtuous action, as in some measure preferable alternatives to Pleasure\nor Happiness--even though we admit that Happiness must be included as\na part of Ultimate Good.... I think, however, that this view ought not\nto commend itself to the sober judgment of reflective persons. In order\nto shew this, I must ask the reader to use the same twofold procedure\nthat I before requested him to employ in considering the absolute and\nindependent validity of common moral precepts. I appeal firstly to\nhis intuitive judgment after due consideration of the question when\nfairly placed before it: and secondly to a comprehensive comparison\nof the ordinary judgments of mankind. As regards the first argument,\nto me at least it seems clear after reflection that these objective\nrelations of the conscious subject, when distinguished from the\nconsciousness accompanying and resulting from them, are not ultimately\nand intrinsically desirable; any more than material or other objects\nare, when considered apart from any relation to conscious existence.\nAdmitting that we have actual experience of such preferences as have\njust been described, of which the ultimate object is something that\nis not merely consciousness: it still seems to me that when (to use\nButler\u2019s phrase) we \u2018sit down in a cool hour,\u2019 we can only justify to\nourselves the importance that we attach to any of these objects by\nconsidering its conduciveness, in one way or another, to the happiness\nof sentient beings.\n\n\u201cThe second argument, that refers to the common sense of mankind,\nobviously cannot be made completely cogent; since, as above stated,\nseveral cultivated persons do habitually judge that knowledge, art,\netc.,--not to speak of Virtue--are ends independently of the pleasure\nderived from them. But we may urge not only that all these elements\nof \u2018ideal good\u2019 are productive of pleasure in various ways; but also\nthat they seem to obtain the commendation of Common Sense, roughly\nspeaking, in proportion to the degree of this productiveness. This\nseems obviously true of Beauty; and will hardly be denied in respect\nof any kind of social ideal: it is paradoxical to maintain that any\ndegree of Freedom, or any form of social order, would still be commonly\nregarded as desirable even if we were certain that it had no tendency\nto promote the general happiness. The case of Knowledge is rather more\ncomplex; but certainly Common Sense is most impressed with the value\nof knowledge, when its \u2018fruitfulness\u2019 has been demonstrated. It is,\nhowever, aware that experience has frequently shewn how knowledge, long\nfruitless, may become unexpectedly fruitful, and how light may be shed\non one part of the field of knowledge from another apparently remote:\nand even if any particular branch of scientific pursuit could be shewn\nto be devoid of even this indirect utility, it would still deserve some\nrespect on utilitarian grounds; both as furnishing to the inquirer\nthe refined and innocent pleasures of curiosity, and because the\nintellectual disposition which it exhibits and sustains is likely on\nthe whole to produce fruitful knowledge. Still in cases approximating\nto this last, Common Sense is somewhat disposed to complain of the\nmis-direction of valuable effort; so that the meed of honour commonly\npaid to Science seems to be graduated, though perhaps unconsciously,\nby a tolerably exact utilitarian scale. Certainly the moment the\nlegitimacy of any branch of scientific inquiry is seriously disputed,\nas in the recent case of vivisection, the controversy on both sides is\ngenerally conducted on an avowedly utilitarian basis.\n\n\u201cThe case of Virtue requires special consideration: since the\nencouragement in each other of virtuous impulses and dispositions is a\nmain aim of men\u2019s ordinary moral discourse; so that even to raise the\nquestion whether this encouragement can go too far has a paradoxical\nair. Still, our experience includes rare and exceptional cases in which\nthe concentration of effort on the cultivation of virtue has seemed to\nhave effects adverse to general happiness, through being intensified\nto the point of moral fanaticism, and so involving a neglect of other\nconditions of happiness. If, then, we admit as actual or possible\nsuch \u2018infelicific\u2019 effects of the cultivation of Virtue, I think we\nshall also generally admit that, in the case supposed, conduciveness\nto general happiness should be the criterion for deciding how far the\ncultivation of Virtue should be carried.\u201d\n\nThere we have Prof. Sidgwick\u2019s argument completed. We ought not, he\nthinks, to aim at knowing the Truth, or at contemplating Beauty,\nexcept in so far as such knowledge or such contemplation contributes\nto increase the pleasure or to diminish the pain of sentient beings.\nPleasure alone is good for its own sake: knowledge of the Truth is good\nonly as a means to pleasure.\n\n\n=52.= Let us consider what this means. What is pleasure? It is\ncertainly something of which we may be conscious, and which, therefore,\nmay be distinguished from our consciousness of it. What I wish first to\nask is this: Can it really be said that we value pleasure, except in\nso far as we are conscious of it? Should we think that the attainment\nof pleasure, of which we never were and never could be conscious, was\nsomething to be aimed at for its own sake? It may be impossible that\nsuch pleasure should ever exist, that it should ever be thus divorced\nfrom consciousness; although there is certainly much reason to believe\nthat it is not only possible but very common. But, even supposing\nthat it were impossible, that is quite irrelevant. Our question is:\nIs it the pleasure, as distinct from the consciousness of it, that we\nset value on? Do we think the pleasure valuable in itself, or must\nwe insist that, if we are to think the pleasure good, we must have\nconsciousness of it too?\n\nThis consideration is very well put by Socrates in Plato\u2019s dialogue\n_Philebus_ (21 A).\n\n\u2018Would _you_ accept, Protarchus,\u2019 says Socrates, \u2018to live your whole\nlife in the enjoyment of the greatest pleasures?\u2019 \u2018Of course I would,\u2019\nsays Protarchus.\n\n_Socrates._ Then would you think you needed anything else besides, if\nyou possessed this one blessing in completeness?\n\n_Protarchus._ Certainly not.\n\n_Socrates._ Consider what you are saying. You would not need to be wise\nand intelligent and reasonable, nor anything like this? Would you not\neven care to keep your sight?\n\n_Protarchus._ Why should I? I suppose I should have all I want, if I\nwas pleased.\n\n_Socrates._ Well, then, supposing you lived so, you would enjoy always\nthroughout your life the greatest pleasure?\n\n_Protarchus._ Of course.\n\n_Socrates._ But, on the other hand, inasmuch as you would _not_ possess\nintelligence and memory and knowledge and true opinion, you would, in\nthe first place, necessarily be without the knowledge whether you were\npleased or not. For you would be devoid of any kind of wisdom. You\nadmit this?\n\n_Protarchus._ I do. The consequence is absolutely necessary.\n\n_Socrates._ Well, then, besides this, not having memory, you must\nalso be unable to remember even that you ever were pleased; of the\npleasure which falls upon you at the moment not the least vestige must\nafterwards remain. And again, not having true opinion, you cannot think\nthat you are pleased when you are; and, being bereft of your reasoning\nfaculties, you cannot even have the power to reckon that you will be\npleased in future. You must live the life of an oyster, or of some\nother of those living creatures, whose home is the seas and whose\nsouls are concealed in shelly bodies. Is all this so, or can we think\notherwise than this?\n\n_Protarchus._ How can we?\n\n_Socrates._ Well, then, can we think such a life desirable?\n\n_Protarchus._ Socrates, your reasoning has left me utterly dumb.\u2019\n\nSocrates, we see, persuades Protarchus that Hedonism is absurd. If\nwe are really going to maintain that pleasure alone is good as an\nend, we must maintain that it is good, whether we are conscious of\nit or not. We must declare it reasonable to take as our ideal (an\nunattainable ideal it may be) that we should be as happy as possible,\neven on condition that we never know and never can know that we are\nhappy. We must be willing to sell in exchange for the mere happiness\nevery vestige of knowledge, both in ourselves and in others, both\nof happiness itself and of every other thing. Can we really still\ndisagree? Can any one still declare it obvious that this is reasonable?\nThat pleasure alone is good as an end?\n\nThe case, it is plain, is just like that of the colours[18], only,\nas yet, not nearly so strong. It is far more possible that we should\nsome day be able to produce the intensest pleasure, without any\nconsciousness that it is there, than that we should be able to produce\nmere colour, without its being any particular colour. Pleasure and\nconsciousness can be far more easily distinguished from one another,\nthan colour from the particular colours. And yet even if this were\nnot so, we should be bound to distinguish them if we really wished to\ndeclare pleasure alone to be our ultimate end. Even if consciousness\nwere an inseparable accompaniment of pleasure, a _sine qu\u00e2 non_ of\nits existence, yet, if pleasure is the only end, we are bound to call\nconsciousness a mere _means_ to it, in any intelligible sense that can\nbe given to the word _means_. And if, on the other hand, as I hope is\nnow plain, the pleasure would be comparatively valueless without the\nconsciousness, then we are bound to say that pleasure is _not_ the only\nend, that some consciousness at least must be included with it as a\nveritable part of the end.\n\n  [18] \u00a7 48 _sup._\n\nFor our question now is solely what the end is: it is quite another\nquestion how far that end may be attainable _by itself_, or must\ninvolve the simultaneous attainment of other things. It may well be\nthat the _practical_ conclusions at which Utilitarians do arrive,\nand even those at which they ought logically to arrive, are not far\nfrom the truth. But in so far as their _reason_ for holding these\nconclusions to be true is that \u2018Pleasure alone is good as an end,\u2019 they\nare _absolutely_ wrong: and it is with _reasons_ that we are chiefly\nconcerned in any scientific Ethics.\n\n\n=53.= It seems, then, clear that Hedonism is in error, so far as it\nmaintains that pleasure alone, and not the consciousness of pleasure,\nis the sole good. And this error seems largely due to the fallacy which\nI pointed out above in Mill--the fallacy of confusing means and end.\nIt is falsely supposed that, since pleasure must always be accompanied\nby consciousness (which is, itself, extremely doubtful), therefore\nit is indifferent whether we say that pleasure or the consciousness\nof pleasure is the sole good. _Practically_, of course, it would be\nindifferent at which we aimed, if it were certain that we could not\nget the one without the other; but where the question is of what is\ngood in itself--where we ask: For the sake of what is it desirable to\nget that which we aim at?--the distinction is by no means unimportant.\nHere we are placed before an exclusive alternative. _Either_ pleasure\nby itself (even though we can\u2019t get it) would be all that is desirable,\n_or_ a consciousness of it would be more desirable still. Both these\npropositions cannot be true; and I think it is plain that the latter is\ntrue; whence it follows that pleasure is _not_ the sole good.\n\nStill it may be said that, even if consciousness of pleasure, and\nnot pleasure alone, is the sole good, this conclusion is not very\ndamaging to Hedonism. It may be said that Hedonists have always meant\nby pleasure the consciousness of pleasure, though they have not been at\npains to say so; and this, I think is, in the main, true. To correct\ntheir formula in this respect could, therefore, only be a matter of\npractical importance, if it is possible to produce pleasure without\nproducing consciousness of it. But even this importance, which I think\nour conclusion so far really has, is, I admit, comparatively slight.\nWhat I wish to maintain is that even consciousness of pleasure is\nnot the sole good: that, indeed, it is absurd so to regard it. And\nthe chief importance of what has been said so far lies in the fact\nthat the same method, which shews that consciousness of pleasure is\nmore valuable than pleasure, seems also to shew that consciousness of\npleasure is itself far less valuable than other things. The supposition\nthat consciousness of pleasure is the sole good is due to a neglect of\nthe same distinctions which have encouraged the careless assertion that\npleasure is the sole good.\n\nThe method which I employed in order to shew that pleasure itself\nwas not the sole good, was that of considering what value we should\nattach to it, if it existed in absolute isolation, stripped of all its\nusual accompaniments. And this is, in fact, the only method that can\nbe safely used, when we wish to discover what degree of value a thing\nhas in itself. The necessity of employing this method will be best\nexhibited by a discussion of the arguments used by Prof. Sidgwick in\nthe passage last quoted, and by an exposure of the manner in which they\nare calculated to mislead.\n\n\n=54.= With regard to the second of them, it only maintains that other\nthings, which might be supposed to share with pleasure the attribute\nof goodness, \u2018seem to obtain the commendation of Common Sense, roughly\nspeaking, in proportion to the degree\u2019 of their productiveness of\npleasure. Whether even this rough proportion holds between the\ncommendation of Common Sense and the felicific effects of that which it\ncommends is a question extremely difficult to determine; and we need\nnot enter into it here. For, even assuming it to be true, and assuming\nthe judgments of Common Sense to be on the whole correct, what would it\nshew? It would shew, certainly, that pleasure was a good _criterion_\nof right action--that the same conduct which produced most pleasure\nwould also produce most good on the whole. But this would by no means\nentitle us to the conclusion that the greatest pleasure _constituted_\nwhat was best on the whole: it would still leave open the alternative\nthat the greatest quantity of pleasure was as a matter of fact, _under\nactual conditions_, generally accompanied by the greatest quantity\nof _other goods_, and that it therefore was _not_ the sole good. It\nmight indeed seem to be a strange coincidence that these two things\nshould always, even in this world, be in proportion to one another.\nBut the strangeness of this coincidence will certainly not entitle us\nto argue directly that it does not exist--that it is an illusion, due\nto the fact that pleasure is really the sole good. The coincidence may\nbe susceptible of other explanations; and it would even be our duty\nto accept it unexplained, if direct intuition seemed to declare that\npleasure was not the sole good. Moreover it must be remembered that\nthe need for assuming such a coincidence rests in any case upon the\nextremely doubtful proposition that felicific effects _are_ roughly in\nproportion to the approval of Common Sense. And it should be observed\nthat, though Prof. Sidgwick maintains this to be the case, his detailed\nillustrations only tend to shew the very different proposition that a\nthing is not held to be good, unless it gives a balance of pleasure;\nnot that the degree of commendation is in proportion to the quantity of\npleasure.\n\n\n=55.= The decision, then, must rest upon Prof. Sidgwick\u2019s first\nargument--\u2018the appeal\u2019 to our \u2018intuitive judgment after due\nconsideration of the question when fairly placed before it.\u2019 And here\nit seems to me plain that Prof. Sidgwick has failed, in two essential\nrespects, to place the question fairly before either himself or his\nreader.\n\n(1) What he has to shew is, as he says himself, not merely that\n\u2018Happiness must be included as a part of Ultimate Good.\u2019 This view,\nhe says, \u2018ought not to commend itself to the sober judgment of\nreflective persons.\u2019 And why? Because \u2018these objective relations,\nwhen distinguished from the consciousness accompanying and resulting\nfrom them, are not ultimately and intrinsically desirable.\u2019 Now, this\nreason, which is offered as shewing that to consider Happiness as\na mere part of Ultimate Good does not meet the facts of intuition,\nis, on the contrary, only sufficient to shew that it _is_ a part of\nUltimate Good. For from the fact that no value resides in one part\nof a whole, considered by itself, we cannot infer that all the value\nbelonging to the whole does reside in the other part, considered by\nitself. Even if we admit that there is much value in the enjoyment of\nBeauty, and none in the mere contemplation of it, which is one of the\nconstituents of that complex fact, it does not follow that all the\nvalue belongs to the other constituent, namely, the pleasure which we\ntake in contemplating it. It is quite possible that this constituent\nalso has no value in itself; that the value belongs to the whole state,\nand to that only: so that _both_ the pleasure _and_ the contemplation\nare mere parts of the good, and both of them equally necessary parts.\nIn short, Prof. Sidgwick\u2019s argument here depends upon the neglect of\nthat principle, which I tried to explain in my first chapter and which\nI said I should call the principle of \u2018organic relations[19].\u2019 The\nargument is calculated to mislead, because it supposes that, if we see\na whole state to be valuable, and also see that one element of that\nstate has no value _by itself_, then the other element, _by itself_,\nmust have all the value which belongs to the whole state. The fact\nis, on the contrary, that, since the whole may be organic, the other\nelement need have no value whatever, and that even if it have some, the\nvalue of the whole may be very much greater. For this reason, as well\nas to avoid confusion between means and end, it is absolutely essential\nto consider each distinguishable quality, _in isolation_, in order to\ndecide what value it possesses. Prof. Sidgwick, on the other hand,\napplies this method of isolation only to _one_ element in the wholes\nhe is considering. He does not ask the question: If consciousness\nof pleasure existed absolutely by itself, would a sober judgment be\nable to attribute much value to it? It is, in fact, always misleading\nto take a whole, that is valuable (or the reverse), and then to ask\nsimply: To which of its constituents does this whole owe its value or\nits vileness? It may well be that it owes it to _none_; and, if one of\nthem does appear to have some value in itself, we shall be led into the\ngrave error of supposing that all the value of the whole belongs to it\nalone. It seems to me that this error has commonly been committed with\nregard to pleasure. Pleasure does seem to be a necessary constituent\nof most valuable wholes; and, since the other constituents, into which\nwe may analyse them, may easily seem not to have any value, it is\nnatural to suppose that all the value belongs to pleasure. That this\nnatural supposition does not follow from the premises is certain; and\nthat it is, on the contrary, ridiculously far from the truth appears\nevident to my \u2018reflective judgment.\u2019 If we apply either to pleasure or\nto consciousness of pleasure the only safe method, that of isolation,\nand ask ourselves: Could we accept, as a very good thing, that mere\nconsciousness of pleasure, and absolutely nothing else, should exist,\neven in the greatest quantities? I think we can have no doubt about\nanswering: No. Far less can we accept this as the _sole_ good. Even\nif we accept Prof. Sidgwick\u2019s implication (which yet appears to me\nextremely doubtful) that consciousness of pleasure has a greater\nvalue by itself than Contemplation of Beauty, it seems to me that a\npleasurable Contemplation of Beauty has certainly an immeasurably\ngreater value than mere Consciousness of Pleasure. In favour of this\nconclusion I can appeal with confidence to the \u2018sober judgment of\nreflective persons.\u2019\n\n  [19] pp. 27-30, 36.\n\n\n=56.= (2) That the value of a pleasurable whole does not belong solely\nto the pleasure which it contains, may, I think, be made still plainer\nby consideration of another point in which Prof. Sidgwick\u2019s argument\nis defective. Prof. Sidgwick maintains, as we saw, the doubtful\nproposition, that the _conduciveness_ to pleasure of a thing is in\nrough proportion to its commendation by Common Sense. But he does\nnot maintain, what would be undoubtedly false, that the pleasantness\nof every state is in proportion to the commendation of that state.\nIn other words, it is only when you take into account _the whole\nconsequences of any state_, that he is able to maintain the coincidence\nof quantity of pleasure with the objects approved by Common Sense.\nIf we consider each state by itself, and ask what is the judgment of\nCommon Sense as to its goodness _as an end_, quite apart from its\ngoodness as a means, there can be no doubt that Common Sense holds many\nmuch less pleasant states to be better than many far more pleasant:\nthat it holds, with Mill, that there are higher pleasures, which are\nmore valuable, though less pleasant, than those which are lower. Prof.\nSidgwick might, of course, maintain that in this Common Sense is merely\nconfusing means and ends: that what it holds to be better as an end,\nis in reality only better as a means. But I think his argument is\ndefective in that he does not seem to see sufficiently plainly that,\nas far as intuitions of goodness _as an end_ are concerned, he is\nrunning grossly counter to Common Sense; that he does not emphasise\nsufficiently the distinction between _immediate_ pleasantness and\n_conduciveness_ to pleasure. In order to place fairly before us\nthe question what is good as an end we must take states that are\nimmediately pleasant and ask if the more pleasant are always also the\nbetter; and whether, if some that are less pleasant appear to be so, it\nis only because we think they are likely to increase the number of the\nmore pleasant. That Common Sense would deny both these suppositions,\nand rightly so, appears to me indubitable. It is commonly held that\ncertain of what would be called the lowest forms of sexual enjoyment,\nfor instance, are positively bad, although it is by no means clear that\nthey are not the most pleasant states we ever experience. Common Sense\nwould certainly not think it a sufficient justification for the pursuit\nof what Prof. Sidgwick calls the \u2018refined pleasures\u2019 here and now, that\nthey are the best means to the future attainment of a heaven, in which\nthere would be no more refined pleasures--no contemplation of beauty,\nno personal affections--but in which the greatest possible pleasure\nwould be obtained by a perpetual indulgence in bestiality. Yet Prof.\nSidgwick would be bound to hold that, if the greatest possible pleasure\ncould be obtained in this way, and if it were attainable, such a state\nof things would be a heaven indeed, and that all human endeavours\nshould be devoted to its realisation. I venture to think that this view\nis as false as it is paradoxical.\n\n\n=57.= It seems to me, then, that if we place fairly before us the\nquestion: Is consciousness of pleasure the sole good? the answer must\nbe: No. And with this the last defence of Hedonism has been broken\ndown. In order to put the question fairly we must isolate consciousness\nof pleasure. We must ask: Suppose we were conscious of pleasure only,\nand of nothing else, not even that we _were_ conscious, would that\nstate of things, however great the quantity, be very desirable? No one,\nI think, can suppose it so. On the other hand, it seems quite plain,\nthat we do regard as very desirable, many complicated states of mind\nin which the consciousness of pleasure is combined with consciousness\nof other things--states which we call \u2018enjoyment of\u2019 so and so. If\nthis is correct, then it follows that consciousness of pleasure is not\nthe sole good, and that many other states, in which it is included\nas a part, are much better than it. Once we recognise the principle\nof organic unities, any objection to this conclusion, founded on the\nsupposed fact that the other elements of such states have no value in\nthemselves, must disappear. And I do not know that I need say any more\nin refutation of Hedonism.\n\n\n=58.= It only remains to say something of the two forms in which a\nhedonistic doctrine is commonly held--Egoism and Utilitarianism.\n\nEgoism, as a form of Hedonism, is the doctrine which holds that we\nought each of us to pursue our own greatest happiness as our ultimate\nend. The doctrine will, of course, admit that sometimes the best means\nto this end will be to give pleasure to others; we shall, for instance,\nby so doing, procure for ourselves the pleasures of sympathy, of\nfreedom from interference, and of self-esteem; and these pleasures,\nwhich we may procure by sometimes aiming directly at the happiness of\nother persons, may be greater than any we could otherwise get. Egoism\nin this sense must therefore be carefully distinguished from Egoism\nin another sense, the sense in which Altruism is its proper opposite.\nEgoism, as commonly opposed to Altruism, is apt to denote merely\nselfishness. In this sense, a man is an egoist, if all his actions\nare actually directed towards gaining pleasure for himself; whether\nhe holds that he ought to act so, because he will thereby obtain for\nhimself the greatest possible happiness on the whole, or not. Egoism\nmay accordingly be used to denote the theory that we should always aim\nat getting pleasure for ourselves, because that is the best _means_\nto the ultimate end, whether the ultimate end be our own greatest\npleasure or not. Altruism, on the other hand, may denote the theory\nthat we ought always to aim at other people\u2019s happiness, on the ground\nthat this is the best _means_ of securing our own as well as theirs.\nAccordingly an Egoist, in the sense in which I am now going to talk of\nEgoism, an Egoist, who holds that his own greatest happiness is the\nultimate end, may at the same time be an Altruist: he may hold that\nhe ought to \u2018love his neighbour,\u2019 as the best means to being happy\nhimself. And conversely an Egoist, in the other sense, may at the same\ntime be a Utilitarian. He may hold that he ought always to direct his\nefforts towards getting pleasure for himself on the ground that he is\nthereby most likely to increase the general sum of happiness.\n\n\n=59.= I shall say more later about this second kind of Egoism, this\nanti-altruistic Egoism, this Egoism as a doctrine of means. What I am\nnow concerned with is that utterly distinct kind of Egoism, which holds\nthat each man ought rationally to hold: My own greatest happiness is\nthe only good thing there is; my actions can only be good as means, in\nso far as they help to win me this. This is a doctrine which is not\nmuch held by writers now-a-days. It is a doctrine that was largely\nheld by English Hedonists in the 17th and 18th centuries: it is, for\nexample, at the bottom of Hobbes\u2019 Ethics. But even the English school\nappear to have made one step forward in the present century: they are\nmost of them now-a-days Utilitarians. They do recognise that if my own\nhappiness is good, it would be strange that other people\u2019s happiness\nshould not be good too.\n\nIn order fully to expose the absurdity of this kind of Egoism, it is\nnecessary to examine certain confusions upon which its plausibility\ndepends.\n\nThe chief of these is the confusion involved in the conception of\n\u2018my own good\u2019 as distinguished from \u2018the good of others.\u2019 This is a\nconception which we all use every day; it is one of the first to which\nthe plain man is apt to appeal in discussing any question of Ethics:\nand Egoism is commonly advocated chiefly because its meaning is not\nclearly perceived. It is plain, indeed, that the name \u2018Egoism\u2019 more\nproperly applies to the theory that \u2018my own good\u2019 is the sole good,\nthan that my own pleasure is so. A man may quite well be an Egoist,\neven if he be not a Hedonist. The conception which is, perhaps, most\nclosely associated with Egoism is that denoted by the words \u2018my own\ninterest.\u2019 The Egoist is the man who holds that a tendency to promote\nhis own interest is the sole possible, and sufficient, justification\nof all his actions. But this conception of \u2018my own interest\u2019 plainly\nincludes, in general, very much more than my own pleasure. It is,\nindeed, only because and in so far as \u2018my own interest\u2019 has been\nthought to consist solely in my own pleasure, that Egoists have been\nled to hold that my own pleasure is the sole good. Their course of\nreasoning is as follows: The only thing I ought to secure is my\nown interest; but my own interest consists in my greatest possible\npleasure; and therefore the only thing I ought to pursue is my own\npleasure. That it is very natural, _on reflection_, thus to identify\nmy own pleasure with my own interest; and that it has been generally\ndone by modern _moralists_, may be admitted. But, when Prof. Sidgwick\npoints this out (III. XIV. \u00a7 5, Div. III.), he should have also pointed\nout that this identification has by no means been made in ordinary\nthought. When the plain man says \u2018my own interest,\u2019 he does _not_ mean\n\u2018my own pleasure\u2019--he does not commonly even include this--he means\nmy own advancement, my own reputation, the getting of a better income\netc., etc. That Prof. Sidgwick should not have noticed this, and that\nhe should give the reason he gives for the fact that the ancient\n_moralists_ did not identify \u2018my own interest\u2019 with my own pleasure,\nseems to be due to his having failed to notice that very confusion\nin the conception of \u2018my own good\u2019 which I am now to point out. That\nconfusion has, perhaps, been more clearly perceived by Plato than\nby any other moralist, and to point it out suffices to refute Prof.\nSidgwick\u2019s own view that Egoism is rational.\n\nWhat, then, is meant by \u2018my own good\u2019? In what sense can a thing be\ngood _for me_? It is obvious, if we reflect, that the only thing which\ncan belong to me, which can be _mine_, is something which is good, and\nnot the fact that it is good. When, therefore, I talk of anything I get\nas \u2018my own good,\u2019 I must mean either that the thing I get is good, or\nthat my possessing it is good. In both cases it is only the thing or\nthe possession of it which is _mine_, and not _the goodness_ of that\nthing or that possession. There is no longer any meaning in attaching\nthe \u2018my\u2019 to our predicate, and saying: The possession of this _by me_\nis _my_ good. Even if we interpret this by \u2018My possession of this is\nwhat _I_ think good,\u2019 the same still holds: for _what_ I think is that\nmy possession of it is good _simply_; and, if I think rightly, then\nthe truth is that my possession of it _is_ good simply--not, in any\nsense, _my_ good; and, if I think wrongly, it is not good at all. In\nshort, when I talk of a thing as \u2018my own good\u2019 all that I can mean\nis that something which will be exclusively mine, as my own pleasure\nis mine (whatever be the various senses of this relation denoted by\n\u2018possession\u2019), is also _good absolutely_; or rather that my possession\nof it is _good absolutely_. The _good_ of it can in no possible sense\nbe \u2018private\u2019 or belong to me; any more than a thing can _exist_\nprivately or _for_ one person only. The only reason I can have for\naiming at \u2018my own good,\u2019 is that it is _good absolutely_ that what I\nso call should belong to me--_good absolutely_ that I should _have_\nsomething, which, if I have it, others cannot have. But if it is _good\nabsolutely_ that I should have it, then everyone else has as much\nreason for aiming at _my_ having it, as I have myself. If, therefore,\nit is true of _any_ single man\u2019s \u2018interest\u2019 or \u2018happiness\u2019 that it\nought to be his sole ultimate end, this can only mean that _that_ man\u2019s\n\u2018interest\u2019 or \u2018happiness\u2019 is _the sole good, the_ Universal Good,\nand the only thing that anybody ought to aim at. What Egoism holds,\ntherefore, is that _each_ man\u2019s happiness is the sole good--that a\nnumber of different things are _each_ of them the only good thing there\nis--an absolute contradiction! No more complete and thorough refutation\nof any theory could be desired.\n\n\n=60.= Yet Prof. Sidgwick holds that Egoism is rational; and it will be\nuseful briefly to consider the reasons which he gives for this absurd\nconclusion. \u2018The Egoist,\u2019 he says (last Chap. \u00a7 1), \u2018may avoid the\nproof of Utilitarianism by declining to affirm,\u2019 either \u2018implicitly or\nexplicitly, that his own greatest happiness is not merely the ultimate\nrational end for himself, but a part of Universal Good.\u2019 And in the\npassage to which he here refers us, as having there \u2018seen\u2019 this,\nhe says: \u2018It cannot be proved that the difference between his own\nhappiness and another\u2019s happiness is not _for him_ all-important\u2019 (IV.\nii. \u00a7 1). What does Prof. Sidgwick mean by these phrases \u2018the ultimate\nrational end for himself,\u2019 and \u2018_for him_ all-important\u2019? He does not\nattempt to define them; and it is largely the use of such undefined\nphrases which causes absurdities to be committed in philosophy.\n\nIs there any sense in which a thing can be an ultimate rational end for\none person and not for another? By \u2018ultimate\u2019 must be meant at least\nthat the end is good-in-itself--good in our undefinable sense; and\nby \u2018rational,\u2019 at least, that it is truly good. That a thing should\nbe an ultimate rational end means, then, that it is truly good in\nitself; and that it is truly good in itself means that it is a part of\nUniversal Good. Can we assign any meaning to that qualification \u2018for\nhimself,\u2019 which will make it cease to be a part of Universal Good?\nThe thing is impossible: for the Egoist\u2019s happiness must _either_ be\ngood in itself, and so a part of Universal Good, _or else_ it cannot\nbe good in itself at all: there is no escaping this dilemma. And if\nit is not good at all, what reason can he have for aiming at it? how\ncan it be a rational end for him? That qualification \u2018for himself\u2019 has\nno meaning unless it implies \u2018_not_ for others\u2019; and if it implies\n\u2018not for others,\u2019 then it cannot be a rational end for him, since\nit cannot be truly good in itself: the phrase \u2018an ultimate rational\nend for himself\u2019 is a contradiction in terms. By saying that a thing\nis an end for one particular person, or good for him, can only be\nmeant one of four things. Either (1) it may be meant that the end\nin question is something which will belong exclusively to him; but\nin that case, if it is to be rational for him to aim at it, that he\nshould exclusively possess it must be a part of Universal Good. Or\n(2) it may be meant that it is the only thing at which he ought to\naim; but this can only be, because, by so doing, he will do the most\nhe can towards realising Universal Good: and this, in our case, will\nonly give Egoism as a doctrine of _means_. Or (3) it may be meant that\nthe thing is what he desires or thinks good; and then, if he thinks\nwrongly, it is not a rational end at all, and, if he thinks rightly,\nit is a part of Universal Good. Or (4) it may be meant that it is\npeculiarly appropriate that a thing which will belong exclusively to\nhim should also by him be approved or aimed at; but, in this case,\nboth that it should belong to him and that he should aim at it must be\nparts of Universal Good: by saying that a certain relation between two\nthings is fitting or appropriate, we can only mean that the existence\nof that relation is absolutely good in itself (unless it be so as a\nmeans, which gives case (2)). By no possible meaning, then, that can\nbe given to the phrase that his own happiness is the ultimate rational\nend for himself can the Egoist escape the implication that his own\nhappiness is absolutely good; and by saying that it is _the_ ultimate\nrational end, he must mean that it is the only good thing--the whole of\nUniversal Good: and, if he further maintains, that each man\u2019s happiness\nis the ultimate rational end for _him_, we have the fundamental\ncontradiction of Egoism--that an immense number of different things\nare, _each_ of them, _the sole good_.--And it is easy to see that the\nsame considerations apply to the phrase that \u2018the difference between\nhis own happiness and another\u2019s is _for him_ all-important.\u2019 This can\nonly mean either (1) that his own happiness is the only end which\nwill affect him, or (2) that the only important thing for him (as a\nmeans) is to look to his own happiness, or (3) that it is only his\nown happiness which he cares about, or (4) that it is good that each\nman\u2019s happiness should be the only concern of that man. And none of\nthese propositions, true as they may be, have the smallest tendency to\nshew that if his own happiness is desirable at all, it is not a part\nof Universal Good. Either his own happiness is a good thing or it is\nnot; and, in whatever sense it may be all-important for him, it must\nbe true that, if it is not good, he is not justified in pursuing it,\nand that, if it is good, everyone else has an equal reason to pursue\nit, so far as they are able and so far as it does not exclude their\nattainment of other more valuable parts of Universal Good. In short\nit is plain that the addition of \u2018for him\u2019 \u2018for me\u2019 to such words as\n\u2018ultimate rational end,\u2019 \u2018good,\u2019 \u2018important\u2019 can introduce nothing but\nconfusion. The only possible reason that can justify any action is that\nby it the greatest possible amount of what is good absolutely should be\nrealised. And if anyone says that the attainment of his own happiness\njustifies his actions, he must mean that this is the greatest possible\namount of Universal Good which he can realise. And this again can only\nbe true either because _he_ has no power to realise more, in which\ncase he only holds Egoism as a doctrine of means; or else because his\nown happiness is the greatest amount of Universal Good which can be\nrealised at all, in which case we have Egoism proper, and the flagrant\ncontradiction that every person\u2019s happiness is singly the greatest\namount of Universal Good which can be realised at all.\n\n\n=61.= It should be observed that, since this is so, \u2018the relation of\nRational Egoism to Rational Benevolence,\u2019 which Prof. Sidgwick regards\n\u2018as the profoundest problem of Ethics\u2019 (III. XIII. \u00a7 5, _n._ 1),\nappears in quite a different light to that in which he presents it.\n\u2018Even if a man,\u2019 he says, \u2018admits the self-evidence of the principle\nof Rational Benevolence, he may still hold that his own happiness is\nan end which it is irrational for him to sacrifice to any other; and\nthat therefore a harmony between the maxim of Prudence and the maxim of\nRational Benevolence must be somehow demonstrated, if morality is to\nbe made completely rational. This latter view is that which I myself\nhold\u2019 (last Chap. \u00a7 1). Prof. Sidgwick then goes on to shew \u2018that\nthe inseparable connection between Utilitarian Duty and the greatest\nhappiness of the individual who conforms to it cannot be satisfactorily\ndemonstrated on empirical grounds\u2019 (Ib. \u00a7 3). And the final paragraph\nof his book tells us that, since \u2018the reconciliation of duty and\nself-interest is to be regarded as a hypothesis logically necessary\nto avoid a fundamental _contradiction_ in one chief department of\nour thought, it remains to ask how far this necessity constitutes a\nsufficient reason for accepting this hypothesis[20]\u2019 (Ib. \u00a7 5). To\n\u2018assume the existence of such a Being, as God, by the _consensus_ of\ntheologians, is conceived to be\u2019 would, he has already argued, ensure\nthe required reconciliation; since the Divine Sanctions of such a God\n\u2018would, of course, suffice to make it always every one\u2019s interest to\npromote universal happiness to the best of his knowledge\u2019 (Ib. \u00a7 5).\n\n  [20] The italics are mine.\n\nNow what is this \u2018reconciliation of duty and self-interest,\u2019 which\nDivine Sanctions could ensure? It would consist in the mere fact that\nthe same conduct which produced the greatest possible happiness of\nthe greatest number would always also produce the greatest possible\nhappiness of the agent. If this were the case (and our empirical\nknowledge shews that it is not the case in this world), \u2018morality\u2019\nwould, Prof. Sidgwick thinks, be \u2018completely rational\u2019: we should avoid\n\u2018an ultimate and fundamental contradiction in our apparent intuitions\nof what is Reasonable in conduct.\u2019 That is to say, we should avoid\nthe necessity of thinking that it is as manifest an obligation to\nsecure our own greatest Happiness (maxim of Prudence), as to secure\nthe greatest Happiness on the whole (maxim of Benevolence). But it\nis perfectly obvious we should not. Prof. Sidgwick here commits the\ncharacteristic fallacy of Empiricism--the fallacy of thinking that\nan alteration in _facts_ could make a contradiction cease to be a\ncontradiction. That a single man\u2019s happiness should be _the sole\ngood_, and that also everybody\u2019s happiness should be _the sole good_,\nis a contradiction which cannot be solved by the assumption that the\nsame conduct will secure both: it would be equally contradictory,\nhowever certain we were that that assumption was justified. Prof.\nSidgwick strains at a gnat and swallows a camel. He thinks the Divine\nOmnipotence must be called into play to secure that what gives other\npeople pleasure should also give it to him--that only so can Ethics\nbe made rational; while he overlooks the fact that even this exercise\nof Divine Omnipotence would leave in Ethics a contradiction, in\ncomparison with which his difficulty is a trifle--a contradiction,\nwhich would reduce all Ethics to mere nonsense, and before which the\nDivine Omnipotence must be powerless to all eternity. That _each_\nman\u2019s happiness should be the _sole good_, which we have seen to be\nthe principle of Egoism, is in itself a contradiction: and that it\nshould also be true that the Happiness of all is the _sole good_,\nwhich is the principle of Universalistic Hedonism, would introduce\nanother contradiction. And that these propositions should all be true\nmight well be called \u2018the profoundest problem in Ethics\u2019: it would be\na problem necessarily insoluble. But they _cannot_ all be true, and\nthere is no reason, but confusion, for the supposition that they are.\nProf. Sidgwick confuses this contradiction with the mere fact (in which\nthere is no contradiction) that our own greatest happiness and that\nof all do not seem always attainable by the same means. This fact, if\nHappiness were the sole good, would indeed be of some importance; and,\non any view, similar facts are of importance. But they are nothing but\ninstances of the one important fact that in this world the quantity of\ngood which is attainable is ridiculously small compared to that which\nis imaginable. That I cannot get the most possible pleasure for myself,\nif I produce the most possible pleasure on the whole, is no more _the_\nprofoundest problem of Ethics, than that in any case I cannot get as\nmuch pleasure altogether as would be desirable. It only states that,\nif we get as much good as possible in one place, we may get less on\nthe whole, because the quantity of attainable good is limited. To say\nthat I have to choose between my own good and that of _all_ is a false\nantithesis: the only rational question is how to choose between my own\nand that of _others_, and the principle on which this must be answered\nis exactly the same as that on which I must choose whether to give\npleasure to this other person or to that.\n\n\n=62.= It is plain, then, that the doctrine of Egoism is\nself-contradictory; and that one reason why this is not perceived,\nis a confusion with regard to the meaning of the phrase \u2018my own\ngood.\u2019 And it may be observed that this confusion and the neglect\nof this contradiction are necessarily involved in the transition\nfrom Naturalistic Hedonism, as ordinarily held, to Utilitarianism.\nMill, for instance, as we saw, declares: \u2018Each person, so far as he\nbelieves it to be attainable, desires his own happiness\u2019 (p. 53). And\nhe offers this as a reason why the general happiness is desirable. We\nhave seen that to regard it as such, involves, in the first place,\nthe naturalistic fallacy. But moreover, even if that fallacy were\nnot a fallacy, it could only be a reason for Egoism and not for\nUtilitarianism. Mill\u2019s argument is as follows: A man desires his own\nhappiness; therefore his own happiness is desirable. Further: A man\ndesires nothing but his own happiness; therefore his own happiness is\nalone desirable. We have next to remember, that everybody, according\nto Mill, so desires his own happiness: and then it will follow\nthat everybody\u2019s happiness is alone desirable. And this is simply\na contradiction in terms. Just consider what it means. Each man\u2019s\nhappiness is the only thing desirable: several different things are\n_each_ of them the _only_ thing desirable. This is the fundamental\ncontradiction of Egoism. In order to think that what his arguments tend\nto prove is not Egoism but Utilitarianism, Mill must think that he can\ninfer from the proposition \u2018Each man\u2019s happiness is his own good,\u2019 the\nproposition \u2018The happiness of all is the good of all\u2019; whereas in fact,\nif we understand what \u2018his own good\u2019 means, it is plain that the latter\ncan only be inferred from \u2018The happiness of all is the good of each.\u2019\nNaturalistic Hedonism, then, logically leads only to Egoism. Of course,\na Naturalist might hold that what we aimed at was simply \u2018pleasure\u2019\nnot our own pleasure; and _that_, always assuming the naturalistic\nfallacy, would give an unobjectionable ground for Utilitarianism. But\nmore commonly he will hold that it is his own pleasure he desires, or\nat least will confuse this with the other; and then he must logically\nbe led to adopt Egoism and not Utilitarianism.\n\n\n=63.= The second cause I have to give why Egoism should be thought\nreasonable, is simply its confusion with that other kind of\nEgoism--Egoism as a doctrine of means. This second Egoism has a right\nto say: You ought to pursue your own happiness, sometimes at all\nevents; it may even say: Always. And when we find it saying this we\nare apt to forget its proviso: But only as a means to something else.\nThe fact is we are in an imperfect state; we cannot get the ideal all\nat once. And hence it is often our bounden duty, we often _absolutely_\n\u2018_ought_,\u2019 to do things which are good only or chiefly as means: we\nhave to do the best we can, what is absolutely \u2018right,\u2019 but not what\nis absolutely good. Of this I shall say more hereafter. I only mention\nit here because I think it is much more plausible to say that we ought\nto pursue our own pleasure as a means than as an end, and that this\ndoctrine, through confusion, lends some of its plausibility to the\nutterly different doctrine of Egoism proper: My own greatest pleasure\nis the only good thing.\n\n\n=64.= So much for Egoism. Of Utilitarianism not much need be said; but\ntwo points may seem deserving of notice.\n\nThe first is that this name, like that of Egoism, does not naturally\nsuggest that all our actions are to be judged according to the degree\nin which they are a means to _pleasure_. Its natural meaning is that\nthe standard of right and wrong in conduct is its tendency to promote\nthe _interest_ of everybody. And by _interest_ is commonly meant a\nvariety of different goods, classed together only because they are\nwhat a man commonly desires for himself, so far as his desires have\nnot that psychological quality which is meant by \u2018moral.\u2019 The \u2018useful\u2019\nthus means, and was in ancient Ethics systematically used to mean,\nwhat is a means to the attainment of goods other than moral goods. It\nis quite an unjustifiable assumption that these goods are only good\nas means to pleasure or that they are commonly so regarded. The chief\nreason for adopting the name \u2018Utilitarianism\u2019 was, indeed, merely to\nemphasize the fact that right and wrong conduct must be judged by its\nresults--as a means, in opposition to the strictly Intuitionistic view\nthat certain ways of acting were right and others wrong, whatever their\nresults might be. In thus insisting that what is right must mean what\nproduces the best possible results Utilitarianism is fully justified.\nBut with this correct contention there has been historically, and very\nnaturally, associated a double error. (1) The best possible results\nwere assumed to consist only in a limited class of goods, roughly\ncoinciding with those which were popularly distinguished as the\nresults of merely \u2018useful\u2019 or \u2018interested\u2019 actions; and these again\nwere hastily assumed to be good only as means to pleasure. (2) The\nUtilitarians tend to regard everything as a mere means, neglecting the\nfact that some things which are good as means are also good as ends.\nThus, for instance, assuming pleasure to be a good, there is a tendency\nto value present pleasure only as a means to future pleasure, and\nnot, as is strictly necessary if pleasure is good as an end, also to\n_weigh it against_ possible future pleasures. Much utilitarian argument\ninvolves the logical absurdity that what is here and now, never has\nany value in itself, but is only to be judged by its consequences;\nwhich again, of course, when they are realised, would have no value in\nthemselves, but would be mere means to a still further future, and so\non _ad infinitum_.\n\nThe second point deserving notice with regard to Utilitarianism is\nthat, when the name is used for a form of Hedonism, it does not\ncommonly, even in its description of its _end_, accurately distinguish\nbetween means and end. Its best-known formula is that the result by\nwhich actions are to be judged is \u2018the greatest happiness of the\ngreatest number.\u2019 But it is plain that, if pleasure is the sole good,\nprovided the quantity be equally great, an equally desirable result\nwill have been obtained whether it be enjoyed by many or by few, or\neven if it be enjoyed by nobody. It is plain that, if we ought to aim\nat the greatest happiness of the greatest number, this can only, on the\nhedonistic principle, be because the existence of pleasure in a great\nnumber of persons seems to be the best _means_ available for attaining\nthe existence of the greatest quantity of pleasure. This may actually\nbe the case; but it is fair to suspect that Utilitarians have been\ninfluenced, in their adoption of the hedonistic principle, by this\nfailure to distinguish clearly between pleasure or consciousness of\npleasure and its possession by a person. It is far easier to regard the\npossession of pleasure by a number of persons as the sole good, than so\nto regard the mere existence of an equally great quantity of pleasure.\nIf, indeed, we were to take the Utilitarian principle strictly, and to\nassume them to mean that the possession of pleasure by many persons\nwas good in itself, the principle is not hedonistic: it includes as\na necessary part of the ultimate end, the existence of a number of\npersons, and this will include very much more than mere pleasure.\n\nUtilitarianism, however, as commonly held, must be understood to\nmaintain that either mere consciousness of pleasure, or consciousness\nof pleasure together with the minimum adjunct which may be meant by\nthe existence of such consciousness in at least one _person_, is the\n_sole good_. This is its significance as an ethical doctrine; and as\nsuch it has already been refuted in my refutation of Hedonism. The\nmost that can be said for it is that it does not seriously mislead in\nits practical conclusions, on the ground that, as an empirical fact,\nthe method of acting which brings most good on the whole does also\nbring most pleasure. Utilitarians do indeed generally devote most of\ntheir arguments to shewing that the course of action which will bring\nmost pleasure is in general such as common sense would approve. We\nhave seen that Prof. Sidgwick appeals to this fact as tending to shew\nthat pleasure is the sole good; and we have also seen that it does not\ntend to shew this. We have seen how very flimsy the other arguments\nadvanced for this proposition are; and that, if it be fairly considered\nby itself, it appears to be quite ridiculous. And, moreover, that the\nactions which produce most good on the whole do also produce most\npleasure is extremely doubtful. The arguments tending to shew it are\nall more or less vitiated by the assumption that what appear to be\nnecessary conditions for the attainment of most pleasure in the near\nfuture, will always continue so to be. And, even with this vicious\nassumption, they only succeed in making out a highly problematical\ncase. How, therefore, this fact is to be explained, if it be a fact,\nneed not concern us. It is sufficient to have shewn that many complex\nstates of mind are much more valuable than the pleasure they contain.\nIf this be so, _no form of Hedonism can be true_. And, since the\npractical guidance afforded by pleasure as a _criterion_ is small in\nproportion as the calculation attempts to be accurate, we can well\nafford to await further investigation, before adopting a guide, whose\nutility is very doubtful and whose trustworthiness we have grave reason\nto suspect.\n\n\n=65.= The most important points which I have endeavoured to establish\nin this chapter are as follows. (1) Hedonism must be strictly\ndefined as the doctrine that \u2018Pleasure is the only thing which is\ngood in itself\u2019: this view seems to owe its prevalence mainly to the\nnaturalistic fallacy, and Mill\u2019s arguments may be taken as a type of\nthose which are fallacious in this respect; Sidgwick alone has defended\nit without committing this fallacy, and its final refutation must\ntherefore point out the errors in his arguments (36-38). (2) Mill\u2019s\n\u2018Utilitarianism\u2019 is criticised: it being shewn (_a_) that he commits\nthe naturalistic fallacy in identifying \u2018desirable\u2019 with \u2018desired\u2019;\n(_b_) that pleasure is not the only object of desire. The common\narguments for Hedonism seem to rest on these two errors (39-44). (3)\nHedonism is considered as an \u2018Intuition,\u2019 and it is pointed out (_a_)\nthat Mill\u2019s allowance that some pleasures are inferior in quality to\nothers implies both that it is an Intuition and that it is a false\none (46-48); (_b_) that Sidgwick fails to distinguish \u2018pleasure\u2019\nfrom \u2018consciousness of pleasure,\u2019 and that it is absurd to regard the\nformer, at all events, as the sole good (49-52); (_c_) that it seems\nequally absurd to regard \u2018consciousness of pleasure\u2019 as the sole good,\nsince, if it were so, a world in which nothing else existed might be\nabsolutely perfect: Sidgwick fails to put to himself this question,\nwhich is the only clear and decisive one (53-57). (4) What are commonly\nconsidered to be the two main types of Hedonism, namely, Egoism and\nUtilitarianism, are not only different from, but strictly contradictory\nof, one another; since the former asserts \u2018My own greatest pleasure\nis the _sole_ good,\u2019 the latter \u2018The greatest pleasure of all is the\n_sole_ good.\u2019 Egoism seems to owe its plausibility partly to the\nfailure to observe this contradiction--a failure which is exemplified\nby Sidgwick; partly to a confusion of Egoism as doctrine of end, with\nthe same as doctrine of means. If Hedonism is true, Egoism cannot\nbe so; still less can it be so, if Hedonism is false. The end of\nUtilitarianism, on the other hand, would, if Hedonism were true, be,\nnot indeed the best conceivable, but the best possible for us to\npromote; but it is refuted by the refutation of Hedonism (58-64).\n\n\n\n\nCHAPTER IV.\n\nMETAPHYSICAL ETHICS.\n\n\n=66.= In this chapter I propose to deal with a type of ethical theory\nwhich is exemplified in the ethical views of the Stoics, of Spinoza,\nof Kant, and especially of a number of modern writers, whose views in\nthis respect are mainly due to the influence of Hegel. These ethical\ntheories have this in common, that they use some _metaphysical_\nproposition as a ground for inferring some fundamental proposition of\nEthics. They all imply, and many of them expressly hold, that ethical\ntruths follow logically from metaphysical truths--that Ethics should be\nbased on _Metaphysics_. And the result is that they all describe the\nSupreme Good in _metaphysical_ terms.\n\nWhat, then, is to be understood by \u2018metaphysical\u2019? I use the term,\nas I explained in Chapter II., in opposition to \u2018natural.\u2019 I call\nthose philosophers preeminently \u2018metaphysical\u2019 who have recognised\nmost clearly that not everything which _is_ is a \u2018natural object.\u2019\n\u2018Metaphysicians\u2019 have, therefore, the great merit of insisting that\nour knowledge is not confined to the things which we can touch and\nsee and feel. They have always been much occupied, not only with that\nother class of natural objects which consists in mental facts, but also\nwith the class of objects or properties of objects, which certainly do\nnot exist in time, are not therefore parts of Nature, and which, in\nfact, do not exist at all. To this class, as I have said, belongs what\nwe mean by the adjective \u2018good.\u2019 It is not _goodness_, but only the\nthings or qualities which are good, which can exist in time--can have\nduration, and begin and cease to exist--can be objects of _perception_.\nBut the most prominent members of this class are perhaps numbers. It\nis quite certain that two natural objects may exist; but it is equally\ncertain that _two_ itself does not exist and never can. Two and two\n_are_ four. But that does not mean that either two or four exists.\nYet it certainly means _something_. Two _is_ somehow, although it\ndoes not exist. And it is not only simple terms of propositions--the\nobjects _about_ which we know truths--that belong to this class. The\ntruths which we know about them form, perhaps, a still more important\nsubdivision. No truth does, in fact, _exist_; but this is peculiarly\nobvious with regard to truths like \u2018Two and two are four,\u2019 in which\nthe objects, _about_ which they are truths, do not exist either. It\nis with the recognition of such truths as these--truths which have\nbeen called \u2018universal\u2019--and of their essential unlikeness to what\nwe can touch and see and feel, that metaphysics proper begins. Such\n\u2018universal\u2019 truths have always played a large part in the reasonings of\nmetaphysicians from Plato\u2019s time till now; and that they have directed\nattention to the difference between these truths and what I have\ncalled \u2018natural objects\u2019 is the chief contribution to knowledge which\ndistinguishes them from that other class of philosophers--\u2018empirical\u2019\nphilosophers--to which most Englishmen have belonged.\n\nBut though, if we are to define \u2018metaphysics\u2019 by the contribution\nwhich it has actually made to knowledge, we should have to say that\nit has emphasized the importance of objects which do not exist at\nall, metaphysicians themselves have not recognised this. They have\nindeed recognised and insisted that there are, or may be, objects\nof knowledge which do not _exist in time_, or at least which we\ncannot perceive; and in recognising the _possibility_ of these, as\nan object of investigation, they have, it may be admitted, done a\nservice to mankind. But they have in general supposed that whatever\ndoes not exist in time, must at least _exist_ elsewhere, if it is to\n_be_ at all--that, whatever does not exist in Nature, must exist in\nsome supersensible reality, whether timeless or not. Consequently\nthey have held that the truths with which they have been occupied,\nover and above the objects of perception, were in some way truths\nabout such supersensible reality. If, therefore, we are to define\n\u2018metaphysics\u2019 not by what it has attained, but by what it has\nattempted, we should say that it consists in the attempt to obtain\nknowledge, by processes of reasoning, of what exists but is _not_ a\npart of Nature. Metaphysicians have actually held that they could give\nus such knowledge of non-natural existence. They have held that their\nscience consists in giving us such knowledge as can be supported by\nreasons, of that supersensible reality of which religion professes to\ngive us a fuller knowledge, without any reasons. When, therefore, I\nspoke above of \u2018metaphysical\u2019 propositions, I meant propositions about\nthe existence of something supersensible--of something which is not\nan object of perception, and which cannot be inferred from what is an\nobject of perception by the same rules of inference by which we infer\nthe past and future of what we call \u2018Nature.\u2019 And when I spoke of\n\u2018metaphysical\u2019 terms, I meant terms which refer to qualities of such\na supersensible reality, which do not belong to anything \u2018natural.\u2019\nI admit that \u2018metaphysics\u2019 should investigate what reasons there may\nbe for belief in such a supersensible reality; since I hold that\nits peculiar province is the truth about all objects which are not\nnatural objects. And I think that the most prominent characteristic of\nmetaphysics, in history, has been its profession to _prove_ the truth\nabout non-natural _existents_. I define \u2018metaphysical,\u2019 therefore, by\na reference to supersensible _reality_; although I think that the only\nnon-natural objects, about which it has _succeeded_ in obtaining truth,\nare objects which do not exist at all.\n\nSo much, I hope, will suffice to explain what I mean by the term\n\u2018metaphysical,\u2019 and to shew that it refers to a clear and important\ndistinction. It was not necessary for my purpose to make the\ndefinition exhaustive or to shew that it corresponds in essentials\nwith established usage. The distinction between \u2018Nature\u2019 and a\nsupersensible reality is very familiar and very important: and since\nthe metaphysician endeavours to _prove_ things with regard to a\nsupersensible reality, and since he deals largely in truths which are\n_not_ mere natural facts, it is plain that his arguments, and errors\n(if any), will be of a more subtle kind than those which I have dealt\nwith under the name of \u2018Naturalism.\u2019 For these two reasons it seemed\nconvenient to treat \u2018Metaphysical Ethics\u2019 by themselves.\n\n\n=67.= I have said that those systems of Ethics, which I propose to\ncall \u2018Metaphysical,\u2019 are characterised by the fact that they describe\nthe Supreme Good in \u2018metaphysical\u2019 terms; and this has now been\nexplained as meaning that they describe it in terms of something\nwhich (they hold) does exist, but does not exist in Nature--in terms\nof a supersensible reality. A \u2018Metaphysical Ethics\u2019 is marked by the\nfact that it makes the assertion: That which would be perfectly good\nis something which exists, but is not natural; that which has some\ncharacteristic possessed by a supersensible reality. Such an assertion\nwas made by the Stoics when they asserted that a life in accordance\nwith Nature was perfect. For they did not mean by \u2018Nature,\u2019 what I have\nso defined, but something supersensible which they inferred to exist,\nand which they held to be perfectly good. Such an assertion, again,\nis made by Spinoza when he tells us that we are more or less perfect,\nin proportion as we are more or less closely united with Absolute\nSubstance by the \u2018intellectual love\u2019 of God. Such an assertion is made\nby Kant when he tells us that his \u2018Kingdom of Ends\u2019 is the ideal. And\nsuch, finally, is made by modern writers who tell us that the final and\nperfect end is to realise our _true_ selves--a self different both from\nthe whole and from any part of that which exists here and now in Nature.\n\nNow it is plain that such ethical principles have a merit, not\npossessed by Naturalism, in recognising that for perfect goodness\nmuch more is required than any quantity of what exists here and now\nor can be inferred as likely to exist in the future. And moreover it\nis quite possible that their assertions should be true, if we only\nunderstand them to assert that something which is real possesses all\nthe characteristics necessary for perfect goodness. But this is not\nall that they assert. They also imply, as I said, that this ethical\nproposition _follows_ from some proposition which is metaphysical: that\nthe question \u2018What is real?\u2019 has some logical bearing upon the question\n\u2018What is good?\u2019 It was for this reason that I described \u2018Metaphysical\nEthics\u2019 in Chapter II. as based upon the naturalistic fallacy. To\nhold that from any proposition asserting \u2018Reality is of this nature\u2019\nwe can infer, or obtain confirmation for, any proposition asserting\n\u2018This is good in itself\u2019 is to commit the naturalistic fallacy. And\nthat a knowledge of what is real supplies reasons for holding certain\nthings to be good in themselves is either implied or expressly\nasserted by all those who define the Supreme Good in metaphysical\nterms. This contention is part of what is meant by saying that Ethics\nshould be \u2018based\u2019 on Metaphysics. It is meant that some knowledge\nof supersensible reality is necessary _as a premise_ for correct\nconclusions as to what ought to exist. This view is, for instance,\nplainly expressed in the following statements: \u2018The truth is that the\ntheory of Ethics which seems most satisfactory has a metaphysical\nbasis.... If we rest our view of Ethics on the idea of the development\nof the ideal self or of the rational universe, the significance of\nthis cannot be made fully apparent without a metaphysical examination\nof the nature of self; _nor can its validity be established except\nby a discussion of the reality of the rational universe_[21].\u2019 The\nvalidity of an ethical conclusion about the nature of the ideal, it\nis here asserted, cannot be established except by considering the\nquestion whether that ideal is _real_. Such an assertion involves\nthe naturalistic fallacy. It rests upon the failure to perceive that\nany truth which asserts \u2018This is good in itself\u2019 is quite unique in\nkind--that it cannot be reduced to any assertion about reality, and\ntherefore must remain unaffected by any conclusions we may reach\nabout the nature of reality. This confusion as to the unique nature\nof ethical truths is, I have said, involved in all those ethical\ntheories which I have called metaphysical. It is plain that, but for\nsome confusion of the sort, no-one would think it worth while even to\ndescribe the Supreme Good in metaphysical terms. If, for instance,\nwe are told that the ideal consists in the realisation of the \u2018true\nself,\u2019 the very words suggest that the fact that the self in question\nis _true_ is supposed to have some bearing on the fact that it is\ngood. All the ethical truth which can possibly be conveyed by such\nan assertion would be just as well conveyed by saying that the ideal\nconsisted in the realisation of a particular kind of self, which might\nbe either real or purely imaginary. \u2018Metaphysical Ethics,\u2019 then,\ninvolve the supposition that Ethics can be _based_ on Metaphysics; and\nour first concern with them is to make clear that this supposition must\nbe false.\n\n  [21] Prof. J. S. Mackenzie, _A Manual of Ethics_, 4th ed., p.\n  431. The italics are mine.\n\n\n=68.= In what way can the nature of supersensible reality possibly have\na bearing upon Ethics?\n\nI have distinguished two kinds of ethical questions, which are far too\ncommonly confused with one another. Ethics, as commonly understood, has\nto answer both the question \u2018What ought to be?\u2019 and the question \u2018What\nought we to do?\u2019 The second of these questions can only be answered\nby considering what effects our actions will have. A complete answer\nto it would give us that department of Ethics which may be called the\ndoctrine of _means_ or practical Ethics. And upon this department of\nethical enquiry it is plain that the nature of a supersensible reality\nmay have a bearing. If, for instance, Metaphysics could tell us not\nonly that we are immortal, but also, in any degree, what effects our\nactions in this life will have upon our condition in a future one,\nsuch information would have an undoubted bearing upon the question\nwhat we ought to do. The Christian doctrines of heaven and hell are\nin this way highly relevant to practical Ethics. But it is worthy\nof notice that the most characteristic doctrines of Metaphysics are\nsuch as either have no such bearing upon practical Ethics or have\na purely negative bearing--involving the conclusion that there is\nnothing which we ought to do at all. They profess to tell us the\nnature not of a future reality, but of one that is eternal and which\ntherefore no actions of ours can have power to alter. Such information\n_may_ indeed have relevance to practical Ethics, but it must be of a\npurely negative kind. For, if it holds, not only that such an eternal\nreality exists, but also, as is commonly the case, that nothing else\nis real--that nothing either has been, is now, or will be real in\ntime--then truly it will follow that nothing we can do will ever bring\nany good to pass. For it is certain that our actions can only affect\nthe future; and if nothing can be real in the future, we can certainly\nnot hope ever to make any good thing real. It would follow, then,\nthat there can be nothing which we ought to do. We cannot possibly do\nany good; for neither our efforts, nor any result which they may seem\nto effect, have any real existence. But this consequence, though it\nfollows strictly from many metaphysical doctrines, is rarely drawn.\nAlthough a metaphysician may say that nothing is real but that which\nis eternal, he will generally allow that there is some reality also in\nthe temporal: and his doctrine of an eternal reality need not interfere\nwith practical Ethics, if he allows that, however good the eternal\nreality may be, yet some things will also exist in time, and that the\nexistence of some will be better than that of others. It is, however,\nworth while to insist upon this point, because it is rarely fully\nrealised.\n\nIf it is maintained that there is any validity at all in practical\nEthics--that any proposition which asserts \u2018We ought to do so and\nso\u2019 can have any truth--this contention can only be consistent with\nthe Metaphysics of an eternal reality, under two conditions. One of\nthese is, (1) that the true eternal reality, which is to be our guide,\ncannot, as is implied by calling it true, be the _only_ true reality.\nFor a moral rule, bidding us realise a certain end, can only be\njustified, if it is possible that that end should, at least partially,\nbe realised. Unless our efforts can effect the _real_ existence of some\ngood, however little, we certainly have no reason for making them.\nAnd if the eternal reality is the sole reality, then nothing good\ncan possibly exist in time: we can only be told to try to bring into\nexistence something which we know beforehand cannot possibly exist. If\nit is said that what exists in time can only be a manifestation of the\ntrue reality, it must at least be allowed that that manifestation is\nanother true reality--a good which we really can cause to exist; for\nthe production of something quite unreal, even if it were possible,\ncannot be a reasonable end of action. But if the manifestation of that\nwhich eternally exists is real, then that which eternally exists is not\nthe sole reality.\n\nAnd the second condition which follows from such a metaphysical\nprinciple of Ethics, is (2) that the eternal reality cannot be\nperfect--cannot be the sole good. For just as a reasonable rule of\nconduct requires that what we are told to realise should be capable of\nbeing truly real, so it requires that the realisation of this ideal\nshall be truly good. It is just that which _can_ be realised by our\nefforts--the appearance of the eternal in time, or whatsoever else is\nallowed to be attainable--which must be truly good, if it is to be\nworth our efforts. That the eternal reality is good, will by no means\njustify us in aiming at its manifestation, unless that manifestation\nitself be also good. For the manifestation is different from the\nreality: its difference is allowed, when we are told that it can be\nmade to exist, whereas the reality itself exists unalterably. And the\nexistence of this manifestation is the only thing which we can hope to\neffect: that also is admitted. If, therefore, the moral maxim is to be\njustified, it is the existence of this manifestation, as distinguished\nfrom the existence of its corresponding reality, which must be truly\ngood. The reality may be good too: but to justify the statement that we\nought to produce anything, it must be maintained, that just that thing\nitself, and not something else which may be like it, is truly good.\nIf it is not true that the existence of the manifestation will add\nsomething to the sum of good in the Universe, then we have no reason to\naim at making it exist; and if it is true that it will add something to\nthe sum of good, then the existence of that which is eternal cannot be\nperfect by itself--it cannot include the whole of possible goods.\n\nMetaphysics, then, will have a bearing upon practical Ethics--upon the\nquestion what we ought to do--if it can tell us anything about the\nfuture consequences of our actions beyond what can be established by\nordinary inductive reasoning. But the most characteristic metaphysical\ndoctrines, those which profess to tell us not about the future but\nabout the nature of an eternal reality, can either have no bearing upon\nthis practical question or else must have a purely destructive bearing.\nFor it is plain that what exists eternally cannot be affected by our\nactions; and only what is affected by our actions can have a bearing\non their value as means. But the nature of an eternal reality either\nadmits no inference as to the results of our actions, except in so far\nas it can _also_ give us information about the future (and how it can\ndo this is not plain), or else, if, as is usual, it is maintained to\nbe the sole reality and the sole good, it shews that no results of our\nactions can have any value whatever.\n\n\n=69.= But this bearing upon practical Ethics, such as it is, is not\nwhat is commonly meant when it is maintained that Ethics must be based\non Metaphysics. It is not the assertion of this relation which I have\ntaken to be characteristic of Metaphysical Ethics. What metaphysical\nwriters commonly maintain is not merely that Metaphysics can help us to\ndecide what the effects of our actions will be, but that it can tell us\nwhich among possible effects will be good and which will be bad. They\nprofess that Metaphysics is a necessary basis for an answer to that\nother and primary ethical question: What ought to be? What is good in\nitself? That no truth about what is real can have any logical bearing\nupon the answer to this question has been proved in Chapter I. To\nsuppose that it has, implies the naturalistic fallacy. All that remains\nfor us to do is, therefore, to expose the main errors which seem to\nhave lent plausibility to this fallacy in its metaphysical form. If\nwe ask: What bearing can Metaphysics have upon the question, What is\ngood? the only possible answer is: Obviously and absolutely none. We\ncan only hope to enforce conviction that this answer is the only true\none by answering the question: Why has it been supposed to have such a\nbearing? We shall find that metaphysical writers seem to have failed to\ndistinguish this primary ethical question: What is good? from various\nother questions; and to point out these distinctions will serve to\nconfirm the view that their profession to base Ethics on Metaphysics is\nsolely due to confusion.\n\n\n=70.= And, first of all, there is an ambiguity in the very question:\nWhat is good? to which it seems some influence must be attributed.\nThe question may mean either: Which among existing things are good?\nor else: What _sort of_ things are good, what are the things which,\nwhether they _are_ real or not, ought to be real? And of these two\nquestions it is plain that to answer the first, we must know both\nthe answer to the second and also the answer to the question: What\nis real? It asks us for a catalogue of all the good things in the\nUniverse; and to answer it we must know both what things there are in\nthe Universe and also which of them are good. Upon this question then\nour Metaphysics would have a bearing, if it can tell us what is real.\nIt would help us to complete the list of things which are both real\nand good. But to make such a list is not the business of Ethics. So\nfar as it enquires What is good? its business is finished when it has\ncompleted the list of things which ought to exist, whether they do\nexist or not. And if our Metaphysics is to have any bearing upon this\npart of the ethical problem, it must be because the fact that something\nis real gives a reason for thinking that it or something else is good,\nwhether it be real or not. That any such fact can give any such reason\nis impossible; but it may be suspected that the contrary supposition\nhas been encouraged by the failure to distinguish between the assertion\n\u2018This is good,\u2019 when it means \u2018_This sort of thing_ is good,\u2019 or \u2018This\nwould be good, if it existed,\u2019 and the assertion \u2018This existing thing\nis good.\u2019 The latter proposition obviously cannot be true, unless\nthe thing exists; and hence the proof of the thing\u2019s existence is a\nnecessary step to its proof. Both propositions, however, in spite\nof this immense difference between them, are commonly expressed in\nthe same terms. We use the same words, when we assert an ethical\nproposition about a subject that is actually real, and when we assert\nit about a subject considered as merely possible.\n\nIn this ambiguity of language we have, then, a possible source of error\nwith regard to the bearing of truths that assert reality upon truths\nthat assert goodness. And that this ambiguity is actually neglected by\nthose metaphysical writers who profess that the Supreme Good consists\nin an eternal reality may be shewn in the following way. We have seen,\nin considering the possible bearing of Metaphysics upon Practical\nEthics, that, since what exists eternally cannot possibly be affected\nby our actions, no practical maxim can possibly be true, if the sole\nreality is eternal. This fact, as I said, is commonly neglected by\nmetaphysical writers: they assert both of the two contradictory\npropositions that the sole reality is eternal and that its realisation\nin the future is a good too. Prof. Mackenzie, we saw, asserts that we\nought to aim at the realisation of \u2018the true self\u2019 or \u2018the rational\nuniverse\u2019: and yet Prof. Mackenzie holds, as the word \u2018true\u2019 plainly\nimplies, that both \u2018the true self\u2019 and \u2018the rational universe\u2019 are\neternally real. Here we have already a contradiction in the supposition\nthat what is eternally real can be realised in the future; and it is\ncomparatively unimportant whether or not we add to this the further\ncontradiction involved in the supposition that the eternal is the sole\nreality. That such a contradiction should be supposed valid can only be\nexplained by a neglect of the distinction between a real subject and\nthe character which that real subject possesses. _What_ is eternally\nreal may, indeed, be realised in the future, if by this be only meant\nthe _sort of thing_ which is eternally real. But when we assert that\na thing is good, what we mean is that its existence or reality is\ngood; and the eternal existence of a thing cannot possibly be the\nsame good as the existence in time of what, in a necessary sense,\nis nevertheless the _same_ thing. When, therefore, we are told that\nthe future realisation of the _true_ self is good, this can at most\nonly mean that the future realisation of a self _exactly like_ the\nself, which is true and exists eternally, is good. If this fact were\nclearly stated, instead of consistently ignored, by those who advocate\nthe view that the Supreme Good can be defined in these metaphysical\nterms, it seems probable that the view that a knowledge of reality\nis necessary to a knowledge of the Supreme Good would lose part of\nits plausibility. That that at which we ought to aim cannot possibly\nbe that which is eternally real, even if it be exactly like it; and\nthat the eternal reality cannot possibly be the sole good--these two\npropositions seem sensibly to diminish the probability that Ethics must\nbe based on Metaphysics. It is not very plausible to maintain that\nbecause one thing is real, therefore something like it, which is not\nreal, would be good. It seems, therefore, that some of the plausibility\nof Metaphysical Ethics may be reasonably attributed to the failure to\nobserve that verbal ambiguity, whereby \u2018This is good\u2019 may mean either\n\u2018This real thing is good\u2019 or \u2018The existence of this thing (whether it\nexists or not) would be good.\u2019\n\n\n=71.= By exposing this ambiguity, then, we are enabled to see more\nclearly what must be meant by the question: Can Ethics be based on\nMetaphysics? and we are, therefore, more likely to find the correct\nanswer. It is now plain that a metaphysical principle of Ethics\nwhich says \u2018This eternal reality is the Supreme Good\u2019 can only mean\n\u2018Something like this eternal reality would be the Supreme Good.\u2019 We\nare now to understand such principles as having the only meaning which\nthey can consistently have, namely, as describing the kind of thing\nwhich ought to exist in the future, and which we ought to try to bring\nabout. And, when this is clearly recognised, it seems more evident\nthat the knowledge that such a kind of thing is also eternally real,\ncannot help us at all towards deciding the properly ethical question:\nIs the existence of that kind of good thing? If we can see that an\neternal reality is good, we can see, equally easily, once the idea\nof such a thing has been suggested to us, that it _would_ be good.\nThe metaphysical construction of Reality would therefore be quite as\nuseful, for the purposes of Ethics, if it were a mere construction\nof an imaginary Utopia: provided the kind of thing suggested is the\nsame, fiction is as useful as truth, for giving us matter, upon\nwhich to exercise the judgment of value. Though, therefore, we admit\nthat Metaphysics may serve an ethical purpose, in suggesting things,\nwhich would not otherwise have occurred to us, but which, when they\nare suggested, we see to be good; yet, it is not as Metaphysics--as\nprofessing to tell us what is real--that it has this use. And, in\nfact, the pursuit of truth must limit the usefulness of Metaphysics\nin this respect. Wild and extravagant as are the assertions which\nmetaphysicians have made about reality, it is not to be supposed but\nthat they have been partially deterred from making them wilder still,\nby the idea that it was their business to tell nothing but the truth.\nBut the wilder they are, and the less useful for Metaphysics, the\nmore useful will they be for Ethics; since, in order to be sure that\nwe have neglected nothing in the description of our ideal, we should\nhave had before us as wide a field as possible of suggested goods. It\nis probable that this utility of Metaphysics, in suggesting possible\nideals, may sometimes be what is meant by the assertion that Ethics\nshould be based on Metaphysics. It is not uncommon to find that which\nsuggests a truth confused with that on which it logically depends; and\nI have already pointed out that Metaphysical have, in general, this\nsuperiority over Naturalistic systems, that they conceive the Supreme\nGood as something differing more widely from what exists here and now.\nBut, if it be recognised that, in this sense, Ethics should, far more\nemphatically, be _based on_ fiction, metaphysicians will, I think,\nadmit that a connection of this kind between Metaphysics and Ethics\nwould by no means justify the importance which they attribute to the\nbearing of the one study on the other.\n\n\n=72.= We may, then, attribute the obstinate prejudice that a knowledge\nof supersensible reality is a necessary step to a knowledge of what\nis good in itself, partly to a failure to perceive that the subject\nof the latter judgment is not anything _real_ as such, and partly to\na failure to distinguish the cause of our perception of a truth from\nthe reason why it is true. But these two causes will carry us only a\nvery little way in our explanation of why Metaphysics should have been\nsupposed to have a bearing upon Ethics. The first explanation which\nI have given would only account for the supposition that a thing\u2019s\nreality is a _necessary condition_ for its goodness. This supposition\nis, indeed, commonly made; we find it commonly presupposed that unless\na thing can be shewn to be involved in the constitution of reality, it\ncannot be good. And it is, therefore, worth while to insist that this\nis not the case; that Metaphysics is not even necessary to furnish\n_part_ of the basis of Ethics. But when metaphysicians talk of basing\nEthics on Metaphysics they commonly mean much more than this. They\ncommonly mean that Metaphysics is the _sole_ basis of Ethics--that it\nfurnishes not only one necessary condition but _all_ the conditions\nnecessary to prove that certain things are good. And this view may,\nat first sight, appear to be held in two different forms. It may be\nasserted that merely to prove a thing supersensibly real is sufficient\nto prove it good: that the truly real must, for that reason alone,\nbe truly good. But more commonly it appears to be held that the real\nmust be good because it possesses certain characters. And we may, I\nthink, reduce the first kind of assertion to no more than this. When\nit is asserted that the real must be good, because it is real, it is\ncommonly also held that this is only because, in order to be real, it\nmust be of a certain kind. The reasoning by which it is thought that a\nmetaphysical enquiry can give an ethical conclusion is of the following\nform. From a consideration of what it is to be real, we can infer that\nwhat is real must have certain supersensible properties: but to have\nthese properties is identical with being good--it is the very meaning\nof the word: it follows therefore that what has these properties is\ngood: and from a consideration of what it is to be real, we can again\ninfer what it is that has these properties. It is plain that, if such\nreasoning were correct, any answer which could be given to the question\n\u2018What is good in itself?\u2019 could be arrived at by a purely metaphysical\ndiscussion and by that alone. Just as, when Mill supposed that \u2018to\nbe good\u2019 _meant_ \u2018to be desired,\u2019 the question \u2018What is good?\u2019 could\nbe and must be answered solely by an empirical investigation of the\nquestion what was desired; so here, if to be good means to have some\nsupersensible property, the ethical question can and must be answered\nby a metaphysical enquiry into the question, What has this property?\nWhat, then, remains to be done in order to destroy the plausibility of\nMetaphysical Ethics, is to expose the chief errors which seem to have\nled metaphysicians to suppose that to be good _means_ to possess some\nsupersensible property.\n\n\n=73.= What, then, are the chief reasons which have made it seem\nplausible to maintain that to be good must _mean_ to possess some\nsupersensible property or to be related to some supersensible reality?\n\nWe may, first of all, notice one, which seems to have had some\ninfluence in causing the view that good must be defined by _some_\nsuch property, although it does not suggest any _particular_ property\nas the one required. This reason lies in the supposition that the\nproposition \u2018This is good\u2019 or \u2018This would be good, if it existed\u2019 must,\nin a certain respect, be of the same type as other propositions.\nThe fact is that there is one type of proposition so familiar to\neveryone, and therefore having such a strong hold upon the imagination,\nthat philosophers have always supposed that all other types must be\nreducible to it. This type is that of the objects of experience--of all\nthose truths which occupy our minds for the immensely greater part of\nour waking lives: truths such as that somebody is in the room, that I\nam writing or eating or talking. All these truths, however much they\nmay differ, have this in common that in them both the grammatical\nsubject and the grammatical predicate stand for something which exists.\nImmensely the commonest type of truth, then, is one which asserts a\nrelation between two existing things. Ethical truths are immediately\nfelt not to conform to this type, and the naturalistic fallacy arises\nfrom the attempt to make out that, in some roundabout way, they do\nconform to it. It is immediately obvious that when we see a thing\nto be good, its goodness is not a property which we can take up in\nour hands, or separate from it even by the most delicate scientific\ninstruments, and transfer to something else. It is not, in fact, like\nmost of the predicates which we ascribe to things, a _part_ of the\nthing to which we ascribe it. But philosophers suppose that the reason\nwhy we cannot take goodness up and move it about, is not that it is\na different _kind_ of object from any which can be moved about, but\nonly that it _necessarily_ exists together with anything with which\nit does exist. They explain the type of ethical truths by supposing\nit identical with the type of scientific laws. And it is only when\nthey have done this that the naturalistic philosophers proper--those\nwho are empiricists--and those whom I have called \u2018metaphysical\u2019 part\ncompany. These two classes of philosophers do, indeed, differ with\nregard to the nature of scientific laws. The former class tend to\nsuppose that when they say \u2018This always accompanies that\u2019 they mean\nonly \u2018This has accompanied, does now, and will accompany that in these\nparticular instances\u2019: they reduce the scientific law quite simply\nand directly to the familiar type of proposition which I have pointed\nout. But this does not satisfy the metaphysicians. They see that when\nyou say \u2018This would accompany that, _if_ that existed,\u2019 you don\u2019t mean\nonly that this and that have existed and will exist together so many\ntimes. But it is beyond even their powers to believe that what you do\nmean is merely what you say. They still think you must mean, somehow\nor other, that something does exist, since that is what you generally\nmean when you say anything. They are as unable as the empiricists to\nimagine that you can ever mean that 2 + 2 = 4. The empiricists say this\nmeans that so many couples of couples of things have in each case been\nfour things; and hence that 2 and 2 would not make 4, unless precisely\nthose things had existed. The metaphysicians feel that this is wrong;\nbut they themselves have no better account of its meaning to give\nthan either, with Leibniz, that God\u2019s mind is in a certain state, or,\nwith Kant, that your mind is in a certain state, or finally, with Mr\nBradley, that something is in a certain state. Here, then, we have the\nroot of the naturalistic fallacy. The metaphysicians have the merit\nof seeing that when you say \u2018This would be good, if it existed,\u2019 you\ncan\u2019t mean merely \u2018This has existed and was desired,\u2019 however many\ntimes that may have been the case. They will admit that some good\nthings have not existed in this world, and even that some may not have\nbeen desired. But what you can mean, except that _something_ exists,\nthey really cannot see. Precisely the same error which leads them to\nsuppose that there must _exist_ a supersensible Reality, leads them to\ncommit the naturalistic fallacy with regard to the meaning of \u2018good.\u2019\nEvery truth, they think, must mean somehow that something exists; and\nsince, unlike the empiricists, they recognise some truths which do not\nmean that anything exists here and now, these they think must mean\nthat something exists _not_ here and now. On the same principle, since\n\u2018good\u2019 is a predicate which neither does nor can exist, they are bound\nto suppose either that \u2018to be good\u2019 means to be related to some other\nparticular thing which can exist and does exist \u2018in reality\u2019; or else\nthat it means merely \u2018to belong to the real world\u2019--that goodness is\ntranscended or absorbed in reality.\n\n\n=74.= That such a reduction of _all_ propositions to the type of those\nwhich assert either that something exists or that something which\nexists has a certain attribute (which means, that both exist in a\ncertain relation to one another), is erroneous, may easily be seen by\nreference to the particular class of ethical propositions. For whatever\nwe may have proved to exist, and whatever two existents we may have\nproved to be necessarily connected with one another, it still remains\na distinct and different question whether what thus exists is good;\nwhether either or both of the two existents is so; and whether it is\ngood that they should exist together. To assert the one is plainly and\nobviously _not_ the same thing as to assert the other. We understand\nwhat we mean by asking: Is this, which exists, or necessarily exists,\nafter all, good? and we perceive that we are asking a question which\nhas _not_ been answered. In face of this direct perception that the\ntwo questions are distinct, no proof that they _must_ be identical can\nhave the slightest value. That the proposition \u2018This is good\u2019 is thus\ndistinct from every other proposition was proved in Chapter I.; and I\nmay now illustrate this fact by pointing out how it is distinguished\nfrom two particular propositions with which it has commonly been\nidentified. That so and so _ought to be done_ is commonly called a\nmoral _law_, and this phrase naturally suggests that this proposition\nis in some way analogous either to a natural law, or to a law in the\nlegal sense, or to both. All three are, in fact, really analogous in\none respect, and in one respect only: that they include a proposition\nwhich is _universal_. A moral law asserts \u2018This is good _in all\ncases_\u2019; a natural law asserts \u2018This happens _in all cases_\u2019; and a\nlaw, in the legal sense, \u2018It is commanded that this be done, or be left\nundone, _in all cases_.\u2019 But since it is very natural to suppose that\nthe analogy extends further, and that the assertion \u2018This is good in\nall cases\u2019 is equivalent to the assertion \u2018This happens in all cases\u2019\nor to the assertion \u2018It is commanded that this be done in all cases,\u2019\nit may be useful briefly to point out that they are _not_ equivalent.\n\n\n=75.= The fallacy of supposing moral law to be analogous to natural\nlaw in respect of asserting that some action is one which is always\nnecessarily done is contained in one of the most famous doctrines of\nKant. Kant identifies what ought to be with the law according to which\na Free or Pure Will _must_ act--with the only kind of action which is\npossible for it. And by this identification he does not mean merely to\nassert that the Free Will is _also_ under the necessity of doing what\nit ought; he means that what it ought to do _means_ nothing but its own\nlaw--the law according to which it must act. It differs from the human\nwill just in that, what _we_ ought to do, is what _it_ necessarily\ndoes. It is \u2018autonomous\u2019; and by this is meant (among other things)\nthat there is no separate standard by which it can be judged: that the\nquestion \u2018Is the law by which this Will acts a good one?\u2019 is, in its\ncase, meaningless. It follows that what is necessarily willed by this\nPure Will is good, not _because_ that Will is good, nor for any other\nreason; but merely because it is what is necessarily willed by a Pure\nWill.\n\nKant\u2019s assertion of the \u2018Autonomy of the Practical Reason\u2019 thus has\nthe very opposite effect to that which he desired; it makes his\nEthics ultimately and hopelessly \u2018heteronomous.\u2019 His Moral Law is\n\u2018independent\u2019 of Metaphysics only in the sense that according to him\nwe can _know_ it independently; he holds that we can only infer that\nthere is Freedom, from the fact that the Moral Law is true. And so far\nas he keeps strictly to this view, he does avoid the error, into which\nmost metaphysical writers fall, of allowing his opinions as to what is\nreal to influence his judgments of what is good. But he fails to see\nthat on his view the Moral Law is dependent upon Freedom in a far more\nimportant sense than that in which Freedom depends on the Moral Law. He\nadmits that Freedom is the _ratio essendi_ of the Moral Law, whereas\nthe latter is only _ratio cognoscendi_ of Freedom. And this means that,\nunless Reality be such as he says, no assertion that \u2018This is good\u2019 can\npossibly be true: it can indeed have no meaning. He has, therefore,\nfurnished his opponents with a conclusive method of attacking the\nvalidity of the Moral Law. If they can only shew by some other means\n(which he denies to be possible but leaves theoretically open) that the\nnature of Reality is not such as he says, he cannot deny that they will\nhave proved his ethical principle to be false. If that \u2018This ought to\nbe done\u2019 _means_ \u2018This is willed by a Free Will,\u2019 then, if it can be\nshewn that there is no Free Will which wills anything, it will follow\nthat nothing ought to be done.\n\n\n=76.= And Kant also commits the fallacy of supposing that \u2018This ought\nto be\u2019 means \u2018This is commanded.\u2019 He conceives the Moral Law to be an\nImperative. And this is a very common mistake. \u2018This ought to be,\u2019 it\nis assumed, must mean \u2018This is commanded\u2019; nothing, therefore, would\nbe good unless it were commanded; and since commands in this world\nare liable to be erroneous, what ought to be in its ultimate sense\nmeans \u2018what is commanded by some real supersensible authority.\u2019 With\nregard to this authority it is, then, no longer possible to ask \u2018Is it\nrighteous?\u2019 Its commands cannot fail to be right, because to be right\nmeans to be what it commands. Here, therefore, law, in the moral sense,\nis supposed analogous to law, in the legal sense, rather than, as in\nthe last instance, to law in the natural sense. It is supposed that\nmoral obligation is analogous to legal obligation, with this difference\nonly that whereas the source of legal obligation is earthly, that of\nmoral obligation is heavenly. Yet it is obvious that if by a source of\nobligation is meant only a power which binds you or compels you to do\na thing, it is not because it does do this that you ought to obey it.\nIt is only if it be itself so good, that it commands and enforces only\nwhat is good, that it can be a source of moral obligation. And in that\ncase what it commands and enforces would be good, whether commanded and\nenforced or not. Just that which makes an obligation legal, namely the\nfact that it is commanded by a certain kind of authority, is entirely\nirrelevant to a moral obligation. However an authority be defined, its\ncommands will be _morally_ binding only if they are--morally binding;\nonly if they tell us what ought to be or what is a means to that which\nought to be.\n\n\n=77.= In this last error, in the supposition that when I say \u2018You\nought to do this\u2019 I must mean \u2018You are commanded to do this,\u2019 we\nhave one of the reasons which has led to the supposition that the\nparticular supersensible property by reference to which good must\nbe defined is Will. And that ethical conclusions may be obtained by\nenquiring into the nature of a fundamentally real Will seems to be by\nfar the commonest assumption of Metaphysical Ethics at the present\nday. But this assumption seems to owe its plausibility, not so much\nto the supposition that \u2018ought\u2019 expresses a \u2018command,\u2019 as to a far\nmore fundamental error. This error consists in supposing that to\nascribe certain predicates to a thing is the same thing as to say\nthat that thing is the object of a certain kind of psychical state.\nIt is supposed that to say that a thing is real or true is the same\nthing as to say that it is known in a certain way; and that the\ndifference between the assertion that it is good and the assertion\nthat it is real--between an ethical, therefore, and a metaphysical\nproposition--_consists_ in the fact that whereas the latter asserts its\nrelation to Cognition the former asserts its relation to Will.\n\nNow that this is an error has been already shewn in Chapter I. That\nthe assertion \u2018This is good\u2019 is _not_ identical with the assertion\n\u2018This is willed,\u2019 either by a supersensible will, or otherwise, nor\nwith any other proposition, has been proved; nor can I add anything\nto that proof. But in face of this proof it may be anticipated that\ntwo lines of defence may be taken up. (1) It may be maintained that,\nnevertheless, they really are identical, and facts may be pointed out\nwhich seem to prove that identity. Or else (2) it may be said that an\n_absolute_ identity is not maintained: that it is only meant to assert\nthat there is some special connection between will and goodness, such\nas makes an enquiry into the real nature of the former an essential\nstep in the proof of ethical conclusions. In order to meet these two\npossible objections, I propose first to shew what possible connections\nthere are or may be between goodness and will; and that none of these\ncan justify us in asserting that \u2018This is good\u2019 is identical with \u2018This\nis willed.\u2019 On the other hand it will appear that some of them may be\neasily confused with this assertion of identity; and that therefore the\nconfusion is likely to have been made. This part of my argument will,\ntherefore, already go some way towards meeting the second objection.\nBut what must be conclusive against this is to shew that any possible\nconnection between will and goodness _except_ the _absolute_ identity\nin question, would not be sufficient to give an enquiry into Will the\nsmallest relevance to the proof of any ethical conclusion.\n\n\n=78.= It has been customary, since Kant\u2019s time, to assert that\nCognition, Volition, and Feeling are three fundamentally distinct\nattitudes of the mind towards reality. They are three distinct ways\nof experiencing, and each of them informs us of a distinct aspect\nunder which reality may be considered. The \u2018Epistemological\u2019 method of\napproaching Metaphysics rests on the assumption that by considering\nwhat is \u2018implied in\u2019 Cognition--what is its \u2018ideal\u2019--we may discover\nwhat properties the world must have, if it is to be _true_. And\nsimilarly it is held that by considering what is \u2018implied in\u2019 the fact\nof Willing or Feeling--what is the \u2018ideal\u2019 which they presuppose--we\nmay discover what properties the world must have, if it is to be good\nor beautiful. The orthodox Idealistic Epistemologist differs from the\nSensationalist or Empiricist in holding that what we directly cognise\nis neither all true nor yet the whole truth: in order to reject the\nfalse and to discover further truths we must, he says, not take\ncognition merely as it presents itself, but discover what is _implied_\nin it. And similarly the orthodox Metaphysical Ethicist differs from\nthe mere Naturalist, in holding that not everything which we actually\nwill is good, nor, if good, completely good: what is really good is\nthat which is implied in the essential nature of will. Others again\nthink that Feeling, and not Will, is the fundamental _datum_ for\nEthics. But, in either case, it is agreed that Ethics has some relation\nto Will or Feeling which it has not to Cognition, and which other\nobjects of study have to Cognition. Will or Feeling, on the one hand,\nand Cognition, on the other, are regarded as in some sense coordinate\nsources of philosophical knowledge--the one of Practical, the other of\nTheoretical philosophy.\n\nWhat, that is true, can possibly be meant by this view?\n\n\n=79.= First of all, it may be meant that, just as, by reflection on our\nperceptual and sensory experience, we become aware of the distinction\nbetween truth and falsehood, so it is by reflection on our experiences\nof feeling and willing that we become aware of ethical distinctions.\nWe should not know what was meant by thinking one thing better than\nanother unless the attitude of our will or feeling towards one thing\nwas different from its attitude towards another. All this may be\nadmitted. But so far we have only the psychological fact that it is\nonly _because_ we will or feel things in a certain way, that we ever\ncome to think them good; just as it is only because we have certain\nperceptual experiences, that we ever come to think things true. Here,\nthen, is a special connection between willing and goodness; but it is\nonly a _causal_ connection--that willing is a necessary condition for\nthe cognition of goodness.\n\nBut it may be said further that willing and feeling are not only the\norigin of cognitions of goodness; but that to will a thing, or to have\na certain feeling towards a thing, is the _same thing_ as to think it\ngood. And it may be admitted that even this is _generally_ true in a\nsense. It does seem to be true that we hardly ever think a thing good,\nand never very decidedly, without at the same time having a special\nattitude of feeling or will towards it; though it is certainly not the\ncase that this is true universally. And the converse may possibly be\ntrue universally: it may be the case that a perception of goodness is\nincluded in the complex facts which we mean by willing and by having\ncertain kinds of feeling. Let us admit then, that to think a thing good\nand to will it are _the same thing_ in this sense, that, wherever the\nlatter occurs, the former also occurs as a _part_ of it; and even that\nthey are _generally the same thing_ in the converse sense, that when\nthe former occurs it is generally a part of the latter.\n\n\n=80.= These facts may seem to give countenance to the general assertion\nthat to think a thing good is to prefer it or approve it, in the\nsense in which preference and approval denote certain kinds of will\nor feeling. It seems to be always true that when we thus prefer or\napprove, there is included in that fact the fact that we think good;\nand it is certainly true, in an immense majority of instances, that\nwhen we think good, we also prefer or approve. It is natural enough,\nthen, to say that to think good is to prefer. And what more natural\nthan to add: When I say a thing is good, I _mean_ that I prefer it?\nAnd yet this natural addition involves a gross confusion. Even if it\nbe true that to think good is the same thing as to prefer (which, as\nwe have seen, is _never_ true in the sense that they are absolutely\nidentical; and not _always_ true, even in the sense that they occur\ntogether), yet it is not true that _what_ you think, when you think a\nthing good, is _that_ you prefer it. Even if your thinking the thing\ngood is the same thing as your preference of it, yet the goodness\nof the thing--that _of_ which you think--is, for that very reason,\nobviously _not_ the same thing as your preference of it. Whether you\nhave a certain thought or not is one question; and whether what you\nthink is true is quite a different one, upon which the answer to the\nfirst has not the least bearing. The fact that you prefer a thing does\nnot tend to shew that the thing is good; even if it does shew that you\nthink it so.\n\nIt seems to be owing to this confusion, that the question \u2018What\nis good?\u2019 is thought to be identical with the question \u2018What is\npreferred?\u2019 It is said, with sufficient truth, that you would never\nknow a thing was good unless you preferred it, just as you would never\nknow a thing existed unless you perceived it. But it is added, and this\nis false, that you would never know a thing was good unless you _knew_\nthat you preferred it, or that it existed unless you _knew_ that you\nperceived it. And it is finally added, and this is utterly false, that\nyou cannot distinguish the fact that a thing is good from the fact\nthat you prefer it, or the fact that it exists from the fact that you\nperceive it. It is often pointed out that I cannot at any given moment\ndistinguish what is true from what I think so: and this is true. But\nthough I cannot distinguish _what_ is true from _what_ I think so, I\nalways can distinguish what I mean by saying _that_ it is true from\nwhat I mean by saying _that_ I think so. For I understand the meaning\nof the supposition that what I think true may nevertheless be false.\nWhen, therefore, I assert that it is true I mean to assert something\ndifferent from the fact that I think so. _What_ I think, namely _that_\nsomething is true, is always quite distinct from the fact that I think\nit. The assertion that it is true does not even _include_ the assertion\nthat I think it so; although, of course, whenever I do think a thing\ntrue, it is, as a matter of fact, also true that I do think it. This\ntautologous proposition that for a thing to be thought true it is\nnecessary that it should be thought, is, however, commonly identified\nwith the proposition that for a thing to _be_ true it is necessary\nthat it should be thought. A very little reflection should suffice to\nconvince anyone that this identification is erroneous; and a very\nlittle more will shew that, if so, we must mean by \u2018true\u2019 something\nwhich includes no reference to thinking or to any other psychical fact.\nIt may be difficult to discover precisely _what_ we mean--to hold the\nobject in question before us, so as to compare it with other objects:\nbut that we do mean something distinct and unique can no longer be\nmatter of doubt. That \u2018to be true\u2019 _means_ to be thought in a certain\nway is, therefore, certainly false. Yet this assertion plays the most\nessential part in Kant\u2019s \u2018Copernican revolution\u2019 of philosophy, and\nrenders worthless the whole mass of modern literature, to which that\nrevolution has given rise, and which is called Epistemology. Kant held\nthat what was unified in a certain manner by the synthetic activity of\nthought was _ipso facto_ true: that this was the very meaning of the\nword. Whereas it is plain that the only connection which can possibly\nhold between being true and being thought in a certain way, is that\nthe latter should be a _criterion_ or test of the former. In order,\nhowever, to establish that it is so, it would be necessary to establish\nby the methods of induction that what was true was always thought in a\ncertain way. Modern Epistemology dispenses with this long and difficult\ninvestigation at the cost of the self-contradictory assumption that\n\u2018truth\u2019 and the criterion of truth are one and the same thing.\n\n\n=81.= It is, then, a very natural, though an utterly false supposition\nthat for a thing to _be_ true is the same thing as for it to be\nperceived or thought of in a certain way. And since, for the reasons\ngiven above, the fact of preference seems roughly to stand in the same\nrelation to thinking things good, in which the fact of perception\nstands to thinking that they are true or exist, it is very natural that\nfor a thing to _be_ good should be supposed identical with its being\npreferred in a certain way. But once this coordination of Volition and\nCognition has been accepted, it is again very natural that every fact\nwhich seems to support the conclusion that being true is identical with\nbeing cognised should confirm the corresponding conclusion that being\ngood is identical with being willed. It will, therefore, be in place to\npoint out another confusion, which seems to have had great influence\nin causing acceptance of the view that to be true is the same thing as\nto be cognised.\n\nThis confusion is due to a failure to observe that when we say we have\na _sensation_ or _perception_ or that we _know_ a thing, we mean to\nassert not only that our mind is cognitive, but _also_ that that which\nit cognises is true. It is not observed that the usage of these words\nis such that, if a thing be untrue, that fact alone is sufficient to\njustify us in saying that the person who says he perceives or knows it,\ndoes not _perceive_ or _know_ it, without our either enquiring whether,\nor assuming that, his state of mind differs in any respect from what\nit would have been had he perceived or known. By this denial we do not\naccuse him of an error in introspection, even if there was such an\nerror: we do not deny that he was aware of a certain object, nor even\nthat his state of mind was exactly such as he took it to be: we merely\ndeny that the object, of which he was aware, had a certain property.\nIt is, however, commonly supposed that when we assert a thing to be\nperceived or known, we are asserting one fact only; and since of the\ntwo facts which we really assert, the existence of a psychical state is\nby far the easier to distinguish, it is supposed that this is the only\none which we do assert. Thus perception and sensation have come to be\nregarded as if they denoted certain states of mind and nothing more; a\nmistake which was the easier to make since the commonest state of mind,\nto which we give a name which does not imply that its object is true,\nnamely imagination, may, with some plausibility, be supposed to differ\nfrom sensation and perception not only in the property possessed by\nits object, but also in its character as a state of mind. It has thus\ncome to be supposed that the only difference between perception and\nimagination, by which they can be defined, must be a merely psychical\ndifference: and, if this were the case, it would follow at once that\nto _be_ true was identical with being cognised in a certain way; since\nthe assertion that a thing is perceived does certainly _include_ the\nassertion that it is true, and if, nevertheless, that it is perceived\nmeans _only_ that the mind has a certain attitude towards it, then\nits truth must be identical with the fact that it is regarded in this\nway. We may, then, attribute the view that to be true _means_ to\nbe cognised in a certain way partly to the failure to perceive that\ncertain words, which are commonly supposed to stand for nothing more\nthan a certain kind of cognitive state, do, in fact, _also_ include a\nreference to the truth of the object of such states.\n\n\n=82.= I will now sum up my account of the apparent connections between\nwill and ethical propositions, which seem to support the vague\nconviction that \u2018This is good\u2019 is somehow identical with \u2018This is\nwilled in a certain way.\u2019 (1) It may be maintained, with sufficient\nshow of truth, that it is only because certain things were originally\nwilled, that we ever came to have ethical convictions at all. And it\nhas been too commonly assumed that to shew what was the cause of a\nthing is the same thing as to shew what the thing itself is. It is,\nhowever, hardly necessary to point out that this is not the case. (2)\nIt may be further maintained, with some plausibility, that to think a\nthing good and to will it in a certain way are _now_ as a matter of\nfact identical. We must, however, distinguish certain possible meanings\nof this assertion. It may be admitted that when we think a thing good,\nwe _generally_ have a special attitude of will or feeling towards it;\nand that, perhaps, when we will it in a certain way, we do always think\nit good. But the very fact that we can thus distinguish the question\nwhether, though the one is always accompanied by the other, yet this\nother may not always be accompanied by the first, shews that the two\nthings are not, in the strict sense, identical. The fact is that,\nwhatever we mean by will, or by any form of will, the fact we mean by\nit certainly always includes something else _beside_ the thinking a\nthing good: and hence that, when willing and thinking good are asserted\nto be identical, the most that can be meant is that this other element\nin will always both accompanies and is accompanied by the thinking\ngood; and this, as has been said, is of very doubtful truth. Even,\nhowever, if it were strictly true, the fact that the two things can be\ndistinguished is fatal to the assumed coordination between will and\ncognition, in one of the senses in which that assumption is commonly\nmade. For it is only in respect of the _other_ element in will, that\nvolition differs from cognition; whereas it is only in respect of the\nfact that volition, or some form of volition, _includes_ a _cognition_\nof goodness, that will can have the same relation to ethical, which\ncognition has to metaphysical, propositions. Accordingly the fact of\nvolition, _as a whole_, that is, if we include in it the element which\nmakes it volition and distinguishes it from cognition, has _not_ the\nsame relation to ethical propositions which cognition has to those\nwhich are metaphysical. Volition and cognition are _not_ coordinate\nways of experiencing, since it is only in so far as volition denotes\na _complex_ fact, which includes in it the one identical simple fact,\nwhich is meant by _cognition_, that volition is a way of experiencing\nat all.\n\nBut, (3) if we allow the terms \u2018volition\u2019 or \u2018will\u2019 to stand for\n\u2018thinking good,\u2019 although they certainly do not commonly stand for\nthis, there still remains the question: What connection would this fact\nestablish between volition and Ethics? Could the enquiry into what was\nwilled be identical with the ethical enquiry into what was good? It\nis plain enough that they could not be identical; though it is also\nplain why they should be thought so. The question \u2018What is good?\u2019 is\nconfused with the question \u2018What is thought good?\u2019 and the question\n\u2018What is true?\u2019 with the question \u2018What is thought true?\u2019 for two main\nreasons. (1) One of these is the general difficulty that is found\nin distinguishing what is cognised from the cognition of it. It is\nobserved that I certainly cannot cognise anything that is true without\ncognising it. Since, therefore, whenever I know a thing that is true,\nthe thing is certainly cognised, it is assumed that for a thing to _be_\ntrue at all is the same thing as for it to be cognised. And (2) it is\nnot observed that certain words, which are supposed to denote only\npeculiar species of cognition, do as a matter of fact _also_ denote\nthat the object cognised is true. Thus if \u2018perception\u2019 be taken to\ndenote only a certain kind of mental fact, then, since the object of it\nis always true, it becomes easy to suppose that to be true means only\nto be object to a mental state of that kind. And similarly it is easy\nto suppose that to be truly good differs from being falsely thought\nso, solely in respect of the fact that to be the former is to be the\nobject of a volition differing from that of which an apparent good is\nthe object, in the same way in which a perception (on this supposition)\ndiffers from an illusion.\n\n\n=83.= Being good, then, is not identical with being willed or felt\nin any kind of way, any more than being true is identical with being\nthought in any kind of way. But let us suppose this to be admitted: Is\nit still possible that an enquiry into the nature of will or feeling\nshould be a necessary step to the proof of ethical conclusions? If\nbeing good and being willed are _not_ identical, then the most that can\nbe maintained with regard to the connection of goodness with will is\nthat what is good is always _also_ willed in a certain way, and that\nwhat is willed in a certain way is always _also_ good. And it may be\nsaid that this is all that is meant by those metaphysical writers who\nprofess to base Ethics upon the Metaphysics of Will. What would follow\nfrom this supposition?\n\nIt is plain that if what is willed in a certain way were always _also_\ngood, then the fact that a thing was so willed would be a _criterion_\nof its goodness. But in order to establish that will is a criterion\nof goodness, we must be able to shew first and separately that in a\ngreat number of the instances in which we find a certain kind of will\nwe also find that the objects of that will are good. We might, then,\nperhaps, be entitled to infer that in a few instances, where it was not\nobvious whether a thing was good or not but was obvious that it was\nwilled in the way required, the thing was really good, since it had the\nproperty which in all other instances we had found to be accompanied by\ngoodness. A reference to will might thus, just conceivably, become of\nuse towards the end of our ethical investigations, when we had already\nbeen able to shew, independently, of a vast number of different objects\nthat they were really good and in what degree they were so. And against\neven this conceivable utility it may be urged (1) That it is impossible\nto see why it should not be as easy (and it would certainly be the more\nsecure way) to prove that the thing in question was good, by the same\nmethods which we had used in proving that other things were good, as by\nreference to our criterion; and (2) That, if we set ourselves seriously\nto find out what things are good, we shall see reason to think (as\nwill appear in Chapter VI.) that they have _no_ other property, both\ncommon and peculiar to them, beside their goodness--that, in fact,\nthere is no criterion of goodness.\n\n\n=84.= But to consider whether any form of will is or is not a\ncriterion of goodness is quite unnecessary for our purpose here;\nsince none of those writers who profess to base their Ethics on\nan investigation of will have ever recognised the need of proving\ndirectly and independently that all the things which are willed in\na certain way are good. They make no attempt to shew that will is a\n_criterion_ of goodness; and no stronger evidence could be given that\nthey do not recognise that this, at most, is all it can be. As has\nbeen just pointed out, if we are to maintain that whatever is willed\nin a certain way is also good, we must in the first place be able to\nshew that certain things have one property \u2018goodness,\u2019 and that the\nsame things _also_ have the other property that they are willed in a\ncertain way. And secondly we must be able to shew this in a very large\nnumber of instances, if we are to be entitled to claim any assent\nfor the proposition that these two properties _always_ accompany one\nanother: even when this was shewn it would still be doubtful whether\nthe inference from \u2018generally\u2019 to \u2018always\u2019 would be valid, and almost\ncertain that this doubtful principle would be useless. But the very\nquestion which it is the business of Ethics to answer is this question\nwhat things are good; and, so long as Hedonism retains its present\npopularity, it must be admitted that it is a question upon which there\nis scarcely any agreement and which therefore requires the most careful\nexamination. The greatest and most difficult part of the business\nof Ethics would therefore require to have been already accomplished\nbefore we could be entitled to claim that anything was a _criterion_\nof goodness. If, on the other hand, to be willed in a certain way\nwas _identical_ with being good, then indeed we should be entitled\nto start our ethical investigations by enquiring what was willed in\nthe way required. That this is the way in which metaphysical writers\nstart their investigations seems to shew conclusively that they are\ninfluenced by the idea that \u2018goodness\u2019 is _identical_ with \u2018being\nwilled.\u2019 They do not recognise that the question \u2018What is good?\u2019 is a\n_different_ one from the question \u2018What is willed in a certain way?\u2019\nThus we find Green explicitly stating that \u2018_the_ common characteristic\nof the good is that it satisfies some desire[22].\u2019 If we are to take\nthis statement strictly, it obviously asserts that good things have no\ncharacteristic in common, except that they satisfy some desire--not\neven, therefore, that they are good. And this can only be the case,\nif being good is _identical_ with satisfying desire: if \u2018good\u2019 is\nmerely another name for \u2018desire-satisfying.\u2019 There could be no plainer\ninstance of the naturalistic fallacy. And we cannot take the statement\nas a mere verbal slip, which does not affect the validity of Green\u2019s\nmain argument. For he nowhere either gives or pretends to give any\nreason for believing anything to be good in any sense, except that\nit is what would satisfy a particular kind of desire--the kind of\ndesire which he tries to shew to be that of a moral agent. An unhappy\nalternative is before us. Such reasoning would give valid reasons for\nhis conclusions, if, and only if, being good and being desired in a\nparticular way were identical: and in this case, as we have seen in\nChapter I., his conclusions would not be ethical. On the other hand,\nif the two are not identical, his conclusions may be ethical and may\neven be right, but he has not given us a single reason for believing\nthem. The thing which a scientific Ethics is required to shew, namely\nthat certain things are really good, he has assumed to begin with, in\nassuming that things which are willed in a certain way are always good.\nWe may, therefore, have as much respect for Green\u2019s conclusions as for\nthose of any other man who details to us his ethical convictions: but\nthat any of his arguments are such as to give us any reason for holding\nthat Green\u2019s convictions are more likely to be true than those of any\nother man, must be clearly denied. The _Prolegomena to Ethics_ is quite\nas far as Mr Spencer\u2019s _Data of Ethics_, from making the smallest\ncontribution to the solution of ethical problems.\n\n  [22] _Prolegomena to Ethics_, p. 178.\n\n\n=85.= The main object of this chapter has been to shew that\nMetaphysics, understood as the investigation of a supposed\nsupersensible reality, can have no logical bearing whatever upon the\nanswer to the fundamental ethical question \u2018What is good in itself?\u2019\nThat this is so, follows at once from the conclusion of Chapter I.,\nthat \u2018good\u2019 denotes an ultimate, unanalysable predicate; but this truth\nhas been so systematically ignored, that it seemed worth while to\ndiscuss and distinguish, in detail, the principal relations, which do\nhold, or have been supposed to hold, between Metaphysics and Ethics.\nWith this view I pointed out:--(1) That Metaphysics may have a bearing\non _practical_ Ethics--on the question \u2018What ought we to do?\u2019--so far\nas it may be able to tell us what the future effects of our action\nwill be: what it can _not_ tell us is whether those effects are good\nor bad in themselves. One particular type of metaphysical doctrine,\nwhich is very frequently held, undoubtedly has such a bearing on\n_practical_ Ethics: for, if it is true that the sole reality is an\neternal, immutable Absolute, then it follows that no actions of ours\ncan have any real effect, and hence that no _practical_ proposition\ncan be true. The same conclusion follows from the ethical proposition,\ncommonly combined with this metaphysical one--namely that this eternal\nReality is also the sole good (68). (2) That metaphysical writers,\nas where they fail to notice the contradiction just noticed between\nany _practical_ proposition and the assertion that an eternal reality\nis the sole good, seem frequently to confuse the proposition that\none particular existing thing is good, with the proposition that the\nexistence of that kind of thing _would_ be good, wherever it might\noccur. To the proof of the former proposition Metaphysics might be\nrelevant, by shewing that the thing existed; to the proof of the latter\nit is wholly irrelevant: it can only serve the _psychological_ function\nof suggesting things which may be valuable--a function which would be\nstill better performed by pure fiction (69-71).\n\nBut the most important source of the supposition that Metaphysics is\nrelevant to Ethics, seems to be the assumption that \u2018good\u2019 _must_\ndenote some _real_ property of things--an assumption which is mainly\ndue to two erroneous doctrines, the first _logical_, the second\n_epistemological_. Hence (3) I discussed the _logical_ doctrine that\nall propositions assert a relation between existents; and pointed\nout that the assimilation of ethical propositions either to natural\nlaws or to commands are instances of this _logical_ fallacy (72-76).\nAnd finally (4) I discussed the _epistemological_ doctrine that to be\ngood is equivalent to being willed or felt in some particular way; a\ndoctrine which derives support from the analogous error, which Kant\nregarded as the cardinal point of his system and which has received\nimmensely wide acceptance--the erroneous view that to be \u2018true\u2019 or\n\u2018real\u2019 is equivalent to being thought in a particular way. In this\ndiscussion the main points to which I desire to direct attention\nare these: (_a_) That Volition and Feeling are _not_ analogous to\nCognition in the manner assumed; since in so far as these words denote\nan attitude of the mind towards an object, they are themselves merely\ninstances of Cognition: they differ only in respect of the kind of\nobject of which they take cognisance, and in respect of the other\nmental accompaniments of such cognitions: (_b_) That universally the\n_object_ of a cognition must be distinguished from the cognition of\nwhich it is the object; and hence that in no case can the question\nwhether the object is _true_ be identical with the question how it is\ncognised or whether it is cognised at all: it follows that even if the\nproposition \u2018This is good\u2019 were always the object of certain kinds of\nwill or feeling, the _truth_ of that proposition could in no case be\nestablished by proving that it was their object; far less can that\nproposition itself be identical with the proposition that its subject\nis the object of a volition or a feeling (77-84).\n\n\n\n\nCHAPTER V.\n\nETHICS IN RELATION TO CONDUCT.\n\n\n=86.= In the present chapter we have again to take a great step in\nethical method. My discussion hitherto has fallen under two main\nheads. Under the first, I tried to shew what \u2018good\u2019--the adjective\n\u2018good\u2019--_means_. This appeared to be the first point to be settled in\nany treatment of Ethics, that should aim at being systematic. It is\nnecessary we should know this, should know what good means, before we\ncan go on to consider what is good--what things or qualities are good.\nIt is necessary we should know it for two reasons. The first reason\nis that \u2018good\u2019 is the notion upon which all Ethics depends. We cannot\nhope to understand what we mean, when we say that this is good or that\nis good, until we understand quite clearly, not only what \u2018this\u2019 is or\n\u2018that\u2019 is (which the natural sciences and philosophy can tell us) but\nalso what is meant by calling them good, a matter which is reserved\nfor Ethics only. Unless we are quite clear on this point, our ethical\nreasoning will be always apt to be fallacious. We shall think that we\nare proving that a thing is \u2018good,\u2019 when we are really only proving\nthat it is something else; since unless we know what \u2018good\u2019 means,\nunless we know what is meant by that notion in itself, as distinct from\nwhat is meant by any other notion, we shall not be able to tell when\nwe are dealing with it and when we are dealing with something else,\nwhich is perhaps like it, but yet not the same. And the second reason\nwhy we should settle first of all this question \u2018What good means?\u2019\nis a reason of method. It is this, that we can never know on what\n_evidence_ an ethical proposition rests, until we know the nature of\nthe notion which makes the proposition ethical. We cannot tell what\nis possible, by way of proof, in favour of one judgment that \u2018This or\nthat is good,\u2019 or against another judgment \u2018That this or that is bad,\u2019\nuntil we have recognised what the nature of such propositions must\nalways be. In fact, it follows from the meaning of good and bad, that\nsuch propositions are all of them, in Kant\u2019s phrase, \u2018synthetic\u2019: they\nall must rest in the end upon some proposition which must be simply\naccepted or rejected, which cannot be logically deduced from any other\nproposition. This result, which follows from our first investigation,\nmay be otherwise expressed by saying that the fundamental principles\nof Ethics must be self-evident. But I am anxious that this expression\nshould not be misunderstood. The expression \u2018self-evident\u2019 means\nproperly that the proposition so called is evident or true, _by itself_\nalone; that it is not an inference from some proposition other than\n_itself_. The expression does _not_ mean that the proposition is\ntrue, because it is evident to you or me or all mankind, because in\nother words it appears to us to be true. That a proposition appears\nto be true can never be a valid argument that true it really is. By\nsaying that a proposition is self-evident, we mean emphatically that\nits appearing so to us, is _not_ the reason why it is true: for we\nmean that it has absolutely no reason. It would not be a self-evident\nproposition, if we could say of it: I cannot think otherwise and\ntherefore it is true. For then its evidence or proof would not lie in\nitself, but in something else, namely our conviction of it. That it\nappears true to us may indeed be the _cause_ of our asserting it, or\nthe reason why we think and say that it is true: but a reason in this\nsense is something utterly different from a logical reason, or reason\nwhy something is true. Moreover, it is obviously not a reason of the\nsame thing. The _evidence_ of a proposition to us is only a reason\nfor _our holding it_ to be true: whereas a logical reason, or reason\nin the sense in which self-evident propositions have no reason, is a\nreason why _the proposition itself_ must be true, not why we hold it\nso to be. Again that a proposition is evident to us may not only be\nthe reason why we do think or affirm it, it may even be a _reason_\nwhy we ought to think it or affirm it. But a reason, in this sense\ntoo, is not a logical reason for the truth of the proposition, though\nit is a logical reason for the rightness of holding the proposition.\nIn our common language, however, these three meanings of \u2018reason\u2019 are\nconstantly confused, whenever we say \u2018I have a reason for thinking that\ntrue.\u2019 But it is absolutely essential, if we are to get clear notions\nabout Ethics or, indeed, about any other, especially any philosophical,\nstudy, that we should distinguish them. When, therefore, I talk of\nIntuitionistic Hedonism, I must not be understood to imply that my\ndenial that \u2018Pleasure is the only good\u2019 is _based_ on my Intuition of\nits falsehood. My Intuition of its falsehood is indeed _my_ reason for\n_holding_ and declaring it untrue; it is indeed the only valid reason\nfor so doing. But that is just because there is _no_ logical reason\nfor it; because there is no proper evidence or reason of its falsehood\nexcept itself alone. It is untrue, because it is untrue, and there\nis no other reason: but I _declare_ it untrue, because its untruth\nis evident to me, and I hold that that is a sufficient reason for my\nassertion. We must not therefore look on Intuition, as if it were\nan alternative to reasoning. Nothing whatever can take the place of\n_reasons_ for the truth of any proposition: intuition can only furnish\na reason for _holding_ any proposition to be true: this however it must\ndo when any proposition is self-evident, when, in fact, there are no\nreasons which prove its truth.\n\n\n=87.= So much, then, for the first step in our ethical method, the\nstep which established that good is good and nothing else whatever,\nand that Naturalism was a fallacy. A second step was taken when we\nbegan to consider proposed self-evident principles of Ethics. In this\nsecond division, resting on our result that good means good, we began\nthe discussion of propositions asserting that such and such a thing\nor quality or concept was good. Of such a kind was the principle of\nIntuitionistic or Ethical Hedonism--the principle that \u2018Pleasure alone\nis good.\u2019 Following the method established by our first discussion,\nI claimed that the untruth of this proposition was self-evident. I\ncould do nothing to _prove_ that it was untrue; I could only point\nout as clearly as possible what it means, and how it contradicts other\npropositions which appear to be equally true. My only object in all\nthis was, necessarily, to convince. But even if I did convince, that\ndoes not prove that we are right. It justifies us in _holding_ that\nwe are so; but nevertheless we may be wrong. On one thing, however,\nwe may justly pride ourselves. It is that we have had a better chance\nof answering our question rightly, than Bentham or Mill or Sidgwick\nor others who have contradicted us. For we have _proved_ that these\nhave never even asked themselves the question which they professed to\nanswer. They have confused it with another question: small wonder,\ntherefore, if their answer is different from ours. We must be quite\nsure that the same question has been put, before we trouble ourselves\nat the different answers that are given to it. For all we know, the\nwhole world would agree with us, if they could once clearly understand\nthe question upon which we want their votes. Certain it is, that in\nall those cases where we found a difference of opinion, we found also\nthat the question had _not_ been clearly understood. Though, therefore,\nwe cannot prove that we are right, yet we have reason to believe that\neverybody, unless he is mistaken as to what he thinks, will think the\nsame as we. It is as with a sum in mathematics. If we find a gross and\npalpable error in the calculations, we are not surprised or troubled\nthat the person who made this mistake has reached a different result\nfrom ours. We think he will admit that his result is wrong, if his\nmistake is pointed out to him. For instance if a man has to add up 5\n+ 7 + 9, we should not wonder that he made the result to be 34, if he\nstarted by making 5 + 7 = 25. And so in Ethics, if we find, as we did,\nthat \u2018desirable\u2019 is confused with \u2018desired,\u2019 or that \u2018end\u2019 is confused\nwith \u2018means,\u2019 we need not be disconcerted that those who have committed\nthese mistakes do not agree with us. The only difference is that in\nEthics, owing to the intricacy of its subject-matter, it is far more\ndifficult to persuade anyone either that he has made a mistake or that\nthat mistake affects his result.\n\nIn this second division of my subject--the division which is occupied\nwith the question, \u2018What is good in itself?\u2019--I have hitherto only\ntried to establish one definite result, and that a negative one: namely\nthat pleasure is _not_ the sole good. This result, if true, refutes\nhalf, or more than half, of the ethical theories which have ever been\nheld, and is, therefore, not without importance. It will, however, be\nnecessary presently to deal positively with the question: What things\nare good and in what degrees?\n\n\n=88.= But before proceeding to this discussion I propose, first, to\ndeal with the _third_ kind of ethical question--the question: What\nought we to do?\n\nThe answering of this question constitutes the third great division\nof ethical enquiry; and its nature was briefly explained in Chap. I.\n(\u00a7\u00a7 15-17). It introduces into Ethics, as was there pointed out, an\nentirely new question--the question what things are related as _causes_\nto that which is good in itself; and this question can only be answered\nby an entirely new method--the method of empirical investigation; by\nmeans of which causes are discovered in the other sciences. To ask what\nkind of actions we ought to perform, or what kind of conduct is right,\nis to ask what kind of effects such action and conduct will produce.\nNot a single question in practical Ethics can be answered except by a\ncausal generalisation. All such questions do, indeed, _also_ involve\nan ethical judgment proper--the judgment that certain effects are\nbetter, in themselves, than others. But they _do_ assert that these\nbetter things are effects--are causally connected with the actions in\nquestion. Every judgment in practical Ethics may be reduced to the\nform: This is a cause of that good thing.\n\n\n=89.= That this is the case, that the questions, What is right? what is\nmy duty? what ought I to do? belong exclusively to this third branch of\nethical enquiry, is the first point to which I wish to call attention.\nAll moral laws, I wish to shew, are merely statements that certain\nkinds of actions will have good effects. The very opposite of this view\nhas been generally prevalent in Ethics. \u2018The right\u2019 and \u2018the useful\u2019\nhave been supposed to be at least _capable_ of conflicting with one\nanother, and, at all events, to be essentially distinct. It has been\ncharacteristic of a certain school of moralists, as of moral common\nsense, to declare that the end will never justify the means. What I\nwish first to point out is that \u2018right\u2019 does and can mean nothing but\n\u2018cause of a good result,\u2019 and is thus identical with \u2018useful\u2019; whence\nit follows that the end always will justify the means, and that no\naction which is not justified by its results can be right. That there\nmay be a true proposition, meant to be conveyed by the assertion \u2018The\nend will not justify the means,\u2019 I fully admit: but that, in another\nsense, and a sense far more fundamental for ethical theory, it is\nutterly false, must first be shewn.\n\nThat the assertion \u2018I am morally bound to perform this action\u2019 is\nidentical with the assertion \u2018This action will produce the greatest\npossible amount of good in the Universe\u2019 has already been briefly shewn\nin Chap. I. (\u00a7 17); but it is important to insist that this fundamental\npoint is demonstrably certain. This may, perhaps, be best made evident\nin the following way. It is plain that when we assert that a certain\naction is our absolute duty, we are asserting that the performance of\nthat action at that time is unique in respect of value. But no dutiful\naction can possibly have unique value in the sense that it is the sole\nthing of value in the world; since, in that case, _every_ such action\nwould be the _sole_ good thing, which is a manifest contradiction.\nAnd for the same reason its value cannot be unique in the sense that\nit has more intrinsic value than anything else in the world; since\n_every_ act of duty would then be the _best_ thing in the world, which\nis also a contradiction. It can, therefore, be unique only in the sense\nthat the whole world will be better, if it be performed, than if any\npossible alternative were taken. And the question whether this is so\ncannot possibly depend solely on the question of its own intrinsic\nvalue. For any action will also have effects different from those of\nany other action; and if any of these have intrinsic value, their value\nis exactly as relevant to the total goodness of the Universe as that of\ntheir cause. It is, in fact, evident that, however valuable an action\nmay be in itself, yet, owing to its existence, the sum of good in the\nUniverse may conceivably be made less than if some other action, less\nvaluable in itself, had been performed. But to say that this is the\ncase is to say that it would have been better that the action should\nnot have been done; and this again is obviously equivalent to the\nstatement that it ought not to have been done--that it was not what\nduty required. \u2018Fiat iustitia, ruat caelum\u2019 can only be justified on\nthe ground that by the doing of justice the Universe gains more than it\nloses by the falling of the heavens. It is, of course, possible that\nthis is the case: but, at all events, to assert that justice _is_ a\nduty, in spite of such consequences, is to assert that it is the case.\n\nOur \u2018duty,\u2019 therefore, can only be defined as that action, which will\ncause more good to exist in the Universe than any possible alternative.\nAnd what is \u2018right\u2019 or \u2018morally permissible\u2019 only differs from this, as\nwhat will _not_ cause _less_ good than any possible alternative. When,\ntherefore, Ethics presumes to assert that certain ways of acting are\n\u2018duties\u2019 it presumes to assert that to act in those ways will always\nproduce the greatest possible sum of good. If we are told that to \u2018do\nno murder\u2019 is a duty, we are told that the action, whatever it may be,\nwhich is called murder, will under no circumstances cause so much good\nto exist in the Universe as its avoidance.\n\n\n=90.= But, if this be recognised, several most important consequences\nfollow, with regard to the relation of Ethics to conduct.\n\n(1) It is plain that no moral law is self-evident, as has commonly been\nheld by the Intuitional school of moralists. The Intuitional view of\nEthics consists in the supposition that certain rules, stating that\ncertain actions are always to be done or to be omitted, may be taken as\nself-evident premisses. I have shewn with regard to judgments of what\nis _good in itself_, that this is the case; no reason can be given for\nthem. But it is the essence of Intuitionism to suppose that rules of\naction--statements not of what ought to _be_, but of what we ought to\ndo--are in the same sense intuitively certain. Plausibility has been\nlent to this view by the fact that we do undoubtedly make immediate\njudgments that certain actions are obligatory or wrong: we are thus\noften intuitively certain of our duty, _in a psychological sense_. But,\nnevertheless, these judgments are not self-evident and cannot be taken\nas ethical premisses, since, as has now been shewn, they are capable of\nbeing confirmed or refuted by an investigation of causes and effects.\nIt is, indeed, possible that some of our immediate intuitions are true;\nbut since _what_ we intuit, _what_ conscience tells us, is that certain\nactions will always produce the greatest sum of good possible under the\ncircumstances, it is plain that reasons can be given, which will shew\nthe deliverances of conscience to be true or false.\n\n\n=91.= (2) In order to shew that any action is a duty, it is necessary\nto know both what are the other conditions, which will, conjointly with\nit, determine its effects; to know exactly what will be the effects of\nthese conditions; and to know all the events which will be in any way\naffected by our action throughout an infinite future. We must have all\nthis causal knowledge, and further we must know accurately the degree\nof value both of the action itself and of all these effects; and must\nbe able to determine how, in conjunction with the other things in the\nUniverse, they will affect its value as an organic whole. And not\nonly this: we must also possess all this knowledge with regard to the\neffects of every possible alternative; and must then be able to see by\ncomparison that the total value due to the existence of the action in\nquestion will be greater than that which would be produced by any of\nthese alternatives. But it is obvious that our causal knowledge alone\nis far too incomplete for us ever to assure ourselves of this result.\nAccordingly it follows that we never have any reason to suppose that an\naction is our duty: we can never be sure that any action will produce\nthe greatest value possible.\n\nEthics, therefore, is quite unable to give us a list of duties:\nbut there still remains a humbler task which may be possible for\nPractical Ethics. Although we cannot hope to discover which, in a given\nsituation, is the best of all possible alternative actions, there may\nbe some possibility of shewing which among the alternatives, _likely to\noccur to any one_, will produce the greatest sum of good. This second\ntask is certainly all that Ethics can ever have accomplished: and it is\ncertainly all that it has ever collected materials for proving; since\nno one has ever attempted to exhaust the possible alternative actions\nin any particular case. Ethical philosophers have in fact confined\ntheir attention to a very limited class of actions, which have been\nselected because they are those which most commonly occur to mankind\nas possible alternatives. With regard to these they may possibly have\nshewn that one alternative is better, _i.e._ produces a greater total\nof value, than others. But it seems desirable to insist, that though\nthey have represented this result as a determination of _duties_, it\ncan never really have been so. For the term duty is certainly so used\nthat, if we are subsequently persuaded that any possible action would\nhave produced more good than the one we adopted, we admit that we\nfailed to do our duty. It will, however, be a useful task if Ethics can\ndetermine which among alternatives _likely to occur_ will produce the\ngreatest total value. For, though this alternative cannot be proved to\nbe the best possible, yet it may be better than any course of action\nwhich we should otherwise adopt.\n\n\n=92.= A difficulty in distinguishing this task, which Ethics may\nperhaps undertake with some hope of success, from the hopeless task\nof finding duties, arises from an ambiguity in the use of the term\n\u2018possible.\u2019 An action may, in one perfectly legitimate sense, be said\nto be \u2018impossible\u2019 solely because the idea of doing it does not occur\nto us. In this sense, then, the alternatives which do actually occur\nto a man would be the only _possible_ alternatives; and the best of\nthese would be the best possible action under the circumstances, and\nhence would conform to our definition of \u2018duty.\u2019 But when we talk of\nthe best _possible_ action as our duty, we mean by the term any action\nwhich no _other_ known circumstance would prevent, _provided_ the idea\nof it occurred to us. And this use of the term is in accordance with\npopular usage. For we admit that a man may fail to do his duty, through\nneglecting to think of what he _might_ have done. Since, therefore,\nwe say that he _might_ have done, what nevertheless did not occur to\nhim, it is plain that we do not limit his _possible_ actions to those\nof which he thinks. It might be urged, with more plausibility, that we\nmean by a man\u2019s duty only the best of those actions of which he _might_\nhave thought. And it is true that we do not blame any man very severely\nfor omitting an action of which, as we say, \u2018he could not be expected\nto think.\u2019 But even here it is plain that we recognise a distinction\nbetween what he might have done and what he might have thought of\ndoing: we regard it as a pity that he did not do otherwise. And \u2018duty\u2019\nis certainly used in such a sense, that it would be a contradiction in\nterms to say it was a pity that a man did his duty.\n\nWe must, therefore, distinguish a possible action from an action of\nwhich it is possible to think. By the former we mean an action which\nno known cause would prevent, _provided_ the idea of it occurred to\nus: and that one among such actions, which will produce the greatest\ntotal good, is what we mean by duty. Ethics certainly cannot hope to\ndiscover what kind of action is always our duty in this sense. It may,\nhowever, hope to decide which among one or two such possible actions is\nthe best: and those which it has chosen to consider are, as a matter of\nfact, the most important of those with regard to which men deliberate\nwhether they shall or shall not do them. A decision with regard to\nthese may therefore be easily confounded with a decision with regard\nto which is the best possible action. But it is to be noted that even\nthough we limit ourselves to considering which is the better among\nalternatives likely to be thought of, the fact that these alternatives\nmight be thought of is not included is what we mean by calling them\npossible alternatives. Even if in any particular case it was impossible\nthat the idea of them should have occurred to a man, the question we\nare concerned with is, which, if it had occurred, would have been the\nbest alternative? If we say that murder is always a worse alternative,\nwe mean to assert that it is so, even where it was impossible for the\nmurderer to think of doing anything else.\n\nThe utmost, then, that Practical Ethics can hope to discover is which,\namong a few alternatives possible under certain circumstances, will, on\nthe whole, produce the best result. It may tell us which is the best,\nin this sense, of certain alternatives about which we are likely to\ndeliberate; and since we may also know that, even if we choose none of\nthese, what we shall, in that case, do is unlikely to be as good as one\nof them, it may thus tell us which of the alternatives, among which\nwe _can_ choose, it is best to choose. If it could do this it would be\nsufficient for practical guidance.\n\n\n=93.= But (3) it is plain that even this is a task of immense\ndifficulty. It is difficult to see how we can establish even a\nprobability that by doing one thing we shall obtain a better total\nresult than by doing another. I shall merely endeavour to point out how\nmuch is assumed, when we assume that there is such a probability, and\non what lines it seems possible that this assumption may be justified.\nIt will be apparent that it has never yet been justified--that no\nsufficient reason has ever yet been found for considering one action\nmore right or more wrong than another.\n\n(_a_) The first difficulty in the way of establishing a probability\nthat one course of action will give a better total result than another,\nlies in the fact that we have to take account of the effects of both\nthroughout an infinite future. We have no certainty but that, if we\ndo one action now, the Universe will, throughout all time, differ in\nsome way from what it would have been, if we had done another; and,\nif there is such a permanent difference, it is certainly relevant to\nour calculation. But it is quite certain that our causal knowledge is\nutterly insufficient to tell us what different effects will probably\nresult from two different actions, except within a comparatively short\nspace of time; we can certainly only pretend to calculate the effects\nof actions within what may be called an \u2018immediate\u2019 future. No one,\nwhen he proceeds upon what he considers a rational consideration of\neffects, would guide his choice by any forecast that went beyond a few\ncenturies at most; and, in general, we consider that we have acted\nrationally, if we think we have secured a balance of good within a few\nyears or months or days. Yet, if a choice guided by such considerations\nis to be rational, we must certainly have some reason to believe that\nno consequences of our action in a further future will generally be\nsuch as to reverse the balance of good that is probable in the future\nwhich we can foresee. This large postulate must be made, if we are\never to assert that the results of one action will be even probably\nbetter than those of another. Our utter ignorance of the far future\ngives us no justification for saying that it is even probably right\nto choose the greater good within the region over which a probable\nforecast may extend. We do, then, assume that it is improbable that\neffects, after a certain time, will, in general, be such as to reverse\nthe comparative value of the alternative results within that time. And\nthat this assumption is justified must be shewn before we can claim\nto have given any reason whatever for acting in one way rather than\nin another. It may, perhaps, be justified by some such considerations\nas the following. As we proceed further and further from the time at\nwhich alternative actions are open to us, the events of which either\naction would be part cause become increasingly dependent on those\nother circumstances, which are the same, whichever action we adopt.\nThe effects of any individual action seem, after a sufficient space of\ntime, to be found only in trifling modifications spread over a very\nwide area, whereas its immediate effects consist in some prominent\nmodification of a comparatively narrow area. Since, however, most\nof the things which have any great importance for good or evil are\nthings of this prominent kind, there may be a probability that after\na certain time all the effects of any particular action become so\nnearly indifferent, that any difference between their value and that\nof the effects of another action, is very unlikely to outweigh an\nobvious difference in the value of the immediate effects. It does in\nfact appear to be the case that, in most cases, whatever action we now\nadopt, \u2018it will be all the same a hundred years hence,\u2019 so far as the\nexistence at that time of anything greatly good or bad is concerned:\nand this might, perhaps, be _shewn_ to be true, by an investigation\nof the manner in which the effects of any particular event become\nneutralised by lapse of time. Failing such a proof, we can certainly\nhave no rational ground for asserting that one of two alternatives is\neven probably right and another wrong. If any of our judgments of right\nand wrong are to pretend to probability, we must have reason to think\nthat the effects of our actions in the far future will not have value\nsufficient to outweigh any superiority of one set of effects over\nanother in the immediate future.\n\n\n=94.= (_b_) We must assume, then, that if the effects of one action are\ngenerally better than those of another, so far forward in the future as\nwe are able to foresee any probable difference in their effects at all,\nthen the total effect upon the Universe of the former action is also\ngenerally better. We certainly cannot hope directly to compare their\neffects except within a limited future; and all the arguments, which\nhave ever been used in Ethics, and upon which we commonly act in common\nlife, directed to shewing that one course is superior to another, are\n(apart from theological dogmas) confined to pointing out such probable\nimmediate advantages. The question remains, then: Can we lay down any\ngeneral rules to the effect that one among a few alternative actions\nwill generally produce a greater total of good in the immediate future?\n\nIt is important to insist that this question, limited as it is, is the\nutmost, to which, with any knowledge we have at present or are likely\nto have for a long time to come, Practical Ethics can hope to give an\nanswer. I have already pointed out that we cannot hope to discover\nwhich is the _best_ possible alternative in any given circumstances,\nbut only which, among a few, is better than the others. And I have also\npointed out that there is certainly no more than a probability, even\nif we are entitled to assert so much, that what is better in regard to\nits immediate effects will also be better on the whole. It now remains\nto insist that, even with regard to these immediate effects, we can\nonly hope to discover which, among a few alternatives, will _generally_\nproduce the greatest balance of good in the immediate future. We can\nsecure no title to assert that obedience to such commands as \u2018Thou\nshalt not lie,\u2019 or even \u2018Thou shalt do no murder,\u2019 is _universally_\nbetter than the alternatives of lying and murder. Reasons why no more\nthan a _general_ knowledge is possible have been already given in Chap.\nI. (\u00a7 16); but they may be recapitulated here. In the first place,\nof the effects, which principally concern us in ethical discussions,\nas having intrinsic value, we know the causes so little, that we can\nscarcely claim, with regard to any single one, to have obtained even\na _hypothetical_ universal law, such as has been obtained in the exact\nsciences. We cannot even say: If this action is performed, under\nexactly these circumstances, and if no others interfere, this important\neffect, at least, will _always_ be produced. But, in the second\nplace, an ethical law is not merely hypothetical. If we are to know\nthat it will always be better to act in a certain way, under certain\ncircumstances, we must know not merely what effects such actions will\nproduce, _provided_ no other circumstances interfere, but also that no\nother circumstances will interfere. And this it is obviously impossible\nto know with more than probability. An ethical law has the nature not\nof a scientific law but of a scientific _prediction_: and the latter is\nalways merely probable, although the probability may be very great. An\nengineer is entitled to assert that, if a bridge be built in a certain\nway, it will probably bear certain loads for a certain time; but he can\nnever be absolutely certain that it has been built in the way required,\nnor that, even if it has, some accident will not intervene to falsify\nhis prediction. With any ethical law, the same must be the case; it can\nbe no more than a generalisation: and here, owing to the comparative\nabsence of accurate hypothetical knowledge, on which the prediction\nshould be based, the probability is comparatively small. But finally,\nfor an ethical generalisation, we require to know not only what effects\nwill be produced, but also what are the comparative values of those\neffects; and on this question too, it must be admitted, considering\nwhat a prevalent opinion Hedonism has been, that we are very liable to\nbe mistaken. It is plain, then, that we are not soon likely to know\nmore than that one kind of action will _generally_ produce better\neffects than another; and that more than this has certainly never been\nproved. In no two cases will _all_ the effects of any kind of action be\nprecisely the same, because in each case some of the circumstances will\ndiffer; and although the effects, that are important for good or evil,\nmay be generally the same, it is extremely unlikely that they will\nalways be so.\n\n\n=95.= (_c_) If, now, we confine ourselves to a search for actions which\nare _generally_ better as means than any probable alternative, it\nseems possible to establish as much as this in defence of most of the\nrules most universally recognised by Common Sense. I do not propose to\nenter upon this defence in detail, but merely to point out what seem to\nbe the chief distinct principles by the use of which it can be made.\n\nIn the first place, then, we can only shew that one action is\ngenerally better than another as a means, provided that certain other\ncircumstances are given. We do, as a matter of fact, only observe its\ngood effects under certain circumstances; and it may be easily seen\nthat a sufficient change in these would render doubtful what seem the\nmost universally certain of general rules. Thus, the general disutility\nof murder can only be proved, provided the majority of the human race\nwill certainly persist in existing. In order to prove that murder, if\nit were so universally adopted as to cause the speedy extermination of\nthe race, would not be good as a means, we should have to disprove the\nmain contention of pessimism--namely that the existence of human life\nis on the whole an evil. And the view of pessimism, however strongly we\nmay be convinced of its truth or falsehood, is one which never has been\neither proved or refuted conclusively. That universal murder would not\nbe a good thing at this moment can therefore not be proved. But, as a\nmatter of fact, we can and do assume with certainty that, even if a few\npeople are willing to murder, most people will not be willing. When,\ntherefore, we say that murder is in general to be avoided, we only\nmean that it is so, so long as the majority of mankind will certainly\nnot agree to it, but will persist in living. And that, under these\ncircumstances, it is generally wrong for any single person to commit\nmurder seems capable of proof. For, since there is in any case no hope\nof exterminating the race, the only effects which we have to consider\nare those which the action will have upon the increase of the goods\nand the diminution of the evils of human life. Where the best is not\nattainable (assuming extermination to be the best) one alternative may\nstill be better than another. And, apart from the immediate evils which\nmurder generally produces, the fact that, if it were a common practice,\nthe feeling of insecurity, thus caused, would absorb much time, which\nmight be spent to better purpose, is perhaps conclusive against it. So\nlong as men desire to live as strongly as they do, and so long as it is\ncertain that they will continue to do so, anything which hinders them\nfrom devoting their energy to the attainment of positive goods, seems\nplainly bad as a means. And the general practice of murder, falling so\nfar short of universality as it certainly must in all known conditions\nof society, seems certainly to be a hindrance of this kind.\n\nA similar defence seems possible for most of the rules, most\nuniversally enforced by legal sanctions, such as respect of property;\nand for some of those most commonly recognised by Common Sense, such\nas industry, temperance and the keeping of promises. In any state of\nsociety in which men have that intense desire for property of some\nsort, which seems to be universal, the common legal rules for the\nprotection of property must serve greatly to facilitate the best\npossible expenditure of energy. And similarly: Industry is a means\nto the attainment of those necessaries, without which the further\nattainment of any great positive goods is impossible; temperance merely\nenjoins the avoidance of those excesses, which, by injuring health,\nwould prevent a man from contributing as much as possible to the\nacquirement of these necessaries; and the keeping of promises greatly\nfacilitates cooperation in such acquirement.\n\nNow all these rules seem to have two characteristics to which it is\ndesirable to call attention. (1) They seem all to be such that, in any\nknown state of society, a _general_ observance of them _would_ be good\nas a means. The conditions upon which their utility depends, namely the\ntendency to preserve and propagate life and the desire of property,\nseem to be so universal and so strong, that it would be impossible to\nremove them; and, this being so, we can say that, under any conditions\nwhich could actually be given, the general observance of these rules\nwould be good as a means. For, while there seems no reason to think\nthat their observance ever makes a society worse than one in which they\nare not observed, it is certainly necessary as a means for any state of\nthings in which the greatest possible goods can be attained. And (2)\nthese rules, since they can be recommended as a means to that which is\nitself only a necessary condition for the existence of any great good,\ncan be defended independently of correct views upon the primary ethical\nquestion of what is good in itself. On any view commonly taken, it\nseems certain that the preservation of civilised society, which these\nrules are necessary to effect, is necessary for the existence, in any\ngreat degree, of anything which may be held to be good in itself.\n\n\n=96.= But not by any means all the rules commonly recognised combine\nthese two characteristics. The arguments offered in defence of Common\nSense morality very often presuppose the existence of conditions,\nwhich cannot be fairly assumed to be so universally necessary as the\ntendency to continue life and to desire property. Such arguments,\naccordingly, only prove the utility of the rule, so long as certain\nconditions, which may alter, remain the same: it cannot be claimed of\nthe rules thus defended, that they would be generally good as means in\nevery state of society: in order to establish this _universal_ general\nutility, it would be necessary to arrive at a correct view of what is\ngood or evil in itself. This, for instance, seems to be the case with\nmost of the rules comprehended under the name of Chastity. These rules\nare commonly defended, by Utilitarian writers or writers who assume as\ntheir end the conservation of society, with arguments which presuppose\nthe necessary existence of such sentiments as conjugal jealousy and\npaternal affection. These sentiments are no doubt sufficiently strong\nand general to make the defence valid for many conditions of society.\nBut it is not difficult to imagine a civilised society existing without\nthem; and, in such a case, if chastity were still to be defended,\nit would be necessary to establish that its violation produced evil\neffects, other than those due to the assumed tendency of such violation\nto disintegrate society. Such a defence may, no doubt, be made; but it\nwould require an examination into the primary ethical question of what\nis good and bad in itself, far more thorough than any ethical writer\nhas ever offered to us. Whether this be so in this particular case or\nnot, it is certain that a distinction, not commonly recognised, should\nbe made between those rules, of which the social utility depends upon\nthe existence of circumstances, more or less likely to alter, and those\nof which the utility seems certain under all possible conditions.\n\n\n=97.= It is obvious that all the rules, which were enumerated above as\nlikely to be useful in _almost any_ state of society, can _also_ be\ndefended owing to results which they produce under conditions which\nexist only in particular states of society. And it should be noticed\nthat we are entitled to reckon among these conditions the sanctions of\nlegal penalties, of social disapproval, and of private remorse, where\nthese exist. These sanctions are, indeed, commonly treated by Ethics\nonly as motives for the doing of actions of which the utility can be\nproved independently of the existence of these sanctions. And it may\nbe admitted that sanctions _ought_ not to be attached to actions which\nwould not be right independently. Nevertheless it is plain that, where\nthey do exist, they are not only motives but also justifications for\nthe actions in question. One of the chief reasons why an action should\nnot be done in any particular state of society is that it will be\npunished; since the punishment is in general itself a greater evil than\nwould have been caused by the omission of the action punished. Thus\nthe existence of a punishment may be an adequate reason for regarding\nan action as generally wrong, even though it has no other bad effects\nbut even slightly good ones. The fact that an action will be punished\nis a condition of exactly the same kind as others of more or less\npermanence, which must be taken into account in discussing the general\nutility or disutility of an action in a particular state of society.\n\n\n=98.= It is plain, then, that the rules commonly recognised by Common\nSense, in the society in which we live, and commonly advocated as if\nthey were all equally and universally right and good, are of very\ndifferent orders. Even those which seem to be most universally good\nas means, can only be shewn to be so, because of the existence of\nconditions, which, though perhaps evils, may be taken to be necessary;\nand even these owe their more obvious utilities to the existence of\nother conditions, which cannot be taken to be necessary except over\nlonger or shorter periods of history, and many of which are evils.\nOthers seem to be justifiable _solely_ by the existence of such more or\nless temporary conditions, unless we abandon the attempt to shew that\nthey are means to that preservation of society, which is itself a mere\nmeans, and are able to establish that they are directly means to things\ngood or evil in themselves, but which are not commonly recognised to be\nsuch.\n\nIf, then, we ask what rules are or would be useful to be observed in\nthe society in which we live, it seems possible to prove a definite\nutility in most of those which are in general both recognised and\npractised. But a great part of ordinary moral exhortation and social\ndiscussion consists in the advocating of rules, which are _not_\ngenerally practised; and with regard to these it seems very doubtful\nwhether a case for their general utility can ever be conclusively made\nout. Such proposed rules commonly suffer from three main defects. In\nthe first place, (1) the actions which they advocate are very commonly\nsuch as it is impossible for most individuals to perform by any\nvolition. It is far too usual to find classed together with actions,\nwhich can be performed, if only they be willed, others, of which the\npossibility depends upon the possession of a peculiar disposition,\nwhich is given to few and cannot even be acquired. It may, no doubt,\nbe useful to point out that those who have the necessary disposition\nshould obey these rules; and it would, in many cases, be desirable that\neverybody should have this disposition. But it should be recognised\nthat, when we regard a thing as a moral rule or law, we mean that it\nis one which _almost everybody can_ observe by an effort of volition,\nin that state of society to which the rule is supposed to apply. (2)\nActions are often advocated, of which, though they themselves are\npossible, yet the proposed good effects are not possible, because the\nconditions necessary for their existence are not sufficiently general.\nA rule, of which the observance would produce good effects, if human\nnature were in other respects different from what it is, is advocated\nas if its general observance would produce the same effects now and\nat once. In fact, however, by the time that the conditions necessary\nto make its observance useful have arisen, it is quite as likely that\nother conditions, rendering its observance unnecessary or positively\nharmful, may also have arisen; and yet this state of things may be a\nbetter one than that in which the rule in question would have been\nuseful. (3) There also occurs the case in which the usefulness of a\nrule depends upon conditions likely to change, or of which the change\nwould be as easy and more desirable than the observance of the proposed\nrule. It may even happen that the general observance of the proposed\nrule would itself destroy the conditions upon which its utility depends.\n\nOne or other of these objections seems generally to apply to proposed\nchanges in social custom, advocated as being better rules to follow\nthan those now actually followed; and, for this reason, it seems\ndoubtful whether Ethics can establish the utility of any rules\nother than those generally practised. But its inability to do so is\nfortunately of little practical moment. The question whether the\ngeneral observance of a rule not generally observed, would or would not\nbe desirable, cannot much affect the question how any individual ought\nto act; since, on the one hand, there is a large probability that he\nwill not, by any means, be able to bring about its general observance,\nand, on the other hand, the fact that its general observance would\nbe useful could, in any case, give him no reason to conclude that he\nhimself ought to observe it, in the absence of such general observance.\n\nWith regard, then, to the actions commonly classed in Ethics, as\nduties, crimes, or sins, the following points seem deserving of notice.\n(1) By so classing them we mean that they are actions which it is\npossible for an individual to perform or avoid, if he only _wills_ to\ndo so; and that they are actions which _everybody_ ought to perform\nor avoid, when occasion arises. (2) We can certainly not prove of\nany such action that it ought to be done or avoided under _all_\ncircumstances; we can only prove that its performance or avoidance will\n_generally_ produce better results than the alternative. (3) If further\nwe ask of what actions as much as this can be proved, it seems only\npossible to prove it with regard to those which are actually generally\npractised among us. And of these some only are such that their general\nperformance would be useful in any state of society that seems\npossible; of others the utility depends upon conditions which exist\nnow, but which seem to be more or less alterable.\n\n\n=99.= (_d_) So much, then, for moral rules or laws, in the ordinary\nsense--rules which assert that it is generally useful, under more or\nless common circumstances, for _everybody_ to perform or omit some\ndefinite kind of action. It remains to say something with regard to the\nprinciples by which _the individual_ should decide what he ought to\ndo, (\u03b1) with regard to those actions as to which some general rule is\ncertainly true, and (\u03b2) with regard to those where such a certain rule\nis wanting.\n\n(\u03b1) Since, as I have tried to shew, it is impossible to establish\nthat any kind of action will produce a better total result than\nits alternative _in all cases_, it follows that in some cases the\nneglect of an established rule will probably be the best course of\naction possible. The question then arises: Can the individual ever be\njustified in assuming that his is one of these exceptional cases? And\nit seems that this question may be definitely answered in the negative.\nFor, if it is certain that in a large majority of cases the observance\nof a certain rule is useful, it follows that there is a large\nprobability that it would be wrong to break the rule in any particular\ncase; and the uncertainty of our knowledge both of effects and of their\nvalue, in particular cases, is so great, that it seems doubtful whether\nthe individual\u2019s judgment that the effects will probably be good in\nhis case can ever be set against the general probability that that\nkind of action is wrong. Added to this general ignorance is the fact\nthat, if the question arises at all, our judgment will generally be\nbiassed by the fact that we strongly desire one of the results which we\nhope to obtain by breaking the rule. It seems, then, that with regard\nto any rule which is _generally_ useful, we may assert that it ought\n_always_ to be observed, not on the ground that in _every_ particular\ncase it will be useful, but on the ground that in _any_ particular\ncase the probability of its being so is greater than that of our being\nlikely to decide rightly that we have before us an instance of its\ndisutility. In short, though we may be sure that there are cases where\nthe rule should be broken, we can never know which those cases are,\nand ought, therefore, never to break it. It is this fact which seems\nto justify the stringency with which moral rules are usually enforced\nand sanctioned, and to give a sense in which we may accept as true the\nmaxims that \u2018The end never justifies the means\u2019 and \u2018That we should\nnever do evil that good may come.\u2019 The \u2018means\u2019 and the \u2018evil,\u2019 intended\nby these maxims, are, in fact, the breaking of moral rules generally\nrecognised and practised, and which, therefore, we may assume to be\ngenerally useful. Thus understood, these maxims merely point out that,\nin any particular case, although we cannot clearly perceive any balance\nof good produced by keeping the rule and do seem to see one that would\nfollow from breaking it, nevertheless the rule should be observed. It\nis hardly necessary to point out that this is so only because it is\ncertain that, in general, the end does justify the means in question,\nand that therefore there is a _probability_ that in this case it will\ndo so also, although we cannot see that it will.\n\nBut moreover the universal observance of a rule which is generally\nuseful has, in many cases, a special utility, which seems deserving of\nnotice. This arises from the fact that, even if we can clearly discern\nthat our case is one where to break the rule is advantageous, yet, so\nfar as our example has any effect at all in encouraging similar action,\nit will certainly tend to encourage breaches of the rule which are not\nadvantageous. We may confidently assume that what will impress the\nimagination of others will not be the circumstances in which our case\ndiffers from ordinary cases and which justify our exceptional action,\nbut the points in which it resembles other actions that are really\ncriminal. In cases, then, where example has any influence at all, the\neffect of an exceptional right action will generally be to encourage\nwrong ones. And this effect will probably be exercised not only on\nother persons but on the agent himself. For it is impossible for any\none to keep his intellect and sentiments so clear, but that, if he has\nonce approved of a generally wrong action, he will be more likely to\napprove of it also under other circumstances than those which justified\nit in the first instance. This inability to discriminate exceptional\ncases offers, of course, a still stronger reason for the universal\nenforcement, by legal or social sanctions, of actions generally useful.\nIt is undoubtedly well to punish a man, who has done an action, right\nin his case but generally wrong, even if his example would not be\nlikely to have a dangerous effect. For sanctions have, in general,\nmuch more influence upon conduct than example; so that the effect\nof relaxing them in an exceptional case will almost certainly be an\nencouragement of similar action in cases which are not exceptional.\n\nThe individual can therefore be confidently recommended _always_\nto conform to rules which are both generally useful and generally\npractised. In the case of rules of which the general observance _would_\nbe useful but does not exist, or of rules which are generally practised\nbut which are not useful, no such universal recommendations can be\nmade. In many cases the sanctions attached may be decisive in favour\nof conformity to the existing custom. But it seems worth pointing out\nthat, even apart from these, the general utility of an action most\ncommonly depends upon the fact that it is generally practised: in a\nsociety where certain kinds of theft are the common rule, the utility\nof abstinence from such theft on the part of a single individual\nbecomes exceedingly doubtful, even though the common rule is a bad one.\nThere is, therefore, a strong probability in favour of adherence to an\nexisting custom, even if it be a bad one. But we cannot, in this case,\nassert with any confidence that this probability is always greater\nthan that of the individual\u2019s power to judge that an exception will be\nuseful; since we are here supposing certain one relevant fact--namely,\nthat the rule, which he proposes to follow, _would_ be better than\nthat which he proposes to break, _if_ it were generally observed.\nConsequently the effect of his example, so far as it tends to break\ndown the existing custom, will here be for the good. The cases, where\nanother rule would certainly be better than that generally observed,\nare, however, according to what was said above, very rare; and cases of\ndoubt, which are those which arise most frequently, carry us into the\nnext division of our subject.\n\n\n=100.= (\u03b2) This next division consists in the discussion of the\nmethod by which an individual should decide what to do with regard to\npossible actions of which the general utility cannot be proved. And\nit should be observed, that, according to our previous conclusions,\nthis discussion will cover almost all actions, except those which,\nin our present state of society, are generally practised. For it has\nbeen urged that a proof of general utility is so difficult, that it\ncan hardly be conclusive except in a very few cases. It is certainly\nnot possible with regard to all actions which _are_ generally\npractised; though here, if the sanctions are sufficiently strong,\nthey are sufficient by themselves to prove the general utility of the\nindividual\u2019s conformity to custom. And if it is possible to prove a\ngeneral utility in the case of some actions, _not_ generally practised,\nit is certainly not possible to do so by the ordinary method, which\ntries to shew in them a tendency to that preservation of society, which\nis itself a mere means, but only by the method, by which in any case,\nas will be urged, the individual ought to guide his judgment--namely,\nby shewing their direct tendency to produce what is good in itself or\nto prevent what is bad.\n\nThe extreme improbability that any general rule with regard to the\nutility of an action will be correct seems, in fact, to be the chief\nprinciple which should be taken into account in discussing how the\nindividual should guide his choice. If we except those rules which are\nboth generally practised and strongly sanctioned among us, there seem\nto be hardly any of such a kind that equally good arguments cannot be\nfound both for and against them. The most that can be said for the\ncontradictory principles which are urged by moralists of different\nschools as universal duties, is, in general, that they point out\nactions which, for persons of a particular character and in particular\ncircumstances, would and do lead to a balance of good. It is, no\ndoubt, possible that the particular dispositions and circumstances\nwhich generally render certain kinds of action advisable, might to\nsome degree be formulated. But it is certain that this has never\nyet been done; and it is important to notice that, even if it were\ndone, it would not give us, what moral laws are usually supposed to\nbe--rules which it would be desirable for every one, or even for most\npeople, to follow. Moralists commonly assume that, in the matter of\nactions or habits of action, usually recognised as duties or virtues,\nit is desirable that every one should be alike. Whereas it is certain\nthat, under actual circumstances, and possible that, even in a much\nmore ideal condition of things, the principle of division of labour,\naccording to special capacity, which is recognised in respect of\nemployments, would also give a better result in respect of virtues.\n\nIt seems, therefore, that, in cases of doubt, instead of following\nrules, of which he is unable to see the good effects in his particular\ncase, the individual should rather guide his choice by a direct\nconsideration of the intrinsic value or vileness of the effects\nwhich his action may produce. Judgments of intrinsic value have this\nsuperiority over judgments of means that, if once true, they are always\ntrue; whereas what is a means to a good effect in one case, will not\nbe so in another. For this reason the department of Ethics, which it\nwould be most useful to elaborate for practical guidance, is that\nwhich discusses what things have intrinsic value and in what degrees;\nand this is precisely that department which has been most uniformly\nneglected, in favour of attempts to formulate rules of conduct.\n\nWe have, however, not only to consider the relative goodness of\ndifferent effects, but also the relative probability of their being\nattained. A less good, that is more likely to be attained, is to be\npreferred to a greater, that is less probable, if the difference in\nprobability is great enough to outweigh the difference in goodness.\nAnd this fact seems to entitle us to assert the general truth of three\nprinciples, which ordinary moral rules are apt to neglect. (1) That\na lesser good, for which any individual has a strong preference (if\nonly it be a good, and not an evil), is more likely to be a proper\nobject for him to aim at, than a greater one, which he is unable to\nappreciate. For natural inclination renders it immensely more easy to\nattain that for which such inclination is felt. (2) Since almost every\none has a much stronger preference for things which closely concern\nhimself, it will in general be right for a man to aim rather at goods\naffecting himself and those in whom he has a strong personal interest,\nthan to attempt a more extended beneficence. Egoism is undoubtedly\nsuperior to Altruism as a doctrine of means: in the immense majority of\ncases the best thing we can do is to aim at securing some good in which\nwe are concerned, since for that very reason we are far more likely to\nsecure it. (3) Goods, which can be secured in a future so near as to be\ncalled \u2018the present,\u2019 are in general to be preferred to those which,\nbeing in a further future, are, for that reason, far less certain of\nattainment. If we regard all that we do from the point of view of\nits rightness, that is to say as a mere means to good, we are apt to\nneglect one fact, at least, which is certain; namely, that a thing that\nis really good in itself, if it exist now, has precisely the same value\nas a thing of the same kind which may be caused to exist in the future.\nMoreover moral rules, as has been said, are, in general, not directly\nmeans to positive goods but to what is necessary for the existence of\npositive goods; and so much of our labour must in any case be devoted\nto securing the continuance of what is thus a mere means--the claims of\nindustry and attention to health determine the employment of so large\na part of our time, that, in cases where choice is open, the certain\nattainment of a present good will in general have the strongest claims\nupon us. If it were not so, the whole of life would be spent in merely\nassuring its continuance; and, so far as the same rule were continued\nin the future, that for the sake of which it is worth having, would\nnever exist at all.\n\n\n=101.= (4) A fourth conclusion, which follows from the fact that what\nis \u2018right\u2019 or what is our \u2018duty\u2019 must in any case be defined as what is\na means to good, is, as was pointed out above (\u00a7 89), that the common\ndistinction between these and the \u2018expedient\u2019 or \u2018useful,\u2019 disappears.\nOur \u2018duty\u2019 is merely that which will be a means to the best possible,\nand the expedient, if it is really expedient, must be just the same. We\ncannot distinguish them by saying that the former is something which\nwe ought to do, whereas of the latter we cannot say we \u2018_ought_.\u2019 In\nshort the two concepts are not, as is commonly assumed by all except\nUtilitarian moralists, simple concepts ultimately distinct. There is\nno such distinction in Ethics. The only fundamental distinction is\nbetween what is good in itself and what is good as a means, the latter\nof which implies the former. But it has been shewn that the distinction\nbetween \u2018duty\u2019 and \u2018expediency\u2019 does not correspond to this: both\nmust be defined as means to good, though both _may also_ be ends in\nthemselves. The question remains, then: What is the distinction between\nduty and expediency?\n\nOne distinction to which these distinct words refer is plain\nenough. Certain classes of action commonly excite the specifically\nmoral sentiments, whereas other classes do not. And the word\n\u2018duty\u2019 is commonly applied only to the class of actions which\nexcite moral approval, or of which the omission excites moral\ndisapproval--especially to the latter. Why this moral sentiment should\nhave become attached to some kinds of actions and not to others is\na question which can certainly not yet be answered; but it may be\nobserved that we have no reason to think that the actions to which\nit was attached were or are, in all cases, such as aided or aid the\nsurvival of a race: it was probably originally attached to many\nreligious rites and ceremonies which had not the smallest utility in\nthis respect. It appears, however, that, among us, the classes of\naction to which it is attached also have two other characteristics in\nenough cases to have influenced the meaning of the words \u2018duty\u2019 and\n\u2018expediency.\u2019 One of these is that \u2018duties\u2019 are, in general, actions\nwhich a considerable number of individuals are strongly tempted to\nomit. The second is that the omission of a \u2018duty\u2019 generally entails\nconsequences markedly disagreeable to _some one else_. The first of\nthese is a more universal characteristic than the second: since the\ndisagreeable effects on other people of the \u2018self-regarding duties,\u2019\nprudence and temperance, are not so marked as those on the future\nof the agent himself; whereas the temptations to imprudence and\nintemperance are very strong. Still, on the whole, the class of actions\ncalled duties exhibit both characteristics: they are not only actions,\nagainst the performance of which there are strong natural inclinations,\nbut also actions of which the most obvious effects, commonly considered\ngoods, are effects on other people. Expedient actions, on the other\nhand, are actions to which strong natural inclinations prompt us\nalmost universally, and of which all the most obvious effects, commonly\nconsidered good, are effects upon the agent. We may then roughly\ndistinguish \u2018duties\u2019 from expedient actions, as actions with regard to\nwhich there is a moral sentiment, which we are often tempted to omit,\nand of which the most obvious effects are effects upon others than the\nagent.\n\nBut it is to be noticed that none of these characteristics, by which\na \u2018duty\u2019 is distinguished from an expedient action, gives us any\nreason to infer that the former class of actions are more useful than\nthe latter--that they tend to produce a greater balance of good.\nNor, when we ask the question, \u2018Is this my duty?\u2019 do we mean to ask\nwhether the action in question has these characteristics: we are asking\nsimply whether it will produce the best possible result on the whole.\nAnd if we asked this question with regard to expedient actions, we\nshould quite as often have to answer it in the affirmative as when we\nask it with regard to actions which have the three characteristics\nof \u2018duties.\u2019 It is true that when we ask the question, \u2018Is this\nexpedient?\u2019 we are asking a different question--namely, whether it will\nhave certain kinds of effect, with regard to which we do not enquire\nwhether they are good or not. Nevertheless, if it should be doubted\nin any particular case whether these effects were good, this doubt is\nunderstood as throwing doubt upon the action\u2019s expediency: if we are\nrequired to _prove_ an action\u2019s expediency, we can only do so by asking\nprecisely the same question by which we should prove it a duty--namely,\n\u2018Has it the best possible effects on the whole?\u2019\n\nAccordingly the question whether an action is a duty or merely\nexpedient, is one which has no bearing on the ethical question whether\nwe ought to do it. In the sense in which either duty or expediency\nare taken as ultimate _reasons_ for doing an action, they are taken\nin exactly the same sense: if I ask whether an action is _really_ my\nduty or _really_ expedient, the predicate of which I question the\napplicability to the action in question is precisely the same. In both\ncases I am asking, \u2018Is this event the best on the whole that I can\neffect?\u2019; and whether the event in question be some effect upon what\nis _mine_ (as it usually is, where we talk of expediency) or some other\nevent (as is usual, where we talk of duty), this distinction has no\nmore relevance to my answer than the distinction between two different\neffects on me or two different effects on others. The true distinction\nbetween duties and expedient actions is not that the former are actions\nwhich it is in any sense more useful or obligatory or better to\nperform, but that they are actions which it is more useful to praise\nand to enforce by sanctions, since they are actions which there is a\ntemptation to omit.\n\n\n=102.= With regard to \u2018interested\u2019 actions, the case is somewhat\ndifferent. When we ask the question, \u2018Is this really to my interest?\u2019\nwe appear to be asking exclusively whether its _effects upon me_ are\nthe best possible; and it may well happen that what will effect me in\nthe manner, which is really the best possible, will not produce the\nbest possible results on the whole. Accordingly _my true interest_ may\nbe different from the course which is really expedient and dutiful. To\nassert that an action is \u2018to my interest,\u2019 is, indeed, as was pointed\nout in Chap. III. (\u00a7\u00a7 59-61), to assert that its effects are really\ngood. \u2018My own good\u2019 only denotes some event affecting me, which is good\nabsolutely and objectively; it is the thing, and not its goodness,\nwhich is _mine_; everything must be either \u2018a part of universal good\u2019\nor else not good at all; there is no third alternative conception\n\u2018good for me.\u2019 But \u2018my interest,\u2019 though it must be something truly\ngood, is only one among possible good effects; and hence, by effecting\nit, though we shall be doing _some_ good, we may be doing less good\non the whole, than if we had acted otherwise. Self-sacrifice may be a\nreal duty; just as the sacrifice of any single good, whether affecting\nourselves or others, may be necessary in order to obtain a better\ntotal result. Hence the fact that an action is really to my interest,\ncan never be a sufficient reason for doing it: by shewing that it is\nnot a means to the best possible, we do not shew that it is not to my\ninterest, as we do shew that it is not expedient. Nevertheless there is\nno necessary conflict between duty and interest: what is to my interest\nmay also be a means to the best possible. And the chief distinction\nconveyed by the distinct words \u2018duty\u2019 and \u2018interest\u2019 seems to be not\nthis source of possible conflict, but the same which is conveyed by\nthe contrast between \u2018duty\u2019 and \u2018expediency.\u2019 By \u2018interested\u2019 actions\nare _mainly_ meant those which, whether a means to the best possible\nor not, are such as have their most obvious effects on the agent;\nwhich he generally has no temptation to omit; and with regard to which\nwe feel no moral sentiment. That is to say, the distinction is not\nprimarily ethical. Here too \u2018duties\u2019 are not, in general, more useful\nor obligatory than interested actions; they are only actions which it\nis more useful to praise.\n\n\n=103.= (5) A fifth conclusion, of some importance, in relation to\nPractical Ethics concerns the manner in which \u2018virtues\u2019 are to be\njudged. What is meant by calling a thing a \u2018virtue\u2019?\n\nThere can be no doubt that Aristotle\u2019s definition is right, in the\nmain, so far as he says that it is an \u2018habitual disposition\u2019 to\nperform certain actions: this is one of the marks by which we should\ndistinguish a virtue from other things. But \u2018virtue\u2019 and \u2018vice\u2019 are\nalso ethical terms: that is to say, when we use them seriously, we mean\nto convey praise by the one and dispraise by the other. And to praise\na thing is to assert either that it is good in itself or else that it\nis a means to good. Are we then to include in our definition of virtue\nthat it must be a thing good in itself?\n\nNow it is certain that virtues are commonly regarded as good in\nthemselves. The feeling of moral approbation with which we generally\nregard them partly consists in an attribution to them of intrinsic\nvalue. Even a Hedonist, when he feels a moral sentiment towards\nthem, is regarding them as good-in-themselves; and Virtue has been\nthe chief competitor with Pleasure for the position of _sole_ good.\nNevertheless I do not think we can regard it as part of the definition\nof virtue that it should be good in itself. For the name has so far\nan independent meaning, that if in any particular case a disposition\ncommonly considered virtuous were proved not to be good in itself, we\nshould not think that a sufficient reason for saying that it _was_ not\na virtue but was only _thought_ to be so. The test for the ethical\nconnotation of virtue is the same as that for duty: What should we\nrequire to be proved about a particular instance, in order to say\nthat the name was wrongly applied to it? And the test which is thus\napplied both to virtues and duties, and considered to be final, is the\nquestion: Is it a means to good? If it could be shewn of any particular\ndisposition, commonly considered virtuous, that it was generally\nharmful, we should at once say: Then it is not really virtuous.\nAccordingly a virtue may be defined as an habitual disposition to\nperform certain actions, which generally produce the best possible\nresults. Nor is there any doubt as to the kind of actions which it is\n\u2018virtuous\u2019 habitually to perform. They are, in general, those which\nare duties, with this modification that we also include those which\n_would_ be duties, if only it were possible for people in general to\nperform them. Accordingly with regard to virtues, the same conclusion\nholds as with regard to duties. If they are really virtues they must be\ngenerally good as means; nor do I wish to dispute that most virtues,\ncommonly considered as such, as well as most duties, really are means\nto good. But it does not follow that they are a bit more useful than\nthose dispositions and inclinations which lead us to perform interested\nactions. As duties from expedient actions, so virtues are distinguished\nfrom other useful dispositions, not by any superior utility, but by the\nfact that they are dispositions, which it is particularly useful to\npraise and to sanction, because there are strong and common temptations\nto neglect the actions to which they lead.\n\nVirtues, therefore, are habitual dispositions to perform actions which\nare duties, or which would be duties if a volition were sufficient on\nthe part of most men to ensure their performance. And duties are a\nparticular class of those actions, of which the performance has, at\nleast generally, better total results than the omission. They are, that\nis to say, actions generally good as means: but not all such actions\nare duties; the name is confined to that particular class which it is\noften difficult to perform, because there are strong temptations to the\ncontrary. It follows that in order to decide whether any particular\ndisposition or action is a virtue or a duty, we must face all the\ndifficulties enumerated in section (3) of this chapter. We shall\nnot be entitled to assert that any disposition or action is a virtue\nor duty except as a result of an investigation, such as was there\ndescribed. We must be able to prove that the disposition or action in\nquestion is generally better as a means than any alternatives possible\nand likely to occur; and this we shall only be able to prove for\nparticular states of society: what is a virtue or a duty in one state\nof society may not be so in another.\n\n\n=104.= But there is another question with regard to virtues and duties\nwhich must be settled by intuition alone--by the properly guarded\nmethod which was explained in discussing Hedonism. This is the question\nwhether the dispositions and actions, commonly regarded (rightly or\nnot) as virtues or duties, are good in themselves; whether they have\nintrinsic value. Virtue or the exercise of virtue has very commonly\nbeen asserted by moralists to be either the sole good, or, at least,\nthe best of goods. Indeed, so far as moralists have discussed the\nquestion what is good in itself at all, they have generally assumed\nthat it must be either virtue or pleasure. It would hardly have been\npossible that such a gross difference of opinion should exist, or that\nit should have been assumed the discussion _must_ be limited to two\nsuch alternatives, if the meaning of the question had been clearly\napprehended. And we have already seen that the meaning of the question\nhas hardly ever been clearly apprehended. Almost all ethical writers\nhave committed the naturalistic fallacy--they have failed to perceive\nthat the notion of intrinsic value is simple and unique; and almost all\nhave failed, in consequence, to distinguish clearly between means and\nend--they have discussed, as if it were simple and unambiguous, the\nquestion, \u2018What ought we to do?\u2019 or \u2018What ought to exist now?\u2019 without\ndistinguishing whether the reason why a thing ought to be done or to\nexist now, is that it is itself possessed of intrinsic value, or that\nit is a means to what has intrinsic value. We shall, therefore, be\nprepared to find that virtue has as little claim to be considered the\nsole or chief good as pleasure; more especially after seeing that, so\nfar as definition goes, to call a thing a virtue is merely to declare\nthat it is a means to good. The advocates of virtue have, we shall see,\nthis superiority over the Hedonists, that inasmuch as virtues are very\ncomplex mental facts, there are included in them many things which are\ngood in themselves and good in a much higher degree than pleasure.\nThe advocates of Hedonism, on the other hand, have the superiority\nthat their method emphasizes the distinction between means and ends;\nalthough they have not apprehended the distinction clearly enough to\nperceive that the special ethical predicate, which they assign to\npleasure as _not_ being a mere means, must also apply to many other\nthings.\n\n\n=105.= With regard, then, to the intrinsic value of virtue, it may be\nstated broadly: (1) that the majority of dispositions, which we call\nby that name, and which really do conform to the definition, so far as\nthat they are dispositions generally valuable as means, at least in our\nsociety, have no intrinsic value whatever; and (2) that no one element\nwhich is contained in the minority, nor even all the different elements\nput together, can without gross absurdity be regarded as the sole good.\nAs to the second point it may be observed that even those who hold the\nview that the sole good is to be found in virtue, almost invariably\nhold other views contradictory of this, owing chiefly to a failure to\nanalyse the meaning of ethical concepts. The most marked instance of\nthis inconsistency is to be found in the common Christian conception\nthat virtue, though the sole good, can yet be rewarded by something\nother than virtue. Heaven is commonly considered as the reward of\nvirtue; and yet it is also commonly considered, that, in order to be\nsuch a reward, it must contain some element, called happiness, which\nis certainly not completely identical with the mere exercise of those\nvirtues which it rewards. But if so, then something which is not virtue\nmust be either good in itself or an element in what has most intrinsic\nvalue. It is not commonly observed that if a thing is really to be a\nreward, it must be something good in itself: it is absurd to talk of\nrewarding a person by giving him something, which is less valuable than\nwhat he already has or which has no value at all. Thus Kant\u2019s view that\nvirtue renders us _worthy_ of happiness is in flagrant contradiction\nwith the view, which he implies and which is associated with his name,\nthat a Good Will is the only thing having intrinsic value. It does\nnot, indeed, entitle us to make the charge sometimes made, that Kant\nis, inconsistently, an Eudaemonist or Hedonist: for it does not imply\nthat happiness is the sole good. But it does imply that the Good Will\nis _not_ the sole good: that a state of things in which we are both\nvirtuous and happy is better in itself than one in which the happiness\nis absent.\n\n\n=106.= In order, however, justly to consider the claims of virtue to\nintrinsic value, it is necessary to distinguish several very different\nmental states, all of which fall under the general definition that they\nare habitual dispositions to perform duties. We may thus distinguish\nthree very different states, all of which are liable to be confused\nwith one another, upon each of which different moral systems have laid\ngreat stress, and for each of which the claim has been made that it\nalone constitutes virtue, and, by implication, that it is the sole\ngood. We may first of all distinguish between (_a_) that permanent\ncharacteristic of mind, which consists in the fact that the performance\nof duty has become in the strict sense a habit, like many of the\noperations performed in the putting on of clothes, and (_b_) that\npermanent characteristic, which consists in the fact that what may be\ncalled good motives habitually help to cause the performance of duties.\nAnd in the second division we may distinguish between the habitual\ntendency to be actuated by one motive, namely, the desire to do duty\nfor duty\u2019s sake, and all other motives, such as love, benevolence, etc.\nWe thus get the three kinds of virtue, of which we are now to consider\nthe intrinsic value.\n\n(_a_) There is no doubt that a man\u2019s character may be such that he\nhabitually performs certain duties, without the thought ever occurring\nto him, when he wills them, either that they are duties or that any\ngood will result from them. Of such a man we cannot and do not refuse\nto say that he possesses the virtue consisting in the disposition to\nperform those duties. I, for instance, am honest in the sense that\nI habitually abstain from any of the actions legally qualified as\nthieving, even where some other persons would be strongly tempted to\ncommit them. It would be grossly contrary to common usage to deny\nthat, for this reason, I really have the virtue of honesty: it is\nquite certain that I have an habitual disposition to perform a duty.\nAnd that as many people as possible should have a like disposition\nis, no doubt, of great utility: it is good as a means. Yet I may\nsafely assert that neither my various performances of this duty, nor\nmy disposition to perform them, have the smallest intrinsic value.\nIt is because the majority of instances of virtue seem to be of this\nnature, that we may venture to assert that virtues have, in general,\nno intrinsic value whatsoever. And there seems good reason to think\nthat the more generally they are of this nature the more useful they\nare; since a great economy of labour is effected when a useful action\nbecomes habitual or instinctive. But to maintain that a virtue, which\nincludes no more than this, is good in itself is a gross absurdity. And\nof this gross absurdity, it may be observed, the Ethics of Aristotle is\nguilty. For his definition of virtue does not exclude a disposition to\nperform actions in this way, whereas his descriptions of the particular\nvirtues plainly _include_ such actions: that an action, in order to\nexhibit virtue, must be done \u03c4\u03bf\u1fe6 \u03ba\u03b1\u03bb\u03bf\u1fe6 \u1f15\u03bd\u03b5\u03ba\u03b1 is a qualification which\nhe allows often to drop out of sight. And, on the other hand, he seems\ncertainly to regard the exercise of _all_ virtues as an end in itself.\nHis treatment of Ethics is indeed, in the most important points, highly\nunsystematic and confused, owing to his attempt to base it on the\nnaturalistic fallacy; for strictly we should be obliged by his words\nto regard \u03b8\u03b5\u03c9\u03c1\u03af\u03b1 as the _only_ thing good in itself, in which case\nthe goodness which he attributes to the practical virtues cannot be\nintrinsic value; while on the other hand he does not seem to regard\nit merely as utility, since he makes no attempt to shew that they are\nmeans to \u03b8\u03b5\u03c9\u03c1\u03af\u03b1. But there seems no doubt that on the whole he regards\nthe exercise of the practical virtues as a good of the same kind as\n(_i.e._ having intrinsic value), only in a less degree than, \u03b8\u03b5\u03c9\u03c1\u03af\u03b1; so\nthat he cannot avoid the charge that he recommends as having intrinsic\nvalue, such instances of the exercise of virtue as we are at present\ndiscussing--instances of a disposition to perform actions which, in\nthe modern phrase, have merely an \u2018external rightness.\u2019 That he is\nright in applying the word \u2018virtue\u2019 to such a disposition cannot be\ndoubted. But the protest against the view that \u2018external rightness\u2019 is\nsufficient to constitute either \u2018duty\u2019 or \u2018virtue\u2019--a protest which is\ncommonly, and with some justice, attributed as a merit to Christian\nmorals--seems, in the main, to be a mistaken way of pointing out an\nimportant truth: namely, that where there is only \u2018external rightness\u2019\nthere is certainly no intrinsic value. It is commonly assumed (though\nwrongly) that to call a thing a virtue means that it has intrinsic\nvalue: and on this assumption the view that virtue does not consist\nin a mere disposition to do externally right actions does really\nconstitute an advance in ethical truth beyond the Ethics of Aristotle.\nThe inference that, if virtue includes in its meaning \u2018good in itself,\u2019\nthen Aristotle\u2019s definition of virtue is not adequate and expresses a\nfalse ethical judgment, is perfectly correct: only the premiss that\nvirtue does include this in its meaning is mistaken.\n\n\n=107.= (_b_) A man\u2019s character may be such that, when he habitually\nperforms a particular duty, there is, in each case of his performance,\npresent in his mind, a love of some intrinsically good consequence\nwhich he expects to produce by his action or a hatred of some\nintrinsically evil consequence which he hopes to prevent by it. In\nsuch a case this love or hatred will generally be part cause of his\naction, and we may then call it one of his _motives_. Where such a\nfeeling as this is present habitually in the performance of duties, it\ncannot be denied that the state of the man\u2019s mind, in performing it,\ncontains something intrinsically good. Nor can it be denied that, where\na disposition to perform duties consists in the disposition to be moved\nto them by such feelings, we call that disposition a virtue. Here,\ntherefore, we have instances of virtue, the exercise of which really\ncontains something that is good in itself. And, in general, we may say\nthat wherever a virtue does consist in a disposition to have certain\nmotives, the exercise of that virtue _may_ be intrinsically good;\nalthough the degree of its goodness may vary indefinitely according to\nthe precise nature of the motives and their objects. In so far, then,\nas Christianity tends to emphasize the importance of motives, of the\n\u2018inward\u2019 disposition with which a right action is done, we may say that\nit has done a service to Ethics. But it should be noticed that, when\nChristian Ethics, as represented by the New Testament, are praised for\nthis, two distinctions of the utmost importance, which they entirely\nneglect, are very commonly overlooked. In the first place the New\nTestament is largely occupied with continuing the tradition of the\nHebrew prophets, by recommending such virtues as \u2018justice\u2019 and \u2018mercy\u2019\nas against mere ritual observances; and, in so far as it does this, it\nis recommending virtues which may be _merely_ good as means, exactly\nlike the Aristotelian virtues. This characteristic of its teaching must\ntherefore be rigorously distinguished from that which consists in its\nenforcement of such a view as that to be angry without a cause is as\nbad as actually to commit murder. And, in the second place, though the\nNew Testament does praise some things which are only good as means, and\nothers which are good in themselves, it entirely fails to recognise\nthis distinction. Though the state of the man who is angry may be\nreally as bad in itself as that of the murderer, and so far Christ\nmay be right, His language would lead us to suppose that it is _also_\nas bad in every way, that it _also causes_ as much evil: and this is\nutterly false. In short, when Christian Ethics approves, it does not\ndistinguish whether its approval asserts \u2018This is a means to good\u2019 or\n\u2018This is good in itself\u2019; and hence it both praises things merely good\nas means, as if they were good in themselves, and things merely good in\nthemselves as if they were also good as means. Moreover it should be\nnoticed, that if Christian Ethics does draw attention to those elements\nin virtues which are good in themselves, it is by no means alone in\nthis. The Ethics of Plato are distinguished by upholding, far more\nclearly and consistently than any other system, the view that intrinsic\nvalue belongs exclusively to those states of mind which consist in love\nof what is good or hatred of what is evil.\n\n\n=108.= But (_c_) the Ethics of Christianity are distinguished from\nthose of Plato by emphasizing the value of one particular motive--that\nwhich consists in the emotion excited by the idea, not of any\nintrinsically good consequences of the action in question, nor even\nof the action itself, but by that of its rightness. This idea of\nabstract \u2018rightness\u2019 and the various degrees of the specific emotion\nexcited by it are what constitute the specifically \u2018moral sentiment\u2019 or\n\u2018conscience.\u2019 An action seems to be most properly termed \u2018internally\nright[23],\u2019 solely in virtue of the fact that the agent has previously\nregarded it as right: the idea of \u2018rightness\u2019 must have been present\nto his mind, but need not necessarily have been among his motives. And\nwe mean by a \u2018conscientious\u2019 man, one who, when he deliberates, always\nhas this idea in his mind, and does not act until he believes that his\naction is right.\n\n  [23] This sense of the term must be carefully distinguished from\n  that in which the agent\u2019s intention may be said to be \u2018right,\u2019 if\n  only the results he intended would have been the best possible.\n\nThe presence of this idea and its action as a motive certainly seem\nto have become more common objects of notice and commendation owing\nto the influence of Christianity; but it is important to observe\nthat there is no ground for the view, which Kant implies, that it is\nthe _only_ motive which the New Testament regards as intrinsically\nvaluable. There seems little doubt that when Christ tells us to \u2018Love\nour neighbours as ourselves,\u2019 He did not mean merely what Kant calls\n\u2018practical love\u2018--beneficence of which the _sole_ motive is the idea of\nits rightness, or the emotion caused by that idea. Among the \u2018inward\ndispositions\u2019 of which the New Testament inculcates the value, there\nare certainly included what Kant terms mere \u2018natural inclinations,\u2019\nsuch as pity, etc.\n\nBut what are we to say of virtue, when it consists in a disposition to\nbe moved to the performance of duties by this idea? It seems difficult\nto deny that the emotion excited by rightness as such has some\nintrinsic value; and still more difficult to deny that its presence\nmay heighten the value of some wholes into which it enters. But, on\nthe other hand, it certainly has not more value than many of the\nmotives treated in our last section--emotions of love towards things\nreally good in themselves. And as for Kant\u2019s implication that it is\nthe sole good[24], this is inconsistent with other of his own views.\nFor he certainly regards it as _better_ to perform the actions, to\nwhich he maintains that it prompts us--namely, \u2018material\u2019 duties--than\nto omit them. But, if better at all, then, these actions must be\nbetter either in themselves or as a means. The former hypothesis would\ndirectly contradict the statement that this motive was _sole_ good,\nand the latter is excluded by Kant himself since he maintains that\nno actions can _cause_ the existence of this motive. And it may also\nbe observed that the other claim which he makes for it, namely, that\nit is _always_ good as a means, can also not be maintained. It is as\ncertain as anything can be that very harmful actions may be done from\nconscientious motives; and that Conscience does not always tell us the\ntruth about what actions are right. Nor can it be maintained even that\nit is _more_ useful than many other motives. All that can be admitted\nis that it is one of the things which are generally useful.\n\n  [24] Kant, so far as I know, never expressly states this view,\n  but it is implied _e.g._ in his argument against Heteronomy.\n\nWhat more I have to say with regard to those elements in some virtues\nwhich are good in themselves, and with regard to their relative degrees\nof excellence, as well as the proof that all of them together cannot be\nthe sole good, may be deferred to the next chapter.\n\n\n=109.= The main points in this chapter, to which I desire to direct\nattention, may be summarised as follows:--(1) I first pointed out how\nthe subject-matter with which it deals, namely, ethical judgments\non conduct, involves a question, utterly different in kind from the\ntwo previously discussed, namely: (_a_) What is the nature of the\npredicate peculiar to Ethics? and (_b_) What kinds of things themselves\npossess this predicate? Practical Ethics asks, not \u2018What ought to be?\u2019\nbut \u2018What ought we to do?\u2019; it asks what actions are _duties_, what\nactions are _right_, and what _wrong_: and all these questions can\nonly be answered by shewing the relation of the actions in question,\nas _causes_ or _necessary conditions_, to what is good in itself. The\nenquiries of Practical Ethics thus fall entirely under the _third_\ndivision of ethical questions--questions which ask, \u2018What is good as\na means?\u2019 which is equivalent to \u2018What is a means to good--what is\ncause or necessary condition of things good in themselves?\u2019 (86-88).\nBut (2) it asks this question, almost exclusively, with regard to\nactions which it is possible for most men to perform, if only they\n_will_ them; and with regard to these, it does not ask merely, which\namong them will have _some_ good or bad result, but which, among all\nthe actions possible to volition at any moment, will produce the best\n_total_ result. To assert that an action is a duty, is to assert that\nit is such a possible action, which will _always_, in certain known\ncircumstances, produce better results than any other. It follows that\nuniversal propositions of which duty is predicate, so far from being\nself-evident, always require a proof, which it is beyond our present\nmeans of knowledge ever to give (89-92). But (3) all that Ethics has\nattempted or can attempt, is to shew that certain actions, possible by\nvolition, _generally_ produce better or worse total results than any\nprobable alternative: and it must obviously be very difficult to shew\nthis with regard to the total results even in a comparatively near\nfuture; whereas that what has the best results in such a near future,\nalso has the best on the whole, is a point requiring an investigation\nwhich it has not received. If it is true, and if, accordingly, we give\nthe name of \u2018duty\u2019 to actions which _generally_ produce better total\nresults in the near future than any possible alternative, it may be\npossible to prove that a few of the commonest rules of duty are true,\nbut _only_ in certain conditions of society, which may be more or less\nuniversally presented in history; and such a proof is only possible\n_in some cases_ without a correct judgment of what things are good\nor bad in themselves--a judgment which has never yet been offered by\nethical writers. With regard to actions of which the _general_ utility\nis thus proved, the individual should _always_ perform them; but in\nother cases, where rules are commonly offered, he should rather judge\nof the probable results in his particular case, guided by a correct\nconception of what things are intrinsically good or bad (93-100).\n(4) In order that any action may be shewn to be a duty, it must be\nshewn to fulfil the above conditions; but the actions commonly called\n\u2018duties\u2019 do not fulfil them to any greater extent than \u2018expedient\u2019\nor \u2018interested\u2019 actions: by calling them \u2018duties\u2019 we only mean that\nthey have, _in addition_, certain non-ethical predicates. Similarly by\n\u2018virtue\u2019 is mainly meant a permanent disposition to perform \u2018duties\u2019\nin this restricted sense: and accordingly a virtue, if it is really\na virtue, must be good _as a means_, in the sense that it fulfils the\nabove conditions; but it is not _better_ as a means than non-virtuous\ndispositions; it generally has no value in itself; and, where it has,\nit is far from being the sole good or the best of goods. Accordingly\n\u2018virtue\u2019 is not, as is commonly implied, an unique _ethical_ predicate\n(101-109).\n\n\n\n\nCHAPTER VI.\n\nTHE IDEAL.\n\n\n=110.= The title of this chapter is ambiguous. When we call a state of\nthings \u2018ideal\u2019 we may mean three distinct things, which have only this\nin common: that we always do mean to assert, of the state of things in\nquestion, not only that it is good in itself, but that it is good in\nitself in a much higher degree than many other things. The first of\nthese meanings of \u2018ideal\u2019 is (1) that to which the phrase \u2018_The_ Ideal\u2019\nis most properly confined. By this is meant the _best_ state of things\n_conceivable_, the Summum Bonum or Absolute Good. It is in this sense\nthat a right conception of Heaven would be a right conception of the\nIdeal: we mean by the Ideal a state of things which would be absolutely\nperfect. But this conception may be quite clearly distinguished from a\nsecond, namely, (2) that of the best _possible_ state of things in this\nworld. This second conception may be identified with that which has\nfrequently figured in philosophy as the \u2018Human Good,\u2019 or the _ultimate_\nend towards which our action should be directed. It is in this sense\nthat Utopias are said to be Ideals. The constructor of an Utopia may\nsuppose many things to be possible, which are in fact impossible; but\nhe always assumes that some things, at least, are rendered impossible\nby natural laws, and hence his construction differs essentially\nfrom one which may disregard _all_ natural laws, however certainly\nestablished. At all events the question \u2018What is the best state of\nthings which we could _possibly_ bring about?\u2019 is quite distinct from\nthe question \u2018What would be the best state of things conceivable?\u2019\nBut, thirdly, we may mean by calling a state of things \u2018ideal\u2019 merely\n(3) that it is good in itself in a high degree. And it is obvious that\nthe question what things are \u2018ideal\u2019 in this sense is one which must\nbe answered before we can pretend to settle what is the Absolute or\nthe Human Good. It is with the Ideal, in this third sense, that this\nchapter will be principally concerned. Its main object is to arrive\nat some positive answer to the fundamental question of Ethics--the\nquestion: \u2018What things are goods or ends in themselves?\u2019 To this\nquestion we have hitherto obtained only a negative answer: the answer\nthat pleasure is certainly not the _sole_ good.\n\n\n=111.= I have just said that it is upon a correct answer to this\nquestion that correct answers to the two other questions, What is the\nAbsolute Good? and What is the Human Good? must depend; and, before\nproceeding to discuss it, it may be well to point out the relation\nwhich it has to these two questions.\n\n(1) It is just possible that the Absolute Good may be entirely\ncomposed of qualities which we cannot even imagine. This is possible,\nbecause, though we certainly do know a great many things that are\ngood-in-themselves, and good in a high degree, yet what is best does\nnot necessarily contain all the good things there are. That this is so\nfollows from the principle explained in Chap. I. (\u00a7\u00a7 18-22), to which\nit was there proposed that the name \u2018principle of organic unities\u2019\nshould be confined. This principle is that the intrinsic value of a\nwhole is neither identical with nor proportional to the sum of the\nvalues of its parts. It follows from this that, though in order to\nobtain the greatest possible sum of values in its parts, the Ideal\nwould necessarily contain all the things which have intrinsic value in\nany degree, yet the whole which contained all these parts might not be\nso valuable as some other whole, from which certain positive goods were\nomitted. But if a whole, which does not contain all positive goods, may\nyet be better than a whole which does, it follows that the best whole\n_may_ be one, which contains _none_ of the positive goods with which we\nare acquainted.\n\nIt is, therefore, _possible_ that we cannot discover what the Ideal\nis. But it is plain that, though this possibility cannot be denied,\nno one can have any right to assert that it is realised--that the\nIdeal _is_ something unimaginable. We cannot judge of the comparative\nvalues of things, unless the things we judge are before our minds. We\ncannot, therefore, be entitled to assert that anything, which we cannot\nimagine, would be better than some of the things which we can; although\nwe are also not entitled to deny the possibility that this may be the\ncase. Consequently our search for the Ideal must be limited to a search\nfor that one, among all the wholes composed of elements known to us,\nwhich seems to be better than all the rest. We shall never be entitled\nto assert that this whole is Perfection, but we shall be entitled to\nassert that it is _better_ than any other which may be presented as a\nrival.\n\nBut, since anything which we can have any _reason_ to think ideal\nmust be composed of things that are known to us, it is plain that\na comparative valuation of these must be our chief instrument\nfor deciding what is ideal. The best ideal we can construct will\nbe that state of things which contains the greatest number of\nthings having positive value, and which contains nothing evil or\nindifferent--_provided_ that the presence of none of these goods,\nor the absence of things evil or indifferent, seems to diminish the\nvalue of the whole. And, in fact, the chief defect of such attempts as\nhave been made by philosophers to construct an Ideal--to describe the\nKingdom of Heaven--seems to consist in the fact that they omit many\nthings of very great positive value, although it is plain that this\nomission does _not_ enhance the value of the whole. Where this is the\ncase, it may be confidently asserted that the ideal proposed is not\nideal. And the review of positive goods, which I am about to undertake,\nwill, I hope, shew that no ideals yet proposed are satisfactory. Great\npositive goods, it will appear, are so numerous, that any whole,\nwhich shall contain them all, must be of vast complexity. And though\nthis fact renders it difficult, or, humanly speaking, impossible, to\ndecide what is The Ideal, what is the absolutely best state of things\nimaginable, it is sufficient to condemn those Ideals, which are formed\nby omission, without any visible gain in consequence of such omission.\nPhilosophers seem usually to have sought only for the _best_ of single\nthings; neglecting the fact that a whole composed of two great goods,\neven though one of these be obviously inferior to the other, may yet be\noften seen to be decidedly superior to either by itself.\n\n(2) On the other hand, Utopias--attempted descriptions of a Heaven\nupon Earth--commonly suffer not only from this, but also from the\nopposite defect. They are commonly constructed on the principle of\nmerely omitting the great positive evils, which exist at present, with\nutterly inadequate regard to the goodness of what they retain: the\nso-called goods, to which they have regard, are, for the most part,\nthings which are, at best, mere means to good--things, such as freedom,\n_without_ which, possibly, nothing very good can exist in this world,\nbut which are of no value in themselves and are by no means certain\neven to produce anything of value. It is, of course, necessary to the\npurpose of their authors, whose object is merely to construct the\nbest that may be possible in this world, that they should include,\nin the state of things which they describe, many things, which are\nthemselves indifferent, but which, according to natural laws, seem\nto be absolutely necessary for the existence of anything which is\ngood. But, in fact, they are apt to include many things, of which\nthe necessity is by no means apparent, under the mistaken idea that\nthese things are goods-in-themselves, and not merely, here and now,\na means to good: while, on the other hand, they also omit from their\ndescription great positive goods, of which the attainment seems to be\nquite as possible as many of the changes which they recommend. That\nis to say, conceptions of the Human Good commonly err, not only, like\nthose of the Absolute Good, in omitting some great goods, but also by\nincluding things indifferent; and they both omit and include in cases\nwhere the limitations of natural necessity, by the consideration of\nwhich they are legitimately differentiated from conceptions of the\nAbsolute Good, will not justify the omission and inclusion. It is, in\nfact, obvious that in order to decide correctly at what state of things\nwe ought to aim, we must not only consider what results it is possible\nfor us to obtain, but also which, among equally possible results, will\nhave the greatest value. And upon this second enquiry the comparative\nvaluation of known goods has a no less important bearing than upon the\ninvestigation of the Absolute Good.\n\n\n=112.= The method which must be employed in order to decide the\nquestion \u2018What things have intrinsic value, and in what degrees?\u2019\nhas already been explained in Chap. III. (\u00a7\u00a7 55, 57). In order to\narrive at a correct decision on the first part of this question, it is\nnecessary to consider what things are such that, if they existed _by\nthemselves_, in absolute isolation, we should yet judge their existence\nto be good; and, in order to decide upon the relative _degrees_ of\nvalue of different things, we must similarly consider what comparative\nvalue seems to attach to the isolated existence of each. By employing\nthis method, we shall guard against two errors, which seem to have\nbeen the chief causes which have vitiated previous conclusions on the\nsubject. The first of these is (1) that which consists in supposing\nthat what seems absolutely necessary here and now, for the existence of\nanything good--what we cannot do without--is therefore good in itself.\nIf we isolate such things, which are mere means to good, and suppose\na world in which they alone, and nothing but they, existed, their\nintrinsic worthlessness becomes apparent. And, secondly, there is the\nmore subtle error (2) which consists in neglecting the principle of\norganic unities. This error is committed, when it is supposed, that,\nif one part of a whole has no intrinsic value, the value of the whole\nmust reside entirely in the other parts. It has, in this way, been\ncommonly supposed, that, if all valuable wholes could be seen to have\none and only one common property, the wholes must be valuable solely\n_because_ they possess this property; and the illusion is greatly\nstrengthened, if the common property in question seems, considered\nby itself, to have more value than the other parts of such wholes,\nconsidered by themselves. But, if we consider the property in question,\n_in isolation_, and then compare it with the whole, of which it forms\na part, it may become easily apparent that, existing by itself, the\nproperty in question has not nearly so much value, as has the whole to\nwhich it belongs. Thus, if we compare the value of a certain amount of\npleasure, _existing absolutely by itself_, with the value of certain\n\u2018enjoyments,\u2019 containing an equal amount of pleasure, it may become\napparent that the \u2018enjoyment\u2019 is much better than the pleasure, and\nalso, in some cases, much worse. In such a case it is plain that the\n\u2018enjoyment\u2019 does _not_ owe its value _solely_ to the pleasure it\ncontains, although it might easily have appeared to do so, when we\nonly considered the other constituents of the enjoyment, and seemed\nto see that, without the pleasure, they would have had no value. It\nis now apparent, on the contrary, that the whole \u2018enjoyment\u2019 owes its\nvalue quite equally to the presence of the other constituents, _even\nthough_ it may be true that the pleasure is the only constituent having\nany value by itself. And similarly, if we are told that all things\nowe their value solely to the fact that they are \u2018realisations of the\ntrue self,\u2019 we may easily refute this statement, by asking whether the\npredicate that is meant by \u2018realising the true self,\u2019 supposing that it\ncould exist alone, would have any value whatsoever. Either the _thing_,\nwhich does \u2018realise the true self,\u2019 has intrinsic value or it has not;\nand if it has, then it certainly does not owe its value solely to the\nfact that it realises the true self.\n\n\n=113.= If, now, we use this method of absolute isolation, and guard\nagainst these errors, it appears that the question we have to answer\nis far less difficult than the controversies of Ethics might have led\nus to expect. Indeed, once the meaning of the question is clearly\nunderstood, the answer to it, in its main outlines, appears to be so\nobvious, that it runs the risk of seeming to be a platitude. By far the\nmost valuable things, which we know or can imagine, are certain states\nof consciousness, which may be roughly described as the pleasures of\nhuman intercourse and the enjoyment of beautiful objects. No one,\nprobably, who has asked himself the question, has ever doubted that\npersonal affection and the appreciation of what is beautiful in Art\nor Nature, are good in themselves; nor, if we consider strictly what\nthings are worth having _purely for their own sakes_, does it appear\nprobable that any one will think that anything else has _nearly_ so\ngreat a value as the things which are included under these two heads.\nI have myself urged in Chap. III. (\u00a7 50) that the mere existence of\nwhat is beautiful does appear to have _some_ intrinsic value; but I\nregard it as indubitable that Prof. Sidgwick was so far right, in the\nview there discussed, that such mere existence of what is beautiful\nhas value, so small as to be negligible, in comparison with that which\nattaches to the _consciousness_ of beauty. This simple truth may,\nindeed, be said to be universally recognised. What has _not_ been\nrecognised is that it is the ultimate and fundamental truth of Moral\nPhilosophy. That it is only for the sake of these things--in order that\nas much of them as possible may at some time exist--that any one can\nbe justified in performing any public or private duty; that they are\nthe _raison d\u2019\u00eatre_ of virtue; that it is they--these complex wholes\n_themselves_, and not any constituent or characteristic of them--that\nform the rational ultimate end of human action and the sole criterion\nof social progress: these appear to be truths which have been generally\noverlooked.\n\nThat they are truths--that personal affections and aesthetic enjoyments\ninclude _all_ the greatest, and _by far_ the greatest, goods we can\nimagine, will, I hope, appear more plainly in the course of that\nanalysis of them, to which I shall now proceed. All the things, which I\nhave meant to include under the above descriptions, are highly complex\n_organic unities_; and in discussing the consequences, which follow\nfrom this fact, and the elements of which they are composed, I may hope\nat the same time both to confirm and to define my position.\n\n\n=114.= I. I propose to begin by examining what I have called aesthetic\nenjoyments, since the case of personal affections presents some\nadditional complications. It is, I think, universally admitted that the\nproper appreciation of a beautiful object is a good thing in itself;\nand my question is: What are the main elements included in such an\nappreciation?\n\n(1) It is plain that in those instances of aesthetic appreciation,\nwhich we think most valuable, there is included, not merely a bare\ncognition of what is beautiful in the object, but also some kind of\nfeeling or emotion. It is not sufficient that a man should merely see\nthe beautiful qualities in a picture and know that they are beautiful,\nin order that we may give his state of mind the highest praise. We\nrequire that he should also _appreciate_ the beauty of that which he\nsees and which he knows to be beautiful--that he should feel and see\n_its beauty_. And by these expressions we certainly mean that he should\nhave an appropriate emotion towards the beautiful qualities which he\ncognises. It is perhaps the case that all aesthetic emotions have some\ncommon quality; but it is certain that differences in the emotion seem\nto be appropriate to differences in the kind of beauty perceived: and\nby saying that different emotions are _appropriate_ to different kinds\nof beauty, we mean that the whole which is formed by the consciousness\nof that kind of beauty _together with_ the emotion appropriate to it,\nis better than if any other emotion had been felt in contemplating that\nparticular beautiful object. Accordingly we have a large variety of\ndifferent emotions, each of which is a necessary constituent in some\nstate of consciousness which we judge to be good. All of these emotions\nare essential elements in great positive goods; they are _parts_ of\norganic wholes, which have great intrinsic value. But it is important\nto observe that these wholes are organic, and that, hence, it does not\nfollow that the emotion, _by itself_, would have any value whatsoever,\nnor yet that, if it were directed to a different object, the whole thus\nformed might not be positively bad. And, in fact, it seems to be the\ncase that if we distinguish the emotional element, in any aesthetic\nappreciation, from the cognitive element, which accompanies it and\nis, in fact, commonly thought of as a part of the emotion; and if\nwe consider what value this emotional element would have, _existing\nby itself_, we can hardly think that it has any great value, even\nif it has any at all. Whereas, if the same emotion be directed to a\ndifferent object, if, for instance, it is felt towards an object that\nis positively ugly, the whole state of consciousness is certainly often\npositively bad in a high degree.\n\n\n=115.= (2) In the last paragraph I have pointed out the two facts,\nthat the presence of some emotion is necessary to give any very high\nvalue to a state of aesthetic appreciation, and that, on the other\nhand, this same emotion, in itself, may have little or no value: it\nfollows that these emotions give to the wholes of which they form a\npart a value far greater than that which they themselves possess.\nThe same is obviously true of the cognitive element which must be\ncombined with these emotions in order to form these highly valuable\nwholes; and the present paragraph will attempt to define what is meant\nby this cognitive element, so far as to guard against a possible\nmisunderstanding. When we talk of seeing a beautiful object, or, more\ngenerally, of the cognition or consciousness of a beautiful object,\nwe may mean by these expressions something which forms no part of\nany valuable whole. There is an ambiguity in the use of the term\n\u2018object,\u2019 which has probably been responsible for as many enormous\nerrors in philosophy and psychology as any other single cause. This\nambiguity may easily be detected by considering the proposition, which,\nthough a contradiction in terms, is obviously true: That when a man\nsees a beautiful picture, he may see nothing beautiful whatever. The\nambiguity consists in the fact that, by the \u2018object\u2019 of vision (or\ncognition), may be meant _either_ the qualities actually seen _or_ all\nthe qualities possessed by the thing seen. Thus in our case: when it\nis said that the picture is beautiful, it is meant that it contains\nqualities which are beautiful; when it is said that the man sees the\npicture, it is meant that he sees a great number of the qualities\ncontained in the picture; and when it is said that, nevertheless,\nhe sees nothing beautiful, it is meant that he does _not_ see those\nqualities of the picture which are beautiful. When, therefore, I speak\nof the cognition of a beautiful object, as an essential element in a\nvaluable aesthetic appreciation, I must be understood to mean only the\ncognition of _the beautiful qualities_ possessed by that object, and\n_not_ the cognition of other qualities of the object possessing them.\nAnd this distinction must itself be carefully distinguished from the\nother distinction expressed above by the distinct terms \u2018seeing the\nbeauty of a thing\u2019 and \u2018seeing its beautiful qualities.\u2019 By \u2018seeing\nthe beauty of a thing\u2019 we commonly mean the having an emotion towards\nits beautiful qualities; whereas in the \u2018seeing of its beautiful\nqualities\u2019 we do not include any emotion. By the cognitive element,\nwhich is equally necessary with emotion to the existence of a valuable\nappreciation, I mean merely the actual cognition or consciousness of\nany or all of an object\u2019s _beautiful qualities_--that is to say any or\nall of those elements in the object which possess any positive beauty.\nThat such a cognitive element is essential to a valuable whole may be\neasily seen, by asking: What value should we attribute to the proper\nemotion excited by hearing Beethoven\u2019s Fifth Symphony, if that emotion\nwere entirely unaccompanied by any consciousness, either of the notes,\nor of the melodic and harmonic relations between them? And that the\nmere _hearing_ of the Symphony, even accompanied by the appropriate\nemotion, is not sufficient, may be easily seen, if we consider what\nwould be the state of a man, who should hear all the notes, but should\n_not_ be aware of any of those melodic and harmonic relations, which\nare necessary to constitute the smallest beautiful elements in the\nSymphony.\n\n\n=116.= (3) Connected with the distinction just made between \u2018object\u2019\nin the sense of the qualities actually before the mind, and \u2018object\u2019\nin the sense of the whole thing which possesses the qualities actually\nbefore the mind, is another distinction of the utmost importance for\na correct analysis of the constituents necessary to a valuable whole.\nIt is commonly and rightly thought that to see beauty in a thing which\nhas no beauty is in some way inferior to seeing beauty in that which\nreally has it. But under this single description of \u2018seeing beauty\nin that which has no beauty,\u2019 two very different facts, and facts\nof very different value, may be included. We may mean _either_ the\nattribution to an object of really beautiful qualities which it does\nnot possess _or_ the feeling towards qualities, which the object does\npossess but which are in reality not beautiful, an emotion which is\nappropriate only to qualities really beautiful. Both these facts are\nof very frequent occurrence; and in most instances of emotion both\nno doubt occur together: but they are obviously quite distinct, and\nthe distinction is of the utmost importance for a correct estimate of\nvalues. The former may be called an error of judgment, and the latter\nan error of taste; but it is important to observe that the \u2018error\nof taste\u2019 commonly involves a false judgment _of value_; whereas the\n\u2018error of judgment\u2019 is merely a false judgment _of fact_.\n\nNow the case which I have called an error of taste, namely, where the\nactual qualities we admire (whether possessed by the \u2018object\u2019 or not)\nare ugly, can in any case have no value, except such as may belong\nto the emotion _by itself_; and in most, if not in all, cases it is\na considerable positive evil. In this sense, then, it is undoubtedly\nright to think that seeing beauty in a thing which has no beauty is\ninferior in value to seeing beauty where beauty really is. But the\nother case is much more difficult. In this case there is present all\nthat I have hitherto mentioned as necessary to constitute a great\npositive good: there is a cognition of qualities really beautiful,\ntogether with an appropriate emotion towards these qualities. There\ncan, therefore, be no doubt that we have here a great positive good.\nBut there is present also something else; namely, a belief that these\nbeautiful qualities exist, and that they exist in a certain relation\nto other things--namely, to some properties of the object to which\nwe attribute these qualities: and further the object of this belief\nis false. And we may ask, with regard to the whole thus constituted,\nwhether the presence of the belief, and the fact that what is believed\nis false, make any difference to its value? We thus get three different\ncases of which it is very important to determine the relative values.\nWhere both the cognition of beautiful qualities and the appropriate\nemotion are present we may _also_ have either, (1) a belief in the\nexistence of these qualities, of which the object, _i.e._ that they\nexist, is true: or (2) a mere cognition, without belief, when it is\n(_a_) true, (_b_) false, that the object of the cognition, _i.e._ the\nbeautiful qualities, exists: or (3) a belief in the existence of the\nbeautiful qualities, when they do not exist. The importance of these\ncases arises from the fact that the second defines the pleasures of\nimagination, including a great part of the appreciation of those\nworks of art which are _representative_; whereas the first contrasts\nwith these the appreciation of what is beautiful in Nature, and the\nhuman affections. The third, on the other hand, is contrasted with\nboth, in that it is chiefly exemplified in what is called misdirected\naffection; and it is possible also that the love of God, in the case of\na believer, should fall under this head.\n\n\n=117.= Now all these three cases, as I have said, have something in\ncommon, namely, that, in them all, we have a cognition of really\nbeautiful qualities together with an appropriate emotion towards\nthose qualities. I think, therefore, it cannot be doubted (nor is it\ncommonly doubted) that all three include great positive goods; they\nare all things of which we feel convinced that they are worth having\nfor their own sakes. And I think that the value of the second, in\neither of its two subdivisions, is precisely the same as the value\nof the element common to all three. In other words, in the case of\npurely imaginative appreciations we have merely the cognition of really\nbeautiful qualities together with the appropriate emotion; and the\nquestion, whether the object cognised exists or not, seems here, where\nthere is no belief either in its existence or in its non-existence, to\nmake absolutely no difference to the value of the total state. But it\nseems to me that the two other cases do differ in intrinsic value both\nfrom this one and from one another, even though the object cognised and\nthe appropriate emotion should be identical in all three cases. I think\nthat the additional presence of a belief in the reality of the object\nmakes the total state much better, if the belief is true; and worse,\nif the belief is false. In short, where there is belief, in the sense\nin which we _do_ believe in the existence of Nature and horses, and do\n_not_ believe in the existence of an ideal landscape and unicorns, the\n_truth_ of what is believed does make a great difference to the value\nof the organic whole. If this be the case, we shall have vindicated\nthe belief that _knowledge_, in the ordinary sense, as distinguished\non the one hand from belief in what is false and on the other from\nthe mere awareness of what is true, does contribute towards intrinsic\nvalue--that, at least in some cases, its presence as a part makes a\nwhole more valuable than it could have been without.\n\nNow I think there can be no doubt that we do judge that there is\na difference of value, such as I have indicated, between the three\ncases in question. We do think that the emotional contemplation of a\nnatural scene, supposing its qualities equally beautiful, is in some\nway a better state of things than that of a painted landscape: we\nthink that the world would be improved if we could substitute for the\nbest works of representative art _real_ objects equally beautiful. And\nsimilarly we regard a misdirected affection or admiration, even where\nthe error involved is a mere error of judgment and not an error of\ntaste, as in some way unfortunate. And further, those, at least, who\nhave a strong respect for truth, are inclined to think that a merely\npoetical contemplation of the Kingdom of Heaven _would_ be superior\nto that of the religious believer, _if_ it were the case that the\nKingdom of Heaven does not and will not really exist. Most persons,\non a sober, reflective judgment, would feel some hesitation even in\npreferring the felicity of a madman, convinced that the world was\nideal, to the condition either of a poet imagining an ideal world, or\nof themselves enjoying and appreciating the lesser goods which do and\nwill exist: But, in order to assure ourselves that these judgments are\nreally judgments of intrinsic value upon the question before us, and\nto satisfy ourselves that they are correct, it is necessary clearly to\ndistinguish our question from two others which have a very important\nbearing upon our total judgment of the cases in question.\n\n\n=118.= In the first place (_a_) it is plain that, where we believe, the\nquestion whether what we believe is true or false, will generally have\na most important bearing upon the value of our belief _as a means_.\nWhere we believe, we are apt to act upon our belief, in a way in which\nwe do not act upon our cognition of the events in a novel. The truth\nof what we believe is, therefore, very important as preventing the\npains of disappointment and still more serious consequences. And it\nmight be thought that a misdirected attachment was unfortunate solely\nfor this reason: that it leads us to count upon results, which the\nreal nature of its object is not of a kind to ensure. So too the Love\nof God, where, as usual, it includes the belief that he will annex to\ncertain actions consequences, either in this life or the next, which\nthe course of nature gives no reason to expect, may lead the believer\nto perform actions of which the actual consequences, supposing no\nsuch God to exist, may be much worse than he might otherwise have\neffected: and it might be thought that this was the sole reason (as it\nis a sufficient one) why we should hesitate to encourage the Love of\nGod, in the absence of any proof that he exists. And similarly it may\nbe thought that the only reason why beauty in Nature should be held\nsuperior to an equally beautiful landscape or imagination, is that\nits existence would ensure greater permanence and frequency in our\nemotional contemplation of that beauty. It is, indeed, certain that\nthe chief importance of most _knowledge_--of the truth of most of the\nthings which we believe--does, in this world, consist in its extrinsic\nadvantages: it is immensely valuable _as a means_.\n\nAnd secondly, (_b_) it may be the case that the existence of that which\nwe contemplate is itself a great positive good, so that, for this\nreason alone, the state of things described by saying, that the object\nof our emotion really exists, would be intrinsically superior to that\nin which it did not. This reason for superiority is undoubtedly of\ngreat importance in the case of human affections, where the object of\nour admiration is the mental qualities of an admirable person; for that\n_two_ such admirable persons should exist is greatly better than that\nthere should be only one: and it would also discriminate the admiration\nof inanimate nature from that of its representations in art, in so far\nas we may allow a small intrinsic value to the existence of a beautiful\nobject, apart from any contemplation of it. But it is to be noticed\nthat this reason would not account for any difference in value between\nthe cases where the truth was believed and that in which it was merely\ncognised, without either belief or disbelief. In other words, so far\nas this reason goes, the difference between the two subdivisions of\nour second class (that of imaginative contemplation) would be as great\nas between our first class and the second subdivision of our second.\nThe superiority of the mere _cognition_ of a beautiful object, when\nthat object also happened to exist, over the same cognition when the\nobject did not exist, would, on this count, be as great as that of the\n_knowledge_ of a beautiful object over the mere imagination of it.\n\n\n=119.= These two reasons for discriminating between the value of the\nthree cases we are considering, must, I say, be carefully distinguished\nfrom that, of which I am now questioning the validity, if we are to\nobtain a correct answer concerning this latter. The question I am\nputting is this: Whether the _whole_ constituted by the fact that\nthere is an emotional contemplation of a beautiful object, which\nboth is believed to be and is _real_, does not derive some of its\nvalue from the fact that the object _is_ real? I am asking whether\nthe value of this whole, _as a whole_, is not greater than that of\nthose which differ from it, _either_ by the absence of belief, with\nor without truth, _or_, belief being present, by the mere absence of\ntruth? I am not asking _either_ whether it is not superior to them as\na means (which it certainly is), _nor_ whether it may not contain a\nmore valuable _part_, namely, the existence of the object in question.\nMy question is solely whether the existence of its object does not\nconstitute an addition to the value of the whole, quite distinct from\nthe addition constituted by the fact that this whole does contain a\nvaluable part.\n\nIf, now, we put this question, I cannot avoid thinking that it should\nreceive an affirmative answer. We can put it clearly by the method of\nisolation; and the sole decision must rest with our reflective judgment\nupon it, as thus clearly put. We can guard against the bias produced\nby a consideration of value _as a means_ by supposing the case of an\nillusion as complete and permanent as illusions in this world never can\nbe. We can imagine the case of a single person, enjoying throughout\neternity the contemplation of scenery as beautiful, and intercourse\nwith persons as admirable, as can be imagined; while yet the whole of\nthe objects of his cognition are absolutely unreal. I think we should\ndefinitely pronounce the existence of a universe, which consisted\nsolely of such a person, to be _greatly_ inferior in value to one in\nwhich the objects, in the existence of which he believes, did really\nexist just as he believes them to do; and that it would be thus\ninferior _not only_ because it would lack the goods which consist in\nthe existence of the objects in question, but _also_ merely because\nhis belief would be false. That it would be inferior _for this reason\nalone_ follows if we admit, what also appears to me certain, that the\ncase of a person, merely imagining, without believing, the beautiful\nobjects in question, would, _although these objects really existed_,\nbe yet inferior to that of the person who also believed in their\nexistence. For here all the additional good, which consists in the\nexistence of the objects, is present, and yet there still seems to be\na great difference in value between this case and that in which their\nexistence is believed. But I think that my conclusion may perhaps be\nexhibited in a more convincing light by the following considerations.\n(1) It does not seem to me that the small degree of value which we may\nallow to the existence of beautiful inanimate objects is nearly equal\nin amount to the difference which I feel that there is between the\nappreciation (accompanied by belief) of such objects, when they really\nexist, and the purely imaginative appreciation of them when they do not\nexist. This inequality is more difficult to verify where the object\nis an admirable person, since a _great_ value must be allowed to his\nexistence. But yet I think it is not paradoxical to maintain that the\nsuperiority of reciprocal affection, where both objects are worthy and\nboth exist, over an unreciprocated affection, where both are worthy\nbut one does not exist, does not lie solely in the fact that, in the\nformer case, we have two good things instead of one, but also in the\nfact that each is such as the other believes him to be. (2) It seems to\nme that the important contribution to value made by true belief may be\nvery plainly seen in the following case. Suppose that a worthy object\nof affection does really exist and is believed to do so, but that there\nenters into the case this error of fact, that the qualities loved,\nthough exactly like, are yet not the _same_ which really do exist.\nThis state of things is easily imagined, and I think we cannot avoid\npronouncing that, _although_ both persons here exist, it is yet not so\nsatisfactory as where the very person loved and believed to exist is\nalso the one which actually does exist.\n\n\n=120.= If all this be so, we have, in this third section, added to our\ntwo former results the third result that a true belief in the reality\nof an object greatly increases the value of many valuable wholes.\nJust as in sections (1) and (2) it was maintained that aesthetic and\naffectionate emotions had little or no value apart from the cognition\nof appropriate objects, and that the cognition of these objects had\nlittle or no value apart from the appropriate emotion, so that the\nwhole, in which both were combined, had a value greatly in excess of\nthe sum of the values of its parts; so, according to this section, if\nthere be added to these wholes a true belief in the reality of the\nobject, the new whole thus formed has a value greatly in excess of the\nsum obtained by adding the value of the true belief, considered in\nitself, to that of our original wholes. This new case only differs from\nthe former in this, that, whereas the true belief, by itself, has quite\nas little value as either of the two other constituents taken singly,\nyet they, taken together, seem to form a whole of very great value,\nwhereas this is not the case with the two wholes which might be formed\nby adding the true belief to either of the others.\n\nThe importance of the result of this section seems to lie mainly in two\nof its consequences. (1) That it affords some justification for the\nimmense intrinsic value, which seems to be commonly attributed to the\nmere _knowledge_ of some truths, and which was expressly attributed\nto some kinds of knowledge by Plato and Aristotle. Perfect knowledge\nhas indeed competed with perfect love for the position of Ideal. If\nthe results of this section are correct, it appears that knowledge,\nthough having little or no value by itself, is an absolutely essential\nconstituent in the highest goods, and contributes immensely to their\nvalue. And it appears that this function may be performed not only\nby that case of knowledge, which we have chiefly considered, namely,\nknowledge of the reality of the beautiful object cognised, but also\nby knowledge of the numerical identity of this object with that which\nreally exists, and by the knowledge that the existence of that object\nis truly good. Indeed all knowledge, which is directly concerned with\nthe nature of the constituents of a beautiful object, would seem\ncapable of adding greatly to the value of the contemplation of that\nobject, although, by itself, such knowledge would have no value at\nall.--And (2) The second important consequence, which follows from\nthis section, is that the presence of true belief may, in spite of a\ngreat inferiority in the value of the emotion and the beauty of its\nobject, constitute with them a whole equal or superior in value to\nwholes, in which the emotion and beauty are superior, but in which a\ntrue belief is wanting or a false belief present. In this way we may\njustify the attribution of equal or superior value to an appreciation\nof an inferior real object, as compared with the appreciation of a\ngreatly superior object which is a mere creature of the imagination.\nThus a just appreciation of nature and of real persons may maintain its\nequality with an equally just appreciation of the products of artistic\nimagination, in spite of much greater beauty in the latter. And\nsimilarly though God may be admitted to be a more perfect object than\nany actual human being, the love of God may yet be inferior to human\nlove, _if_ God does not exist.\n\n\n=121.= (4) In order to complete the discussion of this first class\nof goods--goods which have an essential reference to _beautiful_\nobjects--it would be necessary to attempt a classification and\ncomparative valuation of all the different forms of beauty, a task\nwhich properly belongs to the study called Aesthetics. I do not,\nhowever, propose to attempt any part of this task. It must only be\nunderstood that I intend to include among the essential constituents of\nthe goods I have been discussing, every form and variety of beautiful\nobject, if only it be truly beautiful; and, _if_ this be understood,\nI think it may be seen that the consensus of opinion with regard to\nwhat is positively beautiful and what is positively ugly, and even with\nregard to great differences in degree of beauty, is quite sufficient to\nallow us a hope that we need not greatly err in our judgments of good\nand evil. In anything which is thought beautiful by any considerable\nnumber of persons, there is probably _some_ beautiful quality; and\ndifferences of opinion seem to be far more often due to exclusive\nattention, on the part of different persons, to different qualities\nin the same object, than to the positive error of supposing a quality\nthat is ugly to be really beautiful. When an object, which some think\nbeautiful, is denied to be so by others, the truth is _usually_ that it\nlacks some beautiful quality or is deformed by some ugly one, which\nengage the exclusive attention of the critics.\n\nI may, however, state two general principles, closely connected with\nthe results of this chapter, the recognition of which would seem to\nbe of great importance for the investigation of what things are truly\nbeautiful. The first of these is (1) a definition of beauty, of what\nis meant by saying that a thing is truly beautiful. The naturalistic\nfallacy has been quite as commonly committed with regard to beauty\nas with regard to good: its use has introduced as many errors into\nAesthetics as into Ethics. It has been even more commonly supposed that\nthe beautiful may be _defined_ as that which produces certain effects\nupon our feelings; and the conclusion which follows from this--namely,\nthat judgments of taste are merely _subjective_--that precisely the\nsame thing may, according to circumstances, be _both_ beautiful _and_\nnot beautiful--has very frequently been drawn. The conclusions of this\nchapter suggest a definition of beauty, which may partially explain\nand entirely remove the difficulties which have led to this error. It\nappears probable that the beautiful should be _defined_ as that of\nwhich the admiring contemplation is good in itself. That is to say: To\nassert that a thing is beautiful is to assert that the cognition of it\nis an essential element in one of the intrinsically valuable wholes\nwe have been discussing; so that the question, whether it is _truly_\nbeautiful or not, depends upon the _objective_ question whether the\nwhole in question is or is not truly good, and does not depend upon\nthe question whether it would or would not excite particular feelings\nin particular persons. This definition has the double recommendation\nthat it accounts both for the apparent connection between goodness\nand beauty and for the no less apparent difference between these two\nconceptions. It appears, at first sight, to be a strange coincidence,\nthat there should be two _different_ objective predicates of value,\n\u2018good\u2019 and \u2018beautiful,\u2019 which are nevertheless so related to one\nanother that whatever is beautiful is also good. But, if our definition\nbe correct, the strangeness disappears; since it leaves only one\n_unanalysable_ predicate of value, namely \u2018good,\u2019 while \u2018beautiful,\u2019\nthough not identical with, is to be defined by reference to this,\nbeing thus, at the same time, different from and necessarily connected\nwith it. In short, on this view, to say that a thing is beautiful\nis to say, not indeed that it is _itself_ good, but that it is a\nnecessary element in something which is: to prove that a thing is truly\nbeautiful is to prove that a whole, to which it bears a particular\nrelation as a part, is truly good. And in this way we should explain\nthe immense predominance, among objects commonly considered beautiful,\nof _material_ objects--objects of the external senses; since these\nobjects, though themselves having, as has been said, little or no\nintrinsic value, are yet essential constituents in the largest group of\nwholes which have intrinsic value. These wholes themselves may be, and\nare, also beautiful; but the comparative rarity, with which we regard\nthem as themselves _objects_ of contemplation, seems sufficient to\nexplain the association of beauty with external objects.\n\nAnd secondly (2) it is to be observed that beautiful objects are\nthemselves, for the most part, organic unities, in this sense, that\nthey are wholes of great complexity, such that the contemplation of\nany part, by itself, may have no value, and yet that, unless the\ncontemplation of the whole includes the contemplation of that part, it\nwill lose in value. From this it follows that there can be no single\ncriterion of beauty. It will never be true to say: This object owes its\nbeauty _solely_ to the presence of this characteristic; nor yet that:\nWherever this characteristic is present, the object must be beautiful.\nAll that can be true is that certain objects are beautiful, _because_\nthey have certain characteristics, in the sense that they would not\nbe beautiful _unless_ they had them. And it may be possible to find\nthat certain characteristics are more or less universally present in\nall beautiful objects, and are, in this sense, more or less important\nconditions of beauty. But it is important to observe that the very\nqualities, which differentiate one beautiful object from all others,\nare, if the object be truly beautiful, as _essential_ to its beauty,\nas those which it has in common with ever so many others. The object\nwould no more have the beauty it has, without its specific qualities,\nthan without those that are generic; and the generic qualities, _by\nthemselves_, would fail, as completely, to give beauty, as those which\nare specific.\n\n\n=122.= II. It will be remembered that I began this survey of great\nunmixed goods, by dividing all the greatest goods we know into the two\nclasses of aesthetic enjoyments, on the one hand, and the pleasures of\nhuman intercourse or of personal affection, on the other. I postponed\nthe consideration of the latter on the ground that they presented\nadditional complications. In what this additional complication\nconsists, will now be evident; and I have already been obliged to\ntake account of it, in discussing the contribution to value made by\ntrue belief. It consists in the fact that in the case of personal\naffection, the object itself is not _merely_ beautiful, while possessed\nof little or no intrinsic value, but is itself, in part at least,\nof great intrinsic value. All the constituents which we have found\nto be necessary to the most valuable aesthetic enjoyments, namely,\nappropriate emotion, cognition of truly beautiful qualities, and true\nbelief, are equally necessary here; but here we have the additional\nfact that the object must be not only truly beautiful, but also truly\ngood in a high degree.\n\nIt is evident that this additional complication only occurs in so far\nas there is included in the object of personal affection some of the\n_mental_ qualities of the person towards whom the affection is felt.\nAnd I think it may be admitted that, wherever the affection is most\nvaluable, the appreciation of mental qualities must form a large part\nof it, and that the presence of this part makes the whole far more\nvaluable than it could have been without it. But it seems very doubtful\nwhether this appreciation, by itself, can possess as much value as the\nwhole in which it is combined with an appreciation of the appropriate\n_corporeal_ expression of the mental qualities in question. It is\ncertain that in all actual cases of valuable affection, the bodily\nexpressions of character, whether by looks, by words, or by actions, do\nform a part of the object towards which the affection is felt, and that\nthe fact of their inclusion appears to heighten the value of the whole\nstate. It is, indeed, very difficult to imagine what the cognition of\nmental qualities _alone_, unaccompanied by _any_ corporeal expression,\nwould be like; and, in so far as we succeed in making this abstraction,\nthe whole considered certainly appears to have less value. I therefore\nconclude that the importance of an admiration of admirable mental\nqualities lies chiefly in the immense superiority of a whole, in which\nit forms a part, to one in which it is absent, and not in any high\ndegree of intrinsic value which it possesses by itself. It even appears\nto be doubtful, whether, in itself, it possesses so much value as the\nappreciation of mere corporeal beauty undoubtedly does possess; that is\nto say, whether the appreciation of what has great intrinsic value is\nso valuable as the appreciation of what is merely beautiful.\n\nBut further if we consider the nature of admirable mental qualities,\nby themselves, it appears that a proper appreciation of them involves\na reference to purely material beauty in yet another way. Admirable\nmental qualities do, if our previous conclusions are correct, consist\nvery largely in an emotional contemplation of beautiful objects;\nand hence the appreciation of them will consist essentially in the\ncontemplation of such contemplation. It is true that the most valuable\nappreciation of persons appears to be that which consists in the\nappreciation of their appreciation of other persons: but even here a\nreference to material beauty appears to be involved, _both_ in respect\nof the fact that what is appreciated in the last instance may be the\ncontemplation of what is merely beautiful, _and_ in respect of the fact\nthat the most valuable appreciation of a person appears to _include_\nan appreciation of his corporeal expression. Though, therefore, we\nmay admit that the appreciation of a person\u2019s attitude towards other\npersons, or, to take one instance, the love of love, is far the most\nvaluable good we know, and far more valuable than the mere love of\nbeauty, yet we can only admit this if the first be understood to\n_include_ the latter, in various degrees of directness.\n\nWith regard to the question what _are_ the mental qualities of which\nthe cognition is essential to the value of human intercourse, it is\nplain that they include, in the first place, all those varieties of\naesthetic appreciation, which formed our first class of goods. They\ninclude, therefore, a great variety of different emotions, each of\nwhich is appropriate to some different kind of beauty. But we must now\nadd to these the whole range of emotions, which are appropriate to\npersons, and which are different from those which are appropriate to\nmere corporeal beauty. It must also be remembered that just as these\nemotions have little value in themselves, and as the state of mind in\nwhich they exist may have its value greatly heightened, or may entirely\nlose it and become positively evil in a great degree, according as the\ncognitions accompanying the emotions are appropriate or inappropriate;\nso too the appreciation of these emotions, though it may have some\nvalue in itself, may yet form part of a whole which has far greater\nvalue or no value at all, according as it is or is not accompanied by\na perception of the appropriateness of the emotions to their objects.\nIt is obvious, therefore, that the study of what is valuable in human\nintercourse is a study of immense complexity; and that there may be\nmuch human intercourse which has little or no value, or is positively\nbad. Yet here too, as with the question what is beautiful, there seems\nno reason to doubt that a reflective judgment will in the main decide\ncorrectly both as to what are positive goods and even as to any _great_\ndifferences in value between these goods. In particular, it may be\nremarked that the emotions, of which the contemplation is essential\nto the greatest values, and which are also themselves appropriately\nexcited by such contemplation, appear to be those which are commonly\nmost highly prized under the name of affection.\n\n\n=123.= I have now completed my examination into the nature of those\ngreat positive goods, which do not appear to include among their\nconstituents anything positively evil or ugly, though they include\nmuch which is in itself indifferent. And I wish to point out certain\nconclusions which appear to follow, with regard to the nature of the\nSummum Bonum, or that state of things which would be the most perfect\nwe can conceive. Those idealistic philosophers, whose views agree\nmost closely with those here advocated, in that they deny pleasure to\nbe the sole good and regard what is completely good as having some\ncomplexity, have usually represented a purely spiritual state of\nexistence as the Ideal. Regarding matter as essentially imperfect, if\nnot positively evil, they have concluded that the total absence of\nall material properties is necessary to a state of perfection. Now,\naccording to what has been said, this view would be correct so far\nas it asserts that any great good must be _mental_, and so far as it\nasserts that a purely material existence, _by itself_, can have little\nor no value. The superiority of the spiritual over the material has,\nin a sense, been amply vindicated. But it does not follow, from this\nsuperiority, that a perfect state of things must be one, from which\nall material properties are rigidly excluded: on the contrary, if our\nconclusions are correct, it would seem to be the case that a state of\nthings, in which they are included, must be vastly better than any\nconceivable state in which they were absent. In order to see that this\nis so, the chief thing necessary to be considered is _exactly what it\nis_ which we declare to be good when we declare that the appreciation\nof beauty in Art and Nature is so. That this appreciation _is_ good,\nthe philosophers in question do not for the most part deny. But, if we\nadmit it, then we should remember Butler\u2019s maxim that: Everything is\nwhat it is, and not another thing. I have tried to shew, and I think\nit is too evident to be disputed, that such appreciation is an organic\nunity, a complex whole; and that, in its most undoubted instances,\npart of what is included in this whole is _a cognition of material\nqualities_, and particularly of a vast variety of what are called\n_secondary_ qualities. If, then, it is _this_ whole, which we know to\nbe good, and not another thing, then we know that material qualities,\neven though they be perfectly worthless in themselves, are yet\nessential constituents of what is far from worthless. What we know to\nbe valuable is the apprehension of just these qualities, and not of any\nothers; and, if we propose to subtract them from it, then what we have\nleft is _not_ that which we know to have value, but something else. And\nit must be noticed that this conclusion holds, even if my contention,\nthat a true belief in the existence of these qualities adds to the\nvalue of the whole in which it is included, be disputed. We should\nthen, indeed, be entitled to assert that the _existence_ of a material\nworld was wholly immaterial to perfection; but the fact that what we\nknew to be good was a cognition of _material qualities_ (though purely\nimaginary), would still remain. It must, then, be admitted on pain of\nself-contradiction--on pain of holding that things are not what they\nare, but something else--that a world, from which material qualities\nwere wholly banished, would be a world which lacked many, if not all,\nof those things, which we know most certainly to be great goods. That\nit _might_ nevertheless be a far better world than one which retained\nthese goods, I have already admitted (\u00a7 111 (1)). But in order to shew\nthat any such world _would_ be thus better, it would be necessary to\nshew that the retention of these things, though good in themselves,\nimpaired, in a more than equal degree, the value of some whole, to\nwhich they might belong; and the task of shewing this has certainly\nnever been attempted. Until it be performed, we are entitled to assert\nthat material qualities are a necessary constituent of the Ideal;\nthat, though something utterly unknown _might_ be better than any\nworld containing either them or any other good we know, yet we have no\nreason to suppose that anything whatever would be better than a state\nof things in which they were included. To deny and exclude matter,\nis to deny and exclude the best we know. That a thing may retain its\nvalue, while losing some of its qualities, is utterly untrue. All that\nis true is that the changed thing may have more value than, or as much\nvalue as, that of which the qualities have been lost. What I contend is\nthat nothing, which we _know_ to be good and which contains no material\nqualities, has such great value that we can declare it, _by itself_, to\nbe superior to the whole which would be formed by the addition to it of\nan appreciation of material qualities. That a _purely_ spiritual good\nmay be the _best_ of single things, I am not much concerned to dispute,\nalthough, in what has been said with regard to the nature of personal\naffection, I have given reasons for doubting it. But that by adding\nto it some appreciation of material qualities, which, though perhaps\ninferior by itself, is certainly a great positive good, we should\nobtain a greater sum of value, which no corresponding decrease in the\nvalue of the whole, as a whole, could counterbalance--this, I maintain,\nwe have certainly no reason to doubt.\n\n\n=124.= In order to complete this discussion of the main principles\ninvolved in the determination of intrinsic values, the chief remaining\ntopics, necessary to be treated, appear to be two. The first of these\nis the nature of great intrinsic _evils_, including what I may call\n_mixed_ evils; that is to say, those evil wholes, which nevertheless\ncontain, as essential elements, something positively good or beautiful.\nAnd the second is the nature of what I may similarly call _mixed_\ngoods; that is to say, those wholes, which, though intrinsically good\n_as wholes_, nevertheless contain, as essential elements, something\npositively evil or ugly. It will greatly facilitate this discussion,\nif I may be understood throughout to use the terms \u2018beautiful\u2019 and\n\u2018ugly,\u2019 not necessarily with reference to things of the kind which\nmost naturally occur to us as instances of what is beautiful and ugly,\nbut in accordance with my own proposed definition of beauty. Thus I\nshall use the word \u2018beautiful\u2019 to denote that of which the admiring\ncontemplation is good in itself; and \u2018ugly\u2019 to denote that of which the\nadmiring contemplation is evil in itself.\n\nI. With regard, then, to great positive evils, I think it is evident\nthat, if we take all due precautions to discover _precisely what_ those\nthings are, of which, _if they existed absolutely by themselves_, we\nshould judge the existence to be a great evil, we shall find most of\nthem to be organic unities of exactly the same nature as those which\nare the greatest positive goods. That is to say, they are cognitions of\nsome object, accompanied by some emotion. Just as neither a cognition\nnor an emotion, _by itself_, appeared capable of being greatly good,\nso (with one exception), neither a cognition nor an emotion, _by\nitself_, appears capable of being greatly evil. And just as a whole\nformed of both, even without the addition of any other element,\nappeared undoubtedly capable of being a great good, so such a whole,\n_by itself_, appears capable of being a great evil. With regard to the\n_third_ element, which was discussed as capable of adding greatly to\nthe value of a good, namely, _true belief_, it will appear that it has\ndifferent relations towards different kinds of evils. In some cases the\naddition of true belief to a positive evil seems to constitute a far\nworse evil; but in other cases it is not apparent that it makes any\ndifference.\n\nThe greatest positive evils may be divided into the following three\nclasses.\n\n\n=125.= (1) The first class consists of those evils, which seem always\nto include an enjoyment or admiring contemplation of things which\nare themselves either evil or ugly. That is to say these evils are\ncharacterised by the fact that they include precisely the same emotion,\nwhich is also essential to the greatest unmixed goods, from which they\nare differentiated by the fact that this emotion is directed towards an\ninappropriate object. In so far as this emotion is either a slight good\nin itself or a slightly beautiful object, these evils would therefore\nbe cases of what I have called \u2018mixed\u2019 evils; but, as I have already\nsaid, it seems very doubtful whether an emotion, completely isolated\nfrom its object, has either value or beauty: it certainly has not much\nof either. It is, however, important to observe that the very same\nemotions, which are often loosely talked of as the greatest or the only\ngoods, may be essential constituents of the very worst wholes: that,\naccording to the nature of the cognition which accompanies them, they\nmay be conditions either of the greatest good, or of the greatest evil.\n\nIn order to illustrate the nature of evils of this class, I may take\ntwo instances--cruelty and lasciviousness. That these are great\nintrinsic evils, we may, I think, easily assure ourselves, by imagining\nthe state of a man, whose mind is solely occupied by either of these\npassions, in their worst form. If we then consider what judgment we\nshould pass upon a universe which consisted _solely_ of minds thus\noccupied, without the smallest hope that there would ever exist in it\nthe smallest consciousness of any object other than those proper to\nthese passions, or any feeling directed to any such object, I think\nwe cannot avoid the conclusion that the existence of such a universe\nwould be a far worse evil than the existence of none at all. But, if\nthis be so, it follows that these two vicious states are not only, as\nis commonly admitted, bad as means, but also bad in themselves.--And\nthat they involve in their nature that complication of elements,\nwhich I have called a love of what is evil or ugly, is, I think, no\nless plain. With regard to the pleasures of lust, the nature of the\ncognition, by the presence of which they are to be defined, is somewhat\ndifficult to analyse. But it appears to include both cognitions of\norganic sensations and perceptions of states of the body, of which\nthe enjoyment is certainly an evil in itself. So far as these are\nconcerned, lasciviousness would, then, include in its essence an\nadmiring contemplation of what is ugly. But certainly one of its\ncommonest ingredients, in its worst forms, is an enjoyment of the same\nstate of mind in other people: and in this case it would therefore also\ninclude a love of what is evil. With regard to cruelty, it is easy\nto see that an enjoyment of pain in other people is essential to it;\nand, as we shall see, when we come to consider pain, this is certainly\na love of evil: while, in so far as it also includes a delight in\nthe bodily signs of agony, it would also comprehend a love of what\nis ugly. In both cases, it should be observed, the evil of the state\nis heightened not only by an increase in the evil or ugliness of the\nobject, but also by an increase in the enjoyment.\n\nIt might be objected, in the case of cruelty, that our disapproval of\nit, even in the isolated case supposed, where no considerations of\nits badness as a means could influence us, may yet be really directed\nto the pain of the persons, which it takes delight in contemplating.\nThis objection may be met, in the first place, by the remark that it\nentirely fails to explain the judgment, which yet, I think, no one, on\nreflection, will be able to avoid making, that even though the amount\nof pain contemplated be the same, yet the greater the delight in its\ncontemplation, the worse the state of things. But it may also, I think,\nbe met by notice of a fact, which we were unable to urge in considering\nthe similar possibility with regard to goods--namely the possibility\nthat the reason why we attribute greater value to a worthy affection\nfor a _real_ person, is that we take into account the additional good\nconsisting in the existence of that person. We may I think urge, in\nthe case of cruelty, that its intrinsic odiousness is equally great,\nwhether the pain contemplated really exists or is purely imaginary. I,\nat least, am unable to distinguish that, in this case, the presence\nof _true belief_ makes any difference to the intrinsic value of the\nwhole considered, although it undoubtedly may make a great difference\nto its value _as a means_. And so also with regard to other evils of\nthis class: I am unable to see that a true belief in the _existence_\nof their objects makes any difference in the degree of their positive\ndemerits. On the other hand, the presence of another class of beliefs\nseems to make a considerable difference. When we enjoy what is evil\nor ugly, in spite of our knowledge that it is so, the state of things\nseems considerably worse than if we made no judgment at all as to the\nobject\u2019s value. And the same seems also, strangely enough, to be the\ncase when we make a false judgment of value. When we admire what is\nugly or evil, believing that it is beautiful and good, this belief\nseems also to enhance the intrinsic vileness of our condition. It must,\nof course, be understood that, in both these cases, the judgment in\nquestion is merely what I have called a judgment of taste; that is to\nsay, it is concerned with the worth of the qualities actually cognised\nand not with the worth of the object, to which those qualities may be\nrightly or wrongly attributed.\n\nFinally it should be mentioned that evils of this class, _beside_ that\nemotional element (namely enjoyment and admiration) which they share\nwith great unmixed goods, appear always also to include some specific\nemotion, which does not enter in the same way into the constitution\nof any good. The presence of this specific emotion seems certainly\nto enhance the badness of the whole, though it is not plain that, by\nitself, it would be either evil or ugly.\n\n\n=126.= (2) The second class of great evils are undoubtedly mixed\nevils; but I treat them next, because, in a certain respect, they\nappear to be the _converse_ of the class last considered. Just as it\nis essential to this last class that they should include an emotion,\nappropriate to the cognition of what is good or beautiful, but directed\nto an inappropriate object; so to this second class it is essential\nthat they should include a cognition of what is good or beautiful, but\naccompanied by an inappropriate emotion. In short, just as the last\nclass may be described as cases of the love of what is evil or ugly, so\nthis class may be described as cases of the hatred of what is good or\nbeautiful.\n\nWith regard to these evils it should be remarked: First, that the vices\nof hatred, envy and contempt, where these vices are evil in themselves,\nappear to be instances of them; and that they are frequently\naccompanied by evils of the first class, for example, where a delight\nis felt in the pain of a good person. Where they are thus accompanied,\nthe whole thus formed is undoubtedly worse than if either existed\nsingly.\n\nAnd secondly: That in their case a true belief in the existence of the\ngood or beautiful object, which is hated, does appear to enhance the\nbadness of the whole, in which it is present. Undoubtedly also, as in\nour first class, the presence of a true belief as to the _value_ of the\nobjects contemplated, increases the evil. But, contrary to what was the\ncase in our first class, a _false_ judgment of value appears to lessen\nit.\n\n\n=127.= (3) The third class of great positive evils appears to be the\nclass of _pains_.\n\nWith regard to these it should first be remarked that, as in the case\nof pleasure, it is not pain itself, but only the consciousness of pain,\ntowards which our judgments of value are directed. Just as in Chap.\nIII., it was said that pleasure, however intense, which no one felt,\nwould be no good at all; so it appears that pain, however intense, of\nwhich there was no consciousness, would be no evil at all.\n\nIt is, therefore, only the consciousness of intense pain, which can\nbe maintained to be a great evil. But that this, _by itself_, may be\na great evil, I cannot avoid thinking. The case of pain thus seems to\ndiffer from that of pleasure: for the mere consciousness of pleasure,\nhowever intense, does not, _by itself_, appear to be a _great_ good,\neven if it has some slight intrinsic value. In short, pain (if we\nunderstand by this expression, the consciousness of pain) appears to\nbe a far worse evil than pleasure is a good. But, if this be so, then\n_pain_ must be admitted to be an exception from the rule which seems to\nhold both of all _other_ great evils and of _all_ great goods: namely\nthat they are all organic unities to which _both_ a cognition of an\nobject _and_ an emotion directed towards that object are essential. In\nthe case of pain and of pain alone, it seems to be true that a mere\ncognition, by itself, may be a great evil. It is, indeed, _an_ organic\nunity, since it involves both the cognition and the object, neither of\nwhich, by themselves, has either merit or demerit. But it is a less\ncomplex organic unity than any other great evil and than any great\ngood, _both_ in respect of the fact that it does not involve, _beside_\nthe cognition, an emotion directed towards its object, _and also_ in\nrespect of the fact that the _object_ may here be absolutely simple,\nwhereas in most, if not all, other cases, the object itself is highly\ncomplex.\n\nThis want of analogy between the relation of pain to intrinsic evil and\nof pleasure to intrinsic good, seems also to be exhibited in a second\nrespect. Not only is it the case that consciousness of intense pain\nis, by itself, a great evil, whereas consciousness of intense pleasure\nis, by itself, no great good; but also the _converse_ difference\nappears to hold of the contribution which they make to the value of\nthe whole, when they are combined respectively with another great evil\nor with a great good. That is to say, the presence of pleasure (though\nnot in proportion to its intensity) does appear to enhance the value\nof a whole, in which it is combined with any of the great unmixed\ngoods which we have considered: it might even be maintained that it\nis _only_ wholes, in which _some_ pleasure is included, that possess\nany great value: it is certain, at all events, that the presence of\npleasure makes a contribution to the value of good wholes greatly in\nexcess of its own intrinsic value. On the contrary, if a feeling of\npain be combined with any of the evil states of mind which we have been\nconsidering, the difference which its presence makes to the value of\nthe whole, _as a whole_, seems to be rather for the better than the\nworse: in any case, the only additional evil which it introduces, is\nthat which it, by itself, intrinsically constitutes. Thus, whereas pain\nis _in itself_ a great evil, but makes no addition to the badness of a\nwhole, in which it is combined with some other bad thing, except that\nwhich consists in its own intrinsic badness; pleasure, conversely, is\nnot _in itself_ a great good, but does make a great addition to the\ngoodness of a whole in which it is combined with a good thing, quite\napart from its own intrinsic value.\n\n\n=128.= But finally, it must be insisted that pleasure and pain are\ncompletely analogous in this: that we cannot assume either that the\npresence of pleasure always makes a state of things better _on the\nwhole_, or that the presence of pain always makes it worse. This is the\ntruth which is most liable to be overlooked with regard to them; and\nit is because this is true, that the common theory, that pleasure is\nthe only good and pain the only evil, has its grossest consequences in\nmisjudgments of value. Not only is the pleasantness of a state _not_\nin proportion to its intrinsic worth; it may even add positively to\nits vileness. We do not think the successful hatred of a villain the\nless vile and odious, because he takes the keenest delight in it; nor\nis there the least need, in logic, why we should think so, apart from\nan unintelligent prejudice in favour of pleasure. In fact it seems to\nbe the case that wherever pleasure is added to an evil state of either\nof our first two classes, the whole thus formed is _always_ worse than\nif no pleasure had been there. And similarly with regard to pain. If\npain be added to an evil state of either of our first two classes, the\nwhole thus formed is _always_ better, _as a whole_, than if no pain\nhad been there; though here, if the pain be too intense, since that\nis a great evil, the state may not be better _on the whole_. It is in\nthis way that the theory of vindictive punishment may be vindicated.\nThe infliction of pain on a person whose state of mind is bad may, if\nthe pain be not too intense, create a state of things that is better\n_on the whole_ than if the evil state of mind had existed unpunished.\nWhether such a state of things can ever constitute a _positive_ good,\nis another question.\n\n\n=129.= II. The consideration of this other question belongs properly to\nthe second topic, which was reserved above for discussion namely the\ntopic of \u2018mixed\u2019 goods. \u2018Mixed\u2019 goods were defined above as things,\nwhich, though positively good _as wholes_, nevertheless contain, as\nessential elements, something intrinsically evil or ugly. And there\ncertainly seem to be such goods. But for the proper consideration\nof them, it is necessary to take into account a new distinction the\ndistinction just expressed as being between the value which a thing\npossesses \u2018_as a whole_,\u2019 and that which it possesses \u2018_on the whole_.\u2019\n\nWhen \u2018mixed\u2019 goods were defined as things positively good _as wholes_,\nthe expression was ambiguous. It was meant that they were positively\ngood _on the whole_; but it must now be observed that the value which\na thing possesses _on the whole_ may be said to be equivalent to the\nsum of the value which it possesses _as a whole_, _together with_ the\nintrinsic values which may belong to any of its parts. In fact, by the\n\u2018value which a thing possesses as a whole,\u2019 there may be meant two\nquite distinct things. There may be meant either (1) That value which\narises solely _from the combination_ of two or more things; or else (2)\nThe total value formed by the addition to (1) of any intrinsic values\nwhich may belong to the things combined. The meaning of the distinction\nmay perhaps be most easily seen by considering the supposed case of\nvindictive punishment. If it is true that the combined existence of two\nevils may yet constitute a less evil than would be constituted by the\nexistence of either singly, it is plain that this can only be because\nthere arises from the combination a positive good which is greater than\nthe _difference_ between the sum of the two evils and the demerit of\neither singly: this positive good would then be the value of the whole,\n_as a whole_, in sense (1). Yet if this value be not so great a good\nas the sum of the two evils is an evil, it is plain that the value of\nthe whole state of things will be a positive evil; and this value is\nthe value of the whole, _as a whole_, in sense (2). Whatever view may\nbe taken with regard to the particular case of vindictive punishment,\nit is plain that we have here _two distinct things_, with regard to\n_either_ of which a separate question may be asked in the case of every\norganic unity. The first of these two things may be expressed as _the\ndifference_ between the value _of the whole thing_ and the sum of the\nvalue of its parts. And it is plain that where the parts have little or\nno intrinsic value (as in our first class of goods, \u00a7\u00a7 114, 115), this\ndifference will be nearly or absolutely identical with the value of\nthe whole thing. The distinction, therefore, only becomes important in\nthe case of wholes, of which one or more parts have a great intrinsic\nvalue, positive or negative. The first of these cases, that of a\nwhole, in which one part has a great _positive_ value, is exemplified\nin our 2nd and 3rd classes of great unmixed goods (\u00a7\u00a7 120, 122); and\nsimilarly the Summum Bonum is a whole of which _many_ parts have a\ngreat _positive_ value. Such cases, it may be observed, are also very\nfrequent and very important objects of Aesthetic judgment; since the\nessential distinction between the \u2018classical\u2019 and the \u2018romantic\u2019 styles\nconsists in the fact that the former aims at obtaining the greatest\npossible value for the whole, _as a whole_, in sense (1), whereas\nthe latter sacrifices this in order to obtain the greatest possible\nvalue for some _part_, which is itself an organic unity. It follows\nthat we cannot declare either style to be necessarily superior, since\nan equally good result _on the whole_, or \u2018as a whole\u2019 in sense (2),\nmay be obtained by either method; but the distinctively _aesthetic_\ntemperament seems to be characterised by a tendency to prefer a good\nresult obtained by the classical, to an equally good result obtained by\nthe romantic method.\n\n\n=130.= But what we have now to consider are cases of wholes, in which\none or more parts have a great _negative_ value--are great positive\nevils. And first of all, we may take the _strongest_ cases, like that\nof retributive punishment, in which we have a whole, exclusively\ncomposed of two great positive evils--wickedness and pain. Can such a\nwhole ever be positively good _on the whole_?\n\n(1) I can see no reason to think that such wholes ever are positively\ngood _on the whole_. But from the fact that they may, nevertheless, be\nless evils, than either of their parts taken singly, it follows that\nthey have a characteristic which is most important for the correct\ndecision of practical questions. It follows that, quite apart from\n_consequences_ or any value which an evil may have as a mere means,\nit may, _supposing_ one evil already exists, be worth while to create\nanother, since, by the mere creation of this second, there may be\nconstituted a whole less bad than if the original evil had been left\nto exist by itself. And similarly, with regard to all the wholes which\nI am about to consider, it must be remembered, that, even if they are\nnot goods _on the whole_, yet, where an evil already exists, as in this\nworld evils do exist, the existence of the other part of these wholes\nwill constitute a thing desirable _for its own sake_--that is to say,\nnot merely a means to future goods, but one of the _ends_ which must\nbe taken into account in estimating what that best possible state of\nthings is, to which every right action must be a means.\n\n\n=131.= (2) But, as a matter of fact, I cannot avoid thinking that there\nare wholes, containing something positively evil and ugly, which are,\nnevertheless, great positive goods on the whole. Indeed, it appears\nto be to this class that those instances of virtue, which contain\nanything intrinsically good, chiefly belong. It need not, of course,\nbe denied that there is sometimes included in a virtuous disposition\nmore or less of those unmixed goods which were first discussed--that\nis to say, a real love of what is good or beautiful. But the typical\nand characteristic virtuous dispositions, so far as they are not\nmere means, seem rather to be examples of mixed goods. We may take\nas instances (_a_) Courage and Compassion, which seem to belong to\nthe second of the three classes of virtues distinguished in our last\nchapter (\u00a7 107); and (_b_) the specifically \u2018moral\u2019 sentiment, by\nreference to which the third of those three classes was defined (\u00a7 108).\n\nCourage and compassion, in so far as they contain an intrinsically\ndesirable state of mind, seem to involve essentially a cognition of\nsomething evil or ugly. In the case of courage the object of the\ncognition may be an evil of any of our three classes; in the case of\ncompassion, the proper object is pain. Both these virtues, accordingly,\nmust contain precisely the same cognitive element, which is also\nessential to evils of class (1); and they are differentiated from these\nby the fact that the emotion directed to these objects is, in their\ncase, an emotion of the same kind which was essential to evils of class\n(2). In short, just as evils of class (2) seemed to consist in a hatred\nof what was good or beautiful, and evils of class (1) in a love of what\nwas evil or ugly; so these virtues involve a _hatred_ of what is evil\nor ugly. Both these virtues do, no doubt, also contain other elements,\nand, among these, each contains its specific emotion; but that their\nvalue does not depend solely upon these other elements, we may easily\nassure ourselves, by considering what we should think of an attitude of\nendurance or of defiant contempt toward an object intrinsically good\nor beautiful, or of the state of a man whose mind was filled with pity\nfor the happiness of a worthy admiration. Yet pity for the undeserved\nsufferings of others, endurance of pain to ourselves, and a defiant\nhatred of evil dispositions in ourselves or in others, seem to be\nundoubtedly admirable in themselves; and if so, there are admirable\nthings, which must be lost, if there were no cognition of evil.\n\nSimilarly the specifically \u2018moral\u2019 sentiment, in all cases where it has\nany considerable intrinsic value, appears to include a hatred of evils\nof the first and second classes. It is true that the emotion is here\nexcited by the idea that an action is right or wrong; and hence the\nobject of the idea which excites it is generally not an intrinsic evil.\nBut, as far as I can discover, the emotion with which a conscientious\nman views a real or imaginary right action, contains, as an essential\nelement, the same emotion with which he views a wrong one: it seems,\nindeed, that this element is necessary to make his emotion specifically\n_moral_. And the specifically moral emotion excited by the idea of a\nwrong action, seems to me to contain essentially a more or less vague\ncognition of the kind of intrinsic evils, which are usually caused\nby wrong actions, whether they would or would not be caused by the\nparticular action in question. I am, in fact, unable to distinguish, in\nits main features, the moral sentiment excited by the idea of rightness\nand wrongness, wherever it is intense, from the total state constituted\nby a cognition of something intrinsically evil together with the\nemotion of hatred directed towards it. Nor need we be surprised that\nthis mental state should be the one chiefly associated with the idea\nof rightness, if we reflect on the nature of those actions which are\nmost commonly recognised as duties. For by far the greater part of the\nactions, of which we commonly think as duties, are _negative_: what we\nfeel to be our duty is to _abstain_ from some action to which a strong\nnatural impulse tempts us. And these wrong actions, in the avoidance\nof which duty consists, are usually such as produce, very immediately,\nsome bad consequence in pain to others; while, in many prominent\ninstances, the inclination, which prompts us to them, is itself an\nintrinsic evil, containing, as where the impulse is lust or cruelty,\nan anticipatory enjoyment of something evil or ugly. That right action\ndoes thus so frequently entail the suppression of some evil impulse,\nis necessary to explain the plausibility of the view that virtue\n_consists_ in the control of passion by reason. Accordingly, the truth\nseems to be that, whenever a strong moral emotion is excited by the\nidea of rightness, this emotion is accompanied by a vague cognition of\nthe kind of evils usually suppressed or avoided by the actions which\nmost frequently occur to us as instances of duty; and that the emotion\nis directed towards this evil quality. We may, then, conclude that the\nspecific moral emotion owes almost all its intrinsic value to the fact\nthat it includes a cognition of evils accompanied by a hatred of them:\nmere rightness, whether truly or untruly attributed to an action, seems\nincapable of forming the object of an emotional contemplation, which\nshall be any great good.\n\n\n=132.= If this be so, then we have, in many prominent instances of\nvirtue, cases of a whole, greatly good in itself, which yet contains\nthe cognition of something, whereof the existence would be a great\nevil: a great good is absolutely dependent for its value, upon its\ninclusion of something evil or ugly, although it does not owe its\nvalue _solely_ to this element in it. And, in the case of virtues,\nthis evil object does, in general, actually exist. But there seems no\nreason to think that, when it does exist, the whole state of things\nthus constituted is therefore the better _on the whole_. What seems\nindubitable, is only that the feeling contemplation of an object, whose\nexistence _would_ be a great evil, or which is ugly, may be essential\nto a valuable whole. We have another undoubted instance of this in the\nappreciation of tragedy. But, in tragedy, the sufferings of Lear, and\nthe vice of Iago may be purely imaginary. And it seems certain that,\nif they really existed, the evil thus existing, while it must detract\nfrom the good consisting in a proper feeling towards them, will add\nno positive value to that good great enough to counterbalance such a\nloss. It does, indeed, seem that the existence of a true belief in the\nobject of these mixed goods does add _some_ value to the whole in which\nit is combined with them: a conscious compassion for real suffering\nseems to be better, _as a whole_, than a compassion for sufferings\nmerely imaginary; and this may well be the case, even though the evil\ninvolved in the actual suffering makes the total state of things bad\n_on the whole_. And it certainly seems to be true that a _false_ belief\nin the actual existence of its object makes a worse mixed good than\nif our state of mind were that with which we normally regard pure\nfiction. Accordingly we may conclude that the only mixed goods, which\nare positively good _on the whole_, are those in which the object is\nsomething which _would_ be a great evil, if it existed, or which _is_\nugly.\n\n\n=133.= With regard, then, to those mixed goods, which consist in an\nappropriate attitude of the mind towards things evil or ugly, and which\ninclude among their number the greater part of such virtues as have any\nintrinsic value whatever, the following three conclusions seem to be\nthose chiefly requiring to be emphasized:--\n\n(1) There seems no reason to think that where the object is a thing\nevil in itself, which _actually exists_, the total state of things is\never positively _good on the whole_. The appropriate mental attitude\ntowards a really existing evil contains, of course, an element which\nis absolutely identical with the same attitude towards the same evil,\nwhere it is purely imaginary. And this element, which is common to\nthe two cases, may be a great positive good, on the whole. But there\nseems no reason to doubt that, where the evil is _real_, the amount of\nthis real evil is always sufficient to reduce the total sum of value\nto a negative quantity. Accordingly we have no reason to maintain the\nparadox that an ideal world would be one in which vice and suffering\nmust exist in order that it may contain the goods consisting in the\nappropriate emotion towards them. It is not a positive good that\nsuffering should exist, in order that we may compassionate it; or\nwickedness, that we may hate it. There is no reason to think that any\nactual evil whatsoever would be contained in the Ideal. It follows that\nwe cannot admit the actual validity of any of the arguments commonly\nused in Theodicies; no such argument succeeds in justifying the fact\nthat there does exist even the smallest of the many evils which this\nworld contains. The most that can be said for such arguments is that,\nwhen they make appeal to the principle of organic unity, their appeal\nis valid _in principle_. It _might_ be the case that the existence of\nevil was necessary, not merely as a means, but analytically, to the\nexistence of the greatest good. But we have no reason to think that\nthis _is_ the case in any instance whatsoever.\n\nBut (2) there _is_ reason to think that the cognition of things evil\nor ugly, which are purely imaginary, is essential to the Ideal. In this\ncase the burden of proof lies the other way. It cannot be doubted that\nthe appreciation of tragedy is a great positive good; and it seems\nalmost equally certain that the virtues of compassion, courage, and\nself-control contain such goods. And to all these the cognition of\nthings which would be evil, if they existed, is analytically necessary.\nHere then we have things of which the existence must add value to\nany whole in which they are contained; nor is it possible to assure\nourselves that any whole, from which they were omitted, would thereby\ngain more in its value _as a whole_, than it would lose by their\nomission. We have no reason to think that any whole, which did not\ncontain them, would be so good _on the whole_ as some whole in which\nthey were obtained. The case for their inclusion in the Ideal is as\nstrong as that for the inclusion of material qualities (\u00a7 123, above).\n_Against_ the inclusion of these goods nothing can be urged except a\nbare possibility.\n\nFinally (3) it is important to insist that, as was said above, these\nmixed virtues have a great practical value, in addition to that which\nthey possess either in themselves or as mere means. Where evils do\nexist, as in this world they do, the fact that they are known and\nproperly appreciated, constitutes a state of things having greater\nvalue _as a whole_ even than the same appreciation of purely imaginary\nevils. This state of things, it has been said, is never positively\ngood _on the whole_; but where the evil, which reduces its total value\nto a negative quantity, already unavoidably exists, to obtain the\nintrinsic value which belongs to it _as a whole_ will obviously produce\na better state of things than if the evil had existed by itself,\nquite apart from the good element in it which is identical with the\nappreciation of imaginary evils, and from any ulterior consequences\nwhich its existence may bring about. The case is here the same as with\nretributive punishment. Where an evil already exists, it is well that\nit should be pitied or hated or endured, according to its nature; just\nas it may be well that some evils should be punished. Of course, as in\nall practical cases, it often happens that the attainment of this good\nis incompatible with the attainment of another and a greater one. But\nit is important to insist that we have here a real intrinsic value,\nwhich must be taken into account in calculating that greatest possible\nbalance of intrinsic value, which it is always our duty to produce.\n\n\n=134.= I have now completed such remarks as seemed most necessary to\nbe made concerning intrinsic values. It is obvious that for the proper\nanswering of this, the fundamental question of Ethics, there remains\na field of investigation as wide and as difficult, as was assigned\nto Practical Ethics in my last chapter. There is as much to be said\nconcerning what results are intrinsically good, and in what degrees,\nas concerning what results it is possible for us to bring about: both\nquestions demand, and will repay, an equally patient enquiry. Many of\nthe judgments, which I have made in this chapter, will, no doubt, seem\nunduly arbitrary: it must be confessed that some of the attributions\nof intrinsic value, which have seemed to me to be true, do not display\nthat symmetry and system which is wont to be required of philosophers.\nBut if this be urged as an objection, I may respectfully point out\nthat it is none. We have no title whatever to assume that the truth on\nany subject-matter will display such symmetry as we desire to see--or\n(to use the common vague phrase) that it will possess any particular\nform of \u2018unity.\u2019 To search for \u2018unity\u2019 and \u2018system,\u2019 at the expense\nof truth, is not, I take it, the proper business of philosophy,\nhowever universally it may have been the practice of philosophers.\nAnd that all truths about the Universe possess to one another all\nthe various relations, which may be meant by \u2018unity,\u2019 can only be\nlegitimately asserted, when we have carefully distinguished those\nvarious relations and discovered what those truths are. In particular,\nwe can have no title to assert that ethical truths are \u2018unified\u2019 in\nany particular manner, except in virtue of an enquiry conducted by\nthe method which I have endeavoured to follow and to illustrate. The\nstudy of Ethics would, no doubt, be far more simple, and its results\nfar more \u2018systematic,\u2019 if, for instance, pain were an evil of exactly\nthe same magnitude as pleasure is a good; but we have no reason\nwhatever to assume that the Universe is such that ethical truths must\ndisplay this kind of symmetry: no argument against my conclusion,\nthat pleasure and pain do _not_ thus correspond, can have any weight\nwhatever, failing a careful examination of the instances which have\nled me to form it. Nevertheless I am content that the results of this\nchapter should be taken rather as illustrating the method which must\nbe pursued in answering the fundamental question of Ethics, and the\nprinciples which must be observed, than as giving the correct answer\nto that question. That things intrinsically good or bad are many and\nvarious; that most of them are \u2018organic unities,\u2019 in the peculiar and\ndefinite sense to which I have confined the term; and that our only\nmeans of deciding upon their intrinsic value and its degree, is by\ncarefully distinguishing exactly what the thing is, about which we ask\nthe question, and then looking to see whether it has or has not the\nunique predicate \u2018good\u2019 in any of its various degrees: these are the\nconclusions, upon the truth of which I desire to insist. Similarly, in\nmy last chapter, with regard to the question \u2018What ought we to do?\u2019\nI have endeavoured rather to shew exactly what is the meaning of the\nquestion, and what difficulties must consequently be faced in answering\nit, than to prove that any particular answers are true. And that these\ntwo questions, having precisely the nature which I have assigned to\nthem, are _the_ questions which it is the object of Ethics to answer,\nmay be regarded as the main result of the preceding chapters. These\nare the questions which ethical philosophers have always been mainly\nconcerned to answer, although they have not recognised what their\nquestion was--what predicate they were asserting to attach to things.\nThe practice of asking what things are virtues or duties, without\ndistinguishing what these terms mean; the practice of asking what ought\nto be here and now, without distinguishing whether as means or end--for\nits own sake or for that of its results; the search for one single\n_criterion_ of right or wrong, without the recognition that in order\nto discover a criterion we must first know what things _are_ right or\nwrong; and the neglect of the principle of \u2018organic unities\u2019--these\nsources of error have hitherto been almost universally prevalent in\nEthics. The conscious endeavour to avoid them all, and to apply to\nall the ordinary objects of ethical judgment these two questions and\nthese only: Has it intrinsic value? and Is it a means to the best\npossible?--this attempt, so far as I know, is entirely new; and its\nresults, when compared with those habitual to moral philosophers, are\ncertainly sufficiently surprising: that to Common Sense they will not\nappear so strange, I venture to hope and believe. It is, I think,\nmuch to be desired that the labour commonly devoted to answering such\nquestions as whether certain \u2018ends\u2019 are more or less \u2018comprehensive\u2019\nor more or less \u2018consistent\u2019 with one another--questions, which, even\nif a precise meaning were given to them, are wholly irrelevant to the\nproof of any ethical conclusion--should be diverted to the separate\ninvestigation of these two clear problems.\n\n\n=135.= The main object of this chapter has been to define roughly\nthe class of things, among which we may expect to find either great\nintrinsic goods or great intrinsic evils; and particularly to point\nout that there is a vast variety of such things, and that the simplest\nof them are, with one exception, highly complex wholes, composed of\nparts which have little or no value in themselves. All of them involve\nconsciousness of an object, which is itself usually highly complex,\nand almost all involve also an emotional attitude towards this object;\nbut, though they thus have certain characteristics in common, the vast\nvariety of qualities in respect of which they differ from one another\nare equally essential to their value: neither the generic character\nof all, nor the specific character of each, is either greatly good\nor greatly evil by itself; they owe their value or demerit, in each\ncase, to the presence of both. My discussion falls into three main\ndivisions, dealing respectively (1) with unmixed goods, (2) with evils,\nand (3) with mixed goods. (1) Unmixed goods may all be said to consist\nin the love of beautiful things or of good persons: but the number of\ndifferent goods of this kind is as great as that of beautiful objects,\nand they are also differentiated from one another by the different\nemotions appropriate to different objects. These goods are undoubtedly\ngood, even where the things or persons loved are imaginary; but it\nwas urged that, where the thing or person is real and is believed to\nbe so, these two facts together, when combined with the mere love of\nthe qualities in question, constitute a whole which is greatly better\nthan that mere love, having an additional value quite distinct from\nthat which belongs to the existence of the object, where that object\nis a good person. Finally it was pointed out that the love of mental\nqualities, by themselves, does not seem to be so great a good as that\nof mental and material qualities together; and that, in any case, an\nimmense number of the best things are, or include, a love of material\nqualities (113-123). (2) Great evils may be said to consist either\n(_a_) in the love of what is evil or ugly, or (_b_) in the hatred of\nwhat is good or beautiful, or (_c_) in the consciousness of pain.\nThus the consciousness of pain, if it be a great evil, is the only\nexception to the rule that all great goods and great evils involve both\na cognition and an emotion directed towards its object (124-128). (3)\nMixed goods are those which include some element which is evil or ugly.\nThey may be said to consist either in hatred of what is ugly or of\nevils of classes (_a_) and (_b_), or in compassion for pain. But where\nthey include an evil, which actually exists, its demerit seems to be\nalways great enough to outweigh the positive value which they possess\n(129-133).\n\n\n\n\nINDEX.\n\n\n  Aesthetic\n    enjoyments 189-202, 203\n    judgment 215\n    temperament 216\n\n  Aesthetics 200\n\n  Affection\n    beauty of 204-5\n    misdirected 195, 198\n    reciprocal 198\n    value of 188-9, 203-5\n\n  Altruism 96-7, 167\n\n  Analytic judgments 7, 29, 33-4, 35, 220-1\n\n  Appreciation 189-90, 200, 204-5, 221\n\n  Approval 131\n\n  Approve 60\n\n  Approbation 171\n\n  Appropriate, inappropriate 192, 199, 204-5, 209, 211, 220\n    defined 190\n\n  Aristotle 4\n    definition of virtue 171\n    valuation of virtues 176-7\n    valuation of knowledge 199\n\n  Art\n    value of 188\n    _representative_, value of 193, 195, 196, 200\n\n  Autonomy 127\n\n\n  Bad 5, 27, 28, 95, 140, 143, 157, 178, 181, 188, 209, 210, 213, 214,\n    216, 218\n\n  \u2018Based on\u2019 38, 49, 54, 114, 115, 118, 120, 122, 144\n\n  Beauty\n    corporeal 203-4\n    no criterion of 202\n    definition of 201-2, 208\n    mental 203-5\n    \u2018seeing\u2019 of 190-1\n    value of 28, 81-2, 83-5, 86, 94, 188-9, 201-2, 209, 211, 224\n\n  Being, dist. from existence 110-11\n\n  Belief, value of 193-200, 208, 210-11, 212, 219, 224-5\n\n  Benevolence, Sidgwick\u2019s \u2018principle of Rational\u2019 102-3\n\n  Bentham 145\n    naturalistic fallacy 17-19\n    quantity of pleasure 77-8\n\n  Bradley, F. H.\n    pleasure and desire 70\n    theory of judgment 125\n\n  Butler, Bishop 86, 206\n\n\n  Casuistry 4-5\n\n  Causal judgments\n    relation to Ethics 21-7, 36, 146-8, 149, 180\n\n  Causal relations 31-3, 34-6\n\n  Chastity 158\n\n  Classical style 215-16\n\n  Christ\n    on value of motives 178\n    on love 179\n\n  Christian Ethics 178\n    on \u2018external\u2019 rightness 177\n    on \u2018internal\u2019 rightness 178-9\n    on value of motives 177-9\n    on value of virtue 174\n\n  Clifford, W. K. 40\n\n  Cognition\n    of evil 217-19\n    dist. from knowledge 194\n    relation to will and feeling 129-30, 133, 135-6, 141\n    value of 85, 189-92, 194, 199, 208, 212, 224, 225\n\n  Commands, confused with moral laws 128-9, 141\n\n  Common sense 224\n    on value of pleasure 86, 91-2, 94-5\n    on duties 156-9\n\n  Compassion 217, 219, 220, 225\n\n  Conduct, relation of to Ethics 2-3, 146, 180\n\n  Conscience\n    defined 178\n    not infallible 149, 180\n\n  Conscientiousness 218\n    defined 179\n    utility of 180\n\n  Contempt 211, 217\n\n  Corporeal beauty 203-4\n\n  Courage 217\n\n  Crimes 161\n\n  Criterion\n    of beauty 202\n    evolution as 46, 50, 55-6\n    of goodness 137-8\n    pleasure as 91-2, 94-5, 108\n    of right and wrong 223\n    will as 137-8\n    of truth 133\n\n  Cruelty 209-11, 218\n\n\n  Darwin 47\n\n  Definition, nature of 6-9, 18-20\n\n  Desirable, meaning of 65-7, 73\n\n  Desire, cause and object of 68-70, 73-4\n\n  Duty\n    = cause of or means to good 24-5, 105, 146-8, 167, 180, 223\n    fuller definitions of 148, 161, 180-1, 222\n    incapable of being known 149-50, 181\n    mainly negative 218\n    object of _psychological_ intuition 148\n    relations to expediency 167-70, 181\n      interest 170-1, 181\n      possibility 150-2\n      rightness 148\n      utility 146-7, 167-70\n      virtue 172\n      will 160, 161\n    not self-evident 148, 181\n    self-regarding 168\n\n\n  Egoistic Hedonism 18\n\n  Egoism, as doctrine of end 18, 96-105, 109\n    contradiction of 99, 101-5, 109\n    relation to Hedonism 97-8\n    relation to Naturalistic Hedonism 104-5\n    Sidgwick\u2019s \u2018Rational\u2019 98-9, 102-4\n\n  Egoism, as doctrine of means 96-7, 105, 167\n\n  Emotion\n    aesthetic 190\n    value of 189-92, 199, 203, 204-5, 208, 209, 211, 212, 217, 224, 225\n\n  Empirical 39, 111, 123\n\n  Empiricism 103, 124-5, 130\n\n  End = effect 32\n\n  End = good in itself 18, 24, 64-6, 72, 73, 79-81, 83, 85, 94-5, 184,\n    216\n    dist. from \u2018good as means\u2019 24, 72, 74, 79-81, 89, 90, 94-5, 106-7,\n      173-4, 178, 216, 223\n    \u2018ultimate\u2019 51, 83, 85, 96-7, 99-102, 183, 189\n    \u2018never justifies means\u2019 147, 163\n\n  End = object of desire 68, 70, 71, 72\n\n  Enjoyment 77, 96, 188, 208\n    aesthetic 188-9, 203\n    of evil and ugly 208-11, 218\n    sexual 95\n\n  Envy 211\n\n  Epistemology 133, 140-1\n\n  Ethics\n    Evolutionistic 46, 50, 54, 58\n    Metaphysical 39, 58, 113-15\n    Naturalistic 39-41, 58, 59\n    Practical 115-18, 140, 146, 149, 151, 154, 180, 222\n    province of 1-6, 21, 24, 26-7, 36, 37, 77, 115, 118, 142-6, 184,\n      222-4\n\n  Eudaemonist 175\n\n  Evil 153, 156, 158, 160, 186, 193, 205, 207-14, 224, 225\n    mixed 208, 209, 211\n    positive value of 216-22, 225\n\n  Evolution 46-8, 54-8\n\n  Evolutionistic 46, 50, 52, 54, 58\n\n  Existence\n    dist. from being 110-12\n    judgments about 123-5\n    relation to value 115-18, 118-22, 125-6, 194, 196, 197-9, 206, 210,\n      216, 219, 220, 221, 225\n\n  Expediency 167-70, 181\n\n\n  Feeling\n    supposed analogy to cognition 129-31, 141\n    supposed bearing on Ethics 129-31, 141\n\n  Fiction 121-2\n\n  Freedom, value of 86, 186\n\n  Freedom (of Will) 127\n\n\n  God 82, 102-4\n    love of 113, 194, 195-6, 200\n\n  Good\n    indefinable 6-16, 41, 79, 110-11, 142-4\n    = means to good 21, 24\n    the Absolute 183, 184, 186\n    the Human 183, 184, 186\n    mixed and unmixed 208, 209, 214, 215, 217, 219-20, 224\n    my own 97-9, 101, 170\n    \u2018private\u2019 99\n    _the_ 8-9, 18\n    \u2018Universal\u2019 99-102\n    Will 174-5, 179 _n._ 2, 180\n\n  Green, T. H. 139\n\n  Guyau, M. 46\n\n\n  Habit 171, 175-6, 177\n\n  Hatred 211, 214\n    of beautiful and good 211, 217, 225\n    of evil and ugly 178, 217, 218, 220, 221, 225\n\n  Health 42-3, 65, 157, 167\n\n  Heaven 115, 174, 183, 185, 195\n    upon Earth 186\n\n  Hedonism 39, 52, 59-63, 90-1, 96, 108-9, 174\n    Egoistic 18\n    Ethical 70, 144\n    Intuitionistic 59, 74-6, 144\n    Naturalistic 46, 50, 53, 54, 68, 104, 105\n    Psychological 18, 68, 69, 70, 73\n    Universalistic 103\n\n  Hegel 30, 34, 110\n\n  Heteronomous 127\n\n  Higher 48-9, 78\n\n  Hobbes 97\n\n  Honesty 175-6\n\n  Hypothetical laws 22, 155\n\n\n  Ideal\n    three meanings of 183-4\n    _the_ 183, 185, 205-7, 220-1\n\n  Idealistic 130, 205\n\n  Imagination, value of 193, 194, 196, 197, 210, 219, 220, 221, 224\n\n  Imperative 128\n\n  Industry 157, 167\n\n  Intention 179 _n._ 1\n\n  Interest 102\n    meaning of 97-8, 106, 170-1\n    dist. from \u2018duty\u2019 170-1, 181\n\n  Intrinsic\n    evil 207, 213, 218, 224\n    value 17, 21, 25-30, 36, 147, 173-7, 187, 189, 207, 214-16, 222-4\n\n  Intuition\n    = proposition incapable of proof 59, 77, 108\n    in psychological sense 75, 79, 85, 92, 108, 144, 148-9, 173\n\n  Intuitionism\n    in Sidgwick\u2019s sense 59, 76, 144\n    in proper sense 106, 148\n\n\n  Judgment\n    error of 192-3\n    two types of ethical 21, 23-7, 115, 146, 148, 222, 224\n\n  Justice 178\n\n  Justify 97, 101, 147, 163\n\n\n  Kant 110, 129\n    \u2018Copernican revolution\u2019 133\n    value of Good Will 174-5, 179 _n._ 2, 180\n    value of Happiness 174-5\n    theory of judgment 125\n    \u2018Kingdom of Ends\u2019 113\n    \u2018practical love\u2019 179\n    connection of \u2018goodness\u2019 with \u2018will\u2019 126-8\n\n  Knowledge\n    involves truth of object 132, 134\n    involves belief 194\n    value of 82, 86, 194, 195, 196, 197, 199, 211, 221\n\n\n  Lasciviousness 209-10\n\n  Law\n    ethical 155\n    hypothetical 22, 155\n    legal 126, 128\n    moral 126-8, 146, 148, 160, 162, 165\n    natural 26, 29, 57, 126, 183, 186\n    scientific 22-3, 124, 155\n\n  Legal 126, 128\n\n  Leibniz 125\n\n  Life 15, 46, 50, 52, 156\n\n  Logical\n    dependence 61, 110, 118, 122, 139, 143-4\n    fallacy 140-1\n\n  Love\n    Christ and Kant on 179\n    of beautiful and good 177-9, 199, 204, 217, 224\n    of evil and ugly 209, 210, 211, 217, 225\n\n  Lucian 45\n\n  Lust 209-10, 218\n\n  Lying 154\n\n\n  Mackenzie, Prof. J. S. 114, 120\n\n  Material qualities, value of 204, 205-7, 221, 225\n\n  Matter, value of 205-7\n\n  Meaning, \u2018to have no\u2019 31, 34-5\n\n  Means = cause or necessary condition 18, 21-3, 89, 180\n    dist. from \u2018part of organic whole\u2019 27, 29-30, 32, 220\n    goodness as, dist. from intrinsic value 21, 24, 26, 27, 37, 72, 74,\n      79-81, 89, 90, 94-5, 106-7, 115, 118, 173-4, 178, 187, 195-6,\n      197-8, 216, 223\n    \u2018not justified by end\u2019 147, 163\n\n  Mental\n    beauty of 203-5, 225\n    value of 205-7\n\n  Mercy 178\n\n  Metaphysical 39, 58, 110-15, 139-40\n\n  Method\n    of discovering intrinsic value 20, 36, 59-60, 64, 89, 91, 92, 93, 95,\n      142-5, 173, 185-6, 187-8, 195, 197-8, 206-7, 209, 223\n    of discovering value as means 22-3, 146, 148-54, 172-3\n\n  Mill, J. S. 145\n    Hedonism 63-81, 108\n    naturalistic fallacy 40, 66-7, 69, 72-3, 74, 104, 108\n    Psychological Hedonism 68, 72, 73-4\n    quality of pleasure 77-81, 108\n    Utilitarianism 104-5\n\n  Moral\n    approbation 171\n    law 126-8, 146, 148, 160, 162, 165\n    obligation 128\n    sentiment 168, 178, 217-19\n\n  Motive 67, 70, 177, 178-80\n\n  Murder 148, 151, 154, 156-7, 178\n\n\n  Natural\n    laws 26, 29, 57, 126, 183, 186\n    objects and properties 13-14, 39-41, 58, 110-11\n    selection 47\n\n  Natural = normal 42-4, 58\n\n  Natural = necessary 44-5, 58\n\n  Naturalism 20, 40, 58, 144\n\n  Naturalistic\n    Ethics 39-41, 58, 59\n    fallacy 10, 13-14, 18-20, 38-9, 48, 57, 58, 61, 64, 66-7, 69, 72-3,\n      74, 104, 108, 114, 118, 124, 125, 139, 173, 176, 201\n    Hedonism 46, 50, 53, 54, 68, 104, 105\n\n  Nature 40-1, 110, 111, 112\n\n  Nature, life according to 41-2, 113\n\n  Nature, value of 188, 193, 195, 200, 206\n\n  Necessity\n    analytic 22, 33-4, 35, 220, 221\n    causal or natural 29, 31-2, 34, 186, 187\n\n  New Testament 177, 178, 179\n\n\n  Object\n    of cognition 141, 191, 192, 193, 211\n    of desire 68-70\n    natural 13-14, 39-41, 58, 110-11\n\n  Objective 82, 201\n\n  Obligation\n    moral 103, 128, 147\n\n  Obligatory 25, 148, 170\n\n  Organic relation, unity, whole\n    common usage 30-6\n    my own usage 27-31, 32-3, 36, 93, 96, 149, 184, 187, 189, 190, 202,\n      206, 208, 212, 215, 220, 223\n\n  Ought\n    to aim at 24-6, 100\n    to do 26, 105, 115, 116, 117, 127, 128, 140, 146, 148, 173, 180, 223\n    to be or exist 17, 115, 118, 127, 128, 148, 173, 180, 223\n\n\n  Pain 64, 65, 210, 212-4, 217, 222-3, 225\n\n  Particular 3-4\n\n  Perception 111, 112, 134, 136\n\n  Pessimism 51, 53, 156\n\n  Plato\n    on Egoism 98\n    on goods 178\n    on Hedonism 88\n    on value of Knowledge 199\n    on universal truths 111\n\n  Pleasure 12-13, 16\n    consciousness of 87-91, 109, 212\n    as criterion 91-2, 108\n    and desire 68-71, 73-4\n    and \u2018pleasures\u2019 79\n    \u2018quality of\u2019 77-81\n    value of 39, 46, 50-4, 59-66, 71-2, 74-5, 79-81, 83, 85-96, 144,\n      146, 171, 173, 174, 188, 205, 212-14, 222-3\n\n  Pity 217, 221\n\n  Positive science 39\n\n  Possible action 150-1\n\n  Practical 216, 221\n    Ethics 115-18, 140, 146, 149, 151, 154, 180, 222\n    Philosophy 2\n\n  Practice 2, 20\n\n  Praise 171\n\n  Preference 77-9, 131\n\n  Promises 157\n\n  Property, respect of 157\n\n  Propositions, types of 123-6\n\n  Prove 11, 65, 66, 74, 75-7, 99, 112, 137, 141, 143, 145, 169, 181\n\n  Prudence 168\n    \u2018Maxim of\u2019 102-4\n\n  Psychological 11, 130, 140, 148\n    Hedonism 18, 68, 69, 70, 73\n\n  Punishment 164\n    retributive or vindictive 214, 215, 216, 221\n\n\n  Reason 143-4\n\n  Representative art 193\n\n  Reward 174\n\n  Right 18, 24-5, 105, 146, 180, 216, 218, 223\n    dist. from \u2018duty\u2019 148\n    relation to expediency 167\n    externally 176-7\n    internally 179 _n._ 1\n\n  Romantic style 215-16\n\n  Rousseau 42\n\n\n  Sanctions 159, 164\n\n  Secondary qualities 206\n\n  Self-evidence 143, 144, 148, 181\n\n  Self-realisation 113, 114, 120, 188\n\n  Self-sacrifice 170\n\n  Sensation 134\n\n  Sensationalist 130\n\n  Sidgwick, Henry 145\n    value of beauty 81-4, 85-7\n    on Bentham 17-19\n    rationality of Egoism 99-103\n    \u2018good\u2019 unanalysable 17\n    Hedonism 59, 63, 64, 81-7, 91-6, 108-9\n    \u2018method\u2019 of Intuitionism 59, 92-4\n    value of knowledge 82, 86\n    neglects principle of organic wholes 93\n    pleasure as criterion 91-2, 94-5\n    quality of pleasure 77, 81\n    value of unconscious 81-4\n\n  Sins 161\n\n  Spencer, Herbert 46, 48-58\n\n  Spinoza 110, 113\n\n  Spiritual, value of 205-6\n\n  Summum Bonum 183, 205\n\n  Stoics 41, 110\n\n  Synthetic 7, 58, 143\n\n\n  Taste, error of 192-3, 211\n\n  Taylor, A. E. 60\n\n  Temperance 157, 168\n\n  Theodicies 220\n\n  Tragedy 219, 221\n\n  Truth\n    relation to existence 111, 124-5\n      cognition 130, 132-4, 136, 141, 196\n      knowledge 134, 194\n    types of 111-12, 124-5\n    value of 85-6, 193-200, 208, 210, 211, 212\n\n  Tyndall 40\n\n\n  Ugly 208, 209-11, 214, 216-19, 221\n\n  Ultimate end 51, 83, 85, 96-7, 99-102, 183, 189\n\n  Unity 222\n    organic, see \u2018Organic\u2019\n\n  Universal\n    Good 99-102\n    truths 21-3, 27, 57, 111, 126, 154-5, 181\n\n  Universalistic Hedonism 103\n\n  Useful 106, 146, 167\n\n  Utilitarianism 63, 96, 99, 104-7, 109\n\n  Utopias 183, 186\n\n\n  Value\n    intrinsic 17, 21, 25-30, 36, 147, 173-7, 187, 189, 207, 214-16, 222-4\n    as means 21, 174, 195-6\n    negative 215, 216\n\n  Vice 171, 209, 211\n\n  Virtue\n    definition of 171-3, 181, 223\n    three kinds of 175\n    mixed 221\n    relation to \u2018duty\u2019 172\n    value of 83, 85, 86, 87, 173-80, 181-2, 217-19, 221-2\n\n  Volition\n    supposed coordination with cognition 129-30, 133, 135-6, 141\n    supposed bearing on Ethics 130, 136, 141\n\n\n  Whole\n    good as a 208, 214-16, 219, 221\n    good on the 214-16, 219, 220, 221\n    organic, see \u2018Organic\u2019\n\n  Wickedness 220\n\n  Will\n    as criterion of value 137-8\n    relation to duty 160, 161, 180\n    the Good 174-5, 179 _n._ 2, 180\n    supposed analogy to cognition 129-30, 135-6\n    supposed bearing on Ethics 126-7, 128-31, 135-9, 141\n\n  Wrong 180, 218, 223\n\n\n\n\n  PRINTED IN ENGLAND BY J. B. PEACE, M.A.\n  AT THE CAMBRIDGE UNIVERSITY PRESS","title":"Principia Ethica","tradition":"philosophy-ethics","translated_text":null,"word_count":100104}