3° A Refutation 01 Morals

John Mackie

I N THIS PAPER I DO NOT PRETEND TO BB

advancing any particularly new ideas: hardly any of the arguments areoriginal, and indeed most are the stock instruments of all modem discussionsof morals. But I think I am justified in offering this re-statement of them,because it is seldom realized how they may be brought together .~dinterrelated, or how radically destructive they are of all common views of

.morality,when this is done. . .

We all have moral feelings: all of us find that there are human actionsand states of affairs of which we approve and disapprove, and which wetherefore try to encourage and develop or to oppose. (This emotion ofapproval is different from liking, one difference being that its object is moregeneral. If someone stands me a pint, I like it: if someonestandsan enemyof mine a pint, I dislike it: but I should approve of a state of society whichprovided free beer all round. So if I hear of someone whom I have nevermet and to whom I am personally indifferent being stood a pint, I shouldnot say that I like it, for I am not directly affected, but I may well approveof it, because it is an instance of the sort of thing I want to see everywhere.A thorough distinction of approval from liking and other relations wouldrequire further discussion, .but perhaps this will serve to indicate a contrastbetween classes witlr which we are all in ~act acquainted. I shall suggestlater a possiblesourceof these generalizedemotions.) But most of us do .

not merely admit that we have such feelings, we think we can also judgethat actions and states are right and good, just as we judge about othermatters of fact, that these judgments are either true or false, and that thequalities with which they deal exist objectively. This view, which almosteveryone holds, may be crudely called "believing in morals." A fewsceptics, however, think that there are only feelings of approval, no ob-

[The article from which this selection is taken originally appeared in The Australasianlournal of Philosophy, 1946. It is reprinted here with the kind permission of the authorand editor.]

,.346"

~ "~-- -~" - ~.. - - -- - -- - --- -----" n_- ~

A REFUT AnON OF MORALS »347«

jective moral facts. (Of course the existence of a feeling is an objective fact,but not what is commonly called a moral fact.) One of their main argu-

" mentsis that moral facts wouldbe "queer," in that unlikeother facts theycannot be explained in terms of arrangements of matter, or logical con-structions out of sense-data, or whatever the particular theorist takes to bethe general form of real things. This argument is not in itself very strong, oreven very plausible, for unless we have good a priori grounds for whateveris taken as the basic principle of criticism, the criterion of reality, the merefact that we seem to observe moral qualities and facts would be a reason formodifying that principle. Their other main argument, which is both olderand more convincing, though not logically conclusive, is that although atanyone time, in a particular social group, there is fairly complete agreementabout what is right, in other classes, other countries, and above all in otherperiods of history and other cultures, the actual moral judgments or feelingsare almost completely different, though perhaps there are a few feelings sonatural to man that they are found everywhere. Now feelings may wellchange with changing conditions, but a judgment about objective factshould be everywherethe same: if we have a faculty of moral perception, itmust be an extremely faulty one, liable not only to temporary illusions, assight is, but to great and lasting error. Of course it may be that everysociety except our own is mistaken, that savages""are morally backwardbecause they lack our illuminating exp~rience of the .long-term effects ofvarious kinds of action, and so on. But this complacent view "(not indeedvery popular now) is shaken by the observation that the variations in moralfeelings"can be explained much more plausibly not as being due tomistakes, but as reflections of social habits. This moral relativity would beless alarming if we could say that the varying judgments were not"ultimate,but were applications to different circumstances of a single principle or asmall number of principles, which were everywhere recognized-for ex-ample, that whatever produces pleasure is good, that whatever societycommands is right, or, at the very least, that we should always do what webelieve to be right. But these principles are not commonly laid down first,and the particular judgments deduced from them: rather the particularjudgments are made by ordinary people, whereas the principles are laterinvented by philosophers and manipulated in order to explain them. In anycase there is just as little agreement about principles as about particularjudgments. "

We find on further inquiry that most, perhaps all, actual moral judg-ments are fairly closely correlated with what we may call social demands:any society or social group has regular ways of working, and, in or~er tomaintain these, requires that its members should act in certain ways: themembers-from whatever motive, perhaps mainly habit, which has com-pelled them to adapt their desires to the established customs-obey theserequirements themselves and force their fellows to do so, or at least feelobliged to obey and approve of others obeying. They call "right" and"good" whatever accords with these ways of working. Moreover, as the

- --------- - -- ---

»348« MORAL JUDGMENTS

science of social history develops, it is more and more strongly suggestedthat ways of working and institutions have their own laws of growth, andthat the desires or moral views of individuals do not so much control the

history of society as arise out of it.Belief in the objectivity of moral qualities is further undermined when

we remark that whenever anyone calls an action .or activity or state- ofaffairs right or good (unless he is speaking in an ironical tone or puts thesewords in inverted commas) he himself either has a feeling of approval, ordesires that the action should be done or the activity pursued or the stateof affairs come into existence. (Only one of these alternatives is necessary,but they are often found together.)

None of these considerations is conclusive, but each has a certainweight: together they move the moral sceptic (who is often of a scientificand inductive turn of mind, and less devoted than some others to the clear

. lightof intuitionor the authorityof reason) to concludethat in all proba-bility we do not recognize moral facts, but merely have feelings of approvaland disapproval, which arise in general from social demands arid thereforevary from one society to another. This view I intend to examine. and re-state, and to advance what I regard as decisive arguments for one of itsmore important aspects.

The simplest formulation of this view is that when someone says "thisact is right" he means merely "I approve of tm.sact." The well-known replysimply leaps into the reader's mind: when one person says that an act isright, another that the same act is wrong, they would not on this theory bedisagreeing, whereas in fact they think they are. It will not do to say, withStevenson,1 that there is a disagreementin attitude,but not in belief: theythink, at any rate, that they disagree in belief. Nor does one me~ that"society approves of this act," since we frequently meet people who say, "Iknow society approves of this, but it is wrong all the same." But there isno need for argument: direct introspection shows that when we use theterms "right," "good," and the rest, we never intend merely to state thatthere are feelings of approval. An improved formulation of the scepticalview is that in saying "this is right," and so on, we are not stating anyapproval, but only expressing one, that words like "right" and "wrong,""good" and "bad" are to be compared not with "red" and "square" butwith explanations or ejaculations like "ow!" "boo!" and "hurray!" This iscertainly nearer the truth, and avoid~ the previous difficulties, but is, inanother way, just as implausible. For we do not think that we are merelyejaculating when we talk in moral terms. If we did, and if someone dis-agreed with us, we should merely disapprove of his approvals, -and eithertry to coax him into a different emotional attitude, or if he proved obstinate,knock him down. In fact we reason with him. These facts, and the logicaltangles that we get into when we try to re-state fairly complex moralsituations in the "boo-hurray" language, prove that we think, at least, thatweare not merely expressing our emotions but are describing objective facts,

1. Ethics and Language. Chapter 1.

-- ---u - ----

A REFUTATION OF MORALS }).:>~y«

and therefore that the meaning of moral terms is not parallel with that ofejaculations. Many refutations of the "boo-hurray" theory have beenworked out, but they all depend upon and illustrate the fact that we thinkthat we are doing things of quite different sorts when we sai "right" andwhen we say "ow!" Now if philosophy could do no more than elucidate themeaning of the terms of common speech, remove confusions and rationalizethe thought of ordinary men, there would be nothing more to be said.Moral terms do mean objective qualities, and everyone who uses them doesso because he believes in objective moral facts. But if the very terms ofcommon speech may include errors and confusions within themselves, sothat they cannot be used at all without falsity, if, we may add, philosophymay be permItted to inquire into these errors by observing a few facts foritself and founding inductive conclusions on them, the moral sceptic neednot be so soon disheartened.

But he must modify his view again, and say that in using moral termswe are as it were objectifying our own feelings, thinking them into qualitiesexisting independently of us. For example, we may see a plant, say a fungus,-that fillsus with disgust, but instead of stating that we have this feeling, ormerely expressing and relieving it by an exclamation, we may ascribe tothe fungus a semi-moral quality of foulness, over and above all the qualitiesthat a physical scientist could find in it. Of course, in objectifying our feel-ings we are also turning them inside out: our feeling about the fungus isone of being disgusted, while the foulness we ascribe-to the fungus meansthat it is disgusting. The supposed objective quality is not simply the feel-ing itself transferred to an external object, but is something that wouldinevitably arouse that feeling. (No one would say, "That fungus is foul, butI feel no disgust at it.") The feeling and the supposed quality are relatedas a sealor stamp and its impression. -

This process of objectification is, I think, well known to psychologistsand is not new in philosophy. I believe that it resembles what Hume sayswe do when we manufacture the idea of necessary connection out of ourfeeling of being compelled, by the association of ideas, to pass from causeto effect, though here the process of turning inside out does not occur.

There are strong influences which might lead us thus to objectify moralfeelings. As I have mentioned, our moral judgments seem to arise fromapprovals borrowed from society, or from some social group, and theseare felt by the individual as external to himself. It is for this reason thatthey are universal in form, applying equally to himself and to others. Theyare _thusformally capable of being objective laws, in contrast to the "selfish"desires of the individual. This generality or universality, which I mentionedas characteristic of the emotion of approval, is reflected in Rousseau's doc-trine that the general will and therefore law must be general in their object,and inKant's criterion of the possibility of universalization of a moral law.Since we inevitably tend to encourage what we approve of, and to imposeit upon others, we want everyone to adopt our approvals, -and this willmost surely come about if they have only to perceive a genuinely existing

- --------- -n-n n__- n --- -- -------

»350« MORAL JUDGMENTS

objective fact, for what we feel is in general private, what we perceive maybe common to all. Suppose that we approve of hard work: then if as well as

- a feelingof approvalin our own minds there were an objectivefact like"hard work is good," such that everyone could observe the fact and suchthat the mere observation would arouse in him a like feeling of approval,and even perhaps stimulate him to work, we should eventually get whatwe want done: people would work hard. And since what we want does notexist in fact, we naturally construct it in imagination: we objectify ourfeelings so thoroughly that we completely deceive ourselves. I imagine thatthis is the reason why our belief in moral objectivity is so firm: we muchmore readily admit that the foulness of a fungus is an objectification thanthat the depravity of people who break our windows is~If moral predicateswere admitted t~ be what the moral sceptic says they are, we should neverbe able to extol a state of affairs as good in any sense which would inducepeople to bring it about, unless they already wanted it, though we mightpoint out that this state had features which in fact they did desire, thoughthey had not realized this: we should never be able to recommend anycourse of action, except in such terms, as, "if you want to be rich, beeconomical"; nor could we give commands by any -moral authority, thoughwe might again advise "if you don't want a bullet through your brains, comequietly"; and we should never be able to lecture anyone on his wickedness-an alarming prospect. The temptations to objectify feelings of approval,and to retain our belief in morals, are clearly strong ones.

This process of objectifying our feelings is, then, neither impossible norimprobable: there is also abundant evidence that it is just what has oc-curred. It is commonly believed by moralists that good means desirable ina sense such that the mere recognition that a thing is good makes us desireit, and similarly the conclusion of the practical syllogism is both "this isright" and the performance of the action. This is what we should expect if"right" were the objectificaJion of a tendency to compel or command thekind of act so described, and "good" of desire and approval. This is againindicated by the use of the term "value" which is dearly borrowed fromspheres like economics where value is created by demand-in fact aquality manufactured in imagination out of the relation of being demandedby someone, the abstraction b~ing the easier because the demand is notessentially that of a single buyer, but of an indeterminate crowd of po-tential buyers: the analogy with the objectification of moral feelings, aidedby their generality,is veryplain. . . . -

In attempting to give an account of the origin of ploral terms in thisprocess of objectification, I do not, of course, claim that it is complete orprecise in all respects.-It is still open to discussion and correction on em-pirical grounds. We might go on to consider this process as a psychologicalprocess, investigating its causes, its similarities and contrasts with othermental processes, and the steps of which it is made up. We might askwhether -"objectification" or some other name is really the most suitable,and ~so what are the precise motives objectified: we might consider, for -

-- - u u- ---

A REFUTATION OF MORALS »:1:>1«

example, Westermarck's argumerit2 that "ought" normally expresses aconation, is sometimes but not necessarily or essentially imperative, andhas its origin in disapproval rather than approval.

My discussion in this paper is intended to open the way for such dis-cussions, not to settle them once and for all. What I am concerned toestablish is simply the logical status of moral terms, not the psychologicaldetails of their origin; in effect I am asserting only that there are no factsof the form "this is right," that when we use such words the only fact isthe existence of some feelings in ourselves or in others or in both, but thatin using these terms we are falsely postulating or asserting something of thesimple, objective form "this is right." . . .

2. The Origin and Development of the Moral Ideas, Chapter VI.