Royal Institute ofPhilosophy Lectureshttp://journals.cambridge.org/PHS

Additional services for Royal Institute ofPhilosophy Lectures:

Email alerts: Click hereSubscriptions: Click hereCommercial reprints: Click hereTerms of use : Click here

Human Nature

Peter Winch

Royal Institute of Philosophy Lectures / Volume 4 / March 1970, pp 1 - 13DOI: 10.1017/S0080443600000042, Published online: 08 January 2010

Link to this article: http://journals.cambridge.org/abstract_S0080443600000042

How to cite this article:Peter Winch (1970). Human Nature. Royal Institute of PhilosophyLectures, 4, pp 1-13 doi:10.1017/S0080443600000042

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/PHS, IP address: 217.112.157.7 on 20 Sep 2014

1HUMAN NATURE

Peter Winch

T H E concept of human nature usually enters discussions of thenature and implications of the social sciences in connection with oneor another form of 'relativism'. Confronted with the enormous andapparently conflicting variety of phenomena of human life atdifferent places and times, we are inclined to ask whether there isnot something which holds these phenomena together and unifiesthem. Stated thus baldly this question is no doubt so vague as toapproach meaninglessness; it will have to be posed in differentforms - and probably answered differently - according to the par-ticular phenomena of human life which we happen to have in mind.In this lecture I shall concentrate my attention on some questionsabout the relevance of sociological investigations to our understand-ing of ethics and about the treatment of ethics in such investigations.I shall be particularly interested in the way in which the concept ofhuman nature enters into such discussions; and I shall devote agood deal of attention to Professor Alasdair Maclntyre's recentShort History of Ethics.1 It is a large merit of this book that it explicitlyand invigoratingly relates the manner of its historical exposition toa distinctive philosophico-sociological standpoint concerning thenature of morality. I call this a 'merit' and want to stand by thatcharacterisation even though I think that there are important con-fusions enshrined in Maclntyre's approach. A large part of the taskwhich I want to set myself in this lecture is to make clear the natureand importance of these confusions.

The concept of human nature has been thought important inat least three different, though related, ways by those who have

1 Alasdair Maclntyre, A Short History of Ethics (New York, 1966).

I

HadiHighlight

2 Peter Winch

considered morality from a sociological standpoint: in connectionwith questions about the identification of those features of a society'slife which are to be subsumed under the concept of morality; in con-nection with the explanation of the features thus classified; and inconnection with attempts at the justification of particular moralchoices. Though these types of question run into each other andcannot always be held apart, it will be convenient for me to developmy argument by considering each type of question in turn. I startwith questions about identification.

The phenomena of human life do not reach the sociologicalinvestigator in a pre-packaged state, labelled 'economic', 'religious','aesthetic', 'ethical', and the like. He has to decide, often with con-siderable difficulty, what he is going to count as belonging to one orthe other of such categories. Sometimes he will feel that the facts heis confronted with do not fit without distortion into any of thecategories he is already familiar with and that he must invent newones. Problems of this sort must obviously arise in cases where theinvestigator - if, for instance, he is a social anthropologist - isstudying a society or a culture which looks very different from thatwithin which he has learnt to apply such categories hitherto(probably the one he grew up in). But not only there. It requireslittle sophistication to be struck by the problematic application ofsuch categories nearer home too. Rush Rhees gives us a strikingexample of this.' He refers to a reported interview between a localcommittee of a 'Moral Welfare Council' and an unmarried expec-tant mother, in which the line of the committee's chairman was:'Do you know you can be locked up if you go on like this ? Do youwant to be locked up?', etc. Rhees contrasts the 'moral' outlookexpressed here with what he himself feels inclined to say: that 'theactivities of those moral enthusiasts are foul and filthy'. For mypurposes (as for Rhees's in the essay cited) the point of the contrast isnot to raise the question which of these outlooks is the 'morally right'one, which if either of the two we ought to endorse; it is rather topoint out difficulties in saying that both these outlooks either are ornot forms of'moral' outlook. Rhees comments:

Suppose I dispute the remark that 'their activities are foul andfilthy' is a moral judgement - what would be said to show that itis ? I think the differences are just as important as the similarities.And if you offered reasons for saying that both are forms of moraljudgement, I think there are reasons just as strong for saying theyare not.

1 Without Answers (London, 1969) pp. 97-8.

Human Nature 3Rhees does not particularise the differences he has in mind. But Itake it that one important difference would be that the committeechairman would take herself to be committed to doing somethingabout, intervening to remove, the evils (as she regards them) con-fronting her. Rhees on the other hand might well, I imagine, disownany commitment to do anything about the council's activitiesbeyond saying what he thinks of them - or even just thinking it.

However that may be, there is undoubtedly a difficulty here aboutwhat cases we are going to call cases of a 'moral outlook' and aboutwhat is involved in so calling them. One very common approach tosuch difficulties - exemplified, for example, in some of the writingsof Mrs Philippa Foot - is to try to locate the moral in certain allegedfeatures of human nature, to say what human needs moralityanswers to and to refer to such needs as criteria for what can beaccepted as a moral concern and what cannot. These needs are usedto give content to a notion of 'human good and harm', and theclaim is that all genuine moral judgements involve a reference,though sometimes more explicitly than at other times, to such anotion. But the undoubted attractiveness of such a move is con-siderably lessened if we notice what has been suggested by somecritics of the 'neo-naturalist' movement, that the identification ofthese human needs - at least in many important cases - may itselfbe a matter for dispute of a kind which it is hard not to characteriseas a moral dispute. One of the interesting features of AlasdairMaclntyre's Short History is that it attempts to reconcile this sort ofappeal to human needs in determining what belongs to the moralwith a recognition of the extent to which what we are prepared torecognise as being human needs varies pari passu with the moraloutlook we are prepared to embrace. I think Maclntyre's attemptfails; and indeed I think the failure inevitable. I will now try todevelop my reasons for this by discussing what Maclntyre writesabout the 'explanation' of the existence of such a feature as moralityin the social life of human beings.

One context in which Maclntyre develops this theme is an im-portant criticism which he makes of Professor R. M. Hare's accountof moral discourse. Hare makes a distinction between the languagein which we express our personal likes and dislikes and our privatefeelings about people and actions, our individual choices and thecommands and injunctions which we may at times issue, and thelanguage in which we express our moral evaluations. Maclntyreaccepts this general distinction but complains that Hare's accountstill leaves us with a question to which it is unable to provide ananswer, namely 'why there should exist any specifically moral

grouperHighlight

4 Peter Winchlanguage over and above the ordinary language of feelings, liking,choice and imperatives.1 He himself offers an answer to his questionwhich, taken by itself, is puzzling:

My attitudes and my imperatives have authority for me justbecause they are mine. But when I invoke words such as ought andgood I at least seek to appeal to a standard which has other andmore authority. If I use these words to you, I seek to appeal to youin the name of those standards and not in my own name.2

To understand the force of what Maclntyre is saying here, wemust bring more explicitly into the light his conception of a humannature which is not an invariable datum but which changes inresponse to new social and historical conditions. It is a pervasivetheme of his book that moral concepts can be made sense of only intheir particular historical setting, that they change as forms of sociallife change and are indeed 'embodied in and . . . partially constitu-tive of forms of social life'.3 He concludes from this that moral con-cepts may be expected to play different roles in different socialsettings and accordingly that, when we find moral philosophersgiving different accounts of 'morality' it is dangerous for us toassume that they are talking about the same subject-matter. Here wesee how the difficulties concerning the identification of the socialphenomena we are going to count as 'moral', which I raised earlier,merge with the problems about the 'explanation' of such phenomena,which Maclntyre claims that Hare is unable to deal with.

The difficulties which Maclntyre feels over Hare's account cannow be expressed as follows. Hare's formalistic account of morallanguage seems to leave him with a problem about the source of thespecific content of people's moral evaluations. He tries to overcomethis (in Freedom and Reason) by taking an agent's own interests,desires and inclinations as a datum and then transmuting them intogenuine moral judgements by means of the mechanism of universa-lisation. But, leaving aside other difficulties about this transmutation,Maclntyre asks why one agent's moral evaluations, the content ofwhich springs from a consideration of his own interests and inclina-tions, should carry any authority at all for another agent whoperhaps has quite different inclinations and interests. Will it not besimply a lucky chance if two agents manage to arrive at even roughlythe same evaluations in this way? Moreover, even if they do manageto agree, will not each man's evaluations have authority for himonly in so far as they are backed by his own interests and inclina-tions? Maclntyre's move at this point is to shift the centre of gravity

1 Maclntyre, Short History, p. 264. 2 Ibid., p. 265. 8 Ibid., p. 7.

ha3gHighlight

ha3gSticky-NoteAuthority 1) ) (power). [ (force) ]. ) : ( ) .2) : . ( authorities .3) : .4)

ha3gSticky-Notestandard ( ) measure criterion

ha3gSticky-NoteIn the name of1. 2. = for the sake of3.

By the name of1. ~

In (my/your/...) own name ( ....) In behalf of

ha3gSticky-Notedata. Plural: datum1- ( )2- ( )

ha3gSticky-Noteformalityformalism. ( ). ( ): formalism=formulation=

ha3gSticky-Notesource ( source)

. source : informant

ha3gSticky-Notemechanism

. . ( ). () # spontaneit. - ( ). ( )

: .

ha3gSticky-Noteto make sense of ( )

ha3gSticky-Noteembed

ha3gSticky-Note value judgment

ha3gSticky-Notereason. ( ). . ( ( ) . . ( ). ( ). . Ali has reason in this point give s.o. reason:

ha3gHighlight

ha3gSticky-Note ing form :. : .... : x . : x (= x )[ ing (to have left aside) : having left aside...): : ...

ha3gSticky-Notelucky. : . guess: ( ). chance: . answer or reply: .

ha3gSticky-Notemanage + to: ( )

Human Mature 5from the purely personal concerns of any individual agent to theinstitutions and values generally accepted in the society to which heand those with whom he is in communication equally belong. Inthat society 'there is a recognised list of virtues, an established set ofmoral rules, an institutionalised connection between obedience andrules, the practice of virtues, and the attainment of ends'.1 Thus,when I express a moral judgement to someone else, I appeal to an^authority which is not my authority and one which I can expect himto recognise as well as myself.

Now Maclntyre's account does so far seem to me to be superior toHare's in its suggestion about where we should locate the source ofthe particular contents of people's moral judgements at particulartimes and places: namely in a historical tradition at a particularstage of its development. The point perhaps comes into clearer focusif we compare Hare with Kant in this connection. Kant's attempt,in the Tugendlehre, to derive particular duties from his account of thegeneral form of morality, runs into very similar difficulties. And sinceour views about many practices have changed a good deal sinceKant's day, it may be easier for us to see, in his case, to what a greatextent his claims about specific duties are drawn from ideas whichhappened to be current in the life of his time - contrary, of course,to what he himself claimed for them.

However, Maclntyre seems to claim more for his criticism of Harethan what I have just tried to bring out. He suggests that hisaccount, unlike Hare's, offers us an explanation 'why there should existany specifically evaluative language'. I do not think this claim canbe sustained; and I think that Maclntyre only makes it because hehas not fully seen the implications of what he has said about theconcept of human nature. What Maclntyre in fact offers us is areminder, valuable enough in itself, that an evaluative use oflanguage does exist within the context of such and such social insti-tutions and practices. To speak of a 'recognised' list of virtues and an'established' set of moral rules is simply to assert that the standards Iappeal to in making moral evaluations also carry authority for otherpeople. This is so; but it has certainly not been 'explained'. And it ishard to see how, once human nature is no longer thought of assomething at least relatively permanent and unchanging, it possiblycan provide any explanation of morality of the sort Maclntyreseems to be seeking. If you take the view that it belongs to the natureof human beings to have more or less fixed needs and wants, you mayseem able to 'explain' particular moral codes as different ways inwhich those needs may be ministered to in different social contexts.

1 Ibid., p. 265; my italics.

ha3gSticky-Note : institution: : ( instituting )institute: " " ( )

institute ( - ( ))institution (

ha3gSticky-Noteas well as :. 2. ( [ ] . - ( ) as [ : ]3.

ha3gSticky-Notenow- : - . .- .- : ... (

ha3gHighlight

ha3gHighlight .

ha3gSticky-Notehair's acount

ha3gHighlight

ha3gHighlight : ...

ha3gHighlight

ha3gHighlightdescriptive evaluative. ( )factual normative

ha3gHighlight . . . why . (() ): . : ...

ha3gHighlight ...

ha3gHighlight

ha3gHighlight

ha3gHighlight

ha3gHighlight () --

ha3gHighlight

ha3gHighlight to to be (linking verb) to to

: ( )

ha3gHighlight

ha3gHighlightin= making : to make an evaluation=evaluate= to make making ( )

ha3gHighlight

ha3gHighlight ( how)

ha3gHighlight ...

ha3gHighlightno longer

ha3gHighlight ( ...

ha3gHighlight can can't cant at all, in no means can

ha3gHighlight explanation

6 Peter Winch

But once you allow the conception of human needs to float freely,like a currency with no fixed parity, alongside the changing moralcodes, then those needs will as much (or as little) require explana-tion as the codes themselves.

I have, though, still not said enough about the kind of connectionMaclntyre thinks exists between adherence to a morality and the'attainment of ends'. This matter can be best discussed if I turn nowto the third sort of question about morality which I distinguished atthe beginning of this lecture: about the: justification of moral choices.

Talk about the different moralities or diversity of standards oftencauses uneasiness; and not least among those whom Lord Devlinlikes to call 'rationalist philosophers'. It may be the case, they wantto object, that people's moral beliefs are diverse, but the importantquestion for a philosopher is which, if any, of those beliefs are right.Now the conception of a human nature involving certain needs andwants has very often been brought into play to answer questions ofthis sort too. Thus, it has been thought, you determine what are tocount as 'moral' beliefs by asking whether the beliefs in questionhave to do with what is thought to satisfy human needs; you explainthe existence of such beliefs by pointing out the urgency for humanbeings to satisfy those needs; and you distinguish correct from in-correct moral judgements by asking whether they are right in theirimplications about what will satisfy the most important humanneeds.

It is in his treatment of this last sort of view that the incoherencesin the way Maclntyre thinks of human nature become most obvious.I shall now try to expound briefly what I take his view to be, thencriticise it and finally conclude with some general reflections on whatI take to be matters of very great philosophical importance whichare connected with the difficulties in Maclntyre's position.

Maclntyre takes the goal of a 'theory of morals' to be to makeclear the 'kind of backing [which] is logically appropriate to moralrules',1 and thinks that this 'backing' will have to be accounted forin terms of a 'theory of human nature'. But since he wants to insistthat moralities are diverse not merely in respect of their content buteven in respect of the logical form of their judgements, he holds thattheir respective systems of 'rules' will require backing in terms ofdifferent conceptions of human nature. And when he speaks ofdifferent conceptions of human nature here, I think he means notmerely different views about the specific needs human beings have,but also different views about the sense in which human beings canbe said to have needs at all, about the extent to which there is

Ibid., p. 148.

Human Nature 7anything constant about those needs, and about the kind of impor-tance in human life which those needs have. If, then, a man has tochoose (as far as his own life is concerned) between alternativemoral systems, he will need at the same time and by the same tokento decide between rival conceptions of human nature. For a reasonedchoice between different moralities requires a vocabulary in whichthe reasoning may be expressed; this vocabulary is provided by aman's 'social past', i.e. by the way in which moral conceptions havegrown up and developed in the context of historically conditioned'forms of social and moral practice'. Each such form of practice'carries with it its own picture of human nature. The choice of aform of life and the choice of a view of human nature go together.'1Thus the conception of human nature provides us with no 'neutralstandard' by reference to which such choices may be guided. Andin line with this conclusion Maclntyre maintains (on p. 148) that itis 'arbitrary and illegitimate' in examining the logical structure ofdifferent moralities to specify as the logical form of moral argumentthat form which is characteristic of argument within any one of them.

If this were all Maclntyre wanted to say I should not find muchcause to quarrel, beyond expressing a doubt about whether in thatcase the discussion were really helped by introducing the conceptionof human nature into it. But it turns out that this is not by any meansall that Maclntyre wants to say. He follows his argument for theillegitimacy of saying that any particular form constitutes the logicalform of moral argument with a discussion designed to show the'superiority' of the Aristotelian view of human nature and of itsrelation to morality in comparison with, e.g., those to be found inChristianity, the Sophists and Hobbes. And a question which criesout to be asked at this point is whether we are to take this view as anexpression of Maclntyre's own 'choice' - a choice of which it wouldbe improper to say that it is logically in any better shape than wouldbe a contrary choice - or whether, rather, he is making a logicalcomment about what can be said in general about the nature of suchchoices. This latter interpretation would be hard to reconcile withthose assertions of Maclntyre's to which I have already referred;nevertheless I think it is clear from the larger context of his argumentin the book as a whole that this must be the interpretation intended.

For Maclntyre the virtue of the Aristotelian view is twofold. Onthe one hand it does not take human needs and desires as a fixeddatum, but rather as a subject for criticism and for modification bysuch criticism. On the other hand - more to our immediate point - itdoes squarely link up duties with specific social roles and with the

1 Ibid., p. 268.

8 Peter Winchsocial rewards and satisfactions provided for one who fulfils thoseroles worthily within the life of his society. One of the main over-allcontentions of Maclntyre's book is that the notion of a moral dutyis originally at home, and is only fully intelligible, in the context ofa society offering its members more or less clearly defined rolestogether with definite satisfactions ensuing on their proper fulfilment.Thus the Kantian categorical imperative is interpreted as the pro-duct of a society in which this sort of connection has broken down:the word 'ought' is still used but lacks the social background whichalone would make it intelligible - a background, namely, in whicha man can have a rational expectation that if he does his duty he willbe rewarded with social satisfactions of a sort he can understand andaim at. In this connection Maclntyre refers1 to Kant's view - putforward in a theological context - that 'it would be intolerable if infact duty were not in the end crowned with happiness'. He commentsthat the indeterminacy of the notion of happiness when it is separatedfrom any 'socially established ends' turns this view into 'a tacitadmission that without some such notion, not morality itself, but theKantian interpretation of it scarcely makes sense'.

But the qualification in that last clause is really something of anevasion. For Maclntyre does want to say that Kant's interpretationis applicable to a way in which people really had come to use theword 'ought' in the society to which Kant was implicitly referring.So he must, apparently, think that the morality expressed in this useof the word itself'scarcely makes sense'. This interpretation is borneout by his treatment of our contemporary situation and of therelevance of Hare's theory of morals to it. Hare, so Maclntyre ineffect argues (e.g. on p. 266), is spokesman for that large section ofcontemporary mankind who adhere to no traditional morality andto whom 'those who speak from within [such a morality] appearmerely to be uttering imperatives which express their own liking andtheir private choices'.2 I take it that the reason why it appears likethis to such people is that this is in fact what they themselves aredoing when they use moral language. And it seems clear thatMaclntyre regards this use as one which does not make much senseas anything distinct from an expression of personal likes and dislikes.It appears, for instance, from what I said earlier in the lecture abouthis treatment of Hare, that he would regard such a use of languageas in no way appealing to any authoritative standard in a way whichwould make it rational for one man to be convinced by another thathe should change his 'moral' views.

Thus the prescriptivist account of moral language, on Maclntyre's1 Ibid., p. 196. 2 Ibid., p. 266.

Human Nature 9view, is not a misdescription but a quite accurate representation ofhow a great many people do use moral language: in a way, namely,that Maclntyre regards as not making much sense. What makes thisuse senseless, he thinks, is the absence of any social link betweendefinite roles and definite ends, the attainment of which carries arational expectation of some satisfaction for the agent. It looks asthough Maclntyre were considerably less liberal than once appearedconcerning what can and what cannot be accepted in morality. Heis willing to accept variations in the social goods and personal satis-factions with which the fulfilment of duties is linked; but he is notwilling to accept duties not linked with any such goods or satisfac-tions. And this, I must reiterate, not as a matter of his own moralstandpoint, but as a logical point about what can be understood asbeing a moral standpoint which can be made sense of at all.

A question we should ask at this point is what authority we shouldregard this general requirement of Maclntyre's as having for us. Howdoes he know that the expression of a 'duty' outside such a context issomehow incomplete and incoherent ? A remark of his own, at thebeginning of his book, is germane here. Historical study of ethicalconcepts, he says, may 'break down . . . our too narrow views ofwhat can and cannot be thought, said, and done . . . in face of therecord of what has been thought, said and done'.1 However, it is notthrough mere carelessness that he has got himself into this, as itseems to me, embarrassing position; rather it is due to a veryfundamental philosophical difficulty which confronts us all and aboutwhich I must now try to say something.

'The record of what has been thought, said and done' cannot betaken as meaning just the record of any old thing which anybody hasever felt like saying, thinking or doing. For people sometimes talknonsense and behave in ways which make no sense. As philosophersone of our primary concerns is precisely to distinguish sense fromnonsense. Here we have to ask what this involves and how we are totell when we have done it successfully. One of the chief uses to whichthe concept of human nature has been put by philosophers is in anattempt to answer this question. They have thought that it is bytheir relation, or lack of it, to human nature that we decide whatways of thinking and acting make sense and what do not. What ismost interesting about Maclntyre's book is that it shows somerecognition of the constricting effect of this way of thinking on ourinterpretation of human life and thought without, however, comingto the realisation that the whole matter is being represented herein a topsy-turvy way. Putting the matter briefly, and therefore

1 Ibid., p. 4.

io Peter Winchover-crudely, what I want to say is this: what we can ascribe tohuman nature does not determine what we can and what we cannotmake sense of; rather, what we can and what we cannot make sense ofdetermines what we can ascribe to human nature. It is indeed pre-cisely for this reason that the concept of human nature is not theconcept of something fixed and given; i.e. the reason for this is aphilosophical, not a sociological one.

A child is born within, and grows up into the life of, a particularhuman society. He learns to speak and to engage in various kinds ofactivity in relation to other people. In the course of these activitieshe encounters problems of extremely diverse kinds, problems whichchange in character as he matures, and problems that bring himinto new kinds of relations with other people. Along with thisdevelopment there comes a growth in his understanding of whatconstitute problems and difficulties for them. This growing under-standing manifests itself in the way he comes to treat people in thecourse of his daily life, which will include a development in his ideasof what is permissible in his treatment of them and what is ruled out.This growth in his understanding of other people through his deal-ings with them is at the same time a growth in his understanding ofhimself, which is in its turn a development of the kind of person he is.The way a person develops in these dimensions will be influenced bythe kinds of people, the kinds of situation and the kinds of problemswhich he finds himself confronted with in the course of his life. Butof course it is also true that his growth will depend on what he him-self brings to the situations he faces. Our problem is to understandthe relation between these two 'factors'.

But there is something misleading in this way of formulating theissue. A man is not 'confronted' with situations and problems as he is'confronted' with a brick wall. A brick wall confronts a man whetherhe recognises it or not; but the same is not unqualifiedly true of theother case.1 A passage from R. G. Gollingwood's Autobiography1 willhelp to bring out what I mean.

My father had plenty of books and allowed me to read in them asI pleased. Among others, he had kept the books of classicalscholarship, ancient history, and philosophy which he had used atOxford. As a rule I left these alone; but one day when I was eightyears old curiosity moved me to take down a little black booklettered on its spine 'Kant's Theory of Ethics'. It was Abbott'stranslation of the Grundlegung zur Metaphysik der Sitten; and as I

1 Cf. Peter Winch, Moral Integrity (Oxford, 1968) pp. 10 ff.

2 R. G. Collingwood, An Autobiography (Oxford, 1939) pp. 3-4.

Human Nature 11began reading it, my small form wedged between the bookcaseand the table, I was attacked by a strange succession of emotions.First came an intense excitement. I felt that things of the highestimportance were being said about matters of the utmost urgency:things which at all costs I must understand. Then, with a wave ofindignation, came the discovery that I could not understandthem. Disgraceful to confess, here was a book whose words wereEnglish and whose sentences were grammatical, but whose mean-ing baffled me. Then, third and last, came the strangest emotionof all. I felt that the contents of this book, although I could notunderstand it, were somehow my business: a matter personal tomyself, or rather to some future self of my own. It was not like thecommon boyish intention to 'be an engine driver when I grow up',for there was no desire in it; I did not, in any natural sense of theword, 'want' to master the Kantian ethics when I should be oldenough; but I felt as if a veil had been lifted and my destinyrevealed.

Our first reaction to this passage is likely to be that the youngCollingwood must have been a very remarkable little boy. Thisreaction expresses our perception that for most people of that age -indeed of any age - coming across Kant's Grundlegung would not poseany problem at all, but merely an occasion to put the book down asquickly as possible and look for something more entertaining. Avocation needs someone with ears to hear as well as a call. But ofcourse a call is necessary too and we should ask what are the condi-tions necessary for this. In Collingwood's case it was necessary thatthere should have existed a tradition of thought in which there werecertain live problems, questions which had not been answered.Someone who feels moved to ask himself such questions, who,speaking more generally, finds that his situation confronts him withproblems, discovers something about himself by that very fact - asthe young Collingwood discovered something about himself in dimlyfeeling the force of the questions Kant was asking. But we could notsay what a man discovers about himself in such a situation withoutreferring to the nature of the problems he finds himself faced with;for it is in his response to those problems that he makes the discovery.And we understand what those problems are only in the context ofthe traditions of thought and activity out of which they arise. Thefact that some will see problems where others will not may tempt usto say that the capacity for understanding must be 'in' someonefrom the start. But while it is not exactly wrong to say this, it maymake us overlook the fact that the possibility of such capacities, as we

12 Peter Winchmight say 'the space within which such capacities can exist', comesfrom outside. The question whether or not a given individual has acertain capacity - e.g. to understand higher mathematics - can onlybe asked within the context of the problems of higher mathematics.We may be able to ask of someone from a culture with no mathe-matical traditions whether he is the sort of person who would havebeen able to become a mathematician - though it is a pretty tenuoussort of question - but we, who ask the question, necessarily do havesome relation to such traditions; in the absence of such traditionssuch a question could not be asked at all.

There are two objections to this way of looking at the matter whichcome together in the appeal of 'human nature'. In the first place itmay be said that while what I have called 'the space' within whichhuman capacities of various kinds can exist comes from outside thenature of any given individual, it is still conditioned by humannature in that the systems of ideas and practices within whichindividual capacities may be conceived and may develop are systemsof human ideas and practices and are therefore limited by whathuman beings can attach sense to. The second objection reinforcesthe first. I, as one individual, find other human beings makingvarious claims to be able to do and think certain things which I haveto assess: that is, I have to decide whether or not those claims areintelligible. It is no good saying that the intelligibility of the claimsdepends on whether there is a space within which they can be fitted,because the question of their intelligibility precisely is a questionabout the existence of such a space.

This point is quite valid as an objection to the idea that the con-cept of 'the space within which human capacities exist' provides acriterion for distinguishing sense from nonsense. But then I haveintroduced the concept precisely in the context of an argument forsaying that there is no such general criterion. And in fact the objec-tion is just as fatal to the idea that 'human nature' provides such acriterion. Maclntyre's arguments for the relativity of the conceptionof human nature to particular historical traditions and periods reallyshow this. He tries to avoid the full consequences of his own argu-ment by saying in effect that while men's particular needs vary fromone form of society to another, and while the form of a society'smorality is at the same time a determinant of the needs of thatsociety's members, still what is not variable is that an intelligiblemorality must provide for the satisfaction of its practitioners' needs ifthey do what is morally required of them. Like most philosophicalpositions, this one can be held as a last bastion as long as one is pre-pared to drain it of content. One can, for example, say that in so far

Human Mature 13as a man does feel that something is required of him, he feels a needto do what is required of him and this need is satisfied when he doesit. But this is pretty obviously no more than a juggling act with thewords 'need' and 'satisfaction'; it tells us nothing whatever about theform of a moral requirement and places no limits on what can bemade sense of. For a philosopher will still have to ask about the con-ditions under which he can accept, or must reject, a man's claim tosee that something is required of him. What, then, can we say aboutthis crucial question ?

It is very common in philosophical discussion to find that a wayof speaking and thinking which seems perfectly intelligible andacceptable to oneself is met with incomprehension by other people;and of course vice versa. In such circumstances one is bound to askwhether what one wants to say really does make sense or not.Discussion with other people will often be very important in one'sattempt to settle this question - it was, after all, the incredulous,uncomprehending, reactions of others which prompted the doubt inthe first place (though this need not always be so). Discussion willtake the form of raising difficulties and trying to see one's wayaround them. Sometimes one will think one has dealt with thedifficulties satisfactorily; sometimes that the view under criticismmeets with difficulties so insurmountable as to be really incom-prehensible. But if one recognised the possibility of being mistakenin one's initial belief that one had understood what was being said,or that one had shown it to be unintelligible, one can equally, afterdiscussion, recognise that one may have over- or underestimated thedifficulties which have emerged in its course. But that does not meanthat one's views are subject to the test of some ultimate criterion, thecriterion of what does and what does not belong to human nature.It means only that new difficulties, and perhaps new ways of meetingthe difficulties, are always lurking below the horizon and that dis-cussion continues. Sometimes, if one is lucky, the discussion clarifiesor extends one's conception of what is possible for human beings. Butit is no use saying that this is contingent on what is or is not possiblefor human beings; for our only way of arriving at a view about thatis by continuing to try to deal with the difficulties that arise in thecourse of discussion.