The Psychology of Social Class 
by Michael Argyle.
Routledge, 305 pp., £13.99, December 1993, 0 415 07955 1
Show More
Show More

The sociological method of studying social ‘class’ is hopelessly misconceived. The trouble with it is basic: it has mistaken the nature of the subject. The sociological method flouts two fundamental truths. The first is that the proper way to study ‘class’ is by introspection: by prolonged reflection on what is going on in oneself when one thinks ‘class’ thoughts – a most devious and complex business, full of ruses and logical paradoxes.

The same is true of many gender issues and of ‘race’, which is a far simpler concept than ‘class’. By introspection – that is to say, by asking oneself what ‘race’ one belongs to oneself – one will fairly soon come to realise that, in any socio-political context, the concept of ‘race’ is almost entirely vacuous. Given that, as many people would agree, ‘race’ is a baneful concept, it is the duty of the intellectual to do his or her best to think it through, as a prelude to unthinking it; and introspection seems to be the ideal method. But, if less obviously, ‘class’ also seems to me a baneful concept and one which we need at least to try to unthink, though the task will be enormously harder with this cunning and elusive concept.

Not everyone would agree, of course. There are those who, like John Betjeman, relish (take an ‘unholy pleasure’ in) the game of social ‘class’. And Marx decided that the situation of the exploited could only be bettered by theorising about ‘class’ – i.e. by envisaging underprivileged workers as a ‘class’(the ‘proletariat’), a step which automatically called into being another ‘class’ (the ‘bourgeoisie’) – and it is not certain that he was wrong. Characteristically, though – and here the nature of ‘class’ and ‘class’-thinking is beginning to reveal itself – the history of this piece of entity-creation on Marx’s part is exceedingly complex and chequered. One thinks of his awkward taxonomic jugglings over a sub-class, the ‘Lumpenproletariat’, and the special-pleading he was forced into about the ‘objective’ interests of the ‘proletariat’ as opposed to what its members might at any time think their interests were. One remembers, too, his masterly and mischievous stroke of positing a two-class system rather than a three-class one – leaving dukes to wonder, bewilderedly, how they had come to end up in the bourgeoisie.

The thing to get hold of in ‘class’-thinking – a process in which, by dint of classifying oneself, one is automatically classifying others – is its labyrinthine nature and logical complexity. This brings us to the second of my fundamental truths: that the appropriate discipline for studying ‘class’ is not sociology but rhetoric. To take one example: according to sociological questionnaires, it is usual to find that a very small number of those questioned assigned themselves, unprompted, to the ‘lower middle class’. The reason will appear at once to anyone prepared to be introspective: it is that ‘lower middle class’ is essentially a category for others.

Or again, there is the striking fact that many people, at least in Britain, entertain simultaneously in their head a three-class system (‘upper class’, ‘middle class’ and ‘working class’) and a two-class system (‘bourgeoisie’ and ‘proletariat’), accommodating or masking the mismatch between the two as best they can. This leads us to consider a very important rhetorical figure (or ‘figure of thought’) which we may call the ‘self-excluder’. According to this figure, Sartre or Roland Barthes will heap obloquy on the ‘bourgeoisie’ while leaving quite unanswered the question of what ‘class’ they belong to themselves. The natural inference would be that they are members of the ‘proletariat’; but if so, what becomes of the ‘indicators’ beloved of sociologists? Where are Sartre’s or Barthes’s Gallic equivalent of cloth-cap and whippets? Such intricacies and manoeuvres aren’t accidental anomalies. They constitute the very nature of ‘class’ and ‘class’-thinking, and there is nothing beyond them.

Let us consider another feature of the ‘class’ business. When ordinary people come to discuss ‘class’ they tend to be unstoppable. There passes before their mental eye an endless pageant of evanescent ‘images of class’ – perceptions about pronunciation and clothes and haircuts, choice of newspaper or recreation, table-manners, sexual attitudes and modes of greeting and farewell. They pour forth a torrent of theory, or rather of theories, picked up who knows where: from the newspapers, or the writings of Marx or at their grandmother’s knee. They regard the topic, though proverbially ‘inexhaustible’, as one that can only be studied by the exhaustive method – resembling in this the Renaissance taxonomists described in Foucault’s Les Mots et les choses. To classify the serpent according to this method you are supposed to heap up everything possible to be known about the serpent: puns on its name, etymologies, methods for its capture, remedies for its bite, allegories, adages, dreams and its role in literature and heraldry.

Sociologists are even more unstoppable about ‘class’ than ordinary people, and for a very good reason. According to their method of studying the subject, theories – anybody’s theories – are also ‘facts’, to be quantified and arranged in graphs and tables like other facts. The sociologist, in a word, is doing what no self-respecting ethnographer would dream of doing. He is trying to take up a position simultaneously inside and outside a belief-system; assigning it a truth-value, yet ‘correcting’ it against some supposed principle external to it. An ethnographer studying the animistic beliefs of a Brazilian tribe cannot afford to be a paid-up animist himself.

Thus the reviewer of Michael Argyle’s The Psychology of Class is confronted with a problem: he is forced to blame the author for what is not so much his fault as the fault (as you might say) of the curriculum. All the same, it is not very hard to demonstrate from his book that something is terribly wrong with the sociological approach to ‘class’. ‘Everyone knows what class is,’ he tells us in his first chapter. He asserts this because, in a national poll, 95 per cent of people in Britain said that they belonged to a ‘class’, and named it. But this is rather what you would expect; for if you believe there is such a thing as ‘social class’, you would feel rather foolish in saying you didn’t belong to it, or that, if you did, you didn’t know where.

At all events, the very next sentence or two in Argyle’s text show how little has so far been established. For if you conduct a sampling of sociologists on the question of what ‘class’ is, a small percentage (ought we not to be given a figure, since this is sociology?) will say that ‘class’ is ‘what people think their class is’, and the remainder will give a wide variety of answers: they might hold that it is defined as ‘status’, or as ‘prestige’, or as ‘power’, or as ‘income’, or as ‘wealth and property’, or (this will be the social psychologists speaking) as based on ‘which individuals accept each other as equal and which defer to others’. So either all these remaining sociologists are wrong, or Argyle has misled us in saying: ‘Everyone knows what class is.’ Then we learn that the results of a questionnaire will vary dramatically according to whether you do, or do not, tell those whom you are questioning how many ‘classes’ there are for them to choose from, and whether, in the former case, you tell them there are three, or four. But what else could one possibly expect?

Next we hear that a source of ‘mismatch’ between ‘subjective class’, i.e. the ‘class’ people claim to belong to, and ‘objective class’, i.e. the ‘class’ to which a sociologist assigns them – which is evidently going to differ wildly according to whether the sociologist is of the ‘status’ school or the ‘power’ school or the ‘income’ school – may be an error on the part of the ordinary person: ‘some individuals simply place themselves incorrectly.’ sociologists are allowed more latitude. British sociologists ‘like’ (i.e. prefer) to ‘measure class by occupation’, doing so ‘from theoretical reasons, partly of Marxist origins’, but this is not the practice in the USA, ‘where class is usually assessed by education or income now’. Nevertheless, sociologists, too, make mistakes. ‘If sociologists measure the class of families by husband’s occupation they will be wrong some of the time.’ ‘Wrongness’ seems a strange concept to invoke in this relativistic chaos, this dizzying congeries of optical illusions.

Further, it seems rather hopeless to discuss ‘class’, as Argyle does, as though it were a phenomenon of all places and all times, embracing Hindu caste and feudal tenures as well as the scene outside Sartre’s or Nancy Mitford’s window. No profit whatever can be attained from an ahistorical approach to ‘class’. Indeed, I would say that the only approach which does yield profit is to treat the history of social ‘class’ as, strictly, the history of the term ‘class’ in the social sense – beginning with its advent in England at about the time of the first Reform Bill, when it replaced ‘orders’ and ‘ranks’. The political struggles of this period explain why we speak, unsymmetrically, of ‘upper’ class, ‘middle’ class and ‘working’ (not ‘lower’) class. It is, similarly, a significant fact that, in these early stages, the ‘middle class’ is a purely middle-class concept. One can read one’s way through E.P. Thompson’s The Making of the English Working Class without once fmding a radically-minded worker invoking the ‘middle class’, all his accusations being levelled at ‘the fiend aristocracy’ and ‘Old Corruption’.

When we read Cobden or Matthew Arnold, by whose time the terminology of ‘class’ is beginning to settle down, we are or should be continually amazed at the brilliant and devious manoeuvres they play with ‘class’ terms, especially ‘middle class’. Having, in Culture and Anarchy, said the most frightful things about the ‘middle class’ and its ‘hideous and grotesque illusions’, Arnold turns round and ‘humbly offers himself’ as an illustration of the defects and qualities of this class. Of course he is being waggish; and to the question what ‘class’ he actually thought he belonged to, the answer is, he didn’t think he belonged to any. He ‘placed’ himself socially in another system altogether, the binary system of ‘gentleman’ versus ‘no gentleman’.

By thinking one’s way, like this, into the involved windings of the concept ‘class’, one comes to appreciate its relationship to other favourite ploys of social rhetoric, such as the older one of ‘hierarchy’. Social historians tend to misrepresent ‘hierarchy’, I think, and it is a comedy to see them struggling, with the aid of Gregory King’s table of 1690 and of much fudging, to demonstrate that there was a single social hierarchy in 17th-century England – whereas all that magistrates and divines meant to declare was that there ought to be one (so that everyone should ‘know their place’). Of course, in reality, everyone in England belonged to a criss-cross of hierarchies, occupying a different place in each. This, with all respect to social historians, is how life is.

Another important member of this genre of social ploys is ‘snobbery’. The concept was invented by Thackeray, and its originality and profundity – as well as its touch of perversity – are not always properly appreciated. (It has never really been grasped by any other culture.) In his ‘barbarian’ youth Thackeray had edited a magazine entitled the Snob, in which the term ‘snob’ (literally, a ‘cobbler’) was a synonym for ‘vulgar’ and ‘ill-bred’; but in his Book of Snobs of 16 years later, the abusive term ‘snob’ is reversed, precisely, onto those who would use it in the sense he had given it in his unregenerate youth. According to the new theory, the vulgarest thing you can do is to look down on somebody for being ‘vulgar’ (in the earlier sense) or to curry favour with somebody for not being ‘vulgar’. This new theory has two essential and conflicting features: that for the first time there is a term which will serve equally for servility towards those ‘above’ oneself and arrogance towards those ‘below’; and that there is no escape from the trap of snobbery, since to concern yourself with it is itself snobbish. The logical antinomy between these two features is of a Freudian kind.

Altogether, in ‘class’, ‘hierarchy’, ‘snobbery’ and ‘the gentleman’ (another extraordinarily complex and slippery term-cum-ploy) we have, from a conceptual point of view, the tissue of much of English life and history. Thus it is surprising, now that rhetorical studies are so much to the fore again, that they have not received more analysis of the kind they obviously demand.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.


Vol. 16 No. 5 · 10 March 1994

Craig Brown’s remark, cited by Frank Kermode (LRB, 24 February), that Mark Boxer always ‘homed in on the hair’ in those enjoyable caricatures of his made me think of Aubrey Beardsley’s illustrations to Morte Darthur and wonder whether ‘Marc’ might have been directly influenced in this by Beardsley. Then Morte Darthur led me to wonder why it should ever be supposed, as Tom Shippey, in the same issue, tells us it repeatedly has been, that a man who put the words that Sir Thomas Malory did into the mouth of Sir Launcelot could not be both a knight and a rapist himself: need we look further than the words put into the mouth of Sir Meliagrance confronting Queen Guenever (in the edition which Beardsley illustrated. ‘As for all this language, be it as it may, for wit you well madam, I have loved you many a year, and never or now could I get you at such an advantage as I do now, and therefore I will take you as I find you’)? And Malory then led me to wonder whether P.N. Furbank seriously disputes that in 20th as in 15th-century England there are systematically observable inequalities of economic, ideological and political power to which the contemporary rhetoric relates in all sorts of still understudied ways.

W.G. Runciman
Trinity College, Cambridge

Vol. 16 No. 8 · 28 April 1994

P.N. Furbank (LRB, 24 February) might not have needed elucidation of my letter in the previous issue if the editor hadn’t cut out of it a phrase about ‘the slippery rhetoric of class’. Furbank’s review of Michael Argyle’s The Psychology of Social Class seemed to me to imply that the rhetoric is, in effect, the reality, and that academic sociologists and social psychologists who try to measure ‘class’ objectively only succeed in making themselves look foolish. Of course we don’t always get it right. But I am relieved that Furbank agrees (if he does) that our efforts are not inherently misconceived.

W.G. Runciman
Trinity College, Cambridge

The final sentence of W.G. Runciman’s letter (Letters, 10 March), in its original version, ran as follows: ‘And Malory then led me to wonder whether P.N. Furbank, whose remarks about the slippery rhetoric of “class" can be as well illustrated by reference to Malory as to anyone, seriously disputes that in 20th, as in 15th, century England there are systematically observable inequalities of economic, ideological and political power to which the contemporary rhetoric does indeed relate in all sorts of still understudied ways.’

Editor, ‘London Review’

P.N. Furbank writes: I was puzzled by Lord Runciman’s original letter, but his new one makes all clear, and I am afraid his worst suspicions are confirmed. I do think that, in the matter of social ‘class’, the rhetoric is the reality, and that (to be brutal) academic sociologists and social psychologists who try to ‘measure’ class objectively only succeed, as he puts it, in making themselves look foolish. Let us put the matter this way. The language of ‘class’ only began to be employed in Britain round about the time of the first Reform Bill, being then taken over a decade or two later, to very different effect, by Proudhon and Marx. Now, the uses to which this language was put, and is still put, is a very rich and important study, but one needs to be clear about what one is studying. It is, in the first place, not something objectively ‘out there’, in the world, nor is it something which it makes sense to speak of ‘measuring’. To use ‘class’ language is a form of social (and often also political) action; for it is inherent in the concept of ‘class’ that to classify others in this way is automatically also to be classifying oneself, it being an absurdity to imagine one can both use the system and yet stand outside it. In using ‘class’ terminology one is manoeuvring socially or politically against one’s fellow citizens. Thus, a sociologist or a social historian is bound in logic to give up any idea of using ‘class’ terminology in his profession (what he does in his private life being quite another matter). All he can do, by way of studying ‘class’ scientifically, is to examine what is going on when ‘class’ language is used (i.e. analyse it as a belief system and form of rhetoric). This, to my mind, is a most important study, much neglected; and, as I argued in my review, a great deal of it can best be done by introspection. In this respect the student of ‘class’ has an advantage over the ethnographer, who is less likely to share the belief-system of the community he or she is studying. But to imagine you can use the language of ‘class’ scientifically, and even (as some social historians do) use it about people who did not use it themselves and employed a quite different language, strikes me as a hopeless illusion.

I can’t help feeling that P.N. Furbank must be missing the point. The letter from Lord Runciman to which he refers was surely a brilliant parody? Even if we are not so sure of the exact target.

Sheila Stern

Vol. 16 No. 10 · 26 May 1994

The exchanges between P.N. Furbank and Lord Runciman about the use of ‘class’ as a term for describing social inequalities, past and present, leaves unsettled, indeed undiscussed, the problem of what language could replace it (Letters, 28 April). In his review of Michael Argyle’s The Social Psychology of Class (LRB, 24 February), Furbank argued that the deployment of ‘class’ categories could never claim to be ‘scientific’: the term and its variants were too treacherously rhetorical. Lord Runciman’s attitude to this charge is not easy to describe. Pained bafflement, perhaps, at coming upon such distressing stuff from an LRB reviewer? A fuller answer than he so far has given would be valuable. Meanwhile, the question remains: how should social inequalities be described – unrhetorically? Furbank suggests that since ‘class’ language only emerges in the 1830s, the social historian for earlier periods should use the language of the time. So, when Richard Holmes asserts that, in 1811, Harriet West-brook’s father was ‘achieving that most difficult piece of English social navigation: moving from the lower middle class to the upper middle class’ (Shelley: The Pursuit), the terms are anachronistic. But what others were available? Shelley himself was ‘gentry’ (son of a landowner, knight and MP). Mr West-brook, having made money in ‘trade’ and retired to a house near Belgrave Square, though not ‘gentry’, would have surely thought himself ‘a gentleman’, indignantly rebutting any suggestion that young Harriet was other than a ‘lady’. Hadn’t he educated her at that expensive Clapham school attended by Shelley’s sisters? Yet he would have had to agree that the young man with whom she eloped was socially her superior, and even perhaps his.

Turning to fiction, there was the Bennet family and their friends, shortly to enter public life. The Bingley sisters, carefully forgetting that their late father’s fortune derived from ‘trade’, sneer at the Lucases for their all-too-recent commercial connections. Elizabeth rejects, with confidence, Lady Catherine de Burgh’s assumption of social superiority to her family, Mr Bennet also being a landowner, if on a far smaller scale. But the ‘gentleman’ idea in Austen’s work, here as elsewhere, refers not to the ownership of property, nor to where the money comes from, but to social style, easy manners, a ready flow of small talk, together with the self-discipline required by codes of etiquette, and (presumably) also to dress. This seems to be the force of ‘gentlemanlike’, as in: ‘Bingley was good looking and gentlemanlike; he had a pleasant countenance, and easy, unaffected manners.’ The implied similitude – ‘gentlemanlike’ – is a little puzzling. Does it point to a reservation? Despite Wickham’s scoundrelly behaviour, his qualifications as a ‘gentleman’ are not brought into question. On the whole, in interesting contrast with Victorian usage, ‘gentleman’ seems to have little or no ethical or intellectual content. Bingley’s brother-in-law, the empty-headed Hurst, ‘merely looked the gentleman’. Austen, that is, is less interested in social stratifications, save as an observable fact about the world she depicts, than in the moral and intellectual qualities that particular ‘gentlemen’ might or might not possess. For all its keen and detailed observation of early 19th-century Home Counties life, her novel thus offers no general language for describing social changes that were in process, and which Richard Holmes attempts to describe in his remarks about Shelley’s father-in-law. Becoming ‘Lady Shelley’ would have been thought by Harriet, as also by her circle, as socially advantageous a move as Mrs Bennet thought Elizabeth’s marriage to Darcy. If the language of ‘class’ is anachronistic as a general description of such events, the language of the period (as used by Austen) is no substitute. Furbank’s argument thus seems to condemn the social historian, if not to silence, then to the mere accumulation of detailed cases incapable of being generalised.

The other aspect of his analysis (that to use ‘class’ language is always to engage in rhetorical manoeuvring) is harder to cope with. For if ‘class’ is allowed its full claim, the user is contemplating our species as a botanist or zoologist contemplates plants or animals, allotting him or herself the authority of a more powerful ‘kind’, and – here is the rhetorical move – inviting readers to share the flattering illusion. But ‘class’ is commonly used to mean no more than ‘group’, rough-and-ready perhaps, but usage often is.

Graham Martin
London SW3

Vol. 16 No. 6 · 24 March 1994

Lord Runciman asks (Letters, 10 March) whether I would seriously dispute that in 20th as in 15th-century England there are ‘systematically observable inequalities of economic, ideological and political power to which the contemporary rhetoric relates in all sorts of still understudied ways’. The answer is that I certainly would not – how could I? To be honest, I do not understand why he asks, and wish he would explain.

P.N. Furbank
London NW5

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN

Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences