Copyright © John Scott 1996

The right of John Scott to be identified as author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

First published in 1996 by Polity Press in association with Blackwell Publishers Ltd.

Reprinted 2004, 2007

Polity Press
65 Bridge Street
Cambridge CB2 1UR, UK

Polity Press
350 Main Street
Malden, MA 02148, USA

All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the publisher.

Except in the United States of America, this book is sold subject to the condition that it shall not, by way of trade or otherwise, be lent, re-sold, hired out, or otherwise circulated without the publisher’s prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser.

ISBN: 978-0-7456-1041-2
ISBN: 978-0-7456-1042-9 (pbk)

A CIP catalogue record for this book is available from the British Library and the Library of Congress.

For further information on Polity, visit our website: www.polity.co.uk

Figures

Figure 1 Social classes and social estates

Figure 2 Social classes, social estates and social blocs

Figure 3 A typology of class situations

Figure 4 A translation of normative functionalist terminology

Figure 5 Warner’s social class schema

Figure 6 Caste and class in America

Figure 7 Stratification in the Chicago ghetto

Figure 8 Mosca on elite recruitment

Figure 9 Pareto’s model of the governing elite

Figure 10 Social classes in Germany, 1965

Figure 11 Wright’s initial four-class schema

Figure 12 Contradictory class locations

Figure 13 Wright’s later class schema

Figure 14 Class locations in the United States and Britain, 1980

Figure 15 Goldthorpe’s eight social classes

Figure 16 Social mobility in Britain, 1972

Preface

The framework presented in this book has been developing, in one form or another, for a number of years. Indeed, my use of the contrast between ‘class’ and ‘status’ dates back to an undergraduate essay and to my PhD thesis, completed in 1976. Thanks to a suggestion from David Held that I should write the book, I gradually came to clarify the central idea of a three-dimensional approach to social stratification. In particular, work on the book helped to remove the lingering reservations that I had about the value of the concept of ‘elite’. This had always seemed to me to be a confused and unnecessary idea, and it was not until the nature of the third dimension of stratification was clarified that I was able to see how it fitted into a comprehensive framework. Many confusions do, indeed, surround the use of the word ‘elite’, but I do now believe that the concept must play a central part in the analysis of social stratification. The key to this rehabilitation of the idea of ‘elite’ was to recognise authority and command relations as an autonomous dimension of stratification alongside the more familiar recognition of ‘class’ and ‘status’ relations.

Some of the arguments of the book can be seen as extended, and rather belated, responses to questions that I was asked at job interviews. When I was interviewed for a lectureship at Leicester in 1975, the economic historian the late Ralph Davis asked me – as he did all the candidates – what I understood by the phrase ‘working class’. I cannot now recall my answer, but I discovered later that Davis felt that none of the candidates had given a satisfactory answer and that this was fairly typical of sociologists. I hope that my discussion of the working class in chapters 1 and 8 goes some way towards answering Davis’s question and vindicating sociologists. When I was interviewed for a chair at Essex in 1993, Tony Giddens asked me how I would justify my emphasis on ‘class’ to the person in the street who claims that we live in a classless society. Again, my answer was inadequate – though I blame the inadequacy, in part, on the brevity of the time allowed to answer. The whole of my discussion in chapter 1 is a preamble to the more systematic argument of the rest of the book that, I hope, answers Giddens’s question more adequately.

The emerging ideas of the book have benefited from discussions at a number of institutions. They have been used in courses at Leicester and at Essex, and the preparation and delivery of these courses has helped me to refine the framework and to explore its applications. Informal discussions with colleagues at both institutions have helped to shape the ideas and the book, as have seminar and workshop discussions at a number of institutions. Most recently, colleagues at Plymouth and Reading Universities have provided helpful comments on the developed version of the framework. Numerous individuals have contributed to the development of the ideas over the years, both in conversation and in writing. A number of these have kindly commented on drafts of various parts of the book, and I would particularly like to thank Barry Barnes, Fiona Devine, David Lee, Nirmal Puwar, Garry Runciman and Malcolm Waters. Anonymous readers for Polity Press provided very useful comments.

The final version of the manuscript was produced while I was still ‘in limbo’ after moving to Essex University, spending weekdays away from home for over a year. The long quiet evenings at West Lodge, on the Essex campus, provided the opportunity to get on with the work, while the depression of living away from home and trying to negotiate a way through a collapsing housing market made it all but impossible to make best use of this opportunity. I hope to have completed my long period of transition by the time that this book appears.

John Scott
West Lodge, University of Essex

1

Images of Stratification

The social stratification of a society can be most straightforwardly defined as its internal division into a hierarchy of distinct social groups, each having specific life chances and a distinctive style of life. In contemporary societies, social stratification has most typically been described in the language of ‘class’ and, in Britain in particular, ‘class’ divisions and ‘class’ distinctions have been a perennial topic of both popular and political discussion. The concept of ‘class’ has also been central to sociological discourse. Indeed, it has often been seen by critics of sociology as a defining characteristic of the discipline: sociologists, they hold, reduce everything to class. While this criticism is overstated, there is an element of truth in it. The sociological emphasis on class can be traced back to the ideas of Karl Marx, who saw the history of all societies as grounded in the revolutionary struggles of social classes. Weber and Durkheim were no less convinced of the centrality of class conflict to the struggles of their times, and it was the ideas of these ‘founding fathers’ that shaped contemporary sociological concerns (Dahrendorf 1957; Aron 1964; Bottomore 1965; Giddens 1973a).

American sociologists have tended to put less emphasis on class than have their European counterparts, reflecting a popular view that American society is more ‘open’ and less divided by class. England, it is often claimed, is a peculiarly ‘class-ridden’ society, its members being obsessed with the minutiae of accent, schooling, dress and behaviour. America, by contrast, has invariably been depicted in popular commentary as being a particularly ‘open’ society: even a ‘classless’ society. In such a society – a society of ‘opportunity’ – people can move up and down the social hierarchy with great ease, and there are no marked differences of culture or life style. This image of ‘classlessness’ has served as a foil for critics of the snobbery and ‘class distinction’ that are alleged to deform British society and to disadvantage many of its members.

This image of ‘openness’ can be found behind the claims of many American commentators that class is a factor of declining salience in all contemporary societies. ‘Class’, such commentators hold, is an outmoded nineteenth-century idea that has little relevance for understanding an advanced industrial or post-industrial society (see Nisbet 1959). The drive towards full modernity, it is argued, eliminates outmoded class distinctions and leads to a society in which merit and ability count for more than social background. ‘Class’ is ceasing to have any relevance for individual and social identity, having been supplanted by the more salient divisions of gender, ethnicity and sexuality. ‘Class’ is dead, and new identities have arisen (see the debate in Lee and Turner 1996).

The increasing acceptance of this view has produced something of a crisis for class analysis. Once this was the mainstream of the discipline, but now its practitioners seem to be stuck in a backwater. Paradoxically, this has been associated with the appearance of numerous texts on class and stratification (Scase 1992; Edgell 1993; Crompton 1993; Breen and Rottman 1995; Devine 1996) and a continuing stream of monographs (Erikson and Goldthorpe 1993; Westergaard 1995). What is striking, however, is the great diversity in this output, perhaps reflecting the crisis in class analysis. My intention in this book is, in the words of a group of American sociologists, that of ‘bringing class back in’ (McNall et al. 1991). I seek to return the analysis of social stratification to the mainstream of the discipline by providing a revamped set of conceptual tools that can make sense of popular views on ‘class’ and can show how the contemporary malaise in the sociological analysis of stratification can be seen as a misreading of contemporary trends. While people in their everyday lives may, indeed, now be less likely to identify themselves in ‘class’ terms, this does not mean that class relations, as objective realities, have disappeared.

I will argue, however, that the apparently simple word ‘class’ has been overloaded with meaning and has been stretched beyond its defensible, core meaning. I will also show the relationship between class structure and the consciousness of class to be empirically quite variable. Much popular and academic discussion of class ignores this distinction between ‘structure’ and ‘consciousness’. Indeed, most discussions of ‘class distinctions’ and ‘classlessness’ are not concerned with ‘class’ at all, but with what Max Weber termed ‘status’. They focus on issues of prestige and social honour rather than those of differences in economic power. The distinction between class and status is, I hold, fundamental to any viable investigation of social stratification, and a return to Max Weber’s ideas is the means through which the current crisis can be resolved.

The distinction between class and status has a long history. Medieval writers had generally described their social worlds using an imagery and vocabulary of estates, legal or quasi-legal categories of people that were defined by their social functions and responsibilities and that occupied distinct positions in a social hierarchy of status. In modern thought, by contrast, it was the imagery and vocabulary of classes that seemed to offer a more plausible basis for social understanding. Classes were seen as economic categories that were defined by their position in the system of production and that formed themselves into groups that entered into political struggle with one another. Classes were seen as rooted in inequalities of property and income that cross-cut ‘traditional’ status distinctions and created new forms of social division. The transition from medieval to modern societies, then, was seen as a process of social change in which stratification by ‘status’ was giving way to stratification by ‘class’.

The concept of ‘class’ first emerged as a central theoretical concept in the socialist tradition of political thought, where it was used to describe economically founded social divisions. It was particularly through Marx and Marxism that this view had a major impact on sociological ideas and on popular and official discourse. Very early on, however, the concept was stretched from a purely economic idea to one that grasped political and ideological divisions as well, ‘classes’ coming to be seen as collective historical actors. Weber sought to reappropriate the concept’s core meaning, restricting its reference to the role of economic power and resources in the generation of advantages and disadvantages. This conceptualisation of ‘class’ was contrasted with that of ‘status’, which Weber saw as referring to moral judgements of relative social standing and differences of life style. Taken together, he believed, the concepts of class and status provided powerful analytical tools that had a greater purchase on the social realities that political and popular discourse had attempted to understand through the single word ‘class’.

This theoretical distinction between class and status was not original to Weber, being found in many of the leading German sociologists, though it was Weber who gave it a particularly clear expression. Sombart (1902), for example, used the distinction in his contrast between the ‘organic’ societies of the medieval past and the ‘mechanistic’ societies of the modern era. In organic societies, distinctions of status separated groups that each had a common way of life and a specific legal and political identity. In a mechanical society, on the other hand, class divisions were rooted in individualised differences of economic interest. Tönnies (1931) drew a similar contrast as one feature of his distinction between gemeinschaftlich and gesellschaftlich societies. Weber’s particular contribution was to have allied this historical perspective to his methodology of the ideal type and, in so doing, to convert the concepts into analytical distinctions that could be used in the analysis of all societies. While there may, indeed, be ‘status societies’ and ‘class societies’, status and class coexist, in varying combinations, as features of all societies.

Not all sociological discourse has followed Weber’s usage. Marxist writers have generally continued to rely on an ‘economic’ concept of class alone and to see ‘status’ – to the extent that it is considered at all – as an aspect of the ideological mystification of class relations. The mainstream of American sociology, on the other hand, has tended to follow popular discourse and has conflated the two ideas into a single concept that emphasises social standing and relative ‘prestige’ and that minimises economic divisions. Thus, in much American social thought the word ‘class’ is used to designate the social rankings and judgements of relative social standing that Weber had termed ‘status’. This confusion reflects the reluctance of American commentators to see ‘classes’ as collectively organised social groups. Instead, the ‘open’ character of American society has been emphasised, and its stratification system has been depicted as a social hierarchy with numerous grades and no sharp boundaries. Vance Packard’s enormously popular book on The Status Seekers (1959), for example, thoroughly mixed class and status ideas, arguing that stratification in American society was defined by patterns of education and consumption that underpinned social mobility and status attainment.

This conceptual confusion in academic and popular discourse on social stratification has given credence to the views of those commentators who have suggested that the idea of ‘class’ should be abandoned. The concept is, they argue, purely rhetorical and has no scientific value for the study of social reality. Furbank, for example, has argued that

the terms ‘middle class’, ‘upper class’, ‘working class’ work most unproblematically not as nouns but as epithets – impressionistic epithets … For their power and attraction seem to lie, partly, precisely in the scope that they offer for prevarication, deviousness and the playing of social and political games. They are, essentially, rhetorical concepts. (1986: 5)

The solution to the crisis in stratification research does not, however, consist in abandoning the concept of class. My argument is that the crisis can be overcome if researchers return to the analytical distinctions that were made by Weber.

A coherent and systematic conceptual framework can be built from Weber’s distinction between ‘class’ and ‘status’ and from his related analysis of ‘authority’. Relations of authority establish powers of command among the members of a society, and they are a frequent accompaniment to class and status relations. They must, however, be distinguished from them for analytical purposes. The discussion of command has, in fact, proceeded in virtual isolation from the discussion of class, though not a few writers – most notably Mosca – have attempted to redefine ‘class’ in terms of the holding of powers of political command. The framework that I derive from Weber’s work provides a basis for integrating the arguments of those writers who have tended to concentrate their attention on one or other of the concepts in the Weberian framework. Marxist theories of ‘class’, American functionalist theories of ‘status’, and the more diverse writings of those concerned with the powers of ‘command’ can all find their place in the sociological toolbox. They provide essential and complementary analytical points of view on social stratification, and they allow us to understand why popular discourse has, from a sociological point of view, appeared confused. Popular discourse grasps the concrete interdependence of these elements of stratification in particular societies, but academic discourse must also attempt to isolate them in order to assess their relative salience in those societies.

This was the concern that lay behind the work of Weber. He developed his own ideas in order to clarify what he saw as the central developmental trend in Western societies, the development from medieval ‘status societies’ to modern ‘class societies’, each of which also involved distinctive patterns of authority. Indeed, the founding and pioneer sociologists of the nineteenth and early twentieth centuries were all concerned, in various ways, with this same transition. Those who lived through the transition began to develop a new language of analysis that could properly grasp the novel and distinctive features of modern patterns of social stratification. This contrast between ‘traditional’ and ‘modern’ systems of stratification lay at the core of both the academic and the popular discourses aimed at understanding the transition to modernity.

The language of ‘class’ has persisted for most of the twentieth century, though the recent sociological debates over the apparent ‘death of class’ have occurred because some have claimed that the language of ‘class’ – understood in Weber’s sense – may have lost its purchase on contemporary forms of stratification. The increasing reluctance of many people to employ the language and imagery of class to describe their own social situation, it has been held, may signal a fundamental social transformation. The very conditions that gave rise to modern forms of class stratification may, themselves, have given way to new and fundamentally different social circumstances. Some have described this as a transition from the conditions of modernity to those of post-modernity, a transition that matches in scale and significance that from medievalism to modernity (Bauman 1992). This suggestion raises critical questions about the direction of social change, and answers to these questions go well beyond my immediate concerns. I will suggest, however, that while claims concerning the death of class have been much exaggerated, there have, indeed, been important shifts in patterns of social stratification during the last fifty years.

Pre-modern hierarchies: the language of status

Beginning at least as early as the eleventh century, official and intellectual social thought depicted European societies as being strongly hierarchical. This social hierarchy was most typically described as comprising three estates: a religious estate of priests, a military and political estate of knights or lords and the ‘common’ estate of the ordinary people. With minor variations in terminology, this image of a tripartite hierarchy prevailed throughout the whole of medieval Europe (Mohl 1933; Duby 1978). Actual patterns of social stratification were, of course, more complex than this simple imagery suggested, and there was a general awareness that each of the estates was internally sub-divided. The clergy, for example, were differentiated by their position in the Church hierarchy into cardinals, abbots, priests, and so on. Similarly, the knightly estate was differentiated into various grades of peerage (duke, earl, viscount and baron) that were all distinguished from the ‘mere’ knights by their various roles in systems of royal administration. In some respects, these divisions cross-cut the official categories. The commons, for example, were widely seen as divided along the lines of wealth and status into a hierarchy that ran from the ‘rich’ through the mass of the commons to the ‘poor’. Peers and church leaders, by virtue of their wealth, would often be assimilated to the category of the rich. A subsidiary imagery, then, introduced fine distinctions within the overall social hierarchy. This imagery further defined common people by their agricultural function or their type of residence, and it allowed more nuanced identifications to be made in the everyday face-to-face contexts in which most people lived their lives. The official tripartite imagery was generally employed in public contexts and in legal documents, such as wills and leases, while the subsidiary imagery provided the terminology of day-to-day popular discourse. Whatever specific designations might be used, however, the social strata were seen in status terms as ‘estates’ characterised by specific privileges and life styles.

In England it was in the early modern period that this imagery and vocabulary began to alter (Wrightson 1991; see also Burke 1992). Agriculture had become more commercial and ‘capitalistic’ in orientation, and the growing importance of urban market centres had generated new social divisions that were more difficult to assimilate to the established tripartite model of society. While official and intellectual discourse continued to employ the language of ‘estates’, these came to be seen in a more complex and more differentiated way than before. In part, this involved an incorporation of the kinds of distinctions that had been made in the subsidiary popular imagery, but it also went beyond this. In addition to the clergy, the knights and the commons there were estates of merchants, lawyers, physicians, yeomen, schoolmasters and numerous other professional and occupational groups. Alongside these specialised groups were other recognised social categories, such as those of labourers, cottagers, servants and paupers. Behind this growing complexity of status distinctions was the growing significance of commercial activity, the growing visibility of new sources of economic division and inequality, and consequent shifts in collective identities. The early modern period, then, was characterised by a proliferation of categories that did not always fit into the traditional social hierarchy.

In the face of this growing social complexity, the language of ‘estates’ began to give way to a looser vocabulary of ‘orders’, ‘degrees’ or ‘ranks’ to reflect the more differentiated pattern of stratification that was emerging. This complexity was particularly marked at the upper levels of the social hierarchy, where these distinctions had become so complex that an official scale of status precedence for public occasions was codified by law in the mid sixteenth century. This was subsequently revised and updated on a number of occasions and, in its developed form, this official scale of ‘nobility’ and ‘gentry’ indicates clearly the many modifications that had been made to the traditional hierarchy of estates. Headed by the monarch and the royal family, the scale listed the varying ‘degrees’ of the peerage, the peerage itself being defined as a specific ‘rank’ of nobility. It showed the relative standing of those with official positions at Court and in the church, and of the sons of peers of various types. After these in the official scale of precedence came knights and, later on, baronets, followed by the commanders, members and officers of the various orders of knighthood. At the lowest level of the highest order were the ‘esquires’, a rank that possessed a ‘name of dignity’ that set them apart from mere ‘gentlemen’. Esquires were entitled to heraldic arms by virtue of the prestige of their occupations (as, for example, barristers or army officers) or their holding of public office. They were entitled to be addressed in writing as ‘Esq.’ rather than using the ‘Mr’ with which the ordinary gentlemen had to make do. These ‘gentlemen’ were a particularly important sign of social change. They were, in terms of the traditional classification, merely ‘commoners’, but their importance as landowners, farmers and merchants led them to be recognised by the Court and the nobility as a rather superior type of commoner. Indeed, many of them obtained official positions, leading to much confusion between ‘gentlemen’ and ‘esquires’ and to the awarding of titles of knighthood to many of them.

Outside the public and official sphere of the state and its concern for precedence, these new inequalities began to be grasped in a new vocabulary. This was particularly noticeable from the middle of the sixteenth century, and centred on the idea of a society divided into distinct ‘sorts’ of people. Economic inequalities associated with the expansion of capitalist agriculture and trade created social differences that ran counter to the traditional status distinctions. Market relations had not, of course, been absent from medieval society, but they had been relatively insignificant as sources of social stratification. With the growth of capitalism this was no longer the case, and the market achieved a much expanded role in the generation of propertied wealth. Conflicts and divisions rooted in economic differences of resources and market power were initially expressed in terms of an opposition between the ‘better sort’ and the ‘poorer sort’ of person (Wrightson 1991: 48). The better sort comprised the wealthy gentry and farmers who were dominant in the towns and parishes, while the poorer sort – alternatively described as the vulgar, common, meaner or ruder sort – were those who owned little or nothing in the way of resources and who had to support themselves through their own labour.

From the middle of the seventeenth century, a ‘middling sort’ of person was often identified. This middling sort consisted mainly of urban merchants, tradesmen and artisans, who were growing in numbers and wealth, though yeomen, tradesmen and freeholders in rural areas also came to be seen in the same way as a middling sort of people. The phrase was, most significantly, used to describe the manufacturers that were appearing in ever larger numbers during the seventeenth and eighteenth centuries. Thus, a popular social imagery of ‘better’, ‘poorer’ and ‘middling’ sorts was established, with the nobility, the gentry and paupers coming to be seen as mere elements or fractions of these larger categories. This popular imagery of ‘sorts’ led eventually to a recasting of the dominant imagery of stratification. This same terminology was taken to North America by English settlers and adapted to new circumstances. ‘Negro slaves’, for example, were added as an additional category at the bottom of the hierarchy (Main 1965).

By the eighteenth century, then, medieval status certainties had given way to a confusion of terminology in which competing discourses made themselves felt. The discourse of status was apparent in the widespread use of such terms as ‘estates’, ‘orders’, ‘degrees’ and ‘ranks’, while the language of ‘sorts’ reflected the growing significance of a more modern discourse of economic division. It was from this confusion that a new language of stratification was to emerge. This language of ‘class’ first appeared in the seventeenth and eighteenth centuries. It was taken up in the works of the political economists, and was eventually to prevail in intellectual and popular usage and to make itself felt in official discourse. This was not, however, a simple change in language. The new discourse of class emerged as an attempt to describe the very forces that had brought it into being. ‘Class’ was not a new term for old structures, but a term that identified the appearance of radically new forms of social division and collective action (Bauman 1982: 38). Thus, in the United States, class terminology developed first and most rapidly in the north. In the southern states, permeated by the experience of slavery, older styles of thought persisted for much longer. The language of class was a response to the new conditions of modernity that had been unleashed by capitalist development.

Modernity and the language of class

The Latin word classis first appeared in English during the sixteenth century, when it was used in historical writings to describe the economic and political differentiation of Roman citizens. It was not until the eighteenth century, however, that it was used to describe the contemporary social divisions of English society. This change in usage seems to have been inspired, in great part, by the successful scientific ‘classifications’ that had been produced in biology and in geology. The Latin word had been adapted by natural scientists and philosophers to refer to categories within theoretical schemes, and advances in the natural sciences had involved the development of highly refined classifications of, for example, animal species. Corresponding advances in social understanding seemed possible if human populations could also be ‘classified’ according to their social types. A social classification appeared, to the political economists, to be the essential requirement for the social investigation of the new economic forms of modern society (Calvert 1982: chapter 1). Classical political economy was underpinned by the individualistic social imagery of Hobbes and Locke (Macpherson 1962), but it began to employ such terms as rank, order and class to refer to categories of individuals with similar economic characteristics in the modern agrarian and industrial society. In the work of Adam Smith (1766), classes were seen as integral and interdependent elements in the economic structure of modern society, while Adam Ferguson (1767) was the first of these writers to employ the word ‘class’ within a systematic framework of historical analysis. A similar move occurred in France, where Quesnay and Turgot began to distinguish productive from unproductive classes. Unlike the words ‘rank’ and ‘order’, ‘class’ was unencumbered with connotations of feudal hierarchy and set an appropriately modern and scientific tone for economic analysis.

According to the classical economists, the three fundamental classes of capitalist society were landlords, capitalists and labourers, rooted respectively in landed rent, industrial capital and wage labour. Each class was seen as having a distinct position in the system of production. This theoretical position helped to popularise the ideas of class and class division. The term ‘class’ was taken up in liberal and socialist political discourse, where it was used to designate groups with a distinct social consciousness and involvement in political struggle: classes were seen as collective agents. A popular social imagery of ‘higher’, ‘middling’ and ‘lower’ classes replaced the earlier language of ‘sorts’, and this paved the way for the classic nineteenth-century imagery of the ‘upper’, ‘middle’ and ‘working’ classes (Corfield 1991: 123–6).

It was from the 1760s that the poorer, subordinate strata of the capitalist social order came to be described as the ‘industrious’ or ‘labouring’ classes and then, from around 1789, as the ‘working class’. This new terminology marked a shift from ‘lowliness’ in a status hierarchy to economic role in a system of production as the principal criterion of their social position. By the 1820s the term ‘working class’ was very widely used to describe an economic category defined by its dependence on manual work. Liberals and those who were opposed to socialist politics tended to see this working class merely as an aggregate of individuals who shared a similar economic position, while socialists saw the working class as a progressive political force and emphasised its unity and consciousness as a collective actor. These political differences were reflected in the language used, socialists preferring the singular term ‘working class’ and liberals preferring the plural term ‘working classes’. Collective class unity was emphasised more forcefully in France, where the singular classe ouvrière was far more widely used than was the plural classes ouvrières (Dubois 1962). The great strength of liberal individualism in Britain made many reluctant to use a language that suggested the idea of collective agency. This reluctance was manifest in a tendency to restrict the concept of class to the classification of individuals: ‘class’ was treated as an aggregate term rather than a group term. For this reason, British usage showed a greater preference for the plural term ‘working classes’ (Crossick 1991: 151, 156).

It is somewhat unclear when the phrase ‘middle class’ first came into use in written English. The term ‘middling classes’ seems to have appeared around 1748, and Mary Wollstonecraft (1792) was using the phrase ‘middle class’ just over 40 years later. It was certainly a widely recognised term by 1812, and the plural variant of ‘middle classes’ developed slightly later. The term was used to describe those property owners who could not rely simply on rentals and other propertied income but had to put their property to use and to work at producing an income from it. Capitalist entrepreneurs, for example, owned workshops and machinery, and they were actively occupied in ‘managing’ their capital and their employees. The middle classes, then, were distinguished from both the ‘leisured’ aristocracy and the working class (Seed 1992).

The language of ‘class’ spread across the whole of Europe. In France, for example, the transition to capitalist modernity was seen as a change from a society of états to a society of classes (Sewell 1981), while in Germany it was seen as a transition from one of Stände to one of Klassen. Capitalist development in Germany was much slower than in Britain, and the language of Stand persisted until late in the eighteenth century, when the particular privileges and occupations of the Adelsstand (nobility), Bauernstand (peasantry) and Handwerkstand (artisans) were recognised (Melton 1991: 134). Klasse began to appear in German usage in the 1830s as a way of contrasting a pauperised lower class (niedere Klasse) with the higher or educated classes. In Britain, however, a ‘class’ vocabulary was well established by the 1840s and, though it had not become the dominant language of stratification, it had begun to make inroads in official discourse.

The great diversity of occupations in the new industrial system meant that occupational designations acquired an importance as a means of social placement. Though occupations were now less likely to be defined as ‘estates’, differences in social standing were apparent in the usage of such terms as ‘profession’ to describe particularly privileged occupations. This lay behind the attempt of the Census Office to devise a classification of occupations that would reconcile occupational distinctions and class divisions. The 1841 Census made little attempt to classify occupations, but from 1851 the Census Office did attempt to bring some order to the analysis of occupations. It was not until the 1880s, however, that the language of class became a central characteristic of official discourse. An important factor behind this was the concentration of economic activity into larger financial and technical units and a corresponding centralisation of the whole national economy. As capital became more ‘organised’ economically, so labour took a more organised form in large national trade unions. The growing political assertiveness of organised labour was a major factor responsible for the wider recognition of the need to see social differences as divisions of class. This acceptance of class terminology is clear in the official occupational schema that was devised by the Registrar General for use in the 1911 Census. Reflecting the advances that had been made in the great social surveys undertaken by Booth and Rowntree, who had developed relatively sophisticated class schemata (Scott 1994a), the official class schema – the Registrar General’s Classification – grouped occupations together into larger social classes. This schema assumed that specific economic and housing resources were associated with each occupational category and that the relative social standing of each class was also reflected in its members’ incomes and culture (Marwick 1980: 62; Scott 1990a).

Official acceptance of the language of class is apparent in its increasing usage in parliamentary discussion and legislation. Acts of Parliament concerned with housing and welfare made reference to the ‘working class’, the ‘working classes’ or, on occasion still, the ‘labouring classes’. Legislation on welfare and housing rights, for example, treated the working classes as a group in need of specific legal status entitlements that need not be extended to those in the middle classes, who were presumed to be able to look after themselves. Most typically, the working classes were seen as those who were working in manual trades for a wage and were not employing others. Those on a weekly wage, rather than an annual salary, were seen as being in less secure employment, and perhaps as being subject to periods of unemployment and casual labour. The precise meaning of ‘working class’ was rarely spelled out, as it was assumed that the words were so widely used that their meaning could be taken for granted.

This combination of economic divisions with legal and social standing was central to the discourse of ‘class’. While this involved a valuable recognition that ‘class’ and ‘status’ were closely related in the modern social order, use of the single term ‘class’ to describe patterns of social stratification contained the seeds of later confusion. Crossick (1991: 154), for example, has shown that the equating of the landlords, capitalists and labourers of political economy with the upper, middle and lower classes of public discourse involved a conflation of distinct ideas that was to ‘bedevil conceptions of social order’ throughout the nineteenth century. Specifically, he sees it as conflating an economic conception of classes as positions in the system of production and the more traditional conception of estates as positions in a vertical hierarchy of superiority and inferiority. Nineteenth-century social thought, he argues, failed to clarify the relationship between these differing views of stratification. The confusion could only be resolved when writers such as Weber provided the appropriate analytical tools for recognising the interplay of class and status elements, along with their associated powers of command, in all systems of social stratification.

Usage of the word ‘class’, then, combined the two elements that Weber was to distinguish as ‘class’ and ‘status’. On the one hand, its root meaning had come to refer to the differences of economic power and resources that arose in modern, capitalist society and to the differences in life chances (health, mortality, income, and so on) that were associated with them. On the other hand, it also included the moral judgements that were attached to these differences and that clothed them in conceptions of social prestige. This is apparent not only in the official ‘class’ schema that was used in the Census, but also in the wider public discourse of stratification. Through the latter part of the nineteenth century and into the twentieth century, popular thought separated the ‘respectable’ from the ‘rough’ working class, the ‘deserving’ from the ‘undeserving’ poor, and the ‘respectable poor’ from the ‘residuum’ (Crossick 1991: 161–2; Joyce 1991: 57). Unskilled workers who were subject to casual employment or to protracted periods of unemployment became the objects of a moralising derogation of their life style and of their supposed unwillingness to adopt the ‘respectable’ life style of skilled workers. Those whose circumstances forced them into poverty were held, by middle class and respectable working class opinion, to be capable of helping themselves to rise above their circumstances, and those who did so were regarded as ‘deserving’ of help to attain proper respectability. Economic differences, then, were fused with moral judgements of status, and the language of ‘class’, as a result, became infused with moral connotations. ‘Class’ became a particularly sensitive matter for many in the upper and middle classes, for whom it was, like sex and religion, almost a taboo subject in polite conversation.

For those who were committed to a greater degree of equality, the moral dimension to the language of class made this language itself an obstacle to social change. The ‘class-ridden’ character of British society came to be seen as a matter of attitudes alone. ‘Snobbish’ moralising about the life styles of those regarded as social ‘inferiors’, and ‘envious’ criticism of those regarded as putting on airs of ‘superiority’, were increasingly seen as central to ‘class’. Only in the labour movement, and in the working class communities that sustained it, did a conception of class as rooted in economic power persist. For most others in British society, class was seen simply as a matter of accent, dress, education and manners.

The conflation of class and status in the popular discourse of ‘class’, then, is the reason why contemporary discussions of class have been so confused. Instead of using Weber’s concept to dissect the language of everyday life, many sociologists and commentators have simply taken over that language and have incorporated the same confusions into their own work. Mainstream American sociology, for example, equated class with status and simply lost sight of any real concern for structured social inequalities rooted in economic divisions. Equally, it lost sight of the linkages that exist between these divisions and the powers of command. ‘Stratification’ came to be seen as an exclusively normative matter of invidious status distinctions. The claim that America is an ‘open’ or ‘classless’ society rests as much on this conceptual blindspot as it does on any empirical evidence. This is not, however, to prejudge the issue of whether class – as defined by Weber – is still the most salient characteristic of social stratification in contemporary societies. Beginning with Nisbet (1959), the view has taken firm root that class is, indeed, dead.

A post-modern discourse of stratification?

The economic trends of the post-war period have given a particular thrust to the argument that ‘class’ as a source of economic division is no longer relevant. The growth of mass consumerism and a prolonged period of economic growth and relative affluence during the 1950s and 1960s have been seen as betokening the demise of the very economic conditions that gave birth to class relations in the rise of modernity. Most recently, this has been interpreted as a shift from a modern society of production to a ‘post-modern’ society of consumption (Bauman 1992). Where modern societies are, indeed, class societies, their successor societies, it is claimed, are societies of a fundamentally new kind. While they are not status societies in the sense of the traditional ‘estate’ societies of the feudal past, they do distinguish people on the basis of their life styles. In a post-modern society, social distinctions are based on consumer life styles and, as these are highly differentiated, they produce a fragmentary, kaleidoscopic differentiation of life styles rather than rigid and bounded social strata (Baudrillard 1981). The conception of post-modern society envisages, in a very real sense, not only the end of class but the end of stratification itself.

This has been stated in a particularly clear and radical way by Beck (1986: chapter 3), who argues that the conditions that some have described as ‘post-modern’ are, in fact, radicalised expressions of modernity that appear when the final vestiges of traditionalism have disappeared. The solidarities of social class, he argues, resulted from the fusion of economic differentiation with cultural conceptions of status. These status conceptions are no mere ‘traditional’ survivals, but are essential cultural supports of capitalist economic forms. As a result of rising living standards with the continued expansion of modernity, the communal bonds of communities of producers are dissolved and their subcultural distinctiveness disappears. The labour market is finally emancipated from status restrictions and exclusions, and occupational attainment becomes a matter simply of individual competition. People are motivated to acquire the educational credentials that will help them to achieve upward mobility (or to avoid downward mobility) in a competitive system that is completely individualised. As a result, orientations towards social inequalities involve an attitude of privatism that undermines collective identities of a class kind. Problems of attainment in a competitive market are regarded as individual failings that have to be explained in psychological terms. Class and class politics are dissolved, and individualised life styles are diversified. The society of late modernity is an inegalitarian society without social stratification. It is ‘a capitalism without classes, but with individualized social inequality and all the related social and political problems’ (1986: 88).

In these circumstances, the experience and the language of class no longer seem relevant to people, and Beck holds that they are more likely to acquire their identities from ‘ascribed’ differences and personal relations than they are from the sphere of work and employment. Differences of ethnicity, gender and age become important sources of social identity, as do such matters as sex and disability. New social movements arise around these identities and around the growing perception of environmental ‘risk’, and these social movements replace those based around class. Concerns over ‘risk’ indicate, for Beck, the fact that the era of late modernity is one in transition towards a ‘risk society’ in which inequalities will be rooted in ‘risk positions’ rather than class or status positions. His wider speculation goes beyond my immediate concerns, but Beck’s central claims about the demise of class are at the heart of current debates (see also Eder 1993).