The Class Structure of Capitalist Societies

Volume 1: A Space of Bounded Variety

Will Atkinson

• Publisher: Routledge
• Year: 2020
• ISBN:
  9780429800870
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Political Ideologies
• Index: Politics & International Relations

This first volume of The Class Structure of Capitalist Societies offers a bold and wide-ranging assessment of the shape and effects of class systems across a diverse range of capitalist nations. Plumbing a trove of data and deploying cutting-edge techniques, it carefully maps the distribution of the key sources of power and documents the major convergences and divergences between market societies old and new.

Establishing that the multidimensional vision of class proposed decades ago by Pierre Bourdieu appears to hold good throughout Europe, parts of the wider Western world and Eastern Asia, the book goes on to examine a number of significant themes: the relationship between class and occupation; the intersection of class with gender, religion, geography and age; the correspondences between social position and political attitudes; self-positioning in the class structure; and the extent of belief in meritocracy. For all the striking cross-national commonalities, however, the book unearths consistent variations seemingly linked to distinct politico-economic regimes.

This title will appeal to scholars and advanced undergraduate and postgraduate students interested in sociology, politics and demography, and is essential reading for all those interested in social class across the globe.

Chapter 3 of this book is freely available as a downloadable Open Access PDF at http://www.taylorfrancis.com under a Creative Commons Attribution-Non Commercial-No Derivatives (CC-BY-NC-ND) 4.0 license.

Chapter 1

Introduction

The degree to which the structure and effects of class are ‘the same’ or ‘different’ across capitalist nations has occupied scholarly minds for centuries. Witness Marx and Engels’s eager dissections of the events that exploded across Europe and America in the nineteenth century, with all their considerations of the size and power of this or that class, or Max Weber ranging over East and West in his meticulous documentation of the relationship between class and religion. Towards the latter part of the twentieth century, moreover, it seemed a relatively easy question to answer. Most assumed it to be a case of determining how many people fall into separate pre-defined ‘slots’ in the capitalist production process or occupational structure – a case of jobs and industries, in other words. Perhaps the nature of the economy in one nation, thanks to its position within the global division of labour, or the ‘world-system’, means it has a larger and more diversified body of capitalists than others, or specific industrial specialisations like agriculture, finance, hi-tech manufacturing and so on swell or shrink the ranks of various socio-professional categories. If one has the data available, and a preferred means of assigning people to boxes, then all that needs to be done is to work out how many people fit into each box. From there follows scrutiny of the socio-political consequences – political polarisation following economic polarisation, for example, or liberalism flourishing with the expansion of the educated ‘middle class’. Others less keen on the logic and language of ‘class’, whether on methodological or political grounds, but still covering the same intellectual ground, might take a slightly different tack, reducing ‘stratification’ down to a universal scale of socio-economic standing whose components – relative income, educational attainment, occupational prestige, etc. – are always so closely connected that capitalist social orders, or ‘industrial societies’ as they preferred to call them, can, for practical purposes, be defined by a single dimension of difference.

Forty years ago, however, a different way of seeing things emerged. A fat tome summarising a decade or more of research and reflection on French society fell from the printing press. The book was Distinction, and though it was a collaborative effort, implicating the labours of many colleagues at the Parisian research centre from which it sprang, it carried the name of just one man, its self-styled ‘stage-manager’ (le metteur en scene): Pierre Bourdieu.¹ At its core, underpinning the sundry assertions and asides with which it brimmed, it proffered a novel, distinctive vision of social class. Occupation was not the backbone, as Frank Parkin (1972: 18) had famously summed up the prevailing orthodoxy of the day, but simply a proxy for possession of other properties. If among those properties were economic resources, moreover, then these were reducible to neither wealth disparities, as those of a radical bent might have it, or income differentials, as the advocates of scales would have it, but incorporated both. Not only that, but there were other resources the dispossession of which rendered suffering and constraint. These included the cachet of connections and club membership but also, and especially, mastery of the cultural or symbolic codes deemed legitimate within the social order and practically approximated, but not exhausted, by education level. Yet those various resources, or ‘capitals’, were not directly correlated, as generations of research on social stratification claimed. They constituted sources of polarisation and antagonism, distributing people within a multidimensional ‘social space’ on the basis of not just their overall stocks of capital but the degree to which either cultural or economic capital form the bulk of those stocks. Ultimately, this novel reimaging of class was rooted in the founding premise that the concept made famous by Karl Marx and Max Weber was not, at base, about exploitation or life chances, even if they are rolled up in it, but about misrecognition. That is to say, certain arbitrary properties – money and what it buys, symbolic mastery and its associated goods and practices – and their possessors are seen as inherently worthy, legitimate, authoritative and so on. Having those properties or not thus become the fundamental shapers of conditions of existence and, through the experiences and opportunities attending them, of tastes, desires and ethos – or, to use Bourdieu’s celebrated term, habitus – but also of one’s sense of place and value relative to others.

Surprisingly, however, research to test whether and to what extent this multipolar image of the class structure applies beyond late twentieth-century France is patchy to say the least. There has been some work in Europe, most notably the Nordic countries, but, despite the frequent reference to the concept among those inspired by Bourdieu (e.g. Wacquant, 2009), little has been done elsewhere to scrutinise empirically whether class takes the form of a social space, whether its axes are the same as those found in Bourdieu’s home country and whether it translates into cross-cutting divergences of ethos and distributions of self-worth. Most scholars engaging with Bourdieu on class tend to ignore or dismiss the notion, instead showing interest only in testing his supposed thesis that there is a one-to-one correspondence between class and lifestyles using their own ill-fitting measures of class or stratification and suboptimal analytical techniques. At the same time this dearth means we have no way of knowing just how representative of capitalist nations those few European studies – which have generally supported the Bourdieusian model – actually are, especially as most of them have been conducted in nations with very particular political economies and ‘welfare regimes’, to use Esping-Andersen’s (1990) terms. Would the same conclusions be reached in starkly inegalitarian nations like the US, rapidly industrialised and culturally distinct East Asia, or the nations of Eastern Europe transitioning to capitalism only 30 years ago?

The intention in this book is to go some way towards providing a fuller picture by assessing, as best as possible with the data available, the broad generalisability of Bourdieu’s image of social space, its homologies and its consequences for self-perception to capitalist nations while flagging potential sources of country-specific deviation. Just as advocates of previous images of class, once they had worked out an empirically verifiable proxy model in one nation, generally saw the next logical step to be comparison across a wider range of countries, so too it is now time to take Bourdieu global. Matching philosophical principles and operational procedures, the methodological foundation for all that follows will be the specific form of data analysis known as geometric data analysis, but it will be used – for the first time – for widescale comparative analysis. First, however, there is a need to unpack the context, concepts and outstanding questions at stake.

From scales to classes

Human thought tends to organise around all kinds of oppositions – this we know from Hegel and Durkheim to Bachelard – so it is no surprise that research on class and stratification, and the countless suggestions on how to depict them, has too. A double debate has, in fact, run through Western scholarship – and ‘folk’ thought too (Cannadine, 1998) – on these age-old concerns. On the one hand, are power and inequality best understood as being distributed between large, clearly defined classes or as being arranged in a scalar fashion, with no clear breaks? On the other hand, is there one overriding property that defines the great chain of social being, or a multiplicity of properties which either sit alongside or co-define class? For Marx the overall answer was simple: there are two classes defined by a single relation, namely, ownership or not of the means of production, and thus whether one is an exploiter or exploited. For Max Weber (1968) the distribution of power within a social order also ran between classes and groups, but he injected multidimensional thinking into the sociology of domination by distinguishing class, defined by property and market situations, from status, or esteem and prestige.

After the classics, into the mid-twentieth century, came the scales. Pioneered in the US, where talk of ‘class’ had too much of a ring of the Old World left behind to it and smacked of an unwelcome political radicalism, occupational prestige, as ranked by the general public, was the initial focal constituent. Income and education level were subsequently held to be efficient joint predictors, and thus indicators, of status, since they correlated so strongly with it, leading Otis Dudley Duncan (1961) to devise the first true index of socio-economic status usable across studies. Rolling prestige, education and income together in a single measure like this jarred with the calls from others, echoing Weber, to recognise the distinct existence of several scales in society – of income, authority and prestige, the latter being based on education as well as other factors – since occupying discrepant positions in each can generates all kinds of psychological effects manifest not least in deviant political attitudes (Lipset and Bendix, 1959; Lipset, 1960; Lenski, 1966). Updated socio-economic indexes constructed by Trieman (1977) and Ganzeboom et al. (1992) nevertheless continued the effort to collapse multiple facets of inequality into a single scalar measure, until Hauser and Warren (1997) concluded in their comprehensive appraisal of scales that the failure to decompose the effects of education and earnings renders them analytically deficient and thus, in their words, ‘scientifically obsolete’ (see also Bukodi et al., 2011).

Others, meanwhile, held fast to the principle that distinguishing bounded classes on the basis of set criteria revealed more than scales ever could. Some classifications were multidimensional, like Erik Olin Wright’s (1997) Marxist categorisation of jobs on the basis of property ownership, skill and authority, and some tried not to over-homogenise by homing in on ‘micro-classes’, like David Grusky and his colleagues (see, e.g. Grusky and Galescu, 2005; Grusky and Weeden, 2001). The leading measure of class in sociology today, however, is the neo-Weberian Erikson-Goldthorpe-Portocarero (EGP) scheme. First devised by John Goldthorpe (1980), who had once flirted with prestige scales (Goldthorpe and Hope, 1974), and honed over the next 30 years of research into social mobility and educational inequality (Erikson and Goldthorpe, 1992; Goldthorpe, 2007a), the EGP schema distinguishes classes on the basis of property possession and employment relations, the latter of which are determined by the ‘difficulty of monitoring’ and the ‘specificity of human capital’ associated with different jobs. Effectively, however, the EGP scheme is unidimensional, since its whole premise is to produce a hierarchy of boxes stacked in order of life chances – whether they be in relation to health, educational attainment or social mobility. The explanatory limitations of this schema soon revealed themselves, therefore: in explaining educational inequalities, it transpired that parental education was a crucial factor, not just bare economic constraint or enablement (Bukodi and Goldthorpe, 2018); when it came to lifestyles, it was plain that differences in education level played a fundamental role – much more so than class (Chan and Goldthorpe, 2007a, 2007b, 2010a); and Goldthorpe’s (1982, 1995) repeated prediction that the ‘service class’ or ‘salariat’ at the top of the stack should be conservative in political orientation was frequently belied by research that showed it to be fractured along educational lines, with the highly educated professions being more left-liberal in their outlook and the well-heeled but not so well-educated managers being somewhat more conservative (Evans, 1999; Houtman, 2003; see also debates in Clark and Lipset, 2001). On top of that it seems to jar disconcertingly with practical perceptions of social – and even occupational – difference (Filhon et al., 2016).

Some have suggested these limitations spring from the EGP scheme’s ill-fit with post-industrial economies where the service sector and educational levels have mushroomed dramatically (Esping-Andersen, 1993; Oesch, 2006), and some have even tried to tweak the EGP scheme to make space for intra-class divisions (Güveli et al., 2007; Kriesi et al., 2008), watering down the original logic of construction in the process. Goldthorpe’s response, however, has been to return to the prestige scales he had once rejected and put the differences down to status, defined by patterns of differential association between occupational groups, though in most cases the evidence suggested the key is actually education level (Chan and Goldthorpe, 2004, 2007a, 2007b, 2010a). The goal, it seems, was to defend what he dubbed a ‘narrow’ definition of class which could be played off of other apparently autonomous forms of stratification in multivariate analysis (cited in Lareau, 2008: 11–12). Perhaps the intention was to stay true to the Weberian spirit, but the definition of status is somewhat non-Weberian (see Flemmen et al., 2019). Why did Chan and Goldthorpe not just directly define it in terms of lifestyle practices, rather than try and find an underpinning causal mechanism (or predictive variable) in the occupational structure, since it is lifestyle that is the causal m...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Table of Contents
• List of figures
• List of tables
• Acknowledgements
• 1 Introduction
• 2 Constructing the spaces
• 3 Social spaces
• 4 The division of labour of domination
• 5 Homologies
• 6 Trajectories
• 7 Political position-takings
• 8 Class sense and symbolic violence
• 9 Conclusion
• References
• Index