There have been countless efforts to test and ‘update’ Pierre Bourdieu's thesis that there is a correspondence between the space of social positions and the space of lifestyles. The best known of these targeting the UK are the Cultural Capital and Social Exclusion project and, more recently, the Great British Class Survey, but their conceptual and methodological limitations mean their findings are questionable and hinder closer investigation of an oft-sidelined piece of the puzzle one of the projects specifically highlighted: the significance of gender in structuring taste. Drawing on the 2012 wave of the British Cohort Study, which included a battery of questions on cultural consumption, and deploying a logic and measure of class closer to Bourdieu's own, I thus seek to offer an alternative examination of not only the nature and degree of correspondence between the social space and lifestyles but its entwinement with masculinity and femininity.

In the preface to the English-language edition of Distinction, that weighty tome painstakingly documenting the correspondence between class and lifestyles in 1960s and 1970s France, Pierre Bourdieu extended an invitation to scholars outside his country to ‘join in the game’ and ‘search for equivalents’ in their own nations (Bourdieu 1984: xii). Eventually they took up the call in droves, surely making the thesis presented in Distinction one of the most tested, modified and qualified in recent cultural sociology, if not, indeed, in sociology more widely. Not all of them have been sympathetic by any means, and some have even taken it upon themselves to actively try and disprove Bourdieu's position, either on the grounds that the particular cultural forms corresponding with class positions are no longer those identified by Bourdieu or on the basis that lifestyles are not associated with class at all. Most of the time, however, these claims rest on conceptions of culture, class, methodology, the nature of theory and ultimately the sociological craft quite different from Bourdieu's own, and studies deploying a logic and technique closer to that of Distinction tend to be more favourable. From Denmark to Mexico, Norway to Serbia, the particular practices may well be different in tune with national historical development, but the general principles of differentiation are remarkably consistent (Cveticanin 2012; Coulangeon and Duval 2014).

Research on the UK occupies a peculiar position. Notwithstanding some who are plainly sceptical of Bourdieu's work (Chan and Goldthorpe 2007), a strand of scholarship has claimed to follow Bourdieu's approach closely and uncover some confirmations and some divergences across the Channel (Bennett et al. 2009; Savage et al. 2013). In actual fact, however, similarities of statistical technique and focus obscure very real differences of logic and methodology, meaning a suitable effort to search for contemporary British equivalents to the cultural practices of France during Les Trente Glorieuses is still somewhat lacking. What makes this all the more unsatisfactory is that this line of research – along with many feminists in Britain – has rightly tried to emphasise the intersection of class with gender in producing symbolic struggle and domination more than others and, indeed, more overtly than Bourdieu himself. An important issue, therefore, is simply not done justice. In what follows, as part of a wider project of clarifying, applying and adding to Bourdieu's analyses for the UK today (see Atkinson 2017), I seek to remedy the situation. Drawing on the latest wave of the British Cohort Study (BCS), remaining close to the logic and methodology of Bourdieu's sociology and deploying a specially designed measure of class to approximate the fault lines of the British social space, I offer an analysis of the correspondence between class and culture sensitive to the complicating role of gender. First, however, let me elaborate on the current state of affairs.

Bourdieu's (1984) thesis on the relationship between class and lifestyles is often summarised in terms of ‘homology’, but a trove of nuances specify exactly what that means. First of all, class is not a simple binary relation or vertical stack of categories, as the dominant traditions of thought have it, but a multidimensional social space defined by possession of economic capital (money, wealth), cultural capital (symbolic mastery, measured by education level) and social capital (connections and networks). In a later work Bourdieu (2005) also broached the notion of ‘technical capital’, or particular forms of practical and technical mastery, usually measurable through vocational qualifications, securing limited and localised recognition. In any case the primary dimension of the space is volume of capital in all its forms, distinguishing a dominant and a dominated class and an intermediate class in between. The second dimension is capital composition, separating out the fractions of each class richer in economic capital (e.g. business owners and private sector executives) from those richer in cultural capital (e.g. teachers and intellectuals), again with more ‘balanced’ fractions nestling in the middle. The third dimension is time, usually taken to encompass the rise and fall of specific class fractions as well as individual social mobility.

Capital possession generates certain conditions of existence – specifically, greater or lesser distance from material necessity – and the experiences and possibilities that go with them generate distinct class habitus, manifest in tastes for certain forms of music, film, sports and so on as well as inclination to visit certain venues and events. In fact these cultural goods and activities map into a space of their own, the ‘symbolic space’ or space of lifestyles, with each element being defined against all others within the system – as rare or common, distinct or vulgar and so on – and, if one projects this onto the social space, one sees a clear correspondence between certain practices and certain class fractions, or a homology. The practices and tastes deemed common and vulgar are associated with the dominated class, their proximity to necessity inclining them towards the economical, the practical and the accessible, while the practices considered rare and distinct – whether because they are economically exclusive (i.e. expensive) or culturally exclusive (i.e. dependent on mastery of specific symbolic systems) – are associated with the dominant class. In fact it is only by virtue of their possession of capital that the latter are misrecognised as bearers and definers of ‘legitimate’ culture and the dominated cast as tasteless, uncouth and common in opposition.

So many studies have been carried out to test this vision of class and culture, in so many countries across the globe, that it is impossible to summarise them all. There are, however, a few broad lines of development, the first being the emergence of the ‘omnivore thesis’ in the early 1990s. Originating in the US and kick-started by the late Richard Peterson (1992, 1997), the idea was that dominant, legitimate culture, and therefore cultural capital, is no longer premised on exclusive, obscure cultural forms or knowledge as opposed to popular culture, but on engagement, breadth, variety and openness as opposed to disengagement or narrow interests. In the UK the most high-profile defenders of this thesis are probably Chan and Goldthorpe (2007), even if, good Weberians that they are, they insist it is a phenomenon of status rather than class. After a period of being unearthed just about everywhere, however, the existence of the cultural omnivore is no longer quite as indubitable as it was. Scrupulous researchers began to realise it may well be little more than a statistical artefact – a product of extraordinarily broad ‘genre’ categories, measures of class rooted in wholly different theoretical assumptions to Bourdieu's and particular statistical techniques (Holt 1997; Wuggenig 2007; Atkinson 2011). Instead, using Bourdieu's own favoured technique of multiple correspondence analysis (MCA), minimising reliance on genre categories and remaining faithful to his multidimensional and relational view of class, a recent wave of studies, led by Prieur et al. (2008) and Rosenlund (2009), has begun to confirm and update Bourdieu's original model (see Prieur and Savage 2011, 2013). Practices have certainly changed with the times, they readily admit, but omnivorousness is rather less apparent and the capital composition principle, generally ignored in the pro-omnivore research, is found alive and well.

In the UK this latter current of scholarship could be said to be represented by the Cultural Capital and Social Exclusion (CCSE) project undertaken in 2003 (Bennett et al. 2009) – at least that is the impression Prieur and Savage (2011, 2013) have tried to give. It did, after all, seek to construct the space of lifestyles using MCA, and omnivorousness played a relatively minor role. The conclusions were, however, quite different from Bourdieu's. For one thing, the major principle of cultural difference corresponding with class was deemed to be that between those who frequently engage in all sorts of cultural activities – from going to rock concerts to attending the opera – and those who are somewhat more disengaged, while the capital composition principle barely figured at all. Yet these divergences were more the result of procedural shortcomings than social change or national specificity. An overreliance on genre categories and measures of frequency of cultural activity; a dearth of survey questions tapping into the tastes of the dominated; a lack of indicators designed to bring out the tastes of the economically rich; and dependence on a slightly modified version of the official UK Office for National Statistics Socio-Economic Classification (NS-SEC), ultimately derived from the Weberian Erikson-Goldthorpe-Portacarero (EGP) scheme, as a measure of ‘class’, even though it is ill-equipped to pick up homologies with capital composition – all of these problems diluted any claim that the CCSE project constituted a robust test of Bourdieu's thesis in the UK, and careful re-analysis of their data has shown Britain may be more like the France of Distinction than the team claimed (see further Atkinson 2011, 2017; Atkinson and Rosenlund 2014).

The CCSE dataset is now also ageing. True enough, when it comes to studying classed aesthetics (rather than consumption of particular cultural products), 10- or 15-year-old datasets are certainly tolerable, but there are newer data out there. Indeed, many might say the CCSE project has now been superseded by the Great British Class Survey (GBCS), published to great fanfare a few years ago, since it purported to do a similar thing and was led by a key member of the original CCSE team (Savage et al. 2013). Unfortunately, however, the GBCS repeated the same basic errors of the CCSE project while adding some new ones, the most egregious of which is the decision to define cultural capital, and class fractions, directly in terms of lifestyle practices rather than indicators of symbolic mastery such as education level. Collapsing together the social space and the space of lifestyles – the field and the space of position-takings, in other words, or a social structure and a symbolic structure – ignores the relative autonomy of the two spaces, makes it impossible to assess their degree of homology (an empirical question, after all) and effectively neuters genetic analysis (Mills 2014; Atkinson 2017).

The GBCS also seemed to backtrack on one of the CCSE project's more interesting insights: that gender played a key role in differentiating taste. Adding to the research already consistently showing that women tend to consume ‘highbrow’ culture at a greater rate than men (Lizardo 2006; Purhonen et al. 2011; Christin 2012), the latter, Bennett et al. (2009) found, tend to favour practices and activities which are, as they put it, ‘outwardly oriented’, like sports, documentaries and current affairs programmes, while women are inclined towards practices the researchers describe as more ‘inwardly oriented’, like romance novels, television soaps and drama programmes. Of course this raises the question of how class and gender interact in the production of taste and symbolic domination – a question Bourdieu has been accused of sidelining or, at the very least, answering badly (see Adkins and Skeggs 2004). If gender forms a distinct axis of the symbolic space, could it not form a separate axis of the social space too, defined by a specific ‘gender capital’ (McCall 1992)? Or might gender even form a field of its own, with relative autonomy from class (Coles 2007)? More concretely, could it be that the increasing feminisation of work, or at least of certain fields, in the twentieth century and beyond has transformed or blurred gendered dispositions (McNay 1999; Adkins 2002)?

In fact Bourdieu's view in Distinction was even more ‘intersectional’ that his feminist critics tend to suppose: the social space is highly gendered, with some class fractions being (increasingly) more female-dominated than others, and this will shape the objective and perceptual association of social positions with lifestyle practices and the way in which they are denigrated or valorised. Specifically, the pole of social space richer in cultural than economic capital tends to be more feminised, though the fractions of the dominant class richest in cultural capital are the most ‘androgynous’ in their tastes as women and men positioned there are most likely to challenge their respective gender stereotypes (Bourdieu 1984: 382–3; Rosenlund 2015). This does not, as some critics might fear (e.g. Anthias 2001), collapse gender into class, however. As Bourdieu (2001) argued in his later work, the male/female binary is a scheme of perception shaped by and shaping the positioning of those labelled ‘men’ and ‘women’ in a multitude of fields, including the social space, as well as differential participation and strength of libidinal investment in certain fields. The latter – to elaborate a little – include the fields of art, politics, business and so on, but also the familial field as a site of struggle for love (Bourdieu 1998, 2000), the sexual field as a domain in which individuals struggle to define and embody the legitimate definition of attractiveness (George 2014), and certain fields in which the physical capital of force – of being tough or intimidating, which can also be channelled into sporting prowess – is at stake, the last of these tending to be monopolised by men (Atkinson 2016a).

Studies of symbolic spaces inspired by Bourdieu have not always done the intersection with gender justice. Cveticanin (2012) and his colleagues, for example, expend just one line on its relationship to cultural consumption in Serbia, principally because it seems very muted in their statistical model, while both Rosenlund (2009) and Prieur et al. (2008) noted the patently gendered nature of the social space in Norway and Denmark respectively but opted not to examine how gender complicates or interacts with class fractions in shaping lifestyles. In Coulangeon and Duval's (2014) collection, moreover, the contributions seeking to map symbolic spaces à la Distinction either side-line gender or, like the CCSE team, distinguish it as a third separate axis, sometimes based on a fact-versus-fiction (or outer-versus-inner) dichotomy of taste, without exploring how femininity and masculinity differentially run through, define and are defined by specific class fractions.1 In the case of the CCSE team this is exacerbated by the measure of class deployed: because the EGP scheme makes no space for capital composition, it could not tap into the gendered character of class fractions. In fact, the re-analysis of their data using different measures already referred to above found instead that gender was inextricably bound up with the capital composition principle in the dispersion of lifestyles, mirroring the social space (Atkinson 2017), while cautious analysis of British literary tastes using relatively disaggregated genre categories confirmed Bourdieu's finding that specifically gendered tastes become less pronounced the higher up and further towards the cultural pole of social space one goes (Atkinson 2016b). The question remains, however, as to whether these findings will still stand with newer and broader-ranging data on lifestyles in the UK.

To test, or rather re-test, the applicability of Bourdieu's thesis on class, gender and lifestyles to twenty-first-century Britain, I draw on the latest wave (2012) of the BCS, a longitudinal panel survey following 17,000 people born in 1970.2 Included in the survey was a self-completion component, filled in by 8734 respondents, containing a whole raft of questions relating to cultural consumption and participation. Tastes in literature, newspapers, television programmes and sports as well as frequency of visiting a variety of cultural and entertainment establishments or events were all covered. Our attention will zone in on the propensity to frequent those establishments and events, as well as tastes in sport and television programmes – literary taste has been analysed separately elsewhere. This gives us a total of 43 variables to work with. Of course the data is not without its limits: we get no insight into the mode of consumption of the different practices, for example, and taste in television programmes is, unfortunately, clumped into genres, though these are at least relatively discriminating. There is also something of a bias towards measuring practices and events which one would associate with possession of cultural capital (museums, art galleries, opera, etc.) – a telling assumption about what counts as ‘legitimate’ culture in the minds of the researchers who designed the survey. More popular pastimes are also included, but the practices that might be conjectured to correspond with the economically rich (boat ownership, luxury holidays, car value, etc.) are somewhat underrepresented, potentially effecting the capacity to bring out the capital composition principle as fully as one might like. Nevertheless, the hypothesis is that the correspondence between class and culture is strong enough that it will still emerge in analysis even with these limitations of the data. We may not get an insight into the different ways in which reality television or a soap opera can be interpreted and used (see Atkinson 2017), but we should reap a broad picture of the statistical association of certain types of people with watching certain types of programme, or taking part in certain activities and sports, underpinning everyday perception and judgement.

The BCS, being a cohort study, automatically removes physical age from analysis: all respondents were aged 42 at the time of completing the questionnaire. Differentiation of so-called youth culture and traditional culture of the kind often found in Bourdieu-influenced studies of cultural consumption nowadays is thus bracketed out (for how it figures, compare Atkinson 2017). The same cannot be said, however, for the effects of what Bourdieu (1984) called social age, or the degree to which class fractions, no matter the physical ages of their members, are new or old, emerging or established, ascending or declining, and thus subversive or conservative. Forty-two is, moreover, a particularly apt age for analysis. Trajectories through the social space, class habitus and cultural tastes will be well established, plus the majority of respondents (73 per cent) have children of varying ages living at home, yielding food for thought on the effects of any patterns found for everyday, familial, domestic experience and, with it, social reproduction.

The primary technique employed to chart the relationship between class and culture will be MCA, the method made famous by Bourdieu and rapidly becoming established among Bourdieusian quantitative researchers nowadays. Specifically, MCA will be deployed to construct the space of lifestyles using 35 of the variables from the self-completion package of the BCS and explicative variables projected into the resulting model as supplementary modalities to check their homology.3 A little transformation of the sport and cultural activities variables has been necessary to facilitate this, however. In the data they are recorded in terms of frequency of participation/attendance. This is not ideal on conceptual grounds, but neither is it ideal on technical grounds since it tends to produce models in which the key axes are not fully independent: the first axis opposes frequent activity against infrequent activity, the second moderate activity versus frequent and infrequent activity. Of course a line needs to be drawn to ensure we are tapping into a durable taste – we need to distinguish those who go to art galleries because they like going to art galleries from those who might pop in one day out of curiosity, or for something to do with children on a rainy day. The decision taken here, therefore, has been to simply focus on whether the practice is done at all or not, unless over 50 per cent of the sample ever do it in which case a criterion of regularity has been applied – at least once or twice a month for some practices, at least once a week for others, depending on the nature of the practice. Going to the theatre or art galleries several times a year might qualify one as a ‘regular’, but to be counted as such when going to the pub or dining out, or to distinguish those who do DIY out of necessity and those who do it because they have a taste for it, a rather more stringent yardstick needs to be applied. This means these variables have been converted into binaries, matching the variables recording television tastes which are logged in the dataset as simply genres ‘mentioned’ or ‘not mentioned’ as watched.

To confirm or confute the homology with the social space, various available measures of capital possession can be used. In order to overcome some of the limitations of previous research, however, and as part of a broader endeavour to facilitate consistent statistical analysis across datasets from a Bourdieusian point of view, a class scheme has been designed to approximate maximum similarities and differences in not only capital volume but capital composition in the British social space. The scheme is an aggregation of the unit-level Standard Occupational Classification codes found in government and academic datasets in the UK – the same code which underpins the NS-SEC. Occupations were grouped together to form analytical classes – classes ‘on paper’, as Bourdieu would say – on the basis of an examination of capital profiles in successive government Labour Force Surveys. The method of construction, and corroboration of the scheme's criterion validity using MCA, is detailed elsewhere (Atkinson 2017; Atkinson and Rosenlund 2014). Its structure and some basic indicators of capital drawn from the BCS are presented in Figure 1 and Table 1. Crucially, the left-hand side of the social space, containing the class fractions richer in cultural than economic capital, is highly feminised, the more so the lower one goes down in social space, while the fractions rich in economic capital (as well as technical capital) are the least feminised.
Figure 1.

The British social space.

Figure 1.

The British social space.

Close modal
Table 1.
Characterisation of the class fractions (%).
Class fractionSavings and investments over £10kPost-compulsory educationGraduatesRate of feminisationExamples
Business executives 55 59 43 27 Managers and CEOs in finance and manufacturing 
Professions 59 83 76 36 Lawyers, doctors, architects 
White-collar workers 42 58 41 45 Software developers, surveyors, human relations officers 
Cultural dominant 34 85 73 68 Intellectuals, cultural producers, teachers 
Lower managers and proprietors (LMPs) 38 40 22 40 Hotel managers, shopkeepers 
Technicians 33 46 29 24 Engineers, laboratory technicians 
Administrators 30 35 16 81 Secretaries, bookkeepers, clerks 
Cultural intermediaries 28 66 38 75 Nurses, therapists, youth workers 
Skilled trades 23 15 Mechanics, electricians, plumbers 
Sales workers 15 18 77 Retail assistants, cashiers 
Manual workers 13 11 34 Cleaners, assembly-line workers 
Caring services 19 25 89 Hairdressers, dental nurses, travel agents 
Class fractionSavings and investments over £10kPost-compulsory educationGraduatesRate of feminisationExamples
Business executives 55 59 43 27 Managers and CEOs in finance and manufacturing 
Professions 59 83 76 36 Lawyers, doctors, architects 
White-collar workers 42 58 41 45 Software developers, surveyors, human relations officers 
Cultural dominant 34 85 73 68 Intellectuals, cultural producers, teachers 
Lower managers and proprietors (LMPs) 38 40 22 40 Hotel managers, shopkeepers 
Technicians 33 46 29 24 Engineers, laboratory technicians 
Administrators 30 35 16 81 Secretaries, bookkeepers, clerks 
Cultural intermediaries 28 66 38 75 Nurses, therapists, youth workers 
Skilled trades 23 15 Mechanics, electricians, plumbers 
Sales workers 15 18 77 Retail assistants, cashiers 
Manual workers 13 11 34 Cleaners, assembly-line workers 
Caring services 19 25 89 Hairdressers, dental nurses, travel agents 

Source: British Cohort Study 2012.

Such are the preliminaries; now for the analysis. Axis 1 of the MCA accounts for a considerable 64 per cent of the total variance and opposes, on the one hand, taste for almost all cultural activities/venues – the exceptions are DIY, going to the pub and gardening – as well as participation in all sports to, on the other hand, non-participation in a number of these activities, but also watching soaps (Figure 2, Tables 2 and 3). Though they do not contribute above average to the axis, reality television and general entertainment shows nestle at this latter pole too. To those erroneously misreading the legitimated definition of culture as culture per se this axis might be construed in terms of ‘engagement’ versus ‘disengagement’, but given the nature of the constructed variables it can just as easily be interpreted as an opposition between the exclusive – whether by dint of education, as with the cultural venues, or money, as in the case of many of the sports and regular participation in activities with an economic price – and the common and accessible, or public versus home-based leisure. The common and accessible may be defined largely in terms of television tastes, moreover, but it is also interesting to examine the dispersion of the activities along the axis: the most demanding of cultural capital and, to a degree, economic capital are furthest towards the exclusive pole – bookclubs, art galleries, writing, performing arts, photography (not so middlebrow nowadays, obviously), theatre and museums in relation to cultural capital; skiing, watersports, racquet sports, horse riding and regular cinema attendance regarding economic capital. The closer towards the origin one moves, the more accessible the activities might be said to become – textiles/craft work, going to libraries and dancing (which is less demanding of symbolic mastery) vis-à-vis cultural capital; eating out, gym going, DIY and pub visits in relation to economic capital. Those individuals positioned lower down in the symbolic space, then, while more likely to decline from any activity included in the analysis other than television watching, would seem more likely to engage in these practices when they do venture away from ‘the box’.
Figure 2.

MCA model of the space of lifestyles. (a) Modalities with above average contributions to axis 1 have been displayed. Reality TV and general entertainment programmes are included for comparison. Modalities with white boxes are supplementary. (b) Modalities with above average contributions to axis 2 have been displayed. Reality TV and general entertainment programmes are included for comparison. Modalities with white boxes are supplementary.

Figure 2.

MCA model of the space of lifestyles. (a) Modalities with above average contributions to axis 1 have been displayed. Reality TV and general entertainment programmes are included for comparison. Modalities with white boxes are supplementary. (b) Modalities with above average contributions to axis 2 have been displayed. Reality TV and general entertainment programmes are included for comparison. Modalities with white boxes are supplementary.

Close modal
Table 2.
Eigenvalues and percentage of variance explained.
EigenvaluePercentageModified eigenvalueaModified percentage
Axis 1 0.096 9.4 0.005 64.2 
Axis 2 0.062 6.1 0.001 15.8 
EigenvaluePercentageModified eigenvalueaModified percentage
Axis 1 0.096 9.4 0.005 64.2 
Axis 2 0.062 6.1 0.001 15.8 

aModified eigenvalues and percentages are calculated using the formula devised by Benzecri (1992) to correct for underestimation in MCA.

Table 3.
Contributions of the modalities to the axes.
LabelAxis 1Axis 2
DIY at least once a week 
diy yes 0.468 1.127 
diy no 0.067 0.163 
Total 0.534 1.290 
Photography 
photography yes 3.344 1.161 
photography no 1.055 0.394 
Total 4.399 1.555 
Textiles/crafts 
textilecrafts yes 1.563 5.248 
textilecrafts no 0.698 2.291 
Total 2.261 7.539 
Theatre at least several times a year 
theatre yes 4.321 3.050 
theatre no 0.937 0.680 
Total 5.258 3.730 
Opera/classical concert 
opera yes 5.302 2.496 
opera no 2.207 1.002 
Total 7.509 3.497 
Cinema at least once a month 
cinema yes 1.401 0.778 
cinema no 0.133 0.071 
Total 1.534 0.849 
Art galleries at least several times a year 
artgall yes 5.758 3.945 
artgall no 1.047 0.731 
Total 6.805 4.676 
Stately homes/historical sites at least several times a year 
histsite yes 3.935 2.510 
histsite no 1.110 0.725 
Total 5.045 3.236 
Theme parks at least several times a year 
themepark yes 1.180 0.051 
themepark no 0.185 0.010 
Total 1.365 0.062 
Libraries at least several times a year 
library yes 1.314 3.963 
library no 0.640 1.898 
Total 1.954 5.861 
Eat out at least once a week 
eatout yes 1.135 0.047 
eatout no 0.182 0.009 
Total 1.317 0.056 
Pub at least once a week 
pub yes 0.433 2.027 
pub no 0.075 0.359 
Total 0.508 2.385 
Gardening 
garden yes 0.426 0.183 
garden no 0.102 0.045 
Total 0.529 0.228 
Gym at least 2–3 times a month 
gym yes 2.975 0.260 
gym no 1.968 0.143 
Total 4.943 0.403 
Swimming at least 2–3 times a month 
swimming yes 2.933 0.001 
swimming no 0.796 0.000 
Total 3.729 0.001 
Cycling at least 2–3 times a month 
cycling yes 3.401 1.634 
cycling no 0.914 0.390 
Total 4.315 2.024 
Dancing 
dancing yes 1.517 1.645 
dancing no 0.971 0.993 
Total 2.488 2.639 
Jogging at least 2–3 times a month 
jogging yes 4.707 1.620 
jogging no 2.543 0.845 
Total 7.250 2.466 
Walking at least 2–3 times a month 
walking yes 1.562 0.572 
walking no 1.484 0.539 
Total 3.046 1.111 
Racquet sports 
racquet yes 5.859 2.851 
racquet no 1.746 0.842 
Total 7.605 3.693 
Team sports 
team yes 3.242 9.933 
team no 0.783 1.975 
Total 4.026 11.908 
Water sports 
watersports yes 4.927 1.131 
watersports no 0.739 0.163 
Total 5.666 1.294 
Horse riding 
horseriding yes 1.745 0.062 
horseriding no 0.133 0.005 
Total 1.878 0.067 
Yoga 
yoga yes 3.514 2.887 
yoga no 0.608 0.454 
Total 4.122 3.341 
Golf 
golf yes 1.491 10.818 
golf no 0.283 1.822 
Total 1.775 12.640 
Skiing 
skiing yes 3.899 2.190 
skiing no 0.351 0.184 
Total 4.251 2.374 
News programmes 
news yes 0.384 0.079 
news no 0.953 0.198 
Total 1.338 0.277 
Sports programmes 
sportprog yes 0.232 8.238 
sportprog no 0.211 7.690 
Total 0.443 15.929 
Comedy programmes 
comedy yes 0.033 0.148 
comedy no 0.096 0.451 
Total 0.129 0.599 
Documentary programmes 
documentary yes 0.159 0.004 
documentary no 0.478 0.012 
Total 0.637 0.016 
Drama programmes 
dramas yes 0.043 0.936 
dramas no 0.073 1.666 
Total 0.116 2.602 
Soap operas 
soaps yes 1.245 0.623 
soaps no 0.782 0.386 
Total 2.027 1.009 
Music programmes 
musicprog yes 0.001 0.000 
musicprog no 0.001 0.000 
Total 0.002 0.001 
Reality television 
realitytv yes 0.548 0.416 
realitytv no 0.254 0.189 
Total 0.802 0.605 
General entertainment programmes 
genentprog yes 0.157 0.015 
genentprog no 0.236 0.023 
Total 0.394 0.038 
LabelAxis 1Axis 2
DIY at least once a week 
diy yes 0.468 1.127 
diy no 0.067 0.163 
Total 0.534 1.290 
Photography 
photography yes 3.344 1.161 
photography no 1.055 0.394 
Total 4.399 1.555 
Textiles/crafts 
textilecrafts yes 1.563 5.248 
textilecrafts no 0.698 2.291 
Total 2.261 7.539 
Theatre at least several times a year 
theatre yes 4.321 3.050 
theatre no 0.937 0.680 
Total 5.258 3.730 
Opera/classical concert 
opera yes 5.302 2.496 
opera no 2.207 1.002 
Total 7.509 3.497 
Cinema at least once a month 
cinema yes 1.401 0.778 
cinema no 0.133 0.071 
Total 1.534 0.849 
Art galleries at least several times a year 
artgall yes 5.758 3.945 
artgall no 1.047 0.731 
Total 6.805 4.676 
Stately homes/historical sites at least several times a year 
histsite yes 3.935 2.510 
histsite no 1.110 0.725 
Total 5.045 3.236 
Theme parks at least several times a year 
themepark yes 1.180 0.051 
themepark no 0.185 0.010 
Total 1.365 0.062 
Libraries at least several times a year 
library yes 1.314 3.963 
library no 0.640 1.898 
Total 1.954 5.861 
Eat out at least once a week 
eatout yes 1.135 0.047 
eatout no 0.182 0.009 
Total 1.317 0.056 
Pub at least once a week 
pub yes 0.433 2.027 
pub no 0.075 0.359 
Total 0.508 2.385 
Gardening 
garden yes 0.426 0.183 
garden no 0.102 0.045 
Total 0.529 0.228 
Gym at least 2–3 times a month 
gym yes 2.975 0.260 
gym no 1.968 0.143 
Total 4.943 0.403 
Swimming at least 2–3 times a month 
swimming yes 2.933 0.001 
swimming no 0.796 0.000 
Total 3.729 0.001 
Cycling at least 2–3 times a month 
cycling yes 3.401 1.634 
cycling no 0.914 0.390 
Total 4.315 2.024 
Dancing 
dancing yes 1.517 1.645 
dancing no 0.971 0.993 
Total 2.488 2.639 
Jogging at least 2–3 times a month 
jogging yes 4.707 1.620 
jogging no 2.543 0.845 
Total 7.250 2.466 
Walking at least 2–3 times a month 
walking yes 1.562 0.572 
walking no 1.484 0.539 
Total 3.046 1.111 
Racquet sports 
racquet yes 5.859 2.851 
racquet no 1.746 0.842 
Total 7.605 3.693 
Team sports 
team yes 3.242 9.933 
team no 0.783 1.975 
Total 4.026 11.908 
Water sports 
watersports yes 4.927 1.131 
watersports no 0.739 0.163 
Total 5.666 1.294 
Horse riding 
horseriding yes 1.745 0.062 
horseriding no 0.133 0.005 
Total 1.878 0.067 
Yoga 
yoga yes 3.514 2.887 
yoga no 0.608 0.454 
Total 4.122 3.341 
Golf 
golf yes 1.491 10.818 
golf no 0.283 1.822 
Total 1.775 12.640 
Skiing 
skiing yes 3.899 2.190 
skiing no 0.351 0.184 
Total 4.251 2.374 
News programmes 
news yes 0.384 0.079 
news no 0.953 0.198 
Total 1.338 0.277 
Sports programmes 
sportprog yes 0.232 8.238 
sportprog no 0.211 7.690 
Total 0.443 15.929 
Comedy programmes 
comedy yes 0.033 0.148 
comedy no 0.096 0.451 
Total 0.129 0.599 
Documentary programmes 
documentary yes 0.159 0.004 
documentary no 0.478 0.012 
Total 0.637 0.016 
Drama programmes 
dramas yes 0.043 0.936 
dramas no 0.073 1.666 
Total 0.116 2.602 
Soap operas 
soaps yes 1.245 0.623 
soaps no 0.782 0.386 
Total 2.027 1.009 
Music programmes 
musicprog yes 0.001 0.000 
musicprog no 0.001 0.000 
Total 0.002 0.001 
Reality television 
realitytv yes 0.548 0.416 
realitytv no 0.254 0.189 
Total 0.802 0.605 
General entertainment programmes 
genentprog yes 0.157 0.015 
genentprog no 0.236 0.023 
Total 0.394 0.038 

Note: Modalities with above average contributions have been put in bold.

The second axis accounts for 16 per cent of the variance – bringing the total for the two dimensions to 80 per cent and making interpretation of further axes unnecessary – but examination of the space of individuals suggests this relatively low figure may be because the differentiation it reveals is more pronounced towards the exclusive pole (Figure 3).4 In any case it polarises, at one end, photography, textiles/crafts, theatre, opera, art galleries, historical sites, libraries, dancing and yoga to, at the other, undertaking regular DIY, cycling, jogging, racquet sports, team sports, watersports, golf, skiing and watching sports programmes.
Figure 3.

The space of individuals.

Figure 3.

The space of individuals.

Close modal
Examination of the supplementary variables indicates a slightly skewed homology with the social space (Figure 4). The first axis seems to correspond fairly neatly with volume of capital, though it appears that the white-collar workers have relatively petit-bourgeois tastes – fitting with the position and recent trajectory of this class (see Atkinson 2017) – and the administrators’ tastes are closer to sections of the dominated class than the rest of the intermediate class. The dominant class fractions are also perhaps a little lower, or less dispersed, than one might have imagined, even if that does not take away from the fact that the cultural dominant are the most closely associated with the activities demanding high levels of cultural capital, and the business executives and professions with those demanding economic capital. Internal variation plays a part here – as Figure 4 shows, for example, certain sections of the cultural dominant (cultural producers and intellectuals) which are not always numerous enough in samples to warrant disaggregation correspond much more closely with the highbrow cultural activities than do teachers, who make up a large proportion of this class fraction.
Figure 4.

Supplementary modalities in the space of lifestyles.

Figure 4.

Supplementary modalities in the space of lifestyles.

Close modal

One interpretation of the second axis, given the distribution of the class fractions, would be that it represents capital composition, with activities demanding and consolidating symbolic mastery, or catering towards the related dispositions of asceticism and expressivity, being opposed to those demanding economic capital (e.g. paying for equipment and club membership), building social capital (especially in relation to golf and team sports) and/or in synch with the overtly competitive ethos of the economic field (as with the sports and watching sports programmes). However, since it is the most feminised class fractions which correspond with the cultural pole – including the administrators and sales workers, who otherwise have relatively balanced capital composition – and the least with the economic pole – including the technicians, who, like the administrators, have a balanced capital composition – this division could also (or instead) reflect gendered tastes – the outer and the inner, as Bennett et al. (2009) had it. Projection of recorded sex into the space seems to confirm this.

The knot can, in fact, be untangled a little bit by examining the different rates of participation in the key activities on this axis by class fraction for both men and women. If we do this then the overall pattern which seems to emerge is one of significant interaction or, to use the fashionable term, intersection of capital composition and gender. In relation to sports or physical activities, for example, it can clearly be seen that some are taken up by one sex more than the other across the social space (Table 4). Dance and yoga are more feminine; golf, skiing, racquet, team and water sports and cycling are favoured more by men. Yet the feminised activities are, among men, undertaken at greater rates by men relatively richer in cultural capital than economic capital – the cultural dominant, the cultural intermediaries and the caring services – while those sports seemingly more masculine overall are (with the exception of cycling) practiced at a greater rate by those in class fractions richer in economic capital or more balanced in their holdings – the business executives, the professions, LMPs, technicians and skilled trades – to different degrees depending on the relative economic exclusivity of the practice. Among women, however, the same pattern does not really hold: the dominant class, and especially professionals, are most likely to practice both culturally and economically exclusive sports and physical activities, but differences between the two wings of social space are minimal, suggesting that capital composition may be less important than capital volume among women in this regard.

Table 4.
Taste for sports by class fraction and gender.
DanceYogaGolfSkiingRacquet sportsTeam sportsWater sportsCycling
Class fractionMenWomenMenWomenMenWomenMenWomenMenWomenMenWomenMenWomenMenWomen
Business execs +5.2 +9.0 +7.9 +16.6 +21.6 +1.5 +11.1 +7.1 +17.3 +11.7 +13.9 4.5 +14.1 +12.1 +11.5 +4.2 
Professions +11.7 +8.2 +11.5 +2.0 +16.7 +4.0 +20.6 +25.8 +15.7 +18.8 +14.4 +10.7 +19.9 +21.4 +12.4 +17.8 
White collar +14.4 +8.9 +11.2 +1.3 +14.0 +4.6 +6.0 +2.5 +13.6 +13.9 +9.6 +1.8 +8.1 +6.2 +1.4 +8.0 
Cultural dominant +20.7 +7.1 +18.1 +19.1 +6.0 +1.6 +7.1 +8.2 +13.2 +11.8 +6.4 +3.0 +11.5 +9.8 +14.4 +4.7 
LMPs +12.5 +9.7 +9.5 +16.6 +14.6 +0.6 +6.9 +3.6 +9.3 +2.4 +10.9 +1.7 +10.1 +5.3 +6.4 +6.3 
Technicians +1.9 +1.2 +4.6 +14.3 +14.7 +7.8 +5.0 +4.8 +12.7 +4.0 +16.3 +3.7 +11.9 +0.8 +10.9 +7.8 
Administrators +19.5 +4.3 +13.4 +9.4 +3.6 +0.8 +0.2 +0.1 +3.0 +3.5 9.1 +0.3 +1.6 +1.4 +8.9 +0.1 
Cultural intermediaries +18.1 48.2 +19.8 +17.9 +2.4 2.5 +3.9 +3.5 +9.1 +8.5 +1.6 +1.3 +9.2 +7.8 +14.6 +9.0 
Skilled trades 25.9 +2.5 4.4 11.0 +21.7 +2.9 +3.6 +3.5 +6.2 +5.5 +15.2 +7.8 +9.7 +3.1 +5.5 +9.1 
Sales workers +22.4 +4.3 +7.5 +3.9 +4.4 +1.7 +0.4 +0.6 14.8 12.1 +4.2 +2.5 5.4 5.0 22.6 10.0 
Manual workers +7.2 +4.0 +2.5 +2.4 +11.7 +2.2 3.5 +0.2 +2.5 +1.2 +11.4 +0.6 +4.5 +1.6 +0.6 +3.9 
Caring services +25.3 +6.6 +11.6 +6.5 4.5 +0.8 +0.1 3.3 +6.9 +10.9 +0.9 +3.7 +3.0 +3.0 +11.7 +4.6 
DanceYogaGolfSkiingRacquet sportsTeam sportsWater sportsCycling
Class fractionMenWomenMenWomenMenWomenMenWomenMenWomenMenWomenMenWomenMenWomen
Business execs +5.2 +9.0 +7.9 +16.6 +21.6 +1.5 +11.1 +7.1 +17.3 +11.7 +13.9 4.5 +14.1 +12.1 +11.5 +4.2 
Professions +11.7 +8.2 +11.5 +2.0 +16.7 +4.0 +20.6 +25.8 +15.7 +18.8 +14.4 +10.7 +19.9 +21.4 +12.4 +17.8 
White collar +14.4 +8.9 +11.2 +1.3 +14.0 +4.6 +6.0 +2.5 +13.6 +13.9 +9.6 +1.8 +8.1 +6.2 +1.4 +8.0 
Cultural dominant +20.7 +7.1 +18.1 +19.1 +6.0 +1.6 +7.1 +8.2 +13.2 +11.8 +6.4 +3.0 +11.5 +9.8 +14.4 +4.7 
LMPs +12.5 +9.7 +9.5 +16.6 +14.6 +0.6 +6.9 +3.6 +9.3 +2.4 +10.9 +1.7 +10.1 +5.3 +6.4 +6.3 
Technicians +1.9 +1.2 +4.6 +14.3 +14.7 +7.8 +5.0 +4.8 +12.7 +4.0 +16.3 +3.7 +11.9 +0.8 +10.9 +7.8 
Administrators +19.5 +4.3 +13.4 +9.4 +3.6 +0.8 +0.2 +0.1 +3.0 +3.5 9.1 +0.3 +1.6 +1.4 +8.9 +0.1 
Cultural intermediaries +18.1 48.2 +19.8 +17.9 +2.4 2.5 +3.9 +3.5 +9.1 +8.5 +1.6 +1.3 +9.2 +7.8 +14.6 +9.0 
Skilled trades 25.9 +2.5 4.4 11.0 +21.7 +2.9 +3.6 +3.5 +6.2 +5.5 +15.2 +7.8 +9.7 +3.1 +5.5 +9.1 
Sales workers +22.4 +4.3 +7.5 +3.9 +4.4 +1.7 +0.4 +0.6 14.8 12.1 +4.2 +2.5 5.4 5.0 22.6 10.0 
Manual workers +7.2 +4.0 +2.5 +2.4 +11.7 +2.2 3.5 +0.2 +2.5 +1.2 +11.4 +0.6 +4.5 +1.6 +0.6 +3.9 
Caring services +25.3 +6.6 +11.6 +6.5 4.5 +0.8 +0.1 3.3 +6.9 +10.9 +0.9 +3.7 +3.0 +3.0 +11.7 +4.6 

Notes: Rates are percentages within the class fraction. For clarity, the lowest rate within each genre has been presented in bold and then rates of participation among other class fractions recorded in terms of their distance in percentage points from the lowest rate.

Table 5.
Taste for selected practices by class fraction and gender.
Pub once a weekDIY once a weekTextiles/craftsLibrariesSport programmesSoapsReality TV
Class fractionMenWomenMenWomenMenWomenMenWomenMenWomenMenWomenMenWomen
Business execs +9.7 +6.2 +9.1 +2.9 20.6 +1.3 +6.9 +4.8 +32.0 +7.9 +5.8 +12.0 +6.0 +9.0 
Professions +7.8 +6.7 +3.3 +0.8 +6.6 +12.4 +13.6 +14.7 +29.0 +9.6 18.2 25.5 17.9 24.5 
White collar +8.0 +10.1 +3.4 +2.0 +6.6 +8.1 +7.9 +15.2 +25.6 +14.2 +12.2 +22.0 +9.7 +11.2 
Cultural dominant 10.3 +1.3 +3.1 +2.1 +22.0 +20.0 +25.6 +28.2 +9.1 +1.7 +10.5 +8.9 +5.8 +3.0 
LMPs +12.8 +13.5 +6.5 +1.8 +4.6 +2.3 +1.7 27.5 +23.2 +2.0 +19.0 +27.6 +15.6 +18.6 
Technicians +5.0 6.9 +16.3 +4.5 +2.1 +3.3 +8.4 +5.1 +29.5 +6.2 +6.6 +13.1 +6.3 +8.5 
Administrators +1.8 +3.1 5.2 4.7 +8.3 +2.0 +11.4 +11.3 +7.2 +3.4 +29.5 +28.3 +21.7 +20.0 
Cultural intermediaries +1.4 +1.7 +3.8 +3.9 +19.4 +12.9 +20.9 +16.2 +5.5 +1.6 +20.0 +20.1 +12.9 +12.7 
Skilled trades +13.6 +3.9 +22.6 +11.3 +4.9 +17.5 15.6 +1.8 +34.8 +7.4 +12.1 +27.1 +7.5 +9.7 
Manual workers +6.6 +5.6 +10.5 +1.7 +3.4 +2.8 +1.2 +6.1 +26.3 +1.7 +27.4 +41.6 +13.6 +15.9 
Sales workers +4.2 +1.9 +2.5 +2.6 +6.6 31.2 +9.4 +3.9 +10.7 +2.8 +32.8 +32.6 +20.7 +19.3 
Caring services +1.9 +4.7 +3.7 +3.0 +17.9 +10.0 +6.9 +15.8 31.1 26.8 +32.6 +28.2 +25.0 +20.8 
Pub once a weekDIY once a weekTextiles/craftsLibrariesSport programmesSoapsReality TV
Class fractionMenWomenMenWomenMenWomenMenWomenMenWomenMenWomenMenWomen
Business execs +9.7 +6.2 +9.1 +2.9 20.6 +1.3 +6.9 +4.8 +32.0 +7.9 +5.8 +12.0 +6.0 +9.0 
Professions +7.8 +6.7 +3.3 +0.8 +6.6 +12.4 +13.6 +14.7 +29.0 +9.6 18.2 25.5 17.9 24.5 
White collar +8.0 +10.1 +3.4 +2.0 +6.6 +8.1 +7.9 +15.2 +25.6 +14.2 +12.2 +22.0 +9.7 +11.2 
Cultural dominant 10.3 +1.3 +3.1 +2.1 +22.0 +20.0 +25.6 +28.2 +9.1 +1.7 +10.5 +8.9 +5.8 +3.0 
LMPs +12.8 +13.5 +6.5 +1.8 +4.6 +2.3 +1.7 27.5 +23.2 +2.0 +19.0 +27.6 +15.6 +18.6 
Technicians +5.0 6.9 +16.3 +4.5 +2.1 +3.3 +8.4 +5.1 +29.5 +6.2 +6.6 +13.1 +6.3 +8.5 
Administrators +1.8 +3.1 5.2 4.7 +8.3 +2.0 +11.4 +11.3 +7.2 +3.4 +29.5 +28.3 +21.7 +20.0 
Cultural intermediaries +1.4 +1.7 +3.8 +3.9 +19.4 +12.9 +20.9 +16.2 +5.5 +1.6 +20.0 +20.1 +12.9 +12.7 
Skilled trades +13.6 +3.9 +22.6 +11.3 +4.9 +17.5 15.6 +1.8 +34.8 +7.4 +12.1 +27.1 +7.5 +9.7 
Manual workers +6.6 +5.6 +10.5 +1.7 +3.4 +2.8 +1.2 +6.1 +26.3 +1.7 +27.4 +41.6 +13.6 +15.9 
Sales workers +4.2 +1.9 +2.5 +2.6 +6.6 31.2 +9.4 +3.9 +10.7 +2.8 +32.8 +32.6 +20.7 +19.3 
Caring services +1.9 +4.7 +3.7 +3.0 +17.9 +10.0 +6.9 +15.8 31.1 26.8 +32.6 +28.2 +25.0 +20.8 

Notes: Rates are percentages within the class fraction. For clarity, the lowest rate within each genre has been presented in bold and then rates of participation among other class fractions recorded in terms of their distance in percentage points from the lowest rate.

The data for cultural activities and venues displays similar patterns (Table 5). This time, regularly going to the pub and undertaking DIY are generally more masculine, as is watching sports programmes, while doing crafts/textiles, going to libraries and viewing soaps and (since it corresponds with the cultural/feminine pole of the model) reality television are disproportionally pursued by women across the board. Among men, however, crafts/textiles and libraries are most associated with the class fractions richer in cultural capital, particularly in the dominant and intermediate classes, while going to the pub, doing DIY and watching sport – the latter two in particular reflecting valorisation of practical mastery and certain forms of physical capital – are associated with class fractions richer in economic and/or technical capital, more prominently so lower down in social space. The opposition is, therefore, between the ‘north-west’ and ‘south-east’ poles of the social space. Soaps and reality television are not so clearly associated with cultural capital, but are spurned to the greatest extent by class fractions richer in economic and technical capital in the different classes.

Among women, this time some of the same patterns as found among men can be detected, albeit with some notable differences: libraries are visited most by the culturally rich fractions of the classes, and textiles and crafts are undertaken most by the cultural dominant, cultural intermediate and caring services – but also the skilled workers. This could be because women in this class fraction tend to be in specific trades, especially textiles and upholstery, and may thus be applying their specific species of technical capital. Going to the pub regularly is certainly avoided by the culturally rich, and popular among LMPs as well as white-collar workers, whilst DIY is most popular among skilled workers (and perhaps secondarily technicians) but then indistinguishably so among others. Soaps and reality television are, once again, less popular within the dominant class as a whole, though the white-collar workers are more partial, and favoured least by skilled workers and technicians within their classes. As with men, administrators and sales workers also seem drawn to soaps and reality television, perhaps suggesting a relationship between an orientation towards affect – since these programmes are all about human relationships and emotional drama – and jobs foregrounding interpersonal skills which, insofar as they are occupied disproportionately by women, are cast as distinctly feminine (‘soft skills’).

To summarise, when it comes to the correspondence between the social space and lifestyles in the UK, the prime dimension of difference, homologous with volume of capital, situates practices and tastes vis-à-vis a series of binaries: accessible/exclusive, common/rare, popular/restricted, public/home-based. This appears to be crosscut, however, by a second series of binaries reflecting both capital composition and gendered dispositions: asceticism/materialism, expression/competition, inner/outer, affective/physical. Judging from the class and gender distributions of practices, it might be conjectured that the inner/outer and affective/physical oppositions, and thus overtly gendered dispositions, are more distinguishing lower down in social and symbolic space, while higher reaches are cleft more by capital composition. The result is, therefore, a space of lifestyles organised around four polarities – abstract-expressive, competitive-exclusive, physical-technical and concrete-affective – defined not just by their mutual opposition but, in all likelihood, by mutual antagonism too, undergirding the symbolic struggles and symbolic violence of everyday life manifest not least in all the insulting categorisations and labels hurled at the socially distant and their lifestyles: ‘crass’, ‘brute’, ‘sissy’, ‘flashy’, ‘pretentious’, ‘soppy’, ‘airy fairy’ and so on. Moreover, examination of the data suggests that the gendered division between affect and physicality is not reducible to recorded sex but distinguishes, on one side, women and men in the poorly paid but increasingly numerous service jobs of the ‘post-industrial’ economy from, on the other, largely male skilled workers and technicians, for whom physical and technical prowess – paired in perception with certain ideals of masculinity – is a source of value.

Taken together with analyses presented elsewhere (Atkinson 2016b, 2017; Atkinson and Deeming 2015), therefore, the results not only broadly confirm the applicability of Bourdieu's general thesis on the correspondence between class and lifestyles to twenty-first-century Britain. They also confirm, but accentuate and refine, his statement that sex becomes a more salient principle of taste differentiation the lower and further towards the economic pole one goes in social space since cultural capital inclines both men and women to question and challenge traditional conceptions of ‘what men/women should do’. Yes, affect versus physicality in the dominated class may be homologous with femininity versus masculinity, but the division is not strictly between women and men, as culturally defined by possession of certain physiological properties, but between a version of masculinity – with a long historical pedigree – premised on physical strength or capability and orientations and interests which are defined, in opposition, as feminine (‘soft’, ‘mushy’, ‘wimpy’, ‘gay’, etc.), whether they be displayed by someone perceived to be a woman or a man. This refinement of Bourdieu's thesis is likely to be the product of deeper deindustrialisation since Distinction's day – the decline of traditional skilled male labour and the growth of low-paid service sector jobs emphasising skills and dispositions usually thought of as feminine (care, empathy, self-presentation, etc.) to soak up not just increasing amounts of women entering the workforce but young men with a little cultural capital (see Adkins 2002). In any case the horizontal symbolic struggle within the dominated class tends to mask the greater problem for its warring parties, which is that neither of their orientations are institutionalised and widely defined as legitimate. Vertical domination, and the symbolic power and symbolic violence it produces, ensure that it is education and its symbols, or wealth and it signs, not physical and practical mastery or interest in the concrete-affective, which continue to be misrecognised as definitive of one's worth in the UK today.

No potential conflict of interest was reported by the author.

Will Atkinson is Reader in Sociology in the School of Sociology, Politics and International Studies at the University of Bristol, UK. He is the author of Class, Individualization and Late Modernity (2010, Palgrave Macmillan), Class (Polity, 2015), Beyond Bourdieu (Polity, 2016) and Class in the New Millennium (Routledge, 2017), and co-editor of Class Inequality in Austerity Britain (2012, Palgrave Macmillan).

1

Vandebroeck (2016) offers the exception to the muted or compartmentalised treatment of gender to be found among Bourdieu-inspired analysts of consumption, but he focusses on the narrower field of bodily practices, and with it (though he does not put it as such) the intersection of class practices with the sexual field.

2

For more background information on the BCS and its various waves, see the survey website: http://www.cls.ioe.ac.uk/BCS70.

3

The remaining lifestyle variables are set as passive in order to maintain balance in the model in two senses: first, following LeRoux and Rouanet's (2004) advice, to neutralise modalities otherwise registering an excessive contribution on a major axis, usually on account of rarity; and second, transposing Rosenlund's (2015) guidance on constructing a robust model of social space using MCA, to maintain relative parity between modalities hypothesised to correspond with high cultural capital – overrepresented in surveys like the BCS which take them alone to represent ‘culture’ – and those conjectured to correspond more with high economic capital.

4

This is confirmed if the model is rerun excluding the dominated class and again excluding the dominated and intermediate classes. In both cases the general structure of the model is the same as that of the full model, but in the first case the relative strength of the two major axes shifts to 57 per cent and 21 per cent, and in the second case to 50 per cent and 25 per cent. The full models cannot be reported here for lack of space but details can be supplied on request.

Adkins
,
L.
(
2002
)
Revisions: Gender and Sexuality in Late Modernity
,
Buckingham
:
Open University Press
.
Adkins
,
L.
and
Skeggs
,
B.
(eds.) (
2004
)
Feminism After Bourdieu
,
Oxford
:
Blackwell
.
Anthias
,
F.
(
2001
) ‘
The concept of “social division” and theorising social stratification
’,
Sociology
35
(
4
):
835
54
.
Atkinson
,
W.
(
2011
) ‘
The context and genesis of musical tastes: Omnivorousness debunked, Bourdieu buttressed
’,
Poetics
39
(
3
):
169
86
. doi:
Atkinson
,
W.
(
2016a
)
Beyond Bourdieu: From Genetic Structuralism to Relational Phenomenology
,
Cambridge
:
Polity
.
Atkinson
,
W.
(
2016b
) ‘
The structure of literary taste: Class, gender and reading in the UK
’,
Cultural Sociology
10
(
2
):
247
66
. doi:
Atkinson
,
W.
(
2017
)
Class in the New Millennium: The Structure, Homologies and Experience of the British Social Space
,
London
:
Routledge
.
Atkinson
,
W.
and
Deeming
,
C.
(
2015
) ‘
Class and cuisine in contemporary Britain
’,
The Sociological Review
,
63
(
4
):
876
96
. doi:
Atkinson
,
W.
and
Rosenlund
,
L.
(
2014
)
Mapping the British Social Space: Toward a Bourdieusian Class Scheme
, SPAIS working paper no. 02-14, University of Bristol, http://www.bristol.ac.uk/spais/research/workingpapers/
Bennett
,
T.
,
Savage
,
M.
,
Silva
,
E.
,
Gayo-Cal
,
M.
and
Wright
,
D.
(
2009
)
Culture, Class, Distinction
,
London
:
Routledge
.
Benzécri
,
J.-P.
(
1992
)
Correspondence Analysis Handbook
,
New York
:
Dekker
.
Bourdieu
,
P.
(
1984
)
Distinction
,
London
:
Routledge
.
Bourdieu
,
P.
(
1998
)
Practical Reason
,
Cambridge
:
Polity
.
Bourdieu
,
P.
(
2000
)
Pascalian Meditations
,
Cambridge
:
Polity
.
Bourdieu
,
P.
(
2001
)
Masculine Domination
,
Cambridge
:
Polity
.
Bourdieu
,
P.
(
2005
)
The Social Structures of the Economy
,
Cambridge
:
Polity
.
Chan
,
T.W.
and
Goldthorpe
,
J.H.
(
2007
) ‘
Social stratification and cultural consumption: music in England
’,
European Sociological Review
23
:
1
19
. doi:
Christin
,
A.
(
2012
) ‘
Gender and highbrow cultural participation in the United States
’,
Poetics
40
(
5
):
423
43
. doi:
Coles
,
T.
(
2007
) ‘
Negotiating the field of masculinity
’,
Men and Masculinities
12
(
1
):
30
44
. doi:
Coulangeon
,
P.
and
Duval
,
J.
(eds.) (
2014
)
The Routledge Companion to Bourdieu's Distinction
,
London
:
Routledge
.
Cveticanin
,
P.
(ed.) (
2012
)
Social and Cultural Capital in Serbia
,
Belgrade
:
Centre for Empirical Cultural Studies of South-East Europe
.
George
,
M.
(
2014
) ‘Rejecting the specifically sexual: locating the sexual field in the work of Pierre Bourdieu’, in
I.
Green
(ed.),
Sexual Fields
,
Chicago
:
University of Chicago Press
, pp.
101
22
.
Holt
,
D. B.
(
1997
) ‘
Distinction in America? Recovering Bourdieu's Theory of tastes from its critics
’,
Poetics
25
:
93
120
. doi:
LeRoux
,
B.
and
Rouanet
,
H.
(
2004
)
Geometric Data Analysis
,
Dordrecht
:
Kluwer
.
Lizardo
,
O.
(
2006
) ‘
The puzzle of women's ‘highbrow’ cultural consumption
’,
Poetics
34
(
1
):
1
23
. doi:
McCall
,
L.
(
1992
) ‘
Does gender fit?
’,
Theory and Society
21
(
6
):
837
67
. doi:
McNay
,
L.
(
1999
) ‘
Gender, habitus and the field
’,
Theory, Culture and Society
16
(
1
):
95
117
. doi:
Mills
,
C.
(
2014
) ‘
The Great British class fiasco
’,
Sociology
48
(
3
):
437
44
. doi:
Peterson
,
R. A.
(
1992
) ‘
Understanding audience segmentation: from elite and mass to omnivore and univore
’,
Poetics
21
:
243
58
. doi:
Peterson
,
R. A.
(
1997
) ‘
The rise and fall of snobbery as a status marker
’,
Poetics
25
:
75
92
. doi:
Prieur
,
A.
and
Savage
,
M.
(
2011
) ‘
Updating cultural capital theory
’,
Poetics
39
:
566
80
. doi:
Prieur
,
A.
and
Savage
,
M.
(
2013
) ‘
Emerging forms of cultural capital
’,
European Societies
15
:
246
67
. doi:
Prieur
,
A.
,
Rosenlund
,
L.
and
Skjott-Larson
,
J.
(
2008
) ‘
Cultural capital today: a case study from Denmark
’,
Poetics
36
:
45
71
. doi:
Purhonen
,
S.
,
Gronow
,
J.
and
Rakhonen
,
K.
(
2011
) ‘
Highbrow culture in Finland
’,
Acta Sociologica
54
(
4
):
385
402
. doi:
Rosenlund
,
L.
(
2009
)
Exploring the city with Bourdieu
,
Saarbrucken
:
VDM Verlag Dr. Müller
.
Rosenlund
,
L.
(
2015
) ‘Working with distinction: Scandinavian experiences’, in
P.
Coulangeon
and
J.
Duval
(eds.),
The Routledge Companion to Bourdieu's Distinction
,
London
:
Routledge
, pp.
157
86
.
Savage
,
M.
, et al
. (
2013
) ‘
A new model of social class?
’,
Sociology
47
(
2
):
219
50
. doi:
Vandebroeck
,
D.
(
2016
)
Distinctions in the Flesh
,
London
:
Routledge
.
Wuggenig
,
U.
(
2007
) ‘
Pitfalls in testing Bourdieu's homology assumptions using mainstream social science methodology
’,
Poetics
35
:
306
16
. doi:
This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license, which permits unrestricted use, distribution, and reproduction in any medium, provided the use is non-commercial and the original work is properly cited. For a full description of the license, please visit https://creativecommons.org/licenses/by-nc/4.0/legalcode.