Is eclecticism an emerging form of cultural capital in the country of Pierre Bourdieu? Empirical evidences and theoretical questions.

Transcription

1 Is eclecticism an emerging form of cultural capital in the country of Pierre Bourdieu? Empirical evidences and theoretical questions. Philippe Coulangeon Observatoire Sociologique du Changement SciencesPo/CNRS 27, rue Saint-Guillaume Paris F philippe.coulangeon@sciences-po.fr This paper explores the ambiguities surrounding the notion of cultural eclecticism as an emerging form of cultural capital in contemporary France. My contention here is that although the distinctive power of cultural eclecticism is actually empirically grounded, this observation only partially questions Bourdieu s views on cultural capital which is, by and large, buttressed. The first section of the paper reviews some of the contentious issues related to the so-called omnivorous pattern. The second section focus more precisely on the impact of social mobility on taste and cultural practices, to the extent that omnivorousness has so far often been considered as an indirect consequence of the status inconsistency or plural socialization associated with the experience of mobility. The third section deals with the parallel that can be drawn between cultural eclecticism and linguistic registers variety. Finally, the fourth section addresses the link between the social distribution of eclectic dispositions and cosmopolitan resources in the French context. 1. The rising norm of cultural eclecticism in question The traditional mass/elite model of the social stratification of cultural attitudes has long been seriously challenged by the emergence in the mid-1990s of the omnivore/univore paradigm. Originally formulated by Richard Peterson, omnivorousness refers to the amount and diversity of practices people enjoy rather than to their cultural legitimacy (Peterson, 1992; Peterson and Simkus, 1993; Peterson and Kern, 1996; Peterson, 1997). It also refers to the observation that this amount of diversity is correlated with social class. As such, the very notion of omnivorousness seems to challenge the isomorphic interpretation of class-based social differences in cultural habits and tastes, to the extent that what might be termed elite taste is defined by a mixing of practices rather than by a predominant connection with so-called highbrow practices. Indeed, a great amount of empirical evidence supports the growing eclecticism of upper class cultural consumption. In France, this trend has been well documented since its first mention in the mid-1990s (Donnat, 1994). Nonetheless, the very notion of cultural omnivorousness has sparked sharp debate regarding its intrinsic meaning. Three of the conceptual and empirical issues raised by cultural omnivorousness are emphasized in what follows.

2 From the volume to the compositional definition of omnivorousness The first contentious issue raised by the omnivorousness paradigm refers to the contrast between its volume and compositional interpretations (Warde and Gayo-Cal, 2009). With regard to the former, which some authors prefer to call voraciousness (Sullivan and Katz-Gerro, 2007), omnivorousness simply corresponds to the observation that more fortunate people (in terms of economic and cultural resources) tend to engage in a wide variety of practices. In other words, this first understanding of the notion mainly refers to the cumulative feature of cultural consumption and, perhaps more broadly, of leisure activities, which tend to be more addictive in nature (Coulangeon and Lemel, 2009). The more people consume specific cultural goods, the more likely they are to become involved in other forms of cultural consumption. Indeed, current surveys on cultural consumption show that people who are voracious readers also buy records, go to the cinema and theatre, watch DVDs, etc. With regard to the compositional interpretation, omnivorousness relates more to the ability to cross the cultural and symbolic boundaries that separate various cultural practices and overcome their aesthetic differences. Clearly, the compositional interpretation of omnivorousness is much more restrictive and controversial than the volume interpretation. For many, it is far from clear that cultural omnivorousness may constitute a coherent and unambiguous set of cultural dispositions (Warde et al., 2007). While the latter definition refers to the notion of cultural openness and aesthetic tolerance, the former is not exclusive to more segmented forms of omnivorousness. Practices, dispositions and tastes The prevailing confusion about the indicators by which omnivorousness is usually grasped in survey data is a second contentious aspect of the debate on cultural eclecticism. Whereas some indicators refer to practices, others refer to tastes or dispositions. And the meaning attached to the notion of cultural eclecticism or omnivorousness might be strongly affected by this confusion, to the extent that practice is not always related to taste, as when, for instance, participation is due to familial or professional duty. In addition, it has often been noticed that the conventional approach of tastes as well as practices through standardized list of genres or repertoires might conceal a large part of the internal divisions and hierarchies of these genres and repertoires, as it is obviously the case in the field of music (Glévarec and Pinet 2012). But even though hidden hierarchies might be better measured with more detailed lists of genres, artist names and/or pieces of music, this purely nominal approach to taste would fail to grasp more subtle distinctions attached to the way in which people consume and appreciate cultural products rather than the products they consume themselves. In the music field, a crucial part of pertinent cultural distinctions may indeed lie in the different ways of listening to the same kind of music (Atkinson 2011). The extent to which people combining various tastes or practices in which they get involved with similar dispositions can be considered as truly omnivorous is at least questionable. Finally, unless the dispositions underlying practices and tastes can be qualified and measured, it is hardly possible to consider omnivorousness as a generic disposition. Recent attempts were made in this direction, though, relying on indicators that measure aesthetic dispositions rather than practices or tastes (Daenekindt and Roose 2013).

3 The blind-spot of mass-culture A third critical point in the omnivorousness literature relates to a certain misconception of mass-culture and the disruption it creates in the definition of the line between the so-called highbrow and lowbrow repertoires. Many contemporary cultural sociologists, whatever their position in the omnivore/univore versus highbrow/lowbrow debate, probably fail to adequately identify this issue. This failure is largely due to the common aggregation of all non-highbrow cultural genres into one large, indistinct lowbrow category, as it is the case in the musical field with the loose pop music category. What is currently termed lowbrow or pop is indeed an aggregate of very heterogeneous genres, from mass-produced cultural goods to quite underground subcultural repertoires. Omnivorousness cannot be adequately understood when it relies on such a crude categorization. Relying on the Frankfurt School philosophers thoughts on this matter, it must previously be noticed that the production of a large amount of mass-culture items precisely aims to hide class divisions, in so much that the social profile of these items consumers is not surprisingly rather class-neutral (Gartman, 1991). In that sense, the taste for mass culture can hardly be considered to be a social marker itself, although exclusive access to it and taste for it are statistically much more frequent in the lower classes than in the upper classes. Thus, omnivorousness cannot simply be defined as the mixing of highbrow and mass-produced cultural items, to the extent that the former are much more clearly class-related than the latter. Consequently, what is termed as omnivorous might often correspond to the much more trivial observation that mass-culture items are consumed by almost everybody, at least superficially, including people from socially and culturally privileged background. To sum up, true omnivorousness, if any, requires: a) that diversity in practices and/or tastes is not only a matter of volume, but implies to some extent the crossing of some boundary across repertoires; b) that omnivorousness can be identified to some extent as a generic disposition; c) that the cultural eclecticism of the well-to-do does not only consist in a mix of highbrow practices or tastes and the casual consumption of the most common items of mass-culture. 2. Omnivorousness and social mobility: status inconsistency and cultural goodwill The process by which the omnivore/univore pattern has progressively superseded the traditional highbrow/lowbrow divide is often related to the impact of social mobility. As a result of the expansion of the service economy and the relative contraction of the working class (or at least of its strictly blue-collar component), the service-class is welcoming an increasing number of upwardly mobile people who did not grow up in an upper class environment. Thus, upper class omnivorousness would predominantly be the result of these newcomers bringing heterogeneous dispositions, tastes and cultural influences. Consequently, the increasing omnivorousness of the upper class, as it is generally measured at the aggregate level, could be partly a compositional artefact resulting from the aggregation of socially and culturally heterogeneous individuals whose preferences and practices would remain quite coherently univore if considered at the individual level. For example, when a majority of upper class people who favour classical music and opera is challenged by a growing number of people who prefer jazz or rock, this results in the greater omnivorousness of the upper class as a whole even though upper class individuals themselves might remain univores.

4 Existing research partially validates this interpretation: social mobility increases the cultural heterogeneity of social classes. However, the omnivorous trend remains when we control for this compositional effect (Van Eijck, 1999). Therefore, the most common interpretations of the omnivorous phenomenon also consider the impact of social mobility on individual behaviours themselves. Eclectic cultural habits and tastes are then often considered as the consequence of the incongruous influences to which people are exposed when they experience social mobility (Lahire, 2004 and 2008). In that sense, omnivore cultural resources, considered as a specific form of cultural capital of the upwardly mobiles, could be envisaged as the positive counterpart of status inconsistency. However, this assertion is not fully supported empirically, as can be seen in the following example relating to musical tastes. Three profiles of musical listeners The 2008 survey on French Cultural Practices asked respondents about which musical genres they listen to most often. The question, which included a list of 11 highbrow and lowbrow genres and allowed for multiples answers, was well-suited for measuring both the highbrow/lowbrow and omnivore/univore gradient. The list of genres includes French pop, world music, foreign pop, techno, rap, metal, pop and rock, jazz, opera, classical music, and an other genres category that included all genres cited by less than 1% of respondents (see table 1). Table 1: List of the 11 most listened to music genres (percentage) French pop 0.66 Foreign pop 0.34 Classical 0.30 Rock 0.26 World music 0.24 Jazz 0.19 Techno 0.10 Opera 0.10 Other 0.07 Rap 0.06 Metal 0.06 Source: Survey on French Cultural Participation, 2008; active and ex-active people aged 15 and above (N = 3831) In order to asses for the impact of social mobility, the population under consideration is limited to professionally active or formerly active people whose current or last occupation, as well as that of their father when they were 15 years old, are filled in (N = 3831) 1. 1 Of course, one could question this highly male-biased understanding of social mobility. Various studies have demonstrated that cultural transmission is a highly gendered phenomenon (Mohr and DiMaggio 1995), and have insisted on a mother s influence in this process. Accordingly, the mother s occupation, or at least both the mother s and the father s occupation, could preferably be taken into consideration. However, as many of the respondents mothers belong to past generations in which the employment rate of women was rather small,

5 Respondents and their fathers class are coded as a five-category variable: farmers, shopkeepers, craftsmen (Class I); senior managers, professionals, business owners (Class II); middle managers (Class III); routine non-manual workers (Class IV); and manual workers (Class V). Upward mobility is defined as the situation of all people in classes II or III and whose fathers belonged to classes I, IV and V, in addition to people belonging to class II whose father belonged to class III. Downward mobility is symmetrically defined as the situation of people belonging to classes I, IV or V and whose fathers belonged to classes II and III, in addition to people belonging to class III whose father belonged to class II 2. For the analysis of respondents patterns of music consumption, a latent class analysis (LCA) was applied to the distribution of the 11 most listened to music genres mentioned, following a standard analytical strategy (e.g. Chan and Goldthorpe 2007). The aim of LCA is to identify a categorical latent variable that explains correlations between the manifest indicators here, the 11 binary variables of music genres of a given phenomenon. This latent variable is unknown. Its modalities form the latent classes or clusters across which respondents are distributed according to the pattern of associations between their responses on the manifest indicators so that, conditional on membership in these classes, the indicators become statistically independent of one another (McCutcheon 1987) 3. The result of the analysis shows that a model postulating three latent clusters fits the data satisfactorily according to the p-value criterion (p greater than 0.05) (see table 2). However, according to the BIC criterion (the smaller, the better), the six-cluster model would have been preferable. Given that the four-, five- and six-cluster solutions turned out to be unstable (i.e. the successive computations of the model do not always produce the same estimated conditional probabilities and the same estimated cluster sizes), the five percent criterion was favored, thus leading to select the three-cluster solution. the operationalization of this would be quite problematic. Consequently, in what follows, social class of origin only refers to the father s occupation. 2 Another research strategy would have been to focus on educational rather than social mobility by comparing the respondents highest degree with their parents educational achievements. Measuring social mobility based on educational mobility is indeed a fairly common and valid approach, as the net impact of education on cultural attitudes is known to be greater than the impact of all other socio-demographic characteristics (DiMaggio and Useem 1978; Daenekindt and Roose 2013). Although the survey includes questions on the educational attainment of both the respondents and their parents, the high number of missing values as to the latter and the uncertain reliability of many responses make this research strategy inadequate in this case. 3 Formally, a latent class model with T latent classes and three categorical variables A, B, C with I, J, K categories can be expressed as follows:, Where is the probability that a respondent is found simultaneously at level i of variable A, level j of variable B, level k of variable C and belongs to latent class t,, stands for the probability that a respondent belongs to latent class t,, stands for the conditional probability that a respondent belonging to latent class t is found at level i of the manifest variable A, and so on for the two other manifest variables B and C. All models are fitted with Latent Gold 4.5 (Vermunt and Magidson 2005).

6 Table 2: Latent class measurement models fitted to music listening data # clusters L² df p BIC The estimated relative sizes of the clusters and the estimated conditional probabilities are reported in Table 3. The first music listener profile, with an estimated probability of a little over 65%, is characterized by a quite exclusive taste for French pop music, which is the only genre with a conditional probability above 50 %. The second cluster, with an estimated relative size of 25%, is much smaller than the first. Similarly, it displays a high conditional probability for French pop music (70%) but an even higher conditional probability for classical music (85%). The last cluster is even smaller (9%) and displays conditional probabilities above 50% for six of the 11 genres (French pop, foreign pop, rock, world music, jazz and classical music). Table 3: Estimated relative sizes of latent clusters and the conditional probabilities of the most often listened to music genres under the three-cluster model French Pop Univores Highbrow Omnivores (FPU) (H) (O) Relative size French pop Foreign pop Classical Rock World music Jazz Techno Opera Rap Metal Other Source: Survey on French Cultural Participation, 2008; active and ex-active people aged 15 and above (N = 3831) Although the individuals in the first cluster can be unambiguously labeled French pop univores and the third true omnivores, the second cluster is less easy to qualify. To the extent that French pop is equally present in all of three clusters, with an estimated conditional probability of between 63 and 76%, its combination with classical music is hardly sufficient to qualify the second cluster as truly omnivorous. What distinguishes this cluster

7 from the other two is that in addition to the shared taste in French pop music, the people who belong to this cluster also predominantly listen to classical music. In comparison with the third cluster, it corresponds more to a conventional highbrow profile than to an authentically eclectic one. In addition, cluster 2 is nearly three times as frequent as cluster 3. Incidentally, it is worth noticing that this third cluster combines classical music, opera and jazz with French and foreign pop, rock and world music, with no special emphasis on the most stigmatized lowbrow genres, such as rap and metal. This confirms that genuine omnivorousness hardly corresponds with undifferentiated eclecticism (Bryson 1996). Finally, these results support quite strongly the idea that the omnivore/univore divide is not exclusive to an underlying highbrow/lowbrow pattern. This result does not correspond to what is commonly observed in other studies on this topic, though. For example, in their analysis of music tastes and consumption in the United Kingdom, Chan and Goldthorpe do not identify any real highbrow profile and conclude that the omnivore/univore divide gives a better account of taste differentiation than the highbrow/lowbrow divide. However, their conclusion is most likely affected by their very crude categorization of music genres, including an overly aggregated approach to the lowbrow music genres. Indeed, they only differentiate opera and operetta, jazz, classical music, and, due to the design of the survey from which their data is extracted, an all-encompassing pop and rock category (Chan and Goldthorpe 2007). And, indeed, if Chan and Godthorpe s same categorization is used but with French data, very similar results occur: when all music genres but jazz, opera and classical music are aggregated into a single category, the latent class analysis results in a rather clear differentiation between univores and omnivores, with no obvious highbrow/lowbrow divide 4. However, this is no longer the case when the analysis is based on a more disaggregated categorization, as displayed in Table 3. In other words, it is only when a rather detailed list of genres and sub-genres is taken into consideration that true omnivorousness can be correctly distinguished from common access to and taste for mass culture, which can cohabit with more distinctive access to and taste for highbrow culture. To the extent that mass-produced pop music is today pervasively present in mass media and is casually ubiquitous in daily life, pure highbrow music listening habits, defined by the total rejection of mass-produced pop music, becomes increasingly unrealistic. As a result, people listening to both classical music or opera and mass-produced pop music, when contrasted to people who only listen to the latter, are predominantly distinguished by the fact they also listen to classical music or opera. In that respect, it can be argued that truly omnivorous disposition involves more than a mere contamination of elite culture by the most widespread artifacts of mass culture. It may in addition involve the intrinsic appreciation of more unusual subcultural or countercultural items. It is the reason why the adequate measurement of omnivorousness requires a greater disaggregation of musical genres that would enable a more accurate appreciation of the diversity of musical listening habits than a crude distinction between classical and pop music. 4 According to the BIC criteria, a 3-cluster solution is then acceptable. The first cluster includes 75% of the sample and roughly corresponds to the Univore cluster of the analysis displayed in table 3, with 87% of the respondents in this cluster mentioning pop music, whereas none of the other three genres is mentioned by more than 9% of the respondents. They by and large corresponds to two variants of omnivorousness, with a majority of respondents listening to classical music and French pop in the third cluster, together with jazz in the second. In the second as well as in the third cluster, 37% of the respondents also mention opera.

8 Social mobility and musical taste The more interesting, and rather unexpected result of this analysis, relates to the impact of social mobility on musical taste, as it can be inferred from the cross-tabulation of the three aforementioned latent clusters with a 25-category origin/position variable which itself results from the cross-tabulation of the respondents social class of origin and social class of position. This contingency table is then introduced in a correspondence analysis that provides a graphical representation of the associations between the levels of the crosstabulated variables. The graphical representation displays the rows and columns of the contingency table as points in a low-dimensional space, such that the positions of the row and column points are consistent with their associations in the table. As the latent variable, which corresponds to the smaller dimension of the contingency table, contains only three levels, the association between the two variables is perfectly resumed by a two-dimensional space. Figure 1 displays the positions of each row and column mean points. The row and column points that contribute the most to axis 1 are coloured in black, while those contributing the most to axis 2 are coloured in light grey. Figure 1: Plane 12 - map of the row and column points Note : for the class categories, the first number, before the slash (/), indicates the class of origin. The sceond one correponds to the class of destination. As to the column points (latent clusters), the first axis displays a clear-cut contrast between the Highbrow cluster, on the left side of the map, and the French pop univore one on the right. Concerning the row categories, axis 1 appears mainly defined by the opposition of the non-mobile respondents of the upper-class (2/2) and the slightly upwardly mobile respondents from the upper class originated in the middle class (3/2), on the one hand, to respondents from class 4 originated in class 5. More generally, this first axis is noticeably structured by a clear-cut class-gradient.

9 Axis 2 displays a second structuring principle that contrasts, as to the column points, the highbrow cluster, on the top of the graph, to the omnivorous one, on the bottom. According to the hierarchical structure of the axes extracted by the CA algorithm, it is worth noticing that this contrast is a second-rank structuring principle, by comparison with the main contrast displayed by axis 1. As to the row categories, this second axis appears more clearly structured by social mobility than the previous one. Indeed, it mainly contrasts non-mobile (2/2, 3/3) and slightly mobiles (2/3) respondents of upper and middle class on the bottom to upwardly mobile respondents (5/2, 1/4, 4/3). Thereby, upwardly mobiles appear closer to the highbrow cluster than to the omnivorous one, whereas the reverse holds true for the non-mobiles respondents of the upper and middle class. This result does not support the current understanding of eclecticism as a result of the plural socializing environments mobile people get through during their life course (Lahire, 2010 [2001]). Rather, it suggests that the cultural and/or cognitive dissonance (Festinger, 1957) experienced by upwardly mobile people is more likely to be resolved by a gradual realignment with the cultural norms of the upper-class which may turn into an overadaptation process (Wilensky, Edwards 1959; Merton 1968). It also hark back to the kind of cultural over-conformism observed by Bourdieu and Passeron in France in the mid-sixties among upwardly mobile students who tended to display a higher deference to the norms of cultural legitimacy than those of upper-class origin (Bourdieu & Passeron 1979 [1964]). In that sense, upwardly mobiles attitudes also bring to mind what Bourdieu termed as cultural goodwill in relation to middle classes emulation of the high-class manners and lifestyles. On the contrary, CA displays a substantial attraction between omnivorousness and the core members (i.e. non mobiles) of the upper class. It therefore seems that cultural mobility of the upper class, understood as the ability to move at will between cultural realms (Emmison, 2003, 211), conveys a distance to the norm of cultural legitimacy that is less accessible to the newcomers in this class. This ability to play with this norm and to switch from one cultural repertoire to another recalls what has long been observed in sociolinguistics. 3. Cultural variety and code-switching The uneven ability of people to operate in a wider variety of cultural domains probably proceeds from the same kind of social factors than those observed in linguistics skills. It is well-established that an individual s range of vocabulary, the variety of linguistic registers s/he can manipulate and his/her ability to switch between them are highly correlated to social status. This observation essentially stems from Bernstein s classical concepts of restricted and elaborated codes (Bernstein, 1971). Often misunderstood as the substantial difference between the language of the working class and that of the middle and upper classes, this distinction is in fact related to language use. A restricted code occurs more frequently in informal linguistic interactions between people of similar social and cultural backgrounds, whereas an elaborated code is better adapted to more formal situations, such as the kind of linguistic interactions that are typically required at school or in professional and official settings. Given that these kinds of social interactions are highly variable, this distinction partly overlaps with the contrast between working class language, on the one hand and middle and upper class language, on the other. Although the kind of formal interactions that require an elaborated code are more common to service class

10 occupations (i.e. professional, administrative and managerial employees) than they are to working class jobs, what distinguishes the latter from the former is perhaps that members of latter group are far more likely to cross the boundaries that separate both types of code. The ability to use an elaborated code and to switch between an elaborated code and a restricted code therefore appear as distinctive attributes of social power. It is in this sense that the parallel between sociolinguistics and the sociology of culture can, together, inform our understanding of cultural eclecticism, which can then be better understood in terms of culture switch (DiMaggio, 1987). Just as members of the upper and middle classes appear more linguistically mobile than their working class counterparts - in terms of their ability to code-switching - they are also better able to engage with a wide variety of cultural practices. This switching, whether linguistic or cultural, usually has associations with the characteristics of people s social networks. The more diverse the social ties, the more people are encouraged to engage in a greater variety of cultural and linguistic practices, as social diversity requires the mobilization of cultural resources suited to each social interaction (DiMaggio, 1987; Erickson, 1996; Relish, 1997). Moreover, to the extent that people tend to interact more frequently with people of lower status than themselves and less frequently with people of higher status, the breadth and diversity of the social network is itself correlated with social status: the higher the social status is, the higher the opportunity to engage with more socially and culturally diverse people in everyday life (Marsden, 1987; Lin and Dumin, 1986). This is also probably one of the reasons why the upper class tends to engage in more diverse cultural repertoires and linguistic codes than members of other classes. But linguistic code-switching as well as cultural mobility are not synonymous of an equal valuation of all the repertoires. In linguistic matters, it has even been shown that all situations where the symbolic hierarchy of languages was denied by those who mastered the dominant language were paradoxically reinforcing this hierarchy. A famous example of this is given by Pierre Bourdieu regarding an official ceremony during which the mayor of the city of Pau, in southern France, briefly addressed the audience in the local language, Béarnais. According to Bourdieu, far from lessening the hierarchy of language registers, this apparent transgression subtly contributed to its reinforcement. This subversion of the objective hierarchy between different linguistic varieties is actually limited, Bourdieu tells us, to those who master the dominant language well enough to keep this subversion from being interpreted as resulting from ignorance (Bourdieu, 1982: 63). The same is likely to occur in the transgression of symbolic boundaries between cultural registers. In a book published in the late 1980s, Claude Grignon and Jean-Claude Passeron argued along these lines when describing the incursions of upper class members into popular culture as a kind of symbolic droit du seigneur that ensures the dominant members their cultural domination, and makes reverse incursions of the lower classes into highbrow culture extremely unlikely (Grignon and Passeron, 1989). More generally, diversity is not necessarily synonymous of an equal valuation of all the tastes people express and all the practices in which the get involved. Liking a wide variety of musical genres or kinds of movies, for example, does not necessarily mean liking all of them to the same degree (Lahire, 2004). In other words, crossing the boundaries between cultural repertoires does not necessarily obliterate boundaries and, even more, hierarchies.

11 Be that as it may, the parallel established between language code-switching and cultural omnivorousness does help inform the social process by which openness to cultural and linguistic diversity become an upper class social privilege. Indeed, these two manifestations of an arguably common structuring principle may be linked to the structure of assets that people control, of which those related to cosmopolitanism in its broader sense are particularly noteworthy. 4. Openness to diversity and cosmopolitan capital The emergence of the omnivore/univore pattern of cultural distinction cannot be separated from a larger process that tends to value cultural diversity and strengthen the role of the assets that give access to cosmopolitanism and that are understood both as a set of values, attitudes or dispositions, as well as practices and skills (Vertovec and Cohen, 2002). Considering the attitudes, values and dispositions, modern understanding of this notion highlights the simultaneous recognition of similarities (i.e. universalistic humanism) and differences (i.e. multiculturalism) and the combination of tolerance and understanding of otherness (Szerszynski and Urry, 2002; Beck and Grande, 2007; Pichler, 2008). In the context of practices and skills, the focus is mainly on the cultural resources that allow social achievement in a globalized world (Calhoun, 2002 & 2007). Multicultural capital is in fact an increasingly valued form of cultural capital in the class structure of many contemporary societies (Bryson, 1997). Thus, the rising norm of cultural eclecticism and an increasing openness to diversity cannot be separated from the social change brought about by the process of globalization which, in many contemporary societies, serves to alter existing social structures and hierarchies. In such a context, the omnivore/univore, open/closed, global/local patterns of cultural distinction become increasingly prevalent. However, it goes without saying that what may be termed cosmopolitan resources are still predominantly controlled by the upper class. In his famous lectures on social class from the 1930s, Maurice Halbwachs noted that cosmopolitanism, and especially its linguistic components, was a traditional attribute of the aristocracy (Halbwachs, 2006). What we are seeing, then, in contemporary France is probably more the intensification of pre-existing tendencies than a complete renewal of class and power relations. The result of this is a rising contrast between mobiles and nonmobiles (i.e. those able to live at the global level and those restricted to the local). This contrast, highlighted by Merton (1968) in his characterization of local and cosmopolitan influences, appears today to be clearly structured by an underlying class gradient. In contrast to the lifestyle of the cosmopolitan elites, the French working class existence has often been characterized by local resources and local ties, sometimes called autochthonous capital (Retière, 2003; Renahy, 2010). Of course, in a context where globalization brings forth large migratory flows from among the working class in developing countries, international mobility is not as such a property of the elite. What does makes it an intrinsic characteristic of the elite class, however, is their capacity to transform it into a social asset that constitutes specific types of social and cultural capital (Wagner, 1998 and 2010). The positive experience of controlled cosmopolitanism can thus be contrasted with the negative experiences of uprooting. For example, there is no doubt that bilingualism and more specifically speaking English is today highly encouraged among the French elite and is more broadly promoted as a positive

12 asset among the upper class. Conversely, when it comes to immigrant workers and especially their children speaking another language (i.e. their native language) is more often seen as an obstacle to their integration in French society (Wagner, 2010). The unequal distribution of the positive attributes of globalisation and mobility is also particularly salient when considering foreign holidays and travel, which are still significantly more common among the upper class (Wagner, 2007). Cosmopolitanism is promoted among the upper class in various ways, including the early promotion of the virtues of travelling during childhood or the transmission of cosmopolitan skills in education (Weenink, 2008). This is especially the case at elite French schools such as Sciences Po and the most prestigious business schools (Wagner, 2007). It should be noted, however, that most of these cosmopolitan resources, and especially their linguistic components, cannot be acquired and transmitted only at school. Indeed, foreign language fluency has more to do with diffusion and repeated exposure to informal interactions than with scholastic training. Consequently, the symbolic and cultural inequalities that come hand in hand with unequal access to these resources cannot easily be compensated for by the education system and, more broadly, by the aforementioned cultural goodwill of the upwardly mobiles. Moreover, the social value attached to cultural and symbolic resources that are imperfectly accessible through school achievement strongly reinforces the nonegalitarian structure of class relationships. In a sense, exposure to informal situations conducive to the acquisition of these skills involves resources that are not only cultural, but also social and economic. The same largely holds true for cultural omnivorousness given that the likelihood of experiencing cultural otherness is linked to the same kinds of assets, such as, for example, travel opportunities. The specific nature of multicultural and cosmopolitan resources is also apparent in that control over this kind of capital is not strictly correlated with possessing more traditional cultural assets, as displayed in Figure 2. Therein, the reading habits of individuals by occupation are cross-plotted against their proficiency in English, which is defined by the ability to read a newspaper, hold a conversation, write letters and make a phone call in English.

13 Figure 2: Cross-plot of avid readers (i.e more than one book per month) and proficiency in English by social category % of English Speakers Engineers Technicians Senior Managers (Private Sector) Self-employed Professionals Middle-Ranking Managers (Private Sector) Business Owners Clerks (Private Shopkeeper Sector) Policemen & Soldiers Foremen Salesclerks Secondary School Teachers and Academics Senior Managers (Public Sector) Middle-Ranking Managers (Public Sector) Primary School Teachers Clerks (Public Sector) Crafstmen Domestic Employees Manual Workers Farmers % of Avid Readers Source: Insee, Enquête permanente sur les conditions de vie des ménages, The points represented on the graph correspond to the mean percentages observed for each indicator in respect of 20 different social categories. The graph displays an imperfect correlation between both types of indicator. For example, 13% of farmers are so-called avid readers (i.e. they read more than one book a month) and 2% are English speakers (they have at least one of four aforementioned basic abilities in English). More specifically, the graph shows They show that the categories with the highest percentage of so-called of avid readers (i.e. people reading more than one book per month) are not necessarily those whose members show greatest proficiency in English. If the percentage of avid readers may be considered an indicator of the volume of traditional or literate cultural capital and that measuring proficiency in English a measure of cosmopolitan cultural capital, then the graph reveals a far from perfect correlation between these two types of capital, a trend which is particularly accentuated in the upper and middle classes. For example, engineers and private sector senior managers appear to have more cosmopolitan than literate cultural capital, whereas teachers, academics and public sector senior managers display the reverse pattern. Technicians and, to a lesser extent, middle-ranking private sector managers contrast similarly with primary school teachers and middle-ranking public sector managers.

14 Conclusion Much discussion and debate surrounding Bourdieu s main sociological concepts, especially those developed in Distinction (Bourdieu, 1984 [1979]), raise the issue of their transferability beyond the context of the French sixties. Some authors argue that, far from being universal, Bourdieu s theory could be rather idiosyncratic (Lamont, 1992). Similarly, one might question the scope of the rising norm of eclecticism identified by Richard Peterson among American elites in the early nineties. However, twenty years later, Peterson s original claim of the growing distinctive power of omnivorousness considered as the appreciation of all distinctive leisure activities and creative forms along with the appreciation of classic fine arts (Peterson 1992: 252) is notably supported by a lot of empirical evidence, even in Bourdieu s own country. But what is presently at stake is the social implication of this process. What I suggested in the above paragraphs is that the omnivorization of taste and practices does not really challenge Bourdieu s framework, to the extent that: a) eclecticism does not fully obliterate the prevalent highbrow/lowbrow divide; b) eclecticism can really correspond to a generic disposition toward culture, and not only a loose mixing of highbrow and mass-culture items; c) this emerging norm is far from being synonymous of an obliteration of cultural boundaries and hierarchies. In that sense, the parallel established between code-switching and cultural omnivorousness suggests that openness to cultural diversity and aesthetic eclecticism, as well as the ability to handle various linguistic registers and language varieties, may correlate with similar latent dispositions on the part of the upper classes to engage with many different fields of practice (language, taste, political opinion, mores, values, consumption patterns, etc.), in line with Bourdieu s concept of habitus (Bourdieu, 1980). But it also suggests that these dispositions are context-specific (Emmison, 2003, 213), namely that what is at stake is the ability to mobilize the right code or repertoire in the right context. Thereby, just like code-switching does not mean code-mixing, cultural eclecticism does not mean cultural syncretism. Similarly, the parallel established between cosmopolitanism and cultural eclecticism suggests that what is really socially distinctive is not the handling of a variety of cultural assets per se. What is socially distinctive is the ability to consider diversity as a positive experience and to manage eclectic resources as efficient assets in social life. In the same way, considerations on the impact of social mobility on musical taste illustrate the strong symbolic power that the classic definition of cultural legitimacy still exerts on people beliefs and practices. It also suggests that omnivorousness is underlain by an ability to take some distance from this classic definition that the core members of the upper class (i.e. non-mobiles) are far more likely to master than the newcomers in this class. One possible explanation could be that conformist orientations toward the conservatively highbrow cultural norm might be more easily acquired in the school environment than more inclusive definitions of cultural legitimacy, which rely on more informal resources. Therefore, a major misconception of the rising eclecticism of the elites would be to interpret it as a sign of a declining role of cultural capital and as an indicator of an on-going cultural democratization.

15 References Atkinson, W. (2011). The context and genesis of musical tastes: Omnivorousness debunked, Bourdieu buttressed. Poetics 39(3): Beck, U., and Grande, E. (2007). Cosmopolitan Europe. Cambridge: Polity Press. Bernstein, B. (1971). Class, Codes and Control. Volume 1: Theoretical Studies Towards a Sociology of Language. London, Routledge & Kegan Paul. Bourdieu, P. (1980). The Logic of Practice. Stanford, CA: Stanford University Press. Bourdieu, P. (1982). Ce que parler veut dire: l économie des échanges linguistiques. Paris: Fayard. Bourdieu, P. (1984 [1979]). Distinction. A Social Critique of the Judgement of Taste, Harvard University Press. Bourdieu, P. and J. C. Passeron (1979 [1964]). The inheritors : French students and their relation to culture. Chicago, University of Chicago Press. Bryson, B. (1996). Anything But Heavy Metal : Symbolic Exclusion and Musical Dislikes. American Sociological Review: Bryson, B. (1997). What about the univores? Musical dislikes and group-based identity construction among Americans with low levels of education. Poetics, 25(2), Calhoun, C. (2007). Nations matter: Culture, history and the cosmopolitan dream. London: Routledge. Calhoun, C. J. (2002). The class consciousness of frequent travelers: toward a critique of actually existing cosmopolitanism. The South Atlantic Quarterly,101(4), Chan, T. W. and J. H. Goldthorpe (2007). Social stratification and cultural consumption: Music in England. European Sociological Review 23(1): Coulangeon, P., and Lemel, Y. (2009). Les pratiques culturelles et sportives des Français : arbitrage, diversité et cumul. Economie et statistique, 423(1), Daenekindt, S. and H. Roose (2013). A Mise-en-scène of the Shattered Habitus: The Effect of Social Mobility on Aesthetic Dispositions Towards Films. European Sociological Review 29(1): DiMaggio, P. (1987). Classification in art. American Sociological Review, 52(4), DiMaggio, P., & Useem, M. (1978). Social class and arts consumption. Theory and Society, 5(2), Donnat, O. (1994), Les français face à la culture: de l'exclusion à l'éclecticisme. Paris, La découverte Emmison, M. (2003) Social Class and Cultural Mobility Reconfiguring the Cultural Omnivore Thesis. Journal of Sociology 39(3), Erickson, B. (1996). Culture, class, and connections. American Journal of Sociology, 102(1), Festinger, L. (1957). A theory of cognitive dissonance, Stanford University Press. Gartman, D. (1991). Culture as class symbolization or mass reification? A critique of Bourdieu s distinction. American Journal of Sociology: Glévarec, H. and M. Pinet (2012). Tablatures of musical tastes in contemporary France: distinction without intolerance. Cultural Trends 21(1): Grignon, C., & Passeron, J. C. (1989). Le savant et le populaire. Paris: Gallimard. Halbwachs, M. (2006). Les classes sociales. Paris: Presses Universitaires de France. Lahire, B. (2004). La culture des individus: dissonances culturelles et distinction de soi. Paris: La Découverte.

16 Lahire, B. (2008). The individual and the mixing of genres: cultural dissonance and self-distinction. Poetics, 36(2), Lahire, B. (2010 [2001]). The plural actor, Polity. Lamont, M. (1992). Money, Morals, and Manners: The Culture of the French and the American Upper-Middle Class, Chicago: The University of Chicago Press. Lin, N., and Dumin, M. (1986). Access to occupations through social ties. Social networks, 8(4), Marsden, P. V. (1987). Core discussion networks of Americans. American Sociological Review, 52(1), McCutcheon, A. L. (1987). Latent class analysis, Sage Publications, Incorporated. Merton, R. K. (1968). Social theory and social structure. New York, Free Press. Mohr, J. and P. DiMaggio (1995). The intergenerational transmission of cultural capital. Research in social stratification and mobility 14: Peterson, R. A. (1992). Understanding audience segmentation: From elite and mass to omnivore and univore. Poetics 21(4): Peterson, R. A. (1997). The rise and fall of highbrow snobbery as a status marker. Poetics 25(2-3): Peterson, R. A. and R. M. Kern (1996). Changing highbrow taste: from snob to omnivore. American Sociological Review: Peterson, R. A., and Simkus, A. (1993). How musical taste groups mark occupational status groups. In M. Lamont and M. Fournier (Eds.), Cultivating differences: Symbolic boundaries and the making of inequality (pp ). Chicago: University of Chicago Press. Pichler, F. (2008). How Real Is Cosmopolitanism in Europe? Sociology, 42(6), Relish, M. (1997). It s not all education: Network measures as sources of cultural competency. Poetics, 25(2-3), Renahy, N. (2010). Classes populaires et capital d autochtonie. Genèse et usages d une notion. Regards Sociologiques (40), Retière, J. N. (2003). Autour de l autochtonie. Réflexions sur la notion de capital social populaire. Politix, 16(63), Sullivan, O., and Katz-Gerro, T. (2007). The omnivore thesis revisited: Voracious cultural consumers. European Sociological Review, 23(2), Szerszynski, B., and Urry, J. (2002). Cultures of cosmopolitanism. The Sociological Review, 50(4), Van Eijck, K. (1999). Socialization, education, and lifestyle: How social mobility increases the cultural heterogeneity of status groups. Poetics 26(5): Vermunt, J. K. and J. Magidson (2005). Latent GOLD 4.0 User s Guide. Belmont, Massachusetts, Statistical Innovations Inc. Vertovec, S., and Cohen, R. (Eds.). (2002). Conceiving Cosmopolitanism: Theory, Context, and Practice. Oxford: Oxford University Press. Wagner, A. C. (1998). Les nouvelles élites de la mondialisation: une immigration dorée en France. Paris: Presses universitaires de France. Wagner, A. C. (2007). La place du voyage dans la formation des élites. Actes de la recherche en sciences sociales(5),