Academic capital and music tastes among Swedish adolescents

KEITH ROE

Theoretical background

Following the work of Bourdieu & Passeron (1977, 1979) and Bourdieu (1984), considerable attention has been devoted to the role of education in social and cultural reproduction. While Bourdieu's empirical focus has been on France, the heuristic value of his model has been directly or indirectly demonstrated in studies from countries as diverse as the United States (DiMaggio, 1982; Farkas et al. 1990), Greece (Katsillis & Rubinson, 1990), and Sweden (Roe, 1983, 1987, 1990, 1992). The purpose of this article is to contribute to the growing body of empirical evidence on the relationship between school achievement and cultural reproduction. One of the starting points of Bourdieu & Passeron's analysis is the need to relate cultural choices to the social conditionings of their emergence and perpetuation. Cultural behaviours are seen as determined less by individual tastes than by social determinisms and it is the contribution of pedagogic action to these determinisms which forms the dynamic core of their reproduction model. They argue that classical theories tend to ignore the effect of symbolic relations in the reproduction of power relations because they sever the inter-relationships between cultural and social reproduction. This results from the implicit premise that what pedagogic action aims to reproduce is a common cultural capital shared by a whole society whereas

In reality, because they correspond to the material and symbolic interests of groups or classes differently situated within the power relations, this pedagogic action tends to reproduce the structure of the distribution of cultural capital among these groups or classes, thereby contributing to the reproduction of the social structure. (Bourdieu & Passeron, 1977:11)

While they acknowledge that the process of cultural reproduction begins in the family, Bourdieu & Passeron argue that the influence of social advantages and disadvantages on educational careers and cultural life is cumulative. Thus, the habitus acquired in the family structures the reception of classroom messages and the habitus acquired at school conditions the reception of the messages of the culture industries. This implies that, while the school remains the only realistic path to culture for children from lower socio-economic status (SES) backgrounds, it involves being subjected to a process of acculturation, one of the effects of which is the recognition of what constitutes 'legitimate' knowledge and the devaluation of the knowledge which children from the dominated classes possess (ibid, 1977:42; 1979:21-22).

Furthermore, they argue, an implicit prerequisite for success in some subjects is the possession of extra-curricular attributes, such as knowledge, tastes and 'good taste', which only students from privileged backgrounds have. Such cultural privilege is manifested in the familiarity with cultural works which only frequent direct contact can give and,

It is still more manifest in the case of those works, generally the most modern ones, which are the least....scholastic. In every area of culture in which it is measured - be it the theatre, music, painting, jazz, or the cinema - students have a richer and more extensive knowledge the higher their social origin. (ibid, 1979:17)

and,

Far from constituting a parallel, alternative, or compensatory culture, knowledge of cinema or jazz varies in direct proportion to familiarity with the traditional arts....the groups most integrated into the academic universe, and at the highest levels, should have the best scores in jazz and cinema as well as in other fields. (ibid, 1979:41)

This does not mean that the determination of cultural behaviours in the model is mechanical; students from privileged backgrounds may squander their cultural inheritance and some students from less privileged backgrounds are able to overcome their cultural disadvantages as a result of exceptional ability and certain features of their family background. It follows that students are irreducible solely to their class of origin and are better defined by their current relation to it, i.e. in terms of their trajectory in the social structure (ibid:23-25; cf. Rosengren & Windahl et al. 1989)

Thus, Bourdieu's empirical analyses demonstrate, on the one hand, a close relationship between cultural practices, level of education and, secondarily, social origin; on the other hand, at equivalent levels of educational capital, that the weight of social origin increases the further one moves from the most legitimate areas of culture, becoming strongest in avant-garde culture. The relative weight of home background and formal education is found to vary according to the extent to which different cultural practices are recognized and taught by the educational system. The more the competences measured are recognized by the school system, and the more academic the techniques used to measure them, the stronger the relation found between performance and educational qualification. The strongest correlations are obtained from questions about composers of musical works and,

nothing more clearly affirms one's 'class', nothing more infallibly classifies, than tastes in music. (ibid:14ó18)

These classifications are based on a general transposable disposition towards legitimate culture, the cultivation of which, to a variable extent according to students' social background, is largely achieved by the school by means of its function of inculcating and imposing values and by inscribing into the academic qualification a guarantee not only of formal attainment, but also that the student has accumulated the types of cultural accomplishment that are appropriate to the prestige of the qualification. This synchronization is achieved by formal and informal sorting mechanisms which directly and indirectly manipulate students' self-image and self-esteem, and thereby also their aspirations, so that students are channelled towards prestigeous or low status positions which are more or less associated with legitimate practice.

Nevertheless, while variations in educational qualifications may be closely related to variations in cultural competence, the relationship between educational qualifications and cultural practices cannot be explained solely by the operation of the educational system. Academic capital is seen as the guaranteed product of the interaction of cultural transmission by the family and by the school. Thus, Bourdieu finds that, at equivalent levels of academic capital, differences in social origin are associated with differences in taste and competence, and that, at equivalent levels of socio-economic status, such differences are also associated with differences in academic capital.

A general opposition is found within each social class between the fractions richest in cultural capital and poorest in economic capital and those richest in economic capital and poorest in cultural capital. Those who have acquired the bulk of their cultural capital in and for school are found to have more 'classical', safer, cultural investments than those who have received a large cultural inheritance. This is apparent particularly in the case of the relationship which the lower middle class have with culture, especially their propensity to accumulation of knowledge, and the considerable gap manifested by them between this knowledge and cultural recognition (ibid:39, 63-65, 331).

In this article the efficacy of the prediction that academic capital will be related to cultural knowledge and preferences will be tested empirically with respect to adolescents' liking for and knowledge of various types of music. Since the model predicts that academic capital will be related even to modern areas ostensibly outside of the definition of legitimate scholastic culture, the analysis embraces not only classical music and jazz, but also pop and rock genres. This is also motivated by the fact that there is evidence to suggest that some forms of rock may now be going through a process of cultural legitimation. For example, Trondman (1990) argues that tastes in rock now mark social distance between classes and are related to the establishment of social hierarchies.

Method

The data presented here were collected by means of questionnaires administered to 1,334 Swedish adolescents, aged 15-16 years, in the ninth year of school. This is the final year of compulsory schooling, although most students continue with some form of education thereafter. The overall response rate was 98%. The data were collected in 1984. As such, they are obviously in no way intended as a description of the current music tastes of adolescents. The heuristic value of the model lies in the prediction that tastes and knowledge will reveal structural regularities which will be related to the the activities of the educational system. The theoretical classes are therefore, 'constructed for explanatory purposes and do not necessarily exist as such as groups in reality' (Bourdieu, 1990:117).

The grades awarded to each pupil at the end of the autumn term of the 9th year were obtained from school records. These grades were awarded 3-4 months before the administration of the questionnaire. Each student was awarded a grade from one (lowest) to five (highest) for each subject studied. In these analyses each students' overall grade average was employed.

Father's occupation is employed here as the indicator for socio-economic status. These data were coded according to a classification employed by the Swedish Central Statistical Bureau and are here grouped into five categories: lower (unskilled) working class, upper (skilled) working class, lower middle class, middle class, and upper middle class.

The music preference data were obtained from respondents' numerical ratings of music categories on a scale from 0 to 10. Eight music types are analyzed here: pop, disco, synth (synthesizer rock), rock, heavy metal rock, classical, jazz and blues. The music categories were specified in consultation with a major record store in Lund and, in the questionnaire, two examples of artists/composers were given for each category.

The music knowledge data were collected by means of three items designed to measure how much respondents knew about a wide range of music forms from classical to heavy metal rock. The first asked them to name one composer or artist associated with each of the following kinds of music: jazz, blues, folk music, classical music, opera and ballet. The second asked respondents to name 5 artists/bands in the current Swedish top 20. The responses were coded from 0-5 correct answers, with reference to the published hit list. The third asked them to classify 16 rock bands into the appropriate rock categories. The responses were coded from 0-16 correct answers.

Results

Music tastes

The results of an analysis of means of the expressed liking for eight kinds of music by school achievement, for each category of father's occupation, are summarized in Figures 1-5. In each figure the vertical axis represents liking for the music types and the horizontal axis the level of school achievement. The analysis was based on 1171 respondents.

The results for lower working class adolescents are summarized in fig 1. For each type of music except one, high and very high achievers from this social background expressed a higher mean level of liking than did low and very low achievers. The exception is liking for heavy metal which runs in the opposite direction. The steepest curves in a positive direction are those for synth and classical music, followed by disco. Pronounced positive linear tendences can also be traced for pop and, to a lesser extent, jazz. The flatest curve is that for rock, although it is essentially similar in direction to those for pop and jazz. Conversely, liking for heavy metal, by far the highest rated music among the very low achievers, falls steadily with each category of achievement. Among lower working class very low achievers, then, classical music, jazz and blues are liked least; pop, synth, disco and rock occupy intermediate positions; and heavy metal is by far the most liked. By comparison, among the very high achievers, classical music attains fifth position and surpasses liking for heavy metal, as well as jazz and blues.

Figure 1. Liking for varoius kinds of music by school achievement (means). Lower working class adolescents (n=254)

The results for upper working class adolescents are presented in fig 2. In general terms they are similar to those for the lower working class adolescents: Once again, the tendency of all the curves, except that for heavy metal, is in a positive direction, i.e the values for very high achievers are higher than those for very low achievers; the curves for classical music, rock, blues and jazz are similar to those in fig 1; and classical music surpasses heavy metal among the very high achievers to achieve fifth ranked place. However, there are also some noteworthy differences from the previous figure. Heavy metal is not the most liked type of music among very low achievers; the steepest curve is for pop rather than synth, so that pop rises from fifth most liked among very low achievers to second most liked among very high achievers; and the direction of the curve is interrupted between low and average achievers for heavy metal, rock and pop, an anomoly for which there is no ready explanation.

Figure 2. Liking for various kinds of music by school achievement (means). Upper working class adolescents (n=283)

Figure 3. Liking for various kinds of music by school achievements (means). Lower middle class adolescents (n=279)

Since, according to Bourdieu, the lower middle class has the most problematic relation to culture, greater irregularity is to be expected here than in the other categories. This prediction receives some support from the results shown in fig. 3. Firstly, although the mean levels of liking for most of the music types are once again higher for the very high achievers than they are for the very low achievers, there are here two exceptions rather than one with both heavy metal and rock sloping in a negative direction and starting and finishing in close proximity. Secondly, with the exception of synth, the slope of the curves is generally flatter here than in any of the other figures. Thirdly, in five of the eight music types (synth, pop, rock, classical and jazz) the means for the very high achieving group are appreciably lower than those for either the high or the average achieving group; a pattern which was found in no other class category and which was particularly striking in the case of rock and classical music. Indeed, in no other social class category was the mean for classical music lower among very high achievers than high achievers, let alone average achievers.

Figure 4. Liking for various kinds of music by school achievements (means). Middle class adolescents (n=194)

Figure 5. Liking for various kinds of music by school achievements (means). Upper middle class adolescents (n=161)

A number of points distinguish the results for middle class adolescents which are presented in fig. 4. First, the zero rating given to classical music, blues and jazz by the very few very low achievers in this category. Second, as with the lower middle class, all the curves slope in a positive direction except heavy metal and rock. Third, for the first time the steepest sloping curve is that for classical music, especially from the high to the very high achievers (among whom it ranks fourth, above e.g rock). Liking for disco also increases steeply between these two achievement groups. Finally, while lower middle class very high achievers rated five kinds of music lower than did high achievers, middle class very high achievers rated synth, disco, pop, classical and rock higher than their high achieving counterparts.

The results for upper middle class adolescents are presented in fig. 5. Note first that there were no very low achievers from this background. Second, as with the other two middle class categories the direction of all the curves is positive except those for heavy metal and rock. As in the middle class, the steepest curve is the one for classical music; although here the next steepest curve is the negative one for heavy metal, which moves from second highest position among low achievers to bottom position among very high achievers. Note too that, in this group, liking for disco drops as we move from upper middle class high to very high achievers, the only other category where this occurred being the upper working class.

In summary, some clear patterns may be discerned. First, the most popular kind of music overall was synth. Only among lower working class very low and low achievers, lower middle class very low achievers, and middle class low achievers, was synth not the most popular type of music. Synth received very high mean ratings from average and above average achievers in particular. Disco was the second most popular type of music, particularly in the above average achievement groups.

Second, the direction of the curves for heavy metal was negative in all five social class categories. Thus, although liking for heavy metal was at its greatest among lower working class very low achievers, it was also either the most liked or second most liked type of music in all the other very low achieving groups, even the middle and upper middle class ones. Moreover, with respect to heavy metal, the differences between the class categories were smaller than those within each class category, e.g. very high achievers from the lower working class are closer to very high achievers from the upper middle class than they are to low or very low achievers from their own background - supporting the hypothesis that school achievement helps to structure music tastes. It is also worth noting that, among the very low achievers, by far the largest difference with respect to heavy metal was that between the lower working and upper working class groups, a result also supporting the view that, within the broad social classes, tastes in music vary appreciably in different educational groups - in this case presumably because upper working class adolescents doing badly at school try to distinguish themselves, by means of musical style, from their lower working class counterparts.

Third, in contrast to heavy metal, the curves for classical music were generally positive in all social class groups. With the exception of the lower middle class very high achievers, there was a steep rise in liking classical music in all the above average achieving groups, including those from the working classes. Consequently, as with heavy metal, the differences within the social classes are greater than those between classes, e.g. lower working class very high achievers are more like upper middle class high achievers with respect to liking classical music than they are to average or below average achievers from their own background. The group liking for classical music most of all was not upper middle class very high achievers, but very high achievers from the middle class, perhaps indicating an element of aspiration. Note too that, as predicted, the pattern of liking for jazz, and blues, was generally similar to that for classical music, especially among the lower working class, lower middle class and middle class below average achievers, upper working class average achievers and lower middle class very high achievers. Overall, blues (followed by jazz) was liked least.

Finally, with respect to rock, the curves moved in a positive direction in the two working class groups and in a negative direction in the three middle class groups. Apparently rock means different things in these social class contexts, having a generally positive association with cultural legitimacy in the working class and a generally negative association in the middle classes. This result gives some support to Trondman's (ibid) thesis that, in contemporary culture, rock may be used for marking social distance, but it also ind

Knowledge of music

Pearson product moment correlation coefficients were computed between father's occupation, school achievement and knowledge of jazz, blues, folk music, classical music, opera and ballet. The results are presented in table 1. They show that while amount of music knowledge in each category is correlated significantly with both father's occupational status and school achievement, in each case the correlations for school achievement are higher. Moreover, when the scores for music knowledge are added togther, the resulting correlation coefficients are 0.18 with father's occupation and 0.40 with school achievement, an indication of the cumulative nature of the interrelationship between socio-economic background and school achievement.

Table 1. Correlations between knowledge of various types of music. By father's occupation and school achievement (Pearson's product moment coefficients).

Music type	Father's Occupation	School Achievement
Jazz	.15	.34
Blues	.05*	.12
Folk	.08**	.25
Classical	.18	.39
Opera	.13	.27
Ballet	.07*	.15
ALL	.18	.40
n = 1171

All coefficients sig. <.001 except:

** = <.01 * = <0.5

Table 2. Knowledge of music (combined) by father's occupation and school achievements (means).

This interrelationship becomes more apparent in the mean level of knowledge of these types of music by each social class and school achievement group (table 2). Knowledge increases according to both socio-economic background and school achievement but is at its lowest among very low achievers from the lower middle class and lower working class, and at its highest among very high achievers from lower middle class and upper middle class backgrounds. That lower middle class very high achievers should have the highest level of knowledge of any group is at first sight surprising considering the fact that the results in fig 3. indicated that, of the very high achievers, it was precisely this lower middle class group that had by far the lowest mean level of liking for classical music. That this group displays knowledge about cultural forms which they do not like supports Bourdieu's view concerning their cultural insecurity and propensity to invest in accumulating knowledge of legitimate forms of culture.

These results, then, support the postulate that knowledge of these more 'legitimate' forms of music increases by both SES and academic status. The mean level of knowledge was lowest among lower working class adolescents and highest among upper middle class adolescents. At the same time, in every socio-economic category, knowledge was greater among above-average achievers than it was among below-average achievers.

Knowledge of the top 20 was also positively correlated with SES and school achievement and, as for the music types analyzed above, the correlation with achievement was the stronger. In other words, both higher SES and higher achieving pupils tended to know more about the top 20. Moreover, an analysis of means revealed that the relationship between school achievement and top 20 knowledge was linear: Very low achievers had a mean of 1.16 correct answers, low achievers 2.17, average achievers 2.80, high achievers 3.21 and very high achievers 3.23 (Grand mean 2.7, S.D 2.0). There was also a strong positive correlation between knowledge of the top 20 and knowledge of the more legitimate music forms dealt with above i.e those who knew more about the one, tended also to know more about the other. Unlike the results for knowledge of the other forms of music, knowledge of rock was not significantly related to school achievement. However, there was a weak tendency for those from higher status backgrounds to know more about it than those from lower status backgrounds. There was also a moderately strong positive correlation between knowledge of classical music, opera, ballet etc. and knowledge of rock, indicating that those knowing more about the one sphere tended also to know more about the other. With the exception of the lack of correlation between knowledge of rock and school achievement, all the music knowledge results are in line with the predictions of Bourdieu's model.

Summary of results

In terms of the theoretical model, the most significant results may be summarized as follows:

1. There was a generally positive relationship between school achievement and liking for classical music. The only significant aberration from the linear tendency was found among lower middle class very high achievers. Although less regular, similar results were obtained for synth, pop and disco.

2. The tastes expressed by high achievers from lower status backgrounds were more similar to those expressed by high achievers from higher status backgrounds than they were to the tastes expressed by low achievers from lower status backgrounds.

3. There was a generally negative relationship between school achievement and liking for heavy metal. In the three middle class groups there was a similar, though much less negative tendency for rock. Conversely, in the two working class groups there was a positive tendency between achievement and the level of liking for rock.

4. The tastes expressed by low achievers from higher status backgrounds were more similar to those expressed by low achievers from lower status backgrounds than they were to the tastes expressed by high achievers from higher status backgrounds.

5. Jazz and blues manifested patterns closer to that of classical music than to those of pop or rock.

6. Knowledge of more legitimate forms of music increased with higher social status background.

7. Knowledge of more legitimate forms of music increased with higher school achievement.

8. The association between school achievement and music knowledge was stronger than that between social background and music knowledge.

9. The mean level of knowledge of all the types of music analyzed was especially high among middle and upper middle class high and very high achievers.

Discussion

The view that educational status is related to the cultural orientations of individuals and groups is not new: Sorokin noted that, 'The educational status of a person is one of the coordinates of his personality and position in the socio-cultural universe' (1947:232). Since then direct or indirect evidence of links between educational status, subcultural attachment and music tastes has been steadily accumulating (Coleman, 1961; Stinchcombe, 1964; Hargreaves, 1967; Sugarman, 1967; Willis, 1977; Frith, 1983; Bourdieu, 1984; Fornäs et al. 1988; Roe, 1983; 1987; 1990; 1992; Roe & Löfgren, 1988).

A theoretical model of the dynamics of these relationships is provided by Bourdieu and Passeron (ibid). Starting with the assumption that individuals do not randomly move about in social space, they postulate that education, building on the habitus acquired in the home, is the major mediating factor in the differentiation of taste. This mediation makes it possible for some students from higher status backgrounds to be assigned a low educational status and for some students from lower status backgrounds to be assigned a higher status. It is in this way that cultural tastes and knowledge come to reflect the status relations of different groups, and individuals come to experience different tastes as either attractive or repulsive.

As Bourdieu points out (1990:131), this experience is analogous to the development of a 'sense of one's place' (and of the place of others) whereby individuals learn, subjectively, that, 'this is (or is not) for me', to locate a position in social space through a 'stance taken in symbolic space'. In this way the symbolic space and the space of social positions may be conceived of as the independent but homologous spaces within which social and cultural reproduction is achieved (ibid).

This study supports Bourdieu's assertion that the sociology of culture is inseparable from a sociology of education (and vice versa). The mechanics of academic competition create not only academic differences but even cultural differences that can last a lifetime. Whether this is acceptable, or whether measures should be taken to change the rules and procedures under which schools perform the functions which society has assigned them is ultimately a political choice. However, the ways in which the system operates in these respects needs to be made explicit so that those involved in education are forced to acknowledge that instituting academic classifications is neither purely academic, nor based on wholly objective differences, but rather involves imposing social definitions and identities which have distinct, and sometimes undesirable, cultural and behavioural outcomes.

As Foucault (1977:181-2)) has argued, hierarchical ranking serves, in itself, as a reward or punishment. By distributing pupils in a hierarchy, a constant pressure is exercised over them to conform to the pedagogic authority of the educational institution by displaying qualities such as subordination, docility, and attention to more or less arbitrary studies and exercises. In this unequal power struggle some students play by the rules and reap the academic and cultural rewards which conformity brings; the majority become bored and apathetic; and some react with a fierce antagonism to the institution and the culture which it is perceived to represent, and in so doing collaborate in consigning themselves to the future which has been designated for them. One way of expressing this antagonism is to flaunt provocative symbols and activities - and for that purpose some forms of rock have always provided a rich reservoir of possibilities.

References

BOURDIEU, PIERRE (1984) Distinction: A social critique of the judgement of taste London: Routledge & Kegan

BOURDIEU, PIERRE (1990) In other words: Essays towards a reflexive sociology Cambridge: Polity.

BOURDIEU, PIERRE & JEAN-CLAUDE PASSERON (1977) Reproduction in education society and culture London: Sage.

BOURDIEU, PIERRE & JEAN-CLAUDE PASSERON (1979) The inheritors: French students and their relation to culture University of Chicago Press.

COLEMAN, JAMES S (1961) The adolescent society Glencoe: Free Press.

DIMAGGIO, PAUL (1982) Cultural capital and school success: The impact of status culture participation on the grades of U.S. high school students American sociological review 47. April. 189-201.

FARKAS, GEORGE, ROBERT GROBE, DANIEL SHEEHAN & YUAN SHUAN (1990) Cultural resources and school success American sociological review 55. February. 127-142.

FORNÄS, JOHAN, ULF LINDBERG & OVE SERNHEDE (1988) Under rocken: Musikens roll i tre unga band Stockholm: Symposion.

FOUCAULT, MICHEL (1977) Discipline and punish: The birth of the prison Harmondsworth: Penguin

FRITH, SIMON (1983) Sound effects: Youth, leisure and the politics of rock'n'roll London: Constable.

HARGREAVES, DAVID H (1967) Social relations in a secondary school London: Routledge & Kegan

KATSILLIS, JOHN & RICHARD RUBINSON (1990) Cultural capital, student achievement, and educational reproduction: The case of Greece American sociological review 55. April. 270-279.

ROE, KEITH (1983) Mass media and adolescent schooling: Conflict or co-existence? Stockholm: Almqvist & Wiksell Int.

ROE, KEITH (1987) The school and music in adolescent socialization. James Lull (ed) Popular music and communication.Newbury Park: Sage.

ROE, KEITH (1990) Adolescents music use: A structural-cultural approach. Keith Roe & Ulla Carlsson (eds) Popular music research Gothenburg: NORDICOM.

ROE, KEITH (1992) 'Different destinies - different melodies: School achievement, anticipated status, and adolescents' music use' To appear in The European journal of communication Vol. 7. Autumn.

ROE, KEITH & MONICA LÖFGREN (1988) Music video use and educational achievement Popular music 7. 3. 303-314.

ROSENGREN, KARL ERIK & SVEN WINDAHL (1989) Media matter: TV use in childhood and adolescence Norwood: N.J. Ablex.

SOROKIN, PITIRIM A (1947) Society, culture and personality: Their structure and dynamics New York: Harper.

STINCHCOMBE, ARTHUR LEONARD (1964) Rebellion in a high school Chicago: Quadrangle.

SUGARMAN, BARRY (1967) Involvement in youth culture, academic achievement and conformity in school British journal of sociology 18. 151-164.

TRONDMAN, MATS (1990) Rock taste - on rock as symbolic capital. A study of young people's music, taste and music making Keith Roe & Ulla Carlsson (eds) Popular music research Gothenburg: NORDICOM.

WILLIS, PAUL (1977) Learning to labour London: Saxon House.

YOUNG Volume 1 Number 3 1993