Interpretive Approaches to the Sociology of Sport

Peter Donnelly. Handbook of Sports Studies. Editor: Jay Coakley & Eric Dunning. 2000. Sage Publishing.

Interpretation is the basis of all sociology, and all science. If I observe a recurring event, or discover a relationship between two variables, a statistical tendency, or empirical evidence of causality, such discoveries require interpretation. However, within the field of sociology, the term interpretive is used more narrowly to refer to a particular group of sociologies which have as their basis the interpretation and understanding of human meaning and action. Johnson notes that:

A sociological understanding of behaviour must include the meaning that social actors give to what they and others do. When people interact, they interpret what is going on from the meaning of symbols to the attribution of motives to others. (1995: 146)

Interpretive sociology represents, in large part, one of ‘the two sociologies’ (Dawe, 1970). In their task of exploring the relationship between the individual and society sociologists have divided, since the earliest days of sociology, between the ‘system’ approach and the ‘action’ approach. This division is captured well in Thompson and Tunstall’s question: ‘Do the two approaches of social systems and social action theory simply correspond to our own ambivalent experience of society as something that constrains us and yet also something that we ourselves construct?’ (1975: 476). Interpretive sociology fits clearly into the social action side of the divide, a position that is both its strength and its weakness.

Included in interpretive sociology are Weberian sociology, the ‘sociologies of everyday life’ (symbolic interactionism, Goffman’s dramaturgical sociology, labelling theory, phenomenological sociology, ethnomethodology, and existential sociology) (Douglas et al., 1980), and hermeneutics. Marshall (1994) notes that these sociologies differ in two ways. First, in the extent to which they view interpretation as problematic (p. 255)—Weberian sociology and symbolic interactionism take a relatively unproblematic (commonsense) approach to interpretation; phenomenology, ethnomethodology and hermeneutics developed more refined approaches. Second, in the degree to which they go beyond the actor’s own understanding of what he or she is doing (p. 255). As Jary and Jary note:

All social reality is ‘pre-interpreted’ in that it only has form as (and is constituted by) the outcome of social actors’ beliefs and interpretations. Thus it is, or ought to be, a truism that no form of sociology can proceed without at least a preliminary grasp of actors’ meanings. (1995: 336)

Thus, while Weberian sociology takes Verstehen (understanding) as its basis, and distinguishes between ‘descriptive’ and ‘explanatory’ understanding; and Alfred Schutz (phenomenological sociology) developed Weber’s work to distinguish between ‘because’ motives and ‘in order to’ motives; other interpretive sociologies (for example, existential sociology) assume that the actor’s own meanings are the basis for analysis, while the remainder (such as ethnomethodology, Goffman’s dramaturgy) focus more on discovery of the rules of social action and interaction.

Just as interpretive sociology is related to the social systems/social action debate in sociology, it is also connected with, but not congruent to, two other sociological debates—that between macro- and micro-sociology (with the sociologies of everyday life usually being equated with micro-sociology); and that between the so-called quantitative and qualitative methodologies. Since methodology is a key feature of the practice, and the critique, of interpretive sociology, it is important to address it here. Although ‘understanding’ is the key to Weber’s methodology (Weber, 1904-17, 1922), the actual methods of the sociologies of everyday life often approximate those of anthropology—primarily ethnography and in-depth interviewing. Hermeneutic methods are now likely to be termed ‘textual analysis,’ and overlap with ethnographic and interview work at the level of ‘discourse analysis.’ The quantitative—qualitative debate has been largely resolved now by interpretive sociologists who are likely to turn to, or collect appropriate forms of quantitative data (cf. Denzin’s (1970) strategy of ‘triangulation’; or Willis’s (1978) use of a cluster of methods—both qualitative and quantitative).

In sum, interpretive sociology may be defined by its opposite in that it ‘differs from the view that social life is governed by objective cultural and structural characteristics of social systems (external to individuals, and relatively independent of them); and from the view that it is possible to construct rigid scientific laws to explain patterns of social behaviour’ (Johnson, 1995: 146). Thus, interpretive sociology is concerned with the way in which the social world is not just something to be confronted by individuals, but is continually constructed and reinvented by the participants. The following sections deal with the emergence of this perspective in sociology and in the sociology of sport; with criticisms of the perspective and the responses of interpretive sociologists; with examples of sport-related research; and with an examination of the way in which that research has developed our knowledge of the social world in and beyond sport.

Emergence and Development of Interpretive Sociology

Modern sociology is grounded in classical philosophical ideas about the individual versus society. Douglas et al. summarized the key concepts of sociology in two interrelated questions:

First, can human actions be explained in terms of concrete individual factors (such as individual will, choice, or the concrete situations individuals face) or in terms of something outside of the individual (such as culture or social structure) that determines or causes what they will do? Second, are we to determine the answers to our first question by observing individuals in concrete situations of everyday life, or by observing something (such as a social structure), supposedly outside of the individuals, by experimentally controlled methods? (1980: 183)

Douglas suggests that psychology chose the individual perspective and the social sciences generally opted for a more collectivist perspective; but both preferred abstracted empirical methodology. However, interpretive sociology, which is sometimes referred to as ‘social psychology’ or ‘sociological social psychology,’ certainly opted for the observation of individuals in everyday situations and for individual free will and choice. The interpretive sociologies differ in the extent to which they deal with the way that individuals produce culture and social structure, and in the extent to which they also regard individual action as constrained by the structures it produces.

Among the various interpretive sociologies, hermeneutics has the oldest provenance. The term, derived from Hermes (the messenger), originally referred to biblical studies as practised by individuals attempting to divine the ‘true’ meaning of biblical texts. In its modern incarnation, hermeneutics is closely connected to critical media studies, and it is in that context that it is evident in the sociology of sport. Hermeneutics involves the methodology and ‘the theory or philosophy of the interpretation of meaning’ (Bleicher, 1980: 1). Rather than the ‘factual particulars’ of written or visual texts, researchers ‘deconstruct’ texts by ‘look[ing] for recurring themes and messages. Are some issues being given more attention than others? Are certain ideological perspectives being emphasized?’ (Kane and Disch, 1993: 339; see also Duncan, 1986).

A great deal has been written about Max Weber and Weberian sociology (for example, Bendix, 1960; and Gerth and Mills (1946) for a collection of his work). Along with Durkheim and Marx, Weber is considered to be one of the classic founders of sociology. Unlike Marx’s historical materialism, and Durkheim’s attempt to found a positivist science of sociology, Weber’s contribution is rooted in the German philosophical thought of Kant and Rickert, and led him to draw a sharp distinction between the natural and social sciences. ‘For Weber, the aim of sociology was to achieve an interpretative understanding of subjectively meaningful human action which exposed the actors’ motives, at one level “the causes” of actions, to view’ (Jary and Jary, 1995: 726; original emphasis). Weber argued that, ‘action is social insofar as by virtue of the subjective meaning attached to it by the acting individual or individuals, it takes account of the behaviour of others and is thereby oriented in its course’ (1947: 88). Weber’s work has been enormously influential in the development of sociology, but somewhat less so in the sociology of sport.

The most prominent of the interpretive sociologies in the sociology of sport are the sociologies of everyday life, particularly symbolic interactionism and dramaturgical sociology. These forms of interpretive sociology are primarily American in origin, and the Chicago School of urban sociology that flourished between the First and Second World Wars is often cited as the source. Perhaps the most important contributions of the Chicago School were the development of urban ethnographic fieldwork as a methodology, and W.I. Thomas’s fundamental dictum of interpretive sociology—‘if men define situations as real, they are real in their consequences’ (1923). Thus, in seventeenth-century Salem, certain women were defined as witches; while we recognize that such definitions had no basis in reality, they had real consequences for the women who were imprisoned and executed.

Interpretive thinking in American sociology progressed through C. Everett Hughes, George Herbert Mead and Alfred Schutz (who arrived from Austria in 1935), but remained marginal to the main trends in sociology which, in an attempt to establish scientific credibility in the academy, were being modelled on the positivist and empiricist natural sciences. It was not until the late 1960s and 1970s that, in a backlash against the natural science model, interpretive sociologies began to flourish in North American Sociology Departments. Blumer’s collection of essays was published as Symbolic Interactionism (1969); Goffman’s work was beginning to be taught widely; and Garfinkel’s ethnomethodology was beginning to gain recognition. Berger and Luckmann’s The Social Construction of Reality (1967) took the definition of the situation a step further to describe how individuals construct and reconstruct their social worlds. Interpretive sociology has continued to develop and adapt, most recently in its attempts to deal with social structure and attempts to find common ground with cultural studies (for example, Becker and McCall, 1990; Denzin, 1992).

Although the emergence of a distinct sociology of sport was coincident with the 1970s growth of interpretive sociology, interpretive sociology has not, until recently, had a major impact on the development of the sociology of sport. While the influence of Weber has been evident and implicit in the sociology of sport, very little of the work in the sociology of sport has been explicitly Weberian. Only Guttmann’s much praised, and much criticized From Ritual to Record (1978), and the earlier work of Alan Ingham (1975, 1978, 1979), were clearly Weberian. Hermeneutic analyses are coincident with the recent interest in sport media studies, and have been carried out in the sociology of sport for the past ten years or so. Some examples of these are given subsequently.

As noted previously, the sociologies of everyday life have been the most evident form of interpretive sociology in the sociology of sport.3 These began both independently of the sociology of sport as a part of the countercultural growth of interpretive sociology in mainstream sociology (for example, Scott’s (1968) work on horse racing, and Polsky’s (1969) work on pool hustling), and in the sociology of sport itself. Greg Stone, who is recognized as one of the major early contributors to the sociology of sport, was alone among the original group in his involvement in interpretive sociology (for example, Stone, 1955, 1957). Interpretive sociology of sport began to develop at the University of Massachusetts in the 1970s. Friendships between Charles Page and Greg Stone, and John Loy and Donald Ball (see, for example, Ball, 1976), led Page and Loy to encourage several graduate students (Susan Birrell, Peter Donnelly, Alan Ingham and Nancy Theberge) to engage in subcultural studies. Rob Faulkner’s (1974a, 1974b, 1975) presence in the Sociology Department, especially his course on Qualitative Methods, provided additional encouragement.

During this time, there was a great deal of ‘muckraking sports journalism,’ some of the ‘new journalism’ began to deal with sport, and several athletes’ biographies exposed inside information and corrupt practices in sport. While these were not so sociological, they certainly provided rich ‘insider’ information for sociologists of sport interested in the experience and meaning of being an athlete.

The final step in the development of an interpretive sociology of sport began in the 1980s as sport sociologists began to be exposed to the work of the Centre for Contemporary Cultural Studies at Birmingham University in England. For several sociologists of sport this work led them from a rather descriptive and relatively atheoretical form of participant observation, to a much more critical ethnographic approach to interpretive sociology. However, even those who did not shift to critical cultural studies were affected by the anthropologist Clifford Geertz’s (1973) notion of ‘thick description,’ and have generally begun to provide a much more richly textured level of description as a result of their fieldwork. Independent of these North American trends, Pierre Bourdieu and his students in France began to develop an extremely sophisticated interpretive sociology based on fieldwork, and they considered sport to be a part of their mandate from the very beginning. North American sport sociologists began to become aware of Bourdieu’s work in the 1980s and, in combination with the cultural studies approach, interpretive sociology of sport has now become a central aspect of the field of study, making important contributions to our understanding of sport in society.

Critiques and Responses

Critiques of interpretive sociology take issue with both the theoretical and methodological aspects of the approach. The most obvious theoretical critiques are from the forms of sociology that take a social systems approach, or maintain a view that some aspects of social reality are unproblematic in terms of meaning, and may be understood directly (Marshall, 1994), or believe that it is possible to discover standard laws that govern human behaviour. While these critiques only amount to differences in approach to the study of sociology, others zero in on specific aspects of the theoretical approach such as the overemphasis on agency, or cultural relativism.

For example, phenomenology and ethnomethodology emphasize the capacity of individuals (‘agents’) to construct and reconstruct their worlds, which can then only be understood in the agents’ terms (Jary and Jary, 1995). While the approach is obviously criticized by those approaches that emphasize social structure (for example, functionalism, and some forms of Marxism), it is also criticized by those approaches seeking to incorporate both processes (for example, cultural studies) in an attempt to show that there are some structural constraints on freedom to act. With regard to relativism, Jary and Jary note that a distinguishing characteristic of interpretive sociology is ‘the recognition that any statement about the social world is necessarily relative to any other’ (1995: 336).

In an approach that is based on the social construction of reality, there is an obvious concern about what constitutes ‘reality.’ Shotter notes that:

… there is currently something of a ‘flight’ into realism. For one of the major objections to the whole social constructionist movement is as follows. It claims that there is no independent reality to which claims to truth may be compared or referred—for all human ‘realities’ (Umwelten) are only known from within, so to speak—means that there are no independent standards to which to appeal in their adjudication; thus ‘anything goes!,’ and we slide into relativistic nihilism. (1993: 89)

At the cultural level, a relative approach permits sociologists to understand and interpret cultures on their own terms, but when taken to extremes it may find slavery, torture and female circumcision to be ‘justifiable.’ While the evidence for the social construction of characteristics such as gender, and behaviour such as participating in sport, is overwhelming, sociologists continue to struggle with the issues of human values and relativity.

Methodological critiques, which to a great extent overlap with theoretical issues, usually concern the quality of the data produced by fieldwork and in-depth interviewing. Bottomore has captured the methodological dilemma:

The exclusive insistence, in much recent sociology, upon a rigorous ‘scientific method,’ has tended to create an unusually conservative outlook; the existing framework of society is accepted as given, because it is too complex for scientific study, and all the resources of a truly ‘scientific’ sociology are then marshalled for the investigation of small-scale problems carefully isolated from the wider social structure. It is desirable, therefore, to emphasize once more as the distinctive feature of sociological thought that it attempts to grasp every specific problem in its whole social context … (1979: 323)

Insistence on ‘scientific method’ led to much of the criticism of hermeneutics and the sociologies of everyday life. Researchers were accused of subjectivity, of ‘going native,’ and of producing unreplicable results.

Subjectivity is anathema to the ‘scientific method’ where the researcher is considered to be a dispassionate and objective observer and/or manipulator of events. Given systematic observations of human behaviour, particularly if the observer is also a participant; and given in-depth interviews with subjects, advocates of the ‘scientific method’ questioned why the sociologist’s interpretations were any more valid than any other person’s interpretations. The Rashomon phenomenon was invoked (from the Japanese film in which the stories of three participants in an event—a robbery—are quite different), and researchers were also accused of ‘going native’ (that is, losing objectivity by empathizing with their subjects). And, given that it was usually only one researcher engaged in the work, and that studies were rarely repeated (the replicability of results being a benchmark of reliability in the ‘scientific method,’ although one that is not always used), the work was considered to be unreliable.

An initial response to such criticism was the attempt to become more ‘scientific.’ Hermeneutic analysis became a more strictly quantitative ‘content analysis’; Manford Kuhn developed a more quantitative approach to symbolic interactionism that came to be known as ‘the Iowa School’; and methodological texts warned researchers against ‘going native.’ However, more recently there has been a recognition that, in sociology, both the researcher and the focus of research are subjects; thus sociology must be thought of as a subjective or better still, a reflexive science. As Giddens notes:

1) We cannot approach society, or ‘social facts,’ as we do objects or events in the natural world, because societies only exist in so far as they are created and re-created in our own actions as human beings …

2) … Atoms cannot get to know what scientists say about them, or change their behaviour in the light of that knowledge. Human beings can do so. Thus the relation between sociology and its ‘subject-matter’ is necessarily different from that involved in the natural sciences. (1982: 13-15)

Just as some natural scientists are now beginning to recognize the element of subjectivity present in the type of research questions asked and assumptions made, ‘there is now a tendency among field workers to recognize and reveal, rather than deny and conceal, the part that personal interests, preferences and experiences play in the formulation of field-work plans’ (Georges and Jones, 1983: 233). Recognition of this ‘cultural baggage’ now easily extends to open acknowledgement of ‘going native’ and conducting research that is designed to affect social policy and social change.

Willis has suggested that, ‘We are still in need of a method which respects evidence, seeks corroboration and minimizes distortion, but which is without rationalist natural-sciencelike pretence’ (1980: 91; original emphasis), and has gone a long way towards developing that method (for example, Willis, 1978). Researchers using hermeneutic and ethnographic methods now regularly deal with the methodological issues in an open and reflexive manner. For example, with regard to hermeneutics:

At issue in this methodology is the role of the reader. How do we know that readers will interpret the features of photographs in the ways that the researcher uncovers? … A number of authors have argued that texts of all kinds—written and visual—embody multiple realities and suggest multiple meanings. The researcher can never be certain of how a particular individual may perceive a given text because meaning is created in the interaction between the text and the reader … The reader brings his/her personal set of experiences, history, and social and cultural contexts to the text, and all of these influences shape the reader’s interpretation of that text. Although texts may strongly suggest a particular reading … there is always the possibility of oppositional readings …

Responsible textual studies do not assert with absolute certainty how particular texts are interpreted. But they suggest the kinds of interpretations that may take place, based on the available evidence, and likely interpretations of a particular text. Ultimately these interpretations must be judged on the basis of the persuasiveness and logic of the researcher’s discussion. (Duncan, 1990: 27)

Methodological concerns overlap with theoretical concerns precisely in the issues of theoretical assumptions and interpretation. Because of the critical nature of much recent hermeneutic work, an assumption is made that media messages are designed to maintain an unequal status quo in society—hermeneutic analysis attempts to deconstruct those messages. However, a more recent assumption also allows that any ‘reader’ potentially has the power to reject those in-built messages and make an oppositional or alternative reading.

A final critique of research in interpretive sociology concerns the ‘journalistic’ nature of research reports. Critics announce that they find no difference between sociology and journalism in this area. Of course, there are important differences in terms of explicit methodological and theoretical assumptions and techniques, and in terms of the academic review processes and academic responsibility, but there are also important similarities—especially in the ‘new journalism’ (cf. Tomlinson, 1984). There is even an important symbiosis here because journalists are sometimes able to achieve access to areas out of bounds to sociologists, and provide important data and insights useful to sociological research. Donnelly has advocated that, when studying sport subcultures, all sources of data are appropriate and may be useful:

… the researcher should go beyond … traditional forms of data to see what members of the subculture write about themselves for other members (every sport and many leisure subcultures publish at least one newsletter or magazine) or for the general public in biographies, introductory or how-to books, and general books and magazine articles; and what is written about the subculture by journalists and freelance writers …

Nor should the researcher be constrained by nonfiction since many other forms of writing and artistic work are frequently available and offer points of view that are often unique and frequently enlightening. [These include novels] poetry and songs (both of which may supplement the narrative folklore of the subculture), painting and sculpture, cartoons and films … (1985: 568-9)

In the final analysis though, it is not surprising if good journalistic accounts are somewhat like sociological accounts. As Giddens (cited previously) noted, human beings can get to know what scientists say about them, and sociological knowledge is widely reported in the media, and widely available in bookstores and libraries, as well as being taught in university courses. It would be more surprising if good insider journalistic accounts were not similar to good ethnographic research.

Interpretive Studies in the Sociology of Sport

As noted previously, although Weber is frequently included under the umbrella of interpretive sociology, and although his influence on the development of sociology, and to some extent the sociology of sport, is widespread, very few works in the sociology of sport can be considered as explicitly Weberian. Only Guttmann (1978) and Ingham (1975, 1978, 1979) adopt an explicitly Weberian perspective in major works in the sociology of sport. Seppanen (1981) and Stone (1981) employ Weberian concepts in their respective analyses of Olympic success and sport and community.

Also, as noted, hermeneutic analyses of sport have most frequently taken the form of critical analyses of print and television media (see Kinkema and Harris (1992) for a recent review essay on sport and the mass media). The overwhelming majority of such studies have been concerned with the representation of gender in the media, with violence/masculinity a distant, though related, second. Among the more significant analyses of the representation of women are Duncan (1990) on sports photographs, Duncan and Hasbrook (1988) and Duncan, Messner, Williams and Jensen (1990) on televised sports, and MacNeill (1988) on television coverage of women’s bodybuilding and an exercise programme. Trujillo (1995) on televised coverage of (American) football violence, and White and Gillett’s (1994) analysis of print advertisements for bodybuilding products, are examples of work addressing issues of masculinity.

Some of the best work in hermeneutics focuses on specific incidents in sport, but again these are primarily concerned with gender issues. For example, Birrell and Cole (1990) examined the reaction to Renee Richards (formerly Richard Raskind), a ‘constructed-female transexual’ when she/he began to play on the women’s tennis tour; Kane and Disch (1993) examined media accounts of the sexual har-rassment of Lisa Olson, a female sports reporter, in a men’s (American football) locker room; King (1993) and Cole and Denny (1995) explored the public announcement that ‘Magic’ Johnson was HIV positive; McKay and Smith (1995) reviewed media coverage of the O.J. Simpson case; Messner and Solomon (1993) analysed the media coverage of Sugar Ray Leonard’s wife beating incident; Young (1986) examined media responses to the Heysel Stadium incident in which 39 soccer fans died; and Theberge (1989) reviewed newspaper responses to a major incident of sports violence at the World Junior Hockey Championships.

Although there are now many studies of the actual content of media texts, Kinkema and Harris (1992) and others have noted the absence of studies dealing with the production of those texts, and studies dealing with the way in which audiences (rather than researchers) interpret the texts. Hermeneutic analyses and ethnographic analyses combine around these types of studies. Hesling (1986) reviewed the development of sportscasting codes; while Gruneau (1989) and MacNeill (1996) conducted production ethnographies on a World Cup skiing event and Olympic ice hockey coverage respectively; and Theberge and Cronk (1986) examined the production of newspaper sports news. Duncan and Brummett (1993) and Eastman and Riggs (1994) have conducted ‘living room’ ethnographies of television sports spectators in order to explore both oppositional readings and rituals. A great deal more work needs to be done in these areas.

There are distinct overlaps in all of the sociologies of everyday life, in both methodology and theoretical assumptions. With few exceptions, most of the studies in the sociology of sport employing an interpretive approach do not identify a specific sociological theory, and most of those published between the 1960s and 1980s fall into the general category of symbolic interactionism and Goffman’s dramaturgical approach (since that time there has been a clear shift toward a cultural studies approach to ethnographic research). The exceptions include: Howe (1991) and Whitson’s (1976) calls for the use of phenomenology in the sociology of sport, and Pronger (1990) and Rail’s (1992) actual uses of phenomenology in their studies of gay male athletes and female basketball players respectively; and Kew’s series of studies (1986, 1987, 1990, 1992) of the development of rules in sport, employing an eth-nomethodological perspective.

There are two distinct, but overlapping themes of studies employing the remaining sociologies of everyday life (primarily symbolic interactionism and Goffman’s dra-maturgy; but since the 1980s, an increasingly cultural studies approach): (a) descriptions and analyses of sport subcultures (including the themes of careers and cultural production); (b) the process of socialization (which overlaps to a great extent with the theme of career).

Sport Subcultures

Subcultures are ‘any system[s] of beliefs, values and norms … shared and actively participated in by an appreciable minority of people within a particular culture’ (Jary and Jary, 1995: 665), and the study of subcultures is an important aspect of interpretive sociology. Subcultural studies in the United States developed from Chicago School interest in youth, delinquents and deviance, and posited that the formation of subcultures was either a result of ‘differential interaction’ or an ‘environmental response.’ Cohen (1955) combined these views into a powerful explanation of subcultural formation, and it was a short step from examining deviant ‘careers’ to the study of non-deviant careers, eventually including sports careers.

Although Weinberg and Arond’s (1952) study of boxers was the first of many studies of sport subcultures as ‘careers,’ the majority of such studies were not carried out until after the full recognition of a sociology of sport (Loy and Kenyon, 1965). Stone’s studies of wrestlers (Stone, 1972; Stone and Oldenberg, 1967) were followed by Scott’s (1968) work on horse racing, Polsky’s (1969) study of pool hustlers, and Faulkner’s (1974a, 1974b) study of hockey players. This early work came to fruition in 1975 in four major chapters in Ball and Loy’s Sport and Social Order. Ingham (1975) began to develop his theoretical work on occupational subcultures in sport; Haerle (1975) raised the issue of career patterns and career contingencies (for baseball players); while Faulkner (1975) and Rosenberg and Turowetz (1975) carried out comparative work—hockey players and Hollywood musicians in the former, professional wrestlers and physicians in the latter. The use of ‘career’ as a defining concept carried on for some time after 1975—for example, Theberge (1977) on women professional golfers, and Birrell and Turowetz (1979) in another Goffman-inspired comparative study (professional wrestlers and female university gymnasts)—but by the time Prus (1984) summarized the notion of career contingencies the concept of ‘career’ and work on subcultures had both changed.

Careers were being thought of as any time spent progressing in a sport—a competitive swimmer who began at age 6, retired at age 14 and never earned any money, could now be considered to have had a ‘career’ in swimming. And the cultural characteristics of subcultures were now being studied without resort to the concept of ‘career.’ Vaz (1972) anticipated this with his work on young hockey players, and was followed by Thomson (1977) on rugby players, Pearson (1979) on surfers, Donnelly (1980) and Vanreusel and Renson (1982) on rock climbers and other high-risk sport participants, and Albert’s series of studies on racing cyclists (1984, 1990, 1991).

However, during the 1980s, there was a further change in the study of sport subcultures resulting from two sources. First, Geertz’s (1973) methodological refinement of ‘thick description’—‘[i]ntensive, small scale, dense descriptions of social life from observation, through which broader cultural interpretations and generalizations can be made’ (Marshall, 1994: 533)—led to much richer and textured descriptions of social contexts. Secondly, British subculture theory took a radical theoretical/political turn in the direction of critical sociology (for example, Hall and Jefferson, 1976; Hebdige, 1979). As a result, the sociology of subcultures not only shows how sports and other leisure practices are socially constructed and defined activities, meaningful only to the extent that meaning is attached to them by the participants, but also how ‘subcultures, with their various “establishment” and “countercultural” emphases, have been consti-tutively inserted into the struggles, the forms of compliance and opposition, social reproduction and transformation, associated with changing patterns of social development’ (Gruneau, 1981: 10). These changes have had a powerful impact on subcultural studies of sport.

The changes noted above have produced an extremely rich crop of studies, and the following examples have been influenced by one or both of the changes: Williams’s (Williams et al., 1989) and Giulianotti’s (1995a, 1995b) studies of British soccer hooligans; Klein’s observations of bodybuilders (1993) and baseball players in the Dominican Republic (1991); Foley’s (1990) study of the American high school football subculture; the Adlers’ (1991) work on university basketball players; Crosset’s (1995) study of women professional golfers; Sugden’s (1987) study of boxing; Tomlinson’s (1992) study of a British folk game (knur and spell); Curry’s (1991) study of the male locker room subculture; Birrell and Richter’s (1987) study of transformations in women’s softball; Theberge’s (1995) study of women’s ice hockey; Beal’s (1995) study of skateboarding; and Markula’s (1995) work on women aerobics participants.

Most of these studies have been carried out by researchers from the United States, Canada and Britain. However, there is also a French school of subcultural research rooted in the sociology of Pierre Bourdieu and mostly carried out by his students. For example, Wacquant’s series of studies on boxing (1989, 1992, 1995a, 1995b), Clément’s (1984, 1985) studies on the martial arts, Bruant’s (1992) ethnography of running, Pociello’s (1983) work on rugby, Suaud’s (1989) study of tennis clubs, and Midol’s studies (1993; Midol and Broyer, 1995) of extreme (‘whiz’) sports.

Socialization and Sport

Research on who becomes involved in sport, how they become involved and the effect that sport has on them, has been important since the beginning of the sociology of sport. However, the early research was based on survey methods, and it is only since those involved in the interpretive sociology of sport have turned their attention to socialization that a number of rich insights have been made into the process.

Socialization is an active process of learning and social development that occurs as people interact with each other and become acquainted with the social world in which they live, and as they form ideas about who they are, and make decisions about their goals and behaviours. Human beings are not simply passive learners in the socialization process. Instead, they actively participate: they influence those who influence them, they make their own interpretations of what they see and hear, and they accept, revise, or reject the messages they receive about who they are, what the world is all about, and what they should do as they make their way in the world. (Coakley, 1998: 88)

The processes of becoming an athlete and becoming an adult person, and the way these two interact, were addressed to a certain extent in some of the ‘career’ subculture studies noted above. However, primary recognition of the socialization process seems to have occurred as a consequence of the 1980s changes in subcultural studies of sport.

Early involvement in sport and physical activity has been studied by Hasbrook (1993, 1995) in school playgrounds, Fine (1987) in little league baseball and Ingham and Dewar (1989) in PeeWee hockey. But these studies go well beyond the actual processes of involvement to show how sport and physical activity are major sites for the production and reproduction of traditional and stereotypical notions of gender. Grey (1992) has shown how new immigrants fail to become socialized into high school sport; and Chafetz and Kotarba (1995) remind us that socialization is a two-way process when they show how little league baseball affects the players’ mothers.

Once involved in sport, the socialization process continues. Donnelly and Young (1988) examined the way in which rookie athletes constructed appropriate subcultural identities for themselves, to be confirmed (or rejected) by established athletes in the subculture; Coakley and White (1992) showed how English teenagers made the decision to continue, or not continue, sport participation; Stevenson (1990) demonstrated how international athletes began to focus on their particular sport; and Messner (1992) explored the meaning of success and relationships in the lives of elite male athletes. The process continues to be two-way, and Thompson (1992) showed the way in which the involvement of husbands and children in tennis had an impact on the lives of wives/mothers. Many of the subcultural studies noted above also deal with the continuing process of socialization.

Career interruptions and endings (retirements, sometimes referred to as desocialization) form the final phase of sport socialization. Coakley (1992) has studied burnout among adolescent athletes as a problem of social development, and Swain (1991) and Messner (1992) have taken different approaches to the problem in their analyses of the retirement of elite athletes. Young (Young, White and McTeer, 1994; Young and White, 1995) has provided the most focused view of the way that male and female athletes view and deal with injuries. Of course, except in the case of traumatic injury, such retirements rarely signal a complete end to sport involvement—rather, they indicate a change in the way a person participates.

Finally, several new approaches to interpretive sociology should be noted. While sport sociologists have often accepted athletes’ biographies as reasonable sources of data, more recent sophisticated methodologies (including case studies, life histories and narrative sociology) have brought biography and autobiography to the fore in the interpretive sociologist’s repertoire of techniques. Henning Eichberg’s special issue of the International Review for the Sociology of Sport (29 (1), 1994) on ‘Narrative Sociology’ provides a number of examples of this type of work by scholars from Finland and Denmark; while in North America, Tim Curry (1993, 1996; Curry and Strauss, 1994) has been an exponent of the case study/life history technique. Interpretive sociologists have also begun to explore innovative ways of reporting research—see, for example, Curry and Strauss’s (1994) photo-essay; and Wacquant’s (1989) split-page technique. Finally, Cole (1991) and Foley (1992) have reminded interpretive sociologists of sport of the politics of interpretation, perhaps ensuring that reflexivity is maintained in the use of these techniques.

The Impact of Interpretive Sociology on the Sociology of Sport

Studies using an interpretive sociology approach, particularly ethnographic studies, are extremely time-consuming. They usually require enormous commitment on the part of the researcher, but that commitment has paid off in the sociology of sport with a number of rich and rewarding studies. With the refinements in theory (a critical cultural studies approach is now widely used) and methodologies (reflexivity, ‘thick description,’ ethnography), interpretive sociology has recently come to be what many consider the predominant approach in the sociology of sport. The recent publication dates of the majority of studies noted above attest to the growth of interest in interpretive sociology in the sociology of sport and, as an example of this recent prominence, all of the Presidential Addresses given at the North American Society for the Sociology of Sport annual conferences in the 1990s have taken an interpretive sociology approach.

Weberian sociology has given us important insights into the emergence and development of modern sport; and hermeneutics has shown us the ideological underpinnings and dynamics of the sport—media complex, particularly in terms of gender relations. But perhaps the major impact of interpretive sociology has been the way in which it hangs flesh on the skeletons of survey data. For example, surveys have continually shown that boys and men are more involved in sport and physical activity than girls and women, but researchers were obliged to speculate about the reasons for this difference. Recent observations and in-depth interviews have shown the dynamics of gender relations in sport—how sport is a gendering practice, and how males and females make decisions about participation—thus giving a much clearer explanation for the participation differences. Interpretive sociology has also given us a much richer notion of careers, subcultures and group dynamics in sport, and a great many insights into how we become involved in sport, live our lives as athletes and retire from intense participation. In the final analysis, though, interpretive sociology is about meaning, and in the sociology of sport we are beginning to attain a powerful sense of what sport means, and how sport means, in the lives of human beings.