Sociology and ideas about language
Getting things done: language and social interaction
Ethnomethodology
One approach that seeks to overcome this assumption, by insisting that social relations are practical accomplishments, is ethnomethodology. A central source of inspiration for ethnomethodology was the work of Schutz (1943, 1953, 1967). Schutz developed his ideas through a critique of the Weberian conception of interpretive sociology discussed above. While agreeing with Weber that people experience the social world as a socially meaningful reality, he took issue with Weber's insistence that meanings were subjective [that is, Weber's claim that meanings were attached by the individual to their actions, were provided by the individual, as it were, to make sense of, or to orient, their actions towards definite, intentional purposes). The core of Schutz's reservations lay in his affirmation of the intersubjective nature of meanings. By this, Schutz refers both to the emergent nature of meanings, a theme we have already identified as characteristic of symbolic interactionism, and to the ways in which this emergent nature leads us to experience the everyday world as a common, 'shared' world. In our practical doings in the world we come to take for granted these common meanings: we expect to be understood, to be served when we queue for the checkout, to be recognized as a lecturer by students and so on. In short, individuals experience the world as external, orderly and pre-existing.
However, it is also the case that each of us has to negotiate and make sense of this world whose 'reality' at a practical level has simply to be taken for granted. The basis of our ability to interpret the social world routinely, and by and large successfully, is what Schutz calls our 'common-sense knowledge'. This is the knowledge we gain by virtue of living in and being part of the everyday world, and it enables us to categorize and order the world through the stock of typifications it provides us with. Importantly for the argument we wish to advance later, this stock of typifications is not our own invention, argues Schutz, but is embodied in the language we share with others. Such typifications, the shorthand of commonsense knowledge, are enabled by language.
Ethnomethodology was also partly a reaction to the structural functionalism of Parsons, although it continued to share some of its assumptions, particularly those to do with the consensual basis of social action (Williams 1992). Structural functionalism presents a model of society in which the persistence of social stability and the orderliness of social life are accounted for in terms of externally defined norms and values which people learn and incorporate into their behavior. The common, shared nature of social life is attributed to the 'shared' normative commitments that social actors are given by the processes of socialization. Ethnomethodologists reject the deterministic emphasis on an external value system pressure-moulding a stream of social actors predisposed to conformity. Instead, following Schutz, they want to insist that social order is not something guaran teed from without, but is rather generated from within social settings by the practical activity of those involved. It is people's commonsense knowledgeability, their adroitness at interpreting symbolic cues, at assessing paralinguistic signs, at communicating intentions, that makes social life run in a relatively orderly fashion. It is therefore to the means of 'local production', that is, the ways in which social life is sustained and accomplished by people in everyday settings and on an everyday basis, that social research must turn. This emphasis on how people produce their social worlds through negotiation and interaction inevitably brings language to the fore.
The situated, local nature of social meanings, their rootedness in specific social contexts (what ethnomethodologists refer to as the indexical nature of meanings), leads these theorists to reject approaches to language use which focus on the formal aspects of language structure. When, for example, theorists concern themselves with formal sentences there is an implication, directly analogous to that found in sociological accounts which focus on formal norms and rules, that a lexicon exists with obvious meanings for all competent speakers. Such approaches, precisely because they ignore the indexical nature of meanings, obscure the implicit ways in which the standardized character of the culture or society is presumed as obvious to its members. Garfinkel, regarded as the founder of ethnomethodology, would illustrate this for his students by getting them to enact 'breaching experiments', such as pretending when they went home that they were lodgers rather than family members; this often-provoked hostility and anger, demonstrating that there are many aspects of the everyday 'moral order' which are usually taken for granted (Garfinkel 1967).
Ethnomethodology is thus drawn to a methodological focus on how individuals make norms and meanings relevant to themselves, and claims that this is accomplished through social conversations. As we shall see, this leaves the approach with the problem of how to account for constraints on what people are able to accomplish through their social conversations. The issue of constraint, of course, entails a consideration of power and conflict, features of the social world which interactionist sociologies have found notoriously difficult to explain. This problem of accounting for power is compounded in the case of ethnomethodology by the construal of social interaction as language. Methodologically, interaction is accessed by observing conversation. One attempt to attend to these limitations is made by Goffman, who, although he belongs to the broadly interactionist tradition, introduces the notion of the 'interaction order' (Goffman 1983), as a means of linking these micro-processes to wider issues of social order and social structure (Layder 1997: 211). However, like many of the writers discussed here, he pays little detailed attention to language as such.
A writer who is regarded by some ethnomethodologists as departing from the mainstream because he insists on distinguishing between interaction and language, and who has also concerned himself with the linguistic aspects of social action, is Cicourel. Cicourel's concern with the significance of language starts from where symbolic interactionism stops. Whilst concurring with symbolic interactionism on the crucial role of meanings in the negotiation of everyday social life, Cicourel regards this as only the beginning of analysis, since the more fundamental question is 'How does the actor in everyday life order and assign meanings to objects and events in his environment?' (Cicourel 1973: 12). This is also a direct challenge to the structural functionalist tradition which, in viewing the orderliness of social life as a product of normative patterns inducted by socialization processes, fails to get to grips with the issue of how individuals decide what norms are operative or relevant in any situation.
There are two aspects to this claim that are relevant to our present argument. The first is a methodological one and concerns the assumption often made by certain variants of sociological theory that the (sociologist's) identification of patterns of norms and values is also an account of people's attachment to such norms and values. Consensual behavior is seen to be an outcome of shared norms and values. Cicourel regards this a 'gloss', by means of which the sociologist smuggles into the analysis an implicit model of how individuals assign meaning and relevance to symbols, rules and values, interpret them in the context of everyday situations and engage them through appropriate behavioral displays. There are some similarities between Cicourel's arguments here and those of Archer concerning the 'myth of cultural integration'. Both sets of arguments suggest that identifying 'cultural' or 'ethnic' groups on the basis of a supposed commitment to certain values and norms is far from straightforward; not only is there the general problem of how persons establish roles and statuses in everyday action (how does one know when one is conforming to role expectations? which role?), but also the specific problem of how the researcher establishes goals and norms for the individual, let alone for some group or wider community. As Cicourel notes, this 'is not obvious theoretically, nor procedurally clear methodologically' .
The second aspect of the ethnomethodological insistence that sociological research must make the role of interpretive procedures in social life explicit concerns the central role of language in such procedures. Again, this goes beyond the symbolic interactionist position that meanings are the basis of social interaction, and stresses social performance, that is, the question of how, and by what means, people come to assign relevance to meanings. It is individuals themselves who decide what is relevant, what is influential, what they are to take into account in their social behavior. They do this through what Cicourel terms 'interpretive procedures':
Interpretive procedures provide the actor with a developmentally changing sense of social structure that enables him to assign meaning or relevance to an environment of objects. Normative or surface rules enable the actor to link his view of the world to that of others in concerted social action, and to presume that consensus or shared agreement governs interaction.
We are continually modifying our interpretive procedures as we acquire new context-bound lexical items; the study of language use is important for understanding how we do this and thereby 'place' our environments, rendering them comfortable and familiar, or threatening and strange, or something in between. Interpretive procedures are thus the practical methods by which social order is constructed and sustained through everyday interaction. The notion of interpretive procedures clearly indicates the importance for Cicourel of the sociolinguistic element in sociological research.
However, Cicourel is not abandoning the effort of sociological analysis, nor necessarily confining it to the (linguistic) analysis of actor conversation. The purpose of his work, rather, is to suggest 'a more explicit foundation for integrating social process with the structural or institutionalized features of everyday life' (ibid.: 11). The crucial issue here is 'how much we feel we can genuinely know about social structure through the analysis of local practice' (Coupland 2001a: 17).
Cicourel does provide some grounds for developing a model of social life which is not composed solely of social conversations. From his work, there would appear to be two elements in the social world that are irreducible to, although clearly the product of, linguistically mediated social interaction: interpretive procedures and syntactical rules. We have considered the former's objective status above; according to Cicourel they are 'invariant properties of everyday practical reasoning necessary for assigning sense to the substantive rules sociologists usually call norms' (Cicourel 1973: 52). Syntax, too, shares the same property of irreducibility and so has an independence from social actors which allows them to shape the actors' interpretive repertoire:
... formal syntactic rules can be generated in cultural settings or can remain oral traditions: once they are available and used in socially prescribed and sanctioned ways, however, they constrain (although not completely) interpretations assigned by members.
So, although it would seem that the social world comprises more than human beings and their conversations, the nature of the formal properties of language, and of the invariant properties of interpretive procedures, is unclear. Furthermore, the links between them and local practices are not made explicit, making it difficult to grasp the inter play between, say, the systemic features of a language such as its grammar, and the ways in which these might enable or constrain particular sorts of utterances. As we shall argue, a social realist applied linguistics offers the potential for making these links more explicit.
