5. Approaches to Register and Genre Analysis

Approaches to genre and register variation have also been much in evidence in analysis of discourse. The term ‘register’ has been used in other contexts to denote different collections of features, such as those associated with the cline from formal to informal uses and those associated with particular fields of knowledge. It is frequently associated with the work of Douglas Biber and colleagues on computational analysis of texts, who have used the term broadly ‘for varieties defined by their situation characteristics’ and have, as a central research goal, the understanding of ‘how language varies when used for different purposes in different situations’. To investigate register variation, Biber applied a multidimensional (MD) analysis to a corpus of different registers, quantifying variation in features such as tense and aspect markers, pronouns and pro-verbs, questions and subordination features. Following statistical analysis of the correlations of these features, Biber identified five dimensions of variation within the registers examined:

‘involved’ versus ‘informational production’; ‘narrative’ versus ‘non-narrative discourse’; ‘elaborated’ versus ‘situation-dependent reference’; ‘overt expression of argumentation’; ‘impersonal’ versus ‘non-impersonal style’. This enabled registers to be compared along a scale relating to the particular dimensions. For example, in involved versus informational production, telephone and face-to face conversations would be at the involved end of the scale, general fiction somewhere in the middle and academic prose and official documents at the informational end. In impersonal versus non-impersonal features, face-to-face conversation would be at the non-impersonal end of the scale with general fiction nearby. Academic research articles would be towards or at the impersonal end. Biber maintains that MD analysis of registers gives a more comprehensive picture than approaches which focus on a more limited number of linguistic variables.

Genre analysis is the approach favoured by a variety of analysts interested in both linguistic form and social purpose. The substantial body of work on genre by John Swales has examined the typical structures of texts associated with the communicative purposes of particular discourse communities. He has focused on genres typical of university settings, particularly those encountered by students using English for academic purposes (Khine). The empirical research article (RA) is, perhaps, the most analysed of these academic genres and characteristically follows the pattern of introduction–methods–results–discussion (IMRD).

5. Speech and Discourse Communities

A speech community comprises people in habitual contact with each other through language. The term has been variously defined within linguistics. Definitions may emphasise shared language use, frequency of interaction between speakers, shared rules of speaking and interpretation, or shared attitudes and values with respect to language. Dell Hymes (1974) provides a fairly broad definition of a speech community as one whose members share ‘knowledge of rules for the conduct and interpretation of speech. Such sharing comprises knowledge of at least one form of speech, and knowledge also of its patterns of use’.

A discourse community (Swales) is a grouping based on a common interest, with common goals and mechanisms for intercommunication between members. It uses participatory mechanisms to provide information and feedback, some specific lexis and genres and has a threshold level of members with relevant content and discoursal expertise. The strength, but also the weakness, of this tradition of genre analysis is that it draws on much existing terminology, what might be referred to as ‘folk taxonomies’. The research article, the grant proposal, the PhD thesis or dissertation are all recognisable to the community of academics, but their understanding of what the labels mean is likely to be coloured by their disciplinary context and experiences.