Theoretical and methodological approaches: Institutional order versus interaction order
Workplace interaction has been studied from a great variety of theoretical and methodological perspectives within a range of disciplines, including sociology, organisational studies, sociolinguistics and applied linguistics. Sarangi and Roberts’ (1999: 1-10) suggest that most studies of workplace discourse can be situated on a continuum from those that focus more on the institutional order to those that are more interested in the interaction order. Workplace interactions are embedded in the institutional order, which is defined by Berger and Luckman (1967: 83) as the “body of transmitted recipe knowledge, that is, knowledge that supplied the institutionally appropriate rules of conduct”. Goffman (1974), on the other hand, called for study of the interaction order, which is “the structuring of participation in a given social situation” (Sarangi and Roberts 1999: 2, original italics), in its own right. This involves studying in detail the actual words and interactions of the participants in a given situation.
As the present volume focuses on linguistic approaches to business communication, only those approaches to spoken workplace discourse that have been carried out within a sociolinguistic or applied linguistic framework will be considered here.
But even this more limited set of linguistic approaches to analysing workplace interaction represents quite a range of perspectives which can also usefully be plotted on the institutional order to interaction order “cline” or continuum (see Figure 26.1):
Figure 26.1: Institutional order vs. interaction order (Legend: CDA = Critical Discourse Analysis, IS = Interactional Sociolinguitics, CA = Conversation Analysis)
Studies of spoken workplace interaction that go furthest towards incorporating the institutional or social order into their framework are those that take a socio-critical approach, notably critical discourse analysis, which draws on the works of Foucault (Cameron 2000; Mullany 2007) or Bourdieu (Duchene 2009). Such studies explore, and attempt to deconstruct the ways in which power and ideology are enacted through workplace discourse. The detail or depth of linguistic analysis varies greatly in studies adopting a critical approach. While in more ethnographic studies, analysis of conversational or interview data may be limited to selected examples which illustrate certain practices or ideologies within the organisation (e.g., Duchene 2009), more linguistically-focused studies on critical discourse analysis frequently draw on systemic functional linguistics to carry out a detailed lexico-grammatical or clausal analysis. One good example of the latter is Eggins and Slade’s ( 2005:116-166) critical examination of how dominant ideologies are constructed in the informal conversation of male co-workers.
Social constructionist approaches to workplace discourse can be situated somewhere in the middle of the institutional order - interaction order cline, as social categories and discourse are seen as mutually reflexive. As mentioned in Section 2.4, various aspects of workplace identity have been examined using a social constructionist approach, for example gender, leadership and ethnicity (Holmes, Stubbe, and Vine 1999; Holmes 2006; Mullany 2007; Baxter 2010). What these studies have in common is that social and institutional identities are seen as dynamic and subject to negotiation through interaction (Holmes and Stubbe 2003: 8-12).
Interactional sociolinguistics, as the name implies, is interested in the detail of interaction in relation to the participants’ social and cultural interpretation of “contextualisation cues” in talk. These are any linguistic features of talk that index particular social practices (be they prosodic features, formulaic expressions or particular lexical choices) and thus may be subject to different cultural interpretations. Interactional sociolinguistics has a long tradition in the study of workplace talk and was developed by Gumperz (1982) in studying cross-cultural (mis)communi- cation in multi-ethnic contexts. This approach has frequently been adopted in workplace contexts with a high proportion of ethnic minorities in order to study the processes through which workers or job applicants are systematically disadvantaged (Roberts, Davies, and Jupp  2014). For example, Roberts and Campbell (2005,
2006) found that ethnic minority candidates were frequently unsuccessful in job interviews, not as a result of low linguistic proficiency, but because they failed to adopt certain “discourses” that were expected of them.
Where genre analysis is positioned on the institutional order - interaction order cline depends on which genre-analytical approach is adopted. The “new rhetoric” school of genre analysis, which emerged from a socio-constructionist tradition, is more interested in the ways in which professional discourse communities use genres and less in the formal properties of individual genres (Freedman and Medway 1994). In contrast, genre analysis developed within English for Specific Purposes (ESP) and English for Academic Purposes (EAP) (Swales 1990; Bhatia 1993), as well as within a Hallidayan tradition (Martin and Rose 2008), has tended to focus on the language and structure of different genres, frequently identifying a typical “move structure” for a particular genre, such as sales promotion letters (Bhatia 1993). Written workplace genres (e.g., email, letters, reports) have been analysed far more than spoken genres, but there are nevertheless some descriptions of frequently occurring spoken workplace genres, such as meetings (Bargiela-Chiappini and Harris 1997; Handford 2010), decision-making (Koester 2006) or service encounters (Hasan 1985; Ventola 1987; McCarthy 2000). Some of these studies will be discussed in more detail below.
Although corpus-based approaches have a long tradition within English for Specific Purposes and English for Academic Purposes, the use of corpus analysis in analysing spoken workplace discourse is relatively recent. As spoken data must first be transcribed before it can be entered into a computer-based corpus, spoken corpora tend to be much smaller than written ones. Spoken workplace corpora are, therefore, frequently quite small and specialised, for example Adolphs et al.’s (2004) corpus of calls to a health advisory service (comprising approx. 60,000 words) or Koester’s (2006) corpus of American and British office talk (only 34,000 words). The two largest English language spoken corpora of workplace talk are the 1 million- word Cambridge and Nottingham Business English Corpus (CANBEC), consisting mainly of business meetings (Handford 2010), and the 262,000-word business subcorpus of the Hong Kong Corpus of Spoken English (HKCSE-bus), covering a wider range of workplace contexts and activities, including job interviews, telephone conversations and front desk hotel service encounters (Warren 2004). As corpus analysis involves the identification of frequently-occurring lexical items, collocations and phrases, the focus of corpus-based workplace studies is primarily on the interaction order. However, as Handford (2010: 31) argues, “the corpus as an objective, quantifiable record” of linguistic professional practices also links the text back to its social and institutional context.
Conversation analysis, which, as mentioned above, has a well-established tradition of analysing workplace talk, explicitly sets out to explore the interaction order through micro-analysis of speaker turns and sequences. Moreover, “C[onversation] A[nalysis] argues that institutional talk is centrally and actively involved in the accomplishment of the ‘institutional’ nature of institutions themselves” (Hutchby and Woofitt 1998:145), which is in line with the conversation-analytic view that talk creates its own context. Interestingly, in a recent volume on conversation analysis and Languages for Specific Purposes, Seedhouse and Richards (2009: 21-25) propose a tri-dimensional model of context (with “institutional context” and “micro-context” as separate categories) and contend that such a model is not in conflict with a conversation-analytic view of context as talk-intrinsic. Conversation-analytic studies of institutional discourse cover a very wide range of institutional contexts (including health care, legal professions, news media) and interaction types, such as telephone calls (e.g., emergency calls, in Zimmerman 1992, and telephone service encounters, in Bowles 2006 and Varcasia 2009), radio talk shows (Hutchby 1996) and performance appraisal (Aftmus 2008; Clifton 2012). These studies have shown how specific turn-taking and sequential patterns, which are distinct from those of “ordinary conversation”, operate within different institutional forms of interaction (Drew and Heritage 1992: 26).
Three of these approaches to analysing spoken workplace discourse, namely genre analysis, corpus analysis and conversation analysis, will be examined in more detail in the next section. They have been selected in line with the overall focus on “linguistic approaches” in this book because they are situated at the interaction order-end of the cline (see Figure 1), and therefore involve a linguistic and micro-discursive analysis of spoken workplace communication. Nevertheless, as we shall see, findings from such studies link back to the institutional order, thus also providing insights into the social and professional issues in the workplaces investigated.