«Journal of Sociolinguistics 4/3, 2000: 323±347 Styling the worker: Gender and the commodi®cation of language in the globalized service economy1 ...»
Journal of Sociolinguistics 4/3, 2000: 323±347
Styling the worker: Gender and the
commodi®cation of language in the
globalized service economy1
Institute of Education, London
This paper discusses some sociolinguistic characteristics of the speech style
prescribed to workers for interacting with customers in service contexts,
focusing in particular on the linguistic and vocal `styling' prescribed for
operators in telephone call centres in the U.K. Attention is drawn to the similarities between the preferred style of speech and what is popularly thought of as `women's language'. The intensive regulation of service workers' speech and the valorization of `feminine' communication styles are analysed in relation to changes occurring as a consequence of economic globalization.
KEYWORDS: Language and gender, globalization, institutional talk, call centres
The question of control is raised explicitly in the literature of business and management. In her book Corporate Speak: The Use of Language in Business, for # Published by Blackwell Publishers Ltd. 2000 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden MA 02148, USA.
324 CAMERON instance, Fiona Czerniawska (1998) explains that the adoption of new managerial approaches in a context of intensi®ed global competition has sharpened awareness of language as a valuable commodity, potentially a source of `competitive advantage', which therefore needs to be `managed' rather than simply left to take care of itself. Particularly in the service sector of the economy, whose growth is one feature of globalization, one may observe an increasing tendency foremployers to regulate even quite trivial details of workers' talk (Cameron 2000; du Gay 1996).
Here I examine the imposition on one group of English-speaking customer service workers (telephone call centre operators)2 of a particular speech style as the norm or `standard' for interaction on the job. As well as discussing the means used by organizations seeking to exert control over the speech of their employees, I will discuss some of the sociolinguistic characteristics of the speech style that is prescribed as a `standard'. I will argue that its most salient features are not markers of class, region, or nationality/ethnicity, but symbolic markers of feminine gender (though they are not presented explicitly as gendered, and they are prescribed to workers of both sexes). The commodi®cation of language in contemporary service workplaces is also in some sense the commodi®cation of a quasi-feminine service persona.
Before I proceed, my use of certain terms requires clari®cation. When I talk about the imposition of a standard or about the standardization of speech within an organization, this is not intended to mean `the imposition of the lexicogrammatical norms of a standard (national/international) language', but more abstractly, the practice of making and enforcing rules for language-use with the intention of reducing optional variation in performance (Milroy and Milroy 1998). As will be seen in more detail below, the rules in question tend not to target grammatical or phonological variation (these being the prototypical targets for language standardization in the less
sense). They are more concerned to prescribe features of interactive discourse such as prosody and voice quality, the way in which particular speech acts should be performed, the choice of address terms/salutations and the consistent use of certain politeness formulae. In this instance standardization is not prompted by the need to communicate across regional/national boundaries (though in the case of multinational companies it may operate across them), but rather by the need to subordinate individuals to a corporate norm. Employees' verbal behaviour, along with other aspects of their self-presentation such as bodily appearance and dress (cf. Witz, Warhurst, Nickson and Cullen 1998), is treated as a commodity ± part of what organizations are selling to their customers, an element of their `branding' and corporate image. The signi®cance organizations accord to the prescribed style of speaking is evident from the degree of eort they put into its production via training, regulation and surveillance of employees' speech.
Above I used the phrase `prescribed style of speaking', and throughout this paper I will refer to the object/product of linguistic regulation as a `style'. At this # Blackwell Publishers Ltd. 2000 STYLING THE WORKER 325 point it is helpful to clarify what I mean by the term style and how the phenomena discussed below ®t into ongoing discussions of style in sociolinguistics.
STYLE, STYLING, STYLIZATIONClassically in the variationist paradigm of sociolinguistics, `styles' were de®ned along an axis of formality: an increase in the formality of the situation leads to increased self-monitoring by the speaker and therefore, in the typical case, to rising frequencies of prestige variants in that speaker's output (cf. Labov 1972).
Over time, however, there has been a tendency to adopt a less monodimensional view of style and of the meanings or eects produced by stylistic variation. An example of the more multidimensional approach is Allan Bell's in¯uential theory of style as `audience design' (Bell 1984, 1997) in which it is argued that stylistic choices are primarily motivated by the speaker's assessment of the eect certain ways of speaking will have on particular addressees. Bell's account
is informed by accommodation theory (e.g. Giles and Powesland 1975):
audience design commonly takes the form of convergence towards the addressee's way of speaking (for empirical examples see Bell 1984; Coupland 1984). However, Bell also notes the existence of what he calls `initiative' (as opposed to `responsive') styleshift, and of cases in which `the individual speaker makes creative use of language resources often from beyond the immediate speech community' (Bell 1997: 248). An instance which has attracted attention in recent sociolinguistic research is the phenomenon of `crossing' (Rampton 1995) ± appropriating linguistic features that index an identity which is in some salient way `other' (as with the use of variants marked as Black by speakers who are themselves white; see Bucholtz 1999; Cutler 1999).
Crossing is rarely a case of convergence towards the immediate addressee (more usually it reproduces features associated with an absent reference group ± not uncommonly one whose speech lacks prestige by mainstream de®nitions). Allan Bell, following the literary theorist Bakhtin, puts this under the heading of `stylization' ± taking on a voice which is recognizably dierent from one's `normal' or `expected' voice (Bell 1997: 248).
The creative deployment of varied linguistic resources may also be manifested in linguistic behaviour that is not crossing, but rather involves some mixing of elements from dierent sources. Penelope Eckert suggests: `The construction of a style is a process of bricolage: a stylistic agent appropriates resources from a broad sociolinguistic landscape, recombining them to make a distinctive style' (1996: 3). `Style' in Eckert's usage can be a verb as well as a noun: the `stylistic agent' who draws on the meanings made available by linguistic variation and combines these into a distinctive way of speaking can be seen as `styling' her/ himself. Eckert's particular interest is in the self-styling undertaken by adolescents and pre-adolescents as they experiment with various possible positionings within their newly signi®cant peer groups and social networks.
# Blackwell Publishers Ltd. 2000 326 CAMERON My own use of the term style is broadly in the spirit of the post-Labovian work cited above, but there are some signi®cant (and interesting) dierences between the styling practices I am interested in and those studied by Bell, Rampton or Eckert. Their work focuses on practices of self-styling, where the speaker is also what Eckert calls the `stylistic agent', the person who makes choices about her or his own linguistic performance. In the service workplaces investigated here, by contrast, the roles of speaker and stylistic agent are separated to a signi®cant extent. It is of course true that any actual linguistic performance must, in the ®nal analysis, be produced by the speaker her or himself. It is also true that some stylistic choices remain the prerogative of individual speakers, because they involve variables that have not become objects of institutional regulatory zeal (in the call centre case for example, accent is not normally a target for institutional regulation).3 In general, however, service styles are designed by one set of people (managers on site or at head oce, or ± not uncommonly ± outside consultants) to be enacted in speech by a dierent set of people (frontline customer-service workers). Typically a third set of people (supervisors or `team leaders', and sometimes also `mystery shoppers', people employed by companies to carry out spot-checks on service while posing as genuine customers) are charged with ensuring compliance through monitoring, `coaching' (the ongoing provision of critical feedback) and appraisal of workers' linguistic performance.
A further dierence takes us back to the question of `audience design'.
Corporate style designers do, of course, make stylistic choices with an audience in mind, namely the customers with whom service workers interact. In this they resemble the radio presenters discussed by Bell (1984) in his article on style as audience design, who are obliged to imagine their target addressees as a collectivity, and to make guesses about the preferences of those addressees.
But whereas the radio presenters do their own speaking, the corporate style designers' relationship to the audience is indirect, mediated by the workers who actually talk to customers. These workers eectively have a dual audience: they speak to the customer, but at the same time they are also using the prescribed style for the bene®t of the supervisor or manager who enforces linguistic and other norms through surveillance. Some workers I interviewed, though clear that in theory their job was to serve the customer, not their supervisor, reported that in practice they prioritized the requirements of the `in-house' audience, whose judgements on their performance had more direct and immediate consequences. This is an intriguing case where the demands of what Bell (1997: 246±247) refers to as `auditors' and `overhearers' appear capable of overriding those of the actual addressee.
In sum, `styling' in contemporary service workplaces is less a community practice, generated from the bottom up, than a prescriptive or `verbal hygiene' practice (Cameron 1995), imposed from the top down. For this reason, and despite some points of resemblance, it is not wholly comparable either to the selfstyling practices of adolescents (Cutler 1999; Eckert 1996; Rampton 1995) or to # Blackwell Publishers Ltd. 2000 STYLING THE WORKER 327 the stylistic behaviour of workers modifying their speech (whether consciously or unconsciously) to advance their own interests in business transactions (Coupland 1984; Hall 1995; Johnstone 1999). It might be considered a case of `stylization', since it involves speakers giving a performance, the `script' for which has been written by someone else (literally in some instances, as will be discussed further below). Yet it lacks what might be seen as a de®ning feature of stylized utterance, namely the quality of calling attention to itself (however subtly) as a performance, of pointing to some kind of separation between the speaker's self and her/his speech at that moment. Though they may vary in their ability to bring it o, service workers performing standard routines are typically instructed to aim for a `natural' and `authentic' performance.
Workplace styling, then, is a distinctive phenomenon, and as such it prompts various questions, not all of which can be addressed in the space of a single article. The main question I set out to address here concerns the actual choices made by corporate style designers: what are they, and what is the motivation for them? What social meanings do the designers intend to index when they instruct service workers to adopt a particular style of speaking? Of course, one might also want to ask questions about the extent to which workers actually comply with the instructions they are given, and the meanings actually attributed to their speech by those on the receiving end, i.e. customers. These are important issues, but they are beyond the scope of the present paper. What follows, then, is a
description and analysis of a stylistic ideal, or in corporate language the `brand':