Blommaert, Jan (2010) The Sociolinguistics of Globalization. Cambridge: CUP.
2      A messy new marketplace
Sociolinguistics is the study of language as a complex of resources, of their value, distribution, rights of ownership and effects. It is not the study of an abstract language, but the study of concrete language resources in which people make different investments and to which they attribute different values and degrees of usefulness. In the context of globalization, where language forms are perhaps more mobile than before, such patterns of value and use become less predictable and presupposable. Economic metaphors such as those developed by Bourdieu (1991) are particularly useful for a sociolinguistics of globalization. Recall that Bourdieu saw language as a market of symbolic capital and power, with people juggling for profit and with some people structurally having less capital than others. Bourdieu and his contemporaries Bernstein (1971) and Hymes (1980, 1996) all drew our attention to the same phenomenon: that the world of language is not just one of difference but one of inequality; that some of that inequality is temporal and contingent on situations while another part of it is structural and enduring; and that such patterns of inequality affect, and articulate around, actual, concrete, language forms such as accents, dialects, registers and particular stylistic (e.g. narrative) skills.
The symbolic marketplace described by Bourdieu and others was a local and relatively closed one. Its patterns of value attribution and the logic of the economic game were clear to most of the people involved in the transactions -the speaker from la province knew quite well that their speech was 'inferior' to that of the Parisian, and this awareness accounted for their tendency towards hypercorrection. When we address globalization, however, we address trans-local, mobile markets whose boundaries are flexible and changeable. And this is the theoretical challenge now: to imagine ways of capturing mobile resources, mobile speakers and mobile markets. A sociolinguistics of globalization is perforce a sociolinguistics of mobility, and the new marketplace we must seek to understand is, consequently, a less clear and transparent and a messier one. So let us engage with this issue now and see where it brings us theoretically.
28
A messy new marketplace
29
Figure 2.1 Nina's derriere
2.1       Nina's derriere
A few years ago I was visiting an up-market department store in Central Tokyo, and in the very exclusive and expensive food section of that store I noticed a chocolate shop which bore the name Nina s Derriere.
The stylized lettering and the choice of French betray an aspiration to considerable chic, and the prices of the chocolate on sale materialized that aspiration. But the name of the shop was, let us say, a rather unhappy choice, and I hoped that not too many Japanese customers would know enough French to understand the meaning of the name. I confess that I myself found the thought of offering someone a chocolate obtained from Nina's bum intensely entertaining. It was also useful in bringing an important point home to me. This was not French. At least: while the origins of 'derriere' are clearly French, and while its use in the shop's name drew on indexicals of French chic, the word did not function as a linguistic sign. Linguistically it was only French in a minimal sense, as a word whose origins lie in the stock vocabulary of the language we conventionally call French. Its Frenchness was semiotic rather than linguistic: important was not its linguistic function as a denotational sign, but the emblematic function it had in signalling a complex of associative meanings, the things I captured under the term French chic. This is why the chocolate shop was still in business in spite of its dramatically inappropriate name: the sign did not function linguistically in the context of a Tokyo department store - linguistic knowledge of French being a very rare commodity in Tokyo - but it functioned, and functioned well, emblematically. The sign suddenly becomes a linguistic sign only when someone like me, who has
30 The Socialinguistics of Globalization
Figure 2.2 Keikyu phone card
linguistic competence in French, sees it and reads it as an instance of (linguistic) French. Prior to that the sign is not 'French' but 'Frenchness'. At least, as long as the sign remains in its particular environment. When it is 'exported', so to speak, to the environment that someone like me brings along, it changes.
The world is full of examples of signs that shift functions depending on who uses them, where and for what purpose. We see the same phenomenon in figure 2.2, a phone card from Tokyo, that advertises the Keikyu fast rail connection between Haneda Airport and Central Tokyo.
This simple and mundane object (a typical instance of the multitude of 'unimportant' language objects that litter our world) is amazingly complex both linguistically and semiotically. We see three writing systems - Kanji characters, Katakana script and Roman alphabet - and three 'languages' - Japanese, English and French. The French LeTrain is accompanied by a Katakana transliteration that instructs Japanese customers to read it as rutoren ([fstorsn]). Interestingly, the French here is meant to be spoken, and spoken correctly. The English is confined to one word with. It connects two Kanji words and if we translate the whole phrase, it reads 'Haneda with Keikyu'. The English here is strange. There are perfectly adequate Japanese equivalents for such a function word as with, and even icons such as arrows could have worked. The choice, however, was for English, and there it stands: with. The problem, however, is that unless you know the two Kanji words on either side of it, with has no meaning or function. It is not English linguistically, because English competence does not allow you to understand what with stands for there. You need to know Japanese in order to make sense of with in this phrase, and so, somewhat provocatively, one could say that with is linguistically not English but Japanese. It is English emblematically, like Nina's derriere was French emblematically and (one would certainly hope!) not linguistically.
A messy new marketplace
31
Figure 2.3 'Lced Coffee'
In both examples we see how language material shifts meanings and functions when it is mobile. The French and English elements considered here were mobile semiotic, rather than linguistic, resources. In moving from a space where people have sufficient linguistic competence to project linguistic functions onto the signs (e.g. France in the case of French) to a space where such competences cannot be presupposed (e.g. Japan), the sign changes from a linguistic sign to an emblematic one. It ceases to be something that produces linguistic meanings, because the ones consuming it cannot extract such meanings from the sign. When someone like me then, in Japan, comes across such signs, they re-become linguistic signs.
Or the sign can travel. Globalization involves delocalized production in major industries, including printing, and figure 2.3 shows us an instance of delocalized printing work. The picture was taken in London Chinatown. We see a bilingual poster, advertising, in Chinese and English, a range of cold drinks.
Whereas the Chinese is correctly written, the English contains quite spectacular typos: Lced 'iced', dlrink 'drink', coffee 'coffee', and so forth. Since written signs are always traces of human activity, one can speculate about the processes that generated this poster. Imagine a print shop in China, where a handwritten text is handed down to a typesetter. The typesetter does not know English, and so while the handwritten Chinese characters do not present any difficulties, the English handwritten text is not language but a meaningless design, a set of forms to be
32 The Socialinguistics of Globalization
copied in print. Poor handwriting, such as not closing the first e in coffee, can result in it being read as c instead of e, and thus emerges the printed form coffee. The text only becomes English when it is transferred to London, and posted there against the window of a grocery store in Chinatown. It is in London that people can detect the deviations in spelling, question them or find them amusing. Such value judgments were impossible in the printing shop in China. At the same time of course, for many customers in Chinatown the Chinese in the poster is just a meaningless design. In its transfer to London it ceased to be a linguistic representation and it became a set of forms signalling Chineseness and thus fitting in Chinatown.
I am deliberately overstating my case here, because I wish to emphasize a point: that semiotic mobility has all sorts of effects on the signs that are involved in such mobility. Such processes need to be understood because they are at the heart of globalization as a socio linguistic phenomenon. In the context of globalization, linguistic resources change value, function, ownership and so on, because they can be inserted in patterns of mobility. For this, I would suggest, we need a particular set of conceptual tools, and in the remainder of this chapter I will introduce three central concepts: scales, orders of indexicality and polycentricity.
2.2       Sociolinguistic scales
We have seen in our discussion of Fairclough (2006) how the notion of 'scale' was used essentially to denote spatial scope. We have also seen how this raised several problems, the most important of which was that a purely spatial use of the term de-historicizes the processes it is supposed to capture. So let us see what better use we can find for a notion such as 'scale'. We said above that when people or messages move, they move through a space which is filled with codes, norms and expectations. Scale is a metaphor we can use to imagine such moves.
The metaphor of scale is borrowed from fields such as history and social geography (Swyngedouw 1996; Uitermark 2002). Scales and scaling processes are an important part of the theoretical toolkit of World-Systems Analysis (Wallerstein 1983, 2000). According to World-Systems Analysis, social events and processes move and develop on a continuum of layered scales, with the strictly local (micro) and the global (macro) as its extremes and several intermediary scales (e.g. the level of the state) in between (Lefebvre 2003; also Geertz 2004). Events and processes in globalization occur at different scale-levels, and we see interactions between the different scales as a core feature of understanding such events and processes. Appadurai's (1996) notion of 'vernacular globalization' is a case in point: forms of globalization that contribute to new forms of locality. This locality, however, is destabilized - the immigrant neighbourhood no longer looks like the 'traditional' neighbourhood - because of influences from higher-level scales: migration and diaspora, neighbourhood multilingualism, the presence of the homeland in economies of consumption and in public identity display (Mankekar 2002).
A messy new marketplace
33
The point of departure: horizontal and vertical metaphors
The point of departure for what follows is the non-unified nature of socio-linguistic phenomena. The point has often been noted: acts of communication are all uniquely contextualized, one-time phenomena; yet we understand them because of their manifest lack of autonomy: their consistence with previous traditions of making sense, their connection to shared, enduring (i.e. historical) patterns of understanding such as frames. This dual nature of language practices, both as an individual, one-time and unique phenomenon and, simultaneously, as a collective and relatively stable phenomenon, has often been captured under labels such as 'micro' and 'macro'. The connection between such levels has often been described as complex, difficult, unfathomable. Yet, several very useful theoretical tools have been developed, explicitly identifying the instantaneous transition from one level to another in communication: Gumperz's (1982) notion of 'contextualization', Goffman's (1974) 'frames', the Bakhtinian concept of 'intertextuality' (as further developed, e.g., by Fairclough 1992) and Bourdieu's (1990) 'habitus' - to name just the most widely recognized ones.
In all cases, the concepts identify the jump from one scale to another, from the individual to the collective, the temporally situated to the trans-temporal, the unique to the common, the token to the type, the specific to the general. And the connection between such scales is indexical: it resides in the ways in which unique instances of communication can be captured indexically as 'framed', understandable communication, as pointing towards socially and culturally ordered norms, genres, traditions, expectations - phenomena of a higher scale-level. The capacity to achieve understanding in communication is the capacity to lift momentary instances of interaction to the level of common meanings, and the two directions of indexicality (presupposing - the retrieval of available meanings - and entailing - the production of new meanings; Silverstein 2006a: 14) are at the heart of such processes.
Reviewing current theorizing about such scalar phenomena, we see that a lot of thinking has gone into the connections and movements - sophisticated concepts such as 'intertextuality' and 'entextualization' are results of that. What exactly it is that is connected and moved between, however, has by and large been neglected as an area of theorizing. One effect has been that notions of 'contextualization' (the process of conversion) have been better developed than notions of 'context', the spaces in and between which contextualization happens. (See Hanks 2006 for a recent survey.) I have been using the term scale here as an attempt to at least provide a metaphor that suggests that we have to imagine things that are of a different order, that are hierarchically ranked and stratified. The metaphor suggests spatial images; however, these images are vertical metaphors of space rather than horizontal ones, which are implicit in terms such as distribution, spread, and even community and culture, among
34 The Socialinguistics of Globalization
others. Scales offer us a vertical image of space, of space as stratified and therefore power-invested; but they also suggest deep connections between spatial and temporal features. In that sense, scale may be a concept that allows us to see sociolinguistic phenomena as non-unified in relation to a stratified, non-unified image of social structure. Note that the introduction of 'scale' does not reject horizontal images of space; it complements them with a vertical dimension of hierarchical ordering and power differentiation. Let us look at these aspects of scales in some detail.
Scales as semiotized space and time
To be sure, a notion such as 'scale' is the construction of an analytic image -something which Wallerstein warns us is an invention of social-scientific, traditional thought (Wallerstein 1997, 2001: chapter 10). In particular, our current attempt at 'spatializing' sociolinguistic theory risks being flawed by that institutional problem inscribed in the division of labour between the social sciences: the separation of time and space as different aspects of social life and social phenomena. Against this separation, Wallerstein pits the notion of TimeSpace - a 'single dimension' which locks together time and space (Wallerstein 1997: 1; also Fals Borda 2000). Every social event develops simultaneously in space and in time, often in multiply imagined spaces and timeframes. So here is one critical qualification: a notion such as scale refers to phenomena that develop in TimeSpace. Scale is not just a spatial metaphor.
Talk about 'time' and 'space', however, is slippery, and we must add a second necessary qualification. The phenomena that develop in TimeSpace are social phenomena, and the TimeSpace in which they develop is consequently an 'objective' (physical) context made social. It is an often repeated assertion: people make physical space and time into controlled, regimented objects and instruments, and they do so through semiotic practices; semiotized TimeSpace is social, cultural, political, historical, ideological TimeSpace (Lefebvre 2003; also Haviland 2003; Goodwin 2002). A third necessary qualification to be added follows from the previous one. The semiotization of TimeSpace as social contexts always involves more than just images of space and time. As we shall see, a move from one scale-level to another invokes or indexes images of society, through socially and culturally constructed (semiotized) metaphors and images of time and space. The general direction of such moves can be formulated as follows:
	Lower scale	Higher scale
Time	Momentary	Timeless
Space	Local, situated	Translocal, widespread
A messy new marketplace
35
In social interaction, such TimeSpace moves - 'scale-jumping', as they are called by Uitermark (2002: 750) - are converted into interactional patterns that index norms, expectations and degrees of generalness of positions. They are converted, in other words, into statements that index social order, and the TimeSpace imagery provides rich indexicals (sometimes iconically) for aspects of a real or imagined social order. Consider, by way of illustration, the following bit of (imagined) interaction between a tutor (T) and a PhD student (S):
s: I'll start my dissertation with a chapter reporting on my fieldwork t: We start our dissertations with a literature review chapter here.
The tutor performs a scale-jump here, in which he moves from the local and situated to the translocal and 'general', invoking practices that have validity beyond the here-and-now - normative validity. This 'upscaling' is articulated through a change from personal and situated to impersonal and general -compare S's use of / and my, as well as his use of the future tense, with T's we and our, T's use of the timeless present and his invocation of here: a community larger than just the student and the tutor. The student's utterance was centred on his own work and plans; the tutor's response re-centres it on a higher scale-level: that of the larger academic community and institutional environment of which both are part. The individual plan of the student is countered by an invocation of general rules and norms, valid 'here' (i.e. valid for the particular student as well). The tutor's move is a vertical move performed in a stratified, hierarchically layered system, in which higher scale-levels (institutional and community norms and rules) prevail over lower scale-levels (the individual concerns of the student). It is, of course, a power move in which a higher level of relevance, truth, validity or value is called in to cancel the suggestion made by the student, in which individuals have been replaced by institutionally circumscribed roles, and in which the specific case is measured against categories of cases: from token to type, from contextualized to decontextualized. The scale-jump thus made is a complex one, in which various kinds of semiotic transformations occur:
Lower scale
Higher scale
Momentary
Local, situated
Personal, individual
Contextualized
Subjective
Specific
Token
Individual
Diversity, variation
Timeless
Translocal, widespread
Impersonal, collective
Decontextualized
Objective
General, categorial
Type
Role
Uniformity, homogeneity
36 The Socialinguistics of Globalization
And all of this is produced through simple grammatical, stylistic and generic operations in the utterance: small formal cues that release dense indexical meanings.
The fact that these operations are performed here by the tutor and not by the student is, of course, not accidental. As Uitermark (2002) notes, some people or groups can jump scales while others cannot, and 'outscaling' is a frequent power tactic: lifting a particular issue to a scale-level which is inaccessible to the other, as when a lawyer shifts into legalese or a doctor into medical jargon. Jumping scales depends on access to discursive resources that index and iconicize particular scale-levels, and such access is an object of inequality. As Conley and O'Barr's (1990) work on small-claims courts demonstrated some time ago, discursive resources that are empowering at one scale-level (e.g. issue-centred emotive discourses) can be disempowering at higher scale-levels (where a law-centred rational discourse dominates). Power and inequality are features of scaling, of the asymmetrical capacity to invoke particular scale-levels in the interpretation of an act; scales provide contexts with possible regulations of access.
The simple lexical and grammatical operations performed by the tutor, thus, trigger a whole range of indexical shifts, redefining the situation, the participants, the topic, the scope of 'acceptable' statements on the topic, and so forth; they also firmly set the event in a normative, general norm-oriented frame. This complex indexical shift can now be described not as a series of individual operations, but as one vertical move within a stratified social meaning system, enabling and mobilizing the various forms of indexical re-ordering of the statement. Introducing a notion such as 'scale' for describing current phenomena in communicative action has the advantage of introducing a layered, stratified model of society as a frame for the interpretation of such phenomena. Power and inequality thus become incorporated into our ways of imagining such phenomena, and rather than seeing them as an exceptional aberration in social life (as in many analyses focused on power), they can be seen as an integral feature of every social event. It is the new image of society introduced by the tutor's statement that organizes the new indexical order: he introduces a rigid, norm-oriented, trans-personal social space - a different power regime for the interaction which reorganizes the 'footing' (Goffman 1981) on which the participants can interact with one another.
In sum, scales need to be understood as 'levels' or 'dimensions' (Lefebvre 2003: 136-150) at which particular forms of normativity, patterns of language use and expectations thereof are organized. Scalar processes are processes of shifts between such scales, and we should recall that such shifts involve complex re-semiotizations of TimeSpace: new images of time and space, new patterns of acting upon them. In this more complex understanding, the notion
A messy new marketplace
37
of scale' may allow us to understand the dynamics between local and translocal forces discussed in chapter 1 as well as in the examples given earlier in this chapter. Different scales can interact, collaborate and overlap or be in conflict with one another, because each time there are issues of normativity at play. To these issues we now need to turn.
2.3       Orders of indexicality
Different scales organize different patterns of normativity, of what counts as language. This may quickly be turned into an image of chaos and of fragmentation (and quite a lot of post-modernist literature would make this interpretation), but for reasons that are discussed later in this book, such interpretations are not helpful. The processes we see are not chaotic but ordered, although they are of considerable complexity. Normativity is, by its very essence, a form of organization and order. So we need to look for conceptual tools that help us imagine this complex form of organization and order, and I will turn to the notion of 'orders of indexicality' for that (cf. Blommaert 2005: 69ff).
The point of departure is quite simple: indexicality, even though largely operating at the implicit level of linguistic/semiotic structuring, is not unstructured but ordered. It is ordered in two ways, and these forms of indexical order account for 'normativity' in semiosis. The first kind of order is what Silverstein (2003a) called 'indexical order': the fact that indexical meanings occur in patterns offering perceptions of similarity and stability that can be perceived as 'types' of semiotic practise with predictable (presupposable/entailing) directions (see also Agha 2003, 2005). 'Register' is a case in point: clustered and patterned language forms that index specific social personae and roles can be invoked to organize interactional practices (e.g. turns at talk, narrative), and have a prima facie stability that can sometimes be used for typifying or stereotyping (e.g. 'posh' accents - see Rampton 2003). Speaking or writing through such registers involves insertion in recognizable (normative) repertoires of 'voices': one then speaks as a man, a lawyer, a middle-aged European, an asylum seeker and so forth, and if done appropriately, one will be perceived as speaking as such (Agha 2005). Thus, indexical order is the metapragmatic organizing principle behind what is widely understood as the 'pragmatics' of language.
Such forms of indexical order sometimes have long and complex histories of becoming (Silverstein 2003a and Agha 2003 offer excellent illustrations). These histories are often connected to the histories of becoming of nation states and to their cultural and socio linguistic paraphernalia - the notion of a' standard language' and its derivative, a particular 'national' ethnolinguistic identity (Silverstein 1996, 1998; Errington 2001). Yet, they also display a significant degree of variability and change, they can erupt and fade under pressure of
38 The Socialinguistics of Globalization
macro-developments such as capitalist consumer fashions, as is evident from Silverstein's (2003a, 2006b) oenologia - the register of contemporary wine connoisseurs (see also Agha 2005, 2007). Indexical order of this sort is a positive force, it produces social categories, recognizable semiotic emblems for groups and individuals, a more or less coherent semiotic habitat.
It does so, however, within the confines of a stratified general repertoire in which particular indexical orders relate to others in relations of mutual valuation - higher/lower, better/worse. This is where we meet another kind of order to indexicalities, one that operates on a higher plane of social structuring: an order in the general systems of meaningful semiosis valid in groups at any given time. This kind of ordering results in what I call orders of indexicality - a term obviously inspired by Foucault's 'order of discourse'. Recall that Foucault was interested in the general rules for the production of discourses: their positive emergence as well as their erasure and exclusion. He started from the hypothesis
that in every society the production of discourse is at once controlled, selected, organized and redistributed by a certain number of procedures whose role is to ward off its powers and dangers, to gain mastery over its chance events, to evade its ponderous, formidable materiality. (Foucault 1984 [1971]: 109; see also his notion of'archive', Foucault 2002 [1969]: chapter 5)
If we now paraphrase Foucault's hypothesis we see that ordered indexicalities operate within large stratified complexes in which some forms of semiosis are systemically perceived as valuable, others as less valuable and some are not taken into account at all, while all are subject to rules of access and regulations as to circulation. That means that such systemic patterns of indexicality are also systemic patterns of authority, of control and evaluation, and hence of inclusion and exclusion by real or perceived others. This also means that every register is susceptible to a politics of access. And it also means that there is an economy of exchange, in which the values attached by some to one form of semiosis may not be granted by others: the English spoken by a middle-class person in Nairobi may not be (and is unlikely to be) perceived as a middle-class attribute in London or New York.
'Order of indexicality' is a sensitizing concept that should index ('point a finger to') important aspects of power and inequality in the field of semiosis. If forms of semiosis are socially and culturally valued, these valuation processes should display traces of power and authority, of struggles in which there were winners as well as losers, and in which, in general, the group of winners is smaller than the group of losers. The concept invites different questions -sociolinguistic questions on indexicality - and should open empirical analyses of indexicality to higher-level considerations about relations within sociolinguistic repertoires, the (non-)exchangeability of particular linguistic or semiotic
A messy new marketplace
39
resources across places, situations and groups, and so forth. It invites, in sum, different questions of authority, access and power in this field.
2.4 Polycentricity
One such question is: how do we imagine these patterns of authority and power? One way of answering that question is to suggest that authority emanates from real or perceived 'centres', to which people orient when they produce an indexical trajectory in semiosis. That is, whenever we communicate, in addition to our real and immediate addressees we orient towards what Bakhtin (1986) called a 'super-addressee': complexes of norms and perceived appropriateness criteria, in effect the larger social and cultural body of authority into which we insert our immediate practices vis-a-vis our immediate addressees. And very often, such authorities have names, faces, a reality of their own; they can be individuals (teachers, parents, role models, the coolest guy in class), collectives (peer groups, sub-cultural groups, group images such as 'punk', 'gothic' etc.), abstract entities or ideals (church, the nation state, the middle class, consumer culture and its many fashions, freedom, democracy), and so on: the macro- and micro-structures of our everyday world. The point is: we often project the presence of an evaluating authority through our interactions with immediate addressees, we behave with reference to such an evaluative authority, and I submit we call such an evaluating authority a 'centre'.
The authority of centres is evaluative, and it often occurs as an authority over clusters of semi otic features, including thematic domains, places, people (roles, identities, relationships) and semiotic styles (including linguistic varieties, modes of performance etc.). Thus, broaching a particular topic will trigger a particular semiotic style and suggest particular roles and relationships between participants, and certain types of communicative events require appropriate places and occasions - Not here! Not now! Not while the children are listening! (Scollon and Scollon 2003, Blommaert, Collins and Slembrouck 2005a). One speaks differently and as a different person about cars or music than about the economy or about sex. In one instance, one can speak as an expert using a particular register indexing membership of expert groups, in other instances one can speak as a novice; one can shift from a very masculine voice on a particular topic (e.g. sex or cars) to a gender-neutral voice (e.g. when discussing the war in Iraq), each time also shifting registers, often even accents, pace, tone and rhythm (a declarative tone on one topic, a hesitant one on another). And topics, styles and identities belong to specific places and are excluded from other places (a thing that becomes apparent during after-hours escapades at scientific conferences). Each time one orients towards other centres of authority offering
40 The Socialinguistics of Globalization
ideal-types of norms or appropriateness criteria, as it is called in pragmatics: the places where 'good' discourse about these topics is made.
It is the packaging of topic, place, style and people that makes up the index-ical direction of communication: the fact that certain topics require specific semiotic modes and environments, and so organize identities and roles (Agha 2005). Goffman (1981), as we already saw, called such patterns 'shifts in footing': delicate changes in speaker position that are accompanied by shifts in linguistic and semiotic modes and redefine the participant roles in the interaction. We are now in a position to empirically 'dissect' footing and bring it in line with larger organizational features of life in society.
It is obvious that even though places impose rules and restrictions on what can happen in communication there, every environment in which humans convene and communicate is almost by definition polycentric, in the sense that more than one possible centre can be distinguished. One can follow norms or violate them at any step of the process, and sometimes this is wilfully done while on other occasions it comes about by accident or because of the impossibility of behaving in a particular way. Again, Goffman's descriptions of the multiple layers that characterize mundane interaction scenes are informative. For instance, Goffman distinguishes between 'focal' and 'non-focal' activities occurring in the same event - as when a pupil in class produces an offensive reaction to a teacher's question, giving off negative impressions (focal, for the teacher) as well as positive ones (non-focal, towards his peer group who studiously try to avoid being classified as 'nerds'). In our own research on asylum seekers' narratives, we often found that 'truthful' accounts by the applicant were interpreted as 'implausible' (i.e. untruthful) by the interviewers, because describing the chaotic and often paradoxical realities truthfully often iconically resulted in a chaotic and paradoxical story. Interviewees oriented towards 'the truth' as defined by situated, densely con-textualized realities in countries like Africa, for example, while interviewers oriented towards a particular textual (bureaucratic) ideal of decontextualizable coherence, linearity and factuality (Blommaert 2001a; chapter 6 below). Both centres were always present in such a polycentric interview situation, although the interviewers' centre was often 'non-focal', kept in the background during the interview itself. Thus, in telling 'the truth', the applicants were often 'wrong-footed' by the interviewers; in the real world, the dominant order of indexicality is that of the interviewer and their bureaucratic apparatus.
Polycentricity is a key feature of interactional regimes in human environments: even though many interaction events look 'stable' and monocentric (e.g. exams, wedding ceremonies), there are as a rule multiple - though never unlimited - batteries of norms to which one can orient and according to which one can behave (as when the bride winks at the groom when she says 'I do'). This multiplicity has been previously captured under terms such as 'polyphony'
A messy new marketplace
41
or 'multivocality'. A term such as 'polycentricity' moves the issue from the descriptive to the interpretive level. Again, my attempt here is aimed at sensitizing others to the fact that behind terms such as 'polyphony', social structures of power and inequality are at work. Such structures - orders of indexicality -account for the fact that certain forms of polyphony never occur while other forms of polyphony miraculously seem to assume similar shapes and directions. The bride can wink at her groom, but baring her breasts would be highly unusual. Certain voices, like the bureaucratic one in the asylum system, systemi-cally prevail over others, because the impact of certain centres of authority is bigger than that of others. The multiplicity of available batteries of norms does not mean that these batteries are equivalent, equally accessible or equally open to negotiation. Orders of indexicality are stratified and impose differences in value onto the different modes of semiosis, systematically give preference to some over others, and exclude or disqualify particular modes.
Both concepts, 'order of indexicality' and 'polycentricity', thus suggest a less innocent world of linguistic, social and cultural variation and diversity, one in which difference is quickly turned into inequality, and in which complex patterns of potential-versus-actual behaviour occur. They also enable us to move beyond the usual sociolinguistic units - homogeneous speech communities - and to consider situations in which various 'big' sociolinguistic systems enter the picture, as when people migrate in the context of globalization, or when in the same context messages start moving across large spaces. In both cases, people do not just move across space; given what has been said above, we also realize that they move across different orders of indexicality. Consequently, what happens to them in communication becomes less predictable than what would happen in 'their own' environment. Sociolinguistics in the age of globalization needs to look way beyond the speech community, to sociolinguistic systems and how they connect and relate to one another. Big things matter if we want to understand the small things of discourse.
2.5       A sociolinguistics of mobile resources
Power and mobility
Scales, orders of indexicality, polycentricity: we now have the basis for a little vocabulary that will enable us to talk about language in globalization in a different way. The three concepts I introduced had several things in common. One such thing was their emphasis on power: all of the concepts suggest layering and stratification in hierarchical systems of value for semiotic resources. In any and every social interaction, specific semiotic forms will be expected and be valued as the 'best possible' resources. In pragmatics, such phenomena of selection for preference have been described for decades in terms
42 The Socialinguistics of Globalization
ranging from 'felicity conditions' to 'appropriateness criteria' (see e.g. Levinson 1983), and let us not forget 'politeness' research as a culmination point in such studies (cf. Eelen 2000). Likewise, the Hymesian notion of 'communicative competence' has often been described as the capacity for 'adequate' linguistic performance in a given social situation. Understanding what such forms of adequacy and appropriateness may be in a context which we are now compelled to imagine as complex and mobile, requires a new vocabulary. The three terms I propose here can be assessed with regard to usefulness.
A second feature shared by the three concepts is their spatiotemporal sensitivity. They need to be read in sequence to be clearly understood like that. In other words, sociolinguistic phenomena in a globalization context need to be understood as developing at several different scale-levels, where different orders of indexicality dominate, resulting in a polycentric 'context' where communicative behaviour is simultaneously pushed and pulled in various directions. The French on the Keikyu phone card was both strictly local (a Tokyo sign) as well as translocal (a French sign); the translocal scale-level, however, was not equally accessible to every consumer of the sign. For local people operating within a local order of indexicality, the French made good sense as an emblem of 'Frenchness', with its connotations of chic. For people such as me, who brought along a practical linguistic competence in French, it made sense in a very different way: it became a 'typically Japanese' way of using French, that is, a form that is detached from its conventional 'French' (i.e. translocal and linguistic) functions and is relocated in a local, Tokyo semiotic economy. It became, in short, an exotic French form, and note that the qualification of 'exotic' is an assessment made at one particular scale-level and within one particular order of indexicality - that of the translocal scale. Resetting this Keikyu phone card from its local, Tokyo, context to its translocal, 'French', context involved an operation in which the sign was resemiotized, subject to very different procedures of meaning attribution. The sign was made to travel, it was mobilized by a feature of globalization: a tourist (me) picking it up in its place of origin (Tokyo) and subjecting it to my own (Francophone) procedures of 'reading' and 'decoding'. By doing so, I lifted the sign out of its TimeSpace frame and brought it into quite a different TimeSpace frame, one for which it was not intended, and one in which it could, consequently, acquire 'exotic' meanings.
It should be clear that we are relatively far removed from a traditional sociolinguistic discourse of language mixing and shifting here. Conventional approaches to code-switching would not be able to tell us much about signs such as the 'Haneda with Keikyu' or the 'LeTrain' examples given here. For one thing, there is far more than 'language' (in the sense of 'Japanese', 'English' and 'French') at play here. As I noted in the beginning, the Keikyu phone card contains three such 'languages' (even if in minimal, one-word forms) as well as
A messy new marketplace
43
three scripts and a compound image of a plane and a train. It is, of course, a multimodal sign, and shifts occur here in a vastly more complex field than that of 'language' alone. What we observe is clearly not just a linguistic problem, it is a semiotic one. Consequently, we need to look at resources, actual situated resources as deployed by real people in real contexts, and recontextualized by other real people such as myself. We are witnessing repertoires here, constructed out of bits and pieces of conventionally defined 'languages' and concretely assuming the shape of registers and genres, of specific patterns of language in communicative forms such as a phone card, a poster or a shop sign. The target of our analysis is resources, and even if such resources can be conventionally tagged as 'belonging' to language X or Y, it is good to remember that the whole point is about the dislodging of such resources from their conventional origins. The French on the Keikyu phone card becomes a problem not because of its linguistic features - not because it is 'French' - but because of the particular ways in which it has penetrated the semiotic repertoire of people in Tokyo and has acquired meanings and functions there. This, I would say, is a sociolinguistics of mobile resources, no longer a sociolinguistics of immobile languages.
Locked in space: on linguistic rights
Naturally, this theoretical stance comes at a price. Such forms of inter-language penetration have often been captured in a totalizing but comfortable discourse of language(s), with which we now have to disagree. There is by now a well-entrenched and very respectable branch of sociolinguistics which is concerned with describing the world of globalization from the perspective of linguistic imperialism and 'linguicide' (Phillipson 1992; Skutnabb-Kangas 2000), often based on particular ecological metaphors. These approaches start from a sociolinguistics of distribution and oddly assume that wherever a 'big' and 'powerful' language such as English 'appears' in a foreign territory, small indigenous languages will 'die'. There is, in this image of sociolinguistic space, place for just one language at a time. In general, there seems to be a serious problem with the ways in which space is imagined in such work. In addition, the actual sociolinguistic details of such processes are rarely spelled out - languages can be used in vernacular or in lingua franca varieties and so create different sociolinguistic conditions for mutual influencing; English sometimes 'threatens' other former colonial languages such as French, Spanish or Portuguese, rather than the indigenous languages (a phenomenon noted primarily in former exploitation colonies, and less prominent in former settlement colonies; see Mufwene 2005, 2008); or sometimes the 'threat' to indigenous languages can come from dominant local ('indigenous') languages rather than English, as we shall see further in this book. So there are several major problems with the literature on linguistic rights.
44 The Socialinguistics of Globalization
One major problem is the way in which authors appear to assume the spatial 'fixedness' of people, languages and places. The discourse of minority rights is in general a discourse of strict locality, and the first lines of the UN Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities read: 'States shall protect the existence and the national or ethnic, cultural, religious and linguistic minorities within their respective territories, and shall encourage conditions for the promotion of that identity' (quoted in Skutnabb-Kangas and Phillipson 1999). This Declaration is an agreement between states, which are here presented as territorially bounded entities in the space of which a particular regime can and should be developed with respect to 'minorities', defined in the same move as minorities within that particular ('state') territory. The rights granted by this Declaration are territorially bounded and organized rights, and distinctions between groups evolve along the classic Herderian triad of territory-culture-language (Blommaert and Verschueren 1998).
This discourse of locality is usually couched in environmental-ecological metaphors: a particular place is characterized by specific features ranging from climate through biodiversity to people, cultures and languages. The relationship between these different components is seen as a form of synergy: it is through human variability that diversity in the environment is sustained, for the languages and cultures of local people provide unique views on this environment and help sustain it. See, for example, the point of view articulated by one of the most vocal advocates of linguistic rights, Terralingua (1999, from their website http://cougar.ucdavis.edu/nas/terralin/learn.html):
We know that a diversity of species lends stability and resilience to the world's ecosystems. Terralingua thinks that a diversity of languages does the same for the world's cultures - and that these manifestations of the diversity of life are interrelated.
This diversity is invariably seen as something that needs to be preserved, consequently. It literally needs to be 'kept in place'. To go by the words of Skutnabb-Kangas and Phillipson (1995: 84): '[t]he perpetuation of linguistic diversity can ... be seen as a recognition that all individuals and groups have basic human rights, and as a necessity for the survival of the planet, in a similar way to biodiversity' (see also Skutnabb-Kangas and Phillipson 1999; Nettle and Romaine 2000; see May 2001, Blommaert 2001b and Mufwene 2002 for a critique).
There is a linguistic-ideological dimension to this, in which it is assumed that language functions in a community because it provides local meanings: meanings that provide frames for understanding the local environment, to categorize and analyse the (strictly) local world. References to the unique worldviews enshrined in these languages often revolve around local functionality as well: the worldviews are expressed in terms and grammatical relations that address or articulate a local decoding of the world. Let us return to Skutnabb-Kangas and Phillipson 1995: 89):
A messy new marketplace
45
Linguistic diversity at local levels is a necessary counterweight to the hegemony of a few 'international' languages. The 'World Languages' should, just as roads and bridges, be seen as tools for communication of ideas and matter, but the creation of authentic ideas and products (instead of mass-products) is in most cases necessarily best done locally.
The worldviews are invariably local (or territorialized) worldviews, linked to particular regional surroundings. A people's language localizes these people, it sets them within a particular, spatially demarcated ecology.
It is this view of local functionality that underpins the strong claims, cited above, that the survival of minority languages is crucial for the survival of the planet, for with every language that disappears a uniquely functional local set of meanings about the environment is lost. Languages are seen as local repositories of knowledge, and such local forms of knowledge are essential for understanding the (local) world. Consequently, when people are moved into a different environment, the language may lose (part of) its functions. Conversely, when another language is introduced in a particular environment, it may as well be dysfunctional for it does not articulate the particular local meanings required for the sustenance of this environment. This idea in turn underpins the idea of linguistic imperialism, invariably conceived as a non-local language (usually the ex-colonial language, and usually English) penetrating or invading local spaces and disturbing the ecological balance that existed between people, their language and culture, and their environment (Skutnabb-Kangas 2000; Skutnabb-Kangas and Phillipson 1995, 1999 and Heugh 1999 provide examples).
In sum, what we see here is how language functions are territorialized, tied to particular local environments apparently constructed as static. Language apparently works excellently in its own, original place, and it loses functions as soon as the stable, original, 'autochthonous' (or 'native', 'aboriginal') link between language and place is broken. Consequently, a programme aimed at stimulating or promoting these local languages (invariably mother tongues of apparently inherently monolingual and monocultural people) ties the speakers of these languages to a place and reinforces the presumed fixed connection between people and their environment - a clear reflex of the Saussurean synchrony.
All of this sounds more or less acceptable, at least when some aspects of reality are conveniently overlooked. A rather disturbing aspect of contemporary reality, as we know, is mobility. In contemporary social structures, people tend to move around, both in real geographical space and in symbolic, social space. And all of these processes of mobility appear to display complex connections with language (Rampton 1995, 1999), including language attitudes and language planning.
Language as a social thing, i.e. something in which people have made investments and to which they have attributed values, seems to have awkward relations to space, the main axes of which are those of territorialization and deterritorialization. Territorialization stands for the perception and attribution of
46 The Socialinguistics of Globalization
values to language as a local phenomenon, something which ties people to local communities and spaces. Customarily, people's mother tongue (LI) is perceived as 'territorialized language', alongside orality and the use of dialects. All of these forms of language emanate locality. Conversely, deterritorialization stands for the perception and attribution of values to language as something which does not belong to one locality but which organizes translocal trajectories and wider spaces ('scale-jumps' in the terminology developed earlier in this chapter). Second or other languages (L2) as well as lingua francas and diaspora varieties, standardized varieties and literacy are seen as 'deterritorialized language', language that does not exclusively belong to one place (Jacquemet 2000, 2005; Maryns and Blommaert 2001).
Language variation allows, defines and organizes spatial trajectories. Literacy allows a text to be moved, both physically across spaces in the world, as well as symbolically, across social spheres and scales. A standard variety of a language allows moving to adjacent places where people speak similar dialects, as well as across social spaces, into the elite. International languages such as French or English allow insertion in large transnational spaces and networks as well as access to the elites. The different 'types' of languages, in short, allow access to different scales. All of these scalar patterns of mobility are real, they revolve often around life-chances and opportunities, and consequently people often articulate relations between language or code choices and spaces. The choice for English or French rather than indigenous languages in education is at the grassroots level often motivated by means of discourses of 'getting out of here' and towards particular centres - metropolitan areas - where upward social mobility at least looks possible.
Moving through the various levels of education often involves moving through layered, scaled regimes of language, each time seen as enabling deterritorialization and hence social as well as geographical mobility. Senses of belonging to a particular community conversely often go hand in hand with the creation (or re-creation) of particular varieties that tie people to that community while at the same time indexing displacement and deterritorialization. 'Gangsta' English, for instance, is widespread in African urban centres as a language of the townships and the slums, where particular, often imaginarily violent youth cultures develop (see chapter 7 below for a more extensive discussion). Such linguistic ideologies connecting language varieties to dynamics of locality and mobility, active both at the folk and at institutional levels, often foster resistance to the promotion of indigenous, minority languages - a point often reported by fieldworkers, but rarely written down in publications.
Although one might deplore this, the reasons are usually sound enough. Symbolic marginalization is often just one correlate of real, material margin-alization (Fraser 1995, Stroud 2001); LI promotion is a form of symbolic upgrading of marginalized resources, and resistance is often based on an acute
A messy new marketplace
47
awareness of the persistence of real marginalization. If performed within a monoglot strategy (i.e. a strategy aimed at constructing 'full monolingualism' and rejecting bilingualism as a road to language attrition or language death), LI promotion, is thus seen as an instrument preventing a way out of real marginalization and amounting to keeping people in their marginalized places and locked into one scale-level: the local. Imagine a family in the very marginalized and poor north-eastern parts of South Africa, speaking Venda. Education in Venda is likely to be perceived as keeping people in the marginalized region, as long as good, white-collar jobs and higher education are in effect concentrated in places like Johannesburg - and require access to English and/or Afrikaans. If the family wants to offer its children upward social mobility, then, it needs to offer them geographical mobility and consequently linguistic mobility as well. Language shift, under such conditions, is a strategy for survival. In the eyes of the speakers, the upgrading of marginalized symbolic goods may still be seen as less empowering than the creation of access to the real prestige goods. Mufwene (2002: 377) captures the core of this 'wicked problem' well: '[i]t sometimes boils down to a choice between saving speakers from their economic predicament and saving a language'.
The crux of the matter is that we need to think of issues such as linguistic inequality as being organized around concrete resources, not around languages in general but specific registers, varieties, genres. And such concrete resources follow the predicament of their users: when the latter are socially mobile, their resources will follow this trajectory; when they are socially marginal, their resources will also be disqualified. In both cases, the challenge is to think of language as a mobile complex of concrete resources. If we fail to do that, we risk drawing a caricature of social realities and becoming very upset about that caricature rather than about an accurate replica of social processes. This, needless to say, is a pointless exercise. The matter is of fundamental importance and is easily misunderstood, and this is why I am emphatic about it. In what follows I will try to provide a clear illustration of what we are talking about: the mobility of concrete semiotic resources (not 'languages') in a globalized context. There is always a tension between an ideologically perceived 'language' and socio-linguistically perceived 'resources', as we shall see presently; globalized economic forces exacerbate and exploit such tensions.
2.6       Selling accent
Language policy revolves around the production and enforcement of norms for language use, and its success is measured by the degree to which policy-preferred norms are accepted and spread. Traditionally, the state is the major player in the field of language policy. It regulates which language(s) and forms of literacy are 'official' and 'national', and it imposes rules and constraints on
48 The Socialinguistics of Globalization
the use of languages and scripts in its realm. Usually, the state was and is concerned with standardized 'languages', that is, with one layer of linguistic variation. The state, however, has always had to share the space of norm definition and normative conduct with other actors - the family is one very important such actor, while civil society actors such as churches are others. Media, by and large, traditionally supported the 'official' norms imposed by the state. Intralanguage variation - dialects, accents - was and is within the purview of the individual citizens or of groups using them to flag particular identities. At the national level, they are very often seen as the fabric of national identity and local (sub-state) authenticity and often cherished as such (see e.g. Elmes 2005).
I have to cut quite a few corners here, but the point I wish to make is that (1) the state is traditionally a very powerful actor in the field of language norms and traditionally has the monopoly of formal policy making; but (2) the state has never been the only player in the field of language norms. It has always been a major player, but never a completely hegemonic player; there is always a form of polycentricity, a division of labour between the state and other actors in this field, and formal language policies compete with the language politics of other actors in complex webs of language policing activities. (3) The language policies of the state are traditionally aimed at 'languages' only; the state is usually tolerant when it comes to the co-existence of a multitude of dialects and accents in the national language(s) on its territory. In Belgium, for instance, the state makes no effort to combat regional variation in the national languages French, Dutch and German, even if the education system forcefully promotes standard varieties of these languages, and even if a modest complaints culture exists about the use of regionalisms and dialects in the media.
I will use these general introductory remarks as the backdrop for an argument that runs as follows. The traditional tolerance of state policies towards intralanguage variation such as dialects and accents is not matched by the politics of language of globalized, private enterprise actors. While the state focuses on language, new actors of language commodification focus on accent and discourse, thus creating a market in which sharp distinctions between speaking right and speaking wrong are articulated. Such distinctions draw on globalized orders of indexicality, normative complexes in which imageries of global success and failure are used, and English - the language that defines globalization - is of course the core of such orders. The outcome of this is a competitive market not just of English but of English accents which defies the traditional tolerance of the state policies, as well as popular (and academic) perceptions of accent as producing authenticity. It creates a new commodified dialectology and raises quite complex issues on normativity and identity, as well as on the shifting balance between formal language policies and equally formalized language politics in an age of globalization. It is an instance of Foucaultian 'policing' - the rational production of order - and it works through
A messy new marketplace
49
an infinitely detailed attention to conduct (what Foucault (2005, 2007) called 'the care of the self, the perpetual micro-practices of subjectivity). Accent courses produce a regimented subject that is subjected to rules of 'normal' speech - speech that is invisible because it is uniform and homogeneous. Let us now examine the procedures by means of which this subjectification is effected.
The Internet and the commodification of accent
Replacing to some extent the older industry of correspondence courses, the Internet offers a wide and virtually uncontrolled space for language learning packages. They come in all shapes and sizes, and I will focus on websites offering courses in American accent. I do this because they manifestly bank on two different things: (1) the worldwide 'stampede towards English' (de Swaan 2001) inspired by the global perception of English as the language that defines upwardly mobile trajectories; (2) a particular imagery of the USA and American cultural symbols as being in the forefront of globalization and of upward global mobility. In other words, we get the so-called 'McDonaldization' of the world caught metonymically in packages for acquiring an American-sounding variety of English. What is spread is, of course, not just the product but also the adjectives used to qualify it, not just the language as linguistic structure but the language as a densely loaded ideological format, something that is far more than a language but also an acquirable imagery of the self as being 'in the world'.
Let me note from the outset that, although American accent websites may offer us the most outspoken examples of such dense ideological packaging, websites offering packages in other languages and varieties do the same, be it slightly less explicitly. They all offer language and the social trajectories it is supposed to provide or open. American accent websites are not unique but illustrative of a broader pattern of ideological packaging of commodified languages and varieties. We should also observe right from the start that these websites never offer 'a' language in its totality. What they offer is a register - a specific bit of language tailored to the immediate needs or desires of the customers. What is offered is something that gives the impression of language, a pragmatic and metapragmatic component to language competence that indexically emanates the right ideological package. The point is not to learn American English, but to sound like an American. The language policing here operates on sub-language objects.
Note also that the real US society is of course very much a 'multi-accent' society. Apart from the well-known regional varieties (e.g. Texan, Appalachian) a visit to any big hotel in the USA teaches us that Latino, Eastern European and Asian accents are very much accepted as working varieties of American English. The 'American accent' sold by the Internet companies we consider here is often
50
The Socialinguistics of Globalization
Just check off all the ways your non-American accent prevents you from achieving your professional and personal goals.
n People think I'm less smart than I am.
□ I have been denied a job promotion,
□ I find it very hard to socialize with Americans in personal and business settings.
□ People sometimes laugh at the way I talk.
My clients {or patients or colleagues) often misunderstand me.
! I failed to get a job I was qualified for.
My boss (or teacher) doesn't give me the respect I deserve.
I make less money than less experienced people who have better pronunciation.
What do you believe is the single biggest problem with your accent?
OPTIONAL: To be added to our Newsletter/Email list you
may include the following:_
Your name: _
Your email address: Your native language:
Submit
You will be returned to this page after submitting your survey. Thank you I
Figure 2.4 'American Accent Now'
based on the East-Midwest (Michigan, Illinois) accent, presented here as the neutral, 'accent-less' American accent. That in itself deserves more discussion space than I can spend on it here. The point is: Internet providers sell a 'standard (ized) accent', a regional accent ('American') that carries all the ideological features of a standard(ized) language. With these two caveats in mind, let us now turn to the websites themselves.
Consider figure 2.4, taken from a website called 'American Accent Now' (www.American-accent-now.com).1
A messy new marketplace
51
We see several things here. First, of course, we see two figures: a manifestly happy young white man, and a manifestly frustrated Asian woman. We can assume that the man has acquired an American accent (or had one right from the start) and that the Asian woman has not. The reason why she is frustrated can be read from the boxes one can check to demonstrate the ways in which a 'non-American accent prevents you from achieving your professional and personal goals'. Respondents - prospective customers - are requested to identify the 'single biggest problem with your accent'. The two categories can be divided as follows:
Professional:
• I have been denied a job promotion
• My clients (or patients or colleagues) often misunderstand me
• I failed to get a job I was qualified for
• My boss (or teacher) doesn't give me the respect I deserve
• I make less money than less experienced people who have a better pronunciation
Personal:
• People think I'm less smart than I am
• I find it very hard to socialize with Americans in personal and business settings
• People sometimes laugh at the way I talk
We now begin to discover the strange world we have entered. It is a world in which indexical values and social effects of language - usually tacit ideological features of language use - are made very explicit both in text and in images. The key to getting the jobs one deserves and to earning the money one is entitled to is pronunciation. An image of personality is invoked that makes this reduction possible: the prospective customer is already smart, entitled to better jobs and a superior income, and entitled to respect from their superiors. So acquiring the accent will, so to speak, release all of these qualities to be seen for everyone. Acquiring an American accent will, in the eyes of the interlocutors, turn you into the person you really are.
Not everyone is a prospective customer of course. As we can see in the images used by American Accent Now, the frustrated subject was Asian and the happy one was what is known in the USA as a 'Caucasian'. Websites make general targeted statements about 'strong foreign accents':
A strong foreign accent can prevent you from achieving your professional or personal goals and reaching your full potential. People shouldn't have to ask you to repeat what you said. In today's competitive corporate environment, clear pronunciation and correct grammar are a must. (Accurate English, www.accurateenglish.com)
What is meant by 'strong foreign' accent is often documented in the FAQ or Testimonials pages of such websites, and almost invariably this is done by
52 The Socialinguistics of Globalization
reference to 'national' accents. Some European accents such as French, German and even British accents are seen as particularly prone to misunderstanding, Asian accents such as Indian, Chinese and Japanese accents naturally move in the danger zone of globalized comprehensibility, and Middle Eastern accents are problematic throughout. They are problematic in contacts with 'native speakers':
Our accent reduction and American pronunciation courses will teach you to create the sounds of Standard American English and give you greater confidence in your communication skills with native speakers of English. (Accurate English)
[Testimonial from an 'Iranian dental student'] Hi, my name is Sanaz. About a year ago when I came to the US, I was very confused. I was in culture shock. Thanks to God, it happened to me to meet very lovely people. They gave me courage to deal with all new things, as well as they helped me to improve my English, and gave me confidence to speak English. Among them two people were so special for me - Sheri and Mark [the Accent Workshop trainers]. Not only they were fabulous teachers, they were very helpful friends. You can always trust their opinion. I'll never forget the great influence that Sheri and Mark have on my English. I'm really happy to meet them. (The American Accent Workshop, www.accentworkshop.com)
Sanaz experienced 'confusion' and 'culture shock' upon his arrival in the USA, and the American accent course gave him 'courage' and 'confidence' to speak American English, and this capacity clearly helped him overcome his initial feelings. Note the frequent references to 'feel-good' factors such as 'confidence'.2 Speaking in an American accent makes people more confident in contacts with their American native-speaker counterparts:
[Testimonial from a Spanish male Laboratory manager] The Accurate English accent reduction courses should be a 'must' for all professionals whose foreign accent gets in the way of clear communication. Taking the courses has greatly improved my pronunciation of American English sounds and made a big difference in my speech. My presentations at work are much better. I speak with more confidence and authority. (Accurate English)
[Testimonial from an Israeli male computer consultant] Before I started taking the classes, I couldn't pronounce certain vowel sounds correctly. Lisa [the trainer] identified the issues and taught me to listen and to express myself in a native way. Her customized method allowed me to quantify my weekly improvements. After a relatively short time, I noticed major progress in the way I speak. This lead [sic] to an unexpected increase in my confidence when communicating. I highly recommend the American Pronunciation course. (Accurate English)
The benefits of accent modification can be yours!
• Clear, understandable speech
• Efficient, effective communication
• Career opportunities
• Improved job performance
• Successful public speaking
• More confidence
(Advance American Accent Training, www.advanceamericanaccent.com)
A messy new marketplace
53
The 'confidence' promised or reported in these examples is a mixture of personal and professional features. People are more confident in general, they feel better after having taken the courses, and they also perform better in their jobs. The Spanish Laboratory Manager reported significant improvements in his presentations at work, and Advance American Accent Training promises not just 'more confidence' but also 'career opportunities' as well as 'improved job performance'. The implicit image is that of a 'professional', someone whose job is a central part of their life, and for whom professional unhappiness is equal to personal unhappiness. Most websites, thus, would suggest mixtures of professional and personal results for their accent courses, and we see a sort of continuum here, with some websites suggesting more professional benefits than personal ones. Consider the next example, from the Accent Reduction Institute (www.lessaccent.com), a website that shouts 'Lose your accent in 28 days!':
The Accent Reduction Institute (ARI) is the industry leader in American Accent Training, providing non-native English speakers with proven techniques to quickly master English pronunciation. ARI provides the tools to help people maintain their unique cultural identity while:
Eliminating language barriers and miscommunications
Increasing sales and profitability
Communicating expertise to customers
Building strong teams
Increasing efficiencies
Raising self-confidence
'Self-confidence' has been pushed into a small corner here, and the main advertisement claims are about professional aims. Note, however, that customers can 'maintain their unique cultural identity'. I shall come back to that below. Communicaid (www.communicaid.com), in the meantime, is clear about its business-oriented efficiency:
Why American Accent Training? With the proliferation of offshore operations in locations as diverse as India, the Philippines, South Africa and China, organizations need to ensure that their offshore employees are able to communicate effectively with customers and colleagues alike. A vital ingredient of successful communication for your overseas staff is their accent when speaking English.
The impact of First Language Influence (FLI) on an offshore employee's accent can not be underestimated. American Accent training from Communicaid will help your overseas staff to minimise the influence of their first language while maximising their communication with your customers through a neutral accent. Whether through online American Accent Training courses, virtual American Accent training or face-to-face accent instruction, Communicaid offers scaleable American accent training solutions for your organization's offshore operations.
54 The Socialinguistics of Globalization
A Communicaid American Accent Training course will provide your offshore personnel with the ability to:
• Communicate more efficiently and effectively with customers and colleagues by reducing their first language influenced accent
• Build rapport and empathy and strengthen relationships with customers and colleagues through more successful communication
• Enhance your customer experience and satisfaction
The target groups are here, clearly, the ever-increasing workforce of international call centres, and most of the American accent websites make explicit or implicit reference to call centres as a target or a success story. Thus, the Testimonials page of the Pronunciation Workshop (www.pronunciationwork-shop.com) reports:
This is by far, the MOST EFFECTIVE English Pronunciation programme I have ever seen. I am an American Accent Trainer in India and have trained 22,000 agents. After taking this programme, I wish there was a way to go back and start all over again. (Oorvakx Boyce, American Accent Trainer, India)
I am the Head of the Training Department of the World's Second Largest Call Center Company in India. We have footprints in over 40 countries around the world including many centers in United States as well. I have been working in the Call Center Industry in India for the past 8 years. Because we cater to America, for years we have been searching for a cost effective training method to learn the American style of English Pronunciation, however, have had no luck...that is, until we found the Pronunciation Workshop programme. While searching through Google one day, I
came across the Pronunciation Workshop website and watched the free demo......We
found that demo to be SIMPLY AMAZING STUFF !!!!!!! Instead of showing diagrams,   visuals   combined   with   phonetics,   there   was   Paul..   A LIVE
TRAINER......       SHOWING HIS MOUTH FORMATION and the TECHNIQUES
to get a clearer sound. I quickly showed it to my CEO and he was amazed as well. What's unique about Paul's course is how simple it is...as well as fun to watch. My staff always looks forward to working on this programme...We seem to learn something new every time we watch another video, and it has made remarkable changes in our speech and English skills. We have implemented this training approximately six months ago to improve our customer satisfaction scores. Not only have we seen significant improvements with our scores, but our Average Call Handle Time per customer has reduced approximately by 29% in the last 4 months as the agents do not need to repeat themselves. Looking into the success of the programme at our centre we have also asked Paul to develop a certification programme. I myself, as well as members of my staff have gone through telephonic training with Paul, one-on-one. Paul makes learning so much FUN and has changed the way we speak English!!!. Today I am proud to say that the entire training department has become Certified by Pronunciation Workshop. I cannot praise this course enough (...) It is truly quite an achievement and I know you will be extremely pleased as well. (Joy Deb Mukherjee, Director of Training, India)
A messy new marketplace
55
The truly remarkable thing about Paul's [the trainer] programme is how simple and effective it is...and his engaging persona really pulls you in. Here in India, using this programme we are able to take an individual who is UNEMPLOYABLE for the offshore booming call centre industry, and make him EMPLOYABLE, with a good job which pays well, in less than two and half months... That's the Power of Paul's programme!!! He is changing lives on the opposite side of the world! (Sanjay Mehta, Managing Director, Teleperformance India)
And American Accent (www.americanaccent.com) sees a direct link between its online teaching methodology and its prospective customers, the call centre workers:
It Works
Quick, Easy and Fully Automated
Given the powerful combination of high aptitude and a proven methodology, designated trainees easily and quickly pick up on the accent.
People used to think that classroom training is the best method, but for the ideal call centre candidate, you want someone who is completely comfortable in the virtual environment - from initial training through to long-term job satisfaction.
Global indexicals of success
American accent, personal happiness and self-confidence, smooth and efficient communication with Americans, job satisfaction, business opportunities and money: this is the package sold by these dot-com businesses. The package consists of 'language' itself (or register-bits thereof) as well as representations of it and of what it has to offer its speakers, and, at a very implicit level, representations of 'America', of what America is as a society and of the socio-cultural preferences and expectations of 'Americans'.3 Following Silverstein (2006b: 485) we can call this 'semiotic consubstantiality': you are (or become) what you speak, and speaking it (mysteriously) transforms you into what is indexically suggested by the speech. The providers sell these consubstantial-ities to the people who are at the heart of globalization processes: expatriate (i.e. deterritorialized or peripatetic) professionals hungry for success, opportunity and money, and call centre operatives in 'delocalized' and inherently mobile areas of the business world. And they sell them by means of that defining technology of globalization: the Internet. The target audiences for these websites are not the masses of poor migrants from the margins of the world, not the Philippine domestic workers who have become one of their nation's main export industries, and not the housewives who join their expatriate husbands. The target audiences are the scale-jumpers who embody globalization as a success story - they are, in other words, a very small segment of the flows of people that characterize contemporary globalization. And one of the advantages of these materials is that they begin to show us a picture of what 'success' means
56 The Socialinguistics of Globalization
within this imagery of globalization. For what the websites do is to abundantly flag what Silverstein (2006b: 486) calls (with respect to dictionary cautions such as 'slang' or 'obscene') 'register alerts'. Such alerts 'give normative indexical properties of a lexeme's appropriateness-to and effectiveness-in co(n)texts ... of occurrence: where to use it, and what, socially speaking, will happen when you do'. If we replace 'lexeme' by 'accent' we have, I believe, an accurate description of what these websites try to achieve: to overtly specify the rich indexicals that come with the language in normative terms; to explicitly describe, in sum, an order of indexicality for the use of such an accent: you must use this accent if you want to become the person you intend to be. Naturally, there is a hierarch-ization involved in this: not every kind of speech is adequate and only this kind of speech will do.
We see this clearly from the way in which the websites define the exercise they expect from and offer to coach to their customers. First, it is not just about acquiring a new accent, but even more about getting rid of another accent ('Lose your accent in 28 days!'). Hence the often-used label of 'accent reduction' as a descriptor of the courses offered. The existing accents are wrong:
We focus on correcting your biggest speaking errors first. After the first lesson, you will already feel more confident about your speech. (Accurate English)
Nobody had ever been able to tell me what exactly was wrong with my speech until I took the Accurate English accent reduction course. There was a noticeable difference in my speech after only a few lessons. I see this course as an investment in my future. I learned how to strengthen my 'Chinese tongue' to create the sounds of good American English. (Testimonial page, Accurate English)
They are also, as we saw in several of the examples above, an obstacle to personal and business success, a source of frustration (recall the image of the Asian woman!), something that can hamper individuals and corporations in their ambition. So what is at stake is, one could say, linguistic purification as the first step to linguistic readiness for the globalized world. And at the core of this process of purification we see an image of the regimented subject, someone who can face the challenges of post-modern, globalized existence provided they submit to the process of purification and, consequently, sacrifice their individual agency in a quest for uniformity and homogeneity.
At this point, the locality and authenticity we encounter in popular, political and academic discourses on dialects and accents are far away. There is little value to the linguistic signs of belonging and authenticity that usually go under the label of 'accent'. Or at least: there is clearly a differential value to different accents. While foreign accents are remarkable, audible and problematic, and hence need to be reduced or eliminated, American accent itself is unmarked, unremarkable, unnoticed. Once you acquire American accent, your speech becomes 'normal', invisible, unremarkable, and can so become a vehicle
A messy new marketplace
57
for 'efficient', 'smooth', 'clear', 'confident' and 'convincing' communication (see Accurate English's website banner: 'Helping foreign professionals to communicate with confidence, clarity and accuracy'). American accent is not really an 'accent' like a German-English or Indian-English accent, for example -it is a neutral tool, a purely linguistic-communicative instrument. Thus, in the examples above, we saw how Communicaid projected trained employees as 'maximising their communication with your customers through a neutral accent' while they minimize 'the influence of their first language' (italics added). And in their FAQ section, American Accent reassures its customers as follows:4
Is this slang?
No, of course not. Standard American English pronunciation is different from spelling, but it is not slang.
American accent, thus, is the 'exnominated' accent, to use a term coined by Barthes (1957) to denote the bourgeoisie. The bourgeoisie, Barthes said, was so hegemonic that they ceased to be perceived and named as bourgeoisie; they were just 'the people'. American accent, then, is the accent that is no longer an accent but just a vehicle for doing stuff. It is just a 'normal', expected, customary and efficient language. (This is where we see how American accent is ideologically represented as a standard(ized) language.) If you use it well, it helps you achieve the goals you have set in life: prosperity, success, happiness. And while, as the Accent Reduction Institute claims, customers can 'maintain their unique cultural identity', if they wish to be the globalized actors they aspire to be, they need to change and become sociolinguistically invisible. Their 'unique cultural identity' should not transpire from their speech. As for American cultural identities, they can also be studied in courses provided by the websites:
The American Psyche
Typical culture pieces seek to inform a trainee of facts about America. This information, however, is readily available via the media and the Internet, and can be easily acquired once the student has a standard accent. The AAT culture piece, on the other hand, deals with the American psyche. Trainees learn how Americans think, what is important to them, and how best to get them to respond in the desired way. (American Accent)
Thus, one remains 'Indian' while one can sound 'American'. And to the extent that accents betray the life-histories of their users, and in particular the histories of institutionalized and mundane language usage and learning, sounding American means that all of this has to be backgrounded. An Indian accent in English is the result of language learning processes in India, and the accent displays all the sociolinguistic diacritics we know: age, gender, class, educational background and so on. An American accent course removes all of this
58 The Socialinguistics of Globalization
identificatory uniqueness and replaces it by an exnominated and translocal, presumed neutral and uniform accent.
The effort that trainees should invest in this process of removal and replacement is considerable. Apart from a significant financial divestment in registering for the courses, they are expected to buy books, CDs and DVDs, have telephone tutorials with their trainers, record their own speeches and listen to them, seek informal communication opportunities with 'native speakers', and even attentively watch TV:
When you are watching TV look at the mouth movements of the speakers. Repeat what they are saying, imitating the intonation and rhythm of their speech (Accurate English)
But as we saw in an earlier example, the effort pays: 'our Average Call Handle Time per customer has reduced approximately by 29% in the last 4 months as the agents do not need to repeat themselves'. We also saw in several examples given earlier that individuals referred to their experience with such courses as events that made them gain self-confidence, the respect of their peers or superiors, and upward social mobility. Here is an ideologically rich and well-matured sign: the idea that hard work to change oneself will result in material as well as symbolic rewards. The American dream is here projected onto the symbolic currency marketed by these businesses: American accent.
Summarizing, we see the following picture. Customers who subscribe to these courses buy a package consisting of language as well as of representations of language, society and selves. They are expected to change themselves by 'reducing' their non-native accents in English, and by adopting, with great investments of effort, an accent that makes them socio linguistically inconspicuous and changes their speech from something that contains side-tracking 'noise' (their 'foreign accent') to a normal, uniform, unpeculiar and thus no longer distracting tool of communication. This effort is rewarding, because it will remove the frustration resulting from (repeated) misunderstandings due to the 'noise' in their speech, it will allow Americans, thus, to see them for what they truly are, and this then will offer socially upward mobility. This, I would venture, is the cultural semantics of these courses, the ordered complex of indexicals suggested by American accent courses as a way to global success.
Discussion
This cultural semantics is a bit disconcerting, of course. We live in a world in which language learning environments, especially for English, become highly diversified and now include purely formal environments (state-controlled official curricula and school programmes) as well as purely informal ones (the global media and popular culture empires), with several hybrid forms in between such as the websites examined here. In this highly polycentric learning
A messy new marketplace
59
environment, some actors are subject to formal and sometimes rather rigid state policies - as the formal learning environments typically are - while others escape such forms of imposed normativity. The websites discussed here, I suggest, contain a very clear and rather transparent language policing which, while being informal (in the sense of 'not controlled by external authorities'), is stable, predictable and hegemonic. It taps into widespread and very powerful images of globalization as centred on the USA and revolving around English, corporate culture and individual spatial and social mobility. These images show traces of one of Appadurai's (1996) 'idioscapes' - globally distributed socio-cultural and ideological scenarios - notably of what Fairclough (2006) called 'globalism'. This is an effect undoubtedly of the particular composition of their projected audiences: prospective customers, as said earlier, do not belong to the 'losers' of globalization but rather to its 'winners': globally mobile professionals who have the wherewithal to make the best of the opportunities offered by globalization processes. But its remarkable uniformity betrays the self-evidence of a strong hegemony for such views among the community of users of these websites.
This, of course, should warn us against making quick generalizations about language in globalization. We are not facing the emergence of a wholesale new global order of discourse here, but rather a highly niched phenomenon that affects the lives of relatively restricted groups of people, and that is an effect of larger and slower globalization processes - the deregulation of international business activity and labour, the development of virtual spaces for communication, commodification and learning such as the Internet, and the power of more general ideologies and discourses such as globalism. Sociolinguistic globalization is not the engine behind globalization, but an epiphenomenon of larger processes that are of a far more fundamental nature and have a far greater historical depth. Most people in the world do not want to spend money to learn American accents. In effect, most people could not care less.
Having said that, phenomena such as the ones examined here still do teach us a thing or two about language in globalization. One thing we see here is that the object of globalized commodification is accent and not language. This is a commodified dialectology, not just language learning. Investigations such as the ones reported here provide us with a more fine-grained view of sociolinguistic globalization processes. They are definitely not uniform, nor are they exclusively uniformizing: they are layered processes developing at several, very different scale-levels. The global purchase of English - what de Swaan (2001) called the 'stampede towards English' - is a phenomenon at one scale-level, a very general one and consequently one that, in actual sociolinguistic practise, would appear as a relatively widespread but superficial phenomenon. The marketing of American accents, on the contrary, revolves around specific sociolinguistic registers targeted at specific (rather small) groups. These groups, we
60 The Socialinguistics of Globalization
can assume, already 'know English', but they require a more specific set of linguistic, pragmatic and metapragmatic skills - to sound like an American. The scale of such processes is far smaller than that of 'global Englishes', but the process appears a lot more developed and to have a more profound impact. Given the homogenizing 'semiotic consubstantiality' we witnessed in these dotcom Englishes, we could say that this (rather than the 'imperialist' spread of English tout court) is real 'McDonaldization': customers are expected not just to buy the language but also the whole indexical (that is: ideological) package it contains, and to do so from within a narrowly defined 'globalist' worldview. These are different processes, to be sure - perhaps complementary but of a different order. And I would suggest that a sound sociolinguistics of globalization should not just look at the world and its languages, but also to the world and its registers, genres, repertoires and styles, if it wants to have any empirical grounding. It is in small-scale, niched phenomena such as those considered here that we see real language: language that is invested by real-world interests and language that matters to real people. This, naturally, is no longer the 'linguistic' language, neither is it the language used in official language policies, but it is the sociolinguistic language and the language of the everyday politics of language, produced and articulated in a polycentric environment by a multitude of (often ephemeral) actors.
This leads us to a second point about globalization. Given the intense poly-centricity of learning environments and the fact that globalization processes develop at several different scale-levels, the issue of normativity becomes quite complex. What is the status of language norms, who produces them and who enforces them? These are questions that now require a detailed examination of actors, instruments, goals and resources. American accent websites cater for a market that is not serviced by the national formal learning systems. Indians who have acquired English at school and seek employment in global call centres need to be retrained in American accent, because their school English bears too many traces of what in our examples was called 'strong foreign accent'. The state offers 'English' (with an accent), the Internet companies offer another (a better) accent. Thus we see how actors collaborate in the production of language norms in an interdependent complex of actors: the school system produces 'English', using its own pedagogies and orders of indexicality; but in order to acquire the specific kind of English that offers jobs in the globalized economy, people need to turn to private providers, who impose yet another set of norms and rules of proper speech. These private providers are not tied to the national order of things, their activities are 'border busting' in nature, and interestingly their cultural semantics revolves around uniformity, homogeneity and submission - a very modernist response to globalized, post-modem pressures.
An important feature of globalization processes is the fact that they blend the local and the translocal in complex networks (Castells  1996). Local
A messy new marketplace
61
socio linguistic systems, consequently, are shot through by traces and fragments of translocal ones, without, however, becoming less local (as we shall see in chapter 3). And the language that is 'good' in the local sociolinguistic system may not be good enough in the translocal ones - which is why Indian call centre workers need to learn American English and should not use their local accents. We thus see various orders of indexicality operating in the same polycentric environment, often without manifest overlap or confusion but 'niched' and confined to particular sets of communicative tasks. Highly skilled individuals such as the globalized professionals targeted by the American accent marketeers acquire, and must acquire, the resources that allow them to operate within and across these different orders of indexicality. Predictably, language thus becomes something that requires continuously more investment from its users, and language learning (more specifically, the learning of specific registers tailored to particular communicative tasks) will increasingly become a balancing act between formal and informal learning processes. It becomes, in effect, a permanent 'care of the self in Foucault's (2005) sense, a permanent quest for order by attending to the small details of linguistic conduct. Highly specialized providers such as the ones described here, capable of offering these microscopic definitional registers of subjectification as operational (linguistic) and ideological (indexical) packages may thrive in this brave new world of language.
2.7 Conclusion
So how messy is this new globalized marketplace now? In this chapter, we have covered quite a bit of space, starting from some strange phenomena in which we saw how 'language' surrenders (or at least, dramatically changes) its customary functions due to patterns of mobility. Mobility does some things to language, to be sure. In an attempt towards structuring our understanding of such patterns of mobility, I then offered a small vocabulary, consisting of the concepts of scale, orders of indexicality and polycentricity. I suggested that we should understand communication phenomena as developing at different scales, on which different orders of indexicality operate, resulting in a polycentric context for such communication phenomena - that is, a context in which multiple normative complexes are simultaneously at work, but are of a different order. We saw such forms of polycentricity in the various examples given in this chapter, notably in the ways in which Internet language-learning providers sell American accent: they create an order of indexicality that operates alongside, but not necessarily in conflict with, state-organized normative complexes valid in formal education systems. The local and the translocal appear together, as different forces operating on the same object - language.
The result was complexity, but it was not chaos. It is too easy to see the late-modern globalized world as one in which order is lost and replaced by disorder,
62 The Socialinguistics of Globalization
fragmentation and chaos, as a place where no single instrument of power can fully dominate and impose its rules on the field. Instruments of power now do co-occur in complex patterns of power-sharing (as we saw in the case of the Internet American accent providers), and the simplicity of modernist understandings of the world can no longer be maintained. But there is still order - a more complex kind of order, but a real one. The fact that the state now competes with numerous other actors in the field of language normativity does not abolish the state as a relevant level of power; it couches the state in a wider field of power relations, in which its orders of indexicality now co-exist alongside numerous others in a layered, scaled, polycentric environment. We shall return to the power of the state in chapter 6. The analytical task now is to describe and interpret polycentricity, not simple dominance. We have to come to terms with notions such as micro-hegemonies: restricted, 'niched' hegemonies that co-exist with others in polycentric environments. Such a view can profitably replace older views of hegemony, articulated, for example, in linguistic rights discourses where a single hegemony (that of English) and a single actor (the state) are seen as defining the patterns of sociolinguistic life. There is not much purchase in these views in an age of globalization, I am afraid.