Journal of Sociolinguistics 7/4, 2003: 607623

Commentary: A sociolinguistics of globalization1

Jan Blommaert
Ghent University, Belgium

1. INTRODUCTION Johannes Fabian opens his latest book with an essay entitled `With so much critique and reection around, who needs theory?' (Fabian 2001). The title and the argument are inspiring and capture the spirit of our times: we live in an age e and careful analysis have that of globalization, so to speak in which la pense come under pressure from various sides, yet are more than ever necessary. There is, on the one hand, pressure from `the eld', where urgent calls for immediately applicable solutions to burning problems become louder and louder. And on the other hand, there is pressure from a growing anti litist e lites. intellectualism articulated by politicians, media and other anti-e My argument here, much like Fabian's, is built on the old adage that there is nothing as practical as good theory. Faced with deep transformations in society which demonstrate the failure of older paradigms, we need not to abandon ship but to reconstruct our paradigms, improve them and expand them (Wallerstein 2001). As announced by Nikolas Coupland in his introductory remarks, when sociolinguistics attempts to address globalization, it will need new theory. The rst phase of the process is, therefore, the laborious and often unrewarding phase of trial-and-error: see what works, dene topics, units and elds, and try some analysis. Two issues are on the table. First I want to comment on the dierent papers in this issue and try to distil some general points, useful for what I believe should be our ambition here: to start developing a sociolinguistics of globalization. Second, I will add two suggestions for incorporating particular theoretical instruments into such an exercise: the notion of the world system and that of second linguistic relativity. 2. NECESSARY BUILDING BLOCKS In general, all the papers in this issue address matters of scale: the macro and the micro, the global and the local, the dierent levels at which `language' can be said to exist and at which sociolinguistic processes operate. Various papers treat, for instance, the relationship between a `world language' English and local
# Published by Blackwell Publishing Ltd. 2003 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden MA 02148, USA.

608

BLOMMAERT

speech repertoires or speech communities (House; Heller; Pennycook; Meyerho and Niedzielski), emphasizing the dierent ways in which English permeates speech habits, propagates and organizes genres, or reorganizes functional hierarchies for languages. The scalar processes and phenomena are becoming less predictable and more chaotic and complicated, it seems, in the era of globalization. The paper by Thurlow and Jaworski shows how airlines produce self-imaginings of globality while remaining `based' in one country or place; and Machin and van Leeuwen discuss the ways in which particular discourses of femininity and sexuality get spread globally in the context of media industries. In each of the cases, we see a questioning of the ways in which particular situated events, genres, or styles connect to worldwide patterns and processes. And in both cases, we see how hard it becomes to separate and order `small' versus `large' processes. We are facing `small' genres that are deeply connected to very specic audiences and forms of commodication (both genres address an auent middle class). But they both explicitly articulate a global reach and use similar modes of expression wherever they occur: they explicitly voice and identify globalization as an identity issue for their readership. And yet, they lites and both appeal to and are simultaneously oriented towards national e emphatically ag `national' (or `national-cultural') characteristics. So dierent scales seem to overlap and combine in one genre, and it is hard to determine which scale would hierarchically dominate the others. Scale is denitely the keyword in any analysis of globalization. The term globalization itself suggests a process of lifting events from one level to a higher one, a global one, or vice versa, and a sociolinguistics of globalization will denitely need to explain the various forms of interconnectedness between levels and scales of sociolinguistic phenomena. The complexity and simultaneity we are facing is a challenge, not a danger. But we need to be more precise, and I think two qualications are in order. First, we need to move from Languages to language varieties and repertoires (Hymes 1996: 67; Silverstein 1998). What is globalized is not an abstract Language, but specic speech forms, genres, styles, and forms of literacy practice. And the way in which such globalized varieties enter into local environments is by a reordering of the locally available repertoires and the relative hierarchical relations between ingredients in the hierarchy. Sociolinguistic globalization results in a reorganization of the sociolinguistic stratigraphy, a process which does not necessarily lead to a new solid and lasting hierarchy but may best be seen as an ongoing, highly volatile process cross-cut, again, by matters of scale. The point is forcefully made by Machin and van Leeuwen, who emphasize the domain-related spread of global registers rather than languages. What Cosmopolitan does sociolinguistically is to spread a particular discourse on femininity, success, beauty and sexuality. This particular discourse is globalized and it displays some degree of uniformity in each of the 40-plus local areas of circulation investigated by Machin and van Leeuwen. But it is not a wholesale
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

609

import of a complete, new, set of ideologies of femininity it is niched. Pennycook's Rip Slyme rappers insert globalized slang into an idiosyncratic popular-cultural brew: a particular niche. Heller shows how `authenticity' and locality become commodities in a new, niched, globalized market. Conversely, the argument developed by House, and Meyerho and Niedzielski focuses on the reorganizing eect of `international' speech varieties (House uses `lingua franca' as shorthand for a complex set of particular speech types and genres performed in English) on local repertoires. Again we see that (American) English does not eliminate what was around, it does not bury the local languages, but it enters the repertoire of language users as a resource that fulls both pragmatic functions lingua franca functions, a certain degree of vernacularization and metapragmatic ones. It marks the variety o against others and creates new balances of `value' for the dierent varieties in the repertoire.2 So again we see that the impact is niched and restricted to particular groups (networks, communities of practice, etc.) in societies. Therefore, nding the particular niche on which the globalized ows eventually have an impact may be a crucial part of our sociolinguistic assignment: it would oer us a rst clue about what globalized sociolinguistic phenomena mean to identiable groups of people, and about what these people can actually do with it. It will also oer us deeper insights into the specic role of particular mediating institutions in the new economies that appear to characterize globalized ows. Almost all the papers in this issue mention such mediators: rap artists and the music industry (Pennycook), international English training programs (House), airlines (Thurlow and Jaworski), tourism and service industries (Heller) and the printed press (Machin and van Leeuwen). A realistic look at globalization processes involves questions about whose genres are being globalized, by whom, for whom, when and how? Not everybody appears to be part of the active globalization processes, but particular mediating actors are, and Heller's emphasis on the new economies as driving forces behind the increased commodication of language varieties and discourses may oer us an interesting heuristics of research. A second but closely related qualication is that we need to address the language-ideological level in these processes. The key to understanding the processes of `globalized' insertion of varieties into newly stratied orders of indexicality, is to discover what such reorderings of repertoires actually mean, and represent, to people. There is ample evidence for the assumption that language ideologies aect language change, including forms of transformation now captured under the label `globalized' (Schieelin, Woolard and Kroskrity 1998; Blommaert 1999a; Kroskrity 2000; Gal and Woolard 2001). And there is a lot of evidence for that in the papers as well. The functional division suggested by House between `languages for communication' and `languages for identication' is a metapragmatic dichotomization that allocates specic indexicalities to particular speech varieties. The case is similar in Meyerho and Niedzielski's paper: vernacularization is a metapragmatic complex conveying all
# Blackwell Publishing Ltd. 2003

610

BLOMMAERT

kinds of indexical meanings (including the disappearance of the `Americanness' of particular sociolinguistic items). In Heller's Canadian cases, language shifts from a marker of ethnolinguistic identity to an economically interesting `skill' or a commodity. Rip Slyme's hip-hop slang is pure indexicality I'll come back to it below. The ideological, metapragmatic aspects of language usage lead us to an understanding of meaning and function of `new' ingredients in repertoires: they allow us to understand which functions people assign to such items, and why. In practical terms, they may for instance oer us an understanding of why nonnative English of the kind discussed by House often meets considerable interpersonal tolerance for deviations from `standard' English in pronunciation, syntax, lexis and style.3 Let us take both elements together: we have to deal with niched sociolinguistic phenomena related to the insertion of particular varieties of language in existing repertoires, and also with the language-ideological load both guiding the process and being one of its results. Now, one of the things we see in almost all the papers is how people create semiotic opportunity in globalization processes which, as noted above, may productively be framed in the larger picture of the new economies. This is most outspoken in Pennycook's Rip Slyme case, but Thurlow and Jaworski's airline network maps are also a case in point, and so are House's globalized non-native English-using executives and Heller's heritage tourism workers. They all manage to assign specic new functions to sociolinguistic items (either `global' items or `local' ones), and accomplish specic, targeted (globalized) goals with them; often these forms of identity work could not be done without the potential oered by globalization. Let us take a closer look at the Rip Slyme case discussed by Alastair Pennycook. We have here a hip-hop ensemble who use phonetically `nativized' American English hip-hop slang in a peculiar blend a `fusion' would perhaps be a better term with Japanese, thus constructing a genred semiotic product: rap lyrics. All kinds of functions are accomplished simultaneously here what we see is a bundle of divergent indexical meanings packaged in one speech event (cf. Woolard 1998; Maryns and Blommaert 2001; Blommaert 2003b and in press b discuss similar examples from Tanzania and South Africa). Furthermore, the packaging of indexicalities is rmly anchored in, and enabled by, globalization. The Japanese rap stars manage (at least) to: (a) produce a set of propositionally organized `meanings'; (b) realize a particular genre which is both local and transnational: Japanese versions of rap; (c) suggest locality through the distinct `Japaneseness' of the lyrics; (d) connect with transnational generic, cultural and social networks identied as hip-hop culture with AfroAmerican rap music as its `center'; (e) appeal to, and thus construct a particular subcultural community of hip-hop fans in Japan (in itself a typical globalized sociocultural phenomenon); and (f ) display their own knowledgeability and virtuosity in the stylistically elaborated genre they attempt to realize. In sum, the particular `globalized' linguistic and cultural-stylistic blend allows the semiotization of unique indexicalities that point towards the local-global
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

611

dynamics characterizing and contextualizing the cultural practice of hip-hop. It thus allows the construction of a particular commodied variety of language, a nicely globalized and globalizable language product. Importantly, at this level of language usage, scale proves to be an issue again. We see no fundamental hierarchical dierence between local and global indexicalities: they occur simultaneously. Metaphors such as the `invasion' of English into Japanese, let alone linguistic and cultural `imperialism' or worldwide lingua-cultural homogenization McDonaldization are obviously inadequate for a description of this fantastic semiotic creativity, which allows language users opportunities to represent cultural, social and historical conditions of being.4 It is an important accomplishment if we manage to see sociolinguistic globalization in these terms: as a matter of particular language varieties entering the repertoires of particular groups, creating new semiotic opportunities and commodities for members of such groups and indeed constructing them as groups. We can now move on and focus on mobility as a key feature of sign complexes in globalization: the fact that language varieties, texts, images travel across time and space, and that this is a journey across repertoires and sets of indexicalities attached to ingredients of repertoires. A sociolinguistics of globalization is necessarily a sociolinguistics of mobility. In what follows, I will oer two theoretical suggestions for capturing the nature and structure of such sociolinguistic processes of mobility. The rst one has to do with what we understand by `global' in `globalization', the second one has to do with how we can conceive of the processes of insertion of `globalized' material into repertoires. 3. THE WORLD SYSTEM AS CONTEXT It is a regrettable feature of much discourse on globalization that it seems to present globalization as the creation of worldwide uniformity. Processes are often represented generically, as a universal shift in the nature of societies, semiosis or identities. Terms such as `global ow' suggest a ow across the whole of the globe, a generalized spread of sociocultural and economic patterns, a new universalism. In our eld of study, some recent work on discourse in the late-modern world falls prey to this. Chouliaraki and Fairclough (1999), for instance, suggest that the conditions for the production and circulation of discourse (in general, without qualication) have changed in the present world. Both the nature of signs and discourses as well as their distribution, access and eects have undergone substantial transformations, and `discourse' is now a dierent concept. I would suggest that this is a description of discourse genres in late-modern Britain, but not a general theory of discourse in the world. What is seen as `late-modern discourse' is a new genre, a new variety of language, spread across the globe in the ways mentioned above: as a specic genre, and across specic groups of people in specic contexts. The intricate and layered semiotics we encounter in Cosmopolitan is a specic register, as Machin and van Leeuwen rightly emphasize; Pennycook's Rip Slyme lyrics are also a
# Blackwell Publishing Ltd. 2003

612

BLOMMAERT

generically regimented, highly specic, subcultural format, and so are the airline network maps discussed by Thurlow and Jaworski. Late-modern discourse of the type discussed by Chouliaraki and Fairclough thus needs to be understood in relation to the rest of the semiotic landscape in which it enters and circulates. I will come back to this below. We have to realize that the world is not a uniform space and that consequently, globalization processes need to be understood against the background of the world system. This world system, as Immanuel Wallerstein has extensively argued, is a system built on inequality, on particular, asymmetric divisions of labor between `core regions' and `peripheries', with `semiperipheries' in between (e.g. Wallerstein 1983, 2001). Thus, the system is marked by both the existence of separate spaces (e.g. states) and deep interconnectedness of the dierent spaces, often, precisely, through the exist lites.5 Inequality, not uniformity, organizes the ows and the ence of worldwide e particular nature of such ows across the `globe'. Consequently, whenever sociolinguistic items travel across the globe, they travel across structurally dierent spaces, and will consequently be picked up dierently in dierent places. The interconnectedness of the various parts of the system creates the previously mentioned issues of scale and levels of analysis: what occurs in a particular sovereign state can and must be explained by reference to state-level dynamics, but needs to be set simultaneously against the background of substate and superstate dynamics, and the hierarchical relations between the various levels are a matter of empirical exploration, not positing. Globalization implies that developments at the `top' or the core of the world system have a wide variety of eects at the `bottom' or the periphery of that system. For instance, developments in the eld of sophisticated, multimedial and multimodal internet communication have eects on other, less sophisticated forms of literacy. I will illustrate this below. Important in all of this is that the dierent levels seem to operate at dierent speeds. Fernand Braudel's famous distinctions between slow time, intermediate time, and fast time may be a useful metaphor here (Braudel 1969; see Blommaert 1999d: 36 for comments). Braudel rst observes that historical developments are of dierent orders a climate change (slow time) is something dierent from an economic conjuncture (intermediate time) or a battle (fast time) but he adds that people usually only observe the fast and (parts of ) the intermediate developments, and that historical processes are therefore not necessarily accessible in similar ways to the individual's awareness and agency. One may have degrees of consciousness about and agency over `fast' and `intermediate' processes, while the slow, macro processes are an invisible (yet very real) context, not open to individual agency. A sociolinguistics of globalization will need a holistic and world-systemic view in which local events are read locally as well as translocally, and in which the world system with its structural inequalities is a necessary (but not self-explanatory) context in which language occurs and operates. In my view, this is the main challenge of globalization to intellectual endeavors such as sociolinguistics,
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

613

which claims to contribute to an understanding of society through an understanding of language.6 It is precisely the fragmented but interconnected nature of the world system that accounts for the niched character of sociolinguistic globalization: it occurs not everywhere, but in particular dierent yet interconnected places and not in others, and this is a structural and systemic matter with deep historical roots, not a coincidental one. It is historical, and that means that we have to situate globalization processes in a wider picture of structural `becoming', of processes of worldwide inequality that derive their systemic nature from the long history in which they t. We are never investigating synchronicity, but always a particular stage in a historical process. In World Systems analysis, emphasis is placed on issues of scale, layering, and dierential development within a system dominated by interactions and mutual (though dierential) inuences. This format looks to me to be a highly applicable framework for sociolinguistic investigation. For a sociolinguistics of globalization within a World Systems perspective, I would suggest that emphasis is needed on the relative value of semiotic resources value often being connected to translocally realizable functions, the capacity to perform adequately in and through language in a wide variety of social and geographical spaces and across linguistic economies (something often attributed to English, but also to literacy and internet communication). This capacity is the capacity for mobility, and this emphasis on value as a crucial aspect of function is due to the fact that globalization raises new issues of inequality, both locally and translocally, precisely with respect to the capacity for mobility of resources. Specically, the `weight' of social and cultural forms of capital across spaces (geographical as well as social) appears to vary enormously. What works in one place does not appear to work elsewhere, and the kinds of `ows' usually associated with globalization processes involve important shifts in value and a reallocation of functions (Appadurai 1990; also Bourdieu 1990). When people move across physical as well as social space (and both are usually intertwined), their language practices undergo reevaluation at every step of the trajectory and the functions of their repertoire are redened. And conversely, movements of others, as in Heller's heritage tourism, aect the value and function of local speech repertoires. (One can note, reecting on Heller's case study, that globalization can increase the value of otherwise minorized varieties; the direction of value changes again appears to be unpredictable.) This, to me, looks like a prime target for a sociolinguistics of globalization, and the most adequate way to address it is by looking at the relative (and shiftable) value of linguistic practices as a component of their function. This is the topic of the next section. 4. SECOND LINGUISTIC RELATIVITY AND GLOBALIZATION Let me now introduce a second concept, useful in my opinion for analyses of sociolinguistic globalization phenomena: Dell Hymes' notion of `second linguistic
# Blackwell Publishing Ltd. 2003

614

BLOMMAERT

relativity'. Dell Hymes' voluminous oeuvre is well known and recognized as foundational to sociolinguistics and linguistic anthropology. Yet, some of the theoretical wealth of his work remains largely untapped. A central concern in his work is with function, and it is in this respect that he developed his notion of second relativity. The argument (originally presented in Hymes 1966 but recapitulated in Hymes 1980 and 1996) is of course cast in a performanceview of language, emphasizing variability and thus paying tribute to Whorf 's `rst relativity'. While indispensable to Hymes' view of language, Whorf 's relativity of structure assumed stability in function (`the inference of dierential eect on world view assumed equivalent role in shaping world view', 1966: 116). This is problematic, for:
. . . the role of language may dier from community to community; . . . in general the functions of language in society are a problem for investigation, not postulation . . . If this is so, then the cognitive signicance of a language depends not only on structure, but also on patterns of use. (Hymes 1966: 116)

And consequently:
the type [of relativity] associated with Sapir and Whorf in any case is underlain by a more fundamental kind. The consequences of the relativity of the structure of language depend on the relativity of the function of language. Take, for example, the common case of multilingualism. Inferences as to the shaping eect of some one language on thought and the world must be qualied immediately in terms of the place of the speaker's languages in his biography and mode of life. Moreover, communities dier in the role they assign to language itself in socialization, acquisition of cultural knowledge and performance. . . . This second type of linguistic relativity, concerned with the functions of languages, has more than a critical, cautionary import. As a sociolinguistic approach, it calls attention to the organization of linguistic features in social interaction. Work has begun to show that description of fashions of speaking can reveal basic cultural values and orientations. The worlds so revealed are not the ontological and epistemological worlds of physical relationships, of concern to Whorf, but worlds of social relationships. What are disclosed are not orientations toward space, time, vibratory phenomena, and the like, but orientations towards persons, roles, statuses, rights and duties, deference and demeanor . . . (Hymes 1996: 4445; also Hymes 1980: 38)7

One will be reminded of the comments on indexicalities and language ideologies made above. We will return to this further on. At this point, Hymes' emphasis on the problematic nature of language functions needs to be underscored: according to Hymes, such functions have been taken for granted by linguists while in fact they should be one of the foci of empirical investigation. Even if language forms are similar or identical, the way in which they get inserted in social actions may dier signicantly, and consequently there may be huge dierences in what these (similar or identical) forms do in real societies. Hymes thus shifts the focus of attention away from `linguistic systems' to `sociolinguistic systems': systems that revolve around the concept (classic in
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

615

pre-War American linguistic anthropology) of `fashions/ways of speaking'. What we need to investigate is the way in which language actually works in societies, and function is the key to this. This means, paradoxically at rst sight, that we shall need more ethnography alongside all kinds of other approaches as part of a sociolinguistics of globalization. Ethnography will allow us to unravel the details of how language varieties and discourses work for people, what they accomplish (or fail to) in practice, and how this ts into local economies of resources. It also allows us to check, at the lowest level, how larger patterns and developments are set down in the actual realities of language usage. We obviously need studies of the dierent levels and scales studies of linguistic variation, of history and policy but it would be a fallacy to regard ethnography merely as `the study of small things'. It is an indispensable ingredient of a toolkit for the study of big things. The impact of this relativity of function on sociolinguistic investigation is huge, as is its critical dimension. Part of linguistic inequality in any society and consequently, part of much social inequality depends on the incapacity of speakers to accurately perform certain discourse functions on the basis of available resources. Language functions and the ways in which they are performed by people are constantly assessed and evaluated: function and value are impossible to separate. Consequently, dierences in the use of language are quickly, and quite systematically, translated into inequalities between speakers (the key argument of Hymes 1996; see also Blommaert 2001a and Maryns and Blommaert 2002 for illustrations). This observation holds for what language does in stratied societies and it is central to, for example, Bernstein's and Bourdieu's arguments on language; it accounts for almost any dynamics of prestige and stigma in language, and sociolinguistics has built a remarkable track record of descriptions of such processes in single and synchronically viewed societies or speech communities. But there is more, as soon as we start looking at globalization. Globalization results in intensied forms of ow movements of objects, people and images causing forms of contact and dierence perhaps not new in substance but new in scale and perception. Consequently, key sociolinguistic concepts such as speech community (always carrying problematic suggestions of closure, synchronicity or achronicity, and homogeneity) become more and more dicult to handle empirically (for an early critique see Hymes 1968; Silverstein 1998; see Rampton 1998 for an excellent survey and discussion). Even more disconcerting is the fact that the presupposability of functions for linguistic resources becomes ever more problematic, because the linguistic resources travel across time, space and dierent regimes of indexicalities and organizations of repertoires. The functions that particular ways of speaking will perform, and the functions of the particular linguistic resources by means of which they are accomplished, become less and less a matter of surface inspection, and some of the biggest errors (and injustices) may be committed by simply projecting locally valid functions onto the ways of speaking of people
# Blackwell Publishing Ltd. 2003

616

BLOMMAERT

who are involved in transnational ows. In our work on asylum seekers' narratives, for example, we found that a particular kind of anecdotal subnarrative performed by asylum seekers (`home narratives') were easily dismissed by Belgian ocials as anecdotes that did not matter, whereas for asylum seekers such anecdotes contained crucial contextualizing information without which their story could be easily misunderstood (Blommaert 2001a; Maryns and Blommaert 2001). Whenever discourses travel across the globe, what is carried with them is their shape, but their value, meaning or function do not often travel along. They are a matter of uptake, they have to be granted by others, on the basis of the dominant indexical frames and hierarchies. The fact is that functions performed by particular ways of speaking and particular resources in one place can be altered in another place, and that in such instances the `value' of these linguistic instruments is changed. The English acquired by urban Africans may oer them considerable prestige and access to middle-class identities in African towns. It may be an `expensive' resource to them. But the same variety of English, when spoken in London by the same Africans, may be a crucial object of stigmatization and may qualify them as members of the lower strata of society. What is `expensive' in Lusaka or Nairobi may be very `cheap' in London or New York. What people can actually accomplish with these resources is likewise aected. `Good' and status-carrying English in the periphery may be `bad' and stigma-carrying English in the core of the world system. The opposite can, of course, also occur. Rampton's work on the delicate and complex reshuing of linguistic and stylistic repertoires in contemporary multi-ethnic peer groups has brought us a long way in understanding the relativity (and the renegotiability) of associated `values' to linguistic modes of conduct caused by diaspora or globalization ows in general (Rampton 1995, 2001; Harris, Leung and Rampton 2001). Social identities and the symbolic forms through which they are agged become more and more deterritorialized detached from conventional places and trans-idiomatic detached from `ownership rights' over particular symbolic forms (Jacquemet 2000; Harris, Leung and Rampton 2001; Maryns and Blommaert 2001). A careful re-reading of some outstanding sociolinguistic work along the lines of shiftable and relative value may yield interesting insights in the same direction. What this means for sociolinguistics, I believe, is that we need to revisit our ways of addressing formfunction relations, probably foregrounding them if we want to come to terms with globalization as a sociolinguistic phenomenon. And in doing so, it may be wise to keep in mind that globalization also results in global hierarchies in communication aecting existing local hierarchies and value-scales. As said above, developments at the top have eects at the bottom and vice versa. Consequently, a lot of what happens to linguistic resources in terms of value attribution is beyond the reach of individuals for it happens at macro-levels. It is determined in the Marxian sense of the term (see Williams 1977). In identifying formfunction relationships, a concept such as determination probably deserves more attention than has hitherto been given to it for it
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

617

may be indispensable for an accurate understanding of voice against the background of the world system. The `limits of awareness' of language users (the term is borrowed from Silverstein 2001) may be precisely the take-o point for a sociolinguistics of globalization, for it may be the point where societies come down on individuals' potential to decide and to act to produce voice so as to be heard and read. The more we look at this point, the more dierences and inequalities will appear, and explanations of these will force sociolinguistics to come to terms with, or even contribute to, the construction of theories of society in the world system. 5. WRITING IN/FROM THE MARGIN (OF THE WORLD SYSTEM) Instead of a conclusion I will oer a brief analytic vignette illustrating some of the theoretical points made earlier. I will discuss a small sample of writing produced in sub-Saharan Africa the periphery or margin of the world system but lifted out of its context-of-production and moved into a transnational speech network involving the African author and an addressee from the core of the world system (the `West'). The text is a handwritten letter addressed to me by a 16-year-old girl from Dar es Salaam, Tanzania. The girl, Victoria (a pseudonym), is the daughter of a family I stayed with during eld trips to Tanzania, and I rst met her when she was two years old. Her father is an academic, and Victoria was in secondary school when she wrote the letter. Secondary education in Tanzania is delivered through the medium of English, while the majority of the pupils (and teachers) have either Kiswahili or other African languages as their mother tongue(s). In primary school, Kiswahili is the medium of instruction and pupils are taught English as a subject. Consequently, at the age of 16, Victoria would have had several years of `deep' exposure to English. The girl is denitely a member of the local middle class, a class which uses prociency in English as an emblem of class belonging (Blommaert 1999c). It is, in other words, an `expensive' resource in Dar es Salaam. Let us now take a look at what, and how, she writes. What follows is a transliteration of the handwritten version, in which line breaks and graphic organization are rendered as precisely as possible (all names, except mine, are pseudonyms).

# Blackwell Publishing Ltd. 2003

618 20/9/1999 Dear ! Uncle Jan How are you? I hope you The main aim of this letter is to tell you that, here in Tanzania, we have remember you so much. Dady, Mum, Uzuri Patrick, Furaha, and Veronica and other members like Kazili, Helena, Bahati, Fatima and and others. Other people forget to write for you a letter, geat all your family I don't have much to say. Sorry if you will came Tanzania we will go to beach BYe BYe From VICTORIA MTANGULA

BLOMMAERT

A few comments are in order. Using a punitive reading, the rst thing that strikes the linguist-observer is the frequency of rather severe errors at the level of grammar (`we have remember you so much', `to write for you', `if you will came Tanzania'), as well as at the level of punctuation (absence of periods), orthography (`geat all your family', the alteration of upper and lower case symbols in the concluding line), narrative style and control over literary conventions (the awkward list of names dominating the letter, the separation of `Dear!' and `Uncle Jan', the unnished sentence `I hope you'). Victoria struggles with English literacy, her control over the medium is incomplete. At the same time, her act of writing can best be seen as `language display' in the sense of Eastman and Stein (1993): the mobilization of the best possible resources for a particular act of communication. Given the particular relationship I had with Victoria (and given the references to the other family members not writing to me), the act of writing is loaded with indexicalities, constructing a relational identity of a `good girl', someone who behaves and performs well, is probably among the best pupils in her age-group, and is worthy of compliments from her European Uncle. Her letter also indexicalizes all kinds of things with regard to writing practices and the use of particular codes (English) within a local repertoire. In short, Victoria tries to exploit the semiotic opportunities oered by globalized sociolinguistic phenomena. But she does so under world-systemic constraints. Victoria mobilizes the maximum-status resources within her reach: the best possible (school) English, the language of status and upward social mobility in Tanzania. And it is in that respect that the errors become important: as soon as the document moves
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

619

across the world system and gets transplanted from a repertoire in the periphery to a repertoire in the core of the world system, the resources used by Victoria lite status and prestige. The value of this variety of written would fail to index e English in Europe is deeply dierent from the value it has in Dar es Salaam. The indexicalities of success and prestige, consequently, only work within a local economy of signs, that of Tanzania, an economy in which even a little bit of English could pass as good, prestige-bearing English. We are witnessing, in the process of intercultural/international transfer, a shift in indexical and referential aspects of signs from one `placed' system to another. It is at this point that the critical rereading of Hymes' second relativity may be added to recent insights on indexicality and linguistic ideologies. The reallocation of functions for resources proceeds along indexical and referential lines: we allocate functions to resources on the basis of what we believe to understand by interpreting and contextualizing indexical and referential meanings of signs. We also see huge discrepancies between what linguistic resources and ways of using them mean in local environments that of grassroots literacy in Africa and what they mean in other, transnational environments in which they get inserted. The kind of literacy shown here is, I believe, widespread in Africa, and it characterizes lite strata of many much of what exists in the way of literacy in the sub-e African societies (Blommaert 1999b, 2001b, 2003a, in press a). In these societies the periphery of the world system it may be quite sucient to communicate adequately; in fact, it may even be an object of status and prestige. But lifted out of the periphery and placed into the order of indexicalities of the core of the world system, these forms of literacy lose their functions and receive new ones. From a rather high rank in one's own hierarchies of signs and communication practices, they tumble down to the lowest ranks of the hierarchies of someone else. Consequently, we are facing `placed resources' here: resources that are functional in one particular place but can become dysfunctional as soon as they are moved into other places. The process of mobility creates dierence in value, for the resources are being reallocated dierent functions. The indexical links between signs and modes of communication, and social value scales allowing, for example, identity construction, status attribution and so forth these indexical links are severed and new ones are projected onto the signs and practices. Particular linguistic resources, often those of people in the peripheries of the world system, do not travel well.8 I would claim that such reallocation processes are central to the kinds of mobility that characterize globalization: they dene how mobile resources are or can become and how much opportunity particular resources will oer their users in various places across the world. Consequently, a sociolinguistics of globalization should look carefully into such processes of reallocation, the remapping of forms over function, for it may be central to the various forms of inequality that also characterize globalization processes. For this we need
# Blackwell Publishing Ltd. 2003

620

BLOMMAERT

careful ethnographic work, sustained by a social theory which takes the world system as the highest level of contextualization.

NOTES
1. The main part of this paper is a written version of a lecture given at Cardi University, April 2002, during a workshop of the Leverhulme Trust project on Language and Global Communication. I thank the participants of the workshop, as well as those of the BAAL-panel in Cardi (September 2002) on which this issue is based, for rewarding discussions on the ideas outlined here. I substantially rewrote the rst version during my stay at the Department of Anthropology of the University of Chicago, JanuaryMarch 2003, probably the most generous and stimulating research environment I have ever seen. Finally, Nikolas Coupland provided excellent suggestions and some important caveats that have greatly helped me nalize this text. House perceptively notes that this observation `invalidates the claim that English is an imperialist ``killer language'' which English, we may ask'. The Linguistic Rights issue, as well as issues of language death or attrition, become something completely dierent indeed as soon as `Languages' are replaced by language varieties (see e.g. Silverstein 1998). I would argue, however, that the story is considerably more complex than the distinction between `language for communication' and `language for identication' oered by House. The kind of tolerance for errors in non-native English usage is driven by an ideological perception of this language usage as `instrumental'. Precisely this kind of usage of language denes the context and identies the participants as people who subscribe to this functionalist-referential ideology of language-in-abusiness-setting. See Milroy and Milroy (1985) and Gal and Woolard (2001) for general discussions. Mufwene (2002) provides an insightful discussion of this based on language loss and creolization. He observes, tongue-in-cheek, that `McDonald's outlets around the world operate in the local lingua francas, if not their vernaculars' (2002: 33). McDonaldization thus understood acquires a very dierent, more accurate meaning: the process is exactly that described by Machin and van Leeuwen with regard to Cosmopolitan. lites (the readership of Thus, the existence of such worldwide `globalized' e Cosmopolitan, or customers of McDonald's for instance) does not alter the center periphery structure of the world system, but often reinforces inequalities both locally and translocally. I am surprised by the often myopic nature of social-theoretical reections in our eld. Scholars enthusiastically refer to social theorists such as Habermas and Giddens theorists of the structure and development of First World societies but hardly ever to theory that addresses the world system, (under)development and dependency issues. Wallerstein has already been mentioned, but one could also Gunder Frank, Giovanni Arrighi and think of, for example, Samir Amin, Andre others, scholars whose work consistently emphasizes the interconnectedness of processes across dierent parts of the world, the eects of developments in one part on other parts, and the structural dierences in value of resources from dierent
# Blackwell Publishing Ltd. 2003

2.

3.

4.

5.

6.

A SOCIOLINGUISTICS OF GLOBALIZATION

621

7.

8.

parts of the world system. It is social theory that addresses the world, not just one part of it. The dierence between `real' world and `social' world as directions of orientation in indexicality is somewhat overstated by Hymes in this quote. Work such as that of John Haviland and Charles Goodwin has demonstrated how both are inextricably linked. Hanks (1990) and Gumperz and Levinson (1996) provide excellent discussions of this issue. In another paper, we called this phenomenon `pretextual gaps': gaps that originate when the resources people have fail to match the criteria of expected resources (Maryns and Blommaert 2002). The example given here is merely meant to illustrate such dierences in pretextualities, and I cannot address intricate issues of actual contextual displacement here, the point here being one about potential value rather than actual value. For fuller discussions, see Blommaert (2003a, 2003b).

REFERENCES
Appadurai, Arjun. 1990. Disjuncture and dierence in the global cultural economy. Theory, Culture and Society 7: 295310. Blommaert, Jan (ed.). 1999a. Language Ideological Debates. Berlin: Mouton de Gruyter. Blommaert, Jan. 1999b. Reconstructing the sociolinguistic image of Africa: Grassroots writing in Shaba (Congo). Text 9: 175200. Blommaert, Jan. 1999c. State Ideology and Language in Tanzania. Ko ln, Germany: Ru diger Ko ppe Verlag. Blommaert, Jan. 1999d. The debate is open. In Jan Blommaert (ed.) Language Ideological Debates. Berlin: Mouton de Gruyter. 138. Blommaert, Jan. 2001a. Investigating narrative inequality: African asylum seekers' stories in Belgium. Discourse and Society 12: 413449. Blommaert, Jan. 2001b. The other side of history: Grassroots literacy and autobiography in Shaba, Congo. General Linguistics 38: 133155. Blommaert, Jan. 2003a. Orthopraxy, writing and identity: Shaping lives through borrowed genres in Congo. Pragmatics 13: 3348. Blommaert, Jan. 2003b. Situating language rights: English and Swahili in Tanzania revisited. Paper presented at the African Studies Workshop, University of Chicago, Illinois, March 2003. Blommaert, Jan. In press a. Grassroots historiography and the problem of voice: Tshibumba's `Histoire du Zaire'. To appear in Journal of Linguistic Anthropology. Blommaert, Jan. In press b. In and out of class, codes and control. Globalisation, discourse and mobility. To appear in Mike Baynham and Anna De Fina (eds.) Dislocations/Relocations: Narratives of Displacement. Manchester, U.K.: St. Jerome. Bourdieu, Pierre. 1990. Language and Symbolic Power. Cambridge, U.K.: Polity Press. e. In Braudel, Fernand. 1969 [1958]. Histoire et sciences sociales: La longue dure Fernand Braudel Ecrits sur l'Histoire. Paris: Flammarion. 4183. Chouliaraki, Lilie and Norman Fairclough. 1999. Discourse in Late Modernity: Rethinking Critical Discourse Analysis. Edinburgh: Edinburgh University Press.
# Blackwell Publishing Ltd. 2003

622

BLOMMAERT

Eastman, Carol and R. F. Stein. 1993. Language display. Journal of Multilingual and Multicultural Development 14: 187202. Fabian, Johannes. 2001. Anthropology with an Attitude. Stanford, California: Stanford University Press. Gal, Susan and Kathryn Woolard (eds.). 2001. Languages and Publics: Authority and Representation. Manchester, U.K.: St Jerome. Gumperz, John and Stephen Levinson (eds.). 1996. Rethinking Linguistic Relativity. Cambridge, U.K.: Cambridge University Press. Hanks, William. 1990. Referential Practice: Language and Lived Space among the Maya. Chicago, Illinois: University of Chicago Press. Harris, Roxy, Constant Leung and Ben Rampton. 2001. Globalisation, diaspora and language education in England. Working Papers in Urban Languages and Literacies 17. London: King's College London. Hymes, Dell. 1966. Two types of linguistic relativity (with examples from Amerindian ethnography). In William Bright (ed.) Sociolinguistics: Proceedings of the UCLA Sociolinguistics Conference, 1964. The Hague: Mouton. 114167. Hymes, Dell. 1968. Linguistic problems in dening the concept of `tribe'. In J. Helm (ed.) Essays on the Problem of Tribe. Seattle, Washington: American Ethnological Society and University of Washington Press. 2348. Hymes, Dell. 1980. Language in Education: Ethnolinguistic Essays. Washington, D.C.: Center for Applied Linguistics. Hymes, Dell. 1996. Ethnography, Linguistics, Narrative Inequality: Toward an Understanding of Voice. London: Taylor and Francis. Jacquemet, Marco. 2000. Beyond the speech community. Paper presented at Seventh International Pragmatics Conference, Budapest, Hungary. Kroskrity, Paul (ed.). 2000. Regimes of Language. Santa Fe, New Mexico: SAR Press. Maryns, Katrijn and Jan Blommaert. 2001. Stylistic and thematic shifting as a narrative resource: Assessing asylum seekers' repertoires. Multilingua 20: 6184. Maryns, Katrijn and Jan Blommaert. 2002. Pretextuality and pretextual gaps: On de/rening linguistic inequality. Pragmatics 12: 1130. Milroy, James and Leslie Milroy. 1985. Authority in Language: Investigating Language Prescription and Standardisation. London: Routledge and Kegan Paul. Mufwene, Salikoko. 2002. Colonisation, globalisation, and the future of languages in the twenty-rst century. MOST Journal of Multicultural Societies 4: 148 http://www.unesco.org/most Rampton, Ben. 1995. Crossing: Language and Ethnicity among Adolescents. London: Longman. stman, Jan Rampton, Ben. 1998. Speech community. In Jef Verschueren, Jan-Ola O Blommaert and Chris Bulcaen (eds.) Handbook of Pragmatics 1998. Amsterdam, The Netherlands: John Benjamins. 130. Rampton, Ben. 2001. Critique in interaction. Critique of Anthropology 21: 83107. Schieelin, Bambi, Kathryn Woolard and Paul Kroskrity (eds.). 1998. Language Ideologies: Practice and Theory. New York: Oxford University Press. Silverstein, Michael. 1998. Contemporary transformations of local linguistic communities. Annual Review of Anthropology 27: 401426. Silverstein, Michael. 2001 [1981]. The limits of awareness. In Alessandro Duranti (ed.) Linguistic Anthropology: A Reader. London: Blackwell. 382401. Wallerstein, Immanuel. 1983. Historical Capitalism. London: Verso. Wallerstein, Immanuel. 2001. Unthinking Social Science (second edition). Philadelphia, Pennsylvania: Temple University Press.
# Blackwell Publishing Ltd. 2003

A SOCIOLINGUISTICS OF GLOBALIZATION

623

Williams, Raymond. 1977. Marxism and Literature. Oxford, U.K.: Oxford University Press. Woolard, Kathryn. 1998. Simultaneity and bivalency as strategies in bilingualism. Journal of Linguistic Anthropology 8: 329.

Address correspondence to: Jan Blommaert Ghent University Department of African Languages and Cultures Rozier 44 B-9000 Ghent Belgium jan.blommaert@ugent.be

# Blackwell Publishing Ltd. 2003