Abstract

As a way of tracking popular framing of CMC, this article critically reviews an international corpus of 101 print-media accounts (from 2001 to 2005) of language-use in technologies such as instant messaging and text-messaging. From the combined perspective of folk linguistics and critical discourse analysis, this type of metadiscourse (i.e., discourse about discourse) reveals the conceptual and ideological assumptions by which particular communication practices come to be institutionalized and understood. The article is illustrated with multiple examples from across the corpus in order to demonstrate the most recurrent metadiscursive themes in mediatized depictions of technologically- or computer-mediated discourse (CMD). Rooted in extravagant characterizations of the prevalence and impact of CMD, together with highly caricatured exemplifications of actual practice, these popular but influential (mis)representations typically exaggerate the difference between CMD and non-mediated discourse, misconstrue the 'evolutionary' trajectory of language change, and belie the cultural embeddedness of CMD.

Popular Discourse and Online Discourse

Online language has developed into a shorthand that all but obliterates the Queen's English. Our kids log on and catch the Webspeak virus. This new communicable disease spreads like jam on toast and, presto, Spell-Drek: The Next Generation. {1: 101, headline}

As a dialect, text ('textese'?) is thin and unimaginative. It is bleak, bald, sad shorthand. Drab shrinktalk. The dialect has a few hieroglyphs (codes comprehensible only to initiates) and a range of face symbols. … Linguistically it's all pig's ear. … Texting is penmanship for illiterates. {2: 29}

Much scholarly discourse has been committed to challenging common assumptions that technologically-mediated modes of communication are necessarily impoverished and antisocial. In the case of computer-mediated communication (CMC), a great deal of research evidence now exists that demonstrates the potential for online social interactions to sustain and even enhance human communication (see Walther & Parks, 2002). By the same token, it is also perceived scholarly wisdom that generalizations about CMC are inherently problematic, conflating as they do important differences in the specific affordances and communicative practices of different technologies. As Herring (2001) notes, CMC is clearly affected by technological variables such as synchronicity, granularity (i.e., how long or short text may be), and multimodality (e.g., whether or not graphics, audio, and video are included). There is also a range of social variables that empirical research shows influences the nature and experience of CMC, such as the amount of time participants spend online (Walther & Burgoon, 1992), the nature of their relationship (Walther, Slovacek, & Tidwell, 2001), and their levels of motivation (Utz, 2000). In this regard, contemporary scholarship has come to recognize the relative inseparability of mediated and unmediated communication; both are equally situated and context-dependent, and mediated practices are intricately embedded in the daily lives of users (Howard, 2003).

None of this scholarly insight guarantees, of course, that popular discourse necessarily follows suit. Indeed, for scholars of CMC it is widely accepted that all communication technologies are accompanied by heightened popular reactions regarding their impact on the existing social order (Kling, 1996; Standage, 1999). Whether in terms of people's experiences of community life, their standards of morality, or the way they organize their personal relationships, public discourse about emerging technologies is also typically polarized by judgments of their being either 'all good' or 'all bad.' One major narrative thread in public discourse about emerging technologies involves concerns about the way language is affected (Baron, 2000), and any perceived threats to conventional or standard language practices are invariably met with the same anxiety people have about all language change (Cameron, 1995). Public discourse about language is typically marked by attempts to control the course of language change through the proscription of disfavored forms and the prescription of familiar ones. Given its undeniable influence on language (Baron, 2000), technology often becomes the focus of these public debates about declining standards of 'good' language use. Nowhere is this more manifest than in the print media where, as with the examples given above, journalists' evaluations of the impact of technology on language can be anecdotal, dismissive, and, sometimes, surprisingly vitriolic.

Whereas mainstream CMC research is increasingly able to refute popular and academic misrepresentations of online communication, scholarly attention to computer-mediated discourse (CMD) has been much less forthcoming.¹ With notable exceptions (e.g., Baron, 1998; Herring, 1996, 2001), CMC scholarship has tended to focus on the dynamics of interpersonal and group communication rather than the specifics of linguistic practice. Thurlow (2001; also Thurlow, Lengel, & Tomic, 2004) proposes five ways in which CMD scholarship might be characterized for analytical convenience: multilingualism (studies of national and other language demographics and politics), language change (studies of linguistic forms and orthographic practices), discourse (studies of online conversational practices), stylistic diffusion (studies of the spread and codification of online language), and metalanguage (studies of the representation of online language/discourse in the media). Thus far, scholars have been successful in tackling the first three of these areas. For example, papers in this journal have recently addressed the prevalence and status of different languages online (Danet & Herring, 2003; Warschauer, El Said, & Zohry, 2002). A handful of scholars have also published studies that focus on lexical, syntactic, and grammatical dimensions of language use online (e.g., Androutsopoulos, 2000; Androutsopoulos & Schmidt, 2002; Baron, 1998) and in mobile telephony (e.g., Hård af Segerstad, in press; Thurlow, 2003). Some of the best-known work in CMD has been concerned with the organization of interactional processes in language (e.g., Herring, 2001, 2004b). Far less academic attention, however, has been paid to the way that particular features and practices of CMD have spread into mainstream usage; a rare example is offered by Baron (2002) which, extending her earlier work (Baron, 1998), examines issues of linguistic diffusion and/or stylistic uptake. This excellent research is supplemented by a few other works (e.g., Herring, Kouper, Scheidt, & Wright, 2004; Koutsogiannis & Mitsikopoulou, 2003). Nonetheless, the least written about (in scholarly terms) dimension of CMD continues to be the way it is commonly talked about and represented more widely—in other words, its metalinguistic or metadiscursive construction.

Folk Linguistics and Critical Discourse Analysis

All language and language-use (or discourse) is inherently and self-reflexively metalinguistic insofar as meaning making always relies on the implicit, interpretative-pragmatic framing of language and communication (Silverstein, 1993; see also Gumperz, 1982, on 'metamessages' and Schiffrin, 1987, on 'discourse markers'). However, the term metalanguage is also used to refer to people's explicit, conscious, and articulated reflections about language-in other words, their talk about talk. For some time, scholars have promoted the value of studying lay metalanguage and non-specialists' understanding of language (e.g., Hoenigswald, 1966; McGregor, 1998; Preston, 1996). Preferring the term 'folk linguistics,' Preston (1996, p. 72), for example, argues that metalanguage is "worthy of study not only for its independent scientific value but also for the undeniable importance it has in the language professional's interaction with the public." In this regard, the study of metalanguage is concerned with investigating what people know and say about their own and other people's language practices. This approach starts from the premise that lay interpretations of, or attitudes towards, accents, dialects, grammatical usage, and so on are "entirely legitimate" (McGregor, 1998, p. 33) and evidently meaningful to people in the contexts of their everyday communication. As such, the objective of metalinguistic research is not necessarily to evaluate folk linguistics as right or wrong against scholarly standards, not least because people's inability to articulate their understanding (declarative knowledge) does not necessarily preclude their understanding how to use language effectively (procedural knowledge) (Multhaup, 1997). Similarly, Ryle (1963, p. 29) dismisses what he calls the "intellectualist legend" which holds that anything people do is preceded and steered by a separate, internal act of thinking, considering, etc.

Notwithstanding this, people's beliefs about language and communication do serve to guide their practices (Giles, Coupland, & Wiemann, 1992) and are important for a "theory of human cultural development" (Garrett, Coupland, & Williams, 2004, p. 194). Studies of folk linguistic and metalanguistic comments offer insights into the ways people value language and communication and how they view its role in their lives. It is on this basis that Preston (1996) proposes three objects of metalinguistic analysis:

• the availability of information people have about language
• the degree of 'accuracy' involved in these descriptions
• the nature and specificity of detail provided.

These perspectives of metalanguage are just as easily applied to all communicative practices and need not imply an exclusive concern for formal aspects of language such as perceptual dialectology in Preston's work. It is useful to think in terms of metadiscourse and non-specialist commentary about not only the forms of language-use, but also its social functions and the organization of social interaction through language.

On this point, the theory/method of Critical Discourse Analysis shares a concern for understanding language about language or communication about communication (Cameron, 1995; Fairclough, 1992). Cameron (1995) has shown how popular discourse about language is in fact an integral component of social life. Critical discourse analysts have examined how metadiscourse not only reflects attitudes and beliefs, but is also powerful in constituting ideologies of difference and structures of social inequality (Cameron, 2000; Fairclough, 1992, 1999). This is particularly so in the case of mediatized discourse, which acts as a powerful gatekeeper (e.g., van Dijk, 1993). Where scholars may deliberately avoid judging lay metalanguage, mediatized metadiscourse warrants evaluation and critique precisely because of its institutional power and public influence. Underpinning the work of much CDA is the belief that metalanguage (or metadiscourse) does not merely 'guide' the communication of everyday speakers, but is instrumental in privileging certain ways of communicating over others. Although this may also appear to be done on linguistic grounds and for the sake of communicative transparency, popular metadiscourse also establishes hierarchies of symbolic and economic capital (Bourdieu, 1991).

Current Study: The Metadiscursive Construction of CMD in the Media

As with the examples cited at the start of this article, newspaper and other media reports often seem to portray CMD in a negative light. What is particularly troubling, however, is the tendency for these metadiscursive evaluations also to dovetail with a rising public discourse about the communicative ineptitude of young people (Thurlow, 2003, 2005). Often, it seems, adult anxieties about youth, about technology, and about language merge into a kind of 'triple-whammy' panic about declining standards of morality and the unwinding of the social fabric. Where Thurlow (2003) sought to challenge these popular discourses about CMD by comparing them with a corpus of young people's actual cell-phone text messages, the current article extends this empirical work by undertaking a more systematic examination of the media representations that powerfully frame CMD practice.

Data Generation

With the help of two research assistants, the ProQuest and LexisNexis newspaper databases were searched for any English-language news articles between 2001 and 2005 covering issues related to young people, language, and new technology. (Search terms included: language, teenagers, adolescents, adolescence, youth, young people, technology, email, text messaging, instant messaging.) An initial sample of 156 different news stories was eventually condensed to form a dataset of 101 articles specifically addressing young people's language practices with new media such as the Internet and mobile phones.² While the majority of these articles were from national and regional British (36) and United States (33) newspapers, the rest came from Canada (9), New Zealand (7), Ireland (4), Singapore (3), Malaysia, Philippines, Hong Kong, Australia (2 each), and Indonesia (1). As is typical industry practice, a handful of news stories in the corpus were based on syndicated reports or had picked up on stories reported by other papers.

Analysis of Discourse Data

The analysis of the corpus followed critical discourse analytic principles (e.g., Fairclough, 1999, 2003).

Critical discourse analysis aims to provide a framework for systematically linking properties of …texts with features of their social and cultural circumstances. Particular discursive events … are described in terms of the potentially innovative ways in which they draw upon the orders of discourse which condition them. (Fairclough, 1999, pp. 79-80)

Accordingly, whereas content analysis of texts usually prioritizes quantification, the present analysis favors a more interpretive, critical approach that highlights striking themes rather than statistical patterns. As is true of much qualitative analysis, any interpretations therefore make few direct claims to representativeness but instead appeal to an informed judgment of typicality, supported by the inclusion of multiple examples selected from a wide range of data sources. The current study also works within that style of CDA that orients less to the inner workings of texts and more to the distinctive 'texturing' or discursive organization of social processes (Cameron, 2000; Fairclough, 2003). In other words, the objective is to examine the broad semantic and evaluative fields that are established linguistically and to identify those recurrent narrative resources 'threaded' throughout the corpus. In this way, these media texts are seen to draw on and privilege particular points of view and particular orders of discourse. The ultimate goal of this critical review is to stimulate further research on the metalinguistic dimensions and/or mediatized representations of CMC and CMD.

Focus of Analysis

In a sister publication (Thurlow, in press), I have analyzed and discussed different aspects of this corpus with specific reference to the 'technologization' of young people in adult representations of their new media language use. The current article, in contrast, is focused on the metadiscursive construction of CMD more generally. In this regard, and with reference to the objectives of folk linguistics (Preston, 1996), the corpus has been analyzed with a view toward characterizations of the nature of CMD, descriptions of the spread and impact of CMD, and exemplifications of actual CMD practice.

On this basis, the discussion that follows considers the dominant metadiscursive themes by which CMD appeared to be commonly characterized, described, and exemplified in these newspaper articles. In presenting these inter-related themes, the intention is not necessarily to account comprehensively for the content of the corpus in its entirety; instead, sequentially numbered extracts from newspaper articles are cited as typical examples; these are indicated in bold and cross-referenced to the original newspaper article listed in the appendix of primary resources.

Theme 1—Marks of Distinction: CMD as a Linguistic Revolution

CMD is often framed in popular discourse through the coining of labels such as 'netlingo,' 'weblish,' 'netspeak' and so on (Thurlow, 2001). As a rhetorical strategy, these neologistic naming practices work in part to establish CMD as somehow unique or distinctive from standard English, as a fully-fledged or 'new' language (Extracts 3 & 4). In keeping with this strategy, at least nine articles in the current corpus set the tone by explicitly depicting the emergence of CMD as a form of linguistic 'revolution,' implying a decisive and dramatic break with conventional practice (Extract 5, also Extract 13).

A language all of its own. {3: 22, headline}

A new language of the airwaves has been born. {4: 45}

Not since man uttered his first word and clumsily held a primitive pencil nearly 10,000 years ago has there been such a revolution in language. From tapping abbreviated words into a mobile phone to emailing people on the other side of the Atlantic, today's technology is changing the way in which we communicate at an alarming rate. {5: 24}

By itself, this metaphoric framing of CMD might be passed off as journalistic license; however, the combination of a series of rhetorical devices establishes this as a more dominant theme. Regardless of whether CMD was explicitly labeled revolutionary, a similar rhetoric of uniqueness and distinctiveness was evident throughout. For example, from across the corpus reference was made to:

a text messaging movement; a shorthand language; a nouveau form of communication; a lexicon for electronic communication; a virtual new written language; a new language; a language of its own; a hybrid language; a whole new language; a separate, private language; a language revolution; new truncated language; a new dialect; text messaging lingo; hottest new language; telegraphic shorthand; electronic lingo; a language all of its own; a lingua franca; the lingo of generation text; lingo; technobabble; weblish; textese; NetLingo; an abbreviated language; a second language; new argot; new idiom; mysterious lexicon; a truncated language; digital dialogue; webbish; new language of smileys and abbreviations; new shorthand language; high-tech lingo; a new language called 'globespeak;' a new written language; text-messaging lingo; a new abbreviated language; slanguage; a sub-language.

In many of these cases, the added use of the indefinite article in phrases like 'a new language' as opposed to 'new language' further reiterates the implication of a distinctive variety (see also Extracts 3 & 4 above).³

Distinction through Equivalence

Claims to the extent and distinctiveness of CMD were also made through more subtle rhetorical means, in particular by establishing 'relations of equivalence' (Fairclough, 2003; p. 87 et seq.) between CMD and 'proper' or standard languages. For example, references to bilingualism or the notions of translation and/or fluency were not uncommon (Extracts 6 & 7). By the same token, talk about the diffusion of CMD into mainstream usage, its codification into dictionaries, and other forms of official recognition or acceptance all served to create an image of CMD as a new, distinctive variety with the status and material substance of a language (Extracts 8-10).

• bilingualism, e.g., most texters are, in essence, bilingual {6: 25}
• fluency, e.g., thousands of teens … are fluent in another language {7: 36}
• codification, e.g., hallelujah for the world's first text-messaging dictionary {8: 64}
• diffusion, e.g., text chats are starting to bleed over into others aspects of life {9: 57}
• official recognition, e.g., words from text messaging …find their way into the highest authority on the English language, the Oxford English Dictionary {10: 14}

Together with the broader theme of 'revolution,' these relations of equivalence serve to exaggerate further the nature and extent of CMD.

Privileged Voices of Commerce

Perhaps not surprisingly, heightened claims for the ubiquity and spread of CMD were often a feature of direct quotes from, or the implied reported speech of, people in the telecommunications industry. These are people for whom there is a clear vested, economic interest in promoting the novelty and uptake of new communication technologies (cf. Silver & Garland, 2003). Once again, statements were framed by an exaggerated sense of CMD being markedly different from standard linguistic practice (e.g., in the extracts which follow: 'unique,' 'revolution,' 'new language'), improbable claims for its novelty (Extract 14), or forecasts for its impact (Extract 15).

"They have their own language. It's unique." {11: 13, vice-president of Radio Shack Canada}

"It's really a new language and a new grammar." {12: 48, marketing manager at Qwest}

Messaging by cell phone is the biggest revolution in communications since the advent of e-mail. {13: 28, vice-president of 3G Americas}

No one has really changed the English language like this since Webster's dictionary first came out. {14: 48, president of transl8it.com}

Text messaging might one day be as popular as talking. {15: 43, AT&T spokeswoman}

Although many articles were supplemented by mitigating commentary from scholars and also from everyday user-communicators themselves, commercial sources constituted a dominant voice across the corpus as a whole. This complicity between journalism and big business accounts for some of the over-eager, 'revolutionary' framing of CMD; in turn, these dramatic characterizations of CMD also reinscribe popular misconceptions about the acute (rather than chronic) nature of both technological change and language change. This particular metadiscourse is also premised on a common folk-linguistic misunderstanding about the inevitable hybridity of language varieties and the irregular trajectory of language change (Bauer & Trudgill, 1998).

Theme 2—Statistical Panic: The Rise and Spread of CMD

One key rhetorical resource through which the prevalence of CMD was made vivid was the use of numerous, superlative numerical citations that appeared throughout the corpus.

More than 16 million text messages are sent every day between Britain's 40 million mobile users. That's ten times as many as last year, and experts say the total sent in 2000 could top an amazing SEVEN BILLION. {16: 26}

By 2004, a projected 10.5 billion will be sent in the United States alone. Worldwide, the number of messages is projected to increase from 20 billion last year to 82 billion by 2004. This works out to a rise from $1.73 billion in worldwide text-messaging revenue to $6.6 billion in 2004. {17: 65}

As with Extracts 16 and 17, and in at least 76 other instances across the corpus, the following sample reveals how figure citations usually depicted the volume of messages being sent (usually text but also instant messages; Extracts 18-21), the rate of increase (Extracts 22-24), the revenue generated (Extracts 25 & 26), and/or the demographic profile of users (Extracts 27-29).

Superlative claims of quantity

• more than 1 billion text messages are sent every month in the UK {18: 54}
• billions of text messages are already being sent daily in this country {19: 28}
• in Europe alone more than two billion SMS messages are exchanged each month, while worldwide the figure is put at around 15 billion a month {20: 45}
• 75 million users sending more than 700 million real-time [instant] messages a day {21: 53}

Superlative claims of growth

• text messaging is expanding at the rate of 1,800% a year {22: 94}
• in New Zealand texting has grown from fewer than 60,000 a day in 1999 to more than 10 million a day this year {23: 71}
• the number of texts sent daily by Vodafone costumers in Ireland alone has dramatically increased from 25,000 daily in 1999 to five million a day in 2002. {24: 85}

Superlative claims of monetary value

• an estimated 20 billion text messages sent worldwide last month - worth almost £4 billion to Europe's phone operators. {25: 41}
• US carriers are starting to tap into this market, estimated at $20 billion worldwide. {26: 63}

Superlative claims of usage

• nearly three quarters of online teens use instant messaging {27: 21}
• text totals of 2000 to 3000 a month are common for older teenagers {28: 6}
• a text bomber racked up 27,000 messages a month {29: 71}

On one occasion (two different articles), the same effect of quantity is produced but with the added implication that CMD is disrupting conventional practices—in this case, the sending of Valentines cards (Extracts 30 & 31) and the termination of romantic relationships (Extract 32). In these instances an oppositional discourse of new/old, of mediated/'unmediated' and, usually by implication, of appropriate/inappropriate is invoked.

This Thursday Britain's phone network is expected to [be] swamped by up to 60 million Valentine messages. {30: 58}

On Valentine's Day last year, 78 million text messages were sent. The messages outnumbered Valentine's Day cards by six to one. {31: 55}

13 percent of all cellphone users have ended romantic relationships with text messages {32: 28}

However inaccurate or contradictory figures may be (compare Extracts 18, 19, and 20), they serve important rhetorical, persuasive functions. Most obviously, there is the straightforward display of quantity: The larger the numbers (e.g., '1,800%' in Extract 22), the greater the escalation and ubiquity of CMD. An added claim to legitimization is created through the scientific and/or objectivist connotations of statistics themselves; in other words, we are persuaded that the rise and spread of CMD is fact. The persuasion may, however, be even more subtle than this. Just as detail in everyday conversational narratives fosters the perception of authenticity (Tannen, 1989), the repetitive use of numeric detail in these articles also establishes as 'real' the metadiscursive representation of the nature of CMD as a whole. This taste for excessive, quantitative detail is arguably also symptomatic of what Woodward (1999) characterizes as the "society of the statistic," where a kind of statistical panic instills a perpetual state of concern or fear in people. As such, media representations are again seen to be either generating or at least feeding popular, social anxieties about the impact of new media.

Heightened Narrative Impact

Surprisingly or not, in most cases no organization was specified as the source for figures cited in the articles. Once again, the credibility of figures is less important than their dramatic, narrative effect. Of those sources clearly identified (in 34 cases, less than half of the figure citations), moreover, almost all were supplied by commercial organizations, whether communications providers themselves, industry representatives, or research companies.

• Communications providers: BT Cellnet (x3), Vodaphone (x3), Verizon, AOL (x2), NZ Telecom (x2), Cingular, T-Mobile, Upoc (x2), Tegic Communications
• Industry representatives: Cellular Telecommunications and Internet Association (x5), Mobile Data Association, Mobile Marketing Association
• Commercial research organizations: Nielsen/NetRatings (x2), Teen Research Unlimited, Yankee Group, IDC, Telephia, Forrester Research, Mobile Lifestreams

The only non-commercial organization actually cited (in three cases) was the charitably funded Pew Internet & American Life Project. Not only are industry voices given ample opportunity to frame CMD in accordance with their particular commercial agenda, but accounts for the spread of CMD hinge on notoriously inflated and methodologically questionable commercial 'webmetrics' (Thurlow et al., 2004). Partly as a result of this, the nuanced qualities and social meanings of situated CMD practice are often overshadowed by a journalistic preference for generic, quantitative depictions of its rise and spread (as in Extracts 27-29). Just as CMD has been depicted as creating a whole new culture (or, elsewhere, cr8ts a hul nu cltur) or, indeed, as a cultural revolution, the assumption that underpins this kind of 'statistical panic' is that CMD is not only ubiquitous but also far-reaching in its impact.

Theme 3—Moral Panic: CMD, Literacy, and the Social Order

Manifesting the deterministic view of an unstoppable "technological imperative" that characterizes many popular representations of computer-mediated communication (Thurlow et al., 2004, p. 41), CMD was depicted throughout the corpus as a craze (x5), a mania, a fever, a huge phenomenon, a cultural phenomenon, a meteoric rise, or a youthquake which was then variously described as booming, blooming (x2), expanding, exploding (x4), rocketing, gaining ground, dominating, proliferating, rising (x3), brewing, zooming, leaping, gripping the nation/country (x2), taking by storm, taking the world by storm, changing at an alarming/unprecedented rate (x2), or getting loose. Reported here in isolation from their original contexts, it is not automatically clear whether items from this metaphoric lexicon were intended negatively or positively. Notwithstanding the ambiguity of valence, the semantic field established connotatively depicts CMD's rise and spread as excessive and, once again, revolutionary. In addition, other descriptions of stealth, disease, and inundation (e.g., crept, riddled, spawns (x3), spreading like wildfire, flooding; see also Extract 1 above) were more explicit in establishing a generally pejorative tone.

While it would be untrue to suggest that there were no positive claims made for the effects of CMD (see below), for the most part the nexus of popular discourses about language, about technology, and about young people generates an overwhelmingly pessimistic picture (see also Thurlow, 2003, in press). In addition to being described as reprehensible, frightening, depraved, infamous, criminal, jarring and abrasive, apocalyptic, execrable, pointless, and aberrant, CMD was often held responsible for a number of wider social and educational ills. For example, certain journalists and commentators regarded it as being inflicted on the innocent public {20}, creating a whole new culture in the country {9}, dumbing down the English language {100}, and lowering standards all round {11}.

In this regard, perhaps the most widely and commonly expressed social anxieties in the current corpus were those related to the deleterious impact of CMD on standard English. Indeed, the most dominant theme throughout the corpus of newspaper articles analyzed here was an over-riding sense of moral panic about declining standards of literacy. As one journalist put it, CMD signals the slow death of language {35}.

Fears are growing that today's teenagers are becoming 'Generation Grunt', a section of society that has effectively lost the ability to talk or express itself. {33: 4}

If the already ingrained corruption of the English language is perpetuated, we will soon be a nation made up entirely of grammatical duffers. {34: 19}

The Great Grammar Crusade

What Cameron (1995, p. 78 et seq.) refers to as the "great grammar crusade" is unquestionably a central organizing principle in popular metadiscourse. However unfounded, concerns about threats to the status of language and the protection of linguistic norms invariably center on anxieties about standards of spelling, punctuation, and grammar among young people. As such, the presence of this overarching metadiscursive theme in the corpus as a whole is not in and of itself surprising; what is more surprising was its prevalence and extreme one-sidedness.

By the same token, articles occasionally extrapolated wildly from these concerns about declining standards of literacy to reflect on the wider impact on social order, as in the following examples from two different British newspapers:

Text messaging … is posing a threat to social progress. {35: 1}

"The English language is being beaten up, civilization is in danger of crumbling." {36: 20}

In recognizable journalistic style, the same link between social decay and falling standards of language was discussed in a Vancouver Sun article intertextually headlined as The decline and fall of spellin' & writin' {101}. The ramifications of CMD and declining standards of literacy were taken to extremes in a fourth article which concluded with the following series of unsubstantiated, tenuously connected claims:

Little wonder, then, that researchers have found poor basic skills greatly increase the likelihood of individuals failing to secure or hold down jobs. Those with poor literacy skills are likely to be among the prison population or in sink estates. Poor health and relationship breakdowns are associated with this lifestyle. {37: 40}

In this last extract, the revolutionary rhetoric discussed above is rendered almost apocalyptic and the general tone of negativity most unambiguously established. As before, the listing of these items implies a relation of equivalence and their inclusion in an article on young people, language, and new media encourages an interpretation of causality. While it would be unfair to suggest that the entire corpus was characterized by such extreme depictions of causality, the link between new technologies and declining standards of literacy was evident throughout.

As it happens, there are also tentative research findings that dispute any causal link between declining standards of academic literacy and the rise of CMD (De Vries & Van Der Meij, 2003; Raval, 2002). Most scholarship is certainly inclined to recognize any number of factors contributing to language change rather than singling out new communications technologies as the sole or even primary cause (see Thurlow et al., 2004).

Theme 4—ROTFLMAO: The Fetishization of CMD⁵

Appalled teachers are now presented with essays written not in standard English but in the compressed, minimalist language of mobile phone text-messaging: Gd sAv R grAshz QE2 Gd sAv R nObl QE2 Gd sAv D QE2. {38: 40}

An almost genre-defining feature of the articles in the current corpus was humorous, tokenistic displays of examples of text messaging or text messaging language practices. For example, almost a third (30) of all articles used some example of text messaging or instant messaging style in their headlines (see Appendix). Beyond this, actual CMD practices were either given as in-text examples (with 'translations') or, in some cases, offered as a glossary or 'Do-It-Yourself' listing at the end of articles.

For the most part, the depiction of supposedly real stylistic features relied on a fairly restricted repertoire of hackneyed examples (IMHO, CU, L8R, BRB, LOL, GR8, ASL), most of which tended to conceal the origins of these linguistic practices and to conflate different technologies (e.g., instant messaging, text-messaging, and email). In almost every case, CMD linguistic forms were only ever rendered in terms of abbreviations, acronyms, and emoticons, with little of the scholarly specificity of letter-number homophones, contractions, or accent stylizations which might otherwise reveal something of their subtlety, technique, and variety (Shortis, 2001; see also Androutsopoulos, 2000). Altogether, the exemplification of CMD was greatly simplified and caricatured.

What would Socrates think of instant messaging? {39: 57}

And to think this happened in the land of Shakespeare. If the bard were alive today, he'd probably write, '2B or not 2B…'. {40: 48}

The oppositional rhetoric alluded to above is also evidenced by constantly setting CMD in negative opposition to 'proper' language and received, canonical symbols of acceptability (e.g., the poetry of Shakespeare, in Extract 41, or, in Extract 42, the novels of Jane Austen). No doubt meant humorously, the intention behind these parodic exemplifications was always to depict the subversion of standard, educated language use; by implication CMD is rendered uneducated and therefore unacceptable. That these and many other examples are fabricated exaggerations of real CMD practice is clearly irrelevant.

Shall I compare U 2 a 0's day? U R mo luvlE & mo temperate. Rough winds do shAk d darling buds of mA, & 0's lease hz all 2 sht a D8. {41: 53}

It Is A TrOth UniverslE Aknowledgd, Tht a SngI Man In PoSeSn Of A GOd 4tun, Mst B In 1 nt Of A Wit {42: 23}

What of comparable examples of less comment-worthy acronyms such as ASAP, AKA, BTW, AWOL, SWALK? There is also a lexicon of official acronyms we function with daily from TV to BBC to CNN and from USA to EU to UN. As the editors of any dictionary know, there are always precedents for non-standard forms making their way into standard language use. This is seldom considered in articles about CMD, which appear more concerned with presenting exaggerated and extreme examples of instant- and text-messaging styles. Many of these are not only improbable but counter-intuitive, requiring precisely the kind of effort or space that almost always characterizes the style of online and cell phone discourse (Thurlow, 2001).

Fictionalized Accounts of CMD

In many cases, journalists appeared to have made up their own exaggerated examples of CMD by 'translating' otherwise unlikely phrases into 'textese,' often using the commercially available transl8it.com. While stylistic forms such as the use of capitalization to indicate elongated vowels are promoted in commercial publications such as the Ltl bk of Txt Msgs, these would again seem counter-intuitive for most texters. Certainly these examples of supposedly typical messages seem highly improbable, if not completely fabricated:

Mst f d tym dey usd ds knd f lng'ge 2 tlk 2 1 anthr nt 1ly n txt bt evn n wrtng ltrs 2 {43:9}

lfYaMthWozNEBiGrUWdntHavNEFAcLft2Wsh {44: 23}

The exaggeration of the distinctiveness of CMD is also mirrored by the overstated lack of mutual intelligibility between CMD and standard English (hence 'hieroglyphics' in Extract 2). As such, CMD is represented in a way that may feed existing adult mythologies about the inscrutability of young people's communication in general (Thurlow, 2005). By the same token, it seems unlikely that most messagers would fail to see a qualitative distinction between the widely recognized and increasingly institutionalized :-) smiley and numerous examples of unfamiliar, semantically obscure smileys which were depicted throughout (e.g., from three separate articles: (:-… , :-r or :-)~ for 'a broken heart,' 'a raspberry,' and 'a dribbling idiot,' respectively). In these cases, and in the so called 'glossaries,' examples of instant- and text-messaging style were usually presented as being of the same order in terms of prevalence and applicability; at no point did any article in the current corpus allow for changing fashions, subcultural and age-related variations, or differences in personal style. Without exception across the corpus as whole, there was no evidence of ethnographic or other empirical validity for exemplifications that appeared to be based largely on popular and anecdotal hearsay.

One example of the way in which actual practice may come to be misrepresented is the purported confusion of LOL. In a February 2001 article for the British Daily Telegraph, Professor Jean Aitchison (formerly Chair of Language and Communication at Oxford University) is quoted as having commented on the potential for LOL to be misunderstood as either 'laugh out loud' or 'lots of love.'

"Mostly they are original but sometimes you get a clash of meanings. For example, take LOL which can mean both Laughing Out Loud and Lots Of Love. That could lead to some embarrassing misunderstandings." {45: 14}

This same point was then reported in two subsequent articles, first as a general comment in the Vancouver, WA Columbian (October 2001) and then three years later in a quoted statement for the British Observer (March 2004). On neither occasion was the original source attributed.

… understanding the messages can sometimes be confusing. For example, LOL could mean 'laugh out loud' to one person and 'lots of love' to another. {46: 55}

The problem is when people get so familiar with texting they inflict it on the innocent public. Then it can lead to a breakdown in communication. For instance, in text messaging LOL can mean 'lots of love' or 'laughing out loud'. If, say, you were texting your mother, you can see how things like this can get out of hand. {47: 20}

By no means an isolated incident in the current corpus, this appears to be a revealing indication of the recycling of news stories and putative 'facts' about CMD; as such, it also exposes the mimetic transmission of unfortunate misunderstandings or unwitting misrepresentations more widely. Certainly, it is hard to imagine most young text messagers or instant messagers being confused by the meaning of LOL. (My own students have found the idea risible.) What this story evidently overlooks (or underestimates) is the significance of situated meanings, whereby ambiguity is usually minimized by the context of the sentence and/or conversation itself. By the same token, real CMD practice recognizes the need for basic intelligibility if communication is to be meaningful and efficient; many of the more lengthy examples offered would not qualify in this regard. The LOL story, like most of the exemplifications across the corpus as a whole, ultimately appeared to misrepresent the lived experience of CMD for young people (and others) and its relative 'ordinariness' (cf. Herring, 2004a) in their lives.

Voices of Moderation? Scholarly Commentary

First of all, e-mail replaces face-to-face communication. … because the Internet creates a virtual community, local ones, which is to say, real ones, tend to be eclipsed. … But humans, like ants and whales, are social creatures, and eventually must deal with the consequences of contradicting their own biological nature. {48: 100}

In his op-ed piece for the Montreal Gazette (Extract 48), Barry Cooper (professor of Political Science at Calgary University, Canada) offers comments that exemplify some of the more troubling scholarly opinions and perspectives represented across the corpus. This sort of commentary was by no means typical, however, and most academic commentators in fact sought to frame language, language change, and human communication in a scholarly, measured way. Indeed, the only other uniformly negative commentary came, in Extract 49 below, from Ken Lodge (senior lecturer in Linguistics and Phonetics at the University of East Anglia, UK), Robert Beard (professor emeritus of Linguistics at Bucknell University, USA), and Peter Fernandez (professor of information technology at the Asian Institute of Management, Philippines).

"This is a new kind of slang, a written slang. We've never had anything like it before." {49: 3}

Several other scholars were depicted as evaluating the issues in both negative and positive terms; namely, Judith Donath (professor at the MIT's Media Lab, USA), Brad Mudge (professor of English Literature and Popular Culture at the University of Colorado at Denver, USA) and, in Extract 50, Robert Thompson (professor of Media and Popular Culture at Syracuse University, USA).

Texting can be incredibly simple. You can fill your role of returning calls and keeping in touch with people without any pressure to be creative or witty … We're talking about language in its most stripped down kind of level… {50: 54}

Beyond these seven academics, the remaining 23 instances of scholarly commentary in the corpus were shown to be more unequivocally positive in their evaluation of CMD and CMC more generally. These scholars also represented a range of different academic disciplines: Linguistics (UK: Jean Aitchison, Ron Carter, David Crystal, Geoff Hall, Kon Kuiper, Adam Jaworski; USA: Robert Beard, Scott Kiesling, Donna Napoli, Geoffrey Nunberg, Stephan Reder, John McWhorter), English (USA: Leila Christenbury, Brenda Clarke, Kevin Koch, Neil Randall, George Rasmussen; Philippines: Mildred Rojo-Laurilla), Social Psychology (UK: Cynthia McVey; USA: Robert Kraut; NZ: Anne Weatherall), Education (USA: Erika Karres) and Communication (USA: David Silver).⁴

Voices from Within the Field

With only a handful of exceptions, academic commentators were seldom new media specialists. This may, in part, explain a tendency for scholarly commentary to contribute to the general underestimation of the nature and subtleties of CMC (e.g., 53). Just two examples of this tendency are offered by David Crystal (Extract 51) and Brad Mudge's (Extract 52) comments which, as they were reported, apparently exaggerate ('some people spend all their lives') and oversimplify ('text messaging is about creating a practical language for a specific task') matters in ways that are more consistent with popular metadiscourses about technology and communication. This somewhat reductive accounting of CMD/CMC also occasionally appeared from within the field of new media studies. In the three different articles in which Naomi Baron (professor of linguistics at American University, Washington DC) was asked for comment, she was cited as being surprisingly dismissive ('superficial interaction' Extract 53) given her important contributions to new media scholarship (e.g., Baron, 1998, 2000, 2002).

Some people spend all their lives at their computers without socialising properly. They are in a virtual world where they meet alternative characters. {51: 24}

The language of text messaging is about creating a practical language for a specific task - in this case, communicating in a small space of a little plastic window on a piece of electronic gear. {52: 48}

You're going from place to place, and instead of looking at people or the scenery or thinking about where your life is going, you call someone or text message. … [it's] filling the time gaps with superficial interaction with other people. {53: 74}

It is, of course, important to acknowledge that scholars and other specialists are often loosely quoted, quoted out of context, or misquoted by journalists; it is also possible that these academic commentators may feel compelled to say what they think journalists want to hear, just as journalists usually orient themselves towards what they think mainstream audiences will want to hear.

Undermining the Voices of Moderation

With only three notable exceptions (i.e., Veenal Raval's study at City University London, UK, Mildred Rojo-Laurilla's at De La Salle University in Manila, Philippines, and Bregje de Vries' at the University of Twente, Netherlands), most of what was written relied on apparently anecdotal evidence and, occasionally, official reporting (e.g., the AQA report) rather than empirical research that might more reliably demonstrate and confirm the nature and extent of CMD. For the most part, in spite of their largely positive commentary, scholarly voices of moderation were inevitably over-shadowed by the generally negative tone of articles as a whole.

This happened more abruptly and obviously in some cases than in others. For example, in one article {79} the following carefully qualified observation is offered by an applied linguist (Geoff Hall, Swansea University, UK): 'spelling and language are always changing, the way we spell words is not set in stone.' Immediately after this comment, however, the journalist concludes with the following seemingly contradictory statement: 'the changes we see taking place today in the language will be a prelude to the dying use of good English.' This in an article which, atypically, was headlined English should not fear the rise of text messaging. As Katz & Aakhus (2002) note with regard to mobile telephony, the relative paucity of 'expert framing' of new media issues is rendered all the more noticeable given the quantity and dominance of 'folk framing,' including the framing done by commercial stakeholders.

Making (Up) the News: Two CMD Case Studies

The kind of discourse-content analysis employed here is useful not only insofar as it identifies media representations of a particular issue (in this case, young people's CMD), but also because it reveals something of the institutional processes by which news is fabricated (cf. Bell, 1991). For example, a noticeable feature of many articles in the corpus was the consistent use of unnamed agents, unspecified sources, and unattributed examples.

Some linguists are hailing it as a new hybrid language. {54: 21}

"It has been an accepted fact that the capability in English communications of high school or even college students is quite lacking…" {55: 9}

In other articles, similar types of passive or deindividualized statements were made as follows: everywhere we see evidence of linguistic decline {40}, there is growing concern in some quarters {70}, experts have warned {80}, this makes some educators and linguistics experts nervous {62}.

In addition to revealing ordinary processes of syndication, another empirical benefit of reviewing in detail a corpus of over 100 newspaper articles from around the world has been the opportunity to identify other somewhat more problematic processes of 'informal syndication' and story telling. These processes in turn expose the privileging of certain, usually negative, representations of CMD. The current corpus contained two particularly striking cases of questionable, recycled news making.⁶ These case studies also serve to demonstrate how the primary metadiscursive themes are invariably combined.

Case 1—The National Examiners' Report

On November 7, 2004, the British Daily Telegraph carried an article under the headline Pupils Resort to Text Language in GCSE exams, in which its education correspondent picked up on a report by one of the UK's major secondary school examination authorities (Assessment & Qualifications Alliance, AQA). In what might be regarded either as an over-interpretation or as a gross misrepresentation, the original article contained statements such as the following:

• examiners have given warning that pupils are using text message language in CGSEs
• this is the first official acknowledgement that mobile phone shorthand is undermining standard English
• English GCSE scripts were peppered with abbreviated words which have become second nature to many youngsters
• the examiners' report suggests that such abbreviations are becoming the norm
• "It's quite appalling that school children cannot distinguish between ordinary language and text language." {56: 29}

In point of fact, the 61-page, official examiners' report (AQA, 2004, p. 15) contained only a single statement in the middle of an otherwise long section about general spelling and grammatical issues: "The usual errors with they're/their; are/our; we're/were; your/you're were frequent, and texting spellings such as U for 'you' are increasingly prevalent."⁷ Feeding directly into, and indeed helping to constitute, the continual moral panic about falling standards of literacy, this original article was subsequently picked up in a number of other articles in the current corpus where the original incident was likewise overstated. For example:

Exam papers are 'riddled' with abbreviated words and spellings. Young people seem to be throwing out the dictionary in favour of the quick and easy way of writing. {57: 79}

Examiners have noticed in recent years that the language of text messages has crept into GCSEs and a report published by the largest exam board in the country showed papers were riddled with abbreviated words. {58: 50}

Chief examiners' reports on trends in public examinations have begun to note instances of texting language in exam scripts. Some cases - including a 13-year-old Scottish pupil who wrote an entire description of her summer holidays in text-speak - have provoked concern among some teachers. {59: 70}

In Extract 59 above, the journalist mistakenly conflates the GCSE-examiners story with a completely separate incident in which a Scottish schoolgirl reportedly submitted a classroom essay using text messaging style.

Case 2—The Scottish Girl's School Essay

On March 3, 2003, the British Daily Telegraph ran a story about an anonymous Scottish teacher who claimed to have received from a 13-year-old pupil a composition completed entirely in the style of a text-message. From the original newspaper report, the following is reported:

British education experts have warned of the potentially damaging effect on literacy of cellphone text messaging after a student handed in an essay written in text shorthand. The 13-year-old girl, a student in a secondary school in the west of Scotland, explained that she found it 'easier than standard English.' 'I could not believe what I was seeing,' said her teacher, who asked not to be identified. 'The page was riddled with hieroglyphics, many of which I simply could not translate.' {60: 93}

What is noticeable from this extract is how many of the same metadiscursive tropes arise (e.g., 'damaging effects on literacy,' 'riddled,' 'hieroglyphics,' 'translate'). What is even more telling, however, is how this particular story appeared also in nine other articles in the corpus—in places such as The Scotsman, London Times, and Western Mail (all UK), the Ottawa Citizen and Montreal Gazette (Canada), Reason Magazine (USA), the Irish Times (Ireland), Jakarta Post (Indonesia), and BusinessWorld (Philippines). Reports of the story could also be found in the online versions of both the BBC and CNN. In fact, a cursory web search at the time of writing this article also revealed how the same story was featured as the focus of at least 1,630 different websites. A year and a half after the original Telegraph article, a British Guardian article, headlined Texting is No bar to Literacy, included the longest reported segment of the girl's original essay (together with a 'translation') under the sub-heading SMS: A Textbook Case.⁸

My smmr hols wr CWOT. B4, we used 2 go 2 NY 2C my bro, his GF & thr 3 :-@ kds FTF. ILNY, its gr8. Bt my Ps wr so {:-/ BC o 9/11 tht thay dcdd 2 stay in SCO & spnd 2 wks up N. Up N, WUCIWUG -- 0. I ws vvv brd in MON. 0 bt baas & .

My summer holidays were a complete waste of time. Before, we used to go to New York to see my brother, his girlfriend and their three screaming kids face to face. I love New York, it's a great place. But my parents were so worried because of the terrorism attack on September 11 that they decided we would stay in Scotland and spend two weeks up north. Up north, what you see is what you get - nothing. I was extremely bored in the middle of nowhere. Nothing but sheep and mountains. {61: 70}

Although this was surely an unfortunate misjudgment of register on the part of the young pupil, what news reports uniformly fail to acknowledge is the undeniable creativity, wit, and 'new literacy' of the girl's piece. Instead, as with the original Daily Telegraph article, subsequent articles which ran this story condemned the incident as an indictment of CMD, tending also to exaggerate and/or extrapolate from the original news report—shifting, for example, from an isolated, individual incident to a statement about the 'current generation of teenagers' (Extract 62). In fact, in the case of one article in the Indonesian Jakarta Post (Extract 63), an apparently fabricated excerpt is actually quoted.

This week we learn that the current generation of teenagers is so estranged from real language that a 13-year-old in the west of Scotland has submitted an essay to her teacher in text-message shorthand. {62: 38}

A student that was asked to write a short descriptive piece about a recent holiday included the sentence, 'U can get 2 the hotel straight from the beach', whilst it is possible to decipher what is meant it really is not appropriate for this kind of coding to be encroaching in this way. Grammatical and spelling accuracy are things that are going to be lost if this kind of coding goes on unchecked. {63: 86}

As almost quintessential manifestations of the availability heuristic (see Combs & Slovic, 1979), these articles and others like them appear to overstate the prevalence of CMD practices and their impact on standard, formal language use based on what is relatively minimal, secondhand but certainly sensational evidence. To suggest, however adroitly, that this young person's essay represents a 'textbook case' of CMD is clearly to misconstrue the realities of everyday CMD (cf. Thurlow, 2003). To suggest that it might also epitomize—and internationally too—the literacy and 'real language' of an entire generation is a gross extrapolation from the facts. That adults get away with misrepresenting young people on such a scale says a great deal about the relations of power that structure youth (Griffin, 1993; Thurlow, 2005).

General Discussion: Revolution or Evolution?

… we need to recognize that both our theories and practices have potential consequences for the functions that writing and speaking will have in future decades. These functions aren't set in stone, but evolve in ways particular to an entire nexus of social, religious, political, pedagogical, and technological developments. (Baron, 2000, p. 23; emphasis added)

Few people read the same newspaper more than once; even fewer people regularly read more than one newspaper. It is really only scholars and media analysts who make a point of reading and re-reading dozens of different papers, perhaps even hundreds. The privilege of a study such as the one undertaken here, therefore, is that it offers an otherwise unique opportunity to see how a single issue is reported in many different papers, from many different locations, and over a substantial period of time. This in turn puts the researcher in a better position to identify structural patterns, topical consistencies, and emergent cultural narratives. For a critical scholar, it also helps to reveal discursive "regimes of truth" (Foucault, 1980, p. 131), those ideological assumptions by which social practices come to be institutionally organized and popularly understood. Indeed, as some scholars note, new media practices are as much a product of the cultural narratives about them as they are about the technologies that underpin them (Silver & Garland, 2003; Sterne, 1999). In reviewing mediatized metadiscourse, therefore, the intention is not necessarily to judge the folk-linguistic claims as being necessarily right or wrong even though this may be evidentially warranted. Instead, it is more informative to use these public stories about CMD as a way to understand how lay people make sense (or not) of the role of technology in their lives, and how new technologies come to be discursively constituted and then implicated in wider social debates.

The Revolutionary Framing of CMD

While some journalists and commentators in the print media corpus analyzed here appeared more measured, and arguably more accurate, in characterizing the emergence of CMD as an 'evolutionary' development (e.g., the headlines of articles 2, 21 & 30), the general tendency was for articles to represent CMD in 'revolutionary' terms, whether implicitly (e.g., the exaggeration of linguistic distinction) or explicitly (e.g., the excesses of statistical panic). To be fair, it is by no means only lay writers who are responsible for some of the exaggerated rhetoric about CMD, as evidenced by the following comment by renowned linguist David Crystal (2001, pp. 238-239) in his widely cited book Language and the Internet:

The phenomenon of Netspeak [sic] is going to change the way we think about language in a fundamental way, because it is a linguistic singularity-a genuine new medium. …Netspeak is something completely new. … [It] is a development of millennial significance. A new medium of linguistic communication does not arrive very often, in the history of the race.

It would be hard to deny that linguistic and communicative practices are indeed changing in the face of new communications technologies. It would also be untrue to suggest that there is nothing new or nothing distinctive about the stylistic practices of, for example, text messaging and instant messaging. What is less certain, however, is whether so-called 'Netspeak' does indeed manifest such a decisive or distinctive 'revolutionary' break with conventional standards, whether, for example, in the words of one journalist, suddenly English has become a foreign language {96}. Although it may not appear or feel like this for lay people, it seems far more likely from a sociolinguistic and scholarly point of view that language and communication are changing and evolving as they always have.

As is true of CMC more generally, in spite of Crystal's (2001) millennial rhetoric about 'Netspeak,' the fact remains that new linguistic practices seldom emerge in isolation nor do they neatly replace or break completely with long-standing patterns of interaction and language use. Furthermore, as Baron (2000) notes in the quote above, technology is only ever one of many factors involved in language change. While the new—or at least newer—may feel sudden and unusual to some, this does not presuppose a drastic overthrow or complete rejection of the old ways, either empirically or phenomenologically speaking.

The Negative and Oppositional Framing of CMD

What is perhaps most striking about the print media discourse discussed here is its generally negative and oppositional framing of CMD, by which the emergence of new media language is viewed as inherently contrary and detrimental to more established modes of language and, by implication, the moral order. In this case, the specific (and explicit) object of concern is usually a perceived threat to young people's conventional literacies and/or the status of standard English, even though young people are apparently often criticized for language practices they might never actually use. In academic terms, the juxtaposition and de facto evaluation of lay texts against received literary canons is itself inherently problematic (cf. Lewis & Fabos, 2005). That no lay speakers nowadays actually use the English of Shakespeare is also a point seemingly overlooked by new media language mavens.

Elsewhere, adult journalists and other commentators present a more diffused sense of anxiety and negativity. Even a cursory review of the corpus headlines reveals this slant in terms of the general content, style, and tone of articles (see Appendix). As the one especially pessimistic Montreal Gazette op-ed piece put it, What price technology? {100; see also Extract 48}, the underlying assumption being that a cultural as well as financial price must inevitably be paid for the use of new communication technologies. In short, technology is consistently framed by the print media as subtracting and diminishing rather than adding and enhancing. Voices of enthusiasm from lay and scholarly commentators and even the celebratory claims of big business are ultimately framed by the pervasive narratives of literacy decline and social disorder; in theoretical terms, these function as the complicating actions (Labov, 1972/1999) in many of the articles in the current corpus. In other words, it is these negative issues that apparently motivate most media interest in CMD. This is consistent with news reporters' well-established predilection for negativity regardless of the specific topic under discussion (MacGregor, 2002).⁹

The Disembedding of CMD

It seems from the current survey that media reports about CMD are seldom about CMD per se. Rather, CMD offers itself as the focal point—an idée fixe—for a range of public discourses about other issues, most notably here technology, language, and literacy. In fact, it appears that language and technology are (once again) not only being poorly represented, but also scapegoated for a range of adult anxieties about newness, change, and perceived threats to the status quo. In the process, the caricaturing of instant messaging invariably overstates the difference between online and offline communication, while the fetishization of texting unfairly underestimates the pragmatic subtlety and stylistic diversity of actual practice. Ultimately what this type of misrepresentation does is belie the embeddedness (Howard, 2003) of these technologies in the lives of so many users, especially younger people (cf. McKay, Thurlow, & Toomey Zimmerman, 2005). As a general feature of the print media discourse reviewed here, the exaggeration of the distinctiveness of new media language also functions powerfully to "other" young people by simultaneously exaggerating their differentness; this, in turn, serves to discipline youth and to elevate adulthood (Thurlow, in press).

That mediatized discourse is often anecdotal or empirically unfounded should not be surprising; the procedures and purposes of journalistic writing are understandably different from those of academic writing. Nonetheless, the print media arguably fulfills a more powerful gatekeeping role in shaping public discourse and popular understanding. Newspapers are more typical points of reference and sources of information for lay people than are academic journals. Therefore, just as Baron (2000, quote above) exhorts scholars to reflect on the power of their language theories and practices, the nature of journalistic and other influential lay theorizing about language should not be overlooked or underestimated, particularly insofar as it consistently misrepresents the multiplicity of factors shaping language and the evolutionary nature of language change. With reference to Preston's (1996) objects of folk-linguistic analysis, there appears to be no shortage of information available in the media about CMD; on the contrary, the interplay of language and new technology offers itself as an obviously entertaining and cathartic vehicle for adult commentators.

What is less certain is the degree of accuracy and the specificity of detail offered in media representations of CMD. For this reason, if for no other, future research should pay greater attention to the linguistic and orthographic dimensions of CMC and undertake more situated analyses of actual CMD practice. This might also be done with larger surveys of diverse populations as a means of assessing the validity of claims made in the media, especially as they concern the nature of CMD. It follows, however, that any scholarly research also needs to be accompanied by a more concerted effort to publicize its findings, if rigorous, empirical perspectives are more effectively to challenge the persistent attitude of concern and negativity surrounding the impact of new media on language and social life.

Acknowledgments

I am very grateful to Mary Beth Kaiser and Hazel Lin for their help with collecting and compiling the corpus analyzed here. I am also grateful to the special collaboration of Katrina Barnes and the rest of my COM 482 Computer Mediated Interpersonal Communication class at the University of Washington who, in the Fall quarter of 2005, worked with me in content analyzing some parts of the corpus used in this paper. Last, thanks also go to my colleague Phil Howard for his useful, supportive comments on an earlier version of this paper.

Notes

1. In an article in the journal New Media & Society, Herring (2004a) observes that, given increasing digitization and media convergence, the purview of CMC (as both a label and as field of scholarly research activity) can no longer sensibly be limited to discussions of the Internet and the web. Consistent with this view, CMD is used here as an umbrella term for any situated language practice mediated by a newer or emergent communication technology (see Herring, 2001, 2004b and Schiffrin, 1994 for more on 'discourse' and 'computer-mediated discourse').
2. The articles in the dataset were distributed as follows: 18 from 2001, 10 from 2002, 24 from 2003, 30 from 2004, and 6 from 2005. An additional 13 articles collected previously (from 1999 and 2000) were also included.
3. To avoid the evaluative, sociopolitical connotations of labels like 'language,' 'slang,' 'dialect,' and so on, sociolinguists usually talk about 'varieties' of language in order to respect the linguistic equivalence of different ways of speaking.
4. Working in the area of new technology, communication, and adolescence, I have myself been approached for comment on a number of occasions. Although in one article in the current corpus reference is made to my work (by a colleague elsewhere), I have avoided using articles where I am commenting.
5. Subtitle: 'rolling on the floor laughing my ass off;' Extract 38: 'God save our gracious Queen. God save our noble Queen. God save the Queen' (British national anthem); Extract 44: 'If your mouth was any bigger you wouldn't have any face left to wash.'
6. A more recent example of the mediatized gossip mongering regarding CMC arose in April 2005 following the posting of a Hewlett Packard press release, extracted here:

   Abuse of technology can reduce UK workers' intelligence
   The abuse of 'always-on' technology has led to a nationwide state of 'Info-Mania' where UK workers are literally addicted to checking email and text messages during meetings, in the evening and at weekends. … In a series of tests carried out by Dr Glenn Wilson, Reader in Personality at the Institute of Psychiatry, University of London, an average worker's functioning IQ falls ten points when distracted by ringing telephones and incoming emails. This drop in IQ is more than double the four point drop seen following studies on the impact of smoking marijuana.

   Within a matter of hours and days this already overstated, scientifically questionable press release was picked up by news organizations around the world as evidenced by the following sample of headlines that somewhat predictably sensationalize the IQ or drug links. A notable feature of these headlines is also their stylized (mis)quoting of the original source.
   • 'Infomania' worse than marijuana {BBC News, 22 April}
   • Emails 'pose threat to IQ' {The Guardian, 22 April}
   • E-mails 'hurt IQ more than pot' {CNN, 22 April}
   • Txt and email 'reduce IQ more than cannabis' {Daily Mail, 22 April}
   • Email destroys the mind faster than marijuana {The Register, 22 April}
   • Info-mania dents IQ more than marijuana {New Scientist, 22 April}
   • Low IQ 2day, m8? It's the technology, stupid {The Australian, 23 April}
   • E-mails a big threat to IQ! {Rediff, 23 April}
   • You have mail - and it's making you stupid {South China Morning Post, 24 April}
   • Stoned on E-mail {Information Week, 4 May}
7. A great deal was made of a similar report from the Scottish Qualifications Authority which apparently stated that "text messaging language was used inappropriately" in students' exams; for one article at least this became evidence of just how illiterate Scotland has become {40}. Yet another incident arose in Australia where the cancellation of Ozspell (the national spelling bee) was attributed, at least in part, by one journalist to the slow death of language signaled by the rise of modern communication or CMD {34}.
8. This Guardian newspaper article was one of only three in the corpus that reported the findings of modest, but nonetheless empirical, studies where scholars in the UK {70}, the Netherlands {56}, and the Philippines {46} have disputed the causal linking of poor academic literacies and CMD (i.e., De Vries & Van Der Meij, 2003; Raval, 2002; Rojo-Laurilla, 2005).
9. The same metadiscourse is also to be found in the broadcast media. A recent broadcast of NBC's Today program, for example, featured a special item from journalist Katie Couric about young people's CMD. A recording of this 6-minute item may be downloaded from my website (WMV file 70.6MB) at http://faculty.washington.edu/thurlow/materials/today-clip.wmv.

References

Androutsopoulos, J. K. (2000). Non-standard spellings in media texts: The case of German fanzines. Journal of Sociolinguistics, 4 (4), 514-533.

Androutsopoulos, J. K., & Schmidt, G. (2002). SMS-Kommunikation: Ethnografische Gattungsanalyse am Beispiel einer Kleingruppe [SMS communication: An ethnographic genre analysis of a small group example]. Zeitschrift für Angewandte Linguistik, 36, 49-80. Retrieved March 24, 2006 from http://www.ids-mannheim.de/prag/sprachvariation/tp/tp7/SMS-Kommunikation.pdf

AQA. (2004, June). Report on the Examination: GCSE 2004: English, English Mature and English Literature. Retrieved August 25, 2005 from http://www.aqa.org.uk/qual/pdf/AQA-3701-WRE-Jun04.pdf

Baron, N. (2002). Who sets email style? Prescriptivism, coping strategies, and democratizing communication access. The Information Society, 18 (5), 403-413.

Baron, N. S. (1998). Letters by phone or speech by other means: The linguistics of email. Language and Communication, 18 (2), 133-170.

Baron, N. S. (2000). Alphabet to Email: How Written English Evolved and Where It's Heading. New York: Routledge.

Bauer, L., & Trudgill, P. (1998). Language Myths. London: Penguin Books.

Bell, A. (1991). The Language of News Media. Oxford: Blackwell.

Bourdieu, P. (1991). Language and Symbolic Power. Cambridge: Polity Press.

Cameron, D. (1995). Verbal Hygiene. London: Routledge.

Cameron, D. (2000). Styling the worker: Gender and the commodification of language in the globalized service economy. Journal of Sociolinguistics, 4 (3), 323-347.

Combs, B., & Slovic, P. (1979). Newspaper coverage of causes of death. Journalism Quarterly, 56 (4), 837-843.

Crystal, D. (2001). Language and the Internet. Cambridge: Cambridge University Press.

Danet, B., & Herring, S. (2003). Introduction: The multilingual Internet. Journal of Computer-Mediated Communication, 9 (1). Retrieved March 24, 2006 from http://jcmc.indiana.edu/vol9/issue1/intro.html

De Vries, B., & Van der Meij, H. (2003). Using e-mail to support reflective narration. International Journal of Educational Research, 39 (8), 829-838.

Fairclough, N. (Ed.) (1992). Critical Language Awareness. London: Longman.

Fairclough, N. (1999). Global capitalism and critical awareness of language. Language Awareness, 8 (2), 71-83. Retrieved March 24, 2006 from http://www.multilingual-matters.net/la/008/0071/la0080071.pdf

Fairclough, N. (2003) Analysing Discourse: Textual Analysis for Social Research. London: Routledge.

Foucault, M. (1980). Power/Knowledge: Selected Interviews and Other Writings, 1972-1977 (C. Gordon, Ed.). New York: Pantheon.

Garrett, P., Coupland, N., & Williams, A. (2004). Adolescents' lexical repertoires of peer evaluation: Boring prats and English snobs. In A. Jaworski, N. Coupland, & D. Galasi?ski (Eds.), Metalanguage: Social and Ideological Perspectives (pp. 311-321). Berlin: Mouton de Gruyter.

Giles, H., Coupland, N., & Wiemann, J. M. (1992). 'Talk is cheap' but 'my word is my bond:' Beliefs about talk. In K. Bolton & H. Kwok (Eds.), Sociolinguistics Today: International Perspectives (pp. 218-243). London: Routledge.

Hård af Segerstad, Y. (in press). Language use in Swedish mobile text messaging. In R. Ling & P. Pederson (Eds.), Front-Stage-Back-Stage: Mobile Communication and the Renegotiation of the Social Sphere (pp. 313-333). Dordrecht: Kluwer.

Herring, S. C. (Ed.). (1996). Computer-Mediated Communication: Linguistic, Social, and Cross-Cultural Perspectives. Amsterdam: John Benjamins.

Herring, S. C. (2001). Computer-mediated discourse. In D. Schiffrin, D. Tannen, & H. Hamilton (Eds.), The Handbook of Discourse Analysis (pp. 612-634). Oxford: Blackwell.

Herring, S. C. (2004a). Slouching toward the ordinary: Current trends in computer-mediated communication. New Media & Society, 6 (1), 26-36.

Herring, S. C. (2004b). Computer-mediated discourse analysis: An approach to researching online behavior. In S. A. Barab, R. Kling, & J. H. Gray (Eds.), Designing for Virtual Communities in the Service of Learning (pp. 338-376). New York: Cambridge University Press.

Herring, S. C., Kouper, I., Scheidt, L. A., & Wright, E. L. (2004). Women and children last: The discursive construction of weblogs. In L. J. Gurak, S. Antonijevic, L. Johnson, C. Ratliff, & J. Reyman (Eds.), Into the Blogosphere: Rhetoric, Community, and Culture of Weblogs. Retrieved August 13, 2004 from http://blog.lib.umn.edu/blogosphere/women_and_children.html

Hoenigswald, D. (1966). A proposal for the study of folk-linguistics. In W. Bright (Ed.), Sociolinguistics (pp. 16-26). The Hague: Mouton.

Howard, P. H. (2003). Embedded media: Who we know, what we know, and society online. In P. H. Howard & S. Jones (Eds.), Society Online: The Internet in Context (pp. 1-27). Thousand Oaks, CA: Sage.

Katz, J. E., & Aakhus, M. A. (Eds.). (2002). Perpetual Contact: Mobile Communication, Private Talk, Public Performance. Cambridge: Cambridge University Press.

Kling, R. (1996, February). Hopes and horrors: Technological utopianism and anti-utopianism in narratives of computerization. CMC Magazine. Retrieved April 24, 2006 from http://www.december.com/cmc/mag/1996/feb/kling.html

Koutsogiannis, D., & Mitsikopoulou, B. (2003). Greeklish and Greekness: Trends and discourses of 'glocalness.' Journal of Computer Mediated Communication, 9 (1). Retrieved March 24, 2006 from http://jcmc.indiana.edu/vol9/issue1/kouts_mits.html

Labov, W. (1999). The transformation of experience in narrative. In N. Coupland & A. Jaworski (Eds.), The Discourse Reader (pp. 221-235). London: Routledge. (Original work published 1972)

Lewis, C., & Fabos, B. (2005). Instant messaging, literacies, and social identities. Reading Research Quarterly, 40 (4), 470-501. Retrieved March 24, 2006 from http://www.reading.org/Library/Retrieve.cfm?D=10.1598/RRQ.40.4.5&F=RRQ-40-4-Lewis.pdf

McGregor, G. (1998). Whaddaweknow? Language awareness and non-linguists' accounts of everyday speech activities. Language Awareness, 7 (1), 32-51. Retrieved March 24, 2006 from http://www.multilingual-matters.net/la/007/0032/la0070032.pdf

McGregor, J. (2002). Restating news values: Contemporary criteria for selecting the news. Proceedings of the Australian & New Zealand Communication Association. Retrieved March 24, 2006 from http://www.bond.edu.au/hss/communication/ANZCA/papers/JMcGregorPaper.pdf

McKay, S., Thurlow, C., & Toomey Zimmerman, H. (2005). Wired whizzes or techno slaves? Teens and their emergent communication technologies. In A. Williams & C. Thurlow (Eds.), Talking Adolescence: Perspectives on Communication in the Teenage Years (pp. 185-203). New York: Peter Lang. Retrieved March 24, 2006 from http://faculty.washington.edu/thurlow/papers/McKay,Thurlow,Toomey-Zimmerman(2005)-chapter.pdf

Multhaup, U. (1997). Mental networks, procedural knowledge and foreign language teaching. Language Awareness, 6 (2&3), 75-92.

Preston, D. R. (1996). Whaddayaknow?: The modes of folk linguistic awareness. Language Awareness, 5 (1), 40-74.

Raval, V. (2002). The Negative Impact of SMS Text Messaging on Children's Formal Literacy Skills. Unpublished Bachelor's thesis, City University, London, England.

Rojo-Laurillo, M. (2005). Defining the Filipino Texter and Texting Style: A Sociolinguistic Analysis of Text Messaging in the Philippines. Retrieved April 24, 2006 from http://www.researchsea.com

Ryle, G. (1963). The Concept of Mind. Mitcham, Australia: Penguin.

Schiffrin, D. (1987). Discourse Markers. Cambridge: Cambridge University Press.

Schiffrin, D. (1994). Approaches to Discourse. Oxford: Blackwell.

Shortis, T. (2001). The Language of ICT: Information and Communication Technology. London: Routledge.

Silver, D., & Garland, P. (2003). "sHoP onLiNE!" Advertising female teen cyberculture. In P. H. Howard & S. Jones (Eds.), Society Online: The Internet in Context (pp.157-171). Thousand Oaks, CA: Sage.

Standage, T. (1999). The Victorian Internet: The Remarkable Story of the Telegraph and the Nineteenth Century's On-Line Pioneers. New York: Walker and Company.

Sterne, J. (1999). Thinking the Internet: Cultural studies vs. the millennium. In S. Jones (Ed.), Doing Internet Research (pp. 257-288). Thousand Oaks, CA: Sage.

Tannen, D. (1989). Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse. Cambridge: Cambridge University Press.

Thurlow, C. (2001). The Internet and language. In R. Mesthrie & R. Asher (Eds.), The Concise Encyclopedia of Sociolinguistics. London: Pergamon. Retrieved March 24, 2006 from http://faculty.washington.edu/thurlow/papers/Thurlow(2001)-EofSLX.pdf

Thurlow, C. (2003). Generation Txt? The sociolinguistics of young people's text-messaging. Discourse Analysis Online, 1 (1). Retrieved March 24, 2006 from http://www.shu.ac.uk/daol/articles/v1/n1/a3/thurlow2002003-paper.html

Thurlow, C. (2005). Deconstructing adolescent communication. In A. Williams & C. Thurlow (Eds.), Talking Adolescence: Perspectives on Communication in the Teenage Years (pp. 1-20). New York: Peter Lang. Retrieved March 24, 2006 from http://faculty.washington.edu/thurlow/papers/Thurlow(2005)-chapter.pdf

Thurlow, C. (in press). The technologization of youth in media discourse on language and new communication technologies. S. Johnson & A. Ensslin (Eds.), Language in the Media: Representations, Identities, and Ideologies. Cambridge: Cambridge University Press.

Thurlow, C., Lengel, L., & Tomic, A. (2004). Computer Mediated Communication: Social Interaction and the Internet. London: Sage.

Utz, S. (2000). Social information processing in MUDs: The development of friendships in virtual worlds. Journal of Online Behavior, 1 (1). Retrieved March 24, 2006 from http://www.behavior.net/JOB/v1n1/utz.html

van Dijk, T. A. (1993). Elite Discourse and Racism. Newbury Park, CA: Sage.

Walther, J. B., & Burgoon, J. K. (1992). Relational communication in computer-mediated interaction. Human Communication Research, 19 (1), 50-88.

Walther, J. B., & Parks, M. R. (2002). Cues filtered out, cues filtered in: Computer mediated communication and relationships. In M. L. Knapp & J. A. Daly (Eds.), The Handbook of Interpersonal Communication (pp. 529-563). Thousand Oaks, CA: Sage.

Walther, J. B., Slovacek, C., & Tidwell, L. C. (2001). Is a picture worth a thousand words? Photographic images in long term and short term virtual teams. Communication Research, 28 (1), 105-134.

Warschauer, M., El Said, G. R., & Zohry, A. (2002). Language choice online: Globalization and identity in Egypt. Journal of Computer-Mediated Communication, 7 (4). Retrieved March 24, 2006 from http://jcmc.indiana.edu/vol7/issue4/warschauer.html

Woodward, K. (1999). Statistical panic. differences, 11 (2), 177-203.

Appendix

Complete listing of news articles constituting the full corpus. (Article numbers correspond to extracts used in the article.)

About the Author

Crispin Thurlow is Assistant Professor in the Department of Communication and the Department of Linguistics (adjunct) at the University of Washington. Framed as 'Discourse and Difference,' his research examines the way language and other semiotic modes are used to negotiate boundaries of inequality in everyday interaction. He is particularly interested in sites of difference less commonly investigated in mainstream communication scholarship such as adolescence and young people and tourism and global mobility.
Address: Department of Communication, University of Washington, Box 353740, Seattle, WA 98195, USA