Words: 4102
The students reported lower frequencies for these features in public CMC contexts, while differences by gender were rather insignificant. Second, I attended a comic interpretation of Romeo and Julia recontextualized in the digital era. The plays dialogues were now carried out on facebook walls, with the protagonists’ entries sprinkled with emoticons such as ‘ :-/ laughter acronyms such as rotfl (‘rolling on the floor laughing), expressive punctuation, and the like.
If a discussion of language change and digital media focused on just features of this kind, we could safely assume that a process of change as largely been completed. These anecdotal observations suggest that certain new features of written language are part of the usage of a generation sometimes called the ‘digital natives’, and subject to mediatised stylisation and popular representation. But such a narrow view of language change in digital media is unsatisfactory.
It lacks embedding into a broader picture of sociolinguistic change, which would consider written language in its own right, deconstruct the very notion of ‘language’ into various domains of language practice, and distinguish potential trajectories of change within online written sage, from digital to nan-digital written language, or to spoken usage. Questions and scenarios of this kind circulate in the transnational research literature that has emerged in this area since the mid 1980s.
Its prototypical empirical domain is variably called CMC, computer-mediated discourse or ‘interactive written discourse’ (Ferrara, Brunner and Whittemore 1991). In this paper I also use the term ‘digital ne??orked writing’, a term that emphasises the dialogical and process-oriented character of written language use through technological networks and within social networks (see boyd 2011 All etworked writing is carried out on digital technologies that enable private or public, asynchronous or near-synchronous exchange among individuals and groups on various applications or platforms.
While these technologies enable all sorts of written communication (including carefully drafted, subject, oriented and institutionally framed texts), argue that prototypical networked writing is shaped by four main conditions (Androutsopoulos 2007): (a) it is vernacular, in the sense of non-institutional writing that is located beyond education or professional control; (b) it is interpersonal and relationship ocused rather than subject-oriented; (c) it is unplanned and spontaneous; and (d) it is dialogical and interaction-oriented, carrying expectations of continuous exchange.
These properties set the frame for a prototype of new writing, which first materialised in pre-web applications such as personal emails, newsgroups and chat channels, then carrying on to forums, texting and instant messaging. Written language shaped by these properties captured researchers’ interest and imagination from early on, and virtually all discussion on language change in and through digital media examine etworked writing in the sense outlined here. 2- However, the reach Of CMC has for some time outgrown these conditions, and the relevant literature is full of discrepancies between early and contemporary accounts, visionary scenarios and empirical evidence. An example of the high expectations voiced in early literature is this aphorism by the German linguist Sigurd Wichter from 1 991 : The history of digital networks cannot be written yet, but it is not improbable that these new developments might reach the consequences of the printing press at the beginning of the odern era or of telecommunications technologies in the beginning of the 20th century. 1 Projections of this kind often surface in public discourse, their frequent dystopian versions motivated by ‘a deeper concern: that Internet language is corrupting the way we craft traditional writing or even speak face- to-face’ (Baron 2008: 176). But they have become less common in the ‘Internet linguistics’ literature nowadays, as exemplified by David Crystal’s recent claim: ‘The phenomenon is so recent that we might expect very little to have happened’ (201 1: 57).
This chapter offers a critical synthesis of esearch literature as a backdrop against which to develop a perspective on digital media as sites of sociolinguistic change. 2 1 start by discussing evidence for written-to-spoken and written-to-written effects of CMC language, thereby concluding that findings have been negative, inconclusive, or fairly restricted. Moving to language innovation and change within CMC, three main themes are discussed: the mingling of spoken and written features, strategies of economy, and compensatory means for prosodic and visual cues.
The last part of the chapter outlines a broader perspective on digital media and ociolinguistic change, in which literacy (as a differentiated domain of linguistic practice) and written language (as graphic and visual materiality Of language) feature in their own right. argue that digital media enable an expansion of vernacular writing into new domains of practice, and therefore a diversification of writing styles and pluralisation of written language norms.
The expansion of digital literacy practices affords vernacular written usage more space, visibility and status than ever before, and vernacular usage itself is diversified in what we might call ‘old vernaculars’, representing locally ound ways of speaking that traditionally didnt find their way into (public) writing, and ‘new vernaculars’ ” new patterns of differentiation from written standards, indexing practices and networks of digital culture. In public discourse, however, new media language is discursively constructed as a homogenous and distinct language variety against the backdrop of a technological determinism ideology.
FROM CMC TO WHERE? SCENARIOS OF ‘EFFECTS’ AND ‘INFLUENCE’ Public discourse sometimes raises the effects of digital media on ‘a language’ as a whole (Thurlow 2006, 2007; Squires 2010). But from a research viewpoint, ‘when it comes to speech, the potential effects of the Internet (at least as of now) are negligible at best’ (Baron 2008: 180). The occasional appearance of CMC-typical abbreviations or acronyms, such as LOL, in spoken language is often anecdotally mentioned, in English or other languages.
Apart from that, evidence for effects Of CMC on spoken language are restricted to lexis, an area often neglected by researchers in Internet linguistics. The spread of lexical innovations from the field of information and communication technologies (ICT) in newspaper discourse is well ocumented (e. g. Shortis 2001 , Wichter 1 991 Original: ‘(D)ie Geschichte der Vernetzung kann noch nicht geschrieben werden, aber es ist nicht unwahrscheinlich, dass die neuen Entwicklungen durchaus die Auswirkungen erreichen konnen, die dem Buchdruck zu Beginn der Neuzeit Oder der Fern??bertragungstechnik im Anfang des 20.
Jahrhunderts zukommen. ‘ (Wichter 1 991 : 89, my translation. ). All translations of German excerpts in this chapter are by the author. 2 The extensive use of German-language literature in this chapter reflects the fact that German scholarship addressed relations of digital communication nd language change from early on, and in considerable detail. I integrate it with literature on and in other languages, as my aim is to offer a wider perspective on the vernacularisation of post-standardised (public) written language. 3. In languages other than English, the link be??een technological innovations and Anglicisation was also made early on (e. g. K??niger 1997). In German, for example, English ICT lexis is either morphosyntactically integrated or loan- translated, and variation between these two options may occur. However, these accounts do not specifically distinguish between broader changes and he more specific phenomenon of net neologisms, that is, ‘words that have arisen directly as a result of the Internee (Crystal 201 1: 58).
A methodological challenge here is how to account for the actuation and propagation processes of net neologisms: How can we determine which lexical innovations really emerge in CMC, and what are their paths and trajectories of diffusion across other domains of written usage and modalities of language? David Crystal’s (201 1) approach to net neologisms is to identify areas of technical innovation such as popular platforms and applications and to examine the lexical fields merging in these areas.
He discusses examples of lexical creativity around twitter and blogs, with cautions as to their persistence: ‘Most of these are likely to have a short linguistic life’ (201 1: 59). An alternative procedure for identifying ‘new digital vocabulary is described by SmykBhattacharjee (2006) who studied lexical innovation in blogging. She developed a computeraided analysis comparing blog data with the British National Corpus and the Webster online dictionary, followed by manual verification.
This enabled her to identify new terms coined on blogs, such as blogaholic, which Were neither odified in dictionaries nor attested in large newspaper or spoken language corpora. Such comparisons can help to understand the spread of lexical innovations across domains of written usage. A German example is the productivity of new prefixed and compound verbs around google, such as ergoogeln, a verb roughly meaning ‘to google it out for oneself.
A google search yields 216,000 hits for this item (as per 28 July 2011), but a search in the largest corpus of public written German3 yields only one hit for the infinitive form (set in quotation marks) and seven hits for the participle, ergoogelt. This is a clear, if ough, indicator that a net neologism such as ergoogeln will be around in public net usage for a while before it hits mainstream newspapers. But it does not solve its cross-mode actuation: did this new word first occur in networked typing, or traditional writing, or maybe in talk among net experts?
This question can be raised for each of the numerous net neologisms documented in vernacular lexicography projects such as Urban Dictionary. 4 Strictly speaking, the cross-mode actuation of net neologisms is impossible to determine, unless it is done anecdotally or ethnographically for specific items. On the other hand, the modality of actuation does not predict the cross- media propagation of a net neologism, i. e. the paths and trajectories of its spread across domains of spoken and written usage, and the mediatisation chains that might lead to its eventual codification.
Moreover, we need to consider not only lexis that designates new technologies and applications, but also people’s practices with and negotiations Of digital media. Consider expressions such as facebook stalking (the practice of following someone’s activities on facebook) or the verbs befriend, unfriend, and defriend. The latter two – underlined by my spellchecker as am writing this, but scoring 3,81 0,000 and 550,000 Google hits as per 17 july 2011) – lexicalise a digital literacy practice, whereas befriend is an older form that gains a new mean- 3 See http://www. ids-mannheim. de/cosmas2/.
Ergoogeln is also discussed in the forums of the widely used translation dictionary leo (http://www. leo. org/). 4 The Urban Dictionary features hundreds of word-formation products with google, but not all of these can be expected to be in current usage. For example a search for the word googletowngirl, which is listed in Urban Dictionary as a common noun, produced only a few pages of results, with the word featuring as dictionary entry or user nickname. ing and thus a chance for revival. New lexis of that sort is successful precisely because it lexicalises people’s social practices with digital technology. The influence of CMC on spoken language seems less of a concern to public discourse and popular imagination than its potential effects on other domains of written language production, especially school writing. The idea that pupils might use ‘netspeak’ or text-message style in their school essays is widely publicised linguistic myth on CMC (Thurlovv 2007). Related is the notion that CMC might foster an uninhibited decline of literacy culture (Bei??wenger 2010). Most linguists are very cautious with claims of this sort, but the fact is that robust evidence against them is missing.
There is to my knowledge only one large-scale empirical study specifically comparing digital to non-digital writing. Called ‘How youth write’ (D??rscheid and Wagner 201 0), it was carried out in German-speaking Swiss schools and compared pupils’ school essays to their out-of-school digital writing, based on 1148 digital texts, 53 school essays, and questionnaires to pupils (N=754) and teachers This study draws on a normative conception of salience (Aufflligkeit) as deviation from standard written language norms.
The digital texts are analysed for ‘salient features’ at the levels of punctuation, orthography, morphosyntax, lexicon, and textual organisation, and compared to the school essays. In addition, the ‘writing portfolios’ of nine pupils from different school types are examined in qualitative case studies. The results suggest that out- of-school digital writing does not have any influence on institutional language roduction. Out-of-school digital texts contain some features that do not appear in school essays, but features of networked writing are not transferred to school writing.
Conversely, an orientation to standard language in informal digital writing does not imply normative writing at school. Some of the case studies confirm what would be expected as the default case: normative writing is used at school and ‘deviant’ writing out of school; but other configurations occur too. Young people’s writing is diverse and quite individualised, but ‘interferences’ from informal to institutional writing are not art of the picture. A wider perspective is to ask about the spread of CMC features to other domains of private or public writing.
The use of emoticons in private hand- written texts is sometimes reported, but there certainly are predecessors to such practice, as personal letters were always subject to multimodal enrichment (see e. g. Kataoka 2003). Anecdotally, have seen emoticons and other ‘netspeak’ features used in stylisations of ‘digital youth’ in the press; novels on digital crime using ‘leet speak’ (see below) to decorate their covers; and emoticons finding their way into advertisements, especially in epresentations of young professionals at work.
Such purposeful stylisations of CMC landmarks can be understood as instances of language crossing, with CMC features indexing some (positive or negative, affirmative or distanced) orientation to stereotyped digital-media users and practices, thereby drawing on emerging popular ideologies of new media language. However, what constitutes change here is the availability of new resources for the design of public discourse rather than some new, fixed patterns of non-digital written usage. An even more inclusive approach would centre on the effects of computer- ased writing as opposed to earlier forms of written language production.
Schmitz (2001: 21 70″2171 ) distinguishes four levels at which the computer as a writing-machine changes the nature of writing: (a) monologic (computer- writing enables flexible composition techniques and a ‘less disciplined’ and ‘uninhibited’ writing); (b) dialogic (new writing styles emerging in sites Of public, anonymous participation, a ‘playful anarchy of hybrid, spoken/written patterns); (c) nonlinear (hypertext as new principle of information structure); and (d) interactive (collaborative writing and the fuzzy distinction between uthor and reader).
Clearly, scholarship on language change has concentrated on level (b), to which we now turn. 5 Unfriend was the New Oxford American Dictionary’s word of the year in America, see Savill 2009. Thanks to Sali Tagliamonte for discussion on these verbs. -5- INNOVATION AND CHANGE WITHIN DIGITAL WRITTEN LANGUAGE It seems fair to say that the issues covered so far have often been raised, but rarely systematically studied. What has moved researchers since the mid 1980s was innovation and change in CMC language itself.
Early accounts often proceeded on a ‘butterfly collector’ basis, exploring data from various sources nd often focusing on a single mode, such as e-mail or Internet Relay Chat (IRC). They generally belonged to the ‘first wave’ of CMC linguistics scholarship, focusing on the effect of digital technologies on language (Androutsopoulos 2006; Herring 2003). A key methodological issue in these as well as later studies has been what to compare interactive written discourse with.
The most obvious benchmark, as some researchers have pointed out, would be non-digital vernacular writing, such as private letters or note-taking (Elspa?? 2004; Quasthoff 1997; Ferrara et al. 2001 Others have opted for arge corpora of written or spoken language (Yates 1996; Jucker 2006). However, the mainstream approach has been to draw on frameworks that juxtapose typical features of spoken and written language on situational and linguistic parameters.
While these frameworks differ by language and country,6 they share the analytical foundation of a strong distinction between spoken and written language’ (Squires 2010: 462), leading to a certain idealisation (and implicit normativity) of typical spoken and written language properties, setting a benchmark against which CMC could be conceptualised s a blend or hybrid of written and spoken aspects of language. The main dimensions of innovation in digital written language, as they emerge in research across languages and countries, from early exploratory accounts (e. . Werry 1996) to later textbooks (e. g. Crystal 2006), can be encapsulated in three themes (Androutsopoulos 2007): orality, compensation, and economy. To offer a brief summary: conceptual orality includes all aspects reminiscent of casual spoken language in written discourse. Ulrich Schmitz (2001 : 21 72) coined the term ‘secondary literacy drawing on Walter Ong and Naomi Baron iewed CMC as part of a ‘general tendency for writing to become a transcription of speech’ (1984: 124).
The second theme, the semiotics of compensation, includes any ‘attempt to compensate for the absence of facial expressions or intonation patterns’ (Baron 1 984: 125) by the standardised means of keyboard and typeface. Compensation devices include emoticons, abbreviations that signify various types of laughter, simulations of expressive prosody by iteration of letters and punctuation. The third theme, linguistic economy, includes any strategy of shortening the message form.
This theme is most clearly predicated on technology effects, attributed to the necessity of speed in synchronous exchanges, to financial considerations or to constraints on the size of message. Its counterpart, implicit in the preceding two themes, is the economy of expressiveness, the tendency to contextualize exchanges as informal, engaged and jointly accomplished, drawing on means that often run counter to linguistic economy. These themes are already present in one of the earliest empirical studies in the field, WichteHe observes simplifications, conversational ellipses, representations of colloquial pronunciation, expressive iterations of letters and punctuation signs, and a ‘playful relationship between the phonematic and the graphematic level’. He views mailbox dialogues as ‘a complex meeting of media’ that displays both ‘collaboration and antagonism of orality and literacy, as it is characteristic for phases of media shifts’ (p. 89). A more detailed account of ‘Internet communication and language change’ by Haase et al. 1997) featured a classification of grammatical, lexical and iscourse innovations from Ger 6 In the English-language literature, the categories used by Crystal are based on Chafe, while Bibers framework has also been used. In German and Romance literature, Koch and Oesterreicher’s model of conceptual orality and literacy has been influential (see discussion in Androutsopoulos 2007; Haase 1997; D??rscheid and Wagner 2010). -6- man mailing lists and newsgroups.
Although the authors’ classification of Internet language as ‘group-specific special language of internet users’ is obviously outdated now, their classification illustrates the continuity that xists between early observations and contemporary conceptions of typical internet language. Some of their features directly fit the three themes introduced above. They identify compensatory devices such as emoticons; new means of expressing feelings and affective states, including acronyms such as rotfl and bare verb stems (discussed below); and innovations in punctuation and spelling that serve to ’emulate prosody.
They also identify economy strategies such as a proliferation of clippings and acronyms, and simplifications in punctuation and orthography, such as lack Of noun apitalisation or ‘sloppy punctuation. They further mention spoken-like syntactic constructions such as listbuilding instead of complete sentences and a frequent use of modal particles. Beside these Internet language evergreens, their classification includes phenomena that seem ephemeral and restricted from today’s viewpoint.
They found an ov ergener- alisation of technical and jargon terms, satirical puns on company and sofnna re names, and a so-called ‘P convention’, that is, the transfer of a programming language command, p, to informal networked writing where it is used as interrogative particle. Features like these seem contingent on particular user groups, which at that particular empirical point happened to be among the technology experts that made up a large part of early Internet users. The authors also noted the playful use Of ’emulated whispering, i. . a chat-room command to switch into private chat mode, which was also used in public chat in order to mark a turn as intimate. Such usage again seems characteristic of early Internet users who explored the creative possibilities offered by the reallocation or recontextualisation of particular technology affordances. 7 A third group of features are best described as discourse strategies for new CMC modes and genres. The authors note that new conventions for salutation emerge in newsgroups and chat channels.
They discuss new means of textual cohesion, strategies for quoting and addressing in multi-party environments, and strategies for resolving misunderstandings with deixis, e. g. by means of the acronym, rl ‘real life’. These observations are on new ways of meaning making, creating coherence, and contextualising digitally mediated interaction. They suit an understanding of change that includes genres and iteracy practices. Writers use the resources afforded by a given technology in order to build up and sustain dialogical context, create joint deictic anchoring, and develop appropriate framing.
That said, the boundaries to lexical innovation are fluid, salutations and farewells being a case in point: in some languages, at least, the strongly expected use of salutation and farewell in emails leads both to diversity and innovation in salutation forms and to a heightened awareness of stylistic choices, their appropriateness and their potential for strategic combination in self-presentation and relationship anagement (see Kiesendahl 2011).
Haase et al. (2007) conclude with the insight that innovation and change in CMC entail contradictory tendencies: a loss of morphosyntactic complexity, largely attributed to technical constraints for language production, is counterbalanced by an increase in pragmatic complexity, as writers attempt to contextualize joint production of discourse and manage relationships among spatially (and tempora Ily) distant interlocutors.
More recently, researchers working with larger corpora have pointed out that contrary to popular perception, the frequency Of typical ‘netspeak’ features an be rather low. Tagliamonte and Denis found that abbreviated forms such as nvm ‘nevermind’ in instant messaging are ‘much rarer than the media have led us to believe’ (2008: 12), thereby casting a critical light on media fears of ‘linguistic ruin’. This discrepancy between metadiscourse and empirical evidence is independently confirmed by Squires (2010).
Researchers who compare CMC to earlier vernacular writing ranging from 1 9th century private letters to contemporary popu 7 Another example of this pattern is the use of HTML conventions as a contextualisation cue, or the Twitter hashtag as a marker of a thematic nit outside twitter (see also Crystal 2011: 65). lar culture (Baron 2008; Bergs 2009; Elspa?? 2004; Shortis 2009), conclude that the novelty of digital writing is often exaggerated or lacks historical depth. Moreover, there is a striking lack of systematic micro-diachronic studies within CMC.
While the implicit assumption seems to be that digital language innovations are here to stay, ‘rise and fall’ patterns are just as possible. One recent study pointing to this effect (Henn-Memmesheimer and Eggers 2010) looks at German ‘inflectives’: bare verb stems used without an inflectional morpheme (e. . lach is the stem of lachen ‘to laugh’). Originating in US comics translated into German, inflectives emerged as a feature of youth language in the 1980s, used as exclamations outside the clause structure.
In CMC usage, especially in chat channels, they index affective states and perform ‘virtual’ actions, upon which much playful chat discourse unfolds. Inflectives can be reduplicated or abbreviated, and compound verbs or even verb constructions can be turned into an inflective construction (Schlobinski 2001 Henn- Memmesheimer and Eggers (2010) looked at the ‘career of one popular nflective, grins (verb stem of grinsen ‘grin’). Like other popular inflectives, grins can be clipped to g, which is then again elaborated by iteration, ggg, or typographic mark-up, *g* , and expanded through complements, as in (‘cheekygrin’, p. 9). Based on 24-hour samples from four chat channels and four time slices, from 2002 to 2009, they distinguish three phases in its usage: an early consolidation of chatspecific conventions; then a reorientation toward standard-language usage; and a decline of chat usage. Here, the initial development of a markedly distinctive chat convention is eversed by an orientation to standard norms. This finding seems to echo the sociolinguistic pattern of age grading, in which the linguistic behaviour of young speakers becomes more standardoriented as they grow older.
However, this study lacks an analysis of participant structure and discourse practice in the chat channels. It is therefore not possible to tell whether the decline of inflectives indexes a change of usage by the same writers over time, or a change of activities in the channel, or even a change of participants altogether. Still, the study reminds us of the connection between linguistic hange and discourse that lies at the core of grammaticalisation theory.