2.1Mass media and social media

The term mass media will be understood here as referring mainly to the so-called tertiary and quaternary media (↗0 Preface),162 that is, additional to print media in its different forms and to ‘traditional’ media such as television, radio, cinema and film, on the one hand, and, on the other hand, including electronic media, media often referred to as new, that is, digital media, basically made available through computers or other electronic devices and/or using the Internet (the latter being seen merely as the channel).163 This approach therefore includes social network sites (such as Facebook) and microblogging sites (such as Twitter), communication devices as chat functions and messaging services, other forms of social media, such as YouTube and TripAdvisor, which “also increasingly feature social network capabilities, and thus also give rise to important issues concerning communication and language use” (Seargeant/Tagg 2014a, 3), and other communication media characterized by the usage of electronic devices such as computers or smartphones. According to scholars such as Leppänen et al. (2014), social media can be understood, in a broad definition, as including any digital environment that involves interaction between participants.164 Some functions and media overlap, complicating an analysis according to the medium of communication and devices used: short messages and e-mails can be sent from computers, mobile phones and tablets, for example, and are devices which have distinct medial conditions (key pad characteristics, size of keys, average typing speed, etc.) that can lead to specific usage of written language (cf. Stähli/Dürscheid/Béguelin 2011).

Furthermore, as print media content today is often channeled in parallel through tertiary and quaternary media, it seems especially important to differentiate between medium or mode (essentially for the differentiation between written and spoken language, but which should also include sign language) and channel (via e-mail, by post, etc.) that is often not made in an adequate manner (cf. Sinner 2014, 223).165 Due to the difficulty in clearly distinguishing medium and channel, some authors prefer to speak of communication form as a certain combination of medium, semiotic system, types of signs, and mode of interaction (cf. Bittner 2003, 24), unifying, for instance, all computer-mediated communication forms (cf. Marterer 2006, 9).

2.2Mass media as cause and motor for language change, standardization and diversification: positions

The term language change is usually used with reference to at least four different processes or levels:

Media communication, traditionally, was understood as (and differentiated into) unidirectional communication (one sender and multiple receivers), and bi- or multidirectional media with communication in which participants are, or potentially can be, both senders and receivers (and that allow for more or less feedback of receivers) (cf. Sandbothe 1996; Huberty 2013). This delimitation is constantly becoming fuzzier as channels are becoming more and more intertwined, such as television that allows not only for feedback through other channels, but also for live participation, e. g., via SMS or e-mails, and new media basically based on user participation. Some uses of the newest media devices or services confuse the boundary between interpersonal and broadcast messaging, e. g., the use of short messaging services with interactive television shows (cf. Thurlow/Poff 2013, 164).

Since the beginning of the new century, social network sites on the Internet have become an integral part of modern life worldwide and are considered to be paradigmatic examples of a growing social orientation of online activity; the shift which occurred in the nature of the Internet with the move to the so-called Web 2.0 and its rise of social media has been characterized by growing Internet activity and rapidly increasing user participation (cf. Huberty 2013, 152; Seargeant/Tagg 2014a, 2). Yet, while it is widely accepted that the different online social media are having a profound impact on communication practices, it is disputed whether they actually have an impact on language itself.

Astonishingly, many publications, manuals and textbooks dedicated to media language or on media linguistics, such as Burger (32005), Perrin (2006), Möhn/Roß/Tjarks-Sobhani (2001), Bucher/Straßner (1991), do not even mention the possible influence of mass media on language change. The very contrary can be detected in other publications, e. g., on the new media, where the opinion prevails that the language of the media, for instance, of text messaging, can change, corrupt, and even destroy the language. A search on the Internet for articles concerning language of the new media in any current search machine and in any language results in hundreds of texts with formulations such as “¿Cómo mata la tecnología moderna la escritura y la ortografía?” (‘How does modern technology kill writing and spelling’?), “txt [=text messaging] wrecks the language”, “text messaging ruins the language”, “are text messages destroying our language?”, “sms and IS ruining the language”, “kill the English language”, etc.166 Extreme positions even assert that there is a media language on its own, such as Hjarvard who speaks of medialects: varieties “that arise out of specific media” (2004, 75), or the much disputed, sometimes severely-ridiculed opinion of Crystal (2001) who believes there is a language of the Internet, Netspeak, with different dialects (!) such as e-mail language, chat language, etc., that constitutes a fourth medium to be added to written language, spoken language, and sign language (cf. Sinner 2014, 230). An important aspect is that media is seen as being increasingly present in the language as an infrastructure through which we communicate, and given the fact that since the last century almost all media has been developed in an Anglo-American context, the English language dominates the language specific to certain media (cf. Hjarvard 2004, 92s.).

Although the possibility of a fundamental “multiplying” power of the language used in the media is generally not disputed, there are different positions regarding the extent of influence, i.e., whether there is direct influence, making people actually use certain features of language, or only secondary influence, in the sense that media language consumption only familiarizes the audience with certain uses without actually making it adopt the language it hears or reads. A third, somehow intermediate position only admits the possibility of a restricted impact consisting in the reproduction of single features and catchphrases rather than of more complex features. Furthermore, it is usually admitted by representatives of all positions that it is almost impossible to prove these influences empirically or without leaving certain doubts regarding the impact of other variables.

Some cases mentioned in linguistic studies shall illustrate the position that certain uses are, effectively, propagated through the mass media:

The phrase vivir/estar en un mundo de Bilz y Pap (‘to live/be in a world of Bilz and Pap’, meaning ‘to live in a fantasy world’, ‘to live in a happy world’), resulting from a catch phrase from combined TV ads from the 1980ies for Bilz y Pap, two very popular domestically-produced soft drinks in Chile, even entered Spanish dictionaries (cf. Rivano Fischer 2010, s.v.).

Many cases show that it is often almost impossible to make the assessment that mass media play a role in specific linguistic evolutions without leaving certain reasonable doubts, and that the assumption of a media influence seems to be more a matter of probability.

Chilean school children, for example, are reported to use the noun emparedado instead of pan con queso/carne, etc. (‘bread with cheese/meat, etc.’) or the Anglicisms san(d)wich or sanguchito and alternate torta (‘cake’) with pastel, a word that used to mean only ‘piece of cake’ (cf. Castro et al. 2012). It looks like both features are due to the Spanish used in imported movies dubbed in other Spanish-speaking countries, mostly in Mexican Spanish, because nowhere else are Chilean children constantly exposed to Mexican Spanish than in movies and TV series. As a matter of fact, the described usage can also be observed in children who do not actually watch television or movies and must have acquired these features otherwise. While it is probable that they learn them from their peers who do consume mass media, we have to admit to the impossibility of “blaming” only TV and cinema for having an impact on the children’s language without further studies on individual language acquisition, something much more difficult to do than checking lexical availability. Another similar example is the alleged extension in the use of the gerund (such as fazendo, ‘to be doing’) in European Portuguese instead of the periphrasis estar a + infinitive due to the consumption of Brazilian Portuguese used in imported Brazilian telenovelas (‘serial dramas, soap operas’) screened on Portuguese TV. Both forms exist in the Portuguese language and the shift to gerund could be seen as a “natural” tendency of the language.

Basic problems for determining media impact are, on the one hand, the distinction between media language use as a trigger for certain language uses in other spheres and language acquisition from peers, and on the other, the possibility of the language used in the media being nothing but a reflection of linguistic features or habits already existing in the language. It is impossible to be sure that the increase in the usage of gerund forms in Portugal is not merely reflecting these tendencies, and it could be argued on the basis of simple convergence of European Portuguese with the Brazilian variety; it is equally questionable to exclude the possibility the shift might be caused by the presence of Brazilian immigrants. Thus, mere probability makes the media hypothesis more likely to be true: most Chilean children have access to television, and literally anyone in Portugal has access to TV, while not all the Portuguese people have contact to speakers of Brazilian Portuguese.

The probability approach is widespread in the analysis of mass media impact on language. Many cases of reported media influence on language are the result of mere judgment based on the authors’ intuition and personal perception, recognized by the authors of the studies to differing degrees. Generally, even authors who defend the impact of media on language still qualify their statements, talking of the “strong possibility” or “high probability” of such an influence, and admitting to a great need for further studies. Schneller (2009, 35), who analyzes the language used in German school magazines and student newspapers in print and online format (concentrating on abbreviations, emoticons, reductions, Anglicisms), states that the new media could, at least potentially, have a great impact on the change in adolescents’ written language and that these mass media, due to their technical possibilities, have the ability to spread the typical language of adolescents worldwide and very quickly. Maegaard et al. (2013, 28), in a study on the diffusion of language change in Danish, argue that it is worthwhile to consider the possibility of media being involved in processes of language change. Such prudent, if not vague, formulations are typical for publications on media influence on language.

The difficulty lies in determining cause and effect, because interpreting the appearance as a result of the media impact excludes the possibility that the features are actually to be found in the media due to their presence and availability in the system. According to Gómez-Pablos (1999, 154), the use of chepibe or che pibe in Argentina, a compound of the vocative che (‘hey’, ‘buddy’) and pibe (‘kid’, ‘boy’), came from a TV advertisement from Banco de Galicia in the 1960s; however, how can we be sure the producers of this advert did not use it because they heard it in someone else’s Spanish? Cases where the authors manage to prove that a linguistic trait is clearly traceable to one certain originator without leaving space for doubts are infrequent; one such example is the use of the Catalanism chafardero in European Spanish: While Catalans use it in accordance with the meaning of Catalan xafarder (‘person who spies on someone and/or searches through someone’s things, and who gossips about what s/he found out’), as derived from the Catalan verb xafardejar, speakers in Madrid are only familiar with chafardero as having vague meanings such as ‘incompetent, underachiever, lame duck’, meanings they deduced from the context in which they have read this word, a comic produced in Catalonia and sold all over Spain since the 1940s: El Repórter Tribulete que en todas partes se mete whose main character works in a newspaper called Chafardero indomable (cf. Sinner 2004a).

Isolated words found in the language are often – mistakenly or exaggeratingly – seen as proof of the “mixing” of languages (although the term language mixing itself is seldom explained more than very vaguely). Some authors and sometimes even the mass media themselves simply claim that mass media contribute to the diffusion and even mixing of different varieties, like Schmitz (2004, 31), who asserts that “modern media” contribute to the diffusion and mixing of sociolects, idiolects and other varieties, mentioning examples such as the allegedly German-Turkish “mixed language” called, by some authors, Kanaksprak. It is supposed to be a variety of German influenced by immigrant languages such as Turkish and Arab spoken by some (young) immigrants, and it is said to have influenced, via the media, the language of young Germans (cf. Androutsopoulos 2001; Sinner 2014, 251s.; cf. Fiévet/Podhorná-Polická 2010 and 2011 on the circulation of néologismes identitaires pour les jeunes on the media and the role of the media in the construction of such a generationally claimed argot). The media gladly attend to the alleged “mixing” of languages due to media impact and produce texts on the danger of Anglicisms, “barbarisms”, etc., and in many languages we find puristic columns on language and grammar or heavily sensationalistic stories on “new languages” resulting from media, as, for the example, the varieties used in dubbing or on the Internet (cf. Bagno 532013 on linguistic myths spread by the mass media and Sinner 2007 and 2010 on the coverage of debates on español neutro, ‘neutral Spanish’, or on Spanglish).

In most cases, the reported changes due to mass media only affect single items such as particular words, slogans or catchphrases. This is illustrated by the reported effects of the Brazilian telenovela “Terra Nostra” (‘Our Land’), a serial consisting of 150 episodes about the ups and downs of Italian immigrants in Brazil at the end of the 19th century. The show premiered on Brazilian television at the end of 1999 and had such an impact on the audience that it allegedly sprinkled the language of Brazilians with Italian words (cf. Rivas 2001). No studies have been carried out though on their persistence in Brazilian Portuguese.

As a matter of fact, single words, expressions or constructions are easier to detect than syntactical or morphological changes, especially for non-linguists, and single words or catchphrases can be traced back more easily to a certain radio program, commercial, television personality, movie or songwriter, etc. A good example is the use of the first name Bráulio in Brazil to refer to the male sexual organ as a result of radio and television advertisements broadcast from 1994 onwards featuring a man talking to “his Bráulio” to promote the use of condoms, and, as a consequence, in 1997 Bráulio was said to irritate the unhappy bearers of the name while the advertisement “character” was said to have entered the repertoire of the masses (cf. Sinner 2004a, 98). But a long-lasting presence of “media-born” elements in the language to such an extent seems to be an exception rather than the rule. Chambers (1998) lists examples of different linguistic phenomena reported in US English which originated in TV programs or movies, were used extensively by young people, were picked up by the media who “passed them on” to the public in general, only to decay in frequency after a short while because of being over-used and becoming a fading relic.

Critical positions towards the possible influence of mass media on language consider the dissemination of catchphrases as an exception to the rule, while other elements, especially grammatical features, reportedly do not change as quickly (or even at all) as a result of media influence. Chambers (1998, 125s.) singles out the popularization of words, expressions and catchphrases and denies evidence for media influence on language in other spheres of language:

“If the mass media can popularize words and expressions, the reasoning goes, then presumably they can also spread other kinds of linguistic changes. It comes as a great surprise, then, to discover that there is no evidence for television or the other popular media disseminating or influencing sound changes or grammatical innovations. The evidence against it, to be sure, is indirect. Mostly it consists of a lack of evidence where we would expect to find strong positive effects” (Chambers 1998, 126).

Chambers (1998, 126ss.) gives three main arguments as proof that language change is unlikely to be due to mass media:

Obviously, also in the case of Romance languages such as Spanish and Catalan, local or regional dialects continue to exist despite the exposure of speakers to television, movies and other media.

The assumption that the use of dove instead of the previous standard past tense dived, typical for the northern U.S. and in Canadian English as a result of American influence through the mass media, for example, an observation often repeated in media and literature, has been proven wrong by the author. He especially points out the fact that

“[t]he past tense of the verb dive is not a frequently used word, and so the possibility of Canadians hearing it once in American broadcasts is very slim, let alone hearing it so frequently as to become habituated to it. More important, there is evidence that dove is replacing dived in many other places besides Canada” (Chambers 1998, 128s.).

Yet, in a way, it is precisely the lack of evidence mentioned by Chambers and other authors that also supports the opposite position (i.e., positions that believe in mass media as a motor for change), as many studies do not examine whether the observed changes can actually be seen as such. Only very rarely do authors comment on the duration of the influence and on the durability of the alleged linguistic innovation or change, but only a few cases mentioned in the literature seem to actually last in the respective language.

Other authors seem to have no doubts about the power of the media and strongly believe they have accelerating potential. According to Schmitz (2004, 28s.), mass media accelerate and diversify linguistic changes, even making the pace of changes faster than the succession of generations; any media consumer is subjected to a lifelong language acquisition process (in the case of lexicon, even on a daily basis). The author states that adolescents immediately recognize the linguistic behavior of a twenty or possibly only ten-year-old TV program as outdated. Furthermore, the media are said to contribute to the internationalization or globalization of certain aspects through mass media, as discourse traditions (cf. Cameron 2003, 28), the dissemination of terminology and language for special purposes (cf. Burger 21990, 261ss.), the spread of neologisms (cf. Elsen 2004) and Anglicisms (cf. Kupper 2003), and to the differentiation and rise of koiné phenomena in spelling, phonetics, morphology and other tendencies of language development (cf. Brandt 22000; Zimmermann 1988; Schmitz 2004, 29ss.). The new media in particular are supposed to multiply variants and varieties and at the same time strengthen the effectiveness and scope of norms (cf. Schmitz 2004, 30).

In this sense, the dissemination of standard language (or standard language features) can be promoted by the evolution of printing technology, electronic and digital media. Especially electronic media are said to foster the dialect-standard dynamics through vertical attraction or pressure, even if convergent or divergent evolutions between geographically contiguous dialects (through horizontal dialect leveling) remain significant (cf. Pusch/Kabatek 2009, 1).

According to Schmitz (2004, 31), as a result of mass media production and consumption, the use of any dialect, sociolect, language for specific purposes and group-specific variety can potentially be reinforced in its original usage domain and disseminated beyond this domain, making boundaries between varieties vanish. By doing so, they can contribute to changes of attitude towards a variety (cf. above).

There are examples of structural changes that clearly surpass the lexicon level, seen by authors such as Chambers as the only level where language change can be caused by media use. An example is the innovative function of the suffix -bar in German, due to the success and diffusion of the German word unkaputtbar (literally ‘un+broken+able’, instead of unzerstörbar, ‘indestructible’) as the result of a TV advertising campaign. The word was created for advertising purposes and seems to be particularly memorable because of the (at least initially) strange and remarkable use of the suffix -bar. -bar usually only transforms transitive verbs into adjectives but cannot be added to adjectives (as in unkaputtbar) in order to form adjectives. As a consequence of the “introduction” of the formerly ungrammatical use of the suffix, analogous forms now appear here and there in informal German language, marked by a function of -bar that equals the one it has in unkaputtbar, such as in unplattbar (‘un +flat+ible’), used in an advertisement to promote a bike tire that cannot run flat. While this use is seen, at least by some speakers, to be fresh, innovative and creative, other innovative uses or uses not considered in the standard language are seen as “typical” for the language of mass media, often criticized as corruption, poor language and a threat to the vitality of the language. In many European languages, e. g., the use of Anglicisms, allegedly a phenomenon that the mass media are solely responsible for, is seen as particularly threatening and as a clear motor for the impoverishment of the language.

As Hjarvard (2004, 95) points out, it looks like media, under the influence of globalization, contribute in some ways to the standardization of languages – in his study, Danish – and to linguistic variety and creativity in others. The media spread English (and Anglicisms), often at the expense of other languages, and are, in this sense, a homogenizing factor. At the same time, media are seen as having considerable potential for a new kind of linguistic diversity which transcends other national languages in that innovation is attached to the use of media, not to the individual user, making media a differentiation factor.

Linguistic repetition, and the continuous presence of the written word, are said to strengthen the knowledge of the norm, or to bring the principles of the standard / of the reference norm back to memory (cf. Strubell i Trueta 1982, 17s.). TV viewers and readers, for example, are said to tend to imitate those considered experts or who are idealized or seen as role models for whatever reason, and are therefore often seen as multipliers of language uses (cf. Sinner 2004b and the literature quoted there).

Furthermore, the impact of TV, for example, is said to not only concern individual features of varieties, but also to influence the knowledge, use and status of certain varieties themselves. The knowledge of Standard (High) German by Swiss German-speaking children and their familiarization with the standard even before they enroll at school were attributed to media influence, especially television (cf. Lüdi/Py 1984, 26).167

While mass media are supposed to confirm linguistic models and disseminate standard language features, the same should then hold true for the way norm deviations are established, stabilized, and spread (and eventually turned “normal”) (cf. Sinner 2004a; 2004b). Hjarvard (2004, 81ss., 85ss., 92ss.) relates media usage both with the standardization of Danish and with the presence of Anglicisms in the Danish language (it is interesting to state that most authors seem to immediately consider the use of Anglicisms as the antithesis of a normative, correct language; cf. the debates on Anglicisms in Spanish, Portuguese and French, e. g., in Schmitt 2010 or Helfrich 1993). Apart from disseminating certain features that are consistent with the prescriptive norm in the population, mass media consumption allegedly also confirms non-normative features and makes linguistic models recognizable and authoritative (or not questioned anymore) due to the fact they are being used by authors, journalists or presenters, etc. The prestige of the individuals lends prestige to their language choices. This alleged power of the media is lamented both by professionals and laymen. Boberg (2000) studies the adoption of features of US English in Canadian English (something that is, as seen above, questioned by some authors). He believes that the determining factor for the adoption and diffusion of the US variants via the mass media is the overt prestige of variants seen as correct (cf. Carvalho 2004, 144). Prestige, as a matter of fact, seems to play a very important role in dissemination, something very important in language status planning which includes improvement of prestige of a variety chosen as standard as an important measure for its establishment (cf. chapter 3; cf. Strubell i Trueta 1982, 18 on Catalan).

Taking into account the numerous criteria used by different authors to determine the impact of mass media and media communication and the diverging practice regarding the consideration of extralinguistic variables, it does not come as a surprise that the same media are seen as clearly exercising linguistic influence in one study and as not being involved at all in language change in the next. Saladino (1990) analyzes the diffusion of the Standard Italian norm in rural Southern Italy and its impact on the rural dialect and the factors that lead to the “Italianization” of dialects. While some authors clearly identify television as an important factor in language change, especially the diffusion of normative features, Saladino states television does not exert a big influence. Camps i Giralt/Casals i Martorell (2009), who study the presence of a plural standard – a standard that allows for geographic variation – in Catalan media believe that the production of radio texts following the idea of a plural standard does actually contribute to geolectal leveling.

While Milroy/Milroy (1999, 25) believe that mass media can facilitate awareness about a certain linguistic innovation but can not induce its adoption, Carvalho, in her study on the impact of Brazilian television on the Portuguese spoken in neighboring Uruguay, reveals that “awareness is a precondition for adoption, and without exposure to Brazilian television, both would be unlikely” (Carvalho 2004, 144). The author claims: “[…] a substantial point to be made about the influence of television on people’s speech is that, besides exposure to television, individual motivation to assimilate to a given model is crucial” (Carvalho 2004, 145). One of the important testimonials often referred to is Naro (1981) who stated that the cultural orientation variable and/or the degree of penetration of subjects into higher socio-economic levels around them is the determining factor that allows the viewers of telenovelas to produce variants considered to be part of the standard variety. Naro/Pereira Scherre (1996) later extend the media variable, also taking into account the nature of contact and attitude towards the medium, and show there is a correlation between the use of standard forms and higher integration with the media. The authors’ interpretation of their findings is that the source of the alleged change to the standard is an outside community that the viewers mainly get to “know” via exposure to television. Carvalho (2004, 148) sees this result as countering the belief among sociolinguists that media are not actually able to cause linguistic variation.

Not always do authors highlight the fact that the media-induced language changes they are reporting on actually refer to the use of the language through or on these media, in new genres or text types related and conditioned by these media. Hjarvard (2004) reports on studies such as Hutchby’s (2001), for instance, who, according to him, demonstrates and analyzes

“how a range of media, from the telephone to chatrooms on the net, intrude on and structure the user’s use of language: the taking of turns, linguistic markers of time, space and actors, sentence structure, and so forth” (Hjarvard 2004, 93).

Such characteristics could be seen as changes of the language classified under (c), that is, changes of discourse and text traditions, text type conventions etc., that are clearly conditioned and influenced by the medium; yet, they can involve changes of the system or the norms of a language, such as new lexicon, semantic changes of existing expressions, etc.

The influence that text type-specific norms can exert on the norms of another text type, and which is due to the characteristics of different media, is a well-known possibility, and considered one among many factors of language change (cf. Coulmas 1981, 125). A whole branch of translation studies is dedicated to the description of the impact of translated texts on genres and conventions in the target language and on its whole polysystem.

While investigation has concentrated on the shifts in cultural patterns of living and socialization due to the introduction of telephones, e. g., less attention has been paid to the way people actually behaved on the phone, to the structures of telephone interaction and how they differ from primary patterns of co-present interaction (primary because they precede the phenomenon of phone conversation in temporal terms and outweigh it in terms of global distribution; Hutchby 2001, 6). As Hutchby states, “[p]atterns of talk-in-interaction change as people adapt to developments in the circumstances and possibilities for talk” (2001, 6s.). The same obviously holds true for other new media and their possibilities and restrictions (of space in short messages, for example). The “changes” of patterns actually often seem to be innovations whose characteristics stem from the media conditions they are used in: creations of new patterns for new media, for new text types, etc. Only lately have investigators started to compile larger amounts of text messages (SMS, WhatsApp, etc.) to form corpora in order to be able to describe the language use in these messages, determine characteristic spelling in communication through mobile phones, etc. (cf. Marterer 2006; Stähli/Dürscheid/Béguelin 2011, 6). The scope of most studies of communication via the new media is still the description of the language in these specific domains, as a precondition to then investigate at a later period the possible spread of linguistic characteristics to other spheres of language use. Only then is it feasible to see if and how traces of these uses, typical for these media and their genres or text types, can enter other spheres of language use.

Marterer (2006, 265s.), who compares communication via traditional letters, emails and chat, already states that the linguistic use in e-mails depends much less on the technical conditions of computer-mediated communication than on other factors such as proximity between interlocutors, institutional reference (private, professional, institutional, public), etc.

A look at older communication technology and its modernization also provides insight into what might happen with the effects of the new digital media. The way we talk on the phone traditionally had to do with the formulas implemented at the beginning of the telephone era and that were sometimes even recommended or required by the telephone companies. Therefore, in different languages people develop(ed) and use(d) different habits, e. g., when picking up the phone: saying hello and waiting for the callers to say their names, or saying just their number, etc. Some of these habits – traditions – live on in the mobile phone era. German speakers started using the expression Auf Wiederhören [‘(I hope to) hear you again’] as a greeting formula at the end of a telephone conversation when the telephone was new, as a mere consequence of the fact they could not see their interlocutors, and coined on the existing formula Auf Wiedersehen [‘(I hope to) see you again’]. And Brazilians, even now, answer the phone or mobile phone with the Gallicism alô, while the normal informal greeting is olá. These differences should not necessarily be seen as language change (in comparison to other greeting forms established earlier), but as innovations in certain domains of language use. “Telephone behavior” is even seen as a part of the culture of phone communication in different nations. Both forms, Auf Wiederhören and Alô, can also be used off the phone but would immediately be understood as marked language use, e. g., as joking, or, in the case of Auf Wiederhören, in a semi-offensive context meaning as much as “enough” or “end of discussion”, etc. Auf Wiederhören is also being used on the phone even when new devices actually allow the interlocutors to see each other. The spread of uses from one sphere to another is a secondary shift not caused by the media themselves, but rather due to the will of the speakers and writers to use these linguistic creations and innovations seen as more creative, funnier, younger, more economic, or simply because they spend so much time abbreviating things in text messages they have just got used to the spelling. The use of abbreviations, apparently at first demanded by the need to keep things short in text messages with limited space (cf. above), is not due to such limitations when used “outside” text messaging, but is still recognizable by users as “text message writing”.

Often, laypeople, in heated debates going on in the newspapers, argue for the existence of language change due to text messaging, not seeing that abbreviating is a form of spelling introduced with new techniques and communication possibilities, that it is marked as such and not just a distortion and decay of an existing text type. The use of abbreviations (which is seen as normal in text messaging) in other text types is, therefore, nothing but a mere consequence of the users’ lack of consciousness of text type conventions, of traditions of writing, or their will to not follow these traditions, etc. As with the spread of other habits of language usage, it can also be related to the (lack of) familiarity with certain text types – such as “you do not use certain abbreviations in formal letters”, “you do not use abbreviations in school essays” (or the explicit will to break with such conditions), and the adaptation from characteristics of informal text messaging to written communication in general, etc. What we see is mainly not a change in language in the sense of the modification of existing patterns in existing text traditions, but the creation of new ways of communication accompanied by new ways of using language in written and spoken form.

At the moment, research on communication in the new media basically concentrates on

Much rarer are studies on the impact of new media, the accessibility of new communication media, and studies concentrated on the use of smaller languages focus, mainly, on tendencies of language politics and regulations, translation rules, etc. Using a case study of Irish-language translators, Lenihan (2014) analyzes how and to what extent Facebook gives a certain speech community a voice, and how the issue of language regulation relates to the participatory culture of social media. The author shows that the language policy for the translation of the site to Irish is a result of both top-down (i.e., the company’s regulations) and bottom-up (users’ input) processes. According to the author, this has important consequences for the extent to which social media are driven by a “prosumer” (for producer & consumer) culture and for the users’ capacity to act in the context creation. Language regulations are often not taken into consideration when talking about language use on the new media, forgetting that even online games such as “Forge of Empires”, which allows users to decide which community they want to belong to according to the language chosen as system language, restrict language use by explicitly prohibiting the use of other languages on the respective German, French, Spanish, etc. versions of the online game. Seiler (2013) also analyzes the motivations behind participating in volunteer translation projects for free software, the treatment of normative questions, the efforts to guarantee the internal consistency of the Spanish translations, and social aspects. Like many studies on such issues (e. g., on volunteer translators of Wikipedia content), the study emphasizes the social and ideological function of translated free software, as far as it promotes the diversity, the accessibility and the democratization of knowledge.

A related issue is the impact of state intervention in the media and in language use in the media, e. g., regarding the use of minority languages in the media (cf. Riggins 1992 who identifies five different models of state intervention in minority media), or regulating the language itself, a famous example for the latter being a French “law regarding the use of the French language” from 1994 known as loi Toubon, on the proscription or regulation of the use of foreign languages in certain contexts, such as in advertising (cf. Schmitt 2010 on such laws in Brazil).

While research concentrates on communication patterns and studies of textual types and genres on the net and in the new media, very little can be found on mere linguistic aspects. Time and again we find assertions that media change or can change language, but those claims are constantly not backed up with necessary supporting proof. As an example, we could mention, once again, the volume by Seargeant/Tagg (2014b): while the authors in their introduction explicitly talk about “the language of social media”, the book actually focuses on language use, that is, on the social dimension of language use in social media. None of the articles in the volume actually analyze grammatical or morphological aspects, but concentrate entirely “on the way that social media are offering enhanced means for people to communicate with each other today” (2014a, 17).

Especially in multilingual contexts, studies on new media focus more on language choices than on linguistic structures. This is understandable if we take into consideration findings like those of Siebenhaar (2006), who finds that the “traditional” diglossic medial situation, with Standard German as the written variety and Swiss German as the spoken variety, is not valid in German chat communication in Switzerland anymore, or the conclusion of the Swiss sms4science project whose authors state that 75% of German messages in their corpus were written in Swiss German dialects (cf. Stähli/Dürscheid/Béguelin 2011, 6).

After a long series of selective studies on the characteristics of the language of (or on) the new media, there are now (more or less easily accessible) corpora of growing size that allow for detailed searches. Corpora such as the sms4science corpus (cf. Stähli/Dürscheid/Béguelin 2011, 6), for example, now allow users to find out if earlier findings regarding the characteristics of language in the new media are actually accurate by making it possible to determine the frequency of certain sequences: words, abbreviations, etc. For instance, they can allow users to check if certain spelling forms in text message communication are actually as characteristic as generally stated, or if aspects described as characteristics of certain new media can actually count as universals or are only to be seen as valid for certain languages, etc. Certain features, such as a tendency to use abbreviations, icons, emoticons, etc. were found in SMS communication in the most diverse languages, while many others seem to depend on the language, the given corpus material and setting. Dialect and diglossia play an important role in Swiss SMS communication, and a typical feature of German text messages are inflective constructions, formed with inflectives like hust (‘cough’), seufz (‘sigh’), freu (‘being happy’), as in mich ganz doll freu (‘being very happy’), but neither of them play a role in English or Italian text messaging, for example (cf. the outcome of studies such as Braun 2006; Tagg 2009 or Moretti/Stähli 2011; cf. in this sense Stähli/Dürscheid/Béguelin 2011, 12). The use of hitherto unknown inflectives in Iberoromance language such as Spanish cuelg (‘hang up’) for German aufleg, Spanish lava lava lava (‘wash’) for German wasch wasch wasch or Catalan mam (‘blow’) for German blas can already be found in Spanish and Catalan translations of comics published by the German author Ralf König (cf. König 1997, 23; 1991, 31 and 2008, 36).

An example of aspects repeatedly mentioned as characteristic for short messages is the use of special characters like, e. g., emoticons, spelling relying on rebus techniques such as cu (‘see you’) in English, 9 (both ‘nine’ and ‘new’) in Catalan, or ac (hace, ‘makes’), toma2 (tomados, ‘taken’) in Spanish etc., abbreviations, doubling or repetition of punctuation marks, new or norm-deviating grapheme–phoneme relations such as <x> instead of <ch> in Spanish mxo (mucho, ‘a lot’), probably also used in order to stress the (feigned) orality of the message by trying to represent the depalatalization found in some spoken regional or social varieties). On the stylistic level, among the aspects often presented as characteristic, we could mention the use of colloquial or diatopically-marked forms, ellipses and other means of expression seen as typical for conceptional orality (cf. Koch/Oesterreicher 22011). Still, many of these aspects can also be found in private Internet communication such as chat, which clearly shows their use is not necessarily related to text messaging on mobile phones with limited characters and the (minimum) size of mobile phone keys (cf. Stähli/Dürscheid/Béguelin 2011, 14). Furthermore, Frehner (2008), who compared German, Swiss German and English emails and short messages, showed that the use of abbreviations and other space or time-saving means are also to be found when the writer definitely does not have limitations of space (e. g., when combining several text messages to larger text units). It still has to be analyzed if, besides the aim to economize, there are other factors favoring these characteristic SMS features, such as a certain in-group behavior or the wish to belong to a certain group, the wish to act in conformity with what is believed to be a more prestigious way of communication, as means of creation of a certain identity, etc. (cf. Stähli/Dürscheid/Béguelin 2011, 13s.). For instance, comparisons of SMS from Swiss adolescents with other private daily communication in their corpus inspire Dürscheid/Wagner/Brommer (2010) to believe that spelling deviances such as <sh> instead of <sch> or <x> replacing <s> they found in messages in German were not due to an intent to represent the spoken dialect, but used intentionally to show off knowledge about the rap scene where these alternative spellings are habitual.

Other studies investigate the characteristics of language in the new media from a pragmalinguistic point of view, e. g., analyzing the typical greeting formula in SMS (as in Spagnolli/Gamberini 2007 who analyze short messages in Italian, or Frehner 2008 who dedicates part of her study to greeting formula in German, Swiss German and English used in SMS, MMS, e-mails and, representing “old” means of communication where shortness played a role, telegraphs). Clearly, the majority of studies dedicated to language in the new media analyze aspects described as characteristic for the respective medium beforehand, or try to determine which traces could be seen as characteristic. Apart from such linguistic and pragmalinguistic aspects, studies on new media also look at factors that can be classified more or less precisely as cross-cultural or interactional, such as the analysis of men’s and women’s preferences regarding the use of abbreviations (cf. Bieswanger 2010), or the analysis of SMS exchanges (cf. Stähli/Dürscheid/Béguelin 2011, 13), or can be classified as metalinguistic or ideological, asking, for example, about the impact of SMS communication on standard language or on conventional literacy (cf. Thurlow/Poff 2013, 171; Stähli/Dürscheid/Béguelin 2011, 13). The sensationalist media and puristic authors in particular spread the belief that text messaging influences spelling habits in other spheres of written language, something often demonized as a contribution to language decay. But while only a few scholars believe that the language of the new media, such as text messaging language, actually has a negative impact on language, e. g., a negative influence on the standard language in writing, spelling, and grammar (cf. Siraj/Ullah 2007), most empirical studies that focus on this matter explicitly come to the conclusion that texting does not pose a threat to the standard language itself or to teaching and learning the standard language, and that frequent texting does not have implications for the use of the standard language (cf., e. g., Plester/Wood/Joshi 2009; Dürscheid/Wagner/Brommer 2010; Shafie/Norizul Azida/Nazira 2010; and the overview in Thurlow/Poff 2013 and Stähli/Dürscheid/Béguelin 2011).

Examples given in the literature about the “disastrous” effects of texting for standard language use generally are cases of “texting spelling” – basically, the use of the means of abbreviation mentioned before – or single expressions, often loan words (in most languages Anglicisms or loans from a dominant language, such as Spanish loans in Catalan and Galician, for example). While, as Stähli/Dürscheid/Béguelin (2011, 15) point out, morphological and morpho-syntactic aspects of SMS communication have not yet been studied or have only been looked into marginally, the results of the first detailed studies of such features are remarkable. One of the first such analyses of 400 SMS in Swiss French carried out by Stark (2011) verified that the graphical marking of the subject-verb agreement is closer to standard spelling when co-occurring with lexical subjects (which, however, are infrequent in text messages). Stark interprets this as a hint at their variationist character as a text of communicative immediacy, close to what could be expected in oral discourse (cf. Koch/Oesterreicher 22011). The fact that the marking is intact in more than 90 percent of the cases where it is expected according to the standard leads her to the conclusion that the core syntax is completely unaltered in those otherwise apparently non-standard texts. On these grounds, Stähli/Dürscheid/Béguelin (2011, 15) formulate the hypothesis that there are grammatical structures that are maintained in all cases (such as the marking of subject-verb agreement), while other structures show different and differently-motivated grammatical variations (like the omission of articles and prepositions, negation marking, etc. in French). If their hypotheses are correct, the grammatical structures to be found in SMS texts are a reflection of the language’s own structures that are present even when in other regards the text does not comply with what the standard norms require: linguistic principles not depending on pragmatics.

Among the purposes of SMS corpora mentioned in the literature, time and again we find the opinion that analyzing text messaging will allow analysis of language change. Stähli/Dürscheid/Béguelin, for example, believe their corpus of Swiss SMS will allow researchers “to analyze language contact, linguistic attitudes and language change in the so-called ‘smaller’ languages (Romansh and Italian in Switzerland)” (2011, 14).168 The problem of this view is that the comparison with other text types and genres has only limited value for detecting change, because, as we have already mentioned, new media develop their own conventions, probably based on other traditions of writing or speaking, but the establishment of new text types can only be seen as language change in the sense of (c), “creation of new genres and text types that develop their own traditions”. Corpora of SMS messages will allow linguists to detect changes in a language the same way any other text corpus can do, but will only detect language changes in the system (and norms) (in the sense of (a) given before) in the language of SMS when compared to older or newer texts of the same type or genre. As texting is a rather young writing domain, only longitudinal studies of SMS texts will allow for the determination of changes within the same text type. As the sms4science corpus mentioned by Stähli/Dürscheid/Béguelin (2011) is not yet suitable for such studies, language studied on the grounds of this corpus can only be compared to other sources of written or oral language, thus only enabling researchers to determine the appearance of new conventions in a new medium. Only exceptionally do authors mention the possible lack of comparison of their data with other sources; Cougnon (2010) is one of the few authors to admit to the need of further studies to really use the results of their studies. Analyzing neologisms found in a Belgian French SMS corpus (an outcome of the Faites don de vos SMS à la science campaign), Cougnon explains that her results show that the authors of the analyzed SMS make use of neologisms already attested to in more formal discourse and that they also tend to lexical creation, but concludes, rather unassumingly, questioning what might be the meaning of the use of certain new words by a very small portion of the population in a given moment (their corpus) if it is not compared with other data and results. She proposes to compare her data collected in 2008 with earlier SMS, something that would at least allow for a comparison and determination of actual changes. Cougnon (like many other scholars) does not actually examine SMS in order to prove language change through SMS, but gives examples for already known results of language change she could find in an SMS corpus.

As seen above, much of new media technology comes from English-speaking countries or is developed and marketed first in English, leading to the impression among laypeople and self-declared protectors of the language that the new media especially boost the spread of English loans, something often seen as evidence for ongoing language change or even equated with it. In other contexts, we can state a clear dominance of certain languages that for ideological, political or economic motives possess a dominant position (e. g., the status as official language) over other, smaller, less powerful and therefore minoritized languages. This dominance can also influence the way meta-languages specific for a new medium are formed or develop. A good example is the linguistic situation in Spain, where the regional, co-official languages entered the “cyber world” considerably later than the state language, Spanish. The linguistic minorities literally had to fight for a minimal availability of text and word processors, autocorrect programs, text recognition and predictive text functions, etc. in their languages or suitable for them (cf. Sinner/Wieland 2008; on the presence of Romanian, Picard and Galician in the media ↗14 The Role of Small Languages in the Media I: Presence of Romanian in Medial Communication, ↗15 The Role of Small Languages in the Media II: Presence of Picard in Medial Communication and ↗19 Minority Languages in Media Communication). Since the beginning of the mobile communication era, with the appearance of mobile phones that offered the possibility to send short messages in the 1990s, the meta-language of texting in Spain was Spanish (with obvious English influences due to the origin of the technology), as companies at first only offered the users Spanish-language operating menus. As a result, Spanish (together with its portion of technicisms taken from English) influenced the nascent texting meta-languages in the other languages spoken in Spain (Aranese, Catalan, Galician, etc.) or even hindered the development and spread of a texting meta-language of its own in these languages. The English missed call, for instance, appeared in Spanish as the loan translation llamada perdida; Catalans translated it into their language talking of trucada perduda long before their mobile phones offered the possibility to change the operating language to Catalan, and they even copied the colloquial Spanish reduction of llamada perdida, perdida, as in hacer una perdida (‘make a missed [call]’): fer una perduda.

Only thanks to political pressure on the developers and the legislators, menus in the minority languages were added in the first decade of the new millennium, allowing for the use of these languages on mobile phones, and word recognition functions (such as the T9 predictive text input method) were added in these other languages. These word recognition functions, that (can) make the writing process much quicker, clearly contributed to the use of Catalan in text messaging. Still, unlike in many other languages, Catalans, or at least some Catalans, felt the need for advice on how to write text messages in Catalan. Institutional instances, such as the terminology center TERMCAT, whose mission is to potentialize the diffusion and communication of terminological activities, contributed to the proposal and diffusion of Catalan text messaging terminology, and private commercial enterprises, such as editorial houses, took to publishing guides on text messaging in Catalan (such as Canyelles/Cunill 2004) in order to contribute to a catalanization of text messaging, explaining, for example, Catalan solutions for abbreviations seen as characteristic “tools” used to create text messages. While guides on how to write text messages also appeared in other languages, such as French and German, where they basically exposed the manifold possibilities of creative text messaging in a ludic and entertaining way, the Catalan variants showed a remarkable interest in proposing Catalan solutions for Anglicisms, Hispanisms, for abbreviations based on other languages, particularly Spanish, etc. Maybe for that reason, some Catalan speakers rejected these guides published to contribute to the normalization of their language, regarding them as intrusive and violating their private space, as language planning measures meant to influence personal linguistic behavior in private spheres were seen by them as abusive and politically questionable (cf. Sinner/Wieland 2008). The strong concern for the Catalan language even in these spheres has its roots in the many and powerful movements that exist in Catalan society to normalize their language. Yet, the attempts to normalize the language of text messaging show how the proposal of terminology and the pressure on the industry can contribute to a change in linguistically-adapted technologies that allow for the use of a language in domains where it was not used until then, proving the general capability of influencing the language used in the media “from within”.