1. Introduction
It is an understatement to say that Flanders, the Dutch-speaking northern half of Belgium, has in the last two decades taken an interest in a colloquial and mixed variety of Belgian Dutch. This variety is commonly known as tussentaal, literally ‘interlanguage’ or ‘in between language’ because it cannot be considered either dialect or Standard Dutch but is a mix of the two. To be sure, tussentaal has frequently managed to set this part of the world on discursive fire, its occurrence and spread sparking fierce criticism from linguists, educators, journalists, policy makers of all stripes, literary authors, as well as the general public. Secondary school books teach pupils to disapprove of this “grubby, rotten Dutch.”Footnote 1 Eminent linguists and literary authors categorize it as “lazy Flemish,” “hamburger language,” “soap Flemish,” “crooked language,” or “jabber-Flemish,”Footnote 2 and political party brochures and official linguistic policy documents describe tussentaal as a dangerous force that needs to be cracked down upon to safeguard equal opportunities in education and efficient nationwide communication (for analyses and examples, see Jaspers Reference Jaspers2001, Jaspers & Brisard Reference Jaspers and Brisard2006, De Caluwe Reference De Caluwe2009, Grondelaers & van Hout Reference Grondelaers and van Hout2011, Absillis et al. Reference Absillis, Jaspers and Van Hoof2012, De Schryver Reference De Schryver, Absillis and Van2012, Jaspers & Van Hoof Reference Jaspers2013, Van Hoof, Reference Van Hoofin press). This hostile climate notwithstanding, most Flemish linguists are convinced that the “standard variety of Belgian Dutch … is in any case losing ground” to tussentaal (Grondelaers & van Hout Reference Grondelaers and van Hout2011:221), and many of them are arguing that the dissemination of this colloquial speech style may itself come to represent an autonomous, bottom-up linguistic standardization process (De Caluwe Reference De Caluwe, De Caluwe, Geeraerts, Kroon, Mamadouh, Soetaert, Top and Vallen2002:58, Janssens & Marynissen Reference Janssens and Marynissen2005:196, Vandekerckhove Reference Vandekerckhove2005, Grondelaers & van Hout Reference Grondelaers and van Hout2011:221) or may develop into a regular nation-wide colloquial variety, or umgangssprache (Willemyns 2005:30).
Paradoxically, however, in light of the considerable popular and academic attention it has received, there is ample uncertainty over what tussentaal actually is or what structural features it consists of: “Although tussentaal is immediately recognizable to Belgian listeners, it cannot easily be characterized in terms of necessary and sufficient features” (Grondelaers & van Hout Reference Grondelaers and van Hout2011:222; compare Willemyns Reference Willemyns and Daniëls2003:362). In relation to how this way of speaking has come to be named tussentaal ‘in between language’, some linguists have assumed a tri-polar model that posits tussentaal as a separate, intermediate variety in between the “higher” standard variety and the “lower” basilectal dialects (compare Geeraerts et al. Reference Geeraerts, Grondelaers and Speelman1999, Geeraerts & De Sutter Reference Geeraerts, Sutter and Stroop2003, Taeldeman Reference Taeldeman2008, Cajot Reference Cajot2012).
In this vein, a number of authors have produced descriptions of tussentaal that propose a limited set of usually nonlexical core features (for example, Goossens Reference Belemans and Goossens2000, Plevoets et al. Reference Koen, Speelman, Geeraerts, Elspass, Langer, Scharloth and Vandenbussche2007, Rys & Taeldeman Reference Rys, Taeldeman, Cuvelier, Rymenans, Dominiek and Van Petegem2007, Taeldeman Reference Taeldeman2008), that are argued to “acquir[e] a stable position in this intermediary variety” (Taeldeman Reference Taeldeman2008:26). Others have pointed out that the proposed lists of defining features vary greatly (De Schryver Reference De Schryver, Absillis and Van2012), and that there are few reasons to assume that tussentaal, or any other alternative term for it, necessarily rings any bells with nonlinguists or is recognized by them as a distinct variety they deliberately choose to speak or switch to in interaction (Slembrouck & Van Herreweghe Reference Slembrouck and Van Herreweghe2004: 864, De Caluwe Reference De Caluwe2009:16, Lybaert Reference Chloé2011).
In addition, some empirical analyses indicate that speakers do not switch neatly from tussentaal to Standard Dutch or dialect, but often produce unpredictable, eclectic, “messy” code mixes that range from being more dialectal to fairly standard, complicating the unambiguous classification of utterances as dialect, tussentaal, or Standard Dutch (compare Vandekerckhove & Nobels Reference Vandekerckhove and Nobels2010).Footnote 3 As a result, some authors have argued that tussentaal needs to be seen as part of a diaglossic, continuum without any discrete intermediate strata (Plevoets Reference Plevoets2008, De Caluwe & Van Renterghem Reference De Caluwe and Van Renterghem2011, Ghyselen Reference Ghyselen2011, Grondelaers & van Hout Reference Grondelaers and van Hout2011; also see Auer Reference Auer, Delbecque, van der Auwera and Geeraerts2005) or as “een matrix van taalgebruik, waarvan de dimensies welbekend zijn, maar de gerealiseerde taalpraktijk veel minder voorspelbaar” (Slembrouck & Van Herreweghe Reference Slembrouck and Van Herreweghe2004:862–863; see also Jaspers Reference Jaspers2001, Vandekerckhove & Nobels Reference Vandekerckhove and Nobels2010).Footnote 4
Furthermore, several authors have formulated hypotheses for tussentaal's success, often explaining the occurrence of this way of speaking as oppositional to Standard Dutch. Thus, Van Gijsel et al. (Reference Van, Sophie and Geeraerts2008:220) propose that, for adolescents especially, tussentaal “has young, even somewhat rebellious, connotations, as opposed to the ‘conformist’ norm of the standard language,” or that many Flemings consider tussentaal more “authentic” or “natural” as opposed to an external, namely, standard norm. This is seen to offer a functional motivation for its use in a society where Flemings increasingly view Standard Dutch as an “artificial” or inauthentic variety imposed on them from above.
Plevoets (Reference Plevoets, Vogeleer, Brisard, DeBrabanter, Dendale and LeBruyn2009, Reference Plevoets, Absillis, Jaspers and Van Hoof2012) postulates that tussentaal is a badge of the Flemish elite who have maximally profited from the region's socioeconomic welfare, which has increased considerably after the Second World War. Having reached the top positions in Flemish society, this elite can now resist the pressure of acquiring prestigious linguistic forms, or display indifference to them. In Bourdieu's (Reference Bourdieu1991) words, they can afford to be “hypocorrect,” that is, conspicuously relaxed or intentionally less than correct—similar to the practice Bourdieu observed among the French bourgeois seeking to distinguish themselves from the petitbourgeois, who, in their efforts to imitate them, were becoming hypercorrect, tense, and linguistically anxious (Reference Bourdieu1991:63).
Grondelaers & van Hout (Reference Grondelaers and van Hout2011), in turn, suggest that the triumph of tussentaal may have to be attributed to, among other things, i) an increased Flemish self-consciousness in the wake of the region's augmented political and economic power within the Belgian state, which dimmed erstwhile language-integrationist ambitions and raised ambitions for a separate Flemish standard (also see Goossens Reference Belemans and Goossens2000), ii) an increased democratization of social life that facilitates vernacular language use in public, and iii) an antiauthoritarian, post-1968 grudge against what is, essentially, a foreign and repressively installed linguistic standard. In addition, Grondelaers & van Hout (Reference Grondelaers and van Hout2011) put forward that the “highest” spoken variety in Flemish Belgium is so-called VRT Dutch, the language spoken by newscasters and most presenters on the Flemish public broadcaster VRT, but that this variety is hardly ever spoken outside of broadcasting contexts (see also De Caluwe Reference De Caluwe2009).Footnote 5
Grondelaers et al. (Reference Grondelaers and van Hout2011) in this time frame conducted an evaluation experiment showing that speakers of Belgian Dutch consider none of the other (regionally colored) varieties currently in use as “the best language.” They concluded that whereas tussentaal may be on its way to becoming the new standard of Dutch in Flanders, “Belgian Dutch is currently a standardless variety” (Grondelaers et al. Reference Grondelaers and van Hout2011:200), and consequently “the Belgian standard situation clearly represents a case of Fairclough's (Reference Fairclough1992) destandardization” (2011:218; see also Willemyns Reference Willemyns, Fandrych and Salverda2007, van der Horst Reference Van, Sophie and Geeraerts2008).Footnote 6 In a more recent speaker evaluation experiment reported in Grondelaers & Speelman Reference Grondelaers and Kristiansen.2013, lexical tussentaal features scored high on “dynamism.” Based on this finding, the authors argued “that by using typically Flemish lexemes, speakers project a trendy, assertive image” (2013:184), and hypothesized that these prestige values could be the driving force behind the emergent standardization of tussentaal.
In this paper, we would like to argue that even though tussentaal may, in some contexts, index anti-authoritarian resentment, rebelliousness, assertiveness, or trendiness, these meanings cannot be taken to hold across the board. Drawing on interactional data from a telecinematic corpus (Van Hoof Reference van. Bezooijen, De Caluwe, Van and De2013), we demonstrate that what from the structural point of view can be called mixed Dutch (conventionally, tussentaal) can help index a very diverse range of meanings, including those that are traditionally associated with Standard Dutch or dialect, depending on the local linguistic ecology (Haugen Reference Haugen1972). This finding confounds any suggestion that the use of tussentaal essentially signals rebellion, authenticity, conspicuous relaxation, nonformality, or self-consciousness, and necessitates an interactional analysis of how linguistic features are interpreted by interactants in specific situations. It also complicates analyses that automatically or necessarily associate the use of tussentaal with speakers' negative attitudes toward dialects and/or Standard Dutch (compare Taeldeman Reference Taeldeman2008:29 and others), and with any intention to produce a mixed type of Dutch.
In sum, and following Coupland (Reference Coupland, Junod and Maillat2010:60), we demonstrate that in principle, “we cannot identify what a standard or non-standard variant is, independently of social judgments that are made about its use or users,” and this implies that what looks like tussentaal in etic terms may mean something quite different when examined from an emic perspective. Regardless of its stabilizing or unpredictable nature, we argue therefore that the occurrence of tussentaal cannot simply be taken as an unambiguous sign of destandardization processes.
2. Context: Tussentaal in Flemish Public Discourse
The corpus we draw on is part of a larger study that investigates the evolution of language use in Flemish telecinematic comedy and drama between the early 1980s and the late 2000s (Van Hoof, in press). TV series and films are frequently targeted in the criticism of tussentaal, because they are seen as uncritically promoting it and as having drifted away from the earlier days in which actors still used “proper” Standard Dutch (for an overview of these criticisms, see Van Hoof, in press:12ff). Even if these assertions may be difficult to prove empirically, they are typical for the context in which these TV series and films are produced, and it is worth briefly sketching it before going into an analysis of the data.
One crucially distinctive characteristic of linguistic standardization in Flanders is that stakeholders in the 19th century decided not to develop a Standard Flemish but to adopt the linguistic standard in use in the Netherlands, “Standard Dutch.” In brief, three reasons were behind this decision. First, an endogenous Standard Flemish was not trusted to hold its own next to French in Belgium. Second, Flemish was considered to be too corrupted by French and thus all too symptomatic of French domination. Third, a choice for Standard Dutch would allow Flemings to “set history right” in that it would imply linguistically recreating a unity that the Eighty Years’ War had destroyed through splitting the Low Countries in two (Deprez Reference Deprez1999, Willemyns Reference Willemyns1996, Reference Willemyns2013). Consequently, Flemings needed to be made familiar with an imported linguistic standard, and this gave rise to a blooming tradition of linguistic purification. From the 1950s onwards, this project turned into a large-scale, fiercely propagandistic, and extensively mediatized linguistic standardization campaign that has successfully “enregistered” (Agha Reference Agha2007) Standard Dutch and its linguistic opposites in all corners of Flemish society—a process we have labeled “hyperstandardization” (Van Hoof & Jaspers Reference Van Hoof2012, Jaspers & Van Hoof Reference Jaspers2013).
The Flemish public broadcaster VRT was a key player in this enregisterment process. Its directors considered language a cornerstone of their at the time explicitly civilizational broadcasting mission, and engaged linguists to ensure that all microphone employees spoke an impeccable Standard Dutch. Moreover, for decades the VRT offered prime time slots to Flanders’ linguistic purifiers (Vandenbussche Reference Vandenbussche, Gilles, Scharloth and Zieger2010). The most well-known example is the instructive television program Hier spreekt men Nederlands ‘Here one speaks Dutch’, broadcast three days a week in the 1960s and 1970s.Footnote 7 It focused on speech purification (weeding out dialectal features in civilized speech) and speech typification, linking Standard Dutch and dialect to a set of social personae and stances (Agha Reference Agha2003:237). In particular, Standard Dutch was propagated as the hallmark of a civilized, gentleman-like, stylish, modern, polite speaker, while Flemish dialects were associated with the opposite of these qualities (see, for example, Florquin Reference Florquin1965). To this date, the hope has always been that Standard Dutch would not just remain a variety for special, formal occasions but would eventually anchor itself firmly in Flemings' daily lives as an “informal Standard Dutch” (see, for example, Hendrickx Reference Hendrickx1998).
These standardization efforts notwithstanding, the VRT has recently found itself the target of more and more complaints with regard to the language of its broadcasts. Journalists, teachers, writers, and linguists often protest that language on television is rapidly deteriorating. Especially the “lighter” genres, such as talk shows, comedy shows, or soap operas, regularly find themselves in the line of fire. During the question period at the Flemish parliamentary meeting on November 29, 2012, one MP bemoaned that “up until forty years ago VRT series in tussentaal would have been absolutely impossible” (Vlaams Parlement Reference Vlaams2012), and subsequently complained to the Minister of Culture that Flemish government did far too little to oppose the advance of so much linguistic indolence.Footnote 8 The highly popular nationalist and conservative politician Bart De Wever, who is himself often accused of speaking tussentaal, alluded to deteriorating linguistic standards as a proxy for falling norms in society:Footnote 9
Nivelleren naar boven werd ingeruild voor nivelleren naar beneden, met een enthousiasme alsof diarree de ideale remedie zou zijn voor constipatie. De gevolgen zie je het best aan ons taalgebruik. Het streven naar het doorsijpelen van algemeen Nederlands werd zo goed als opgegeven. Zelfs de openbare omroep vindt het vandaag normaal om populaire series te maken waarin de acteurs een soort tussentaal spreken. … Het resultaat is een veralgemening van een soort rand-stedelijk Antwerps, gelardeerd met exotische klanken en uitdrukkingen uit andere provincies.Footnote 10 (De Wever Reference De Wever2007)
Leveling up has been turned in for leveling down, with an enthusiasm as if diarrhea were the ideal remedy for constipation. Our language use best shows what the consequences are. Attempts to disseminate Standard Dutch in society have been all but relinquished. Even the public broadcaster today considers it normal to make popular series where actors speak some kind of tussentaal. … The result is a generalization of some sort of suburban Antwerp dialect, larded with exotic sounds and expressions from other provinces.
Journalist and historian Marc Reynebeau summarized the whole issue as follows:
[E]r is amper nog AN te horen, niet op school en niet in de massamedia. Behalve in journaals en andere ernstige programma's hanteert zelfs de openbare omroep steeds minder het AN, zeker niet in talk-shows of fictieprogramma's. Mensen moeten zich in fictie kunnen herkennen, heet het dan, en het klopt dat dat niet kon met het bombastische theater-ABN van vroeger. (Reynebeau Reference Reynebeau2006)
[W]e hardly hear Standard Dutch anymore, not at school and not in the mass media. Except for the news and other serious programs, even the public broadcaster is using Standard Dutch less and less, and certainly not in talk shows or TV fiction. People are supposed to recognize them-selves in fiction, so it goes, and it is true that that was impossible with the bombastic theatre Standard Dutch of before.
Quotes such as these are representative of the widespread opinion that before, the language of comedy and drama series on the public broadcaster was Standard (even at times somewhat “bombastic”) Dutch, whereas contemporary series (and films) have all fallen prey to tussentaal. The turning point is usually situated around 1989, the year when the first commercial TV station VTM started broadcasting in Flanders. Janssens & Marynissen (Reference Janssens and Marynissen2005:198) sum up the consequences of the end of the VRT's broadcasting monopoly as follows:Footnote 11
Om hoge kijkcijfers te halen en adverteerders te kunnen aantrekken, programmeerde VTM niet alleen recente speelfilms maar ook veel populaire soaps, (vragen)spelletjes en andere ontspanningsprogram-ma's. De meeste acteurs in deze soaps en vele deelnemers aan de spelletjes en amusementsprogramma's spraken de taalvariëteit die in Vlaanderen gebruikelijk was: tussentaal. … En ook op de [V]RT deed deze variëteit geleidelijk haar intrede in soaps en dergelijke, en op de radio in reclameboodschappen. (emphasis is ours–JJ&SVH)
In order to attain high viewer ratings and advertisers, VTM not only scheduled recent films but also many popular soaps, quiz games, and other entertainment. Most actors in these soaps and the many participants in these quizzes and entertainment shows spoke the common linguistic variety in Flanders: tussentaal. … Gradually this variety also made its entry on the VRT in soaps and other programs, and in radio commercials.
It is no coincidence in this context that “soap Flemish” (Geeraerts Reference Geeraerts1998) has become a popular label for tussentaal. The term refers to the genre of the soap opera, introduced in Flanders by the commercial broadcaster and later also adopted by the VRT. This type of low-cost, quickly produced TV fiction is often frowned upon as low-quality television, a connotation that is easily extended to the language it uses.
3. Tussentaal in Flemish Telecinematic Discourse
That the above analyses require some revision is demonstrated by Van Hoof & Vandekerckhove's (Reference van. Bezooijen, De Caluwe, Van and De2013) quantitative analysis of language use in several TV fiction series broadcast by the VRT around 1980 (that is, before the arrival of commercial television) and in the late 2000s.Footnote 12 In focusing on TV fiction series, Van Hoof & Vandekerckhove align themselves with the burgeoning interest in so-called telecinematic discourse (see, for example, Queen Reference Queen2004, Richardson Reference Richardson2010, Piazza et al. Reference Piazza, Bednarek and Rossi2011, Androutsopoulos Reference Androutsopoulos2012). This interest is driven by the realization that as all language use is essentially constructed and performed, excluding the language of fictional characters from sociolinguistic analysis as “inauthentic” or “unreal” means overlooking an important part of daily language use. It would also mean disregarding that authenticity or “real-ness” is a product of social interaction, and is “always achieved rather than given in social life, although this achievement is often rendered invisible” (Bucholtz Reference Bucholtz2003:407–408).
Under this approach, fictional interaction can be investigated as social interaction in its own right, with no presumption that the social meanings of linguistic features in the nonfictional world would simply be reproduced or can be taken for granted in the fictional world. Indeed, research in this tradition often works from the assumption that linguistic resources and their social meanings are recontextualized in telecinematic discourse, depending on the dramatic conditions on hand, the relations between characters, and the development of the narrative (compare Coupland 2007). In other words, linguistic features contribute to the construction of diegetic worlds, that is, the story worlds of films or TV series (Elsaesser & Hagener Reference Elsaesser and Hagener2010:5), and are as such to be interpreted within these diegetic ecologies.
Telecinematic discourse has proven to be particularly apt for exploring how linguistic variation is associated with certain social stances, activities and identities, or is “enregistered” (Agha Reference Agha2003), and for investigating how such metapragmatic work reproduces, contests or temporarily inflects the larger-scale, real-world metadiscursive regimes in which the fictional product is embedded. While most analyses in this tradition are of a qualitative nature, Van Hoof & Vandekerckhove (Reference van. Bezooijen, De Caluwe, Van and De2013) produced a quantitative analysis in order to obtain a broad appreciation of linguistic evolutions in the abovementioned corpus, and to test the popular claim that language use in TV fiction has shifted from predominantly Standard Dutch to mostly tussentaal.
To this end, Van Hoof and Vandekerckhove selected seven morpho-syntactic variables, each with one or several standard variants and one or several nonstandard variants considered in the literature as typical markers of tussentaal (see Goossens Reference Belemans and Goossens2000, Plevoets et al. Reference Koen, Speelman, Geeraerts, Elspass, Langer, Scharloth and Vandenbussche2007, Rys & Taeldeman Reference Rys, Taeldeman, Cuvelier, Rymenans, Dominiek and Van Petegem2007, Taeldeman Reference Taeldeman2008), and compared their occurrence in the series aired around 1980 and those broadcast in the late 2000s. The seven variables they selected are listed in table 1 (below).
The nonstandard variants, however, cannot be taken as exclusive markers of tussentaal. Without going into too much detail, typical for the speech style that linguists term tussentaal is the frequent use of so-called tertiary, that is, general, or geographically widespread dialect features (Taeldeman Reference Taeldeman and Hinskens2006, Rys & Taeldeman Reference Rys, Taeldeman, Cuvelier, Rymenans, Dominiek and Van Petegem2007), such as the nonstandard variants listed above, and rather limited use of secondary and primary dialect features, which are more strongly regionally or locally colored and can be considered markers of (more) fully-fledged dialect use. Thus, the use of a feature such as double negation in itself may mark both tussentaal and dialect use. Van Hoof and Vandekerckhove (Reference van. Bezooijen, De Caluwe, Van and De2013) therefore also included in their analysis two dialect markers that are typically not transferred to tussentaal (Rys & Taeldeman Reference Rys, Taeldeman, Cuvelier, Rymenans, Dominiek and Van Petegem2007:5) and thus allowed to filter out dialect use. Each marker had a standard variant and several dialect variants. These markers are displayed in table 2.
Van Hoof & Vandekerckhove (Reference van. Bezooijen, De Caluwe, Van and De2013), using mixed-effects logistic regression, assessed the effect of period (1980 versus present) on language variety (standard versus nonstandard), including estimates of the speakers' dialect use as a covariate. Apart from the fixed effects of period and dialect use, the model contained random effects of series, speaker, and marker. Van Hoof & Vandekerckhove found that period had a significant positive effect on language variety, and that exclusion of the covariate dialect use severely weakened the effect of period, due to the higher occurrence of dialect in TV fiction around 1980.
Table 1. Markers of tussentaal.Footnote 13
Table 2. Markers of dialect use.
Rather than evidencing a straightforward decrease in the use of Standard Dutch in favor of tussentaal, these findings indicate that non-standard language use has been an integral part of Flemish TV fiction for at least 30 years. It would be thus difficult to associate the increase in the use of tussentaal with the arrival of commercial television in Flanders, since this type of language use was already used on television long before “changes in the Flemish media landscape” (Janssens & Marynissen Reference Janssens and Marynissen2005:189) took effect.
That said, Van Hoof & Vandekerckhove Reference van. Bezooijen, De Caluwe, Van and De2013 demonstrate that language use in TV series in the 1980s covered a much wider spectrum than it does today. Apart from tussentaal, Standard Dutch and dialect were also used regularly in early 1980s fiction. Today, however, the use of tussentaal dominates the genre and offers its services to a much wider range of characters and situations than in the 1980s corpus, at the expense of Standard Dutch and dialect. In other words, language use in Flemish fiction series has evolved in two directions at the same time, namely, away from Standard Dutch and away from the Flemish dialects.
4. An Interactional Approach
Van Hoof & Vandekerckhove's (Reference van. Bezooijen, De Caluwe, Van and De2013) analysis provides a general characterization of linguistic variation in Flemish TV series, and of its evolution in 30 years’ time. It shows to what extent characters and series make use of linguistic features that are conventionally labeled as indicative of varieties such as Standard Dutch, dialect, or tussentaal. In covering the entire spectrum between Standard Dutch and dialect, language use in a fictional setting attests to the diaglossic nature of the linguistic situation in Flanders (even if contemporary series appear to cover a shorter continuum than the 1980s series).
At the same time, the analysis also shows that certain linguistic features in effect do tend to conglomerate, which explains why tussentaal is often viewed as a distinct, bounded variety: Whereas eclectic mixes are, in principle, quite possible (for example, dialectal realizations of the dialect markers in table 1 could be combined with standard realizations of the tussentaal markers in table 1), such mixes are rarely produced by the characters in the corpora. The mixed, intermediate way of speaking that Flemish linguists have labeled tussentaal tends to come in fairly predictable guises (certain combinations are frequent, others fairly exceptional).
Nevertheless, although it can be applied to a large amount of data to provide a panoramic view, the quantitative analysis described above has a number of shortcomings. First, it uncovers structural tendencies, but it does not explain why the language of fiction has taken the form that it has over the last 30 years. An answer to that question has to be sought in the social meanings carried by linguistic forms in TV series of the 1980s and today. Incorporating more independent variables into the analysis may provide a starting point for identifying those meanings. Investigating the possible effect of the variable genre, for example, might reveal that comedy and drama favor different kinds of language use (presumably the nonstandard kind in the case of the former; compare Androutsopoulos Reference Androutsopoulos2012:146). Series aiming to be realistic would probably display patterns of social stratification, where characters with a higher socioeconomic status score higher on the use of standard variants than lower class characters (compare Geeraerts Reference Geeraerts, Lehr, Kammerer, Konerding, Storter, Thimm and Wilski2001). Thus, the variable social class may also contribute to a more detailed quantitative analysis. But, and second, even the most finegrained quantitative investigation of language use has the shortcoming of attributing an absolute—that is, context-independent—meaning to the variants earmarked as symptoms of tussentaal, dialect, or Standard Dutch. These variants are not allowed to jump ship, and are assumed to have the same meaning across all contexts under consideration.
In contrast, interactional approaches have over the years pointed out that the meaning of linguistic features—and of the sociolinguistic activities they help construct—is relative to the unfolding interaction in a specific context and can be subject to significant reinterpretation when features are transferred across contexts:
[T]he same semiotic acts or processes [can] mean very different things in different environments. Shifts across environments involv[e] shifts in function, the attribution of different meanings and values to the same semiotic action, object or process. (Blommaert Reference Blommaert2005:172).
Coupland (Reference Coupland, Junod and Maillat2010:62) likewise argues that
… judgement/attitudes research has always shown that the social meanings of linguistic varieties are complex and multidimensional … [and that] contextual factors impinge crucially on which social meanings are attributed to varieties.Footnote 14
In other words, the same verbal performances can be intended or interpreted differently, depending on how speakers are perceived (for example, as a conventional dialect or Standard Dutch speaker, as a learner of Dutch), on what relationship speakers have with one another (for example, hierarchical or sociable), on the particular social environment or regime of linguistic expectations they find themselves in (for example, at home or at school), and on how the performance is keyed (jocular or serious). Frequency counts, however, classify inauthentic, caricatured language use as similar to conventional, nonironic language use.
In sum, while it is clear that a quantitative approach has helped to reveal structural tendencies in our corpus that a qualitative analysis would be at pains to provide, such an approach necessarily abstracts away from the specific situations in which various features appear. Consequently, a quantitative approach often fails to determine how particular linguistic features are interpreted in particular situations of use, and ignores whether such features can actually be seen as true representatives of the varieties that the quantitative analysis has already decided that they are. As a result, there is a risk of reading meanings into the use of features that are not warranted by how they are recruited in dialogue by specific characters, and of overlooking how the insertion of these features speaks to the social environment or linguistic ecology (Haugen Reference Haugen1972, Mühlhäusler Reference Mühlhäusler1996) that the series and the characters (re)create.
These considerations do not invalidate the use of quantitative approaches per se. But on the basis of the literature reviewed above, any use of tussentaal signals a rebellious, authentic, conspicuously relaxed, nonformal, or self-conscious interactional stance. In the next section, we discuss the interactional account of two scenes analyzed at length in Van Hoof, in press. Given this account, we suggest that what many Flemish linguists would characterize, based on its structural properties, as an instance of tussentaal “contribute[s] to the construction of different identities, depending on how they are locally contextualized” (Coupland Reference Coupland2009:288), and flags a range of other meanings than what tussentaal has hitherto been allowed to index. If occurrences of tussentaal cannot be trusted to stand for its traditionally attributed meanings, tussentaal research in general (quantitative and qualitative) may have to reconsider some of its assumptions.Footnote 15
5. Tussentaal Functioning as Standard Dutch
Our first example comes from the recent film Groenten uit Balen ‘Vegetables from Balen’ (2011), a screen adaptation of Walter Van den Broeck's play of the same title.Footnote 16 The movie depicts the strike that broke out in the Vieille Montagne zinc factory in Balen-Wezel (some 70 kilometers east of Antwerp, Belgium) in 1971. For this film, director Frank Van Mechelen cast mostly actors originating from the Campine area and asked them to speak in their native dialect, accounting for this choice as a way to achieve realism. According to Van Mechelen, the film was to paint an authentic picture of the working class of a few decades ago:
[De film] speelt zich om te beginnen al in de jaren 70 af, waar dat er nog veel meer dialect gesproken werd dan nu … zeker in die sociale klasse, dus ik heb daar [aan de keuze voor dialect] nooit aan getwijfeld. da moest gewoon, da kon nie anders.Footnote 17
[F]irst of all it [that is, the film] takes place in the 1970s, when dialect was still spoken much more often than it is today … especially in that social class, so I never doubted that [that is, the choice for dialect]. It just had to be like that, there was no other option.
That said, there are a couple of (secondary) characters whose language use diverges from the dialectal baseline, and intentionally so. One of them is the engineer working at the Vieille Montagne factory, whose short appearance in the film we have transcribed in the following extract.
Extract 1.Footnote 18 Across the bridge! January 8, 1971, the Vieille Montagne factory in Balen-Wezel. A group of dissatisfied workers, led by Piet Poppeliers, heads towards the factory's engineer. They would like him to go to the board of directors to discuss their dissatisfaction with wages (workers at other factories are earning 10 francs per hour, while they are only earning eight francs). Emphases are underlined, the “=” sign indicates latching. An English translation is given in italics below each line.
In the social hierarchy of the Vieille Montagne factory, the engineer takes up a higher position. This is why the workers turn to him with their grievances: They know that they have a better chance to be heard if the engineer were to voice their concerns to the board of directors. The higher position of the engineer is signaled by a number of semiotic contrasts. On a visual level, it has to be noted that the engineer's outfit is markedly different from those of the workers (see pictures 1 and 2): He is wearing glasses, a shirt and a tie, and a waistcoat with a beige mackintosh over it. The workers are all wearing the same blue-colored workman's clothes.
Picture 1. Engineer (middle), surrounded by Piet (furthest left),
Kris (furthest right) and the other workmen.
Picture 2. Piet (furthest left), the engineer (right),
Kris (furthest right) and the other workmen.
In addition to the visual contrast, the actors in this scene unmistakably create a linguistic contrast between the engineer and the workforce. Thus, the language of Piet, Kris, and Frans, the workers who have a turn at talk in this extract, is characterized by primary and secondary phonological dialect features (Taeldeman Reference Taeldeman and Hinskens2006) typical for the southern Campine (Belemans & Goossens Reference Belemans and Goossens2000):
• the half-close /e;/ is realized as diphthong [iə] (twieë ‘two’, line 18; ieëne ‘one’, line 19; nieë ‘no’, line 30, 31; menieër ‘sir’, line 33; see table 2 above);
• the half-close /o;/ is realized as diphthong [uə] (g'oerd ‘heard’, line 32; onnoeëzeleir ‘fool’, line 43; see table 2 above);
• the diphthong /εi/ is produced as the long monophthong [a;] (gaa ‘you’, line 13, 14);
• the diphthong /œy/ is pronounced as [ɔ;ə] (boik ‘stomach’, line 19) or [œ] (leusteren ‘listen’, line 15);
• the long /a;/ is produced as [ɔ;] (goan ‘go’, line 13; boazen ‘bosses’, line 19) or [ɔ] (al'mol ‘all’, line 32, 36; mor ‘but’, line 36).
In comparison, the engineer's vowels are much more standard-like:
• he pronounces the half-close /o;/ as a Standard Dutch [o;] (ook ‘also’, line 41; see table 2 above);
• he realizes the long /a;/ consistently as a Standard Dutch [a;] (al'maal ‘all’, line 11; maa ‘but’, line 11; daaraan ‘about that’, line 12; daarboven ‘up there’, line 16; daar ‘there’, line 21; maa ‘but’, line 23; gaan ‘go’, line 24; staan ‘stand’, line 24).
From a structural point of view, his long, monophthongal realizations of /œy/ as [œ;] (in luister ‘listen’, twice on line 21) hover in-between the Brabantic dialectal pronunciation of that vowel (as [œ]) and the Standard Dutch, diphtongal realization of it ([œy]), which would warrant their categorization as tussentaal. His deletions of the word-final /t/ in words such as wa ‘what’ (versus Standard Dutch wat, line 11), nie ‘not’ (versus Standard Dutch niet, line 21) and me ‘with’ (versus Standard Dutch met, line 21) are classic markers of tussentaal (see table 1 above). The engineer's morphosyntactic and lexical choices likewise are typical for tussentaal:
• he addresses the workers with the nonstandard pronoun ge ‘you’ instead of Standard Dutch jij, but instead of pronouncing it with a dialectal vowel (gaa) he opts for a more standard pronunciation (with a schwa, line 22, see table 1 above);
• similarly, he produces an intermediary form of the possessive pronoun used with that last noun (producing inflected uwe ‘your’, instead of uninflected Standard Dutch uw or inflected dialectal aawe, line 23);
• he realizes an enclitic form of the 2nd person pronoun, namely moet(e) ‘you have to’ (line 23), instead of Standard Dutch moet je (see table 1 above);
• he selects a number of vernacular lexical items such as content ‘satisfied’ (a loan from French, line 22) and frak ‘coat’ (line 23).
In structural terms, therefore, the engineer is using tussentaal, but it is the nondialectal character of his language use that appears to be most meaningful here: In accordance with his higher social status, which is also expressed through other semiotic means (namely, his clothes and glasses), he expresses himself in a higher, that is, more standardized variety than the dialect of the factory workers, who occupy a lower social position. A subtle detail that contributes to this contrast is that the engineer realizes the phoneme /r/ as the uvular trill [r], while the workmen use an alveolar [r]. By selecting this originally French uvular trill rather than the original Flemish [r], which is part of his default speech style, the actor might be suggesting that his character is French-speaking, or more precisely, bilingual in Dutch and French, which at the time of the narrative, in the early 1970s in Flemish Belgium, would not be unimaginable.Footnote 19 Also, his pronunciation of the word pakken ‘catch’ with a slightly prolonged and stressed last syllable, which is typical for French-speaking learners of Dutch, hints at phonological interference of French.Footnote 20
Hence, the language use of the engineer in Groenten uit Balen reveals its full social meaning only when analyzed in situ and placed in its local social ecology: In a film where most characters are dialect-speaking workers, the engineer's tussentaal is first and foremost indexical of his higher social position in the Vieille Montagne factory; it also reflects a trait of the time, namely, that members of the elite and those who had enjoyed higher education in Flanders tended to be predominantly French-speaking.
In the previous example, language use that on a structural basis can be termed tussentaal indexes a social meaning traditionally associated with Standard Dutch. In the next example, we demonstrate that language use that is structurally very similar to the engineer's speech style in extract 1 can also carry meanings conventionally associated with dialect use.
6. Tussentaal Functioning as Dialect
In our second example, taken from the 1984 series De burgemeester van Veurne ‘The mayor of Furnes’, we find one character that, on a linguistic level, diverges from the other characters in the series. This character is the mother of the protagonist, Mayor Terlinck. In the next extract, she makes her first appearance in the series.
Extract 2.Footnote 21 A rich mister (The mayor of Furnes). Terlinck walks across the lawn between a couple of small farms. He runs into his mother.
This scene is part of the narrative exposition and provides background information on the main characters, as is typical for the first episode of a series. In the preceding scenes, the audience was introduced to the protagonist Joris Terlinck, who is not just the mayor of Veurne but also the owner of a profitable cigar factory, and who finds his power besieged by the catholic opposition in his town. It soon turns out that Terlinck has a cold and calculating mind. His fortune comes from his marriage of convenience to the ailing Thérèse, a rich architect's daughter, and from the inheritance left to him by a rich widow. Terlinck used to be her bookkeeper and became a beneficiary under her will in a not entirely legitimate way.
In this scene, Terlinck's mother summarizes this biographical information explicitly: He has become a “rich mister” (line 41). The audience now finds out that this was not always the case. In a monologue (lines 44–52) which is redundant for Terlinck himself but clarifying for the viewers, the mother adds that the Terlinck family is quite lowborn, and that Terlinck is the child of a poor shrimper. This information is also visually illustrated through the contrast between Terlinck's exquisite mansion, in which a number of the preceding scenes were set, and the starkly furnished, dark fisherman's cottage, where he pays a visit to his mother (see pictures 3–6).
Picture 3. Terlinck having coffee at his mother's (1).
Picture 4. Terlinck having coffee at his mother's (2).
Picture 5. Terlinck having dinner with Thérèse at home (1).
Picture 6. Terlinck having dinner with Thérèse at home (2).
This social contrast, we would like to argue, is also indexed linguistically. Thus, the language used by Terlinck's mother deviates somewhat from that of the other, relatively more standard speaking characters. From a structural point of view, it is tussentaal: With the exception of one minor character, Mother Terlinck's language displays the highest rate of tussentaal variants and of (primary and secondary) dialect variants in the series, whereas the frequency of occurrence of the latter in her speech is too low to characterize it as a full-fledged dialect.
A particularly salient dialectal variant is the (originally Middle-Dutch) monophthong [i], which she uses instead of the Standard Dutch diphthong [εi] in the place name Kokside ‘Coxyde’ (line 1) and which is a typically West-Flemish dialect feature (Devos & Vandekerckhove Reference Devos and Vandekerckhove2005:43–44). Kokside seems to provide what Coupland (2007:124) has termed a “phono-opportunity”: The place name is particularly apt for “showcasing” a dialectal vowel that indexically reinforces the denotational link with the West-Flemish coastal town. In addition, Mother Terlinck produces two dialectal diphthongs [iə] instead of Standard Dutch monophthongal [e;] (menieër ‘mister’, menieëren ‘misters’, line 41, 42; see table 2). Note, however, that this dialect marker does not occur in every potential environment: In line 54, Mother Terlinck produces meneer with a Standard Dutch [e;]; the same holds for twee (line 48) and burgemeester (line 55). The old woman's speech can therefore hardly be called truly dialectal. It is mainly her use of the typical tussentaal markers that makes her way of speaking different from that of the other characters. Thus, she is the only character in this first episode that uses nonstandard forms of address, ge/gij ‘you’ (line 1, 34, 36); all other characters without exception address one another using the standard forms je ‘you’ and u ‘you’.
Other features generally regarded as typical for tussentaal include:
• the use of t-apocope (for example, nie, line 6; see table 1);
• h-procope (‘ij ‘he’, line 20; zieken'uis ‘hospital’, line 14, 15; see table 1);
• elisions (g[e] in ‘you in’, line 27; m[e] anders ‘me otherwise’, line 35; z[e] ‘ier ‘them here’, line 36);
• inflected articles (nen arme mens instead of standard een arme mens ‘a poor man’, line 19; ne rijke meneer instead of standard een rijke meneer, line 54; see table 1).
Again, these features are not used consistently, however. Sometimes Mother Terlinck does produce word-final [t] (for instance, lines 41, 48) and onset [h] (line 44); similarly, sometimes she does not inflect her adnominals (for example, de dokter instead of nonstandard den dokter ‘the doctor’, line 5). In other words, standard and nonstandard realizations alternate.
On a structural level, Mother Terlinck's language use in extract 2 is different from the engineer's in extract 1, but the two also have similarities. One difference, for example, is that the engineer consistently produces his tussentaal markers in a nonstandard way, whereas Mother Terlinck uses nonstandard features more eclectically. At the same time, both characters’ speech styles have to be described as cases of tussentaal; they can hardly be called dialectal, and they also significantly diverge from Standard Dutch due to the presence of a number of nonstandard linguistic features. Structurally, then, it would be safe to categorize both characters’ language use as tussentaal.
It would not be an exaggeration to claim that the actress playing Mother Terlinck is building a contrast between the rich misters of the series and herself, a poor fisherwoman. To this end she is utilizing both the topic of the conversation and the style of her language use. Thus, her speech contributes implicitly—indeed, indexically—to what she is trying to convey explicitly to Terlinck and to the viewer. Just as in extract 1, where linguistic means are recruited to convey a social gap between the engineer and factory workers, Mother Terlinck's language in extract 2 helps depict a social hierarchy. While Mayor Terlinck rose to the top of his hierarchy, his mother still finds herself at the bottom. However, unlike the engineer's tussentaal in extract 1, Mother Terlinck's tussentaal represents not the “higher” but the “lower” variety relative to the Standard Dutch spoken by the other characters in the series. Put differently, in extract 1, the engineer's tussentaal is meaningful because of its nondialectal quality, which places the speaker higher in the local social hierarchy. In contrast, in extract 2, Mother Terlinck's tussentaal is meaningful because of its nonstandard quality, its deviation from the language of other, higher-placed characters. What structurally is the same kind of language use, namely, tussentaal, can thus be read as dialect or Standard Dutch depending on the circumstances in which it is produced.
7. Discussion
On a structural level, extracts 1 and 2 adequately present tussentaal. But the engineer's tussentaal in extract 1 appears to carry overtones of Standard Dutch, in the sense that it is the higher, or more prestigious variety in that setting. Conversely, Mother Terlinck's tussentaal in extract 2 appears to evoke dialect use, in the sense of Bleichenbacher (Reference Bleichenbacher2008:59): Instead of letting the woman speak an authentic dialect, which an elderly, socially and geographically non-mobile fisherwoman from Coxyde would have spoken in the period between the two world wars, the creators of the series and the actress have decided to use a substitute language that would only display interference from the evoked language, namely, the dialect.Footnote 22 Through selectively incorporating dialect features, the fisherwoman's speech style conjures up social meanings typically as-sociated with dialect use, such as rurality, old age, manual labor, and low social status. Similar to what Vandekerckhove & Nobels (Reference Vandekerckhove and Nobels2010:667) have observed with regard to the chat language of Flemish teenagers, she is “us[ing] a dialect style without consistently using dialect.”
The preceding paragraphs thus demonstrate that it is necessary to analyze language use in its “volle situationaliteit” [full situationality] (Slembrouck & Van Herreweghe Reference Slembrouck and Van Herreweghe2004:874). An analysis based solely on a frequency count of structural elements in many cases fails to tap into the actual meanings that linguistic elements come to evoke in situated interaction, so that a qualitative, context-sensitive, interpretive analysis is needed to unlock what linguistic features in different environments can mean. A closer look at the two extracts reveals that what can be called tussentaal from a structural point of view often turns out to be something quite different seen from fictional characters’ perspectives, to the extent that it may have to be understood as dialect or Standard Dutch. Although different linguistic strategies (tuning up in the engineer's case, tuning down in the fisherwoman's case) may exhibit considerable structural overlap, the effect they produce may be quite different, to the point of evoking diametrically opposed indexical meanings. Tussentaal may even pose as the standard variety it is so often seen to be but a crummy or mal-adroit replacement for.
Viewing tussentaal as a particular register, our data effectively illustrate what Agha (Reference Agha2005:47) argues is a characteristic of all register use, namely, that “effects of register token use are not always consistent with the stereotypic values associated with the register's form types.” Tussentaal may evoke a range of stereotypic values (see above), but the effects of its use in particular ecologies may be entirely different. Tussentaal's bi- or multivalency (Woolard Reference Woolard1998) might, at least in part, account for its persistent structural unpredictability (see above): Whereas its general contours are more or less stable (it consists of a number of elements that distinguish it from Standard Dutch but are not strictly dialectal), its actual realization varies, most probably depending on speakers’ routine speech style, their relationship with others, the activity type, its keying, and the local linguistic ecology. For tuning up and tuning down relative to other participants’ ways of speaking, speakers have a number of options available to them, at every level of linguistic structure (phonology, morphology, syntax, and lexicon). Different choices at any of these levels may sometimes trigger very subtle differences (see, for example, the engineer's uvular trill in extract 1, which is not a regular ingredient of tussentaal, but which is highly meaningful in this particular context).
In any case, it is clear that tussentaal can hardly be attributed context-independent or essential meanings, such as rebelliousness, naturalness, a conspicuously relaxed character, or antiauthoritarian resentment. All of these meanings may be communicated through tussentaal, and in some contexts (such as some media genres) they may have become enregistered, stereotypic attributes of tussentaal. However, the point is that the meaning of a particular set of linguistic forms is, in principle, dependent on local linguistic ecologies and cannot be presumed to hold across the board: The engineer and the fisherwoman above are neither rebellious nor intentionally sloppy, nor are they taking any antiauthoritarian stance.
We suggest that on many occasions, including nonfictional ones, speakers produce meanings through tussentaal that go beyond its stereo-typical appreciation. In principle, this dependency between forms and ecologies is a necessary requirement for speaking of “informal Standard Dutch” altogether (see section 2). If tussentaal were essentially to convey stances such as rebelliousness and conspicuous relaxation in a context where dominant expectations put a premium on Standard Dutch, then the latter variety could only be an emblem of obedience, compliance with expectations, and non-relaxedness, complicating its use as an informal variety in casual conversation. Such determinisms can be avoided if we allow tussentaal as well as Standard Dutch and dialect to be much more multi-dimensional or polyvalent in their indexical significance, and, in particular, when we liberate tussentaal from the small harnessed set of social meanings it has been allowed to index as the ‘anti-language’ within an ideology of standardization. Doing so can help reconsider current claims that tussentaal is one of the symptoms of a pan-European destandardization process.
8. Destandardization?
The term destandardization mostly applies to changing attitudes toward linguistic normativity, which are not only observed in Flanders but also elsewhere in Europe. The current most widely used definition is the one formulated by Coupland & Kristiansen (Reference Coupland, Kristiansen, Kristiansen and Coupland2011:28), who take this term
[T]o refer to a possible development whereby the established standard language loses its position as the one and only ‘best language’. … Such a development would be equal to a radical weakening, and eventual abandonment, of the ‘standard ideology’ itself.
For Flanders in particular, van der Horst (Reference Vandekerckhove and Nobels2010:23) has hypothesized a process of destandardization, pointing at (among other things) a “widening of the norm,” “increased tolerance,” “decreasing fear of variation and ‘foreign elements’,” etc., while Willemyns (Reference Willemyns2013:245–246) has noticed (both for the Netherlands and for Flanders) “an important attitudinal change [that] is upgrading the prestige of intermediate varieties: people seem to take them more seriously and their use is more commendable.” Grondelaers & van Hout (Reference Grondelaers and van Hout2011) and Grondelaers et al. (Reference Grondelaers and van Hout2011) have powerfully argued, on the basis of experimental attitude research, that Flanders experiences a “standard language vacuum” (Grondelaers & van Hout Reference Grondelaers and van Hout2011): Since the erstwhile imported Standard Dutch is all but a virtual variety, exclusively used by Flemish news anchors, and given that no other way of speaking is consistently identified as “best” or most “beautiful” by subjects in the experiments invited to judge the regionally colored but formal spoken Dutch of teachers, no stand-in appears to be ready to replace the virtual norm (Grondelaers et al. Reference Grondelaers and van Hout2011:217–218).
It is undeniable that much more nonstandard language use can be observed in contexts traditionally associated with standard language use (for example, audio-visual media, school, etc.). It is also true that more and more language experts and/or exemplary speakers (authors, politicians, TV presenters, sociolinguists, official authorities) have contributed to an ambivalent attitude toward standard language use, in Flanders as elsewhere, through their explicit criticism of linguistic purism (in the case of sociolinguists) or their abundant recruitment of nonstandard language use in their writing, as well as their oratory or public communication (all others; compare Grondelaers & van Hout Reference Grondelaers and van Hout2011). In addition, the standard language itself has been increasingly portrayed as undesirably intellectualist, posh, and otherwise “uncool” (see Mugglestone Reference Mugglestone2003:280 on how Received Pronunciation in Britain has come to be seen as “talking posh” rather than “talking proper”).
In spite of these changes, however, a number of facts suggest that linguistic standardization, as a widespread, historically rooted view on the use, organization, and application of language (see Bauman & Briggs Reference Bauman and Briggs2003), may not have lost all of its feathers yet. First of all, the sheer increase in nonstandard language use in the public sphere, and on television in particular, ought not, in our view, be taken as an undiluted sign of the dwindling hold of standardization on the public mind or automatically be seen as a denial of standardization (compare Grondelaers & Kristiansen Reference Grondelaers and Kristiansen.2013:10). A quantitative increase does not by itself change the conditions within which these quantities are produced (compare Coupland Reference Coupland and Androutsopoulos2014:90). To be sure, while media allows for much more linguistic diversity in drama and comedy, shows, and other entertainment, the more “important” and authoritative genres (such as the news) are still the exclusive domain of the standard variety. Agha (Reference Agha2003:264) argues in this respect that standard language use may have to be less widespread than nonstandard language use in order for the former to retain social prestige (complete diffusion amounting to the devaluation of the standard).
But also the empirical data we have discussed indicate that nonstandard language use does not simply contradict but may even depend on the existence of a standard language ideal. The engineer and the fisherwoman signal status differences through inserting in their speech linguistic features (next to postural or sartorial ones) that have been firmly enregistered as “high” or “low” in the symbolic hierarchy that the process of linguistic standardization has come to construct (Agha Reference Agha2003, Bauman & Briggs Reference Bauman and Briggs2003). They are, in other words, orienting to a frame of standardization in which it makes sense to treat dialectal and standard features as hierarchically organized and to recruit them for flagging personae or stances that are considered higher or lower within that frame.
What these characters do, therefore, makes no sense unless it is viewed as constructed for a framework that producers and consumers of the discourse take for granted or see as the baseline of their linguistic ecology, namely, the framework of linguistic standardization. Seen from this perspective, tussentaal in Flemish telecinematic discourse in large part reproduces at the same time as it requires an ideology of standardization, rather than, as often suggested, announcing its approaching demise (see Grondelaers & Speelman Reference Grondelaers and Kristiansen.2013 for an analysis that goes beyond this traditional view). This also puts into perspective why many actors accused of “succumbing” to tussentaal in the fiction series they play in quite often turn out to be ardent supporters of Standard Dutch in nonfictional settings (see Van Hoof, in press).Footnote 23
For these reasons, we think it would be more useful, by analogy with analyses of late modernity (Bauman Reference Bauman2000, Giddens Reference Giddens1991), to characterize the current linguistic situation in Flanders as a case of late standardization rather than destandardization. This indicates that while linguistic standardization has been challenged by socioeconomic processes (globalization, migration), changes within public broadcasting that open up markets for nonstandard language use, as well as by linguists who explicitly criticize prescriptivist attitudes, it still appears to set the guidelines for much language use on television, and probably also on the other side of the screen. The term late standardization alludes to the fact that standard and vernacular language use are currently being reconfigured, with the latter becoming more prominent in contexts and genres where the standard language used to be de rigueur, while the tension between standardizing and vernacularizing forces is intensifying and their relationship becoming more complex (Coupland Reference Coupland, Junod and Maillat2010). At the same time, conceptualizing the present situation as a case of late standardization captures the fact that this tension is not entirely new, and that in some contexts, vernacular language use has always had (sometimes even considerable) presence.
Obviously, as a product of human ambition, an ideology of standardization is not impervious to change or an eventual collapse. However, the signs of such a demise, or at least of the diminishing relevance of the standard language for fiction series, would have to include, among other things, a less systematic distribution of dialect and standard language features over characters, situations, and stances that are hierarchically ordered. A fundamental linguistic regime change may also have to imply the nonconspicuous switches to a more dialectal way of speaking when character traits such as beauty, politeness, and refined manners are to be displayed; or, the inconspicuous use, by the same set of characters, of idiomatic dialect and standard language for similar purposes and in similar circumstances. While TV fiction and film would seem to be perfect venues for conjuring up worlds without standard languages, the persistent influence of linguistic standardization is shown in a wide range of contemporary fantasy TV series and films, such as Game of Thrones or The Hobbit. Game of Thrones typically illustrates hierarchical differences between and also within various aristocratic families not only by the lustre of their lodgings and clothes, but also by the amount of nonstandard language features in their speech. Thus, even in a fantasy world, standardization is our guiding light.
9. Conclusion
In this article, we have tried to demonstrate that the meaning of linguistic features needs to be established in situ, and that what looks like tussentaal at the structural level, in specific contexts of use must be interpreted as dialect or Standard Dutch. Consequently, the popularity of tussentaal and its ascendance in Flemish telecinematic discourse may effectively hide that in many cases, much use of this variety is meant and interpreted as Standard Dutch or dialect. Fictional characters’ moves up and down the linguistic ladder show that they orient to, rather than ignore, a long-standing standardization ideology in which these moves make sense. Consequently, the occurrence of tussentaal—or mixed Dutch—is not necessarily at loggerheads with linguistic standardization, but even reproduces it in accepting that it defines the limits of the linguistically possible. This complicates tussentaal's representation as a clear symptom of destandardization. Instead, this view calls for a perspective of late standardization, in which standardizing and vernacularizing forces condition, rather than cancel one another.
